Targeting cytotoxic cells with chimeric receptors for adoptive immunotherapy

Abstract

The present invention provides compositions and methods for regulating the specificity and activity of T cells. In one embodiment, the invention provides a type of chimeric antigen receptor (CAR) wherein the CAR is termed a "KIR-CAR" which is a CAR design comprising a component of a receptor naturally found on natural killer (NK) cells. In one embodiment, the NK receptor includes but is not limited to a naturally occurring activating and inhibitory receptor of NK cells known as a killer cell immunoglobulin-like receptor (KIR).

Description

(19) *EP004696718A2* (11) EP 4 696 718 A2 (12) EUROPEAN PATENT APPLICATION (43) Date of publication: 18.02.2026 Bulletin 2026 / 08 (21) Application number: 25208869.5 (22) Date of filing: 15.03.2014 (51) International Patent Classification (IPC): C07K 19 / 00 (2006.01) (52) Cooperative Patent Classification (CPC): C07K 16 / 18; C07K 14 / 70503; C07K 14 / 70535; C07K 2319 / 03; C07K 2319 / 33; C07K 2319 / 70 (84) Designated Contracting States: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR (30) Priority: 15.03.2013 US 201361793443 P (62) Document number(s) of the earlier application(s) in accordance with Art. 76 EPC: 19187938.6 / 3 623 380 14722891.0 / 2 970 426 (71) Applicants: • Novartis AG 4056 Basel (CH) • The Trustees of The University of Pennsylvania Philadelphia, PA 19104 (US) (72) Inventors: • MILONE, Michael C. Moorestown, NJ 08057 (US) • WANG, Enxiu Wayne, PA 19087 (US) (74) Representative: Mathys & Squire The Shard 32 London Bridge Street London SE1 9SG (GB) Remarks: •This application was filed on 15‑10‑2025 as a divisional application to the application mentioned under INID code 62. •Claims filed after the date of filing of the application / after the date of receipt of the divisional application (Rule 68(4) EPC). (54) TARGETING CYTOTOXIC CELLS WITH CHIMERIC RECEPTORS FOR ADOPTIVE IMMUNOTHERAPY (57) The present invention provides compositions and methods for regulating the specificity and activity of T cells. In one embodiment, the invention provides a typeof chimeric antigen receptor (CAR)wherein theCAR is termeda "KIR-CAR"which is aCARdesign comprising acomponent of a receptor naturally foundonnatural killer (NK) cells. In one embodiment, the NK receptor includes but is not limited to a naturally occurring activating and inhibitory receptor of NK cells known as a killer cell immunoglobulin-like receptor (KIR). EP 4 69 6 71 8 A 2 Processed by Luminess, 75001 PARIS (FR) Description

[0001] This application claims priority to U.S. Serial No.: 61 / 793,443 filed March 15, 2013, the entire contents which is incorporated herein by reference. STATEMENT REGARDING FEDERALLYSPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under PN2 EY016586 awarded by the National Institutes of Health (NIH). The government has certain rights in the invention. BACKGROUND OF THE INVENTION

[0003] With the use of gene transfer technologies, Tcells can be geneticallymodified to stably express antibody binding domains on their surface that endow the Tcells with specificities that are independent of the constraints imposed by the major histocompatibility complex (MHC).Chimericantigen receptors (CARs) represent syntheticproteinsexpressedonT- cells (CARTcells) that fuse an antigen recognition fragment of an antibody (e.g., an scFv, or single-chain variable region fragment) with an intracellular domain of the CD3-zeta chain. Upon interaction with a target cell expressing the scFv’s cognate antigen,CARsexpressed onTcell cells can trigger T-cell activation leading to target cell killing (also referred to as target cell lysis).Whencombinedwith additional costimulatory signals suchas the intracellular domain ofCD137orCD28, these receptors are also capable of generating proliferation. However, some of this proliferation appears to be antigen- independent, unlike normal T cell receptor (TCR) responses (Milone et al., 2009, Mol Ther 17(8):1453‑64). Artificial receptors do not fully reproduce the intracellular signal transduction produced by natural TCRbinding to antigenic peptide complexed with MHC molecules (Brocker, 2000, Blood 96(5):1999‑2001). The signaling defects may limit the long-term survival ofCARTcells uponadoptive transfer in theabsenceofhigh levelsof cytokines like IL‑2 (Loet al., 2010,ClinCancer Res 16(10):2769‑80). They also have altered regulation thatmight be beneficial in someanti-cancer applications (Loskog et al., 2006, Leukemia 20(10):1819‑28), but these regulatory defects also lead to potential challenges to controlling their "off-target" activity against normal tissues that also express antigen, even at extremely low levels. These "off-target" effects are a serious limitation to CAR-based therapeutics, and have resulted in probable deaths during early Phase I evaluation of CAR-modified T cells (Morgan et al., 2010, Mol Ther 18(4):843‑51).

[0004] Thus, there is a need in the art for alternative approaches for constructing CARs that overcome the limitations to current CAR-based therapeutics. The present invention addresses this unmet need in the art. SUMMARY

[0005] In a first aspect, the invention features a purified, or non-naturally occurring,NKR-CARcomprising one, twoor all of an extra-cellular antigen binding domain, a transmembrane domain, e.g., an NKR transmembrane domain, and a cytoplasmic domain, e.g., an NKR cytoplasmic domain.

[0006] In one embodiment, said NKR-CAR comprises an extra-cellular antigen binding domain; a transmembrane domainandanNKRcytoplasmicdomain. Inoneembodiment, saidNKR-CARcomprisesaKIR-CAR,e.g., anactKIR-CAR or inhKIR-CAR, a NCR-CAR, e.g., an actNCR-CAR, a SLAMF-CAR, e.g., an inhSLAMF-CAR, a FcR-CAR, e.g., CD16- CAR, e.g., an actCD16-CAR, or CD64-CAR, e.g., an actCD64-CAR, or a Ly49-CAR, e.g., an actLy49-CAR or inhLy49- CAR. In one embodiment, the NKR-CAR comprises a transmembrane domain and an extra-cellular antigen binding domain, and further comprising a hinge domain disposed between said transmembrane domain and said extra-cellular antigen binding domain.

[0007] In another aspect, the invention features a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, e.g., a nucleicacid comprisingaDNA,orRNA, sequence, e.g., amRNA, comprisingasequence that encodesaNKR-CAR describedherein. Inoneembodiment, thenucleicacid further comprisingasequence that encodesanadaptormoleculeor intracellular signaling domain that interacts with said NKR-CAR.

[0008] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line comprising a NKR-CAR described herein. In one embodiment, the cytotoxic cell further comprises an adaptor molecule or intracellular signaling domain that interacts with said NKR-CAR.

[0009] In another aspect, the invention features a method of making a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line comprising aNKR-CAR, described herein comprising introducing into a cytotoxic cell a nucleic acid, e.g., a mRNA, comprising a sequence that encodes a NKR-CAR, described herein. In one embodiment, the method further comprises making a NKR-CAR, described herein, in the cytotoxic cell.

[0010] In another aspect, the invention features amethod of treating a subject, e.g., amethod of providing an anti-tumor immunity in amammal, comprising administering to themammal an effective amount of a cytotoxic cell, e.g., a naturally or non-naturally occurring T cell, NK cell or cytotoxic T cell or NK cell line comprising a NKR-CAR described herein. 2 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0011] In another aspect, the invention features a purified, or non-naturally occurring, KIR-CAR comprising an extra- cellular antigen binding domain and a transmembrane domain, e.g., a KIR transmembrane domain, or cytoplasmic domain, e.g., an ITIM-containing cytoplasmic domain, or a KIR-cytoplasmic domain. In one embodiment, the KIR-CAR comprises an extra-cellular antigen binding domain, a transmembrane domain, and an ITIM-containing cytoplasmic domain, or a KIR-cytoplasmic domain.

[0012] In one embodiment, said transmembrane domain can interact with, e.g., bind, the transmembrane domain of DAP12. In one embodiment, said transmembrane domain comprises a positively charged moiety, e.g., an amino acid residue comprising a positively charged moiety, e.g., side chain. In one embodiment, said transmembrane domain comprises a KIR-transmembrane domain.

[0013] In one embodiment, said KIR-CAR is an activating KIR-CAR. In one embodiment, said KIR-CAR comprises a KIR-transmembrane domain. In one embodiment, said KIR-CAR is an inhibitory KIR-CAR. In one embodiment, said KIR- CAR comprises a KIR-cytoplasmic domain. In one embodiment, said KIR-CAR comprises an extra-cellular antigen binding domain and a transmembrane domain, e.g., a transmembrane domain comprising a positively charged moiety, e.g., an amino acid residue comprising a positively charged moiety, e.g., side chain, or a KIR-transmembrane domain.

[0014] In one embodiment, a KIR-CAR described herein comprises an antigen binding domain comprising an scFv. In oneembodiment, said antigenbindingdomain comprisesa singleVHdomain, e.g., a camelid, shark, or lamprey singleVH domain, or a single VH domain derived from a human or mouse sequence, or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. In one embodiment, said antigen binding domain comprises ananobody. In one embodiment, said antigen binding domain comprises a camelid VHH domain.

[0015] In one embodiment, a KIR-CAR described herein comprises an extracellular hinge domain. In one embodiment, the extracellular hinge domain is other than a KIR hinge domain, e.g., other than a KIR2DS2 hinge domain. In one embodiment, theextracellular hingedomain isderived fromanaturalmolecule. Inoneembodiment, theextracellular hinge domain is derived fromanaturalmolecule other than aKIR. In one embodiment, the extracellular hinge domain comprises a non-naturally occurring polypeptide sequence. In one embodiment, the extracellular hinge domain comprises the extracellular hinge from human CD8-alpha. In one embodiment, the extracellular hinge domain comprises a synthetic extracellular hinge. In one embodiment, the extracellular hinge domain is less than 50, 20, or 10 amino acids in length. In one embodiment, the extracellular hinge domain has fewer amino acids than a KIR2DS2 hinge domain.

[0016] In one embodiment, the KIR-CAR described herein is an actKIR-CAR. In one embodiment, said actKIR-CAR comprises a transmembrane domain comprising a positively charged moiety, e.g., an amino acid residue comprising a positively chargedmoiety, e.g., a positively charged side chain or an actKIR transmembrane domain. In one embodiment, said actKIR-CAR can interact with and promote signaling from an ITAM-containing polypeptide or adaptor molecule. In one embodiment, said actKIR-CAR can interact with and promote signaling from a DAP12 polypeptide. In one embodi- ment, said actKIR-CARcomprises aKIRDdomain. In one embodiment, said actKIR-CARcomprises aKIRD1domain. In one embodiment, said actKIR-CARcomprises aKIRD2domain. In one embodiment, said actKIR-CARsaid act KIR-CAR doesnot comprise aKIRDdomain. In oneembodiment, saidactKIR-CARcomprisesaKIR2DS2 transmembranedomain. In one embodiment, said actKIR-CAR further comprises a KIR2DS2 cytoplasmic domain. In one embodiment, said actKIR-CAR does not comprise a KIR D domain.

[0017] In one embodiment, the antigen binding domain of a KIR-CAR described herein binds an antigen present on a target cell, e.g., a cancer cell. In one embodiment, said antigen binding domain binds an antigen that is more highly expressedona target cell, e.g., a cancer cell, thananon-target cell, e.g., anon-cancerouscell, e.g., a noncancerouscell of the same type as the target cell. In one embodiment, said antigen binding domain is binds an antigen described herein.

[0018] In one embodiment, the KIR-CAR described herein is an inhKIR-CAR. In one embodiment, the inhKIR-CAR comprises an inhKIR transmembrane domain. In one embodiment, the inhKIR-CAR inhKIR-CAR comprises an ITIM- containing cytoplasmic domain, e.g., an inhKIR cytoplasmic domain, e.g., a KIR2DL or KIR3DL cytoplasmic domain. In one embodiment, the inhKIR-CARcomprises a transmembrane other than aKIR transmembrane, e.g., a transmembrane domain from PD‑1, CTLA4 or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors. In one embodiment, the inhKIR-CAR comprises a cytoplasmic domain from an inhibitory receptor other than a KIR, e.g., from PD‑1, CTLA4 or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors. In one embodiment, the inhKIR-CAR comprises a transmembrane and cytoplasmic domain from an inhibitory receptor other than a KIR, e.g., transmembrane and cytoplasmic domain, independently, from e.g., PD‑1, CTLA4 or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors. In one embodiment, said cytoplasmic domain comprises an ITIM. In one embodiment, the inhKIR-CAR comprises a KIR D domain. In one embodiment, the inhKIR-CAR comprises a KIR D0 domain. In one embodiment, the inhKIR-CAR comprises a KIR D1 domain. In one embodiment, the inhKIR-CAR comprises a KIR D2 domain. In one embodiment, the inhKIR-CAR does not comprise a KIR D domain.

[0019] In one embodiment, the antigen binding domain of the inhKIR-CARs described herein binds an antigen not present on a target cell, e.g., a cancer cell. In one embodiment, said antigen binding domain binds an antigen that is more highly expressed on a non-target cell, e.g., a non-cancer cell, than a target cell, e.g., cancerous cell, e.g., a cancerous cell 3 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 of the same type as the target cell. In one embodiment, said antigen binding domain binds desmoglein1 / 3 (DSG1 / 3). In an embodiment, an inhCAR, e.g., an inhTCAR or inhNKR-CAR, e.g., an inhKIR-CAR, and an actCAR, e.g., an actTCAR or actNKR-CAR, e.g., an actKIR-CAR, are provided in which the inhCAR comprises an antigen binding domain that targets desmoglein1 / 3 (DSG1 / 3) and the actCAR comprises an antigen binding domain that targets an antigen other than DSG1 / 3, e.g., EGFR. In an embodiment, this pair is used to treat anEGFRexpressing cancer, e.g., an adenocarcinomaof the lung or colon. In an embodiment the cancer cells express less DSG1 / 3 than non-cancer cells. In an embodiment this combination can minimize CAR-mediated attack of skin cells or squamous cells of the GI track (i.e. oral mucosa). In one embodiment, said antigen binding domain binds an ephrin receptor or a claudin.

[0020] In another aspect, the invention features a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, e.g., a nucleic acid comprising a DNA, or RNA, sequence, e.g., a mRNA, comprising (a) a sequence that encodes a KIR- CAR, e.g., a first KIR-CAR described herein. In one embodiment, said KIR-CAR, e.g., said first KIR-CAR, is an actKIR- CAR, e.g., an actKIR-CAR described herein. In one embodiment, said KIR-CAR, e.g., said first KIR-CAR, is an inhKIR- CAR, e.g., an inhKIR-CAR described herein. In one embodiment, said nucleic acid comprises a DNA sequence. In one embodiment, said nucleic acid comprises a RNA sequence, e.g., a mRNA sequence.

[0021] In an embodiment, said nucleic acid comprises sequence that encodes a KIR-CAR, e.g., an actKIR-CAR, and sequence that encodesan inhibitorymoleculecomprising:an inhKIRcytoplasmicdomain;a transmembranedomain,e.g., a KIR transmembrane domain; and an inhibitor cytoplasmic domain, e.g., an ITIMdomain, e.g., an inhKIR ITIMdomain. In anembodiment the inhibitorymolecule is anaturally occurring inhKIR, or a sequencesharingat least 50, 60, 70, 80, 85, 90, 95, or 99% homology with, or that differs by no more than 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, or 20 residues from, a naturally occurring inhKIR.

[0022] In an embodiment, said nucleic acid comprises sequence that encodes a KIR-CAR, e.g., an actKIR-CAR, and sequence that encodesan inhibitorymolecule comprising: aSLAM family cytoplasmic domain; a transmembranedomain, e.g., a SLAM family transmembrane domain; and an inhibitor cytoplasmic domain, e.g., a SLAM family domain, e.g., an SLAM family ITIM domain. In an embodiment the inhibitory molecule is a naturally occurring SLAM family member, or a sequencesharingat least 50,60,70,80,85,90,95,or99%homologywith,or thatdiffersbynomore than1,2,3, 4, 5,6, 7, 8, 9, 10, 15, or 20 residues from, a naturally occurring SLAM family member.

[0023] In one embodiment, said nucleic acid described herein further comprises (b) a sequence that encodes a second KIR-CAR described herein, e.g., a second KIR-CAR that is different from said first KIR-CAR. In one embodiment, (a) and (b) are disposed on the same nucleic acid molecule, e.g., the same vector, e.g., the same viral vector, e.g., a lenti-viral vector. In one embodiment, one of (a) and (b) is disposed on a first nucleic acid molecule, e.g., a first vector, e.g., a viral vector, e.g., a lenti-viral vector, and the other is disposed on a second nucleic acid molecule, e.g., a second vector, e.g., a viral vector, e.g., a lenti-viral vector. Inoneembodiment, saidfirstKIR-CARandsaidsecondKIR-CAR isanactKIR-CAR. In an embodiment, engagement of either act KIR-CAR alone is insufficient to trigger substantial levels of activation. In an embodiment, engagement of both the first and second actKIR-CAR gives an additive, or synergistic, level of activation. In one embodiment, said first KIR-CAR and said second KIR-CAR is an inhKIR-CAR. In one embodiment, one of said first KIR-CAR and said second KIR-CAR is an actKIR-CAR and the other is an inhKIR-CAR. In one embodiment, said actKIR- CAR isanactKIR-CARdescribedherein. Inoneembodiment, said inhKIR-CAR isan inhKIR-CARdescribedherein. Inone embodiment, the nucleic acid described herein comprises an actKIR-CAR described herein and an inhKIR-CAR described herein.

[0024] In an embodiment the nucleic further comprises (c) sequence that encodes an intracellular signaling domain, e.g., an adaptormolecule, which can produce an activating signal. In one embodiment, said intracellular signaling domain comprises an ITAM motif. In one embodiment, said sequence encodes a DAP 12 polypeptide comprising a DAP 12 intracellular signaling domain. In one embodiment, saidDAP12 polypeptide further comprises a transmembrane domain. In one embodiment, said DAP 12 polypeptide further comprises an extracellular domain. In one embodiment, each of (a), (b), and (c) are present on the samenucleic acidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector. In one embodiment, oneof (a), (b), and (c) isencodedonafirst nucleicacidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti- viral vector and a second and third of (a), (b), and (c) is encoded on a second nucleic acid molecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector. In one embodiment (a) is present on a first nucleic acidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector, and (b)and (c) arepresent onasecondnucleicacidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector. In another embodiment, (b) is present on a first nucleic acidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector, and (a)and (c) arepresent onasecondnucleicacidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector. In one embodiment, (c) is present on a first nucleic acidmolecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector, and (b) and (a) are present on a second nucleic acid molecule, e.g., a vector, e.g., a viral vector, e.g., a lenti-viral vector. In one embodiment, eachof (a), (b), and (c) are present on different nucleic acidmolecules, e.g., different vectors, e.g., viral vectors, e.g., a lenti-viral vectors.

[0025] In an embodiment, (i) the antigen binding domain of one of said first KIR-CAR said second KIR-CAR does not comprise a light chain variable domain and a heavy chain variable domain, (ii) the antigen binding domain of one of said firstKIR-CARsaidsecondKIR-CAR isanscFv,and theother is other is other thananscFv, (iii)whenpresent on thesurface 4 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 ofacell, theantigenbindingdomainsof saidfirstKIR-CARandsaidsecondKIR-CAR,associatewithoneanother less than if both were scFv antigen binding domains, (iv) wherein, when present on the surface of a cell, binding of the antigen binding domain of said first KIR-CAR to its cognate antigen is not substantially reduced by the presence of said second KIR-CAR, (v) the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence, or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule, (vi) the antigen bindingdomainofoneof saidfirstKIR-CARsaidsecondKIR-CAR, isanscFv,and theother comprisesasingleVHdomain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VHdomain derived fromahumanormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule, (vii) the antigenbindingdomain of one of said first KIR-CAR said second KIR-CAR, is an scFv, and the other comprises a nanobody, or (viii) the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, is an scFv, and the other comprises a camelid VHH domain.

[0026] In one embodiment, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR does not comprisea light chainvariabledomainandaheavychainvariabledomain. Inoneembodiment, theantigenbindingdomain of oneof said firstKIR-CARsaidsecondKIR-CAR isanscFv, and theother is other thananscFv. Inoneembodiment,when present on the surface of a cell, the antigen binding domains of said first KIR-CAR said second KIR-CAR, associate with one another less than if both were scFv antigen binding domains. In one embodiment, when present on the surface of a cell, binding of the antigen binding domain of said first KIR-CAR to its cognate antigen is not substantially reduced by the presence of said second KIR-CAR. In one embodiment, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived froma human ormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-like molecule. In one embodiment, the antigen binding domain of one of said first KIR-CAR said second KIR- CAR, is an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. Inoneembodiment, theantigenbindingdomainofoneof saidfirstKIR-CARsaid secondKIR-CAR, is an scFv, and the other comprises ananobody. In one embodiment, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, is an scFv, and the other comprises a camelid VHH domain.

[0027] Ina embodiment the nucleic acid comprises a sequence that encodes a TCAR. In one embodiment, said TCAR comprisesanantigenbindingdomain andanactivating cytoplasmic domain from theTcell receptor complexwithCD3e.g. CD3 zeta chain, CD3 epsilon chain, CD3 gamma chain, CD3 delta chain. In one embodiment, said TCAR comprises a costimulatory domain from costimulatory receptor e.g. CD28, CD137, CD27, ICOS or OX40.

[0028] In an embodiment (i) the antigen binding domain of one of said KIR-CAR said TCAR does not comprise a light chain variable domain and a heavy chain variable domain, (ii) the antigen binding domain of one of said KIR-CAR said TCAR isanscFv, and theother is other thananscFv, (iii) whenpresent on the surfaceof a cell, theantigenbindingdomains of saidKIR-CARandsaidTCAR,associatewithoneanother less than if bothwerescFvantigenbindingdomains, (iv)when present on the surface of a cell, binding of the antigen binding domain of said KIR-CAR to its cognate antigen is not substantially reduced by the presence of said second TCAR, (v) the antigen binding domain of one of said KIR-CAR said TCAR, comprises a single VHdomain, e.g., a camelid, shark, or lamprey singleVHdomain, or a singleVHdomain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-like molecule, (vi) the antigen binding domain of one of said KIR-CAR said TCAR, is an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule, (vii) the antigen binding domain of one of said KIR-CAR said TCAR, is an scFv, and the other comprises a nanobody, or (viii) wherein, the antigen binding domain of one of saidKIR-CARsaid TCAR, is an scFv, and the other comprises a camelid VHHdomain. In one embodiment, the antigen binding domain of one of said KIR-CAR said TCAR does not comprise a light chain variable domain and a heavy chain variable domain.

[0029] In one embodiment, the antigen-binding domain of one of said KIR-CAR said TCAR is an scFv, and the other is other than an scFv. In one embodiment, when present on the surface of a cell, the antigen binding domains of said KIR- CAR said TCAR, associate with one another less than if both were scFv antigen binding domains. In one embodiment, when present on the surface of a cell, binding of the antigen binding domain of said KIR-CAR to its cognate antigen is not substantially reducedby thepresenceof saidsecondTCAR. Inoneembodiment, theantigenbindingdomainofoneof said KIR-CAR said TCAR, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived froma human ormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. Inoneembodiment, theantigenbindingdomainofoneof saidKIR-CARsaidTCAR, isanscFv,and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. Inoneembodiment, theantigenbindingdomainofoneof saidKIR-CARsaidTCAR, isanscFv,and theother comprises ananobody. In oneembodiment, the antigenbindingdomain of oneof saidKIR-CARsaidTCAR, is an 5 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 scFv, and the other comprises a camelid VHH domain.

[0030] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxicTcell or cell of anNKcell line, e.g., NK92, comprising (a) a first KIR-CARdescribedherein. In oneembodiment, said cytotoxic cell is Tcell. In one embodiment, said cytotoxic cell is an NK cell. In one embodiment, said cytotoxic cell is from an NK cell line, e.g., an NK92 cell. In one embodiment, In one embodiment, said first KIR-CAR is an actKIR-CAR described herein. In one embodiment, In one embodiment, said first KIR-CAR is an inhKIR-CAR described herein.

[0031] In an embodiment, said cytotoxic cell comprises a KIR-CAR, e.g., an actKIR-CAR, and an inhibitory molecule comprising: an inhKIR cytoplasmic domain; a transmembrane domain, e.g., a KIR transmembrane domain; and an inhibitor cytoplasmic domain, e.g., an ITIMdomain, e.g., an inhKIR ITIMdomain. In anembodiment the inhibitorymolecule is a naturally occurring inhKIR, or a sequence sharing at least 50, 60, 70, 80, 85, 90, 95, or 99% homology with, or that differs by no more than 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, or 20 residues from, a naturally occurring inhKIR.

[0032] In an embodiment, said cytotoxic cell comprises a KIR-CAR, e.g., an actKIR-CAR, and an inhibitory molecule comprising: a SLAM family cytoplasmic domain; a transmembrane domain, e.g., a SLAM family transmembrane domain; andan inhibitor cytoplasmicdomain,e.g., aSLAMfamilydomain, e.g., anSLAMfamily ITIMdomain. Inanembodiment the inhibitorymolecule is a naturally occurring SLAM familymember, or a sequence sharing at least 50, 60, 70, 80, 85, 90, 95, or 99% homology with, or that differs by no more than 1, 2, 3, 4, 5,

[0033] In one embodiment, the cytotoxic cell further comprises (b) a second KIR-CAR described herein, e.g., a second KIR-CAR that is different from said first KIR-CAR. In one embodiment, one of said KIR-CAR and said second KIR-CAR is an actKIR-CAR and the other is an inhKIR-CAR. In one embodiment, said actKIR-CAR is an actKIR-CAR described herein. Inoneembodiment, oneof said inhKIR-CAR isan inhKIR-CARdescribedherein. In oneembodiment, the cytotoxic cell described herein comprises actKIR-CAR described herein and an inhKIR-CAR described herein.

[0034] In an embodiment the cytotoxic cell further comprises an intracellular signaling domain, e.g., an adaptor molecule, which can produce an activating signal, e.g., which is exogenous to said cell, which can produce an activating signal. In one embodiment, said intracellular signaling domain comprises an ITAM motif. In one embodiment, said intracellular signaling domain comprises a DAP 12 polypeptide comprising DAP 12 intracellular signaling domain. In one embodiment, said DAP 12 polypeptide further comprises a transmembrane domain. In one embodiment, said DAP 12 polypeptide further comprises an extracellular domain.

[0035] In anembodiment a cytotoxic cell comprises a first and secondKIR-CARdescribedhereinwherein (i) theantigen binding domain of one of said first KIR-CAR said second KIR-CAR does not comprise a light chain variable domain and a heavy chain variable domain, (ii) the antigen binding domain of one of said first KIR-CARsaid secondKIR-CAR is an scFv, and the other is other than an scFv, (iii) when present on the surface of a cell, the antigen binding domains of said first KIR- CARandsaid secondKIR-CAR,associatewith oneanother less than if bothwere scFvantigenbindingdomains, (iv)when present on the surface of a cell, binding of the antigen binding domain of said first KIR-CAR to its cognate antigen is not substantially reducedby thepresenceof said secondKIR-CAR, (v) theantigenbindingdomainofoneof said firstKIR-CAR said secondKIR-CAR, comprises a single VHdomain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VH domain derived froma human ormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-like molecule, (vi) the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived froma human ormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-like molecule, (vii) wherein, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, comprises an scFv, and the other comprises a nanobody, or (viii) the antigen binding domain of one of said first KIR-CAR said second KIR-CAR, comprises an scFv, and the other comprises a camelid VHH domain.

[0036] In one embodiment, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR does not comprise a light chain variable domain and a heavy chain variable domain. In one embodiment, the antigen-binding domain of one of said first KIR-CAR said second KIR-CAR is an scFv, and the other is other than an scFv. In one embodiment, when present on the surface of a cell, the antigen binding domains of said first KIR-CAR said second KIR- CAR, associate with one another less than if both were scFv antigen binding domains. In one embodiment, when present on thesurfaceofacell, bindingof theantigenbindingdomainof said firstKIR-CARto its cognateantigen isnot substantially reduced by the presence of said secondKIR-CAR. In oneembodiment, the antigen bindingdomain of one of said first KIR- CAR said second KIR-CAR, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. Inoneembodiment, theantigenbindingdomainofoneof saidfirstKIR-CARsaid second KIR-CAR, comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., afibronectin type III antibody-likemolecule. Inoneembodiment, theantigen-bindingdomainofoneof said first KIR-CAR said second KIR-CAR, comprises an scFv, and the other comprises a nanobody. In one embodiment, the antigen-binding domain of one of said first KIR-CARsaid secondKIR-CAR, comprises an scFv, and the other comprises a camelid VHH domain. 6 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0037] In an embodiment a cytotoxic cell comprises KIR-CARs as described herein and further comprises a TCAR. In oneembodiment, saidTCARcomprisesanantigenbindingdomainandaprimarystimulationdomain. Inoneembodiment, said TCAR comprises a costimulation domain.

[0038] In an embodiment the cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line, e.g., an NK92 cell, comprises a nucleic acid as described herein; or a KIR-CAR encoded by a nucleic acid described herein. In one embodiment, said cytotoxic cell is Tcell. In one embodiment, said cytotoxic cell is an NK cell. In one embodiment, said cytotoxic cell is from an NK cell line, e.g., NK92.

[0039] Inanotheraspect, the invention featuresmethodsofmakingacell describedherein comprising, introducing intoa cytotoxic cell, a nucleic acid described herein into said cell. In one embodiment, said method comprises forming in a cytotoxic cell, a KIR-CAR of any of claims 14 to 66.

[0040] In another aspect, the invention featuresmethods of treating a subject, e.g., amethod of providing an anti-tumor immunity in a mammal, comprising administering to the mammal an effective amount of a cell described herein. In one embodiment, said cell is autologous. Inoneembodiment, said cell is allogenic. Inoneembodiment, the cell isTcell, e.g., an autologous Tcell. In one embodiment, the cell is an allogeneic Tcell. In one embodiment, the cell is an NK cell, e.g., an autologousNKcell. In one embodiment, the cell is an allogeneicNKcell. In one embodiment, the cell is cell fromanNKcell line, e.g., NK92. In one embodiment, said mammal is a human. In one embodiment, the method further comprises evaluating said mammal, e.g., human, for a side effect of said treatment. In one embodiment, said side effect comprises acute respiratory distress syndrome, febrile neutropenia, hypotension, encephalopathy, hepatic transaminitis, seizure, or macrophage activation syndrome. In one embodiment, the method further comprises treating said human having a side effect with anti-cytokine agent, e.g., a tumor necrosis factor antagonist, e.g., a TNF-Ig fusion, e.g., etanercept, an IL‑6 antagonist, e.g., an IL‑6 receptor antagonist, e.g., an anti-IL6 receptor antibody, e.g., tocilizumab, or a corticosteroid. In one embodiment, treating comprises administering an anti-IL6 receptor antibody to said human. In one embodiment, the methodcomprises treatingamammal, e.g., a human, havingadiseaseassociatedwithexpressionofmesothelin orCD19. Inoneembodiment, themethodcomprises treatingamammal, e.g., a human, havingadisorder associatedwith unwanted cell proliferation, e.g., cancer. In one embodiment, said disorder is pancreatic carcinoma,mesothelioma, lung carcinoma, ovarian carcinoma, leukemia or lymphoma.

[0041] In another aspect, the invention features a purified, or non-naturally occurring, NCR-CAR, e.g., an activating NCR-CAR, comprising an extra-cellular antigen binding domain, a transmembrane domain, e.g., a transmembrane domain comprising apositively chargedmoiety, e.g., an amino acid residue comprising apositively chargedmoiety, e.g., a positively charged side chain or an NCR transmembrane domain, and a cytoplasmic domain, e.g., a NCR cytoplasmic domain. In one embodiment, said NCR-CAR comprises an a transmembrane domain comprising a positively charged moiety, e.g., anaminoacid residue comprisingapositively chargedmoiety, e.g., a positively chargedside chain, e.g., NCR transmembrane domain, e.g., a NKp30, NKp44, or NKp46 cytoplasmic domain. In one embodiment, said NCR-CAR comprises a cytoplasmic domain which can interact with an adaptor molecule or intracellular signaling molecule comprising, e.g., a DAP12, FcRγ or CD3 ζ cytoplasmic domain. In one embodiment, said NCR-CAR, e.g., a NKp30- CAR, comprisesa transmembranedomainwhichcan interactwithanadaptormoleculeor intracellular signalingmolecule, e.g., DAP12. In one embodiment, said NCR-CAR comprises a NKp46-CAR. In one embodiment, said NKp46-CAR, comprises a transmembrane domain comprising a positively charged moiety, e.g., an amino acid residue comprising a positively charged moiety, e.g., a positively charged side chain or, e.g., an NCR transmembrane domain, which can interactwithanadaptormoleculeor intracellular signalingmolecule, e.g., onehavingaFcRγorCD3ζcytoplasmicdomain. In one embodiment, said NCR-CAR described herein further comprises a hinge domain disposed between said transmembrane domain and said an extra-cellular antigen binding domain.

[0042] In another aspect, the invention features a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, e.g., a nucleicacid comprisingaDNA,orRNA,sequence, e.g., amRNA, comprisingasequence that encodesaNCR-CAR described herein. In one embodiment, the nucleic acid comprises a sequence that encodes aNKp30-CARand optionally, an adaptor molecule or intracellular signaling molecule, e.g., DAP12. In one embodiment, said NCR-CAR, e.g., NKp46- CAR,comprisesa transmembranedomaincomprisingapositively chargedmoiety, e.g., anaminoacid residuecomprising a positively charged moiety, e.g., a positively charged side chain or an NCR transmembrane domain which can interact with an adaptor molecule or intracellular signaling molecule, e.g., a FcRγ or CD3 ζ molecule. In one embodiment, the nucleic acid further comprises sequence encoding an adaptor molecule or intracellular signaling molecule, which e.g., comprises a DAP12, FcRγ or CD3 ζ.

[0043] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line comprising a NCR-CAR described herein. In one embodiment, the cytotoxic cell further comprises an adaptor molecule or intracellular signaling molecule, which e.g., comprises a DAP12, FcRγ or CD3 ζ cytoplasmic domain. In one embodiment, the cytotoxic cell comprises a NKp30-CAR and optionally, an adaptor molecule or intracellular signaling molecule, e.g.,DAP12. In one embodiment, said NKp46-CAR comprises a transmembrane domain which can interact with an adaptor molecule or intracellular signaling molecule, e.g., a FcRγ or CD3 ζ molecule.

[0044] In another aspect, the invention features amethod ofmaking a cell described herein comprising, introducing into 7 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 acytotoxic cell, a nucleic acid comprisinga sequence that encodesaNCR-CARdescribedherein. In oneembodiment, the method comprises forming in a cytotoxic cell, a NCR-CAR described herein.

[0045] In another aspect, the invention features amethod of treating a subject, e.g., amethod of providing an anti-tumor immunity in amammal, comprising administering to themammal an effective amount of a cell described herein, e.g., a cell of claim described herein comprising NCR-CAR described herein.

[0046] In another aspect, the invention features a purified, or non-naturally occurring, SLAMF-CAR, e.g., an inhibitory SLAMF-CAR, comprising an extra-cellular antigen binding domain, a transmembrane domain, e.g., a transmembrane domain comprising apositively chargedmoiety, e.g., an amino acid residue comprising apositively chargedmoiety, e.g., a positively charged side chain, e.g., a SLAMF transmembrane domain, and a SLAMF cytoplasmic domain. In one embodiment, said SLAMF-CAR comprises a SLAMF, CD48, CD229, 2B4, CD84, NTB-A, CRACC, BLAME, or CD2F‑10 cytoplasmic domain. In one embodiment, said SLAMF-CAR further comprises a hinge domain, disposed between said transmembrane domain and said an extra-cellular antigen binding domain.

[0047] In another aspect, the invention features a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, e.g., a nucleic acid comprising a DNA, or RNA, sequence, e.g., a mRNA, comprising a sequence that encodes a SLAMF- CAR described herein.

[0048] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic T cell or NK cell line comprising a SLAMF ‑CAR described herein.

[0049] In another aspect, the invention features a method of making a cytotoxic cell comprising a SLAMF ‑CAR described herein, comprising, introducing into a cytotoxic cell a nucleic acid comprising a sequence that encodes a SLAMF-CAR described herein. In one embodiment, the method comprises forming in a cytotoxic cell, a SLAMF ‑CAR described herein.

[0050] In another aspect, the invention features amethod of treating a subject, e.g., amethod of providing an anti-tumor immunity in amammal, comprising administering to themammal an effective amount of a cell comprising a SLAMF ‑CAR described herein.

[0051] In another aspect, the invention features a purified, or non-naturally occurring, FcR-CAR, e.g., CD16-CAR, e.g., an activating CD16-CAR or a CD64-CAR, e.g., an activating CD64-CAR, comprising an extra-cellular antigen binding domain, a transmembrane domain, and aCD16 orCD64 cytoplasmic domain. In one embodiment, said FcR-CAR is aCD 16-CAR. In one embodiment, said FcR-CAR is a CD64-CAR. In one embodiment, said FcR-CAR can interact with an adaptor molecule or intracellular signaling molecule, e.g., a FcRγ or CD3 ζ domain, e.g., via a transmembrane domain, e.g., a transmembranedomain comprising apositively chargedmoiety, e.g., anaminoacid residuecomprising apositively charged moiety, e.g., a positively charged side chain or e.g., a CD16 or CD64 transmembrane domain. In one embodiment, said FcR-CAR further comprises a hinge domain, disposed between said transmembrane domain and said an extra-cellular antigen binding domain.

[0052] In another aspect, the invention features a purified or non-naturally occurring, nucleic acid, e.g., a nucleic acid comprising a DNA, or RNA, sequence, e.g., amRNA comprising a sequence that encodes a FcR -CAR described herein. In one embodiment, the nucleic acid further comprises an adaptormolecule or intracellular signalingmolecule comprising a cytoplasmic activation domain, e.g., FcRγ or CD3 ζ cytoplasmic domain. In one embodiment, said FcR -CAR and said cytoplasmic activation domain are disposed on separated nucleic acid molecules, e.g., separate vectors, e.g., separate viral vectors, e.g., separate lenti-viral vectors.

[0053] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line comprising a FcR - CAR described herein. In one embodiment, the cytotoxic cell further comprises a cytoplasmic activation domain, e.g., FcRγ or CD3 ζ cytoplasmic domain.

[0054] In another aspect, the invention features a method of making a cell comprising a FcR ‑CAR described herein, comprising, introducing into a cytotoxic cell, a nucleic acid comprising a sequence that encodes a FcR ‑CAR described herein. In one embodiment, the method comprises forming in a cytotoxic cell, a FcR ‑CAR described herein.

[0055] In another aspect, the invention features amethod of treating a subject, e.g., amethod of providing an anti-tumor immunity in a mammal, comprising administering to the mammal an effective amount of a cell comprising a FcR ‑CAR described herein.

[0056] In another aspect, the invention features purified, or non-naturally occurring, Ly49-CAR comprising an extra- cellular antigen binding domain, and a transmembrane domain, e.g., a Ly49-transmembrane domain, or a cytoplasmic domain, e.g., an ITIM-containing cytoplasmic domain, e.g., a Ly49-cytoplasmic domain. In one embodiment, the Ly49- CAR comprises a transmembrane domain and a Ly49-cytoplasmic domain. In one embodiment, said Ly49-CAR is an activating Ly49-CAR, e.g., Ly49D or Ly49H. In one embodiment, said Ly49-CAR comprises a positively charged transmembrane domain, e.g., a positively charged Ly49 transmembrane domain. In one embodiment, said Ly49-CAR can interact with an ITAM-containing cytoplasmic domain, e.g., DAP 12. In one embodiment, said Ly49-CAR comprises a Ly49-transmembrane domain. In one embodiment, said KIR-CAR is an inhibitory Ly49-CAR, e.g., Ly49Aor Ly49C. In one embodiment, said Ly49-CARcomprises an ITIM-containing cytoplasmic domain, e.g., a Ly49-cytoplasmic domain. In one embodiment, said Ly49-CAR comprises a Ly49-transmembrane domain or a Ly49-cytoplasmic domain selected, 8 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 independently from Ly49A-Ly49W. In one embodiment, said Ly49-CAR further comprises a hinge domain, disposed between said transmembrane domain and said an extra-cellular antigen binding domain.

[0057] In another aspect, the invention features a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, e.g., a nucleic acid comprisingaDNA,orRNA, sequence, e.g., amRNA, comprisingasequence that encodesaLy49-CAR described herein. In one embodiment, the nucleic acid further comprises a cytoplasmic activation domain, e.g., DAP12 cytoplasmic domain. In one embodiment, said Ly49-CAR and said cytoplasmic activation domain are disposed on separate nucleic acid molecules, e.g., separate vectors, e.g., separate viral vectors, e.g., separate lenti-viral vectors.

[0058] In another aspect, the invention features a cytotoxic cell, e.g., a naturally or non-naturally occurring Tcell, NK cell or cytotoxic Tcell or NK cell line comprising a Ly49-CAR described herein. In one embodiment, the cytotoxic cell further comprises a cytoplasmic activation domain, e.g., DAP12 cytoplasmic domain.

[0059] In another aspect, the invention features a method of making a cell comprising a Ly49-CAR described herein, comprising, introducing into a cytotoxic cell, a nucleic acid comprising a sequence that encodes a Ly49-CAR described herein into said cell.

[0060] In another aspect, the invention features a method of making a cell comprising a Ly49-CAR described herein, comprising, forming in a cytotoxic cell, a Ly49-CAR described herein.

[0061] In another aspect, the invention features amethod of treating a subject, e.g., amethod of providing an anti-tumor immunity in amammal, comprising administering to themammal an effective amount of a cell described herein, e.g., a cell comprising a Ly49-CAR described herein.

[0062] In another aspect, the invention features a cell comprising, e.g., a cytotoxic cell, comprising a first non-naturally occurring chimeric membrane embedded receptor comprising an antigen binding domain and a second non-naturally occurring chimeric membrane embedded receptor comprising an antigen binding domain wherein, (i) the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor does not comprise a light chain variable domain and a heavy chain variable domain, (ii) the antigen binding domain of one of said first and said second non-naturally occurring chimericmembrane embedded receptor comprises an scFv, and the other is other thananscFv, (iii) whenpresent on the surfaceof a cell, the antigenbindingdomains of said first and said secondnon- naturally occurring chimeric membrane embedded receptor, associate with one another less than if both were scFv antigen binding domains, (iv) when present on the surface of a cell, binding of the antigen binding domain of said first non- naturally occurring chimeric membrane embedded receptor to its cognate antigen is not substantially reduced by the presence of said second non-naturally occurring chimeric membrane embedded receptor, (v) the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-like molecule, (vi) the antigen binding domain of one of said first said second non-naturally occurring chimeric membrane embedded receptor, comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule, (vii) theantigenbindingdomainof oneof said first andsaid secondnon-naturally occurring chimericmembrane embedded receptor, comprises an scFv, and the other comprises a nanobody, and (viii) the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor, comprises an scFv, and theother comprises a camelidVHHdomain. In one embodiment, said cell is Tcell. In one embodiment, said cell is an NK cell. In one embodiment, said cell is from an NK cell line, e.g., NK92. In one embodiment, one of said first and said second non-naturally occurring chimeric membrane embedded receptors is a TCAR. In one embodiment, both of said first and said second non-naturally occurring chimeric membrane embedded receptors is a TCAR. In one embodiment, one of said first and said second non-naturally occurring chimeric membrane embedded receptors is a NKR-CAR, e.g., a KIR-CAR. In one embodiment, both of said first and said second non-naturally occurring chimeric membrane embedded receptors is a NKR-CAR, e.g., a KIR-CAR. In one embodiment, the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor does not comprise a light chain variable domain and a heavy chain variable domain. In one embodiment, when present on the surface of a cell, the antigen binding domains of said first and said second non-naturally occurring chimeric membrane embedded receptor, associate with one another less than if bothwere scFv antigen binding domains. In one embodiment, when present on the surface of a cell, binding of the antigen binding domain of said first non-naturally occurring chimeric membrane embedded receptor to its cognate antigen is not substantially reduced by the presence of said second non- naturally occurringchimericmembraneembedded receptor. Inoneembodiment, theantigenbindingdomainofoneof said first and said second non-naturally occurring chimeric membrane embedded receptor, comprises a single VH domain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VHdomain derived fromahumanormouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a fibronectin type III antibody-likemolecule. Inoneembodiment, theantigen binding domain of one of said first said second non-naturally occurring chimeric membrane embedded receptor, comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence or a non-antibody scaffold, e.g., a fibronectin, e.g., a 9 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 fibronectin type III antibody-like molecule. In one embodiment, the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor, comprises an scFv, and the other comprises a nanobody. In one embodiment, the antigen binding domain of one of said first and said second non-naturally occurring chimeric membrane embedded receptor, comprises an scFv, and the other comprises a camelid VHH domain. In one embodiment, the invention comprises a nucleic acid, e.g., a purified or non-naturally occurring, nucleic acid, comprising a sequence that encodes a first and second non-naturally occurring chimericmembrane embedded receptor comprising an antigen binding domain described herein. In one embodiment, the invention comprises a method of making a cell described herein comprised introducing into a cell the nucleic acid described herein. In one embodiment, the invention comprises amethod ofmaking a cell described herein, comprising, forming in a cytotoxic cell, a first and said second non- naturally occurring chimeric membrane embedded receptor described herein. In one embodiment, the invention comprises a method of treating a subject e.g., a method of providing an anti-tumor immunity in a mammal, comprising administering to the mammal an effective amount of a cell described herein.

[0063] In another aspect, the invention features a kit comprising a cell or nucleic acid described herein.

[0064] In another aspect, the invention features an isolated nucleic acid sequence encoding a KIR-CAR (killer cell immunoglobulin receptor-like-chimeric antigen receptor), wherein the isolated nucleic acid sequence comprises the nucleic acid sequence of an antigen binding domain and a KIR or fragment thereof. In one embodiment, the antigen binding domain is selected from the group consisting of a murine antibody, a humanized antibody, a human antibody, a chimeric antibody, and a fragment thereof. In one embodiment, the fragment is a Fab or an scFv. In one embodiment, the KIR is selected from the group consisting of an activating KIR, an inhibitory KIR, and any combination thereof. In one embodiment, at least one hinge region has been removed from the activating KIR.

[0065] In another aspect, the invention features an isolated KIR-CAR (killer cell immunoglobulin-like receptor-chimeric antigen receptor) comprising an antigen binding domain and a KIR or fragment thereof. In one embodiment, the antigen binding domain is selected from the group consisting of a murine antibody, a humanized antibody, a human antibody, a chimeric antibody, and a fragment thereof. In one embodiment, the fragment is a Fab or an scFv. In one embodiment, the KIR is selected from the group consisting of an activating KIR, an inhibitory KIR, and any combination thereof. In one embodiment, at least one hinge region has been removed from the activating KIR.

[0066] In another aspect, the invention features a composition comprising at least two KIR-CARs, wherein the first KIR- CARcomprises an antigen binding domain andan activatingKIRor fragment thereof and the secondKIR-CARcomprises anantigenbindingdomainandan inhibitoryKIRor fragment thereof. Inoneembodiment, theantigenbindingdomain in the first KIR-CAR is specific for an antigen present on a tumor and the antigen binding domain in the second KIR-CAR is specific for an antigen present on a normal cell.

[0067] In another aspect, the invention features a geneticallymodifiedTcell comprising at least twoKIR-CARs,wherein the first KIR-CAR comprises an antigen binding domain and an activating KIR or fragment thereof and the second KIR- CAR comprises an antigen binding domain and an inhibitory KIR or fragment thereof. In one embodiment, the antigen binding domain in the first KIR-CAR is specific for an antigen present on a tumor and the antigen binding domain in the second KIR-CAR is specific for an antigen present on a normal cell. In one embodiment, the cell is a T cell.

[0068] In another aspect, the invention features amethod of providing an anti-tumor immunity in amammal, themethod comprising administering to themammal an effective amount of a cell comprising at least two KIR-CARs, wherein the first KIR-CAR comprises an antigen binding domain and an activating KIR or fragment thereof and the second KIR-CAR comprises an antigen binding domain and an inhibitory KIR or fragment thereof. In one embodiment, the antigen binding domain in the firstKIR-CAR is specific for anantigenpresent ona tumorand theantigenbindingdomain in the secondKIR- CAR is specific for an antigen present on a normal cell, thereby controlling the off-target activity of the cell. In one embodiment, the cell is a T cell. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The following detailed description of preferred embodiments of the inventionwill be better understoodwhen read in conjunction with the appended drawings. For the purpose of illustrating the invention, there are shown in the drawings embodiments which are presently preferred. It should be understood, however, that the invention is not limited to the precise arrangements and instrumentalities of the embodiments shown in the drawings. Figure 1, comprising Figures 1A and 1B, is a series of schematics showing the structure of naturally occurring inhibitory and activating KIRs (Figure 1A) and a scFv-based activating KIR-CAR (Figure 1B). Figure 2 is a schematic representation of the lentiviral vector used to deliver an activating KIR-based CAR in combination with the DAP12 signaling molecule. Figure 3 is an image demonstrating that a mesothelin-specific actKIR-CARs can be efficiently expressed on the surface of primary human T cells. Human T cells were stimulated with anti-CD3 / anti-CD28 microbeads and transduced with the indicated CAR or mock transduced and expanded ex vivo. The expression was detected using 10 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 a biotinylated goat-anti-mouse F(ab)2-specific polyclonal IgG (Jackson Immunologics) followed by staining with streptavidin-PE. Figure 4 is an image demonstrating that Tcells expressing the SS1 actKIR-CAR exhibited cytotoxic activity towards target K562 cells engineered to express the mesothelin ligand (KT-meso). Human Tcells were stimulated with anti- CD3 / anti-CD28 microbeads, transduced with the indicated CAR or mock transduced and expanded ex vivo. 105 CFSE-labeled K562 cells expressing mesothelin (KT-meso) or wild-type control K562 were incubated with varying ratios of CAR-expressing Tcells for 16 hours at 37°C, 5% CO2. The K562 target cells were then enumerated by flow cytometry usingcountbright beadsandaviability stain (7AAD). ThepercentageofK562cells lysed (percent lysis)was calculated by subtracting the number of viable target cells remaining after incubation with effector T cells from the number of viable K562 remaining after overnight culture without effector Tcells, and then dividing by the number of viable K562 remaining after overnight culture without effector T cells. Figure5, comprisingFigure5Aand5B, isa seriesof schematics showinganactivatingKIRCAR inwhich theKIR2DS2 hingewas removed (KIR2SCAR).Basedupon the kinetic segregationmodel of TCRactivation diagrammed inFigure 5A, it is believed that the mesothelin-specific SS1 KIR CAR has a hinge that is too long to permit appropriate segregation. Therefore making the mesothelin-specific KIR CAR hinge shorter is believed to improve the function. Figure 5B is a schematic showing that the SS1 scFvwas fused to the KIR transmembrane domain without the two Ig- like domains from KIR2DS2 as the hinge. Figure6, comprisingFigures6Aand6B, isaseriesof imagesdemonstrating thatSS1scFvbasedKIRS2CARexhibits enhancedcytolytic activity towardsmesothelin-expressing target cells comparedwith theCAR formedby fusionof the SS1scFvonto full lengthwildtypeKIR2DS2.PrimaryhumanTcellswerestimulatedwithCD3 / 28microbeads followed by lentiviral transductionwith either theSS1-KIR2DS2activatingKIR-CAR,SS1-KIRS2activatingKIRCAR, theSS1- zeta CAR.Mock non-transduced Tcells (NTD) were used as a control. The Tcells were expanded until the end of log- phase growth. The surface expression of the SS1-specific CARs was determined by flow cytometry using a biotinylated goat anti-mouse F(ab)2 specific polyclonal antibody followed by streptavidin-PE detection as shown in Figure 6A.Shown in Figure 6B,K562 target cellswith orwithoutmesothelin and stainedwithCFSEweremixedwith theeffectorTcells characterized inFigure6Aas indicatedusingvaryingeffectorTcell to target ratios ranging from10:1 to1:1.TargetK562cell lysiswasassessedusingflowcytometry todetermine the%ofviableCFSE+cellsasdescribed for Figure 4. Data shown is the calculated % target cell lysis compared against target cells without effector cells. Figure7, comprisingFigures7Aand7B, isaseriesof imagesshowingco-expressionof theCD19actKIR-CARand the SS1 inhKIR-CAR. Jurkat NFAT-GFP reporter cells were transduced with the indicated KIR CAR or non-transduced (NDT) andmixed 1: 1 with target cells with or without the CD19 andmesothelin antigens as indicated. Results shows GFP expression at 24 hours following mixing of Jurkat and Target cells (Figure 7A). Figure 7B shows surface expression of themesothelin andCD19 idiotypes as determined by stainingwith amesothelin-Fc fusion protein and a monoclonal antibody specific for the FMC63 anti-CD19 scFv idiotype. Figure 8, comprisingFigures8A through8C, is a series of imagesdemonstrating co-expression ofwild-typePD‑1with both an activating KIR-based CAR or TCR-zeta based CAR targeting mesothelin. Primary human T cells were stimulated with CD3 / 28microbeads followed by lentiviral transduction with either the SS1-KIRS2 activating KIRCAR or the SS1-zeta CAR.Mock non-transduced cells (NTD) were used as a negative control. The Tcells were expanded over 9 days, and surface CAR expression was determined by staining with mesothelin-Fc followed by a goat-anti- humanFcspecificantibodyconjugated toPE (Figure8A).K562cell lines (wildtype [wt],mesothelin expressing [meso] or mesothelin and PD-L1 co-expressing [meso-PDL1]) were stained using the CAK1 anti-mesothelin specific monoclonal antibody to confirm mesothelin expression on the targets (Figure 8B). The primary human T cells transduced as shown in Figure 8Awere electroporatedwith 10 ug of in vitro transcribed RNAencodingwild-type PD1 using a BTX ECM830 electroporator (PD1+) or mock transfected (PD1-). The surface expression of PD‑1 was expressed using an anti-PD1 monoclonal conjugated to APC (Figure 8C). Figure 9 is an image demonstrating that the combination of co-expressingwild-typePD‑1with both an activating KIR- based CAR and TCR-zeta based CAR targeting mesothelin led to PD‑1 ligand 1 (PDL‑1) dependent inhibition of the mesothelin-specific activatingKIR-CAR cytotoxicity. Primary humanTcells were stimulatedwithCD3 / 28microbeads followed by lentiviral transduction with either the SS1-KIRS2 activating KIR CAR, the SS1-zeta CAR or mock transduced (NTD). The Tcells were expanded over 9 days followed by electroporation of 5x106 Tcells with 10 ug of in vitro transcribed RNA encoding wild-type PD1 using a BTX ECM830 electroporator (PD1+) or mock transfected (PD1‑). The surface expression of theSS1-specificCARandPD‑1was determined as shown in Figure 8. K562 target cells with either no mesothelin or expressing mesothelin with or without PDL‑1 were mixed with the different Tcells conditions as indicated using varying effector Tcell to target ratios of 30:1 to 1:1 as shown. Target K562 cell lysis was assessed using a calcein AM dye method to quantify the remaining viable cells following 4 hours of incubation. Data shown is calculated % target cell lysis compared against target cells without effector cells. Figure 10 is an image demonstrating the interferon-gamma (IFN-γ) production by T cells from two different donors expressing amesothelin-specific activatingKIR-basedCAR (SS1-KIR2DS2or SS1-KIRS2) or TCR-zeta basedCAR 11 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 with or without a costimulatory domain (SS1-z, SS1‑28z or SS1-BBz). Primary human Tcells were stimulated with CD3 / 28microbeads followedby lentiviral transductionwith the indicatedactivatingKIRCARorTCR-zetabasedCAR. Mock non-transduced cells (NTD)were used as a negative control. K562 target cells with either nomesothelin (K562) or expressing mesothelin (K562-meso) were mixed with the different Tcells conditions as indicated at a 2:1 ratio of effector Tcells to target cells. Following 16 hours of incubation, IFN-γwasmeasured in the culture supernatants using a human IFN-gamma specific ELISA assay (R&D systems). Figure 11 is an image demonstrating the interleukin‑2 (IL‑2) production by Tcells expressing a mesothelin-specific activating KIR-based CAR (SS1-KIR2DS2 or SS1-KIRS2) or TCR-zeta based CAR with or without a costimulatory domain (SS1-z, SS1‑28z or SS1-BBz). Primary human Tcells were stimulated with CD3 / 28 microbeads followed by lentiviral transduction with the indicated activating KIR CAR or TCR-zeta based CAR. Mock non-transduced cells (NTD) were used as a negative control. K562 target cells with either no mesothelin (K562) or expressing mesothelin (K562-meso)weremixedwith the different Tcells conditions as indicated at a 2:1 ratio of effector Tcells to target cells. Following 16 hours of incubation, IL‑2 was measured in the culture supernatants using a human IL‑2 specific ELISA assay (R&D systems). Figure 12, comprising figures 12A‑12B, depicts construction of a mesothelin-specific KIR-based chimeric antigen receptor (KIR-CAR) engineered T cell with robust cytotoxic activity. Primary human T cells were stimulated with CD3 / 28 microbeads followed by transduction with a lentiviral vector expressing either GFP and dsRed (Control) or DAP12 and dsRed (DAP12). The cells were expanded ex vivo until the end of log phase growth. 5x106 Tcells from each transduced population were electroporated with 10 ug of in vitro transcribed RNA encoding SS1-KIRS2 using a BTXECM830electroporator. Theexpression of both dsRedandSS1-KIRS2wasassessedbyflowcytometrywith the SS1-KIRS2 detected using a biotinylated goat anti-mouse F(ab)2 specific polyclonal antibody followed by strepta- vidin-PE.Theupper panel of Figure12Ashows thegating strategy for identificationof Tcells expressingdsRed,which were thenanalyzed forSS1-KIRS2expression as shown in the lower portionof the panel. Figure12Bshows theability of the cells characterized in Figure 12A to mediate cytotoxicity against wild-type K562 cells (K562-wt) or K562 cells that express mesothelin (K562-mesothelin) as assessed using a 4-hr 51Cr-release assay. Figure 13 shows that the expression of an endogenous TCR is unaffected by SS1-KIRS2 and DAP12 expression. 5x106 primary humanTcellswere electroporatedwith 10 ug of in vitro transcribedRNAencodingSS1-KIRS2ormock transfected using a BTX ECM830 electroporator. After overnight incubation, the transfected Tcells were stained for the expression of SS1-KIRS2 using a biotinylated goat anti-mouse F(ab)2 specific polyclonal antibody followed by streptavidin-PE. The expression of Vβ13.1wasassessed using aPE-conjugatedmonoclonal antibody specific to this Vβ chain of the TCR. Figure 14 illustrates the ability of a mesothelin-specific KIR-based CAR (SS1-KIRS2) to stimulate Tcell proliferation that is antigen-dependent but independent of additional CD28 costimulation. Primary human Tcells were stimulated with CD3 / 28 microbeads followed by lentiviral transduction of SS1-KIRS2 and DAP12 or the mesothelin-specific TCR-zeta CAR (SS1-zeta). Mock non-transduced cells (NTD) were used as a negative control. K562 target cells with either no mesothelin (K562 wt) or expressing mesothelin (K562-mesothelin) were mixed with the different T cells conditions as indicated at a 2: 1 ratio of effector Tcells to target cells. Tcells stimulated with K562-mesothelin were further divided into a condition with or without a monoclonal anti-CD28 agonist antibody (clone 9.3) at 1 ug / mL. The number of viable Tcellswere enumerated by flowcytometry using bead-based counting at the indicated time points to calculate the number of population doublings following antigen stimulation. Figure 15, comprising figures 15A‑15C, demonstrates that mesothelin-specific KIR-CAR modified T cells show enhanced anti-tumor activity in vivo compared with second generation TCR‑ ζ based CARs bearing CD28 or CD137 (4‑1BB) costimulatory domains. Figure 15A shows an experiment in which NOD-SCID-γc‑ / - (NSG) mice were subcutaneously implanted with a mesothelioma-derived cell expressing mesothelin (EM-meso cells). 20 days following tumor implantation, each animal was injected intravenously with 7 million Tcells that were stimulated with anti-CD3 / anti-CD28stimulator beads followed lentiviral transductionwith a series ofCD3ζ-basedCARwith orwithout acostimulatorydomain (SS1-ζ, SS1-BBζandSS1‑28ζ) or themesothelin-specificKIR-basedCARs,SS1-KIRS2with DAP12. Mock transduced Tcells (NTD) were used as a control. Tumor volume was assessed via caliper measure- ment. 8 animals were analyzed for each Tcell condition Figure 15B shows that the in vivo activity of the KIR-CAR is independent of Tcell engraftment in blood, spleen or tumor. The frequency of human CD45+ Tcells was assessed at the end of the experiment by flow cytometry, and data are expressed as a percentage of total viable cells in the blood, spleen and tumor digest. Figure 15C shows that DAP12-modified Tcells require the mesothelin-specific KIR-based CAR for tumoreradication. Thesamemodel as that shown inFigure15Awasused. 4millionTcells expressingDAP12 and dsRed (DAP12), SS1‑28z or SS1-KIRS2 and DAP12 (SS1-KIRS2) were injected intravenously on day 20, and tumor volume was assessed over time via caliper measurement. Figure 16 comprising figures 16A‑16B, demonstrates a KIR-based CAR with CD19 specificity can trigger antigen- specific target cell cytotoxicity. Following anti-CD3 / anti-CD28 bead activation, T cells were transduced with a bicistronic lentiviral vector expressing DAP12 along with either a CD19-specific KIR-based CAR in which the 12 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 FMC63-derived scFv is fused to full length KIR2DS2 (CD19-KIR2DS2) or a KIR-based CAR generated by fusing the FMC63 scFv to the transmembrane and cytoplasmic domain of KIR2DS2 via a short linker [Gly]4‑Ser linker(CD19- KIRS2). The transduced Tcells were cultured until the end of the log phase growth, and the expression of the CD19- specific KIR-based CAR was assessed by flow cytometry using a biotinylated goat-anti-mouse F(ab)2 polyclonal antibody followed bySA-PE. 51Cr-labeledK562 target cellswith (K562‑CD19) orwithout (K562-wt)CD19expression were mixed at varying ratios with T cells to target cells (E:T ratio). Cytotoxicity was determined by measuring the fraction of 51Cr released into the supernatant at 4 hours. Control Tcells that were either mock transduced (NTD) or transducedwith a CD3ζ-basedCAR specific to CD19 (CD19-z) were also included as negative and positive controls, respectively. Figure17comprisingfigures17A‑17BshowsCD19-KIRS2 in vivoactivity.NOD-SCID-γc‑ / -(NSG)micewereengrafted intravenously by tail vein injection on day 0 of 1millionNalm‑6CBG tumor cells, a leukemic cell line expressingCD19. Tcells were stimulated with anti-CD3 / anti-CD28 stimulator beads followed by lentiviral transduction on day 1 with a series of CD19-specfic CD3ζ-based CAR with or without a costimulatory domain (CD19z, 19BBz) or the CD19- specific KIR-based CARs, CD19-KIRS2 with DAP12 (19KIRS2). Mock non-transduced Tcells (NTD) were used as a control. The Tcells were expanded until the end of log-phase growth ex vivo and injected intravenously on day 5 post leukemic cell line injection with 2 million CAR T cells per mouse. Tumor burden was assessed via bioluminescent imaging. 5 animals were analyzed for each Tcell condition. Figure 17A shows the individual bioluminescent photon flux for individual animals on day 5 (baseline prior to Tcell injection) and at day 15 following leukemic cell engraftment. Figure 17B shows the median total flux for each treatment group over time. Figure 18 demonstrates an NKp46-based NCRCARwith mesothelin specificity triggers antigen specific cytotoxicity. Following anti-CD3 / anti-CD28 bead activation, Tcells were transduced with a bi-cistronic lentiviral vector expressing either DAP12 and SS1-KIRS2 (control), or FcεRγ and a mesothelin specific NKp46-based CAR (SS1-NKp46) or FcεRγ and a mesothelin-specific NKp46 CAR in which the natural NKp46 extracellular domain was truncated (SS1- TNKp46). The expression of themesothelin-specificCARswasassessed by flowcytometry using abiotinylated goat- anti-mouse F(ab)2 polyclonal antibody followed by SA-PE as shown in figure 18A. The Tcells were mixed with 51Cr- labeled K562 target cells expressing mesothelin at varying ratios of ef fector T cells to target K562 cells (E:T ratio). Cytotoxicitywasdetermined bymeasuring the fraction of 51Cr released into the supernatant at 4 hours comparedwith spontaneous release as shown in figure 18B. Figure 19 shows a schematic representation of the receptors used in Experiments shown in Figures 21‑23. Figure20demonstrates thegenerationandcharacterizationof aK562-mesocell line that express theKIR2DL3 ligand HLA-Cw. K562 cells (K562) or K562 cells expressing mesothelin (K562-meso) were transduced with the HLA-Cw3 allele followed by fluorescence activated cell sorting to obtain K562 cells expressing HLA-Cw with (K562-meso- HLACw) or without (K562-HLACw) expression of mesothelin. HLA-Cw3 expression was assessed by flow cytometry using an APC-conjugated monoclonal antibody that recognizes HLA-A, B and C alleles (clone W6 / 32). Figure 21 demonstrates co-expression of SS1-KIRS2 and KIR2DL3 in primary human Tcells. Primary human Tcells were stimulated with CD3 / 28 microbeads followed by lentiviral transduction with SS1-KIRS2 and DAP12 (SS1- KIRS2) ormock transduced (NTD) in combinationwithwild-type KIR2DL3. TheTcells were expanded until the end of log-phase growth. The surface expression of themesothelin-specific CAR and KIR2DL3was determined by staining with mesothelin-Fc followed by PE-conjugated goat-anti-human Fc and a monoclonal antibody to the KIR2DL3 ectodomain. Figure 22 demonstrates that KIR2DL3 coexpressed with a KIR CAR can suppress antigen specific cytotoxicity in the presence of HLA-Cw on the target cells. Tcell that were generated and characterized as described in Figure 21 were mixed with 51-Cr labeled target K562 cells that were generated and characterized as described in Figure 22. Cytotoxicity was determined by measuring the fraction of 51Cr released into the supernatant at 4 hours compared target cells without effector cells. Figure 23 shows a schematic representation of the receptors used in Experiments shown in Figure 24.. Figure24demonstrates the inability to co-express twoscFv-basedchimeric receptors on thesurfaceof theTcellwhile retaining each receptors’ respective binding specificity. Jurkat T cells were transduced using a lentiviral vector encoding SS1-KIR2DL3. These cells were subsequently transduced with a second lentiviral vector encoding CD19- KIR2DS2 at varying dilutions of the vector. The expression of the SS1-specific scFv was assessed usingmesothelin- Fc followed by PE-conjugated goat-anti-human Fc. The CD19-specific scFv expression was assessed using a PE- conjugated monoclonal antibody specific to the FMC63 idiotype. Figure 25 is an image demonstrating that expression of a CD19-specific CAR also reduced the expression of mesothelin-binding sites on the surface of cells co-expressing an SS 1-zeta-mCherry fusion CAR. Primary human T cellswerestimulatedwithCD3 / 28microbeads followedby lentiviral transductionwithanSS1scFvzetaCARbearinga C-terminal mCherry fusion (SS1z-mCh) or the FMC63-derived CD19 specific 41BB-zeta CAR (19bbz) alone or combination. Mock-transduced cells were used as a control. The T cells were expanded until the end of log-phase growth, and dsRed as well as surface CAR expression was determined by flow cytometry after staining with 13 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 mesothelin-Fc followed by a goat-anti-human Fc specific polyclonal antibody conjugated to FITC. Figure 26 is an imagedemonstrating thatmutually exclusive expression of binding sites for theSS1scFv is not unique to theFMC63scFv.PrimaryhumanTcellswerestimulatedwithCD3 / 28microbeads followedby lentiviral transduction with either an SS1 scFv zeta CAR or various CD19 specific 41BB-zeta CARs (19BBz [FMC63 scFv, 214d scFv or the BL22 scFvCARswith alternate VHandVL orientations [H2L and L2H]). NTD representsmock-transduced cells used as a staining control. In addition, a separate set of Tcells were co-transduced with the SS1 scFv zeta CAR and the differentCD19specificCARsasabove. TheTcellswereexpandeduntil theendof logphasegrowth, andsurfaceCAR expression was determined by staining with biotinylated protein L (recognizes kappa light chain) followed by streptavidin APC simultaneously with mesothelin-Fc followed by a goat-anti-human Fc specific polyclonal antibody conjugated to PE. The cotransduced cells showed that the mutually exclusive expression observed with FMC63- based CAR is also observed with other scFv-CARs.. Figure 27, comprising figures 27A‑27B, depict the putative mechanism for loss of scFv binding when two scFv molecules are co-expressed on the cell surface (figure 27A) and the putative avoidance of this interaction when a camelid single VHH domain-based CAR is expressed on a T cell surface in combination with a scFv-based CAR. Figure 28 demonstrates a camelid single VHH domain-based CAR can be expressed on a T cell surface in combination with a scFv-based CAR without appreciable receptor interaction. Jurkat T cells expressing GFP under an NFAT-dependent promoter (NF-GFP) were transduced with either a mesothelin-specific activating CAR (SS1- CAR), CD19-specific activating (19-CAR) or a CAR generated using a camelid VHH domain specific to EGFR (VHH- CAR). Following transduction with the activating CAR, the cells were then transduced with an additional inhibitory CAR recognizing CD19 (19-PD1) to generate cells co-expressing both the activating and inhibitory CAR (SS1+19PD1, 19+19PD1 or VHH+19PD1). The transduced Jurkat Tcells were co-cultured for 24 hours with different cell lines that areeither 1) devoid of all target antigens (K562), 2) expressmesothelin (K-meso),CD19 (K‑19) orEGFR (A431) only, 3) express a combination of EGFR and mesothelin (A431-mesothelin) or CD19 (A431-CD19) or 4) express a combination of CD19 and mesothelin (K‑19 / meso). Additional conditions that include either no stimulator cells (no stim) or K562 with 1 ug / mL of OKT3 (OKT3) were also included as negative and positive controls for NFAT activation, respectively. GFP expression, as a marker of NFAT activation, was assessed by flow cytometry. Figure 29 shows a KIR2DS2 Sequence Annotation. Figure 30 shows a KIR2DL3 Sequence Annotation. Figure 31 shows a NKp46 Sequence Annotation. Figure 32 shows a SS1-KIRS2 Sequence Annotation. Figure 33 shows a SS1-KIR2DS2 Sequence Annotation. Figure 34 shows a SS1-tNKp46 Sequence Annotation. Figure 35 shows a SS1-KIRL3 Sequence Annotation. DETAILED DESCRIPTION

[0070] In oneaspect, the present invention provides compositions andmethods for regulating the specificity andactivity of T cells, or other cytotoxic cells, e.g., NK cells. In an embodiment a chimeric antigen receptor (a CAR), e.g., a NK cell receptor CAR (aNKR-CAR) based on anNK cell receptor (a NKR), e.g., a KIR-CAR, aNCR-CAR, a SLAMF-CAR, a FcR- CAR, or a Ly49-CAR is provided. In one embodiment, the invention provides a type of chimeric antigen receptor (CAR) wherein theCAR is termedanNKR, e.g., a "KIR-CAR,"which is aCARdesign comprisingacomponent of a receptor found onnatural killer (NK)cells. Inoneembodiment, theNKreceptor includesbut is not limited toakiller cell immunoglobulin-like receptor (KIR). KIRs can function as an activating KIR or an inhibiting KIR.

[0071] One advantage of the NKR-CARs, e.g., KIR-CARs, of the invention is that a NKR-CAR, e.g., a KIR-CARs provides amethod for regulating cytotoxic cell, e.g., Tcell, specificity to control off-target activity of the engineeredTcell. In some instances, the KIR-CARs of the invention do not require a costimulation to proliferate.

[0072] NKR-CARs can deliver a signal through an adaptor protein, e.g., an ITAM containing adaptor protein. In one embodiment, theKIR-CARsof the inventioncompriseanactivatingKIRwhichdelivers its signal throughan interactionwith the immunotyrosine-based activation motif (ITAM) containing membrane protein, DAP12 that is mediated by residues within the transmembrane domains of these proteins.

[0073] In an embodiment a NKR-CAR can deliver an inhibitory signal by means of an inhibitory motif. In one embodiment, the KIR-CARs of the invention comprise an inhibitory KIR which delivers its signal through an interaction with the immunotyrosine-based inhibitory motifs (ITIMs). KIRs bearing cytoplasmic domains that contain ITIMs abrogate the activating signal leading to inhibition of NK cytolytic and cytokine producing activity. However, the invention should not be limited to inhibitoryKIRs.Rather, any inhibitory protein havingacytoplasmicdomain that is associatedwithan inhibitory signal can be used in the construction of the CARs of the invention.

[0074] Accordingly, the invention provides a composition comprising aNKR-CAR, e.g., a KIR-CAR, vectors comprising the same, compositions comprising a NKR-CAR, e.g., a KIR-CAR, vectors packaged in viral particles, and recombinant T 14 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 cells or other cytotoxic cells comprising a NKR-CAR, e.g., a KIR-CAR. The invention also includes methods of making a genetically modified Tcell or other cytotoxic cell, e.g., a NK cell, or cultured NK cell, e.g., a NK92 cell, expressing a NKR- CAR,e.g., aKIR-CAR(KIR-CART),wherein theexpressedNKR-CAR,e.g., aKIR-CAR, comprisesanantigen recognition domain of a specific antibody with an intracellular signaling molecule from a NKR, e.g., a KIR. For example, in some embodiments, the intracellular signaling molecule includes, but is not limited to, a KIR ITAM, a KIR ITIM, and the like.

[0075] Accordingly, the invention provides compositions andmethods to regulate the specificity and activity of Tcells or other cytotoxic cells modified to express a NKR-CAR, e.g., a KIR-CAR. The present invention also provides cells comprising a plurality of types of NKR-CARs, e.g., KIR-CARs (e.g. activating NKR-CARs, e.g., KIR-CARs and inhibiting NKR-CAR, e.g., a KIR-CAR), wherein the plurality of types of NKR-CARs, e.g., KIR-CARs, participate in signaling to regulate Tcell activation. In this aspect, it is beneficial to effectively control and regulate NKR-CAR cytotoxic cells, e.g., KIR-CAR Tcells, such that they kill tumor cells while not affecting normal bystander cells. Thus, in one embodiment, the present inventionalsoprovidesmethodsof killing cancerous cellswhileminimizing thedepletionof normal non-cancerous cells, thereby improving the specificity of a NKR-CAR, e.g., a KIR-CAR, therapy.

[0076] In one embodiment, the NKR-CAR, e.g., KIR-CAR approach includes the physical separation of a plurality of types of CARs expressed on a cell, wherein binding of a plurality of types of NKR-CARs, e.g., KIR-CARs to their target antigen is required for NKR-CAR cytotoxic cell, e.g., KIR-CAR Tcell, activation. For example in the KIR-CAR approach, each KIR-CAR from the plurality of type of KIR-CARs have different intracellular signaling domain. For example, when a plurality of types of KIR-CARs is used to induceKIR-CARTcell activation, the first type of KIR-CARs can only comprise an intracellular domain froman activatingKIR and the second type of CAR can only comprise an intracellular domain froman inhibiting KIR. In this manner, conditional activation of T cells is generated by engagement of the activating KIR-CAR (actKIR-CAR) to an antigen on a malignant cell of interest. An inhibitory KIR-CAR (inhKIR-CAR) bearing an antigen binding moiety directed against an antigen that is present on a normal, but not malignant cell provides dampening of the activating effects from the actKIR-CAR when the T cell encounters normal cells.

[0077] In one embodiment, the present invention provides a Tcell or other cytotoxic cell engineered to express at least two NKR-CARs, e.g., at least two KIR-CARs, wherein the first NKR-CAR, e.g., a KIR-CA,R is an actNKR-CAR, e.g., an actKIR-CAR, and the second NKR-CAR, e.g., a KIR-CAR, is an inhNKR-CAR, e.g., an inhKIR-CAR. In one embodiment, the invention provides an inhNKR-CAR, e.g., an inhKIR-CAR, wherein binding of the inhNKR-CAR, e.g., an inhKIR-CAR, to a normal cell results in inhibition of the cytotoxic cell, e.g., inhibition of KIR-CAR T cell activity. In one embodiment, bindingof the inhNKR-CAR,e.g., an inhKIR-CAR, toanantigenassociatedwithanon-cancerouscell results in thedeathof the NKR-CAR cytotoxic cell, e.g., a KIR-CAR T cell.

[0078] In one embodiment, an inhNKR-CAR, e.g., an actKIR-CAR, of the invention can be used in combination with existing CARs in order to regulate the activity of the CARs. Exemplary CARs have been described in PCT / US11 / 64191, which is incorporated in its entirety by reference herein.

[0079] It has also been discovered that, in cells having a plurality of chimeric membrane embedded receptors comprising an antigen binding domain (CMERs), interactions between the antigen binding domain of the CMERs can be undesirable, e.g., because interaction inhibits the ability of one or more of the antigen binding domains to bind its cognate antigen or might generate novel binding sites with unknown cognate antigen. Accordingly, disclosed herein are cells having a first and a second non-naturally occurring CMER wherein the antigen binding domains minimizes such interactions. Also disclosed herein are nucleic acids encoding a first and a second non-naturally occurring such CMERs, aswell asmethodsofmakingandusing such cells andnucleic acids. In anembodiment, theantigenbindingdomainof one of said first said second non-naturally occurring CMER, comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VHdomain derived fromahumanormouse sequence or a non-antibody scaffold. DEFINITIONS

[0080] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understoodby oneof ordinary skill in the art towhich the invention pertains. Although anymethodsandmaterials similar or equivalent to those described herein can be used in the practice of and / or for the testing of the present invention, the preferred materials and methods are described herein. In describing and claiming the present invention, the following terminology will be used according to how it is defined, where a definition is provided.

[0081] It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting.

[0082] Thearticles "a"and "an"areusedherein to refer tooneor tomore thanone (i.e., toat least one)of thegrammatical object of the article. By way of example, "an element" means one element or more than one element.

[0083] "About" as used herein when referring to a measurable value such as an amount, a temporal duration, and the like, is meant to encompass variations of±20% or±10%, in some instances±5%, in some instances±1%, and in some instance ±0.1% from the specified value, as such variations are appropriate to perform the disclosed methods. 15 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0084] Adaptor molecule, as that term is used herein, refers to a polypeptide with a sequence that permits interaction with two ormoremolecules, and in embodiments, promotes activation or inactivation of a cytotoxic cell. E.g., in the case of DAP12, this comprises interactions with an activating KIR via charge interactions within the transmembrane domain and interactions with signaling molecules like ZAP70 or Syk via a phosphorylated ITAM sequence within the cytoplasmic domain.

[0085] The term "antigen" or "Ag" as used herein is defined as a molecule that provokes an immune response. This immune responsemay involveeither antibodyproduction, or theactivationof specific immunologically-competent cells, or both. The skilled artisan will understand that any macromolecule, including virtually all proteins or peptides, can serve as anantigen. Furthermore, antigens canbederived from recombinant or genomicDNA.Askilled artisanwill understand that any DNA, which comprises a nucleotide sequences or a partial nucleotide sequence encoding a protein that elicits an immune response therefore encodes an "antigen" as that term is used herein. Furthermore, one skilled in the art will understand that anantigenneednot beencodedsolely bya full lengthnucleotide sequenceof agene. It is readily apparent that the present invention includes, but is not limited to, the use of partial nucleotide sequences ofmore than onegene and that these nucleotide sequences are arranged in various combinations to encode polypeptides that elicit the desired immune response. Moreover, a skilled artisan will understand that an antigen need not be encoded by a "gene" at all. It is readily apparent that an antigen can be generated synthesized or can be derived from a biological sample. Such a biological sample can include, but is not limited to a tissue sample, a tumor sample, a cell or a biological fluid.

[0086] The term "anti-tumor effect" as usedherein, refers to a biological effectwhich canbemanifested by adecrease in tumor volume, a decrease in the number of tumor cells, a decrease in the number of metastases, an increase in life expectancy, or amelioration of various physiological symptoms associated with the cancerous condition. An "anti-tumor effect" can also be manifested by the ability of the peptides, polynucleotides, cells and antibodies of the invention in prevention of the occurrence of tumor in the first place.

[0087] The term "auto-antigen" means, in accordance with the present invention, any self-antigen which is recognized by the immune system as if it were foreign. Auto-antigens comprise, but are not limited to, cellular proteins, phospho- proteins, cellular surface proteins, cellular lipids, nucleic acids, glycoproteins, including cell surface receptors.

[0088] The term "autoimmune disease" as used herein is defined as a disorder that results from an autoimmune response. Anautoimmunedisease is the result of an inappropriate and excessive response to a self-antigen. Examples of autoimmune diseases include but are not limited to, Addision’s disease, alopecia areata, ankylosing spondylitis, autoimmune hepatitis, autoimmune parotitis, Crohn’s disease, diabetes (Type I), dystrophic epidermolysis bullosa, epididymitis, glomerulonephritis, Graves’ disease, Guillain-Barr syndrome, Hashimoto’s disease, hemolytic anemia, systemic lupus erythematosus, multiple sclerosis, myasthenia gravis, pemphigus vulgaris, psoriasis, rheumatic fever, rheumatoid arthritis, sarcoidosis, scleroderma, Sjogren’s syndrome, spondyloarthropathies, thyroiditis, vasculitis, vitiligo, myxedema, pernicious anemia, ulcerative colitis, among others.

[0089] Asusedherein, the term "autologous" ismeant to refer toanymaterial derived from thesame individual towhich it is later to be re-introduced into the individual.

[0090] "Allogeneic" refers to a graft derived from a different animal of the same species.

[0091] CAR, as that term is used herein, refers to a chimeric polypeptide that shares structural and functional properties with a cell immune-function receptor or adaptor molecule, from e.g., a Tcell or a NK cell. CARs include TCARs and NKR- CARs. Upon binding to cognate antigen, a CAR can activate or inactivate the cytotoxic cell in which it is disposed, or modulate the cell’s antitumor activity or otherwise modulate the cells immune response.

[0092] The term "cancer" as used herein is defined as disease characterized by the rapid and uncontrolled growth of aberrant cells. Cancer cells can spread locally or through the bloodstreamand lymphatic system to other parts of the body. Examples of various cancers includebut are not limited to, breast cancer, prostate cancer, ovarian cancer, cervical cancer, skin cancer, pancreatic cancer, colorectal cancer, renal cancer, liver cancer, brain cancer, lymphoma, leukemia, lung cancer, glioma, and the like.

[0093] As used herein, the term "conservative sequencemodifications" is intended to refer to amino acid modifications that do not significantly affect or alter the binding characteristics of the antibody containing the aminoacid sequence.Such conservativemodifications include amino acid substitutions, additions anddeletions.Modifications canbe introduced into an antibody of the invention by standard techniques known in the art, such as site-directed mutagenesis and PCR- mediated mutagenesis. Conservative amino acid substitutions are ones in which the amino acid residue is replaced with an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains have been defined in the art. These families include amino acids with basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine, tryptophan), nonpolar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine), beta-branched side chains (e.g., threonine, valine, isoleucine) and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine). Thus, for example, one or more amino acid residues within the CDR regions of an antibody of the invention canbe replacedwith other aminoacid residues from the sameside chain family and the altered antibody can be tested for the ability to bind FRβ using the functional assays described herein. 16 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0094] A "constitutive" promoter is a nucleotide sequence which, when operably linked with a polynucleotide which encodes or specifies a gene product, causes the gene product to be produced in a cell under most or all physiological conditions of the cell.

[0095] Cytoplasmic and intracellular, as applied to adaptor molecules and signaling domains are used interchangeably herein.

[0096] "Encoding" refers to the inherent property of specific sequences of nucleotides in a polynucleotide, such as a gene, a cDNA, or an mRNA, to serve as templates for synthesis of other polymers and macromolecules in biological processes having either a defined sequence of nucleotides (i.e., rRNA, tRNA andmRNA) or a defined sequence of amino acids and the biological properties resulting therefrom. Thus, a gene encodes a protein if transcription and translation of mRNA corresponding to that gene produces the protein in a cell or other biological system. Both the coding strand, the nucleotide sequenceofwhich is identical to themRNAsequenceand is usually provided in sequence listings, and thenon- coding strand, used as the template for transcription of a gene or cDNA, can be referred to as encoding the protein or other product of that gene or cDNA.

[0097] Unless otherwise specified, a "nucleotide sequence encoding an amino acid sequence" includes all nucleotide sequences that are degenerate versions of each other and that encode the same amino acid sequence. Nucleotide sequences that encode proteins and RNA may include introns.

[0098] "Effective amount" or "therapeutically effective amount" are used interchangeably herein, and refer to an amount of acompound, formulation,material, or composition, asdescribedhereineffective toachieveaparticular biological result. Such resultsmay include, but are not limited to, the inhibition of virus infection as determined by anymeans suitable in the art.

[0099] As used herein "endogenous" refers to anymaterial from or produced inside an organism, cell, tissue or system.

[0100] As used herein, the term "exogenous" refers to any material introduced from or produced outside an organism, cell, tissue or system.

[0101] The term "expression" as used herein is defined as the transcription and / or translation of a particular nucleotide sequence driven by its promoter.

[0102] "Expression vector" refers to a vector comprising a recombinant polynucleotide comprising expression control sequences operatively linked to a nucleotide sequence to be expressed. An expression vector comprises sufficient cis- acting elements for expression; other elements for expression can be supplied by the host cell or in an in vitro expression system. Expression vectors include all those known in the art, such as cosmids, plasmids (e.g., naked or contained in liposomes) andviruses (e.g., lentiviruses, retroviruses, adenoviruses, andadeno-associated viruses) that incorporate the recombinant polynucleotide.

[0103] FcR-CAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a FcR.

[0104] "Fully human" refers to an immunoglobulin, such as an antibody, where the wholemolecule is of human origin or consists of an amino acid sequence identical to a human form of the antibody.

[0105] "Homologous" as used herein, refers to the subunit sequence identity between two polymeric molecules, e.g., between two nucleic acid molecules, such as, two DNA molecules or two RNA molecules, or between two polypeptide molecules. When a subunit position in both of the two molecules is occupied by the same monomeric subunit; e.g., if a position in each of two DNAmolecules is occupied by adenine, then they are homologous at that position. The homology between two sequences is a direct function of the number of matching or homologous positions; e.g., if half (e.g., five positions in a polymer ten subunits in length) of the positions in two sequences are homologous, the two sequences are 50% homologous; if 90% of the positions (e.g., 9 of 10), are matched or homologous, the two sequences are 90% homologous.

[0106] "Humanized" forms of non-human (e.g., murine) antibodies are chimeric immunoglobulins, immunoglobulin chains or fragments thereof (such as Fv, Fab, Fab’, F(ab’)2 or other antigen-binding subsequences of antibodies) which containminimal sequence derived fromnon-human immunoglobulin. For themost part, humanized antibodies are human immunoglobulins (recipient antibody) in which residues from a complementary-determining region (CDR) of the recipient are replaced by residues from a CDR of a non-human species (donor antibody) such as mouse, rat or rabbit having the desired specificity, affinity, and capacity. In some instances, Fv framework region (FR) residues of the human immu- noglobulin are replaced by corresponding non-human residues. Furthermore, humanized antibodies can comprise residues which are found neither in the recipient antibody nor in the imported CDR or framework sequences. These modifications are made to further refine and optimize antibody performance. In general, the humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the CDR regions correspond to those of a non-human immunoglobulin and all or substantially all of the FR regions are those of a human immunoglobulin sequence. The humanized antibody optimally also will comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. For further details, see Jones et al., Nature, 321: 522‑525, 1986; Reichmann et al., Nature, 332: 323‑329, 1988; Presta, Curr. Op. Struct. Biol., 2: 593‑596, 1992. 17 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0107] "Intracellular signaling domain", as used herein, refers to a polypeptide sequence that is a component of a larger integral membrane protein. This polypeptide sequence, through regulated interactions with other cellular proteins, is capable of stimulating or inhibiting immune cell function such as lytic granule release, cytokine production or proliferation.

[0108] As used herein, an "instructionalmaterial" includes a publication, a recording, a diagram, or any othermediumof expression which can be used to communicate the usefulness of the compositions and methods of the invention. The instructional material of the kit of the inventionmay, for example, be affixed to a container which contains the nucleic acid, peptide, and / or composition of the invention or be shipped together with a container which contains the nucleic acid, peptide, and / or composition.Alternatively, the instructionalmaterialmaybeshippedseparately from thecontainerwith the intention that the instructional material and the compound be used cooperatively by the recipient.

[0109] "Isolated" means altered or removed from the natural state. For example, a nucleic acid or a peptide naturally present in a living animal is not "isolated," but the same nucleic acid or peptide partially or completely separated from the coexisting materials of its natural state is "isolated." An isolated nucleic acid or protein can exist in substantially purified form, or can exist in a non-native environment such as, for example, a host cell.

[0110] In the context of the present invention, the following abbreviations for the commonly occurring nucleic acid bases are used. "A" refers to adenosine, "C" refers to cytosine, "G" refers to guanosine, "T" refers to thymidine, and "U" refers to uridine.

[0111] Unless otherwise specified, a "nucleotide sequence encoding an amino acid sequence" includes all nucleotide sequences that are degenerate versions of each other and that encode the same amino acid sequence. The phrase nucleotide sequence that encodes aprotein or anRNAmayalso include introns to theextent that the nucleotide sequence encoding the protein may in some version contain an intron(s).

[0112] KIR-CAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a KIR.

[0113] A "lentivirus" as used herein refers to a genus of the Retroviridae family. Lentiviruses are unique among the retroviruses in being able to infect non-dividing cells; they can deliver a significant amount of genetic information into the DNA of the host cell, so they are one of the most efficient methods of a gene delivery vector. HIV, SIV, and FIV are all examples of lentiviruses. Vectors derived from lentiviruses offer themeans to achieve significant levels of gene transfer in vivo.

[0114] Ly49-CAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a Ly49.

[0115] NCR-CAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a NCR.

[0116] NK cell immune-function receptor (or NKR), as that term is used herein, refers to an endogenous naturally occurring transmembraneprotein expressed inNKcells, which canengagewith a ligand onanantigenpresenting cell and modulate an NK cell immune-function response, e.g., it can modulate the cytolytic activity or cytokine secretion of the NK cell. The NKR can contribute to activation (an activating NKR, or actNKR), or inhibition (an inhibitory NKR, or inhNKR). Typically, an NKR comprises an extracellular ligand-binding domain (ECD), a transmembrane domain (TM) and an intracellular cytoplasmic domain (ICD). NKRs include the Killer Immunoglobulin-like Receptor (KIR) family of receptors such as KIR2DS2, the NK cell receptor (NCR) receptor family of receptors such as NKp46 (NCR1), the signaling lymphocyte activation receptor (SLAM) family (SLAMF) of receptors such as 2B4, and the Fc-binding receptors such as the IgG-binding receptor, CD16 (FcγRIII). Examples of NK cell immune-function responsesmodulated byNKRs comprise target cell killing (often referred to as cytotoxicity or cytolysis), cytokine secretion and / or proliferation. Typically, an NKR suitable for use in the methods and compositions described herein is a human NKR, (or hNKR). In an embodiment, the Ly49 receptor family in Mus musculus, which emerged by convergent evolution to provide the same function as a KIR in murine NK and T cells, is also included.

[0117] NKR-CAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a NKR or adaptor molecule from a NK cell.

[0118] The term "operably linked" refers to functional linkage between a regulatory sequence and a heterologous nucleic acid sequence resulting in expression of the latter. For example, a first nucleic acid sequence is operably linked with a second nucleic acid sequence when the first nucleic acid sequence is placed in a functional relationship with the secondnucleic acid sequence. For instance, a promoter is operably linked to a coding sequence if the promoter affects the transcriptionorexpressionof thecodingsequence.Generally, operably linkedDNAsequencesarecontiguousand,where necessary to join two protein coding regions, in the same reading frame.

[0119] "Parenteral" administration of an immunogenic composition includes, e.g., subcutaneous (s.c.), intravenous (i.v.), intramuscular (i.m.), or intrasternal injection, or infusion techniques.

[0120] The term "polynucleotide" as used herein is defined as a chain of nucleotides. Furthermore, nucleic acids are polymersofnucleotides.Thus, nucleicacidsandpolynucleotidesasusedhereinare interchangeable.Oneskilled in theart has the general knowledge that nucleic acids are polynucleotides, which can be hydrolyzed into the monomeric "nucleotides." The monomeric nucleotides can be hydrolyzed into nucleosides. As used herein polynucleotides include, 18 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 but are not limited to, all nucleic acid sequences which are obtained by any means available in the art, including, without limitation, recombinant means, i.e., the cloning of nucleic acid sequences from a recombinant library or a cell genome, using ordinary cloning technology and PCR™, and the like, and by synthetic means.

[0121] As used herein, the terms "peptide," "polypeptide," and "protein" are used interchangeably, and refer to a compound comprisedof aminoacid residues covalently linkedbypeptide bonds.Aprotein or peptidemust contain at least two amino acids, and no limitation is placed on the maximum number of amino acids that can comprise a protein’s or peptide’s sequence. Polypeptides include any peptide or protein comprising two ormore amino acids joined to each other by peptide bonds. As used herein, the term refers to both short chains, which also commonly are referred to in the art as peptides, oligopeptides and oligomers, for example, and to longer chains, which generally are referred to in the art as proteins, of which there are many types. "Polypeptides" include, for example, biologically active fragments, substantially homologous polypeptides, oligopeptides, homodimers, heterodimers, variants of polypeptides, modified polypeptides, derivatives, analogs, fusion proteins, among others. The polypeptides include natural peptides, recombinant peptides, synthetic peptides, or a combination thereof.

[0122] The term "promoter" as used herein is defined as aDNA sequence recognized by the synthetic machinery of the cell, or introduced synthetic machinery, required to initiate the specific transcription of a polynucleotide sequence.

[0123] As used herein, the term "promoter / regulatory sequence" means a nucleic acid sequence which is required for expression of a geneproduct operably linked to thepromoter / regulatory sequence. In some instances, this sequencemay be the core promoter sequence and in other instances, this sequencemay also include an enhancer sequence and other regulatory elements which are required for expression of the gene product. The promoter / regulatory sequence may, for example, be one which expresses the gene product in a tissue specific manner.

[0124] An "inducible" promoter is a nucleotide sequence which, when operably linked with a polynucleotide which encodes or specifies a gene product, causes the gene product to be produced in a cell substantially only when an inducer which corresponds to the promoter is present in the cell.

[0125] TCAR, as that term is used herein, refers to a CAR which shares functional and structural properties with a cell immune-function receptor or adaptor molecule from a T cell.

[0126] A"tissue-specific"promoter isanucleotidesequencewhich,whenoperably linkedwithapolynucleotideencodes or specifiedbyagene, causes thegeneproduct tobeproduced inacell substantially only if thecell is acell of the tissue type

[0127] A "signal transduction pathway" refers to the biochemical relationship between a variety of signal transduction molecules that play a role in the transmission of a signal from one portion of a cell to another portion of a cell. The phrase "cell surface receptor" includes molecules and complexes of molecules capable of receiving a signal and transmitting signal across the membrane of a cell.

[0128] The term "subject" is intended to include living organisms in which an immune response can be elicited (e.g., mammals).

[0129] As used herein, a "substantially purified" cell is a cell that is essentially free of other cell types. A substantially purified cell also refers to a cell which has been separated from other cell types with which it is normally associated in its naturally occurring state. In some instances, a population of substantially purified cells refers to a homogenous population of cells. In other instances, this term refers simply to cell that have been separated from the cells with which they are naturally associated in their natural state. In some embodiments, the cells are cultured in vitro. In other embodiments, the cells are not cultured in vitro.

[0130] As used herein, a 5’ cap (also termed an RNA cap, an RNA 7-methylguanosine cap or an RNA m7G cap) is a modified guanine nucleotide that has been added to the "front" or 5’ end of a eukaryotic messenger RNA shortly after the start of transcription. The 5’ cap consists of a terminal groupwhich is linked to the first transcribed nucleotide. Its presence is critical for recognition by the ribosome and protection fromRNases. Cap addition is coupled to transcription, and occurs co-transcriptionally, such that each influences the other. Shortly after the start of transcription, the 5’ end of the mRNA being synthesized is bound by a cap-synthesizing complex associated with RNA polymerase. This enzymatic complex catalyzes the chemical reactions that are required for mRNA capping. Synthesis proceeds as a multi-step biochemical reaction. The capping moiety can be modified to modulate functionality of mRNA such as its stability or efficiency of translation.

[0131] As used herein, "in vitro transcribed RNA" refers to RNA, preferably mRNA, that has been synthesized in vitro. Generally, the in vitro transcribed RNA is generated from an in vitro transcription vector. The in vitro transcription vector comprises a template that is used to generate the in vitro transcribed RNA.

[0132] As used herein, a "poly(A)" is a series of adenosines attached by polyadenylation to the mRNA. In the preferred embodiment of a construct for transient expression, the polyA is between 50 and 5000, preferably greater than 64, more preferably greater than 100, most preferably greater than 300 or 400. poly(A) sequences can be modified chemically or enzymatically to modulate mRNA functionality such as localization, stability or efficiency of translation.

[0133] Asusedherein, "polyadenylation" refers to thecovalent linkageofapolyadenylylmoiety, or itsmodifiedvariant, to a messenger RNA molecule. In eukaryotic organisms, most messenger RNA (mRNA) molecules are polyadenylated at the 3’ end. The 3’ poly(A) tail is a long sequence of adenine nucleotides (often several hundred) added to the pre-mRNA 19 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 through theactionofanenzyme,polyadenylatepolymerase. Inhigher eukaryotes, thepoly(A) tail is addedonto transcripts that contain a specific sequence, the polyadenylation signal. The poly(A) tail and the protein bound to it aid in protecting mRNA from degradation by exonucleases. Polyadenylation is also important for transcription termination, export of the mRNA from thenucleus, and translation.Polyadenylationoccurs in thenucleus immediatelyafter transcriptionofDNA into RNA, but additionally can also occur later in the cytoplasm. After transcription has been terminated, the mRNA chain is cleaved through the action of an endonuclease complex associated with RNA polymerase. The cleavage site is usually characterized by the presence of the base sequence AAUAAA near the cleavage site. After themRNA has been cleaved, adenosine residues are added to the free 3’ end at the cleavage site.

[0134] As used herein, "transient" refers to expression of a non-integrated transgene for a period of hours, days or weeks, wherein the period of time of expression is less than the period of time for expression of the gene if integrated into the genome or contained within a stable plasmid replicon in the host cell.

[0135] As used herein, the term "TCAR" comprises an antigen domain, an intracellular signaling domain, and optionally one or more costimulatory domains.

[0136] The term "therapeutic" as used hereinmeans a treatment and / or prophylaxis. A therapeutic effect is obtained by suppression, remission, or eradication of a disease state.

[0137] The term "transfected" or "transformed" or "transduced" as used herein refers to a process by which exogenous nucleic acid is transferredor introduced into thehost cell. A "transfected" or "transformed" or "transduced" cell is onewhich has been transfected, transformed or transduced with exogenous nucleic acid. The cell includes the primary subject cell and its progeny.

[0138] Thephrase "under transcriptional control" or "operatively linked" as usedhereinmeans that the promoter is in the correct location and orientation in relation to a polynucleotide to control the initiation of transcription by RNA polymerase and expression of the polynucleotide.

[0139] A "vector" is a composition of matter which comprises an isolated nucleic acid and which can be used to deliver the isolated nucleic acid to the interior of a cell. Numerous vectors are known in the art including, but not limited to, linear polynucleotides, polynucleotides associated with ionic or amphiphilic compounds, plasmids, and viruses. Thus, the term "vector" includes an autonomously replicating plasmid or a virus. The term should also be construed to include non- plasmid and non-viral compounds which facilitate transfer of nucleic acid into cells, such as, for example, polylysine compounds, liposomes, and the like. Examples of viral vectors include, but are not limited to, adenoviral vectors, adeno- associated virus vectors, retroviral vectors, lentiviral vectors, and the like.

[0140] By the term"specifically binds,"asusedherein, ismeantanantibody, ora ligand,which recognizesandbindswith acognatebindingpartnerproteinpresent inasample, butwhichantibodyor liganddoesnot substantially recognizeorbind other molecules in the sample.

[0141] By the term "stimulation," is meant a primary response induced by binding of a stimulatory molecule with its cognate ligand thereby mediating a signal transduction event, such as, but not limited to, signal transduction via the appropriate NK receptor.

[0142] "Xenogeneic" refers to a graft derived from an animal of a different species.

[0143] Ranges: throughout this disclosure, variousaspectsof the inventioncanbepresented ina range format. It should beunderstood that thedescription in range format ismerely for convenienceandbrevity andshouldnot beconstruedasan inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from1 to 6 should be considered to have specifically disclosed subranges such as from1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 2.7, 3, 4, 5, 5.3, and 6. This applies regardless of the breadth of the range. NKR-CARs

[0144] Disclosedherein are compositions andmethods for regulating the specificity andactivity of cytotoxic cells, e.g., T cells or NK cells, e.g., with a non-naturally occurring chimeric antigen receptor (CAR). In an embodiment the CAR is an NKR-CAR. A NKR-CAR is a CAR which shares functional and structural properties with a NK cell immune-function receptor (or NKR). NKRs andNKR-CARs are described herein, e.g., in the section below. As is discussed below, a variety of NKRs can serve as the basis for an NKR-CAR. NK CELL IMMUNE-FUNCTION RECEPTORS (NKRS) AND NK CELLS

[0145] As discussed herein, NK cell immune-function receptor (or NKR) refers to an endogenous naturally occurring transmembraneprotein expressed inNKcells, which canengagewith a ligandonanantigenpresenting cell andmodulate an NK cell immune-function response, e.g., it can modulate the cytolytic activity or cytokine secretion of the NK cell.

[0146] NK cells are mononuclear cells that develop in the bone marrow from lymphoid progenitors, and morphological 20 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 features and biological properties typically include the expression of the cluster determinants (CDs) CD16, CD56, and / or CD57; the absence of the alpha / beta or gamma / delta TCR complex on the cell surface; the ability to bind to and kill target cells that fail to express "self" major histocompatibility complex (MHC) / human leukocyte antigen (HLA) proteins; and the ability to kill tumor cells or other diseased cells that express ligands for activatingNK receptors. NK cells are characterized by their ability to bind and kill several types of tumor cell lineswithout the need for prior immunization or activation. NKcells can also release soluble proteins and cytokines that exert a regulatory effect on the immune system; and can undergo multiple rounds of cell division and produce daughter cells with similar biologic properties as the parent cell. Upon activation by interferons and / or cytokines, NK cells mediate the lysis of tumor cells and of cells infected with intracellular pathogensbymechanisms that requiredirect, physical contactsbetween theNKcell and the target cell. Lysisof target cells involves the release of cytotoxic granules from theNK cell onto the surface of the bound target, and effector proteins such as perforin and granzymeB that penetrate the target plasmamembrane and induce apoptosis or programmed cell death. Normal, healthy cells are protected from lysis by NK cells. NK cell activity is regulated by a complex mechanism that involves both stimulating and inhibitory signals.

[0147] Briefly, the lytic activity of NK cells is regulated by various cell surface receptors that transduce either positive or negative intracellular signals upon interaction with ligands on the target cell. The balance between positive and negative signals transmitted via these receptors determines whether or not a target cell is lysed (killed) by a NK cell. NK cell stimulatory signals canbemediated byNaturalCytotoxicityReceptors (NCR) suchasNKp30,NKp44, andNKp46; aswell as NKG2C receptors, NKG2D receptors, certain activating killer cell immunoglobulin-like receptors (KIRs), and other activating NK receptors (Lanier, Annual Review of Immunology 2005; 23:225‑74). NK cell inhibitory signals can be mediated by receptors like Ly49, CD94 / NKG2A, as well as certain inhibitory KIRs, which recognize major histocompat- ibility complex (MHC) class I molecules (Karre et al., Nature 1986; 319:675‑8; Ohlen et al, Science 1989; 246:666‑8). These inhibitory receptors bind to polymorphic determinants of MHC class I molecules (including HLA class I) present on other cells and inhibit NK cell-mediated lysis. KIR-CARs

[0148] Disclosedherein is a chimeric antigen receptor (CAR)molecule comprisinganantigenbindingmoiety andakiller cell immunoglobulin-like receptor (KIR-CAR). In one embodiment, the KIR-CAR of the invention is expressed on the surface of a T cell. KIR-CAR BASED NKCARs

[0149] KIRs, referred to as killer cell immunoglobulin-like receptors, have been characterized in humans and non- human primates, and are polymorphic type 1 transmembrane molecules present on certain subsets of lymphocytes, including NK cells and some T cells. KIRs interact with determinants in the alpha 1 and 2 domains of the MHC class I molecules and, as described elsewhere herein, distinct KIRs are either stimulatory or inhibitory for NK cells.

[0150] NKCARs described herein include KIR-CARs, which share functional and structural properties with KIRs.

[0151] KIRs are a family of cell surface proteins found on NK cells. They regulate the killing function of these cells by interactingwithMHC class Imolecules, which are expressed on all cell types. This interaction allows them to detect virally infected cells or tumor cells. Most KIRs are inhibitory, meaning that their recognition of MHC suppresses the cytotoxic activity of the NK cell that expresses them. Only a limited number of KIRs have the ability to activate cells.

[0152] The KIR gene family have at least 15 gene loci (KIR2DL1, KIR2DL2 / L3, KIR2DL4, KIR2DL5A, KIR2DL5B, KIR2DS1,KIR2DS2, KIR2DS3,KIR2DS4,KIR2DS5,KIR3DL1 / S1, KIR3DL2, KIR3DL3and twopseudogenes, KIR2DP1 andKIR3DP1)encodedwithin a100‑200Kb regionof theLeukocyteReceptorComplex (LRC) locatedonchromosome19 (19q13.4). The LRC constitutes a large, 1 Mb, and dense cluster of rapidly evolving immune genes which contains genes encoding other cell surface molecules with distinctive Ig-like extra-cellular domains. In addition, the extended LRC contains genes encoding the transmembrane adaptor molecules DAP10 and DAP12.

[0153] KIR genes vary in length from 4 to 16 Kb (full genomic sequence) and can contain four to nine exons. KIR genes are classified as belonging to one of three groups according to their structural features: (1) Type I KIR2D genes, which encode two extra-cellular domain proteins with a D1 and D2 conformation; (2) The structurally divergent Type II KIR2D genes which encode two extra-cellular domain proteins with a D0 and D2 conformation; and finally (3) KIR3D genes encoding proteins with three extra-cellular Ig-like domains (D0, D1 and D2).

[0154] Type I KIR2D genes, which include the pseudogene KIR2DP1 as well as KIR2DL1‑3 and KIR2DS1‑5 genes, possess eight exonsaswell as a pseudoexon3 sequence. This pseudoexon is inactivated in Type I KIR2D. In somecases this is due to a nucleotide substitution located on the intron 2-exon 3 splice-site where its nucleotide sequence exhibits a high-degreeof identity toKIR3Dexon3 sequencesandpossessesa characteristic threebasepair deletion. In other cases a premature stop codon initiates differential splicing of exon 3. Within the Type I KIR2D group of genes, KIR2DL1 and KIR2DL2 share a common deletion in exon 7 distinguishing them from all other KIR in this exon, which results in a shorter 21 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 exon 7 sequence. Similarly, within Type I KIR2D, KIR2DL1‑3 differ fromKIR2DS1‑5 only in the length of their cytoplasmic tail encoding region in exon 9. The KIR2DP1 pseudogene structure differs from that of KIR2DL1‑3 in that the former has a shorter exon 4 sequence, due to a single base pair deletion.

[0155] Type II KIR2D genes include KIR2DL4 and KIR2DL5. Unlike KIR3D and Type I KIR2D, Type II KIR2D characteristically have deleted the region corresponding to exon 4 in all other KIR. Additionally, Type II KIR2D genes differ from Type I KIR2D genes in that the former possess a translated exon 3, while Type I KIR2D genes have an untranslated pseudoexon 3 sequence in its place.Within the Type II KIR2D genes, KIR2DL4 is further differentiated from KIR2DL5 (as well as from other KIR genes) by the length of its exon 1 sequence. In KIR2DL4, exon 1 was found to be six nucleotides longer and to possess an initiation codon different from those present in the other KIR genes. This initiation codon is in better agreement with the ’Kozak transcription initiation consensus sequence’ than the second potential initiation codon in KIR2DL4 that corresponds to the initiation codon present in other KIR genes.

[0156] KIR3D genes possess nine exons and include the structurally related KIR3DL1, KIR3DS1, KIR3DL2 and KIR3DL3 genes. KIR3DL2 nucleotide sequences are the longest of all KIR genes and span 16,256 bp in full genomic sequencesand1,368bp in cDNA.Within theKIR3Dgroup, the fourKIRgenesdiffer in the lengthof the regionencoding the cytoplasmic tail in exon 9. The length of the cytoplasmic tail of KIR proteins can vary from 14 amino acid residues long (in some KIR3DS1 alleles) to 108 amino acid residues long (in KIR2DL4 proteins). Additionally, KIR3DS1 differs from KIR3DL1orKIR3DL2 in that the former hasashorter exon8sequence.KIR3DL3differs fromotherKIRsequences in that it completely lacks exon 6. The most extreme KIR gene structure difference observed was that of KIR3DP1. This gene fragment completely lacks exons 6 through 9, and occasionally also exon 2. The remaining portions of the genewhich are present (exon 1, 3, 4 and 5) share a high level of sequence identity to other KIR3D sequences, in particular to KIR3DL3 sequences.

[0157] KIRproteinspossesscharacteristic Ig-likedomains on their extracellular regions,which in someKIRproteinsare involved in HLA class I ligand binding. They also possess transmembrane and cytoplasmic regions which are functionally relevant as they define the type of signal which is transduced to the NK cell. KIR proteins can have two or three Ig-like domains (hence KIR2D or KIR3D) as well as short or long cytoplasmic tails (represented as KIR2DS or KIR2DL). Two domain KIR proteins are subdivided into two groups depending on the origin of the membrane distal Ig-like domains present. Type I KIR2D proteins (KIR2DL1, KIR2DL2, KIR2DL3, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4 andKIR2DS5) possessamembrane-distal Ig-likedomainsimilar inorigin to theKIR3DD1 Ig-likedomainbut lackaD0domain.ThisD1 Ig- like domain is encoded mainly by the fourth exon of the corresponding KIR genes. The Type II KIR2D proteins, KIR2DL4 andKIR2DL5, possessamembrane-distal Ig-like domainof similar sequence to theD0domainpresent inKIR3Dproteins, however, Type II KIR2D lack a D1 domain. Long cytoplasmic tails usually contain two Immune Tyrosine-based Inhibitory Motifs (ITIM) which transduce inhibitory signals to the NK cell. Short cytoplasmic tails possess a positively charged amino acid residue in their transmembrane region which allows them to associate with a DAP12 signaling molecule capable of generating an activation signal Exceptions to this is KIR2DL4, which contains only one N-terminus ITIM. In addition, KIR2DL4 also possesses a charged residue (arginine) in its transmembrane domain, a feature which allows this receptor to elicit both inhibitory and activating signals. KIR control the response of human NK cells by delivering inhibitory or activating signals upon recognition of MHC class I ligands on the surface of potential target cells.

[0158] KIR proteins vary in length from 306 to 456 amino acid residues. Although the differences in protein length are mostly theconsequenceof thenumberof Ig-likedomainspresent, cytoplasmic region lengthdiversity isalsoan influencing factor. The leader peptide of most KIR proteins is 21 amino acid residues long. However, the presence of a different initiation codon generates a correspondingly longer leader peptide in KIR2DL4 proteins .

[0159] The D0 Ig-like domain present in Type II KIR2D proteins and KIR3D proteins is approximately 96 amino acid residues in length. The D1 domain of Type I KIR2D and of KIR3D proteins is 102 amino acid residues long, while the D2 domain of all KIR proteins is 98 amino acid residues long. The length of the stem region varies from the 24 amino acid residues present in most KIR proteins, to only seven amino acid residues in the divergent KIR3DL3 protein. The transmembrane region is 20 amino acid residues long for most KIR proteins, but one residue shorter on KIR2DL1 and KIR2DL2 proteins as a result of a three base pair deletion in exon 7. Finally, the cytoplasmic region of KIR proteins exhibits greater length variations, ranging from 23 amino acid residues in some KIR3DS1 alleles to the 96 amino acid residues present in KIR3DL2 proteins.

[0160] Aminoacidsequences forhumanKIRpolypeptides (Homosapiens)areavailable in theNCBIdatabase, seee.g., accession number NP_037421.2 (GI:134268644), NP_703144.2 (GI:46488946), NP_001229796.1 (GI:338968852), NP_001229796.1 (GI:338968852), NP_006728.2 (GI:134268642), NP_065396.1 (GI: 11968154), NP_001018091.1 (GI:66267727), NP_001077008.1 (GI:134133244), NP_036444.1 (GI:6912472), NP_055327.1 (GI:7657277), NP_056952.2 (GI:71143139), NP_036446.3 (GI:116517309), NP_001074239.1 (GI:124107610), NP_002246.5 (GI: 124107606), NP_001074241.1 (GI: 124107604), NP_036445.1 (GI:6912474).

[0161] The nomenclature for KIRs is based upon the number of extracellular domains (KIR2D and KIR3D having two and three extracellular Ig-domains, respectively) and whether the cytoplasmic tail is long (KIR2DL or KIR3DL) or short (KIR2DS or KIR3DS). The presence or absence of a given KIR is variable from one NK cell to another within the NK 22 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 populationpresent ina single individual.Amonghumans, there isalsoa relativelyhigh level of polymorphismofKIRgenes, with certain KIR genes being present in some, but not all individuals. The expression of KIR alleles on NK cells is stochastically regulated, meaning that, in a given individual, a given lymphocyte may express one, two, or more different KIRs, depending on the genoptype of the individual. The NK cells of a single individual typically express different combinations of KIRs, providing a repertoire of NK cells with different specificities for MHC class I molecules.

[0162] Certain KIR gene products cause stimulation of lymphocyte activity when bound to an appropriate ligand. The activating KIRs all have a short cytoplasmic tail with a charged trans-membrane residue that associates with an adapter molecule having an Immunoreceptor Tyrosine-basedActivationMotifs (ITAMs)which transduce stimulatory signals to the NK cell. By contrast, inhibitory KIRs have a long cytoplasmic tail containing Immunoreceptor Tyrosine-based Inhibitory Motif (ITIM), which transduce inhibitory signals to the NK cell upon engagement of their MHC class I ligands. The known inhibitory KIRs includemembers of the KIR2DLandKIR3DL subfamilies. Inhibitory KIRs having two Ig domains (KIR2DL) recognize HLA-C allotypes: KIR2DL2 (formerly designated p58.2) and the closely related, allelic gene product KIR2DL3 both recognize "group 1" HLA-C allotypes (including HLA-Cw1, ‑3, ‑7, and - 8), whereas KIR2DL1 (p58.1) recognizes "group 2"HLA-C allotypes (such asHLA-Cw2, ‑4, ‑5, and ‑6). The recognition by KIR2DL1 is dictated by the presence of a Lys residue at position 80 of HLA-C alleles. KIR2DL2 and KIR2DL3 recognition is dictated by the presence of an Asn residue at position 80 in HLA-C. Importantly, the great majority of HLA-C alleles have either an Asn or a Lys residue at position 80. Therefore, KIR2DL1, ‑2, and ‑3 collectively recognize essentially all HLA-C allotypes found in humans. One KIRwith three Ig domains, KIR3DL1 (p70), recognizes an epitope shared byHLA-Bw4 alleles. Finally, KIR3DL2 (p140), a homodimer of molecules with three Ig domains, recognizes HLA-A3 and -A11.

[0163] However, the invention should not be limited to inhibitory KIRs comprising a cytoplasmic tail containing ITIM. Rather, any inhibitory protein having a cytoplasmic domain that is associated with an inhibitory signal can be used in the constructionof theCARsof the invention.Non-limitingexamplesof an inhibitory protein includebut arenot limitedCTLA‑4, PD‑1, and the like. These proteins are known to inhibit T cell activation.

[0164] Accordingly, the invention provides a KIR-CAR comprising an extracellular domain that comprises a target- specific binding element otherwise referred to as an antigen binding domain fused to a KIR or fragment thereof. In one embodiment, the KIR is an activating KIR that comprises a short cytoplasmic tail that associateswith an adaptermolecule having an Immunoreceptor Tyrosine-based Activation Motifs (ITAMs) which transduce stimulatory signals to the NK cell (referred elsewhere herein as actKIR-CAR). In one embodiment, the KIR is an inhibitory KIR that comprises a long cytoplasmic tail containing Immunoreceptor Tyrosine-based Inhibitory Motif (ITIM), which transduce inhibitory signals (referred elsewhere herein as inhKIR-CAR).In some instances, it is desirable to remove the hinge region for the activating KIRs when construction an actKIR-CAR. This is because the invention is partly based on the discovery that an activating KIR CAR in which the KIR2DS2 hinge was removed to generate the KIR2S CAR, this KIRS2 CAR exhibited enhanced cytolytic activity compared to an actKIR-CAR comprising a full length wildtype KIR2DS2.

[0165] Thenucleic acid sequences coding for thedesiredmoleculesof the invention canbeobtainedusing recombinant methods known in the art, such as, for example by screening libraries from cells expressing the gene, by deriving the gene fromavector known to include the same, or by isolating directly fromcells and tissues containing the same, using standard techniques. Alternatively, the gene of interest can be produced synthetically, rather than cloned.

[0166] The present invention includes retroviral and lentiviral vector constructs expressing a KIR-CAR that can be directly transduced into a cell. The present invention also includes anRNAconstruct that can be directly transfected into a cell. A method for generating mRNA for use in transfection involves in vitro transcription (IVT) of a template with specially designed primers, followed by polyA addition, to produce a construct containing 3’ and 5’ untranslated sequence ("UTR"), a 5’ cap and / or Internal RibosomeEntry Site (IRES), the gene to be expressed, and a polyA tail, typically 50‑2000 bases in length. RNA so produced can efficiently transfect different kinds of cells. In one embodiment, the template includes sequences for the KIR-CAR.

[0167] In an embodiment, a KIR-CAR comprises an antigen binding domain and a KIR transmembrane domain. In an embodiment, aKIR-CARcomprisesanantigenbindingdomainandaKIR intracellular domain, e.g., an inhKIR intracellular domain.

[0168] KIR D domain, as that term is used herein, refers to a D0, D1, or D2 domain of a KIR.

[0169] KIR D domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a D domain of a KIR.

[0170] KIRD0domain, as that term is used herein, refers to aD0domain of a KIR. In an embodiment theKIRD0domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring KIR D0domain or aKIRD0domain described herein. In embodiments theKIRD0domain of aKIR-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurringKIRD0domain or aKIRD0domain described herein. In embodiments the KIR D0 domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIRD0 domain or a KIRD0 domain described herein. In embodiments the KIRD0 domain of a KIR-CAR does not differ from, or shares 100%homologywith, a reference sequence, e.g., a naturally occurring KIR D0 domain or a KIR D0 domain described herein. 23 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0171] KIR D1 domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a D1 domain of a KIR. In an embodiment the KIR D1 domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99%homologywitha referencesequence, e.g., anaturally occurringKIRD1domainoraKIRD1domaindescribedherein. In embodiments theKIRD1domainof aKIR-CARdiffers at nomore than15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring KIR D1 domain or a KIR D1 domain described herein. In embodiments the KIR D1 domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIR D0 domain or a KIR D1 domain described herein. In embodiments the KIR D1 domain of a KIR-CAR does not differ from, or shares 100%homologywith, a reference sequence, e.g., a naturally occurringKIRD1domain or aKIRD1domain described herein.

[0172] KIR D2 domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a D2 domain of a KIR. In an embodiment the KIR D2 domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99%homologywitha referencesequence, e.g., anaturally occurringKIRD2domainoraKIRD2domaindescribedherein. In embodiments theKIRD2domainof aKIR-CARdiffers at nomore than15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring KIR D2 domain or a KIR D2 domain described herein. In embodiments the KIR D2 domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIR D2 domain or a KIR D2 domain described herein. In embodiments the KIR D2 domain of a KIR-CAR does not differ from, or shares 100%homologywith, a reference sequence, e.g., a naturally occurringKIRD2domain or aKIRD0domain described herein.

[0173] KIR hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional propertiesof ahingeor stemdomainof aKIR. Inanembodiment theKIRhingeor stemdomainofaKIR-CARhas at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring KIR hinge or stem domain or a KIR hinge or stem domain described herein. In embodiments the KIR hinge or stem domain of a KIR-CAR differs at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurringKIRhingeor stemdomainoraKIRhingeor stemdomaindescribedherein. Inembodiments theKIRhingeor stemdomainofaKIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIR hinge or stem domainoraKIRhingeor stemdomaindescribedherein. Inembodiments theKIRhingeor stemdomainof aKIR-CARdoes not differ from, or shares 100%homologywith, a referencesequence, e.g., a naturally occurringKIRhingeor stemdomain or a KIR hinge or stem domain described herein.

[0174] KIR transmembrane domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a transmembrane domain of a KIR. In an embodiment the KIR transmembrane domain of a KIR- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring KIR transmembrane domain or a KIR transmembrane domain described herein. In embodiments the KIR transmembrane domain of a KIR-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring KIR transmembrane domain or a KIR transmembrane domain described herein. In embodiments the KIR transmembrane domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIR transmembrane domain or a KIR transmembrane domain described herein. In embodiments the KIR transmembrane domain of a KIR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring KIR transmembrane domain or a KIR transmembrane domain described herein.

[0175] KIR intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of an intracellular domain of a KIR. KIR intracellular domains comprise inhibitory KIR intracellular domains (referred to herein as inhKIR intracellular domains) andactivatingKIR intracellular domains (referred to herein as actKIR intracellular domains). In an embodiment the inhKIR intracellular domain comprises an ITIM sequence. In an embodiment the KIR intracellular domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring KIR intracellular domain or a KIR intracellular domain described herein. In embodiments the KIR intracellular domain of a KIR-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a referencesequence,e.g., anaturallyoccurringKIR intracellular domainoraKIR intracellular domaindescribedherein. In embodiments the KIR intracellular domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring KIR intracellular domain or a KIR intracellular domain described herein. In embodiments the KIR intracellular domain of a KIR-CAR does not differ from, or shares 100%homologywith, a reference sequence, e.g., a naturally occurring KIR intracellular domain or a KIR intracellular domain described herein. NCRs

[0176] NKCARs described herein include NCR-CARs, which share functional and structural properties with NCRs.

[0177] Natural killer (NK) cells are cytotoxic lymphoid cells specialized in destroying tumors and virus-infected cells. Unlike cytotoxic T lymphocytes, NK cells do not express antigen-specific receptors. The recognition of transformed cells occurs via the association of a multitude of cell-surface receptors with surface markers on the target cell. The NK cell surface receptors can be distinguished according to whether they activate or inhibit NK cell-mediated cytotoxicity. 24 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 Numerous interactions between different receptors appear to lead to the formation of synapses between NK and target cells. The integration of activating and inhibiting signals at the synapse dictates whether or not the NK cells exert their cytolytic function on the target cell. Among the activating receptors, the family of Ig-like molecules is termed natural cytotoxicity receptors (NCRs). These natural cytotoxicity receptors include NKp30, NKp44 and NKp46 molecules. The NCRs are key activating receptors for NK cells in tumor cell recognition. All three NCRs are involved in the clearance of both tumor and virus-infected cells. In the latter, the antiviral activity is initiated by the interaction of NKp44 with hemagglutinin of influenza virus or Sendai virus. NKp46 targets virus-infected cells by binding to influenza virus hemagglutinin or Sendai virus hemagglutinin-neuraminidase. In contrast, it has been shown that NK cell-mediated cytotoxicity is inhibited by binding of NKp30 to the human cytomegaloviral protein pp65 (see, e.g., Arnon, et. al., Nat. Immunol. (2005) 6:515‑523).

[0178] Amino acid sequences for a human NCR polypeptides (Homo sapiens) are available in the NCBI database, see e.g., accession number NP_004819.2 (GI:153945782), 014931.1 (GI:47605770), 095944.2 (GI:251757303), 076036.1 (GI:47605775), NP_001138939.1 (GI:224586865), and / or NP_001138938.1 (GI:224586860).

[0179] In an embodiment, aNCR-CARcomprises an antigen binding domain and aNCR transmembrane domain. In an embodiment, a KIR-CAR comprises an antigen binding domain and a NCR intracellular domain.

[0180] NCR extracellular domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a extracellular domain of a NCR. In an embodiment the NCR extracellular domain of a NCR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring NCR extracellular domain or a NCR extracellular domain described herein. In embodiments the NCR extracellular domain of a NCR-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring NCR extracellular domain or a NCR extracellular domain described herein. In embodiments the NCR extracellular domain of a NCR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring NCR extracellular domain or a NCR extracellular domain described herein. In embodiments the NCR extracellular domain of a NCR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring NCR extracellular domain or a NCR extracellular domain described herein.

[0181] NCR hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a hingeor stemdomainof aNCR. In anembodiment theNCRhingeor stemdomain of aNCR-CAR has at least 70, 80, 85, 90, 95, or 99%homologywith a reference sequence, e.g., a naturally occurring NCRhinge or stem domain or a NCRhinge or stem domain described herein. In embodiments the NCR hinge or stem domain of a NCR-CAR differsatnomore than15,10, 5, 2, or1%of its residues froma referencesequence,e.g., anaturally occurringNCRhingeor stemdomain or aNCRhinge or stemdomain described herein. In embodiments theNCRhinge or stemdomain of aNCR- CARdiffers at nomore than5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurringNCRhingeor stem domain or a NCRhinge or stem domain described herein. In embodiments the NCR hinge or stem domain of a NCR-CAR does not differ from, or shares 100%homology with, a reference sequence, e.g., a naturally occurring NCR hinge or stem domain or a NCR hinge or stem domain described herein.

[0182] NCR transmembrane domain, as that term is used herein, refers to a polypeptide domain having structural and functional propertiesof a transmembranedomainofaNCR. Inanembodiment theNCRtransmembranedomainofaNCR- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring NCR transmembrane domain or a NCR transmembrane domain described herein. In embodiments the NCR transmembrane domainof aNCR-CARdiffers at nomore than15, 10, 5, 2, or 1%of its residues froma referencesequence, e.g., a naturally occurring NCR transmembrane domain or a NCR transmembrane domain described herein. In embodiments the NCR transmembrane domain of a NCR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurringNCR transmembranedomainor aNCR transmembranedomaindescribedherein. Inembodiments the NCR transmembrane domain of a NCR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring NCR transmembrane domain or a NCR transmembrane domain described herein.

[0183] NCR intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of an intracellular domain of a NCR. In an embodiment the NCR intracellular domain of a NCR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring NCR intracellular domain or a NCR intracellular domain described herein. In embodiments the NCR intracellular domain of a NCR-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring NCR intracellular domain or a NCR intracellular domain described herein. In embodiments the NCR intracellular domain of a NCR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring NCR intracellular domain or a NCR intracellular domain described herein. In embodiments the NCR intracellular domain of a NCR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring NCR intracellular domain or a NCR intracellular domain described herein. 25 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 SLAM Receptors

[0184] NKCARs described herein include SLAMF ‑CARs, which share functional and structural properties with SLAMFs.

[0185] The signaling lymphocyte activation molecule (SLAM) family of immune cell receptors is closely related to the CD2 family of the immunoglobulin (Ig) superfamily of molecules. The SLAM family (SLAMF) currently includes nine membersnamedSLAM,CD48,CD229, 2B4,CD84,NTB-A,CRACC,BLAME,andCD2F‑10. Ingeneral,SLAMmolecules possess two to four extracellular Ig domains, a transmembrane segment, and an intracellular tyrosine-rich region. The molecules are differentially expressed on a variety of immune cell types. Several are self ligands and SLAM has been identified as the human measles virus receptor. Several small SH2- containing adaptor proteins are known to associate with the intracellular domains of SLAM family members and modulate receptor signaling including SH2D1A (also known as SLAM-associated protein [SAP]) and SH2D1B (also known as EAT2). For example, in Tand NK cells, activated SLAM family receptors become tyrosine phosphorylated and recruit the adaptor SAP and subsequently the Src kinase Fyn. The ensuing signal transduction cascade influences the outcome of T cell-antigen presenting cell and NK cell-target cell interactions.

[0186] Amino acid sequences for human SLAM receptor polypeptides (Homo sapiens) are available in the NCBI database, see e.g., accession number NP_057466.1 (GI: 7706529), NP_067004.3 (GI: 19923572), NP_003028.1 (GI:4506969), NP_001171808.1 (GI: 296434285), NP_001171643.1 (GI:296040491), NP_001769.2 (GI:21361571), NP_254273.2 (GI: 226342990), NP_064510.1 (GI: 9910342) and / or NP_002339.2 (GI: 55925578) In an embodiment, a SLAMF-CAR comprises an antigen binding domain and a SLAMF transmembrane domain. In an embodiment, a SLAMF-CAR comprises an antigen binding domain and a NCR intracellular domain.

[0187] SLAMF extracellular domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a extracellular domain of a SLAMF. In an embodiment the SLAMF extracellular domain of a SLAMF-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring SLAMFextracellular domain or aSLAMFextracellular domain describedherein. In embodiments theSLAMFextracellular domain of a SLAMF-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring SLAMF extracellular domain or a SLAMF extracellular domain described herein. In embodiments the SLAMF extracellular domain of a SLAMF-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring SLAMF extracellular domain or a SLAMF extracellular domain described herein. In embodi- ments theSLAMFextracellular domain of aSLAMF-CARdoes not differ from, or shares 100%homologywith, a reference sequence, e.g., a naturally occurring SLAMF extracellular domain or a SLAMF extracellular domain described herein.

[0188] SLAMF hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a hinge or stem domain of a SLAMF. In an embodiment the SLAMF hinge or stem domain of a SLAMF-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring SLAMF hinge or stem domain or a SLAMF hinge or stem domain described herein. In embodiments the SLAMF hinge or stemdomain of a SLAMF-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurringSLAMFhingeor stemdomainor aSLAMFhingeor stemdomaindescribedherein. In embodiments the SLAMF hinge or stem domain of a SLAMF-CAR differs at nomore than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring SLAMF hinge or stem domain or a SLAMF hinge or stem domain described herein. In embodiments the SLAMF hinge or stem domain of a SLAMF-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring SLAMF hinge or stem domain or a SLAMF hinge or stem domain described herein.

[0189] SLAMF transmembranedomain, as that term isusedherein, refers toapolypeptidedomainhavingstructural and functional properties of a transmembrane domain of aSLAMF. In an embodiment theSLAMF transmembrane domain of a SLAMF-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring SLAMF transmembrane domain or a SLAMF transmembrane domain described herein. In embodiments the SLAMF transmembrane domain of a SLAMF-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring SLAMF transmembrane domain or a SLAMF transmembrane domain described herein. In embodiments theSLAMF transmembranedomainof aSLAMF-CARdiffers at nomore than5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring SLAMF transmembrane domain or a SLAMF transmembrane domain described herein. In embodiments the SLAMF transmembrane domain of a SLAMF-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring SLAMF transmembrane domain or a SLAMF transmembrane domain described herein.

[0190] SLAMF intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of an intracellular domain of a SLAMF. In an embodiment the SLAMF intracellular domain of a SLAMF-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring SLAMF intracellular domain or a SLAMF intracellular domain described herein. In embodiments the SLAMF intracellular domain of a SLAMF-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a 26 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 naturally occurring SLAMF intracellular domain or a SLAMF intracellular domain described herein. In embodiments the SLAMF intracellular domain of a SLAMF-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurringSLAMF intracellular domain or aSLAMF intracellular domain described herein. In embodiments the SLAMF intracellular domain of a SLAMF-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring SLAMF intracellular domain or a SLAMF intracellular domain described herein. Fc-binding Receptors

[0191] NKCARs described herein include CARs based on the Fc receptors, FcR-CARs, e.g., CD16 ‑CARs, and CD64- CARs,which share functional and structural properties with CD16 and CD64.

[0192] Upon activation, NK cells produce cytokines and chemokines abundantly and at the same time exhibit potent cytolytic activity. ActivationofNKcells canoccur through thedirect bindingofNKcell receptors to ligands on the target cell, as seen with direct tumor cell killing, or through the crosslinking of the Fc receptor (CD 16; FcyRIII) by binding to the Fc portion of antibodies bound to an antigen-bearing cell. This CD16 engagement (CD16 crosslinking) initiates NK cell responses via intracellular signals that are generated through one, or both, of the CD16-associated adaptor chains, FcRγ or CD3 ζ. Triggering of CD16 leads to phosphorylation of the γ or ζ chain, which in turn recruits tyrosine kinases, syk and ZAP‑70, initiating a cascade of signal transduction leading to rapid and potent effector functions. The most well-known effector function is the releaseof cytoplasmic granules carrying toxic proteins to kill nearby target cells through theprocess of antibody-dependent cellular cytotoxicity. CD16 crosslinking also results in the production of cytokines and chemokines that, in turn, activate and orchestrate a series of immune responses.

[0193] However, unlike Tand B lymphocytes, NK cells are thought to have only a limited capacity for target recognition using germline-encoded activation receptors (Bottino et al., Curr Top Microbiol Immunol. 298:175‑182 (2006); Stewart et al., Curr Top Microbiol Immunol. 298:1‑21 (2006)). NK cells express the activating Fc receptor CD 16, which recognizes IgG-coated target cells, thereby broadening target recognition (Ravetch & Bolland, Annu Rev Immunol. 19:275‑290 (2001); Lanier Nat. Immunol. 9(5):495‑502 (2008); Bryceson & Long, Curr Opin Immunol. 20(3):344‑352 (2008)).

[0194] The expression and signal transduction activity of several NK cell activation receptors requires physically associated adaptors, which transduce signals through immunoreceptor tyrosine-based activationmotifs (ITAMs). Among these adaptors, FcRγ and CD3 ζ chains can associate with CD16 and natural cytotoxicity receptors (NCRs) as either disulfide-linked homo-dimers or hetero-dimers, and these chains have been thought to be expressed by all mature NK cells.

[0195] Amino acid sequence for CD16 (Homo sapiens) is available in the NCBI database, see e.g., accession number NP_000560.5 (GI: 50726979), NP_001231682.1 (GI: 348041254)

[0196] In an embodiment, a FcR-CAR comprises an antigen binding domain and a FcR transmembrane domain. In an embodiment, a FcR-CAR comprises an antigen binding domain and a FcR intracellular domain.

[0197] CD16 extracellular domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a extracellular domain of a CD16. In an embodiment the CD16 extracellular domain of a CD16- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD16 extracellular domainor aCD16extracellular domaindescribedherein. Inembodiments theCD16extracellular domainof a CD16-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD16 extracellular domain or a CD16 extracellular domain described herein. In embodiments the CD16 extracellular domain of aCD16-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD16 extracellular domain or a CD16 extracellular domain described herein. In embodiments the CD16 extracellular domain of a CD16-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD16 extracellular domain or a CD16 extracellular domain described herein.

[0198] CD16 hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a hinge or stem domain of a CD16. In an embodiment the CD16 hinge or stem domain of a CD16- CARhasat least 70, 80, 85, 90, 95, or 99%homologywith a reference sequence, e.g., a naturally occurringCD16hinge or stem domain or a CD16 hinge or stem domain described herein. In embodiments the CD16 hinge or stem domain of a CD16-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD16 hinge or stem domain or a CD16 hinge or stem domain described herein. In embodiments the CD16 hinge or stem domain of aCD16-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD16 hinge or stem domain or a CD16 hinge or stem domain described herein. In embodiments the CD16 hinge or stem domain of a CD16-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD16 hinge or stem domain or a CD16 hinge or stem domain described herein.

[0199] CD16 transmembrane domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a transmembrane domain of a CD16. In an embodiment the CD16 transmembrane domain of a CD16-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD16 transmembrane domain or aCD16 transmembrane domain described herein. In embodiments theCD16 transmembrane 27 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 domain of a CD16-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring CD16 transmembrane domain or a CD16 transmembrane domain described herein. In embodiments theCD16 transmembranedomainof aCD16-CARdiffersat nomore than5, 4, 3, 2or 1 residue froma referencesequence, e.g., a naturally occurring CD16 transmembrane domain or a CD16 transmembrane domain described herein. In embodiments the CD16 transmembrane domain of a CD16-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD16 transmembrane domain or a CD16 transmembrane domain described herein.

[0200] CD16 intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of an intracellular domain of a CD16. In an embodiment the CD16 intracellular domain of a CD16- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD16 intracellular domain or a CD16 intracellular domain described herein. In embodiments the CD16 intracellular domain of a CD16-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD16 intracellular domain or a CD16 intracellular domain described herein. In embodiments the CD16 intracellular domain of aCD16-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD16 intracellular domain or a CD16 intracellular domain described herein. In embodiments the CD16 intracellular domain of a CD16-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD16 intracellular domain or a CD16 intracellular domain described herein.

[0201] CD64 extracellular domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a extracellular domain of a CD64. In an embodiment the CD64 extracellular domain of a CD64- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD64 extracellular domainor aCD64extracellular domaindescribedherein. Inembodiments theCD64extracellular domainof a CD64-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD64 extracellular domain or a CD64 extracellular domain described herein. In embodiments the CD64 extracellular domain of aCD64-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD64 extracellular domain or a CD64 extracellular domain described herein. In embodiments the CD64 extracellular domain of a CD64-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD64 extracellular domain or a CD64 extracellular domain described herein.

[0202] CD64 hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a hinge or stem domain of a CD64. In an embodiment the CD64 hinge or stem domain of a CD64- CARhasat least 70, 80, 85, 90, 95, or 99%homologywith a reference sequence, e.g., a naturally occurringCD64hinge or stem domain or a CD64 hinge or stem domain described herein. In embodiments the CD64 hinge or stem domain of a CD64-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD64 hinge or stem domain or a CD64 hinge or stem domain described herein. In embodiments the CD64 hinge or stem domain of aCD64-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD64 hinge or stem domain or a CD64 hinge or stem domain described herein. In embodiments the CD64 hinge or stem domain of a CD64-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD64 hinge or stem domain or a CD64 hinge or stem domain described herein.

[0203] CD64 transmembrane domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a transmembrane domain of a CD64. In an embodiment the CD64 transmembrane domain of a CD64-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD64 transmembrane domain or aCD64 transmembrane domain described herein. In embodiments theCD64 transmembrane domain of a CD64-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring CD64 transmembrane domain or a CD64 transmembrane domain described herein. In embodiments theCD64 transmembranedomainof aCD64-CARdiffersat nomore than5, 4, 3, 2or 1 residue froma referencesequence, e.g., a naturally occurring CD64 transmembrane domain or a CD64 transmembrane domain described herein. In embodiments the CD64 transmembrane domain of a CD64-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD64 transmembrane domain or a CD64 transmembrane domain described herein. CD64 intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional propertiesof an intracellular domain of aCD64. In anembodiment theCD64 intracellular domainof aCD64- CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CD64 intracellular domain or a CD64 intracellular domain described herein. In embodiments the CD64 intracellular domain of a CD64-CARdiffers at nomore than 15, 10, 5, 2, or 1%of its residues froma reference sequence, e.g., a naturally occurring CD64 intracellular domain or a CD64 intracellular domain described herein. In embodiments the CD64 intracellular domain of aCD64-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue froma reference sequence, e.g., a naturally occurring CD64 intracellular domain or a CD64 intracellular domain described herein. In embodiments the CD64 intracellular domain of a CD64-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CD64 intracellular domain or a CD64 intracellular domain described herein. 28 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 Ly49 and related Killer Cell Lectin-like Receptors

[0204] NKCARs described herein include Ly49-CARs, which share functional and structural properties with Ly49.

[0205] The Ly49 receptors derive from at least 23 identified genes (Ly49A-W) in mice. These receptors share many of the same roles in mouse NK cells and T cells as that played by KIRs in humans despite their different structure (type II integral membrane proteins of the C-type lectin superfamily), and they also contain a considerable degree of genetic variation like human KIRs. The remarkable functional similarity between Ly49 and KIR receptors suggest that these groups of receptors have evolved independently yet convergently to perform the same physiologic functionals in NK cells and T cells.

[0206] Like KIRs in humans, different Ly49 receptors recognize different MHC class I alleles and are differentially expressed on subsets of NK cells. The original prototypic Ly49 receptors, Ly49A and Ly49C possess a cytoplasmic domain bearing two immunotyrosine-based inhibitory motifs (ITIM) similar to inhibitory KIRs such as KIR2DL3. These domains havebeen identified to recruit the phosphatase, SHP‑1, and like the inhibitoryKIRs, serve to limit the activation of NK cells and Tcells. In addition to the inhibitory Ly49molecules, several family members such as Ly49D and Ly49H have lost the ITIM-containing domains, and have instead acquired the capacity to interact with the signaling adaptor molecule, DAP12 similar to the activating KIRs such as KIR2DS2 in humans.

[0207] Amino acid sequence for Ly49 familymembers are available in theNCBI database, see e.g., accession numbers AAF82184.1 (GI: 9230810), AAF99547.1 (GI: 9801837), NP_034778.2 (GI: 133922593), NP_034779.1 (GI: 6754462), NP_001095090.1 (GI: 197333718), NP_034776.1 (GI: 21327665), AAK11559.1 (GI: 13021834) and / or NP_038822.3 (GI: 9256549).

[0208] In an embodiment, a Ly49-CARcomprises an antigen binding domain and a Ly49 transmembrane domain. In an embodiment, a Ly49-CAR comprises an antigen binding domain and a NCR intracellular domain.

[0209] LY49 extracellular domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a extracellular domain of a LY49. In an embodiment the LY49 extracellular domain of a LY49-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring LY49 extracellular domain or a LY49 extracellular domain described herein. In embodiments the LY49 extracellular domain of a LY49-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring LY49 extracellular domain or a LY49 extracellular domain described herein. In embodiments the LY49 extracellular domain of a LY49-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring LY49 extracellular domain or a LY49 extracellular domain described herein. In embodiments the LY49 extracellular domain of a LY49-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring LY49 extracellular domain or a LY49 extracellular domain described herein.

[0210] LY49 hinge or stem domain, as that term is used herein, refers to a polypeptide domain having structural and functional propertiesof ahingeor stemdomainofaLY49. Inanembodiment theLY49hingeor stemdomainofaLY49-CAR has at least 70, 80, 85, 90, 95, or 99%homologywith a reference sequence, e.g., a naturally occurring LY49 hinge or stem domain or a LY49hinge or stemdomain described herein. In embodiments the LY49hinge or stemdomain of a LY49-CAR differs at nomore than 15, 10, 5, 2, or 1%of its residues from a reference sequence, e.g., a naturally occurring LY49 hinge or stem domain or a LY49 hinge or stem domain described herein. In embodiments the LY49 hinge or stem domain of a LY49-CAR differs at nomore than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring LY49 hinge or stem domain or a LY49 hinge or stem domain described herein. In embodiments the LY49 hinge or stem domain of a LY49-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring LY49 hinge or stem domain or a LY49 hinge or stem domain described herein.

[0211] LY49 transmembrane domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a transmembrane domain of a LY49. In an embodiment the LY49 transmembrane domain of a LY49-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring LY49 transmembrane domain or a LY49 transmembrane domain described herein. In embodiments the LY49 transmembrane domainof a LY49-CARdiffersat nomore than15, 10, 5, 2, or 1%of its residues froma referencesequence, e.g., anaturally occurring LY49 transmembrane domain or a LY49 transmembrane domain described herein. In embodiments the LY49 transmembrane domain of a LY49-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring LY49 transmembrane domain or a LY49 transmembrane domain described herein. In embodiments the LY49 transmembrane domain of a LY49-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring LY49 transmembrane domain or a LY49 transmembrane domain described herein.

[0212] LY49 intracelluar domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of an intracellular domain of a LY49. In an embodiment the LY49 intracellular domain of a LY49-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring LY49 intracellular domain or a LY49 intracellular domain described herein. In embodiments the LY49 intracellular domain of a LY49-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring LY49 intracellular domain or a LY49 intracellular domain described herein. In embodiments the LY49 intracellular domain of a 29 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 LY49-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring LY49 intracellular domain or a LY49 intracellular domain described herein. In embodiments the LY49 intracellular domain of a LY49-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring LY49 intracellular domain or a LY49 intracellular domain described herein. Intracelluar Signaling domains or adaptor molecules, e.g., DAP12

[0213] Some NKR-CARs interact with other molecules, e.g., molecules comprising an intracellular signaling domain, e.g., an ITAM. In an embodiment a intracellular signaling domain is DAP 12.

[0214] DAP12 is sonamedbecauseof its structural features, andpresumed function.Certain cell surface receptors lack intrinsic functionality, which hypothetically may interact with another protein partner, suggested to be a 12 kD protein. The mechanism of the signaling may involve an ITAM signal.

[0215] The DAP12 was identified from sequence databases based upon a hypothesized relationship to CD3 (see Olcese, et al. (1997) J. Immunol. 158:5083‑5086), the presence of an ITAM sequence (see Thomas (1995) J. Exp. Med. 181:1953‑1956), certain size predictions (see Olcese; and Takase, et al. (1997) J. Immunol. 159:741‑747, and other features. In particular, the transmembrane domain was hypothesized to contain a charged residue, which would allow a salt bridge with the corresponding transmembrane segments of its presumed receptor partners, KIR CD94 protein, and possibly other similar proteins. See Daeron, et al. (1995) Immunity 3:635‑646.

[0216] In fact, many of the known KIR, MIR, ILT, and CD94 / NKG2 receptor molecules may actually function with an accessory protein which is part of the functional receptor complex. SeeOlcese, et al. (1997) J. Immunol. 158:5083‑5086; and Takase, et al. (1997) J. Immunol. 159:741‑747.

[0217] A DAP 12 domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a cytoplasmic domain of a DAP 12, and will typically include an ITAM domain. In an embodiment a DAP 12 domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurringDAP12or aDAP12described herein. In embodiments theDAP12domain of aKIR-CARdiffers at nomore than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring DAP 12 or a DAP 12 described herein. In embodiments the DAP 12 domain of a KIR-CAR differs at no more than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurringDAP12 or aDAP12 described herein. In embodiments theDAP12 domain of a KIR- CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring DAP 12 or a DAP 12 described herein.

[0218] TheDAP10was identified partly by its homology to theDAP12, and other features. In particular, in contrast to the DAP12, which exhibits an ITAMactivationmotif, the DAP10 exhibits an ITIM inhibitorymotif. TheMDL‑1 was identified by its functional association with DAP12.

[0219] The functional interaction between, e.g., DAP12 or DAP10, and its accessory receptor may allow use of the structural combination in receptors which normally are not found in a truncated receptor form. Thus, the mechanism of signaling through such accessory proteins as the DAP12 and DAP10 allow for interesting engineering of other KIR-like receptor complexes, e.g., with theKIR,MIR, ILT, andCD94NKG2 type receptors. Truncated formsof intact receptorsmay be constructed which interact with a DAP12 or DAP10 to form a functional signaling complex.

[0220] The primate nucleotide sequence of DAP12 corresponds to SEQ ID NO: 6; the amino acid sequence corre- sponds to SEQ ID NO: 7. The signal sequence appears to run frommet(‑26) to gln(‑1) or a1a1; the mature protein should run from about a1a1 (or gln2), the extracellular domain from about a1a1 to pro14; the extracellular domain contains two cysteines at 7 and 9, which likely allow disulfide linkages to additional homotypic or heterotypic accessory proteins; the transmembrane region runs from about gly15 or va116 to about gly39; and an ITAMmotif from tyr65 to leu79 (YxxL‑6 / 8x- YxxL). The LVA03AESTwas identified andused to extract other overlapping sequences.SeealsoGenbankHumanESTs that are part of human DAP12; some, but not all, inclusive Genbank Accession # AA481924; H39980; W60940; N41026; R49793; W60864; W92376; H12338; T52100; AA480109; H12392; W74783; and T55959 INHIBITORY NKR-CARs

[0221] The present invention provides compositions andmethods for limiting the depletion of non-cancerous cells by a type of CAR Tcell therapy. As disclosed herein, a type of CAR Tcell therapy comprises the use of NK receptors including but is not limited to activating and inhibitory receptors of NK cells known as killer cell immunoglobulin-like receptor (KIR). Accordingly the invention provides compositions andmethods of using a NKR-CAR, e.g., a KIR-CAR, including but is not limited to an activating NKR-CAR (actNKR-CAR), e.g., an activating KIR-CAR (actKIR-CAR) and an inhibitory NKR-CAR (inhNKR-CAR), e.g., an inhibitory KIR-CAR (inhKIR-CAR).

[0222] In some embodiments, the KIR of an inhKIR-CARs is an inhibitory KIR that comprises a long cytoplasmic tail containing Immunoreceptor Tyrosine-based InhibitoryMotif (ITIM),which transduce inhibitory signals (referredelsewhere herein as inhKIR-CAR). 30 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0223] In someembodiments, an inhKIR-CARcomprisesacytoplasmicdomainofan inhibitorymoleculeother thanKIR. These inhibitory molecules can, in some embodiments, decrease the ability of a cell to mount an immune effector response. Cytoplasmic domains of inhibitory molecules may be coupled, e.g., by fusion, to transmembrane domains of KIR. Exemplary inhibitory molecules are shown in table 1: Table 1 Inhibitory molecules CD160 2B4 PD1 TIM3 LAG3 TIGIT CTLA‑4 BTLA LAIR1 PD-L1 VISTA

[0224] In some embodiments, an inhKIR-CAR comprises a PD1 cytoplasmic domain. A PD1 cytoplasmic domain, as that term isusedherein, refers toapolypeptidedomainhavingstructural and functional propertiesof a cytoplasmicdomain of a PD1. In an embodiment the PD1 cytoplasmic domain of a KIR-CAR has at least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring PD1 cytoplasmic domain or a PD1 cytoplasmic domain described herein (SEQ IDNO: 14). In embodiments thePD1cytoplasmic domain of aKIR-CARdiffers at nomore than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring PD1 cytoplasmic domain or a PD1 cytoplasmic domain described herein. In embodiments thePD1 cytoplasmic domain of aKIR-CARdiffers at nomore than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring PD1 cytoplasmic domain or a PD1 cytoplasmic domain described herein. In embodiments the PD1 cytoplasmic domain of a KIR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring PD1 cytoplasmic domain or a PD1 cytoplasmic domain described herein.

[0225] In some embodiments, an inhKIR-CAR comprises a CTLA‑4 cytoplasmic domain. A CTLA‑4 cytoplasmic domain, as that term is used herein, refers to a polypeptide domain having structural and functional properties of a cytoplasmic domain of aCTLA‑4. In an embodiment theCTLA‑4 cytoplasmic domain of aKIR-CARhasat least 70, 80, 85, 90, 95, or 99% homology with a reference sequence, e.g., a naturally occurring CTLA‑4 cytoplasmic domain or a CTLA‑4 cytoplasmic domain described herein (SEQ ID NO: 15). In embodiments the CTLA‑4 cytoplasmic domain of a KIR-CAR differs at no more than 15, 10, 5, 2, or 1% of its residues from a reference sequence, e.g., a naturally occurring CTLA‑4 cytoplasmic domain or aCTLA‑4 cytoplasmic domain described herein. In embodiments theCTLA‑4 cytoplasmic domain of a KIR-CAR differs at nomore than 5, 4, 3, 2 or 1 residue from a reference sequence, e.g., a naturally occurring CTLA‑4 cytoplasmic domain or aCTLA‑4 cytoplasmic domain described herein. In embodiments theCTLA‑4 cytoplasmic domain of a KIR-CAR does not differ from, or shares 100% homology with, a reference sequence, e.g., a naturally occurring CTLA‑4 cytoplasmic domain or a CTLA‑4 cytoplasmic domain described herein.

[0226] In an embodiment, an inhNKR-CAR, e.g., an inhKIR-CAR, upon engagement with an antigen on a non-target or bystander cell, inactivates the cytotoxic cell comprising the inhNKR-CAR. While much of the description below relates to inhKIR-CARs, the invention includes the analogous application of other inhNKR-CARs.

[0227] In one embodiment, Tcells expressing the actKIR-CAR exhibit an antitumor property when bound to its target, whereas Tcells expressing an inhKIR-CAR results in inhibition of cell activity when the inhKIR-CAR is bound to its target.

[0228] Regardless of the type of KIR-CAR, KIR-CARs are engineered to comprise an extracellular domain having an antigen binding domain fused to a cytoplasmic domain. In one embodiment, KIR-CARs, when expressed in a Tcell, are able to redirect antigen recognitionbasedupon theantigenspecificity.Anexemplaryantigen isCD19because this antigen is expressed onB cell lymphoma.However, CD19 is also expressed on normal B cells, and thusCARs comprising an anti- CD19 domainmay result in depletion of normal B cells. Depletion of normal B cells canmake a treated subject susceptible to infection, as B cells normally aid Tcells in the control of infection. The present invention provides for compositions and methods to limit the depletion of normal tissue during KIR-CAR Tcell therapy. In one embodiment, the present invention 31 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 provides methods to treat cancer and other disorders using KIR-CAR Tcell therapy while limiting the depletion of healthy bystander cells.

[0229] In oneembodiment, the inventioncomprises controllingor regulatingKIR-CARTcell activity. In oneembodiment, the invention comprises compositions andmethods related to geneticallymodifyingTcells to express aplurality of types of KIR-CARs, where KIR-CARTcell activation is dependent on the binding of a plurality of types of KIR-CARs to their target receptor. Dependence on the binding of a plurality of types of KIR-CARs improves the specificity of the lytic activity of the KIR-CAR T cell, thereby reducing the potential for depleting normal healthy tissue.

[0230] In another embodiment, the invention comprises compositions and methods related to genetically modifying T cells with an inhibitory KIR-CAR. In one embodiment, the inhibitory KIR-CAR comprises an extracellular antigen binding domain that recognizes an antigen associated with a normal, non-cancerous, cell and an inhibitory cytoplasmic domain.

[0231] In one embodiment, the invention provides a dual KIR-CAR where a Tcell is genetically modified to express an inhKIR-CAR and an actKIR-CAR. In one embodiment, binding of the inhKIR-CAR to a normal, non-cancerous cell results in the inhibition of the dual KIR-CARTcell. For example, in one embodiment, binding of the inhKIR-CAR to a normal, non- cancerous cell results in the death of the dual KIR-CAR Tcell. In another embodiment, binding of the inhKIR-CAR to a normal, non-cancerous cell results in inhibiting the signal transduction of the actKIR-CAR. In yet another embodiment, binding of the inhKIR-CAR to a normal, non-cancerous cell results in the induction of a signal transduction signal that prevents the actKIR-CAR Tcell from exhibiting its anti-tumor activity. Accordingly, the dual KIR-CAR comprising at least one inhKIR-CAR and at least one actKIR-CAR of the invention provides a mechanism to regulate the activity of the dual KIR-CAR T cell.

[0232] In one embodiment, the present invention provides methods for treating cancer and other disorders using KIR- CAR T cell therapies while minimizing the depletion of normal healthy tissue. The cancer may be a hematological malignancy, a solid tumor, a primary or a metastasizing tumor. Other diseases treatable using the compositions and methods of the invention include viral, bacterial and parasitic infections as well as autoimmune diseases. EXTRACELLULAR HINGE DOMAIN

[0233] Extracellular hinge domain, as that term is used herein, refers to a polypeptide sequence of a NKCAR disposed between the transmembrane domain and antigen binding domain. In an embodiment the extracellular hinge domain allows sufficient distance from the outer surface of the cell and the antigen binding domain as well as flexibility tominimize steric hinderance between the cell and the antigen binding domain. In an embodiment the extracellular hinge domain is sufficiently short or flexible that it does not interfere with engagement of the cell that includes the NKCARwith an antigen bearing cell, e.g., a target cell. In an embodiment the extracellular hinge domain is from 2 to 20, 5 to 15, 7 to 12, or 8 to 10 amino acids in length. In an embodiment the hinge domain includes at least 50, 20, or 10 residues. In embodiments the hinge is 10 to 300, 10 to 250, or 10 t 200 residues in length. In an embodiment the distance from which the hinge extends from the cell is sufficiently short that the hinge does not hinder engagement with the surface of a target cell. In an embodiment thehingeextends less than20, 15, or 10nanometers from thesurfaceof the cytotoxic cell. Thus, suitability for a hinge can be influenced by both linear length, the number of amino acid residues and flexibility of the hinge. An IgG4 hingecanbeas longas200aminoacids in length, but thedistance it extends from thesurfaceof the cytotoxic cell is smaller due to Ig-domain folding. A CD8alpha hinge, which is ~43 amino acids is rather linear at ~ 8 nm in length. In contrast, the IgG4C2&C3 hinge) is ~200 amino acids in length, but has a distance from the cytotoxic cell surface comparable to that of the CD8 alpha hinge. While not wishing to be bound by theory, the similarity in extension is influenced by flexibility.

[0234] In some instances, the extracellular hinge domain is, e.g., a hinge from a human protein, a fragment thereof, or a short oligo‑ or polypeptide linker.

[0235] In some embodiments, the hinge is an artificial sequence. In one embodiment, the hinge is a short oligopeptide linker comprising a glycine-serine doublet.

[0236] In some embodiments, the hinge is a naturally occurring sequence. In some embodiments, the hinge can be a human Ig (immunoglobulin) hinge, or fragment thereof. In one embodiment, for example, the hinge comprises (e.g., consists of) the amino acid sequence of the IgG4 hinge (SEQ ID NO: 49). In one embodiment, for example, the hinge comprises (e.g., consists of) the amino acid sequence of the IgD hinge (SEQ IDNO: 50). In someembodiments, the hinge can be a human CD8 hinge, or fragment thereof. In one embodiment, for example, the hinge comprises (e.g., consists of) the amino acid sequence of the CD8 hinge (SEQ ID NO: 51). TCARS

[0237] In some embodiments, theCAR cell therapy of the present invention comprises NKR-CAR in combination with a TCAR. In one embodiment, a TCAR comprises an antigen binding domain fused to an intracellular domain. In embodi- ments, an intracellular signaling domain produces an intracellular signal when an extracellular domain, e.g., an antigen binding domain, towhich it is fused binds a counter ligand. Intracellular signaling domains can include primary intracellular 32 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 signaling domains and costimulatory signaling domains. In an embodiment, a TCAR molecule can be constructed for expression in an immune cell, e.g., a Tcell, such that the TCARmolecule comprises a domain, e.g., a primary intracellular signaling domains, costimulatory signaling domain, inhibitory domains, etc., that is derived from a polypeptide that is typically associatedwith the immunecell. For example, aTCAR for expression in aTcell can comprise a 41BBdomain and anCD3zeta domain. In this instance, both the41BBandCD3zeta domains arederived frompolypeptides associatedwith theTcell. In another embodiment, aTCARmolecule canbeconstructed for expression in an immunecell e.g., aTcell, such that the TCAR molecule comprises a domain that is derived from a polypeptide that is not typically associated with the immune cell. Alternatively, a TCAR for expression in a NK cell can comprise a 41BB domain and a CD3 zeta domain derived from a T cell (See e.g. WO2013 / 033626, incorporated herein by reference). PRIMARY INTRACELLUAR SIGNALING DOMAIN

[0238] In some embodiments a primary intracellular signaling domain produces an intracellular signal when an extracellular domain, e.g., an antigen binding domain, to which it is fused binds cognate antigen. It is derived from a primary stimulatory molecule, e.g., it comprises intracellular sequence of a primary stimulatory molecule. It comprises sufficient primary stimulatorymolecule sequence to produce an intracellular signal, e.g., when an antigen binding domain to which it is fused binds cognate antigen.

[0239] A primary stimulatory molecule, is a molecule, that upon binding cognate ligand, mediates an immune effector response, e.g., in the cell inwhich it is expressed. Typically, it generates an intracellular signal that is dependent onbinding to a cognate ligand that comprises antigen. The TCR / CD3 complex is an exemplary primary stimulatory molecule; it generates an intracellular signal upon binding to cognate ligand, e.g., an MHCmolecule loaded with a peptide. Typically, e.g., in the case of the TCR / CD3 primary stimulatory molecule, the generation of an intracellular signal by a primary intracellular signaling domain is dependent on binding of the primary stimulatory molecule to antigen.

[0240] Primary stimulation can mediate altered expression of certain molecules, such as downregulation of TGF-β, and / or reorganization of cytoskeletal structures, and the like.Stimulation, can, e.g., in the presenceof costimulation, result in an optimization, e.g., an increase, in an immune effector function of the Tcell. Stimulation, e.g., in the context of a Tcell, can mediate a T cell response, e.g., proliferation, activation, differentiation, and the like.

[0241] In an embodiment, the primary intracellular signaling domain comprises a signaling motif, e.g., an immunor- eceptor tyrosine-based activationmotif or ITAMs. A primary intracellular signaling domain can comprise ITAM containing cytoplasmic signaling sequences from TCR zeta (CD3 zeta), FcR gamma, FcR beta, CD3 gamma, CD3 delta, CD3 epsilon, CD5, CD22, CD79a, CD79b, CD278 (also known as "ICOS")and CD66d.

[0242] Exemplary primary intracellular signaling domains are provided in Table 2. Table 2 Primary Intracellular Signaling Domains In embodiments the domain comprises an ITAM TCR zeta FcR gamma FcR beta CD3 gamma CD3 delta CD3 epsilon CD79a CD79b CD66d DAP10 DAP12

[0243] A primary intracellular signaling domain comprises a functional fragment, or analog, of a primary stimulatory molecule (e.g., CD3 zeta - GenBank Acc. No. BAG36664.1) It can comprise the entire intracellular region or a fragment of the intracellular regionwhich is sufficient for generationof an intracellular signalwhenanantigenbindingdomain towhich it is fused, or coupled by a dimerization switch, binds cognate antigen. In embodiments the primary intracellular signaling domain has at least 70, 75. 80. 85, 90, 95, 98, or 99 % sequence identity with a naturally occurring primary stimulatory 33 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 molecule, e.g., a human (GenBank Acc. No. BAG36664.1), or other mammalian, e.g., a nonhuman species, e.g., rodent, monkey, ape or murine intracelluar primary stimulatory molecule. In embodiments the primary intracellular signaling domain has at least 70, 75. 80. 85, 90, 95, 98, or 99 % sequence identity with SEQ ID NO: 13.

[0244] In embodiments the primary intracellular signaling domain, has at least 70, 75, 80, 85, 90, 95, 96, 97, 98, or 99% identity with, or differs by no more than 30, 25, 20, 15, 10, 5, 4, 3, 2, or 1 amino acid residues from the corresponding residues of a naturally occurring human primary stimulatory molecule, e.g., a naturally occurring human primary stimulatory molecule disclosed herein. COSTIMULATORY SIGNALING DOMAIN

[0245] In an embodiment, a costimulatory signaling domain produces an intracellular signal when an extracellular domain, e.g., an antigen binding domain towhich it is fused, or coupled by a dimerization switch, binds cognate ligand. It is derived from a costimulatory molecule. It comprises sufficient primary costimulatory molecule sequence to produce an intracellular signal, e.g.,whenanextracellular domain, e.g., anantigenbindingdomain, towhich it is fused, or coupledbya dimerization switch, binds cognate ligand.

[0246] Costimulatory molecules are cell surface molecules, other than antigen receptors or their counter ligands, that promote an immune effector response. In some cases they are required for an efficient or enhanced immune response. Typically, a costimulatory molecule generates an intracellular signal that is dependent on binding to a cognate ligand that is, in embodiments, other than an antigen, e.g., the antigen recognized by an antigen binding domain of a Tcell. Typically, signaling from a primary stimulatory molecule and a costimulatory molecule contribute to an immune effector response, and in some cases both are required for efficient or enhanced generation of an immune effector response.

[0247] A costimulatory domain comprises a functional fragment, or analog, of a costimulatorymolecule (e.g., 4‑1BB). It can comprise the entire intracellular region or a fragment of the intracellular region which is sufficient for generation of an intracellular signal, e.g., when an antigen binding domain to which it is fused, or coupled by a dimerization switch, binds cognate antigen. In embodiments the costimulatory domain has at least 70, 75. 80. 85, 90, 95, 98, or 99 % sequence identity with a naturally occurring costimulatory molecule, e.g., a human, or other mammalian, e.g., a nonhuman species, e.g., rodent, monkey, ape or murine intracelluar costimulatory molecule. In embodiments the costimatory domain has at least 70, 75. 80. 85, 90, 95, 98, or 99 % sequence identity with SEQ ID NO: 12.

[0248] Exemplary costimulatory signaling domains (intracellular signaling domains) are provided in Table 3. Table 3: Costimulatory Signaling Domains for RCARX (identified by the Costimulatory Molecules from which they are derived) CD27 CD28, 4‑1BB (CD137) OX40 CD30 CD40 ICOS (CD278) ICAM‑1 LFA‑1 (CD11a / CD18) CD2 CD7 LIGHT NKG2C B7-H3 a ligand that specifically binds with CD83

[0249] In embodiments the costimulatory signaling domain, has at least 70, 75, 80, 85, 90, 95, 96, 97, 98, or 99% identity with, or differs by nomore than 30, 25, 20, 15, 10, 5, 4, 3, 2, or 1 amino acid residues from the corresponding residues of a naturally occurring human primary stimulatory molecule, e.g., a naturally occurring human costimulatory molecule 34 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 disclosed herein. ANTIGEN BINDING DOMAIN

[0250] The CARs described herein, e.g., the KIR-CARs described herein, include an antigen binding domain in the extracellular region.An "antigenbindingdomain"as the term isusedherein, refers toamolecule that hasaffinity for a target antigen, typically an antigen on a target cell, e.g., a cancer cell. An exemplary antigen binding domain comprises a polypeptide, e.g., an antibody molecule (which includes an antibody, and antigen binding fragments thereof, e.g., a immunoglobulin, single domain antibody (sdAb), and an scFv), or a non-antibody scaffold, e.g., a fibronectin, and the like. In embodiments, the antigen binding domain is a single polypeptide. In embodiments, the antigen binding domain comprises, one, two, or more, polypeptides.

[0251] Thechoiceofanantigenbindingdomaincandependupon the typeandnumberof ligandsor receptors that define the surface of a target cell. For example, the antigen binding domainmay be chosen to recognize a ligand or receptor that acts as a cell surface marker on target cells associated with a particular disease state. Examples of cell surface markers that may act as ligands or receptors include a cell surface marker associated with a particular disease state, e.g., cell surfacemakers for viral diseases, bacterial diseases parasitic infections, autoimmunediseases and disorders associated with unwanted cell proliferation, e.g., a cancer, e.g., a cancer described herein.

[0252] In the context of the present disclosure, "tumor antigen" or "proliferative disorder antigen" or "antigen associated with a proliferative disorder" refers to antigens that are common to specific proliferative disorders. In certain aspects, the proliferative disorder antigens of the present invention are derived from, cancers including but not limited to primary or metastatic melanoma, thymoma, lymphoma, sarcoma, lung cancer (e.g., NSCLC or SCLC), liver cancer, non-Hodgkin’s lymphoma, Hodgkin’s lymphoma, leukemias, multiple myeloma, glioblastoma, neuroblastoma, uterine cancer, cervical cancer, renal cancer, thyroid cancer, bladder cancer, kidneycancer andadenocarcinomassuchasbreast cancer, prostate cancer, ovarian cancer, pancreatic cancer, colon cancer and the like. In some embodiments, the cancer is B-cell acute lymphoid leukemia ("BALL"), T-cell acute lymphoid leukemia ("TALL"), acute lymphoid leukemia (ALL), acute myelo- genous leukemia (AML); oneormore chronic leukemias includingbut not limited to chronicmyelogenous leukemia (CML), chronic lymphocytic leukemia (CLL); additional hematologic cancersorhematologic conditions including, but not limited to B cell prolymphocytic leukemia, blastic plasmacytoid dendritic cell neoplasm, Burkitt’s lymphoma, diffuse large B cell lymphoma, follicular lymphoma, hairy cell leukemia, small cell‑ or a large cell-follicular lymphoma, malignant lympho- proliferative conditions, MALT lymphoma, mantle cell lymphoma, Marginal zone lymphoma, multiple myeloma, myelo- dysplasia and myelodysplastic syndrome, non-Hodgkin’s lymphoma, plasmablastic lymphoma, plasmacytoid dendritic cell neoplasm, Waldenstrom macroglobulinemia,

[0253] In one embodiment, the tumor antigen comprises one or more antigenic cancer epitopes immunologically recognized by tumor infiltrating lymphocytes (TIL) derived from a cancer tumor of a mammal.

[0254] Tumor antigens are proteins that are produced by tumor cells that elicit an immune response, particularly T-cell mediated immune responses. The selection of the antigen binding domain of the invention will depend on the particular type of cancer to be treated. Tumor antigens are well known in the art and include, for example, a glioma-associated antigen, carcinoembryonic antigen (CEA), EGFRvIII, IL‑11Ra, IL‑13Ra, EGFR, B7H3, Kit, CA-IX, CS‑1, MUC1, BCMA, bcr-abl, HER2, β-human chorionic gonadotropin, alphafetoprotein (AFP), ALK, CD19, CD123, cyclin B1, lectin-reactive AFP, Fos-related antigen 1, ADRB3, thyroglobulin, EphA2, RAGE‑1, RU1, RU2, SSX2, AKAP‑4, LCK, OY-TES1, PAX5, SART3, CLL‑1, fucosyl GM1, GloboH, MN-CA IX, EPCAM, EVT6-AML, TGS5, human telomerase reverse transcriptase, plysialic acid, PLAC1, RU1, RU2 (AS), intestinal carboxyl esterase, lewisY, sLe, LY6K, mut hsp70‑2, M-CSF, MYCN, RhoC, TRP‑2, CYP1B1, BORIS, prostase, prostate-specific antigen (PSA), PAX3, PAP, NY-ESO‑1, LAGE‑1a, LMP2, NCAM, p53, p53 mutant, Ras mutant, gp100, prostein, OR51E2, PANX3, PSMA, PSCA, Her2 / neu, hTERT, HMWMAA, HAVCR1, VEGFR2, PDGFR-beta, legumain, HPV E6,E7, survivin and telomerase, sperm protein 17, SSEA‑4, tyrosi- nase, TARP, WT1, prostate-carcinoma tumor antigen‑1 (PCTA‑1), ML-IAP, MAGE, MAGE-A1,MAD-CT‑1, MAD-CT‑2, MelanA / MART1, XAGE1, ELF2M, ERG (TMPRSS2ETS fusion gene), NA17, neutrophil elastase, sarcoma translocation breakpoints, NY-BR‑1, ephrinB2, CD20, CD22, CD24, CD30, CD33, CD38, CD44v6, CD97, CD171, CD179a, androgen receptor, insulin growth factor (IGF)‑I, IGF-II, IGF-I receptor, GD2, o-acetyl-GD2, GD3, GM3, GPRC5D, GPR20, CXORF61, folate receptor (FRa), folate receptor beta, ROR1, Flt3, TAG72, TN Ag, Tie 2, TEM1, TEM7R, CLDN6, TSHR,UPK2, andmesothelin. Inapreferredembodiment, the tumorantigen is selected from thegroupconsistingof folate receptor (FRa),mesothelin, EGFRvIII, IL‑13Ra,CD123,CD19,CD33,BCMA,GD2,CLL‑1,CA-IX,MUC1,HER2, andany combination thereof.

[0255] In one embodiment, the tumor antigen comprises one or more antigenic cancer epitopes associated with a malignant tumor. Malignant tumors express a number of proteins that can serve as target antigens for an immune attack. These molecules include but are not limited to tissue-specific antigens such as MART‑1, tyrosinase and GP 100 in melanoma and prostatic acid phosphatase (PAP) and prostate-specific antigen (PSA) in prostate cancer. Other target antigens include transformation-relatedmolecules such as the oncogeneHER‑2 / Neu / ErbB‑2. Yet another group of target 35 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 antigens are onco-fetal antigens suchas carcinoembryonic antigen (CEA). In B-cell lymphoma the tumor-specific idiotype immunoglobulin constitutes a truly tumor-specific immunoglobulin antigen that is unique to the individual tumor. B-cell differentiation antigens such as CD19, CD20 and CD37 are other candidates for target antigens in B-cell lymphoma.

[0256] Non-limiting examples of tumor antigens include the following: Differentiation antigens such asMART‑1 / MelanA (MART-I), gp100 (Pmel 17), tyrosinase, TRP‑1, TRP‑2 and tumor-specific multilineage antigens such as MAGE‑1, MAGE‑3, BAGE, GAGE‑1, GAGE‑2, p15; overexpressed embryonic antigens such as CEA; overexpressed oncogenes andmutated tumor-suppressor genes such as p53, Ras, HER‑2 / neu; unique tumor antigens resulting from chromosomal translocations; such as BCR-ABL, E2A-PRL, H4-RET, IGH-IGK, MYL-RAR; and viral antigens, such as the Epstein Barr virus antigens EBVA and the human papillomavirus (HPV) antigens E6 and E7. Other large, protein-based antigens include TSP‑180, MAGE‑4, MAGE‑5, MAGE‑6, RAGE, NY-ESO, p185erbB2, p180erbB‑3, c-met, nm‑23H1, PSA, TAG‑72, CA 19‑9, CA 72‑4, CAM 17.1, NuMa, K-ras, beta-Catenin, CDK4, Mum‑1, p 15, p 16, 43‑9F, 5T4, 791Tgp72, alpha-fetoprotein, beta-HCG,BCA225, BTAA,CA125,CA15‑3\CA27.29\BCAA,CA195,CA242,CA‑50,CAM43,CD68 \P1, CO‑029, FGF‑5, G250, Ga733\EpCAM, HTgp‑175, M344, MA‑50, MG7-Ag, MOV18, NB / 70K, NY-CO‑1, RCAS1, SDCCAG16, TA‑90\Mac‑2 binding protein\cyclophilin C-associated protein, TAAL6, TAG72, TLP, and TPS.

[0257] Depending on the desired antigen to be targeted, the CAR of the invention can be engineered to include the appropriate antigen bind domain that is specific to the desired antigen target. Antigen binding domains derived from an Antibody Molecule

[0258] The antigen binding domain can be derived from an antibody molecule, e.g., one or more of monoclonal antibodies, polyclonal antibodies, recombinant antibodies, human antibodies, humanized antibodies, single-domain antibodies e.g., a heavy chain variable domain (VH), a light chain variable domain (VL) and a variable domain (VHH) from, e.g., humanor camelid origin. In some instances, it is beneficial for the antigenbindingdomain to bederived from the same species in which the CAR will ultimately be used in, e.g., for use in humans, it may be beneficial for the antigen binding domain of the CAR, e.g., the KIR-CAR, e.g., described herein, to comprise a human or a humanized antigen binding domain. Antibodies can be obtained using known techniques known in the art.

[0259] The term "antibody," as used herein, refers to an immunoglobulin molecule which specifically binds with a target antigen. An antibody can be intact immunoglobulin derived from natural sources or from recombinant sources and can be an immunoreactive portionof intact immunoglobulin. Antibodiesare typically tetramersof immunoglobulinmolecules. The antibody molecule described herein may exist in a variety of forms where the antigen binding portion of the antibody is expressed as part of a contiguous polypeptide chain including, for example, a single domain antibody fragment (sdAb), a single chain antibody (scFv) and a humanized or human antibody, e.g., as described herein.

[0260] The term "antibody fragment" refers to a portion of an intact antibody and refers to the antigenic determining variable regionsof an intact antibody.Examplesof antibody fragments include, but arenot limited to, a single chaindomain antibody (sdAb), Fab, Fab’, F(ab’)2, andFv fragments, linear antibodies, scFv antibodies, a linear antibody, single domain antibody such as an sdAb (either VL or VH), a camelid VHH domain, and multispecific antibodies formed from antibody fragments.

[0261] An "antibody heavy chain," as usedherein, refers to the larger of the two types of polypeptide chains present in all antibody molecules in their naturally occurring conformations.

[0262] An "antibody light chain," as used herein, refers to the smaller of the two types of polypeptide chains present in all antibodymolecules in their naturallyoccurringconformations.κandλ light chains refer to the twomajorantibody light chain isotypes.

[0263] By the term "synthetic antibody" as used herein, is meant an antibody molecule which is generated using recombinant DNA technology, such as, for example, an antibody molecule expressed by a bacteriophage as described herein. The term should also be construed tomean an antibodymolecule which has been generated by the synthesis of a DNAmolecule encoding the antibodymolecule andwhichDNAmolecule expresses an antibody protein, or an amino acid sequence specifying the antibody, wherein the DNA or amino acid sequence has been obtained using synthetic DNA or amino acid sequence technology which is available and well known in the art.

[0264] In embodiments, the antigen binding domain comprises a fragment of an antibody that is sufficient to confer recognition and specific binding to the target antigen. Examples of an antibody fragment include, but are not limited to, an Fab, Fab’, F(ab’)2, or Fv fragment, an scFv antibody fragment, a linear antibody, single domain antibody such as an sdAb (either VL or VH), a camelid VHH domain, and multi-specific antibodies formed from antibody fragments.

[0265] In an embodiment, the antigen binding domain is a "scFv," which can comprise a fusion protein comprising a VL chain and a VH chain of an antibody, where the VH and VL are, e.g., linked via a short flexible polypeptide linker, e.g., a linker described herein. The scFv is capable of being expressed as a single chain polypeptide and retains the specificity of the intact antibody fromwhich it is derived.Moreover, theVL andVHvariable chains canbe linked in either order, e.g., with respect to the N-terminal and C-terminal ends of the polypeptide, the scFv may comprise VL-linker-VH or may comprise VH-linker-VL. An scFv that can be prepared according to method known in the art (see, for example, Bird et al., (1988) 36 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 Science 242:423‑426 and Huston et al., (1988) Proc. Natl. Acad. Sci. USA 85:5879‑5883).

[0266] Asdescribedaboveandelsewhere, scFvmolecules canbeproducedby linkingVHandVLchians together using flexible polypeptide linkers. In some embodiments, the scFv molecules comprise flexible polypeptide linker with an optimized length and / or amino acid composition. The flexible polypeptide linker length can greatly affect how the variable regions of a scFv fold and interact. In fact, if a short polypeptide linker is employed (e.g., between 5‑10 amino acids, intrachain folding is prevented. For examples of linker orientation and size see, e.g., Hollinger et al. 1993 Proc Natl Acad. Sci. U.S.A. 90:6444‑6448, U.S. Patent Application Publication Nos. 2005 / 0100543, 2005 / 0175606, 2007 / 0014794, and PCT publication Nos. WO2006 / 020258 and WO2007 / 024715, is incorporated herein by reference. In one embodiment, the peptide linker of the scFv consists of amino acids such as glycine and / or serine residues used alone or in combination, to link variable heavy and variable light chain regions together. In one embodiment, the flexible polypeptide linker is a Gly / Ser linker and, e.g., comprises the amino acid sequence (Gly-Gly-Gly-Ser)n, where n is a positive integer equal to or greater than 1. For example, n=1, n=2, n=3. n=4, n=5 and n=6, n=7, n=8, n=9 and n=10. In one embodiment, the flexible polypeptide linkers include, but are not limited to, (Gly4 Ser)4 or (Gly4 Ser)3. In another embodiment, the linkers include multiple repeats of (Gly2Ser), (GlySer) or (Gly3Ser).

[0267] In some embodiments, the antigen binding domain is a single domain antigen binding (SDAB) molecules. A SDABmolecule includes molecules whose complementary determining regions are part of a single domain polypeptide. Examples include, but are not limited to, heavy chain variable domains, bindingmolecules naturally devoid of light chains, singledomainsderived fromconventional 4-chainantibodies, engineereddomainsandsingledomainscaffoldsother than those derived from antibodies (e.g., described inmore detail below). SDABmoleculesmay be any of the art, or any future single domain molecules. SDABmolecules may be derived from any species including, but not limited to mouse, human, camel, llama, fish, shark, goat, rabbit, and bovine. This term also includes naturally occurring single domain antibody molecules from species other than Camelidae and sharks.

[0268] In one aspect, anSDABmolecule can be derived fromavariable region of the immunoglobulin found in fish, such as, for example, that which is derived from the immunoglobulin isotype known as Novel Antigen Receptor (NAR) found in the serum of shark. Methods of producing single domainmolecules derived from a variable region of NAR ("IgNARs") are described in WO 03 / 014161 and Streltsov (2005) Protein Sci. 14:2901‑2909.

[0269] According to another aspect, anSDABmolecule is a naturally occurring single domain antigen bindingmolecule known as a heavy chain devoid of light chains. Such single domainmolecules are disclosed inWO9404678 andHamers- Casterman, C. et al. (1993) Nature 363:446‑448, for example. For clarity reasons, this variable domain derived from a heavy chain molecule naturally devoid of light chain is known herein as a VHH or nanobody to distinguish it from the conventional VH of four chain immunoglobulins. Such a VHH molecule can be derived from Camelidae species, for example in camel, llama, dromedary, alpaca and guanaco. Other species besides Camelidaemay produce heavy chain molecules naturally devoid of light chain; such VHHs are within the scope of the invention.

[0270] In certain embodiments, the SDABmolecule is a single chain fusion polypeptide comprising one or more single domain molecules (e.g., nanobodies), devoid of a complementary variable domain or an immunoglobulin constant, e.g., Fc, region, that binds to one or more target antigens.

[0271] The SDAB molecules can be recombinant, CDR-grafted, humanized, camelized, de-immunized and / or in vitro generated (e.g., selected by phage display).

[0272] In one embodiment, the antigen biding domain portion comprises a human antibody or a fragment thereof.

[0273] In some embodiments, a non-human antibody is humanized, where specific sequences or regions of the antibody are modified to increase similarity to an antibody naturally produced in a human. In an embodiment, the antigen binding domain is humanized.

[0274] Nonhumanantibodies can be humanized using a variety of techniques known in the art, e.g., CDR-grafting (see, e.g., European Patent No. EP 239,400; International Publication No. WO 91 / 09967; and U.S. Pat. Nos. 5,225,539, 5,530,101, and 5,585,089, eachofwhich is incorporated herein in its entirety by reference), veneering or resurfacing (see, e.g., European Patent Nos. EP 592,106 and EP 519,596; Padlan, 1991, Molecular Immunology, 28(4 / 5):489‑498; Studnicka et al., 1994, Protein Engineering, 7(6):805‑814; and Roguska et al., 1994, PNAS, 91:969‑973, each of which is incorporatedhereinby itsentiretyby reference), chainshuffling (see, e.g.,U.S.Pat.No.5,565,332,which is incorporated herein in its entirety by reference), and techniques disclosed in, e.g., U.S. Patent Application Publication No. US2005 / 0042664, U.S. Patent Application Publication No. US2005 / 0048617, U.S. Pat. No. 6,407,213, U.S. Pat. No. 5,766,886, International Publication No. WO 9317105, Tan et al., 2002, J. Immunol., 169:1119‑25; Caldas et al., 2000, Protein Eng., 13(5):353‑60; Morea et al., 2000, Methods, 20:267‑79; Baca et al., 1997, J. Biol. Chem., 272:10678‑84; Roguska et al., 1996, Protein Eng., 9(10):895‑904; Couto et al., 1995, Cancer Res., 55 :5973s‑5977; Couto et al., 1995, Cancer Res., 55(8):1717‑22; Sandhu 1994Gene, 150(2):409‑10; and Pedersen et al., 1994, J. Mol. Biol., 235(3):959‑73, eachofwhich is incorporatedherein in its entirety by reference.Often, framework residues in the framework regionswill be substituted with the corresponding residue from the CDR donor antibody to alter, for example improve, antigen binding. These framework substitutions are identified bymethods well-known in the art, e.g., bymodeling of the interactions of the CDR and framework residues to identify framework residues important for antigen binding and sequence comparison to 37 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 identify unusual framework residues at particular positions. (See, e.g., Queen et al., U.S. Pat. No. 5,585,089; and Riechmann et al., 1988, Nature, 332:323, which are incorporated herein by reference in their entireties.). In preferred embodiments, the humanized antibody molecule comprises a sequence described herein, e.g., a variable light chain and / or a variable heavy chain described herein, e.g., a humanized variable light chain and / or variable heavy chain described in Table 4.

[0275] Ahumanized antibody has one ormore amino acid residues introduced into it from a sourcewhich is nonhuman. Thesenonhumanaminoacid residues are often referred to as "import" residues,which are typically taken froman "import" variable domain. Thus, humanized antibodies comprise one or more CDRs from nonhuman immunoglobulin molecules and framework regions fromhuman.Humanization of antibodies is well-known in the art and can essentially be performed following the method of Winter and co-workers (Jones et al., Nature, 321:522‑525 (1986); Riechmann et al., Nature, 332:323‑327 (1988); Verhoeyen et al., Science, 239:1534‑1536 (1988)), by substituting rodent CDRs or CDR sequences for the corresponding sequences of a human antibody, i.e., CDR-grafting (EP 239,400; PCT Publication No. WO 91 / 09967; and U.S. Pat. Nos. 4,816,567; 6,331,415; 5,225,539; 5,530,101; 5,585,089; 6,548,640, the contents of which are incorporated herein by reference herein in their entirety). In such humanized chimeric antibodies, substantially less than an intact human variable domain has been substituted by the corresponding sequence from a nonhuman species. In practice, humanized antibodies are typically human antibodies in which some CDR residues and possibly some frame- work (FR) residues are substituted by residues fromanalogous sites in rodent antibodies. Humanization of antibodies can also be achieved by veneering or resurfacing (EP 592,106; EP 519,596; Padlan, 1991, Molecular Immunology, 28(4 / 5):489‑498; Studnicka et al., Protein Engineering, 7(6):805‑814 (1994); and Roguska et al., PNAS, 91:969‑973 (1994)) or chain shuffling (U.S. Pat. No. 5,565,332), the contents of which are incorporated herein by reference herein in their entirety.

[0276] In some embodiments, the antibody of the invention is further prepared using an antibody having one or more of the VH and / or VL sequences disclosed herein can be used as starting material to engineer a modified antibody, which modifiedantibodymayhavealteredpropertiesascompared to thestartingantibody. In variousembodiments, theantibody is engineered by modifying one or more amino acids within one or both variable regions (i.e., VH and / or VL), for example within one or more CDR regions and / or within one or more framework regions.

[0277] In another aspect, the antigen binding domain is a Tcell receptor ("TCR"), or a fragment thereof, for example, a single chain TCR (scTCR).Methods tomake such TCRs is known in the art. See, e.g.,WillemsenRA et al, Gene Therapy 7: 1369‑1377 (2000); Zhang Tet al, Cancer Gene Ther 11: 487‑496 (2004); Aggen et al, Gene Ther. 19(4):365‑74 (2012) (references are incorporated herein by its entirety). For example, scTCR can be engineered that contains the Vα and Vβ genes fromaTcell clone linkedbya linker (e.g., a flexible peptide). This approach is veryuseful to cancerassociated target that itself is intracellular, however, a fragment of such antigen (peptide) is presented on the surface of the cancer cells by MHC. Table 4: Exemplary Antigen Binding Domains Target Antigen Name Amino Acid Sequence CD19 huscFv1 CD19 huscFv2 CD19 huscFv3 38 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 (continued) Target Antigen Name Amino Acid Sequence CD19 huscFv4 CD19 huscFv5 CD19 huscFv6 CD19 huscFv7 CD19 huscFv8 CD19 huscFv9 CD19 Hu scFv10 CD19 Hu scFv11 CD19 Hu scFv12 CD19 muCTL 019 39 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 (continued) Target Antigen Name Amino Acid Sequence CD123 Mu1172 CD123 Mu1176 CD123 huscFv1 CD123 huscFv2 CD123 huscFv3 CD123 huscFv4 CD123 huscFv5 CD123 huscFv6 CD123 huscFv7 40 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 (continued) Target Antigen Name Amino Acid Sequence CD123 huscFv8 EGFRvIII huscFv1 EGFRvIII huscFv2 EGFRvIII huscFv3 EGFRvIII huscFv4 EGFRvIII huscFv5 EGFRvIII huscFv6 EGFRvIII huscFv7 EGFRvIII huscFv8 41 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 (continued) Target Antigen Name Amino Acid Sequence EGFRvIII Mu310C mesothel in ss1 Non-Antibody Scaffolds

[0278] In embodiments, the antigen binding domain comprises a non antibody scaffold, e.g., a fibronectin, ankyrin, domain antibody, lipocalin, small modular immuno-pharmaceutical, maxybody, Protein A, or affilin. The non antibody scaffold has the ability to bind to target antigen on a cell. In embodiments, the antigen binding domain is a polypeptide or fragment thereof of a naturally occurring protein expressed on a cell. In some embodiments, the antigen binding domain comprises a non-antibody scaffold. A wide variety of non-antibody scaffolds can be employed so long as the resulting polypeptide includes at least one binding region which specifically binds to the target antigen on a target cell.

[0279] Non-antibody scaffolds include: fibronectin (Novartis, MA), ankyrin (Molecular Partners AG, Zurich, Switzer- land), domain antibodies (Domantis, Ltd., Cambridge, MA, and Ablynx nv, Zwijnaarde, Belgium), lipocalin (Pieris Proteolab AG, Freising, Germany), small modular immuno-pharmaceuticals (Trubion Pharmaceuticals Inc., Seattle, WA), maxybodies (Avidia, Inc., Mountain View, CA), Protein A (Affibody AG, Sweden), and affilin (gamma-crystallin or ubiquitin) (Scil Proteins GmbH, Halle, Germany).

[0280] Fibronectin scaffolds can be based on fibronectin type III domain (e.g., the tenthmodule of the fibronectin type III (10 Fn3 domain)). The fibronectin type III domain has 7 or 8 beta strands which are distributed between two beta sheets, which themselves pack against each other to form the core of the protein, and further containing loops (analogous to CDRs) which connect the beta strands to each other and are solvent exposed. There are at least three such loops at each edge of the beta sheet sandwich, where the edge is the boundary of the protein perpendicular to the direction of the beta strands (see US 6,818,418). Because of this structure, this non-antibody scaffold mimics antigen binding properties that are similar in nature and affinity to those of antibodies. These scaffolds can be used in a loop randomization and shuffling strategy in vitro that is similar to the process of affinity maturation of antibodies in vivo.

[0281] The ankyrin technology is based on using proteins with ankyrin derived repeat modules as scaffolds for bearing variable regions which can be used for binding to different targets. The ankyrin repeat module is a 33 amino acid polypeptide consisting of two anti-parallel α-helices and a B-turn. Binding of the variable regions is mostly optimized by using ribosome display.

[0282] Avimers arederived fromnatural A-domain containing protein suchasHER3.Thesedomains areusedbynature for protein-protein interactions and in human over 250 proteins are structurally based on A-domains. Avimers consist of a number of different "A-domain"monomers (2‑10) linked via aminoacid linkers. Avimers can be created that can bind to the target antigen using the methodology described in, for example, U.S. Patent Application Publication Nos. 20040175756; 20050053973; 20050048512; and 20060008844.

[0283] Affibody affinity ligands are small, simple proteins composed of a three-helix bundle based on the scaffold of one of the IgG-binding domains of Protein A. Protein A is a surface protein from the bacterium Staphylococcus aureus. This scaffold domain consists of 58 amino acids, 13 of which are randomized to generate affibody libraries with a large number of ligand variants (See e.g., US 5,831,012). Affibodymolecules mimic antibodies, they have amolecular weight of 6 kDa, compared to the molecular weight of antibodies, which is 150 kDa. In spite of its small size, the binding site of affibody molecules is similar to that of an antibody.

[0284] Protein epitopemimetics (PEM) aremedium-sized, cyclic, peptide-likemolecules (MW1‑2kDa)mimicking beta- 42 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 hairpin secondary structures of proteins, the major secondary structure involved in protein-protein interactions MISMATCHED ANTIGEN BINDING DOMAINS

[0285] It has been discovered, that cells having a plurality of chimericmembrane embedded receptors each comprising an antigen binding domain (CMERs) that interactions between the antigen binding domain of the CMER can be undesirable, e.g., because it inhibits the ability of one or more of the antigen binding domains to bind its cognate antigen. Accordingly, disclosed herein are a first and a second non-naturally occurring CMERcomprising antigen binding domains that minimize such interactions when expressed in the same cell. In an embodiment a plurality of CMERs comprises two TCARs. In an embodiment a plurality of CMERs comprises a TCAR and another CMER. In an embodiment a plurality of CMERscomprises twoNKR-CARs. Inanembodiment aplurality ofCMERscomprisesaNKR-CARandanotherCMER. In an embodiment a plurality of CMERs comprises a TCAR and an NKR-CAR.

[0286] In some embodiments, the claimed invention comprises a first and second CMER, wherein the antigen binding domain of one of said first CMER said second CMER does not comprise a variable light domain and a variable heavy domain. In some embodiments, the antigen binding domain of one of said first CMER said second CMER is an scFv, and the other is not an scFv. In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises a single VHdomain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VHdomain derived froma human or mouse sequence or a non-antibody scaffold. In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises a nanobody. In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises a camelid VHH domain.

[0287] In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence. In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises an scFv, and the other comprises a nanobody. In some embodiments, the antigen binding domain of one of said first CMER said second CMER comprises comprises an scFv, and the other comprises a camelid VHH domain.

[0288] In some embodiments, when present on the surface of a cell, binding of the antigen binding domain of said first CMER to its cognate antigen is not substantially reduced by the presence of said second CMER. In some embodiments, binding of the antigen binding domain of said first CMER to its cognate antigen in the presence of said second CMER is 85%, 90%, 95%, 96%, 97%, 98%or 99%of binding of the antigen binding domain of said first CMER to its cognate antigen in the absence of said second CMER.

[0289] In someembodiments,whenpresent on the surfaceof a cell, theantigenbindingdomains of said firstCMERsaid second CMER, associate with one another less than if both were scFv antigen binding domains. In some embodiments, the antigen binding domains of said first CMER said second CMER, associate with one another 85%, 90%, 95%, 96%, 97%, 98% or 99% less than if both were scFv antigen binding domains.

[0290] In someembodiments, theclaimed inventioncomprisesafirst andsecondKIR-CAR,wherein theantigenbinding domain of one of said first KIR-CARsaid secondKIR-CARdoes not comprise a variable light domain and a variable heavy domain. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR is an scFv, and the other is not an scFv. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human ormouse sequence or a non-antibody scaffold. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR comprises a nanobody. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR comprises a camelid VHH domain.

[0291] In someembodiments, the antigen binding domain of one of said first KIR-CARsaid secondKIR-CARcomprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR comprises an scFv, and the other comprises a nanobody. In some embodiments, the antigen binding domain of one of said first KIR-CAR said second KIR-CAR comprises an scFv, and the other comprises a camelid VHH domain.

[0292] In some embodiments, when present on the surface of a cell, binding of the antigen binding domain of said first KIR-CAR to its cognate antigen is not substantially reduced by the presence of said second KIR-CAR. In some embodiments, binding of the antigen binding domain of said first KIR-CAR to its cognate antigen in the presence of said secondKIR-CAR is 85%,90%, 95%, 96%,97%, 98%or 99%of bindingof the antigenbindingdomain of said first KIR- CAR to its cognate antigen in the absence of said second KIR-CAR.

[0293] In some embodiments, when present on the surface of a cell, the antigen binding domains of said first KIR-CAR said second KIR-CAR, associate with one another less than if both were scFv antigen binding domains. In some embodiments, the antigen binding domains of said first KIR-CAR said second KIR-CAR, associate with one another 85%, 90%, 95%, 96%, 97%, 98% or 99% less than if both were scFv antigen binding domains. 43 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0294] In some embodiments, the claimed invention comprises a first and second TCAR, wherein the antigen binding domain of one of said first TCAR said second TCAR does not comprise a variable light domain and a variable heavy domain. In someembodiments, theantigenbindingdomainof oneof said first TCARsaid secondTCAR isanscFv, and the other is not an scFv. In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises a single VHdomain, e.g., a camelid, shark, or lamprey single VHdomain, or a single VHdomain derived froma human or mouse sequence or a non-antibody scaffold. In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises a nanobody. In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises a camelid VHH domain.

[0295] In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises an scFv, and the other comprises a single VH domain, e.g., a camelid, shark, or lamprey single VH domain, or a single VH domain derived from a human or mouse sequence. In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises an scFv, and the other comprises a nanobody. In some embodiments, the antigen binding domain of one of said first TCAR said second TCAR comprises an scFv, and the other comprises a camelid VHH domain.

[0296] In some embodiments, when present on the surface of a cell, binding of the antigen binding domain of said first TCAR to its cognate antigen is not substantially reduced by the presence of said second TCAR. In some embodiments, bindingof theantigenbindingdomainof said firstTCAR to its cognateantigen in thepresenceof saidsecondTCAR is85%, 90%, 95%, 96%, 97%, 98%or 99%of binding of the antigen binding domain of said first TCAR to its cognate antigen in the absence of said second TCAR. In some embodiments, when present on the surface of a cell, the antigen binding domains of said first TCAR said second TCAR,associatewithoneanother less than if bothwere scFvantigenbindingdomains. In someembodiments, theantigen binding domains of said first TCAR said second TCAR, associate with one another 85%, 90%, 95%, 96%, 97%, 98% or 99% less than if both were scFv antigen binding domains. VECTORS

[0297] The present invention also provides vectors in which a DNA of the present invention is inserted. Vectors derived from retroviruses such as the lentivirus are suitable tools to achieve long-term gene transfer since they allow long-term, stable integration of a transgene and its propagation in daughter cells. Lentiviral vectors have the added advantage over vectors derived fromonco-retroviruses such asmurine leukemia viruses in that they can transduce non-proliferating cells, such as hepatocytes. They also have the added advantage of low immunogenicity. In brief summary, the expression of natural or synthetic nucleic acids encodingCARs is typically achievedbyoperably linking anucleic acid encoding theCAR polypeptideorportions thereof toapromoter, and incorporating theconstruct intoanexpressionvector. Thevectors canbe suitable for replication and integration eukaryotes. Typical cloning vectors contain transcription and translation termina- tors, initiation sequences, and promoters useful for regulation of the expression of the desired nucleic acid sequence.

[0298] Thenucleic acid canbecloned into anumber of types of vectors. For example, the nucleic acid canbe cloned into a vector including, but not limited to a plasmid, a phagemid, a phage derivative, an animal virus, and a cosmid. Vectors of particular interest include expression vectors, replication vectors, probe generation vectors, and sequencing vectors.

[0299] Further, the expression vectormay be provided to a cell in the formof a viral vector. Viral vector technology iswell known in the art and is described, for example, in Sambrook et al., 2012, MOLECULAR CLONING: A LABORATORY MANUAL, volumes 1 ‑4, Cold Spring Harbor Press, NY), and in other virology and molecular biology manuals. Viruses, which are useful as vectors include, but are not limited to, retroviruses, adenoviruses, adeno‑ associated viruses, herpes viruses, and lentiviruses. In general, a suitable vector contains anorigin of replication functional in at least one organism, a promoter sequence, convenient restriction endonuclease sites, andoneormore selectablemarkers, (e.g.,WO01 / 96584; WO 01 / 29058; and U.S. Pat. No. 6,326,193).

[0300] Additional promoter elements, e.g., enhancers, regulate the frequency of transcriptional initiation. Typically, these are located in the region 30‑110 bp upstream of the start site, although a number of promoters have recently been shown to contain functional elements downstream of the start site as well. The spacing between promoter elements frequently is flexible, so that promoter function is preserved when elements are inverted ormoved relative to one another. In the thymidine kinase (tk) promoter, the spacing between promoter elements can be increased to 50 bp apart before activity begins to decline.Depending on thepromoter, it appears that individual elements can function either cooperatively or independently to activate transcription.

[0301] An example of a promoter is the immediate early cytomegalovirus (CMV) promoter sequence. This promoter sequence is a strong constitutive promoter sequence capable of driving high levels of expression of any polynucleotide sequenceoperatively linked thereto.However, other constitutivepromoter sequencesmayalsobeused, including, but not limited to the simian virus 40 (SV40) early promoter, mouse mammary tumor virus (MMTV), human immunodeficiency virus (HIV) long terminal repeat (LTR) promoter, MoMuLV promoter, an avian leukemia virus promoter, an Epstein-Barr virus immediateearlypromoter, aRoussarcomaviruspromoter, theelongation factor‑1αpromoter, aswell ashumangene 44 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 promoters such as, but not limited to, the actin promoter, themyosin promoter, the hemoglobin promoter, and the creatine kinase promoter. Further, the invention should not be limited to the use of constitutive promoters. Inducible promoters are alsocontemplatedaspart of the invention.Theuseof an induciblepromoter providesamolecular switchcapableof turning on expression of the polynucleotide sequencewhich it is operatively linkedwhen such expression is desired, or turning off the expression when expression is not desired. Examples of inducible promoters include, but are not limited to a metallothionine promoter, a glucocorticoid promoter, a progesterone promoter, and a tetracycline promoter.

[0302] In one embodiment, the vector is a lentivirus vector. In one embodiment, the vector further comprises apromoter. In one embodiment, the promoter is an EF‑1 promoter.

[0303] In oneembodiment, thevector is an in vitro transcribedvector, e.g., a vector that transcribesRNAofanucleic acid molecule described herein. In one embodiment, the nucleic acid sequence in the vector further comprises a poly(A) tail, e.g., a poly A tail described herein, e.g., comprising about 150 adenosine bases. In one embodiment, the nucleic acid sequence in the vector further comprises a 3’UTR.

[0304] Inorder toassess theexpressionof aCARpolypeptideorportions thereof, theexpressionvector tobe introduced intoacell canalsocontaineitheraselectablemarkergeneor a reporter geneorboth to facilitate identificationandselection of expressing cells from the population of cells sought to be transfected or infected through viral vectors. In other aspects, theselectablemarkermaybecarriedonaseparatepieceofDNAandused inaco‑ transfectionprocedure.Both selectable markers and reporter genesmay be flankedwith appropriate regulatory sequences to enable expression in the host cells. Useful selectable markers include, for example, antibiotic-resistance genes, such as neo and the like.

[0305] Reporter genes are used for identifying potentially transfected cells and for evaluating the functionality of regulatory sequences. In general, a reporter gene is a gene that is not present in or expressed by the recipient organismor tissue and that encodes a polypeptide whose expression is manifested by some easily detectable property, e.g., enzymatic activity. Expression of the reporter gene is assayed at a suitable time after the DNA has been introduced into the recipient cells. Suitable reporter genes may include genes encoding luciferase, beta-galactosidase, chloram- phenicol acetyl transferase, secreted alkaline phosphatase, or the green fluorescent protein gene (e.g., Ui-Tei et al., 2000 FEBS Letters 479: 79‑82). Suitable expression systems are well known andmay be prepared using known techniques or obtainedcommercially. In general, the constructwith theminimal 5’ flanking regionshowing thehighest level of expression of reporter gene is identifiedas thepromoter. Suchpromoter regionsmaybe linked toa reporter geneandused toevaluate agents for the ability to modulate promoter‑ driven transcription.

[0306] Methods of introducing and expressing genes into a cell are known in the art. In the context of an expression vector, the vector can be readily introduced into a host cell, e.g., mammalian, bacterial, yeast, or insect cell by anymethod in theart. For example, the expression vector can be transferred into a host cell by physical, chemical, or biologicalmeans.

[0307] Polynucleotides can be introduced into target cells using any of a number of different methods, for instance, commercially available methods which include, but are not limited to, electroporation (Amaxa Nucleofector-II (Amaxa Biosystems, Cologne, Germany)), (ECM830 (BTX) (Harvard Instruments, Boston, Mass.) or theGene Pulser II (BioRad, Denver, Colo.), Multiporator (Eppendort, Hamburg Germany), cationic liposome mediated transfection using lipofection, polymer encapsulation, peptidemediated transfection, or biolistic particle delivery systems such as "gene guns" (see, for example, Nishikawa, et al. Hum Gene Ther., 12(8):861‑70 (2001).

[0308] Physical methods for introducing a polynucleotide into a host cell include calcium phosphate precipitation, lipofection, particle bombardment, microinjection, electroporation, and the like. Methods for producing cells comprising vectors and / or exogenous nucleic acids arewell-known in the art. See, for example, Sambrook et al., 2012,MOLECULAR CLONING: A LABORATORY MANUAL, volumes 1 ‑4, Cold Spring Harbor Press, NY).

[0309] Biological methods for introducing a polynucleotide of interest into a host cell include the use of DNA and RNA vectors. Viral vectors, andespecially retroviral vectors, havebecome themostwidely usedmethod for inserting genes into mammalian, e.g., human cells. Other viral vectors can be derived from lentivirus, poxviruses, herpes simplex virus I, adenoviruses and adeno-associated viruses, and the like. See, for example, U.S. Pat. Nos. 5,350,674 and 5,585,362.

[0310] Chemical means for introducing a polynucleotide into a host cell include colloidal dispersion systems, such as macromolecule complexes, nanocapsules, microspheres, beads, and lipid-based systems including oil-in-water emul- sions, micelles, mixedmicelles, and liposomes. An exemplary colloidal system for use as a delivery vehicle in vitro and in vivo is a liposome (e.g. , an artificial membrane vesicle).

[0311] Disclosed herein are methods for producing an in vitro transcribed RNA NK-CAR. The present invention also includesanNK-CARencodingRNAconstruct that canbedirectly transfected intoacell.Amethod forgeneratingmRNAfor use in transfection can involve in vitro transcription (IVT) of a template with specially designed primers, followed by polyA addition, to produce a construct containing 3’ and 5’ untranslated sequence ("UTR"), a 5’ cap and / or Internal Ribosome Entry Site (IRES), the nucleic acid to be expressed, and a polyA tail, typically 50‑2000 bases in. RNA so produced can efficiently transfect different kinds of cells. In one aspect, the template includes sequences for the NK-CAR.

[0312] In one aspect theNK-CAR is encoded by amessengerRNA (mRNA). In one aspect themRNAencoding theNK- CAR is introduced into a T cell for production of a NK-CAR cell.

[0313] In one embodiment, the in vitro transcribed RNA NK-CAR can be introduced to a cell as a form of transient 45 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 transfection. TheRNA is produced by in vitro transcription using a polymerase chain reaction (PCR)‑generated template. DNA of interest from any source can be directly converted by PCR into a template for in vitro mRNA synthesis using appropriate primers and RNA polymerase. The source of the DNA can be, for example, genomic DNA, plasmid DNA, phage DNA, cDNA, synthetic DNA sequence or any other appropriate source of DNA. In one embodiment, the desired temple for in vitro transcription is a NK-CAR of the present invention. For example, the template for the RNA NK-CAR comprises an extracellular region comprising a single chain variable domain of an anti-tumor antibody; a hinge region, a transmembrane domain (e.g., a transmembrane domain of KIR). In one embodiment, the desired temples for in vitro transcription comprises KIR-CAR and DAP12 on separate templates. In one embodiment, the desired temple for in vitro transcription comprises KIR-CAR and DAP12 on the same template. The template for DAP12 comprises a transmem- brane domain and an intracellular region.

[0314] In one embodiment, the DNA to be used for PCR contains an open reading frame. The DNA can be from a naturally occurring DNA sequence from the genome of an organism. In one embodiment, the nucleic acid can include some or all of the 5’ and / or 3’ untranslated regions (UTRs). The nucleic acid can include exons and introns. In one embodiment, theDNA to be used for PCR is a human nucleic acid sequence. In another embodiment, theDNA to be used for PCR is a human nucleic acid sequence including the 5’ and 3’ UTRs. The DNA can alternatively be an artificial DNA sequence that is not normally expressed in a naturally occurring organism. An exemplary artificial DNA sequence is one that contains portions of genes that are ligated together to form an open reading frame that encodes a fusion protein. The portions of DNA that are ligated together can be from a single organism or from more than one organism.

[0315] PCR is used to generate a template for in vitro transcription of mRNAwhich is used for transfection. Methods for performing PCR are well known in the art. Primers for use in PCR are designed to have regions that are substantially complementary to regions of the DNA to be used as a template for the PCR. "Substantially complementary," as used herein, refers to sequences of nucleotides where amajority or all of the bases in the primer sequence are complementary, or one or more bases are non-complementary, or mismatched. Substantially complementary sequences are able to anneal or hybridize with the intendedDNA target under annealing conditions used for PCR. The primers can be designed tobesubstantially complementary toanyportionof theDNAtemplate.Forexample, theprimerscanbedesigned toamplify the portion of a nucleic acid that is normally transcribed in cells (the open reading frame), including 5’ and 3’ UTRs. The primers can also be designed to amplify a portion of a nucleic acid that encodes a particular domain of interest. In one embodiment, the primers are designed to amplify the coding region of a human cDNA, including all or portions of the 5’ and 3’ UTRs. Primers useful for PCR can be generated by synthetic methods that are well known in the art. "Forward primers" are primers that contain a region of nucleotides that are substantially complementary to nucleotides on the DNA template that areupstreamof theDNAsequence that is to beamplified. "Upstream" is usedherein to refer to a location5, to theDNA sequence to be amplified relative to the coding strand. "Reverse primers" are primers that contain a region of nucleotides that are substantially complementary to a double-stranded DNA template that are downstream of the DNA sequence that is to be amplified. "Downstream" is used herein to refer to a location 3’ to the DNA sequence to be amplified relative to the coding strand.

[0316] AnyDNApolymeraseuseful forPCRcanbeused in themethodsdisclosedherein. The reagentsandpolymerase are commercially available from a number of sources.

[0317] Chemical structures with the ability to promote stability and / or translation efficiencymay also be used. The RNA preferablyhas5’ and3’UTRs. Inoneembodiment, the5’UTR isbetweenoneand3000nucleotides in length.The lengthof 5’ and 3’UTRsequences to be added to the coding region can bealtered by differentmethods, including, but not limited to, designing primers for PCR that anneal to different regions of the UTRs. Using this approach, one of ordinary skill in the art can modify the 5’ and 3’ UTR lengths required to achieve optimal translation efficiency following transfection of the transcribed RNA.

[0318] The 5’ and 3’ UTRs can be the naturally occurring, endogenous 5’ and 3’ UTRs for the nucleic acid of interest. Alternatively, UTR sequences that are not endogenous to the nucleic acid of interest can be added by incorporating the UTR sequences into the forward and reverse primers or by any other modifications of the template. The use of UTR sequences that arenot endogenous to thenucleic acidof interest canbeuseful formodifying thestability and / or translation efficiency of the RNA. For example, it is known that AU-rich elements in 3’ UTR sequences can decrease the stability of mRNA. Therefore, 3’ UTRs can be selected or designed to increase the stability of the transcribed RNA based on properties of UTRs that are well known in the art.

[0319] In one embodiment, the 5’ UTR can contain the Kozak sequence of the endogenous nucleic acid. Alternatively, when a 5’ UTR that is not endogenous to the nucleic acid of interest is being added by PCR as described above, a consensus Kozak sequence can be redesigned by adding the 5’ UTR sequence. Kozak sequences can increase the efficiency of translation of some RNA transcripts, but does not appear to be required for all RNAs to enable efficient translation. The requirement for Kozak sequences for manymRNAs is known in the art. In other embodiments the 5’ UTR can be 5’UTR of an RNA virus whose RNA genome is stable in cells. In other embodiments various nucleotide analogues can be used in the 3’ or 5’ UTR to impede exonuclease degradation of the mRNA.

[0320] To enable synthesis of RNA from a DNA template without the need for gene cloning, a promoter of transcription 46 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 should beattached to theDNA template upstreamof the sequence to be transcribed.Whenasequence that functions asa promoter for an RNA polymerase is added to the 5’ end of the forward primer, the RNA polymerase promoter becomes incorporated into the PCR product upstream of the open reading frame that is to be transcribed. In one preferred embodiment, the promoter is a T7 polymerase promoter, as described elsewhere herein. Other useful promoters include, but are not limited to, T3 and SP6 RNA polymerase promoters. Consensus nucleotide sequences for T7, T3 and SP6 promoters are known in the art.

[0321] In apreferredembodiment, themRNAhasbotha capon the5’ endanda3’ poly(A) tailwhichdetermine ribosome binding, initiation of translation and stabilitymRNA in the cell.Ona circularDNA template, for instance, plasmidDNA,RNA polymerase produces a long concatameric product which is not suitable for expression in eukaryotic cells. The transcription of plasmid DNA linearized at the end of the 3’ UTR results in normal sized mRNA which is not effective in eukaryotic transfection even if it is polyadenylated after transcription.

[0322] Ona linearDNA template, phageT7RNApolymerasecanextend the3’ endof the transcript beyond the last base of the template (Schenborn and Mierendorf, Nuc Acids Res., 13:6223‑36 (1985); Nacheva and Berzal-Herranz, Eur. J. Biochem., 270:1485‑65 (2003).

[0323] The conventional method of integration of polyA / Tstretches into a DNA template is molecular cloning. However polyA / T sequence integrated into plasmid DNA can cause plasmid instability, which is why plasmid DNA templates obtained from bacterial cells are often highly contaminated with deletions and other aberrations. This makes cloning proceduresnot only laboriousand timeconsumingbut oftennot reliable.That iswhyamethodwhichallowsconstructionof DNA templates with polyA / T 3’ stretch without cloning highly desirable.

[0324] ThepolyA / Tsegment of the transcriptionalDNA template can beproducedduringPCRbyusing a reverse primer containing a polyT tail, such as 100T tail (size can be 50‑5000 T (SEQ ID NO: 111)), or after PCR by any other method, including, but not limited to, DNA ligation or in vitro recombination. Poly(A) tails also provide stability to RNAs and reduce their degradation.Generally, the lengthof a poly(A) tail positively correlateswith the stability of the transcribedRNA. In one embodiment, the poly(A) tail is between 100 and 5000 adenosines.

[0325] Poly(A) tails ofRNAs canbe further extended following in vitro transcriptionwith the use of a poly(A) polymerase, such asE. coli polyA polymerase (E-PAP). In one embodiment, increasing the length of a poly(A) tail from100 nucleotides to between 300 and 400 nucleotides results in about a two-fold increase in the translation efficiency of the RNA. Additionally, the attachment of different chemical groups to the 3’ end can increase mRNA stability. Such attachment can contain modified / artificial nucleotides, aptamers and other compounds. For example, ATP analogs can be incorpo- rated into the poly(A) tail using poly(A) polymerase. ATP analogs can further increase the stability of the RNA.

[0326] 5’ caps on also provide stability to RNAmolecules. In a preferred embodiment, RNAs produced by the methods disclosedherein includea5’ cap.The5’ cap isprovidedusing techniquesknown in theart anddescribedherein (Cougot, et al., Trends inBiochem.Sci., 29:436‑444 (2001); Stepinski, et al., RNA, 7:1468‑95 (2001);Elango, et al., Biochim.Biophys. Res. Commun., 330:958‑966 (2005)).

[0327] The RNAs produced by the methods disclosed herein can also contain an internal ribosome entry site (IRES) sequence. The IRES sequence may be any viral, chromosomal or artificially designed sequence which initiates cap- independent ribosome binding to mRNA and facilitates the initiation of translation. Any solutes suitable for cell electro- poration, which can contain factors facilitating cellular permeability and viability such as sugars, peptides, lipids, proteins, antioxidants, and surfactants can be included. In the case where a non-viral delivery system is utilized, an exemplary delivery vehicle is a liposome. The use of lipid formulations is contemplated for the introduction of the nucleic acids into a host cell (in vitro, ex vivo or in vivo). In another aspect, the nucleic acid may be associated with a lipid. The nucleic acid associated with a lipidmay be encapsulated in the aqueous interior of a liposome, interspersed within the lipid bilayer of a liposome, attached to a liposome via a linkingmolecule that is associated with both the liposome and the oligonucleotide, entrapped ina liposome, complexedwitha liposome,dispersed inasolutioncontaininga lipid,mixedwitha lipid, combined witha lipid, containedasasuspension ina lipid, containedor complexedwithamicelle, or otherwiseassociatedwitha lipid. Lipid, lipid / DNA or lipid / expression vector associated compositions are not limited to any particular structure in solution. For example, theymay be present in a bilayer structure, asmicelles, or with a "collapsed" structure. Theymay also simply be interspersed in a solution, possibly forming aggregates that are not uniform in size or shape. Lipids are fatty substances whichmay benaturally occurring or synthetic lipids. For example, lipids include the fatty droplets that naturally occur in the cytoplasm aswell as the class of compounds which contain long-chain aliphatic hydrocarbons and their derivatives, such as fatty acids, alcohols, amines, amino alcohols, and aldehydes.

[0328] Lipids suitable for use can be obtained from commercial sources. For example, dimyristyl phosphatidylcholine ("DMPC") canbe obtained fromSigma,St. Louis,MO; dicetyl phosphate ("DCP") canbe obtained fromK&KLaboratories (Plainview, NY); cholesterol ("Choi") can be obtained from Calbiochem-Behring; dimyristyl phosphatidylglycerol ("DMPG") and other lipids may be obtained from Avanti Polar Lipids, Inc. (Birmingham, AL.). Stock solutions of lipids in chloroform or chloroform / methanol can be stored at about ‑20°C. Chloroform is used as the only solvent since it ismore readily evaporated than methanol. "Liposome" is a generic term encompassing a variety of single and multilamellar lipid vehicles formed by the generation of enclosed lipid bilayers or aggregates. Liposomes can be characterized as having 47 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 vesicular structures with a phospholipid bilayer membrane and an inner aqueous medium. Multilamellar liposomes have multiple lipid layers separated by aqueous medium. They form spontaneously when phospholipids are suspended in an excess of aqueous solution. The lipid components undergo self-rearrangement before the formation of closed structures and entrap water and dissolved solutes between the lipid bilayers (Ghosh et al., 1991 Glycobiology 5: 505‑10). However, compositions that have different structures in solution than the normal vesicular structure are also encompassed. For example, the lipids may assume a micellar structure or merely exist as nonuniform aggregates of lipid molecules. Also contemplated are lipofectamine-nucleic acid complexes. SOURCES OF T CELLS

[0329] Prior to expansion and genetic modification, a source of Tcells is obtained from a subject. The term "subject" is intended to include living organisms in which an immune response can be elicited (e.g., mammals). Examples of subjects include humans, dogs, cats,mice, rats, and transgenic species thereof. Tcells can be obtained froma number of sources, including peripheral blood mononuclear cells, bone marrow, lymph node tissue, cord blood, thymus tissue, tissue from a site of infection, ascites, pleural effusion, spleen tissue, and tumors. In certain embodiments of the present invention, any number of T cell lines available in the art, may be used. In certain embodiments of the present invention, T cells can be obtained froma unit of blood collected froma subject using any number of techniques known to the skilled artisan, such as Ficoll™ separation. In one preferred embodiment, cells from the circulating blood of an individual are obtained by apheresis. The apheresis product typically contains lymphocytes, including T cells, monocytes, granulocytes, B cells, other nucleated white blood cells, red blood cells, and platelets. In one embodiment, the cells collected by apheresis may be washed to remove the plasma fraction and to place the cells in an appropriate buffer or media for subsequent processing steps. In one embodiment of the invention, the cells are washed with phosphate buffered saline (PBS). In an alternative embodiment, the wash solution lacks calcium and may lack magnesium or may lack many if not all divalent cations. Again, surprisingly, initial activation steps in the absence of calcium lead to magnified activation. As those of ordinary skill in theartwould readily appreciate awashingstepmaybeaccomplishedbymethodsknown to those in theart, such as by using a semi-automated "flow-through" centrifuge (for example, the Cobe 2991 cell processor, the Baxter CytoMate, or the Haemonetics Cell Saver 5) according to themanufacturer’s instructions. After washing, the cells may be resuspended in a variety of biocompatible buffers, such as, for example, Ca-free, Mg-free PBS, PlasmaLyte A, or other saline solution with or without buffer. Alternatively, the undesirable components of the apheresis samplemay be removed and the cells directly resuspended in culture media.

[0330] In another embodiment, Tcells are isolated from peripheral blood lymphocytes by lysing the red blood cells and depleting the monocytes, for example, by centrifugation through a PERCOLL™ gradient or by counterflow centrifugal elutriation. A specific subpopulation of Tcells, such as CD3+, CD28+, CD4+, CD8+, CD45RA+, and CD45RO+Tcells, can be further isolated by positive or negative selection techniques. For example, in one embodiment, Tcells are isolated by incubationwith anti-CD3 / anti-CD28 (i.e.,3x28)‑conjugated beads, suchasDYNABEADS®M‑450CD3 / CD28T, for a time period sufficient for positive selection of the desired Tcells. In one embodiment, the time period is about 30 minutes. In a further embodiment, the timeperiod ranges from30minutes to36hoursor longer andall integer values therebetween. Ina further embodiment, the timeperiod is at least 1, 2, 3, 4, 5, or 6 hours. In yet another preferredembodiment, the timeperiod is 10 to 24 hours. In one preferred embodiment, the incubation time period is 24 hours. For isolation of Tcells frompatients with leukemia, use of longer incubation times, such as 24 hours, can increase cell yield. Longer incubation times may be used to isolate Tcells in any situation where there are few Tcells as compared to other cell types, such in isolating tumor infiltrating lymphocytes (TIL) from tumor tissueor from immunocompromised individuals. Further, useof longer incubation times can increase the efficiency of capture of CD8+ Tcells. Thus, by simply shortening or lengthening the time Tcells are allowed tobind to theCD3 / CD28beadsand / orby increasingordecreasing the ratioof beads toTcells (asdescribed further herein), subpopulations of T cells can be preferentially selected for or against at culture initiation or at other time points during the process. Additionally, by increasing or decreasing the ratio of anti-CD3 and / or anti-CD28 antibodies on the beads or other surface, subpopulations of Tcells can be preferentially selected for or against at culture initiation or at other desired time points. The skilled artisanwould recognize thatmultiple rounds of selection can also be used in the context of this invention. In certain embodiments, it may be desirable to perform the selection procedure and use the "unselected" cells in the activation and expansion process. "Unselected" cells can also be subjected to further rounds of selection.

[0331] Enrichment of a T cell population by negative selection can be accomplished with a combination of antibodies directed to surfacemarkersunique to thenegatively selectedcells.Onemethod is cell sortingand / or selectionvianegative magnetic immunoadherence or flow cytometry that uses a cocktail of monoclonal antibodies directed to cell surface markerspresenton thecellsnegatively selected.Forexample, toenrich forCD4+cellsbynegativeselection, amonoclonal antibody cocktail typically includesantibodies toCD14,CD20,CD11b,CD16,HLA-DR, andCD8. In certain embodiments, it may be desirable to enrich for or positively select for regulatory Tcells which typically express CD4+, CD25+, CD62Lhi, GITR+, and FoxP3+. Alternatively, in certain embodiments, T regulatory cells are depleted by anti-C25 conjugated beads or other similar method of selection. 48 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0332] For isolation of a desired population of cells by positive or negative selection, the concentration of cells and surface (e.g., particles such asbeads) canbe varied. In certain embodiments, itmaybe desirable to significantly decrease the volume in which beads and cells are mixed together (i.e., increase the concentration of cells), to ensure maximum contact of cells and beads. For example, in one embodiment, a concentration of 2 billion cells / ml is used. In one embodiment, a concentration of 1 billion cells / ml is used. In a further embodiment, greater than100million cells / ml is used. Ina further embodiment, a concentrationof cells of 10, 15, 20, 25, 30, 35, 40, 45, or50millioncells / ml isused. In yet another embodiment, a concentration of cells from 75, 80, 85, 90, 95, or 100 million cells / ml is used. In further embodiments, concentrations of 125 or 150million cells / ml can be used. Using high concentrations can result in increased cell yield, cell activation, and cell expansion. Further, use of high cell concentrations allows more efficient capture of cells that may weakly express target antigens of interest, such as CD28-negative Tcells, or from samples where there are many tumor cells present (i.e., leukemic blood, tumor tissue, etc.).Such populations of cellsmay have therapeutic value andwould be desirable to obtain. For example, using high concentration of cells allows more efficient selection of CD8+ T cells that normally have weaker CD28 expression.

[0333] In a related embodiment, it may be desirable to use lower concentrations of cells. By significantly diluting the mixtureofTcells andsurface (e.g., particles suchasbeads), interactionsbetween theparticlesandcells isminimized.This selects for cells that express high amounts of desired antigens to be bound to the particles. For example, CD4+ T cells express higher levels of CD28 and are more efficiently captured than CD8+ T cells in dilute concentrations. In one embodiment, the concentration of cells used is 5 X 106 / ml. In other embodiments, the concentration used can be from about 1 X 105 / ml to 1 X 106 / ml, and any integer value in between.

[0334] In other embodiments, the cells may be incubated on a rotator for varying lengths of time at varying speeds at either 2‑10°C or at room temperature.

[0335] Tcells for stimulation can also be frozen after a washing step.Wishing not to be bound by theory, the freeze and subsequent thaw step provides a more uniform product by removing granulocytes and to some extent monocytes in the cell population. After the washing step that removes plasma and platelets, the cells may be suspended in a freezing solution.Whilemany freezing solutions and parameters are known in the art and will be useful in this context, onemethod involves using PBS containing 20% DMSO and 8% human serum albumin, or culture media containing 10% Dextran 40 and 5%Dextrose, 20%Human SerumAlbumin and 7.5%DMSO, or 31.25%Plasmalyte-A, 31.25%Dextrose 5%, 0.45% NaCl, 10% Dextran 40 and 5% Dextrose, 20% Human Serum Albumin, and 7.5% DMSO or other suitable cell freezing media containing for example, Hespan and PlasmaLyte A, the cells then are frozen to ‑80°C at a rate of 1° per minute and stored in the vapor phase of a liquid nitrogen storage tank. Other methods of controlled freezing may be used as well as uncontrolled freezing immediately at ‑20° C or in liquid nitrogen.

[0336] In certain embodiments, cryopreserved cells are thawed andwashed as described herein and allowed to rest for one hour at room temperature prior to activation using the methods of the present invention.

[0337] Also contemplated in the context of the invention is the collection of blood samples or apheresis product from a subject at a time period prior to when the expanded cells as described herein might be needed. As such, the source of the cells to beexpanded canbe collectedat any timepoint necessary, anddesired cells, suchasTcells, isolatedand frozen for later use in Tcell therapy for any number of diseases or conditions that would benefit from Tcell therapy, such as those described herein. In one embodiment a blood sample or an apheresis is taken from a generally healthy subject. In certain embodiments, a blood sample or an apheresis is taken from a generally healthy subject who is at risk of developing a disease, but who has not yet developed a disease, and the cells of interest are isolated and frozen for later use. In certain embodiments, the T cells may be expanded, frozen, and used at a later time. In certain embodiments, samples are collected from a patient shortly after diagnosis of a particular disease as described herein but prior to any treatments. In a further embodiment, the cells are isolated from a blood sample or an apheresis from a subject prior to any number of relevant treatmentmodalities, including but not limited to treatment with agents such as natalizumab, efalizumab, antiviral agents, chemotherapy, radiation, immunosuppressive agents, such as cyclosporin, azathioprine, methotrexate, myco- phenolate, and FK506, antibodies, or other immunoablative agents such as CAMPATH, anti-CD3 antibodies, cytoxan, fludarabine, cyclosporin, FK506, rapamycin,mycophenolic acid, steroids, FR901228, and irradiation. These drugs inhibit either the calcium dependent phosphatase calcineurin (cyclosporine and FK506) or inhibit the p70S6 kinase that is important for growth factor induced signaling (rapamycin). (Liu et al., Cell 66:807‑815, 1991; Henderson et al., Immun. 73:316‑321, 1991; Bierer et al., Curr. Opin. Immun. 5:763‑773, 1993). In a further embodiment, the cells are isolated for a patient and frozen for later use in conjunction with (e.g., before, simultaneously or following) bone marrow or stem cell transplantation, Tcell ablative therapy using either chemotherapy agents such as, fludarabine, external-beam radiation therapy (XRT), cyclophosphamide, or antibodies such as OKT3 or CAMPATH. In another embodiment, the cells are isolated prior to and canbe frozen for later use for treatment followingB-cell ablative therapy suchas agents that reactwith CD20, e.g., Rituxan.

[0338] Ina furtherembodiment of thepresent invention,Tcellsareobtained fromapatientdirectly following treatment. In this regard, it has been observed that following certain cancer treatments, in particular treatments with drugs that damage the immune system, shortly after treatment during the period when patients would normally be recovering from the 49 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 treatment, the quality of Tcells obtainedmay be optimal or improved for their ability to expand ex vivo. Likewise, following exvivomanipulationusing themethodsdescribedherein, thesecellsmaybe inapreferredstate for enhancedengraftment and in vivo expansion. Thus, it is contemplatedwithin the context of the present invention to collect blood cells, including T cells, dendritic cells, or other cells of the hematopoietic lineage, during this recovery phase. Further, in certain embodi- ments,mobilization (for example,mobilizationwithGM-CSF) and conditioning regimens canbe used to create a condition in a subject wherein repopulation, recirculation, regeneration, and / or expansion of particular cell types is favored, especially during a defined window of time following therapy. Illustrative cell types include Tcells, B cells, dendritic cells, and other cells of the immune system. ACTIVATION AND EXPANSION OF T CELLS

[0339] T cells are activated and expanded generally using methods as described, for example, in U.S. Patents 6,352,694; 6,534,055; 6,905,680; 6,692,964; 5,858,358; 6,887,466; 6,905,681; 7,144,575; 7,067,318; 7,172,869; 7,232,566; 7,175,843; 5,883,223; 6,905,874; 6,797,514; 6,867,041; and U.S. Patent Application Publication No. 20060121005.

[0340] Generally, the Tcells of the invention are expanded by contact with a surface having attached thereto an agent that stimulates a CD3 / TCR complex associated signal and a ligand that stimulates a co-stimulatory molecule on the surfaceof theTcells. Inparticular,Tcell populationsmaybestimulatedasdescribedherein, suchasbycontactwithananti- CD3 antibody, or antigen-binding fragment thereof, or an anti-CD2 antibody immobilized on a surface, or by contact with a protein kinase C activator (e.g., bryostatin) in conjunction with a calcium ionophore. For co-stimulation of an accessory molecule on the surface of the Tcells, a ligand that binds the accessory molecule is used. For example, a population of T cells can be contacted with an anti-CD3 antibody and an anti-CD28 antibody, under conditions appropriate for stimulating proliferationof theTcells. To stimulate proliferation of eitherCD4+Tcells orCD8+Tcells, an anti-CD3antibodyandananti- CD28 antibody. Examples of an anti-CD28 antibody include 9.3, B-T3, XR-CD28 (Diaclone, Besançon, France) can be usedascanothermethodscommonly known in theart (Bergetal., TransplantProc. 30(8):3975‑3977, 1998;Haanenetal., J. Exp. Med. 190(9):13191328, 1999; Garland et al., J. Immunol Meth. 227(1‑2):53‑63, 1999).

[0341] In certainembodiments, theprimarystimulatorysignal and theco-stimulatorysignal for theTcellmaybeprovided by different protocols. For example, the agents providing each signal may be in solution or coupled to a surface. When coupled to asurface, theagentsmaybecoupled to the samesurface (i.e., in "cis" formation) or to separate surfaces (i.e., in "trans" formation). Alternatively, one agent may be coupled to a surface and the other agent in solution. In one embodiment, the agent providing the co-stimulatory signal is bound to a cell surface and the agent providing the primary activation signal is in solution or coupled to a surface. In certain embodiments, both agents can be in solution. In another embodiment, the agentsmay be in soluble form, and then cross-linked to a surface, such as a cell expressing Fc receptors or an antibody or other binding agent whichwill bind to the agents. In this regard, see for example, U.S. Patent Application PublicationNos. 20040101519 and 20060034810 for artificial antigen presenting cells (aAPCs) that are contemplated for use in activating and expanding T cells in the present invention.

[0342] In one embodiment, the two agents are immobilized on beads, either on the same bead, i.e., "cis," or to separate beads, i.e., "trans." By way of example, the agent providing the primary activation signal is an anti-CD3 antibody or an antigen-binding fragment thereof and the agent providing the co-stimulatory signal is an anti-CD28 antibody or antigen- binding fragment thereof; and both agents are co-immobilized to the same bead in equivalent molecular amounts. In one embodiment, a 1:1 ratio of each antibody bound to the beads forCD4+ Tcell expansion andTcell growth is used. In certain aspects of the present invention, a ratio of anti CD3:CD28antibodies bound to the beads is used such that an increase in T cell expansion is observed as compared to the expansion observed using a ratio of 1:1. In one particular embodiment an increase of from about 1 to about 3 fold is observed as compared to the expansion observed using a ratio of 1:1. In one embodiment, the ratio of CD3:CD28 antibody bound to the beads ranges from 100:1 to 1:100 and all integer values there between. In oneaspect of thepresent invention,moreanti-CD28antibody is bound to theparticles thananti-CD3antibody, i.e., the ratio of CD3:CD28 is less than one. In certain embodiments of the invention, the ratio of anti CD28 antibody to anti CD3 antibody bound to the beads is greater than 2:1. In one particular embodiment, a 1:100 CD3:CD28 ratio of antibody bound to beads is used. In another embodiment, a 1:75 CD3:CD28 ratio of antibody bound to beads is used. In a further embodiment, a 1:50 CD3:CD28 ratio of antibody bound to beads is used. In another embodiment, a 1:30 CD3:CD28 ratio of antibody bound to beads is used. In one preferred embodiment, a 1:10 CD3:CD28 ratio of antibody bound to beads is used. In another embodiment, a 1:3CD3:CD28 ratio of antibody bound to the beads is used. In yet another embodiment, a 3:1 CD3:CD28 ratio of antibody bound to the beads is used.

[0343] Ratios of particles to cells from1:500 to 500:1 and any integer values in betweenmay be used to stimulate Tcells or other target cells. As those of ordinary skill in the art can readily appreciate, the ratio of particles to cells may depend on particle size relative to the target cell. For example, small sized beads could only bind a few cells, while larger beads could bind many. In certain embodiments the ratio of cells to particles ranges from 1:100 to 100:1 and any integer values in- between and in further embodiments the ratio comprises 1:9 to 9:1 and any integer values in between, can also be used to 50 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 stimulateTcells. The ratio of anti-CD3-andanti-CD28-coupledparticles toTcells that result inTcell stimulationcanvaryas noted above, however certain preferred values include 1:100, 1:50, 1:40, 1:30, 1:20, 1:10, 1:9, 1:8, 1:7, 1:6, 1:5, 1:4, 1:3, 1:2, 1:1, 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 8:1, 9:1, 10:1, and 15:1 with one preferred ratio being at least 1:1 particles per Tcell. In one embodiment, a ratio of particles to cells of 1:1 or less is used. In one particular embodiment, a preferred particle: cell ratio is 1:5. In further embodiments, the ratio of particles to cells can be varied depending on the day of stimulation. For example, in one embodiment, the ratio of particles to cells is from 1:1 to 10:1 on the first day and additional particles are added to the cells every day or every other day thereafter for up to 10 days, at final ratios of from 1:1 to 1:10 (based on cell countson thedayof addition). Inoneparticular embodiment, the ratioof particles to cells is 1:1on thefirst dayof stimulation andadjusted to 1:5 on the third and fifth days of stimulation. In another embodiment, particles are addedonadaily or every otherdaybasis toafinal ratioof 1:1on thefirst day, and1:5on the thirdandfifthdaysof stimulation. Inanotherembodiment, the ratio of particles to cells is 2:1on the first dayof stimulationandadjusted to 1:10on the third andfifth daysof stimulation. Inanother embodiment, particlesareaddedonadaily or everyother daybasis to afinal ratio of 1:1on the first day, and1:10 on the third and fifth days of stimulation.Oneof skill in the art will appreciate that a variety of other ratiosmaybe suitable for use in the present invention. In particular, ratios will vary depending on particle size and on cell size and type.

[0344] In furtherembodimentsof thepresent invention, thecells, suchasTcells, arecombinedwithagent-coatedbeads, the beads and the cells are subsequently separated, and then the cells are cultured. In an alternative embodiment, prior to culture, the agent-coated beads and cells are not separated but are cultured together. In a further embodiment, the beads and cells are first concentrated by application of a force, such as a magnetic force, resulting in increased ligation of cell surface markers, thereby inducing cell stimulation.

[0345] By way of example, cell surface proteins may be ligated by allowing paramagnetic beads to which anti-CD3 and anti-CD28 are attached (3x28 beads) to contact the Tcells. In one embodiment the cells (for example, 104 to 109 Tcells) and beads (for example, DYNABEADS® M‑450 CD3 / CD28 T paramagnetic beads at a ratio of 1:1) are combined in a buffer, for examplePBS (without divalent cations such as, calciumandmagnesium). Again, thoseof ordinary skill in the art can readily appreciate any cell concentrationmaybe used. For example, the target cellmay be very rare in the sample and comprise only 0.01% of the sample or the entire sample (i.e., 100%)may comprise the target cell of interest. Accordingly, any cell number is within the context of the present invention. In certain embodiments, it may be desirable to significantly decrease the volume in which particles and cells are mixed together (i.e., increase the concentration of cells), to ensure maximum contact of cells and particles. For example, in one embodiment, a concentration of about 2 billion cells / ml is used. In another embodiment, greater than 100million cells / ml is used. In a further embodiment, a concentration of cells of 10, 15, 20, 25, 30, 35, 40, 45, or 50million cells / ml is used. In yet another embodiment, a concentration of cells from75, 80, 85,90, 95,or100millioncells / ml isused. In further embodiments, concentrationsof125or150millioncells / ml canbeused. Using high concentrations can result in increased cell yield, cell activation, and cell expansion. Further, use of high cell concentrations allows more efficient capture of cells that may weakly express target antigens of interest, such as CD28- negative T cells. Such populations of cells may have therapeutic value and would be desirable to obtain in certain embodiments. For example, using high concentration of cells allowsmore efficient selection of CD8+ Tcells that normally have weaker CD28 expression.

[0346] In one embodiment of the present invention, the mixture may be cultured for several hours (about 3 hours) to about 14 days or any hourly integer value in between. In another embodiment, themixturemay be cultured for 21 days. In one embodiment of the invention the beads and the T cells are cultured together for about eight days. In another embodiment, the beads and Tcells are cultured together for 2‑3 days. Several cycles of stimulation may also be desired such that culture time of Tcells can be 60 days or more. Conditions appropriate for Tcell culture include an appropriate media (e.g., Minimal Essential Media or RPMI Media 1640 or, X-vivo 15, (Lonza)) that may contain factors necessary for proliferation and viability, including serum (e.g., fetal bovineor humanserum), interleukin‑2 (IL‑2), insulin, IFN-γ, IL‑4, IL‑7, GM-CSF, IL‑10, IL‑12, IL‑15, TGFβ, and TNF-α or any other additives for the growth of cells known to the skilled artisan. Other additives for the growth of cells include, but are not limited to, surfactant, plasmanate, and reducing agents such as N-acetyl-cysteine and 2-mercaptoethanol.Media can includeRPMI 1640, AIM-V,DMEM,MEM,α-MEM, F‑12, X-Vivo 15, andX-Vivo20,Optimizer,withaddedaminoacids, sodiumpyruvate, andvitamins, either serum-freeor supplementedwith anappropriateamount of serum (or plasma)or adefinedset of hormones, and / or anamount of cytokine(s) sufficient for the growth and expansion of Tcells. Antibiotics, e.g., penicillin and streptomycin, are included only in experimental cultures, not in cultures of cells that are to be infused into a subject. The target cells are maintained under conditions necessary to support growth, for example, an appropriate temperature (e.g., 37° C) and atmosphere (e.g., air plus 5% CO2).

[0347] T cells that have been exposed to varied stimulation times may exhibit different characteristics. For example, typical blood or apheresed peripheral blood mononuclear cell products have a helper Tcell population (TH, CD4+) that is greater than the cytotoxic or suppressor Tcell population (TC, CD8+).Ex vivo expansion of Tcells by stimulating CD3 and CD28 receptors produces a population of Tcells that prior to about days 8‑9 consists predominately of TH cells, while after about days 8‑9, the population of Tcells comprises an increasingly greater population of TC cells. Accordingly, depending on the purpose of treatment, infusing a subject with a T cell population comprising predominately of TH cells may be advantageous. Similarly, if an antigen-specific subset of TC cells has been isolated it may be beneficial to expand this 51 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 subset to a greater degree.

[0348] Further, in addition to CD4 and CD8 markers, other phenotypic markers vary significantly, but in large part, reproducibly during the course of the cell expansion process. Thus, such reproducibility enables the ability to tailor an activated T cell product for specific purposes. THERAPEUTIC APPLICATION

[0349] The present invention encompasses a cell (e.g., T cell) modified to express a plurality of types of KIR-CARs, wherein each KIR-CAR combines an antigen recognition domain of a specific antibody with a component of a KIR.

[0350] In one embodiment, the KIR-CARs of the invention comprise an activating KIR which delivers its signal through an interaction with the immunotyrosine-based activation motif (ITAM) containing membrane protein, DAP12 that is mediated by residues within the transmembrane domains of these proteins.

[0351] In one embodiment, the KIR-CARs of the invention comprise an inhibitory KIR which delivers its signal through one or more immunotyrosine-based inhibitory motifs (ITIMs) that interact directly or indirectly with cytoplasmic signaling proteins such as SHP‑1, SHP‑2 and Vav family of proteins. KIRs bearing cytoplasmic domains that contain (ITIMs) abrogate the activating signal leading to inhibition of NK cytolytic and cytokine producing activity. In some instances, the modifiedTcell expressing aKIR-CARof the invention canelicit aKIR-CAR-mediatedT-cell response. In oneembodiment, the dependence of the binding tomore than one type of antigen allows themodifiedTcell to exhibit a heightened specificity to elicit a response upon binding of a tumor cell rather than a normal bystander cell.

[0352] The invention provides the use of a plurality of types of KIR‑CARs to redirect the specificity of a primary Tcell to a tumor antigen. Thus, the present invention also provides amethod for stimulating aTcell-mediated immune response to a target cell population or tissue in a mammal comprising the step of administering to the mammal a Tcell that expresses a plurality of types of KIR-CARs, wherein each type of KIR-CARcomprises a bindingmoiety that specifically interactswith a predetermined target. In one embodiment, the cell comprises a first KIR-CARcomprising an activatingKIR (actKIR-CAR), and a second KIR-CAR comprising an inhibitory KIR (inhKIR-CAR).

[0353] Withoutwishing to be bound by any particular theory, the anti-tumor immunity response elicited by theKIR-CAR- modified Tcells may be an active or a passive immune response. In addition, the KIR-CAR mediated immune response may be part of an adoptive immunotherapy approach in which KIR-CAR-modified T cells induce an immune response specific to the antigen binding domain in the KIR-CAR.

[0354] Cancers that may be treated include tumors that are not vascularized, or not yet substantially vascularized, as well as vascularized tumors. The cancers may comprise non-solid tumors (such as hematological tumors, for example, leukemias and lymphomas) or may comprise solid tumors. Types of cancers to be treated with the CARs of the invention include, but are not limited to, carcinoma, blastoma, and sarcoma, and certain leukemia or lymphoidmalignancies, benign and malignant tumors, and malignancies e.g., sarcomas, carcinomas, and melanomas. Adult tumors / cancers and pediatric tumors / cancers are also included.

[0355] Hematologic cancers are cancers of the blood or bone marrow. Examples of hematological (or hematogenous) cancers include leukemias, including acute leukemias (such as acute lymphocytic leukemia, acute myelocytic leukemia, acute myelogenous leukemia and myeloblastic, promyelocytic, myelomonocytic, monocytic and erythroleukemia), chronic leukemias (such as chronic myelocytic (granulocytic) leukemia, chronic myelogenous leukemia, and chronic lymphocytic leukemia), polycythemia vera, lymphoma, Hodgkin’s disease, non-Hodgkin’s lymphoma (indolent and high grade forms), multiple myeloma, Waldenstrom’s macroglobulinemia, heavy chain disease, myelodysplastic syndrome, hairy cell leukemia and myelodysplasia.

[0356] Solid tumors are abnormalmasses of tissue that usually do not contain cysts or liquid areas. Solid tumors can be benign or malignant. Different types of solid tumors are named for the type of cells that form them (such as sarcomas, carcinomas, and lymphomas). Examples of solid tumors, such as sarcomas and carcinomas, include fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma, and other sarcomas, synovioma, mesothelioma, Ewing’s tumor, leiomyosarcoma, rhabdomyosarcoma, colon carcinoma, lymphoid malignancy, pancreatic cancer, breast cancer, lung cancers, ovarian cancer, prostate cancer, hepatocellular carcinoma, squamouscell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, medullary thyroid carcinoma, papillary thyroid carcinoma, pheochromocyto- mas sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinomas, medullary carcinoma, bronchogenic carcinoma, renal cell carcinoma, hepatoma, bile duct carcinoma, choriocarcinoma, Wilms’ tumor, cervical cancer, testicular tumor, seminoma, bladder carcinoma, melanoma, and CNS tumors (such as a glioma (such as brainstem glioma and mixed gliomas), glioblastoma (also known as glioblastoma multiforme) astrocytoma, CNS lymphoma, germinoma,medulloblastoma,Schwannomacraniopharyogioma, ependymoma,pinealoma,hemangioblastoma, acous- tic neuroma, oligodendroglioma, menangioma, neuroblastoma, retinoblastoma and brain metastases).

[0357] In one embodiment, the antigen bindmoiety portion of the KIR-CARTcells of the invention is designed to treat a particular cancer. In one embodiment, the KIR-CAR Tcells of the invention are modified to express a first actKIR-CAR targeting a first antigen and a second inhKIR-CAR targeting a second antigen, where the first antigen is expressed on a 52 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55 particular tumor or cancer and the second antigen is not expressed on a particular tumor or cancer. In this manner, conditional activation of Tcells is generated by engagement of actKIR-CAR (or standard TCR-zetaCARbearing a scFv to an antigen on the malignant cell of interest) and the inhKIR-CAR bearing for example a scFv directed against an antigen that ispresentonnormal, butnotmalignant tissueprovides inhibitionof theactivatingsignal from theactKIR-CARwhen the KIR-CARTcell encounters normal cells. Examples of antigens that serve as useful targets for inhibitory CARs include the ephrin receptors (Pasquale, 2010,NatRevCancer 10(3):165‑80)andclaudins (Singhet al., 2010, JOncol, 2010:541957), which are expressed by epithelial cells from normal tissues, but often selectively lost by cancers (e.g. EPHA7).

[0358] With respect toex vivo immunization, at least oneof the followingoccurs in vitroprior to administering the cell into a mammal: i) expansion of the cells, ii) introducing a nucleic acid encoding a KIR-CAR to the cells, and / or iii) cryopreservation of the cells.

[0359] Ex vivo procedures are well known in the art and are discussedmore fully below. Briefly, cells are isolated from a mammal (preferably a human) and geneticallymodified (i.e., transduced or transfected in vitro) with a vector expressing a KIR-CAR disclosed herein. The KIR-CAR-modified cell can be administered to a mammalian recipient to provide a therapeutic benefit. The mammalian recipient may be a human and the KIR-CAR-modified cell can be autologous with respect to the recipient. Alternatively, the cells can be allogeneic, syngeneic or xenogeneic with respect to the recipient.

[0360] The procedure for ex vivo expansion of hematopoietic stem and progenitor cells is described in U.S. Pat. No. 5,199,942, incorporated herein by reference, can be applied to the cells of the present invention. Other suitable methods are known in the art, therefore the present invention is not limited to any particularmethod ofex vivoexpansion of the cells. Briefly, ex vivo culture and expansion of Tcells comprises: (1) collecting CD34+ hematopoietic stem and progenitor cells fromamammal fromperipheral bloodharvest or bonemarrowexplants; and (2) expanding suchcellsexvivo. In addition to the cellular growth factors described in U.S. Pat. No. 5,199,942, other factors such as flt3-L, IL‑1, IL‑3 and c-kit ligand, can be used for culturing and expansion of the cells.

[0361] In addition to using a cell-based vaccine in terms of ex vivo immunization, the present invention also provides compositions andmethods for in vivo immunization to elicit an immune response directed against an antigen in a patient.

[0362] Generally, the cells activated and expanded as described herein may be utilized in the treatment and prevention of diseases that arise in individuals who are immunocompromised. In particular, the KIR-CAR-modified T cells of the invention areused in the treatment of cancer. In certain embodiments, the cells of the inventionare used in the treatment of patients at risk for developing cancer. Thus, the present invention provides methods for the treatment or prevention of cancer comprising administering to a subject in need thereof, a therapeutically effective amount of the KIR-CAR-modified T cells of the invention.

[0363] TheKIR-CAR-modified Tcells of the present inventionmay be administered either alone, or as a pharmaceutical composition in combinationwith diluents and / or with other components such as IL‑2 or other cytokines or cell populations. Briefly, pharmaceutical compositionsof thepresent inventionmaycomprisea target cell populationasdescribedherein, in combination with one or more pharmaceutically or physiologically acceptable carriers, diluents or excipients. Such compositionsmaycomprisebuffers suchasneutral bufferedsaline, phosphatebufferedsalineand the like; carbohydrates such as glucose, mannose, sucrose or dextrans, mannitol; proteins; polypeptides or amino acids such as glycine; antioxidants; chelating agents such as EDTA or glutathione; adjuvants (e.g., aluminum hydroxide); and preservatives. Compositions of the present invention are preferably formulated for intravenous administration.

[0364] Pharmaceutical compositions of the present invention may be administered in a manner appropriate to the disease tobe treated (orprevented). Thequantity and frequencyofadministrationwill bedeterminedbysuch factorsas the condition of the patient, and the type and severity of the patient’s disease, although appropriate dosages may be determined by clinical trials.

[0365] When "an immunologically effective amount", "an anti-tumor effective amount", "an tumor-inhibiting effective amount", or "therapeutic amount" is indicated, the precise amount of the compositions of the present invention to be administered can be determined by a physician with consideration of individual differences in age, weight, tumor size, extent of infection or metastasis, and condition of the patient (subject). It can generally be stated that a pharmaceutical composition comprising the Tcells described herein may be administered at a dosage of 104 to 109 cells / kg body weight, preferably 105 to 106 cells / kgbodyweight, includingall integer valueswithin those ranges. Tcell compositionsmayalsobe administered multiple times at these dosages. The cells can be administered by using infusion techniques that are commonly known in immunotherapy (see, e.g., Rosenberg et al., New Eng. J. of Med. 319:1676, 1988). The optimal dosage and treatment regime for a particular patient can readily be determined by one skilled in the art of medicine by monitoring the patient for signs of disease and adjusting the treatment accordingly.

[0366] In certain embodiments, it may be desired to draw blood (or have an apheresis performed), activate and genetically modify the Tcells therefrom according to the present invention, and reinfuse the patient with these activated and expanded genetically modified T cells. This process can be carried out multiple times every few weeks. In certain embodiments, T cells can be activated from blood draws of from 10cc to 400cc. In certain embodiments, T cells are activated from blood draws of 20cc, 30cc, 40cc, 50cc, 60cc, 70cc, 80cc, 90cc, or 100cc. Not to be bound by theory, using this multiple blood draw / multiple reinfusion protocol may serve to select out certain populations of T cells. 53 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0367] The administration of the subject compositions may be carried out in any convenient manner, including by aerosol inhalation, injection, ingestion, transfusion, implantation or transplantation. The compositions described herein may be administered to a patient subcutaneously, intradermally, intratumorally, intranodally, intramedullary, intramuscu- larly, by intravenous (i.v.) injection, or intraperitoneally. In oneembodiment, theTcell compositions of thepresent invention are administered to a patient by intradermal or subcutaneous injection. In another embodiment, the Tcell compositions of thepresent inventionarepreferablyadministeredby i.v. injection.ThecompositionsofTcellsmaybe injecteddirectly intoa tumor, lymph node, or site of infection.

[0368] In certain embodiments of the present invention, cells activated and expanded using the methods described herein, or othermethods known in the art where Tcells are expanded to therapeutic levels, are administered to a patient in conjunction with (e.g., before, simultaneously or following) any number of relevant treatment modalities, including but not limited to treatmentwith agents suchasantiviral therapy, cidofovir and interleukin‑2,Cytarabine (also knownasARA-C) or natalizumab treatment forMSpatients or efalizumab treatment for psoriasis patients or other treatments for PMLpatients. In further embodiments, the T cells of the invention may be used in combination with chemotherapy, radiation, immunosuppressive agents, such as cyclosporin, azathioprine, methotrexate, mycophenolate, and FK506, antibodies, or other immunoablative agents such asCAMPATH, anti-CD3antibodies or other antibody therapies, cytoxin, fludaribine, cyclosporin, FK506, rapamycin, mycophenolic acid, steroids, FR901228, cytokines, and irradiation. These drugs inhibit either the calcium dependent phosphatase calcineurin (cyclosporine and FK506) or inhibit the p70S6 kinase that is important for growth factor induced signaling (rapamycin) (Liu et al., Cell 66:807‑815, 1991; Henderson et al., Immun. 73:316‑321, 1991;Bierer et al., Curr.Opin. Immun. 5:763‑773, 1993). In a further embodiment, the cell compositionsof the present invention are administered to a patient in conjunctionwith (e.g., before, simultaneously or following) bonemarrow transplantation, Tcell ablative therapy using either chemotherapy agents such as, fludarabine, external-beam radiation therapy (XRT), cyclophosphamide, or antibodies such as OKT3 or CAMPATH. In another embodiment, the cell compositions of the present invention are administered following B-cell ablative therapy such as agents that react with CD20, e.g., Rituxan. For example, in one embodiment, subjects may undergo standard treatment with high dose chemotherapy followed by peripheral blood stem cell transplantation. In certain embodiments, following the transplant, subjects receive an infusion of the expanded immune cells of the present invention. In an additional embodiment, expanded cells are administered before or following surgery.

[0369] The dosage of the above treatments to be administered to a patient will vary with the precise nature of the condition being treated and the recipient of the treatment. The scaling of dosages for human administration can be performedaccording to art-accepted practices. Strategies forCARTcell dosing and scheduling havebeendiscussed (Ertl et al, 2011, Cancer Res, 71:3175‑81; Junghans, 2010, Journal of Translational Medicine, 8:55). EXPERIMENTAL EXAMPLES

[0370] The invention is further described in detail by reference to the followingexperimental examples. Theseexamples are provided for purposes of illustration only, and are not intended to be limiting unless otherwise specified. Thus, the invention should in no way be construed as being limited to the following examples, but rather, should be construed to encompass any and all variations which become evident as a result of the teaching provided herein.

[0371] Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, make and utilize the compounds of the present invention and practice the claimed methods. The following working examples therefore, specifically point out the preferred embodiments of the present invention, and are not to be construed as limiting in any way the remainder of the disclosure. Example 1: Chimeric NK receptors

[0372] The results presented herein demonstrate an alternative approach to constructing CARs for Tcells that can be morefinely regulatedcomparedwith currentCARdesigns.Experimentsweredesigned todevelopanovel, regulatedCAR system that comprises at least two or three chimeric fusion proteins. The primary T cell activating signal and inhibitory signals are based upon naturally occurring activating and inhibitory receptors of NK cells known as killer cell immuno- globulin-like receptors (KIRs).

[0373] KIRs exist as both activating and inhibitory forms that depend upon the intracellular domain of the receptor. Activating KIRs deliver their signals through an interaction with the immunotyrosine-based activation motif (ITAM) containing membrane protein, DAP12 that is recruited by residues within the transmembrane domains of these proteins. Inhibitory KIRs bear a cytoplasmic domain that contains immunotyrosine-based inhibitory motifs (ITIMs), which abrogate theactivating signal leading to inhibitionofNKcytolytic andcytokineproducingactivity.Similar toTCRs,KIRsbelong to the immunoglobulin family of protein receptors, andmany bind to invariant MHCandMHC-like ligands.Without wishing to be boundbyanyparticular theory, it is believed that these interactions are utilized to naturally distinguish normal cells (usually expressing high density MHC class I) from malignant or virally infected cells (often with low or missing MHC class I). 54 EP 4 696 718 A2 5 10 15 20 25 30 35 40 45 50 55

[0374] KIR-like chimeric antigen receptors (KIR-CARs) have been constructed which fuse an scfv to a target antigen of interest with activating and inhibitory KIRs as shown in Figure 1. Conditional activation of T cells is generated by engagement of an activating KIR-CAR (actKIR-CAR) or standard TCR-zeta CAR bearing an scfv to an antigen on the malignant cell of interest.An inhibitoryCAR(inhCAR)bearinganscfvdirectedagainstanantigen that ispresentonnormal, but notmalignant tissuewould provide dampening of the activatingCARprimary signal when the Tcell encounters normal cells. Examples of antigens that serve as useful targets for inhibitory CARs include the ephrin receptors (Pasquale, 2010, Nat RevCancer 10(3):165‑80) and claudins (Singh et al., 2010, JOncol, 2010:541957), which are expressed by epithelial cells from normal tissues, but often selectively lost by cancers (e.g. EPHA7). Example 2: Activating KIR-CAR Construction and Activity

[0375] Experiments were designed to construct activating KIR-CARs based upon fusion of the anti-CD19 or anti- mesothelin scFv (SS‑1) that were previously incorporated into CARs based upon the TCR-zeta cytoplasmic domain that are currently in clinical trials. The human KIR2DS2 activating KIR receptor was chosen as the initial base receptor for the actKIR-CAR. In order to deliver activating signals, the actKIR-CARs required coexpression of DAP12, which is not expressed normally in Tcells. Therefore, a lentiviral vector that expresses both the actKIR-CARwith humanDAP12 using a "bicistronic" gene cassette based upon the 2A ribosomal skip pept...

Claims

1. A pharmaceutical composition comprising a T cell that comprises a killer cell immunoglobulin-like receptor chimeric antigen receptor (KIR-CAR) and optionally a pharmaceutically acceptable carrier, diluent, or excipient, wherein the KIR-CAR comprises: an extra-cellular antigen binding domain; a transmembrane domain, and a cytoplasmic domain, wherein: (i) the KIR-CAR is an activating KIR-CAR (actKIR-CAR) comprising an actKIR transmembrane domain; or (ii) the KIR-CAR is an inhibitory KIR-CAR (inhKIR-CAR) comprising an inhKIR cytoplasmic domain.

2. A kit comprising a T cell that comprises a KIR-CAR, or a nucleic acid encoding the KIR-CAR, wherein the KIR-CAR comprises: an extra-cellular antigen binding domain; a transmembrane domain, and a cytoplasmic domain, wherein: (i) the KIR-CAR is an actKIR-CAR comprising actKIR transmembrane domain; or (ii) the KIR-CAR is an inhKIR-CAR comprising an inhKIR cytoplasmic domain.

3. The pharmaceutical composition of claim 1 or kit of claim 2, wherein: (a) said actKIR transmembrane domain can interact with the transmembrane domain of DAP12; and / or (b) said actKIR transmembrane domain comprises a positively charged moiety.

4. The pharmaceutical composition or kit of claim 3, wherein: (a) said cytoplasmic domain is a KIR-cytoplasmic domain; (b) said antigen binding domain binds an antigen present on a cancer cell; and / or (c) said antigen binding domain binds either: (i) folate receptor (FRa), mesothelin, EGFRvIII, IL-13Ra, CD123, CD19, CD33, BCMA, GD2, CLL-1, CA-IX, MUC1, HER, and any combination thereof; or (ii) CD19, mesothelin, CD123, or BCMA.

5. The pharmaceutical composition or kit of claim 3 or 4, wherein said actKIR-CAR: (i) can interact with and promote signaling from an ITAM-containing polypeptide or adaptor molecule; (ii) can interact with and promote signaling from a DAP12 polypeptide; (iii) either comprises a KIR D, D1, or D2 domain, or does not comprise a KIR D domain; (iv) comprises a KIR2DS2 transmembrane domain; and / or (v) comprises a KIR2DS2 cytoplasmic domain.

6. The pharmaceutical composition or kit of any of claims 3-5, wherein: (I) the nucleic acid comprises: (a) a nucleic acid sequence that encodes the actKIR-CAR, and (b) a nucleic acid sequence that encodes an intracellular signaling domain comprising an ITAM motif, optionally wherein: (i) the sequence that encodes the intracellular signalling domain encodes a DAP12 polypeptide comprising a DAP12 intracellular signaling domain; and / or (ii) each of (a) and (b) are present on the same nucleic acid molecule or wherein each of (a) and (b) are present on different nucleic acid molecules; and / or (II) the T cell further comprises a DAP12 polypeptide comprising a DAP12 intracellular signaling domain.

7. The pharmaceutical composition of claim 1 or kit of claim 2, wherein: (a) said inhKIR cytoplasmic domain comprises an ITIM-containing cytoplasmic domain; (b) said inhKIR-CAR comprises an inhKIR transmembrane domain or a transmembrane domain other than a KIR transmembrane domain; (c) the antigen binding domain of the inhKIR-CAR binds an antigen not present on a cancer cell or an antigen that is more highly expressed on a non-cancer cell relative to a cancer cell of the same cell type as the non-cancer cell; and / or (d) the antigen binding domain of the inhKIR-CAR binds to desmoglein1 / 3, an ephrin receptor, or a claudin.

8. The pharmaceutical composition or kit of claim 7, wherein: (i) the inhKIR cytoplasmic domain comprises a KIR2DL or KIR3DL cytoplasmic domain; (ii) the transmembrane domain other than a KIR transmembrane domain comprises a transmembrane domain from PD-1, CTLA4 or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors; (iii) the inhKIR-CAR comprises a cytoplasmic domain from an inhibitory receptor other than a KIR, such as from PD-1, or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors; (iv) the inhKIR-CAR comprises a transmembrane and cytoplasmic domain from an inhibitory receptor other than a KIR, such as transmembrane and cytoplasmic domain, independently, from PD-1, or ITIM-containing receptors from ILT (CD85), Siglec, LMIR (CD300) and / or SLAM gene families of receptors; and / or (v) the inhKIR-CAR either (a) comprises a KIR D domain, a KIR D0 domain, a KIR D1 domain, and / or a KIR D2 domain; or (b) does not comprise a KIR D domain.

9. The pharmaceutical composition or kit of any of the preceding claims, wherein the extra-cellular antigen binding domain is derived from an antibody molecule or which comprises a non-antibody scaffold that has the ability to bind to a target antigen on a cell, optionally wherein said extra-cellular antigen binding domain from an antibody molecule: (i) comprises a scFv; (ii) comprises a single VH domain; (iii) comprises a nanobody; (iv) comprises a camelid VHH domain; or (v) comprises a single domain antibody.

10. The pharmaceutical composition or kit of any of the preceding claims, wherein the KIR-CAR comprises an extracellular hinge domain that: (i) is other than a KIR hinge domain; (ii) is derived from a natural molecule; (iii) is derived from a natural molecule other than a KIR; (iv) comprises a non-naturally occurring polypeptide sequence; (v) comprises the extracellular hinge from human CD8-alpha or a human immunoglobulin hinge; (vi) comprises a synthetic extracellular hinge; (vii) is of less than 50, 20, or 10 amino acids in length; or (viii) has fewer amino acids than a KIR2DS2 hinge domain.

11. The pharmaceutical composition or kit of any of the preceding claims, wherein: (a) the nucleic acid further comprises a sequence that encodes a TCAR; and / or (b) the T cell comprises: (i) the actKIR-CAR as defined in any of claims 3-6, wherein the actKIR-CAR targets a first antigen that is expressed by the cancer; and (ii) the inhKIR-CAR as defined in claim 7 or 8, wherein the inhKIR-CAR targets a second antigen that is not expressed by the cancer or an antigen that is more highly expressed on a non-cancer cell, relative to a cancer cell of the same type as the non-cancer cell.

12. The pharmaceutical composition or kit of any of the preceding claims, wherein said T cell is: (i) an autologous T cell; or (ii) an allogeneic T cell.

13. The pharmaceutical composition or kit of any of the preceding claims, for use in treating a subject having an autoimmune disease, optionally wherein the autoimmune disease is chosen from Addision's disease, alopecia areata, ankylosing spondylitis, autoimmune hepatitis, autoimmune parotitis, Crohn's disease, diabetes (Type I), dystrophic epidermolysis bullosa, epididymitis, glomerulonephritis, Graves' disease, Guillain-Barr syndrome, Hashimoto's disease, hemolytic anemia, systemic lupus erythematosus, multiple sclerosis, myasthenia gravis, pemphigus vulgaris, psoriasis, rheumatic fever, rheumatoid arthritis, sarcoidosis, scleroderma, Sjogren's syndrome, spondyloarthropathies, thyroiditis, vasculitis, vitiligo, myxedema, pernicious anemia, or ulcerative colitis.

14. The pharmaceutical composition or kit of any of the preceding claims, for use in treating a subject that previously received a treatment, optionally wherein the subject previously received an antiviral therapy, cidofovir, interleukin-2, cytarabine, or natalizumab for MS, efalizumab for psoriasis, chemotherapy, radiation, immunosuppressive agent, cyclosporin, azathioprine, methotrexate, mycophenolate, FK506, antibody (e.g., an anti-CD3 antibody), cytoxan, fludarabine, cyclosporin, FK506, rapamycin, mycophenolic acid, steroid, FR901228, cytokine, irradiation, bone marrow transplantation, T cell ablative therapy (e.g., using either fludarabine, external-beam radiation therapy (XRT), cyclophosphamide, or antibodies such as OKT3 or CAMPATH), B-cell ablative therapy (e.g., an agent that reacts with CD20, e.g., rituxan), or surgery.

15. The pharmaceutical composition or kit for use of claim 13 or 14, wherein: (a) the subject is a human, optionally wherein said human has a side effect and said side effect is treated with an anti-cytokine agent, further optionally wherein said anti-cytokine agent is a tumor necrosis factor antagonist, an IL-6 antagonist, an IL-6 receptor antagonist, or a corticosteroid; and / or (ii) the subject is a mammal that has a cancer.

Images