Method, system, and computer-accessible medium for artificial intelligence-driven chimeric antigen receptors

Abstract

Disclosed herein are methods, systems, computer-implemented methods, and computer-accessible media for designing, modifying and / or redesigning peptide-centric chimeric antigen receptors (PC-CARs) against novel peptide antigens.

Description

Attorney Docket No. 11820-039W01YAROl-HPRO METHOD, SYSTEM, AND COMPUTER-ACCESSIBLE MEDIUM FOR ARTIFICIAL INTELLIGENCE-DRIVEN CHIMERIC ANTIGEN RECEPTORS Cross-Reference to Related Applications

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 720,043, filed on November 13, 2024, the content of which is incorporated by reference herein in its entirety.Field of the Disclosure

[0002] The present disclosure is related to methods, systems, computer-implemented methods, and computer- accessible media for designing, modifying and / or redesigning Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs) (referred to herein as AIR-CARs) that can bind to specific tumor antigens.Background

[0003] Peptide-centric chimeric antigen receptors (PC-CARs) development is restricted by throughput of the receptors engineering process, a multi-year labor intensive screening process that is essential to generate safe and functional therapies.

[0004] Accordingly, there is a need to address and / or at least partially overcome at least some of the deficiencies described herein.Summary

[0005] PC-CAR T cells overcome several obstacles in targeting immunologically cold tumors using CAR T cells. These constructs can precisely target non-immunogenic cancer drivers, selectively eliminating aggressive tumors; these therapies are entering a clinical trial in neuroblastoma in 2024 (See Figure 1A). Numerous PC-CAR targets have been identified in other cancer types, however, PC-CAR development remains restricted by throughput of the receptors engineering process, a multi-year labor intensive screening process that is essential to generate safe and functional therapies. The targets of PC-CARs are tumor-specific peptide presented by human leukocyte antigen (HLA), a challenging target class due to the limited surface area of the peptide as presented by HLA (1-2%). This is further compounded by a complex array of potential cross-reactivities with homologous peptides presented by HLA in normal tissue. Though applications of the generative Al to protein design are still in their nascency, good results have been achieved in the year since their release, demonstrating the ability to robustly generate de novo binders with superior binding than multiple clinical molecules. However, there are still some drawbacks associated with the prior systems.Attorney Docket No. 11820-039W01YAROl-HPRO

[0006] For example, generative Al has transformed fields like natural language processing and is now reshaping protein design, enabling in-silico generation of binders with superior affinity to antibodies generated by conventional methods. Parallel advances in protein structure predictions with Alphafold2 have achieved high-fidelity predictions as compared to experimental validation, revolutionizing the field of structural biology. The field of antibody engineering, including in CAR design has long been dominated by selective panning strategies rather than rational design, currently relying on the ability to identify rare binders with desired properties from large libraries, a labor-intensive and costly experimental screening process.

[0007] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can facilitate the use of peptide-centric (PC)-CAR receptors capable of selectively targeting 1% of a protein complex surface that is specific to tumors. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure provide generative Al, coupled with structural modeling, which can yield binders with superior physicochemical characteristics to those identified from random antibody libraries. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can develop a comprehensive end-to-end in-silico CAR design platform featuring high-throughput binder generation and in silico cross-reactivity screening process, resulting in binder models with superior binding properties than those of current clinical PC-CARs. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure have generated a pool of 100,000 unique binders to the previously described tumor antigen PIIOX2B and a pipeline for prioritizing the most promising candidates, and further applying the same pipeline to additional tumor targets such as ORF2 presented on HLA-A*24:02. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can propose to select a number of these binders (e.g., 300) and generate a library of AI-engineered Receptor (AIR)-CAR constructs to test using high-throughput screening technologies. Suitable AIR-CAR clones can be subsequently isolated by exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure for further in vitro binding and cross-reactivity characterization and validated for anti-tumor activity. In this way, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can revolutionize the development of CARs and PC-CAR T cells for cancer and other diseases.Attorney Docket No. 11820-039W01YAROl-HPRO

[0008] In some exemplary embodiments of the present disclosure, the exemplary systems, methods, and computer accessible medium can be provided for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs), which can include generating, with a deep-learning-empowered structural modeling tool, an optimal binding interface scaffold conforming to the unique structure of a peptide and generating, with a deep-learning -based protein sequencer, a set of binder sequences optimized for one or more proteins that fold into the optimal binding interface scaffold and generate strong interactions with a defined epitope.

[0009] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can also employ a custom peptide / HLA bond scoring system to evaluate the quality and selectivity of the interface, by quantifying possible bonds between the binder and the target peptide and calculating a peptide-contact score and an HLA-contact score for each AIR-CAR binder, and prioritizing binders that have a peptide-contact score at or above a predetermined cutoff (c.g., > 25) while maintaining limited contact with HLA.

[0010] In some implementations, a computer-implemented method for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs) is provided. The computer-implemented method can include: generating, by at least one processor and with a deep-learning-empowered structural modeler, a three-dimensional (3D) structure of a CAR target; generating, by the at least one processor, an optimal binding interface scaffold conforming to a unique peptide structure; generating, by the at least one processor and using a deep-learning -based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predicting, by the at least one processor and using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets; evaluating, by the at least one processor and using a deep-learning-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; and selecting, by the at least one processor, a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

[0011] In some implementations, the 3D structure of the CAR target includes at least one of an IILA molecule, an IILA-presented peptide, and a beta-2-microglobulin (beta-2-m).

[0012] In some implementations, evaluating the selectivity of the binder sequences includes: performing, by the at least one processor and using one or more decoy peptides, aAttorney Docket No. 11820-039W01YAROl-HPRO cross-reactivity test; and eliminating one or more binders from the set of binder sequences based on results of the cross-reactivity test.

[0013] In some implementations, each interaction score (e.g., Predicted Aligned Error (PAE) score) is within a predefined distance threshold.

[0014] In some implementations, evaluating the selectivity of the binder sequences includes: determining, by the at least one processor, a specificity score for each binder sequence; and selecting, by the at least one processor, the subset of the binder sequences based on the determined specificity scores.

[0015] In some implementations, the specificity score for each binder sequence is determined based on possible bonds (e.g., interactions between the CAR target and the binder sequence) that are between 0-6 Angstroms (A).

[0016] In some implementations, the CAR target includes a predicted computer structure model or an experimentally solved structure model.

[0017] In some implementations, the method of generating the AIR-CAR is independent of HLA allele type.

[0018] In some implementations, the one or more decoy peptides have binding residues within a predefined distance threshold of at least one binder sequence.

[0019] In some implementations, the subset of binder sequences is used for in vitro testing.

[0020] In some implementations, a method of treating a cancer in a subject in need thereof is provided, comprising administering to the subject a substance comprising the AIRCAR generated according to any one of claims described herein.

[0021] In some implementations, a method of treating a cancer in a subject in need thereof is provided, comprising administering to the subject an AIR-CAR immune cell comprising at least one binding sequence generated according to any one of the claims described herein.

[0022] In some implementations, the techniques described herein relate to a method, wherein immune cell is a T cell, B cell, natural killer (NK) cell, NK T cell, or macrophage.

[0023] In some implementations, the cancer includes neuroblastoma / glioblastoma, colon adenocarcinoma, breast cancer, ovarian cancer, skin cutaneous melanoma, esophageal cancer, lung squamous cell carcinoma, lung adenocarcinoma, pancreatic adenocarcinoma, or cervical squamous cell carcinoma.

[0024] In some implementations, the techniques described herein relate to a system for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen ReceptorsAttorney Docket No. 11820-039W01YAROl-HPRO (CARs), including: at least one processor; and a memory having instructions thereon, wherein the instructions when executed by the at least one processor, cause the at least one processor to: generate a three-dimensional (3D) structure of a CAR target; generate an optimal binding interface scaffold conforming to a unique peptide structure; generate, using a deep-leaming-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predict, using an artificial intelligence model, a plurality of target crossreactivities to generate a panel of putative cross-reactive targets; evaluate, using a deeplearning-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico: and select a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

[0025] In some implementations, the techniques described herein relate to a non-transitory computer accessible medium including a memory having instructions stored thereon to cause at least one processor to: generate a three-dimensional (3D) structure of a CAR target; generate an optimal binding interface scaffold conforming to a unique peptide structure; generate, using a deep-leaming-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predict, using an artificial intelligence model, a plurality of target cross -reactivities to generate a panel of putative cross-reactive targets; evaluate, using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; and select a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

[0026] These and other objects, features and advantages of the exemplary embodiments of the present disclosure will become apparent upon reading the following detailed description of the exemplary embodiments of the present disclosure, when taken in conjunction with the appended paragraphs.Brief Description of Drawings

[0027] Further objects, features and advantages of the present disclosure will become apparent from the following detailed description taken in conjunction with the accompanying Figures showing illustrative embodiments of the present disclosure, in which:Attorney Docket No. 11820-039W01YAROl-HPRO

[0028] Figures 1A and IB are a set of illustrations providing an exemplary AIR-CAR workflow to recapitulate PC-CAR development process according to an exemplary embodiment of the present disclosure;

[0029] Figure 1C is an exemplary illustration of AIR-CAR binder to PHOX2B pMHC according to an exemplary embodiment of the present disclosure;

[0030] Figure ID is a model of an identified binder sequence, AIR637, bound to the PHOX2B-A2402. Amino acid residues shown in blue are residues in the AIR637 binder sequence that make contact with the PHOX2B peptide. Amino acid residues in magenta are residues in the HLA -presented peptide, PHOX2B, that make contact with the binder sequence. Amino acid residues shown in gray are amino acids in the HLA molecule that make contact with the AIR637 binder sequence. Yellow dashed lines represent hydrogen bonds, green dashed lines represent cation-7i interactions and i-n stacking. Red dashed lines represent salt bridges, orange dashed lines represent hydrophobic interactions;

[0031] Figure IE is an example computing device;

[0032] Figure 2 is an example system in accordance with certain embodiments described herein;

[0033] Figure 3 is a flowchart diagram of an example method in accordance with certain embodiments described herein;

[0034] Figure 4 is an example AIR-CAR pipeline overview in accordance with certain embodiments described herein;

[0035] Figure 5A is an example of specificity of a representative AIR-CAR. in vitro killing assays of a representative AIR-CAR, AIR-CAR 397 show AIR-CAR 397 kills cells presenting ZNF737, but not ARIIGAP18 and MEIS2 peptides. The table shows that all peptides have similar hotspots:

[0036] Figure 5B is an example of the effect of mutations in the CAR target on the ability of a representative AIR-CAR to bind and kill;

[0037] Figure 5C is an example of the characterization of two representative AIR-CARs, AIR-CAR 637 and AIR-CAR 397. Each tested AIR-CAR demonstrates on-target binding;

[0038] Figure 5D is an example of in vitro validation confirming on-target killing ability of a representative AIR-CAR, AIR-CAR 637.

[0039] It should be appreciated that the logical operations described herein with respect to the various figures may be implemented (1) as a sequence of computer-implemented acts or program modules (i.e., software) running on a computing device (e.g., the computing device described in Figure IE), (2) as interconnected machine logic circuits or circuitAttorney Docket No. 11820-039W01YAROl-HPRO modules (i.e., hardware) within the computing device and / or (3) a combination of software and hardware of the computing device.

[0040] Throughout the drawings, the same reference numerals and characters, unless otherwise stated, are used to denote features, elements, components, or portions of the illustrated embodiments. Moreover, while the present disclosure will now be described in detail with reference to the figures, it is done so in connection with the illustrative embodiments and is not limited by the particular embodiments illustrated in the figures and the appended paragraphs.Detailed DescriptionDefinitions

[0041] An "increase" can refer to any change that results in a greater amount of a symptom, disease, composition, condition or activity. An increase can be any individual, median, or average increase in a condition, symptom, activity, composition in a statistically significant amount. Thus, the increase can be a 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100% increase so long as the increase is statistically significant.

[0042] A "decrease" can refer to any change that results in a smaller amount of a symptom, disease, composition, condition, or activity. A substance is also understood to decrease the genetic output of a gene when the genetic output of the gene product with the substance is less relative to the output of the gene product without the substance. Also for example, a decrease can be a change in the symptoms of a disorder such that the symptoms are less than previously observed. A decrease can be any individual, median, or average decrease in a condition, symptom, activity, composition in a statistically significant amount. Thus, the decrease can be a 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100% decrease so long as the decrease is statistically significant.

[0043] "Inhibit," "inhibiting," and "inhibition" mean to decrease an activity, response, condition, disease, or other biological parameter. This can include but is not limited to the complete ablation of the activity, response, condition, or disease. This may also include, for example, a 10% reduction in the activity, response, condition, or disease as compared to the native or control level. Thus, the reduction can be a 10, 20, 30, 40, 50, 60, 70, 80, 90, 100%, or any amount of reduction in between as compared to native or control levels.

[0044] By “reduce” or other forms of the word, such as “reducing” or “reduction,” is meant lowering of an event or characteristic (e.g., tumor growth). It is understood that this isAttorney Docket No. 11820-039W01YAROl-HPRO typically in relation to some standard or expected value, in other words it is relative, but that it is not always necessary for the standard or relative value to be referred to. For example, “reduces tumor growth” means reducing the rate of growth of a tumor relative to a standard or a control.

[0045] By “prevent” or other forms of the word, such as “preventing” or “prevention,” is meant to stop a particular event or characteristic, to stabilize or delay the development or progression of a particular event or characteristic, or to minimize the chances that a particular event or characteristic will occur. Prevent does not require comparison to a control as it is typically more absolute than, for example, reduce. As used herein, something could be reduced but not prevented, but something that is reduced could also be prevented. Likewise, something could be prevented but not reduced, but something that is prevented could also be reduced. It is understood that where reduce or prevent are used, unless specifically indicated otherwise, the use of the other word is also expressly disclosed.

[0046] The term “subject” refers to any individual who is the target of administration or treatment. The subject can be a vertebrate, for example, a mammal. In one aspect, the subject can be human, non-human primate, bovine, equine, porcine, canine, or feline. The subject can also be a guinea pig, rat, hamster, rabbit, mouse, or mole. Thus, the subject can be a human or veterinary patient. The term “patient” refers to a subject under the treatment of a clinician, e.g., physician.

[0047] The term “treatment” refers to the medical management of a patient with the intent to cure, ameliorate, stabilize, or prevent a disease, pathological condition, or disorder. This term includes active treatment, that is, treatment directed specifically toward the improvement of a disease, pathological condition, or disorder, and also includes causal treatment, that is, treatment directed toward removal of the cause of the associated disease, pathological condition, or disorder. In addition, this term includes palliative treatment, that is, treatment designed for the relief of symptoms rather than the curing of the disease, pathological condition, or disorder; preventative treatment, that is, treatment directed to minimizing or partially or completely inhibiting the development of the associated disease, pathological condition, or disorder; and supportive treatment, that is, treatment employed to supplement another specific therapy directed toward the improvement of the associated disease, pathological condition, or disorder.

[0048] "Biocompatible" generally refers to a material and any metabolites or degradation products thereof that are generally non-toxic to the recipient and do not cause significant adverse effects to the subject.Attorney Docket No. 11820-039W01YAROl-HPRO

[0049] “Effective amount” of an agent refers to a sufficient amount of an agent to provide a desired effect. The amount of agent that is “effective” will vary from subject to subject, depending on many factors such as the age and general condition of the subject, the particular agent or agents, and the like. Thus, it is not always possible to specify a quantified “effective amount.” However, an appropriate “effective amount” in any subject case may be determined by one of ordinary skill in the art using routine experimentation. Also, as used herein, and unless specifically stated otherwise, an “effective amount” of an agent can also refer to an amount covering both therapeutically effective amounts and prophylactically effective amounts. An “effective amount” of an agent necessary to achieve a therapeutic effect may vary according to factors such as the age, sex, and weight of the subject. Dosage regimens can be adjusted to provide the optimum therapeutic response. For example, several divided doses may be administered daily or the dose may be proportionally reduced as indicated by the exigencies of the therapeutic situation.

[0050] A "pharmaceutically acceptable" component can refer to a component that is not biologically or otherwise undesirable, i.e., the component may be incorporated into a pharmaceutical formulation provided by the disclosure and administered to a subject as described herein without causing significant undesirable biological effects or interacting in a deleterious manner with any of the other components of the formulation in which it is contained. When used in reference to administration to a human, the term generally implies the component has met the required standards of toxicological and manufacturing testing or that it is included on the Inactive Ingredient Guide prepared by the U.S. Food and Drug Administration.

[0051] "Pharmaceutically acceptable carrier" (sometimes referred to as a “carrier”) means a carrier or excipient that is useful in preparing a pharmaceutical or therapeutic composition that is generally safe and non-toxic and includes a carrier that is acceptable for veterinary and / or human pharmaceutical or therapeutic use. The terms "carrier" or "pharmaceutically acceptable carrier" can include, but are not limited to, phosphate buffered saline solution, water, emulsions (such as an oil / water or water / oil emulsion) and / or various types of wetting agents. As used herein, the term "carrier" encompasses, but is not limited to, any excipient, diluent, filler, salt, buffer, stabilizer, solubilizer, lipid, stabilizer, or other material well known in the art for use in pharmaceutical formulations and as described further herein.

[0052] “Pharmacologically active” (or simply “active”), as in a “pharmacologically active” derivative or analog, can refer to a derivative or analog (e.g., a salt, ester, amide,Attorney Docket No. 11820-039W01YAROl-HPRO conjugate, metabolite, isomer, fragment, etc.) having the same type of pharmacological activity as the parent compound and approximately equivalent in degree.

[0053] “Therapeutic agent” refers to any composition that has a beneficial biological effect. Beneficial biological effects include both therapeutic effects, e.g., treatment of a disorder or other undesirable physiological condition, and prophylactic effects, e.g., prevention of a disorder or other undesirable physiological condition (e.g., a non-immunogenic cancer). The terms also encompass pharmaceutically acceptable, pharmacologically active derivatives of beneficial agents specifically mentioned herein, including, but not limited to, salts, esters, amides, proagents, active metabolites, isomers, fragments, analogs, and the like. When the terms “therapeutic agent” is used, then, or when a particular agent is specifically identified, it is to be understood that the term includes the agent per se as well as pharmaceutically acceptable, pharmacologically active salts, esters, amides, proagents, conjugates, active metabolites, isomers, fragments, analogs, etc.

[0054] The term “therapeutically effective” refers to the amount of the composition used is of sufficient quantity to ameliorate one or more causes or symptoms of a disease or disorder. Such amelioration only requires a reduction or alteration, not necessarily elimination.

[0055] “Therapeutically effective amount” or “therapeutically effective dose” of a composition (e.g. a composition comprising an agent) refers to an amount that is effective to achieve a desired therapeutic result. In some embodiments, a desired therapeutic result is the control of type I diabetes. In some embodiments, a desired therapeutic result is the control of obesity. Therapeutically effective amounts of a given therapeutic agent will typically vary with respect to factors such as the type and severity of the disorder or disease being treated and the age, gender, and weight of the subject. The term can also refer to an amount of a therapeutic agent, or a rate of delivery of a therapeutic agent (e.g., amount over time), effective to facilitate a desired therapeutic effect, such as pain relief. The precise desired therapeutic effect will vary according to the condition to be treated, the tolerance of the subject, the agent and / or agent formulation to be administered (e.g., the potency of the therapeutic agent, the concentration of agent in the formulation, and the like), and a variety of other factors that are appreciated by those of ordinary skill in the art. In some instances, a desired biological or medical response is achieved following administration of multiple dosages of the composition to the subject over a period of days, weeks, or years.

[0056] A “control” is an alternative subject or sample used in an experiment for comparison purposes. A control can be "positive" or "negative."Attorney Docket No. 11820-039W01YAROl-HPRO

[0057] The term “bind” or “interacts with” refers to a first molecule recognizing and adhering to a second molecule, where the first molecule does not substantially recognize and / or adhere to other structurally unrelated molecules. Molecules, for example, antibodies can specifically bind a first antigen, but may not substantially bind or recognize other molecules.

[0058] A “binding protein” or “binder sequence” is a protein or peptide that is able to non-covalently bind another molecule. A binding protein can bind another protein or itself and can have more than one type of binding activity. It is preferred that optimal binder sequences have specific and strong binding. In the present disclosure, strong binding preferably is an interaction between the binder sequence and the target that is sufficient to activate a response in the cell which encodes the binder sequence, for example activation of a T cell.

[0059] The term “amino acid,” includes but is not limited to amino acids contained in the group consisting of alanine (Ala or A), cysteine (Cys or C), aspartic acid (Asp or D), glutamic acid (Glu or E), phenylalanine (The or F), glycine (Gly or G), histidine (His or H), isoleucine (He or 1), lysine (Lys or K), leucine (Leu or L), methionine (Met or M), asparagine (Asn or N), proline (Pro or P), glutamine (Gin or Q), arginine (Arg or R), serine (Ser or S), threonine (Thr or T), valine (Vai or V), tryptophan (Trp or W), and tyrosine (Tyr or Y) residues. Amino acids have the generic H2NCHRCOOH, where R is an organic substituent known as the variable side chain. Amino acids are defined by their variable side chain. The variable side chain defines the chemical and physical properties of each individual amino acid and influences overall protein folding. Positively charged amino acids include Arg, His, and Lys. Negatively charged amino acids include Asp and Glu. Polar uncharged amino acids include Ser, Thr, Asn, and Gin. Hydrophobic amino acids include Ala, Vai, He, Leu, Met, Phe, Tyr, and Trp. Special amino acids include Cys, Gly, and Pro.

[0060] The term “amino acid residue” also may include amino acid residues contained in the group consisting of homocysteine, 2-Aminoadipic acid, N-Ethylasparagine, 3-Aminoadipic acid, Hydroxy lysine, P-alanine, [LAmino-propionic acid, allo-Hydroxylysine acid, 2-Aminobutyric acid, 3-Hydroxyproline, 4-Aminobutyric acid, 4-Hydroxyproline, piperidinic acid, 6-Aminocaproic acid, Isodesmosine, 2-Aminoheptanoic acid, allo-Isoleucine, 2-Aminoisobutyric acid, N-Methylglycine, sarcosine, 3-Aminoisobutyric acid, N-Methylisoleucine, 2-Aminopimelic acid, 6-N-Methyllysine, 2,4-Diaminobutyric acid, N-Methylvaline, Desmosine, Norvaline, 2,2'-Diaminopimelic acid, Norleucine, 2,3-Diaminopropionic acid, Ornithine, Selenocystine, and N-Ethylglycine. Typically, the amideAttorney Docket No. 11820-039W01YAROl-HPRO linkages of the peptides are formed from an amino group of the backbone of one amino acid and a carboxyl group of the backbone of another amino acid.

[0061] The peptides, polypeptides, and proteins disclosed herein may be modified to include non-amino acid moieties. Modifications may include but are not limited to carboxylation (e.g., N-terminal carboxylation via addition of a di-carboxylic acid having 4-7 straight-chain or branched carbon atoms, such as glutaric acid, succinic acid, adipic acid, and 4,4-dimethylglutaric acid), amidation (e.g., C-terminal amidation via addition of an amide or substituted amide such as alkylamide or dialkylamide), PEGylation (e.g., N-terminal or C-terminal PEGylation via additional of polyethylene glycol), acylation (e.g., O-acylation (esters), N-acylation (amides), S-acylation (thioesters)), acetylation (e.g., the addition of an acetyl group, either at the N-terminus of the protein or at lysine residues), formylation lipoylation (e.g., attachment of a lipoate, a C8 functional group), myristoylation (e.g., attachment of myristate, a C14 saturated acid), palmitoylation (e.g., attachment of palmitate, a C16 saturated acid), alkylation (e.g., the addition of an alkyl group, such as an methyl at a lysine or arginine residue), isoprenylation or prenylation (e.g., the addition of an isoprenoid group such as farnesol or geranylgeraniol), amidation at C-terminus, glycosylation (e.g., the addition of a glycosyl group to either asparagine, hydroxylysine, serine, or threonine, resulting in a glycoprotein). Distinct from glycation, which is regarded as a nonenzymatic attachment of sugars, polysialylation (e.g., the addition of polysialic acid), glypiation (e.g., glycosylphosphatidylinositol (GPI) anchor formation, hydroxylation, iodination (e.g., of thyroid hormones), and phosphorylation (e.g., the addition of a phosphate group, usually to serine, tyrosine, threonine or histidine).

[0062] The disclosed nucleic acids are made up of for example, nucleotides, nucleotide analogs, or nucleotide substitutes. Non-limiting examples of these and other molecules are discussed herein. It is understood that for example, when a vector is expressed in a cell, that the expressed mRNA will typically be made up of A, C, G, and U. Likewise, it is understood that if, for example, an antisense molecule is introduced into a cell or cell environment through for example exogenous delivery, it is advantageous that the antisense molecule be made up of nucleotide analogs that reduce the degradation of the antisense molecule in the cellular environment.Protein structure and interaction predictions

[0063] Proteins may be defined or specified by one or more amino acid sequences in two dimensions (2D) or three dimensions (3D). The amino acid sequence can include, for example, a long polypeptide, a short polypeptide, or a peptide. The structure of a proteinAttorney Docket No. 11820-039W01YAROl-HPRO folded in 3D is determined by the physical and chemical properties of the amino acids that make up the protein. The primary forces driving protein folding are hydrophobic interactions, hydrogen bonds, and van der Waals forces. Additionally, the structure of the amino acid itself, H2NCHRCOOH, also limits the folding and structure of a protein. The amino acid chemical structure restricts angles and conformations of an overall protein.

[0064] From a biophysical perspective, folding of proteins is exemplified by the concept of a folding energy funnel within the atomic configuration space, as protein folding must be thermodynamically favorable. Multiple energy minima correspond to alternate stable states for the protein that make transitions with characterizable rates. Thus, protein structure can be inferred from an amino acid sequence. Machine learning methods, such as deep learning methods, can be used for protein structure prediction. Programs are known in the art to predict and characterize protein structure, folding, thermodynamics, kinetics of conformational transitions, ab initio folding, and protein-protein and protein-ligand binding and interactions.

[0065] The overall structure of a protein and the chemical and physical properties of the amino acids that make up the protein also determine protein-protein interactions. Proteinprotein interactions, or protein binding, is the physical contact established between two or more protein molecules. Hydrogen bonds between amino acid side chains or amino acid side chains and the amino or carboxyl groups of the amino acid backbone help mediate protein binding. Hydrophobic interactions between non-polar amino acid side chains also foster protein binding. Additionally, ionic bonds, due to the presence of oppositely charged ions stabilize protein binding. Salt bridges also play an important role in strengthening proteinprotein binding interactions. Other interactions that facilitate protein binding include TI-TI stacking, which occurs when aromatic rings where the p orbital electron clouds overlap, and cation-71 interactions, in which a positively charged cation interacts with an aromatic ring that exhibits a quadrupole moment.

[0066] Different parameters can be measured to characterize protein-protein interactions or binding. One parameter measured in characterizing protein-protein interactions and potential binding strength is size of the binding interface. This is often referred to as buried surface area. A greater binding interface can strengthen protein-protein interactions.Complementary charges between the two interacting interfaces strengthen protein binding, as well as hydrophobicity. The shape of the proteins involved in the binding interaction also influences the strength of the interaction and the ability of a protein to respond or conformationally change in response to the binding.Attorney Docket No. 11820-039WG1YAROl-HPRO

[0067] The discovery system of the present disclosure identifies de novo peptides capable of binding HLA-presented peptides to target cancers. The disclosed method and system utilize artificial intelligence to predict structures of known cancer related proteins and create novel peptides capable of binding and recognizing cancer proteins.Nucleotides and related molecules

[0068] A nucleotide is a molecule that contains a base moiety, a sugar moiety and a phosphate moiety. Nucleotides can be linked together through their phosphate moieties and sugar moieties creating an internucleoside linkage. The base moiety of a nucleotide can be adenin-9-yl (A), cytosin-l-yl (C), guanin-9-yl (G), uracil-l-yl (U), and thymin-l-yl (T). The sugar moiety of a nucleotide is a ribose or a deoxyribose. The phosphate moiety of a nucleotide is pentavalent phosphate. A non-limiting example of a nucleotide would be 3'-AMP (3'-adenosine monophosphate) or 5'-GMP (5'-guanosine monophosphate). There are many varieties of these types of molecules available in the art and available herein.

[0069] A nucleotide analog is a nucleotide which contains some type of modification to either the base, sugar, or phosphate moieties. Modifications to nucleotides are well known in the art and would include for example, 5 -methylcytosine (5-me-C), 5 -hydroxymethyl cytosine, xanthine, hypoxanthine, and 2-aminoadenine as well as modifications at the sugar or phosphate moieties. There are many varieties of these types of molecules available in the art and available herein.

[0070] Nucleotide substitutes are molecules having similar functional properties to nucleotides, but which do not contain a phosphate moiety, such as peptide nucleic acid (PNA). Nucleotide substitutes are molecules that will recognize nucleic acids in a Watson-Crick or Hoogsteen manner, but which are linked together through a moiety other than a phosphate moiety. Nucleotide substitutes are able to conform to a double helix type structure when interacting with the appropriate target nucleic acid. There are many varieties of these types of molecules available in the art and available herein.

[0071] It is also possible to link other types of molecules (conjugates) to nucleotides or nucleotide analogs to enhance for example, cellular uptake. Conjugates can be chemically linked to the nucleotide or nucleotide analogs. Such conjugates include but are not limited to lipid moieties such as a cholesterol moiety. (Letsinger et al., Proc. Natl. Acad. Sei. USA, 1989, 86, 6553-6556). There are many varieties of these types of molecules available in the art and available herein.

[0072] A Watson-Crick interaction is at least one interaction with the Watson-Crick face of a nucleotide, nucleotide analog, or nucleotide substitute. The Watson-Crick face of aAttorney Docket No. 11820-039W01YAROl-HPRO nucleotide, nucleotide analog, or nucleotide substitute includes the C2, Nl, and C6 positions of a purine based nucleotide, nucleotide analog, or nucleotide substitute and the C2, N3, C4 positions of a pyrimidine based nucleotide, nucleotide analog, or nucleotide substitute.

[0073] A Hoogsteen interaction is the interaction that takes place on the Hoogsteen face of a nucleotide or nucleotide analog, which is exposed in the major groove of duplex DNA. The Hoogsteen face includes the N7 position and reactive groups (NH2 or O) at the C6 position of purine nucleotides.Functional Nucleic Acids

[0074] Functional nucleic acids are nucleic acid molecules that have a specific function, such as binding a target molecule or catalyzing a specific reaction. Functional nucleic acid molecules can be divided into the following categories, which are not meant to be limiting. For example, functional nucleic acids include antisense molecules, aptamers, ribozymes, triplex forming molecules, and external guide sequences. The functional nucleic acid molecules can act as affcctors, inhibitors, modulators, and stimulators of a specific activity possessed by a target molecule, or the functional nucleic acid molecules can possess a de novo activity independent of any other molecules.

[0075] In one aspect, disclosed herein are engineered cells (for example, T cells) comprising a nucleic acid encoding a chimeric antigen receptor that binds to an HLA-present peptide or neoantigen.

[0076] It is further understood and herein contemplated that the disclosed neoantigens can not only serve as an active component of a vaccine, but can be a target for a tumor infiltrating lymphocyte (TIL), T cell receptor (TCR), antibodies, scFv, bispecific T cell engagers (BiTEs), nanobodies, diabodies, and / or chimeric antigen receptor (CAR) including, but not limited to CAR T cells, CAR natural killer (NK) cells (CAR NK cells), CAR NK T cells, CAR macrophage (CARMA), and peptide-centric CARs (PC-CARs).Antibodies

[0077] The term “antibodies” is used herein in a broad sense and includes both polyclonal and monoclonal antibodies. In addition to intact immunoglobulin molecules, also included in the term “antibodies” are fragments or polymers of those immunoglobulin molecules, and human or humanized versions of immunoglobulin molecules or fragments thereof. The antibodies can be tested for their desired activity using the in vitro assays described herein, or by analogous methods, after which their in vivo therapeutic and / or prophylactic activities are tested according to known clinical testing methods. There are five major classes of human immunoglobulins: IgA, IgD, IgE, IgG and IgM, and several of these may be further dividedAttorney Docket No. 11820-039W01YAROl-HPRO into subclasses (isotypes), e.g., IgG-1, IgG-2, IgG-3, and IgG-4; IgA-1 and IgA-2. One skilled in the art would recognize the comparable classes for mouse. The heavy chain constant domains that correspond to the different classes of immunoglobulins are called alpha, delta, epsilon, gamma, and mu, respectively.

[0078] The term “monoclonal antibody” as used herein refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies within the population are identical except for possible naturally occurring mutations that may be present in a small subset of the antibody molecules. The monoclonal antibodies herein specifically include "chimeric" antibodies in which a portion of the heavy and / or light chain is identical with or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain(s) is identical with or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies, as long as they exhibit the desired antagonistic activity.

[0079] The disclosed monoclonal antibodies can be made using any procedure which produces mono clonal antibodies. For example, disclosed monoclonal antibodies can be prepared using hybridoma methods, such as those described by Kohler and Milstein, Nature.256:495 (1975). In a hybridoma method, a mouse or other appropriate host animal is typically immunized with an immunizing agent to elicit lymphocytes that produce or are capable of producing antibodies that will specifically bind to the immunizing agent. Alternatively, the lymphocytes may be immunized in vitro.

[0080] The monoclonal antibodies may also be made by recombinant DNA methods. DNA encoding the disclosed monoclonal antibodies can be readily isolated and sequenced using conventional procedures (e.g., by using oligonucleotide probes that are capable of binding specifically to genes encoding the heavy and light chains of murine antibodies). Libraries of antibodies or active antibody fragments can also be generated and screened using phage display techniques, e.g., as described in U.S. Patent No. 5,804,440 to Burton et al. and U.S. Patent No. 6,096,441 to Barbas et al.

[0081] In vitro methods are also suitable for preparing monovalent antibodies. Digestion of antibodies to produce fragments thereof, particularly, Fab fragments, can be accomplished using routine techniques known in the art. For instance, digestion can be performed using papain. Examples of papain digestion are described in WO 94 / 29348 published Dec. 22, 1994 and U.S. Pat. No. 4,342,566. Papain digestion of antibodies typically produces two identical antigen binding fragments, called Fab fragments, each with a single antigen binding site, andAttorney Docket No. 11820-039W01YAROl-HPRO a residual I 'c fragment. Pepsin treatment yields a fragment that has two antigen combining sites and is still capable of cross-linking antigen.

[0082] As used herein, the term “antibody or fragments thereof’ encompasses chimeric antibodies and hybrid antibodies, with dual or multiple antigen or epitope specificities, diabodies, and fragments, such as F(ab’)2, Fab’, Fab, Fv, sFv, scFv, nanobodies, and the like, including hybrid fragments. Thus, fragments of the antibodies that retain the ability to bind their specific antigens are provided. Such antibodies and fragments can be made by techniques known in the art and can be screened for specificity and activity according to the methods set forth in the Examples and in general methods for producing antibodies and screening antibodies for specificity and activity (See Harlow and Lane. Antibodies, A Laboratory Manual. Cold Spring Harbor Publications, New York, (1988)).

[0083] Also included within the meaning of “antibody or fragments thereof’ are conjugates of antibody fragments and antigen binding proteins (single chain antibodies).

[0084] The fragments, whether attached to other sequences or not, can also include insertions, deletions, substitutions, or other selected modifications of particular regions or specific amino acids residues, provided the activity of the antibody or antibody fragment is not significantly altered or impaired compared to the non-modified antibody or antibody fragment. These modifications can provide for some additional property, such as to remove / add amino acids capable of disulfide bonding, to increase its bio-longevity, to alter its secretory characteristics, etc. In any case, the antibody or antibody fragment must possess a bioactive property, such as specific binding to its cognate antigen. Functional or active regions of the antibody or antibody fragment may be identified by mutagenesis of a specific region of the protein, followed by expression and testing of the expressed polypeptide. Such methods are readily apparent to a skilled practitioner in the art and can include site-specific mutagenesis of the nucleic acid encoding the antibody or antibody fragment. (Zoller, M.J. Curr. Opin. Biotechnol. 3:348-354, 1992).

[0085] As used herein, the term “antibody” or “antibodies” can also refer to a human antibody and / or a humanized antibody. Many non-human antibodies (e.g., those derived from mice, rats, or rabbits) are naturally antigenic in humans, and thus can give rise to undesirable immune responses when administered to humans. Therefore, the use of human or humanized antibodies in the methods serves to lessen the chance that an antibody administered to a human will evoke an undesirable immune response.Human antibodiesAttorney Docket No. 11820-039W01YAROl-HPRO

[0086] The disclosed human antibodies can be prepared using any technique. The disclosed human antibodies can also be obtained from transgenic animals. For example, transgenic, mutant mice that are capable of producing a full repertoire of human antibodies, in response to immunization, have been described (see, e.g., Jakobovits et al.. Proc. Natl. Acad. Sci. USA, 90:2551-255 (1993); Jakobovits et al., Nature, 362:255-258 (1993);Bruggemiann et al., Year in Immunol., 7:33 (1993)). Specifically, the homozygous deletion of the antibody heavy chain joining regiongene in these chimeric and germ-line mutant mice results in complete inhibition of endogenous antibody production, and the successful transfer of the human germ-line antibody gene array into such gem- line mutant mice results in the production of human antibodies upon antigen challenge. Antibodies having the desired activity are selected using Env-CD4-co-receptor complexes as described herein.Humanized antibodies

[0087] Antibody humanization techniques generally involve the use of recombinant DNA technology to manipulate the DNA sequence encoding one or more polypeptide chains of an antibody molecule. Accordingly, a humanized form of a non-human antibody (or a fragment thereof) is a chimeric antibody or antibody chain (or a fragment thereof, such as an sFv, Fv, Fab, Fab’, F(ab’)2, or other antigen-binding portion of an antibody) which contains a portion of an antigen binding site from a non-human (donor) antibody integrated into the framework of a human (recipient) antibody.

[0088] To generate a humanized antibody, residues from one or more complementarity determining regions (CDRs) of a recipient (human) antibody molecule are replaced by residues from one or more CDRs of a donor (non-human) antibody molecule that is known to have desired antigen binding characteristics (e.g., a certain level of specificity and affinity for the target antigen). In some instances, Fv framework (FR) residues of the human antibody are replaced by corresponding non-human residues. Humanized antibodies may also contain residues which are found neither in the recipient antibody nor in the imported CDR or framework sequences. Generally, a humanized antibody has one or more amino acid residues introduced into it from a source which is non-human. In practice, humanized antibodies are typically human antibodies in which some CDR residues and possibly some FR residues are substituted by residues from analogous sites in rodent antibodies. Humanized antibodies generally contain at least a portion of an antibody constant region (Fc), typically that of a human antibody (Jones et al., Nature, 321:522-525 (1986), Reichmann et al., Nature, 332:323-327 (1988), and Presta, Curr. Opin. Struct. Biol., 2:593-596 (1992)).Attorney Docket No. 11820-039W01YAR01-11PRO

[0089] Methods for humanizing non-human antibodies are well known in the art. For example, humanized antibodies can be generated according to the methods 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. Methods that can be used to produce humanized antibodies are also described in U.S. Patent No. 4,816,567 (Cabilly et al.), U.S. Patent No. 5,565,332 (Hoogenboom et al.), U.S. Patent No. 5,721,367 (Kay et al.), U.S. Patent No. 5,837,243 (Deo et al.), U.S. Patent No. 5, 939,598 (Kucherlapati et al.), U.S. Patent No. 6,130,364 (Jakobovits et al.), and U.S. Patent No. 6,180,377 (Morgan et al.).Chimeric antigen receptors

[0090] In one aspect, disclosed herein is chimeric antigen receptor (CAR) that binds to one or more peptides (i.e., a peptide centric (PC) CAR (PC-CAR)) (including, but not limited to a CAR that recognizes the same peptide across two or more HLA alleles or a single peptide in the context of a single HLA (i.e., TCR mimics)).

[0091] It is understood and herein contemplated that disclosed CAR can be expressed on any cell capable of expressing said CAR including, but not limited to CAR T cells, CAR natural killer (NK) cells (CAR NK cells), CAR NK T cells, CAR macrophage (CARMA). In some aspects the CAR is a peptide-centric CAR (PC-CAR).

[0092] Chimeric antigen receptors (CARs) are engineered receptors, which graft an arbitrary specificity onto a cell, typically an immune effector cell. The extracellular recognition unit generally contains an antibody-derived recognition domain. The antigen recognition domain of the disclosed CAR is usually an scFv. There are however many alternatives. An antigen recognition domain from native T-cell receptor (TCR) alpha and beta single chains have been described, as have simple ectodomains (e.g. CD4 ectodomain to recognize HIV infected cells) and more exotic recognition components such as a linked cytokine (which leads to recognition of cells bearing the cytokine receptor). In fact almost anything that binds a given target with high affinity can be used as an antigen recognition region.

[0093] The endodomain is the business end of the CAR that after antigen recognition transmits a signal to the immune effector cell, activating at least one of the normal effector functions of the immune effector cell. Effector function of a T cell, for example, may be cytolytic activity or helper activity including the secretion of cytokines. Therefore, the endodomain may comprise the “intracellular signaling domain” of a T cell receptor (TCR)Attorney Docket No. 11820-039W01YAROl-HPRO and optional co-receptors. While usually the entire intracellular signaling domain can be employed, in many cases it is not necessary to use the entire chain. To the extent that a truncated portion of the intracellular signaling domain is used, such truncated portion may be used in place of the intact chain as long as it transduces the effector function signal.

[0094] Cytoplasmic signaling sequences that regulate primary activation of the TCR complex that act in a stimulatory manner may contain signaling motifs which are known as immunoreceptor tyrosine -based activation motifs (ITAMs). Examples of ITAM containing cytoplasmic signaling sequences include those derived from CD8, CD3^, CD36, CD3y, CD3s, CD32 (Fc gamma Rlla), DAP10, DAP12, CD79a, CD79b, FcyRIy, FcyRIIIy, FcsRip (FCERIB), and FcsRIy (FCERIG).

[0095] In particular embodiments, the intracellular signaling domain is derived from CD3 zeta (CD3Q (TCR zeta, GenBank aceno. BAG36664.1). T-cell surface glycoprotein CD3 zeta (CD3Q chain, also known as T-cell receptor T3 zeta chain or CD247 (Cluster of Differentiation 247), is a protein that in humans is encoded by the CD247 gene.

[0096] First-generation CARs typically had the intracellular domain from the CD3 chain, which is the primary transmitter of signals from endogenous TCRs. Second-generation CARs add intracellular signaling domains from various costimulatory protein receptors (e.g., CD28, 41BB, ICOS) to the endodomain of the CAR to provide additional signals to the T cell. Preclinical studies have indicated that the second generation of CAR designs improves the antitumor activity of T cells. More recent, third-generation CARs combine multiple signaling domains to further augment potency. T cells grafted with these CARs have demonstrated improved expansion, activation, persistence, and tumor-eradicating efficiency independent of costimulatory receptor / ligand interaction (Imai C, et al. Leukemia 2004 18:676-84; Maher J, et al. Nat Biotechnol 200220:70-5).

[0097] For example, the endodomain of the CAR can be designed to comprise the CD3C, signaling domain by itself or combined with any other desired cytoplasmic domain(s) useful in the context of the CAR of the invention. For example, the cytoplasmic domain of the CAR can comprise a CD3^ chain portion and a costimulatory signaling region. The costimulatory signaling region refers to a portion of the CAR comprising the intracellular domain of a costimulatory molecule. A costimulatory molecule is a cell surface molecule other than an antigen receptor or their ligands that is required for an efficient response of lymphocytes to an antigen. Examples of such molecules include CD27, CD28, 4- IBB (CD 137), 0X40, CD30, CD40, ICOS, lymphocyte function-associated antigen- 1 (LFA-1), CD2, CD7, LIGHT, NKG2C, B7-H3, and a ligand that specifically binds with CD83, CD8, CD4, b2c, CD80,Attorney Docket No. 11820-039W01YAROl-HPRO CD86, DAP10. DAP12, MyD88, BTNL3, and NKG2D. Thus, while the CAR is exemplified primarily with CD28 as the co-stimulatory signaling element, other costimulatory elements can be used alone or in combination with other co-stimulatory signaling elements. Thus, specifically contemplated herein are CARs comprising any one or combination of two more co-stimulatory signaling elements for the group consisting of CD27, CD28, 4- IBB (CD 137), 0X40, CD30, CD40, ICOS, LFA-1, CD2, CD7, LIGHT, NKG2C, B7-H3, and a ligand that specifically binds with CD83, CD8, CD4, b2c, CD80, CD86, DAP 10, DAP 12, MyD88, BTNL3, and NKG2DCD28 and 4-1 BB, CD28 and 0X40, CD28 and LFA-1, CD28 and CD40. Thus, for example, specifically contemplated herein are CARs comprising co-stimulatory signaling elements for CD28 and CD40, CD28 and 4- IBB, CD28 and 0X40, and CD28 and LFA-1.

[0098] In some embodiments, the CAR comprises a hinge sequence. A hinge sequence is a short sequence of amino acids that facilitates antibody flexibility (see, e.g., Woof et al., Nat. Rev. Immunol., 4(2): 89-99 (2004)). The hinge sequence may be positioned between the antigen recognition moiety (e.g., anti-IL13Ra2 scFv) and the transmembrane domain. The hinge sequence can be any suitable sequence derived or obtained from any suitable molecule. In some embodiments, for example, the hinge sequence is derived from a CD8 alpha molecule or a CD28 molecule.

[0099] The transmembrane domain may be derived either from a natural or from a synthetic source. Where the source is natural, the domain may be derived from any membrane-bound or transmembrane protein. For example, the transmembrane region may be derived from (i.e. comprise at least the transmembrane region(s) of) the alpha, beta or zeta chain of the T-cell receptor, CD28, CD3 epsilon, CD45, CD4, CD5, CD8 (e.g., CD8 alpha, CD8 beta), CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, or CD 154, KIRDS2, 0X40, CD2, CD27, LFA-1 (GDI la, CD 18) , ICOS (CD278) , 4- IBB (CD 137) , GITR, CD40, BAFFR, HVEM (LIGHTR) , SLAMF7, NKp80 (KLRF1) , CD 160, CD 19, IL2R beta, IL2R gamma, IL7R a, ITGA1, VLA1, CD49a, ITGA4, IA4, CD49D, ITGA6, VLA-6, CD49f, ITGAD, CD lid, ITGAE, CD 103, ITGAL, CD Ila, LFA-1, ITGAM, CDllb, ITGAX, CDl lc, ITGB1, CD29, ITGB2, CD18, LFA-1, ITGB7, TNFR2, DNAM1 (CD226) , SLAMF4 (CD244, 2B4) , CD84, CD96 (Tactile) , CEACAM1, CRTAM, Ly9 (CD229) , CD160 (BY55) , PSGL1, CD100 (SEMA4D) , SLAMF6 (NTB-A, Lyl08) , SLAM (SLAMF1, CD 150, IPO-3) , BLAME (SLAMF8) , SELPLG (CD 162) , LTBR, and PAG / Cbp. Alternatively the transmembrane domain may be synthetic, in which case it will comprise predominantly hydrophobic residues such as leucine and valine. In some cases, aAttorney Docket No. 11820-039W01YAROl-HPRO triplet of phenylalanine, tryptophan and valine will be found at each end of a synthetic transmembrane domain. A short oligo- or polypeptide linker, such as between 2 and 10 amino acids in length, may form the linkage between the transmembrane domain and the endoplasmic domain of the CAR. In some embodiments, the linker can be a spacer derived from the same source as the transmembrane domain. For example in some instances, the spacer can and the transmembrane domain are both derived from the CD28 or CD8 alpha, or from any other source for the transmembrane domain listed above including, but not limited to, the alpha, beta or zeta chain of the T-cell receptor, CD3 epsilon, CD45, CD4, CD5, CD8 beta, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, or CD154, KIRDS2, 0X40, CD2, CD27, LFA-1 (CDlla, CD18) , ICOS (CD278) , 4-1BB (CD137) , GITR, CD40, BAFFR, HVEM (LIGHTR) , SLAMF7, NKp80 (KLRF1) , CD 160, CD 19, IL2R beta, IL2R gamma, IL7R a, ITGA1, VLA1, CD49a, ITGA4, IA4, CD49D, ITGA6, VLA-6, CD49f, ITGAD, CDlld, ITGAE, CD103, ITGAL, CDlla, LFA-1, ITGAM, CD1 lb, ITGAX, CDllc, ITGB1, CD29, ITGB2, CD18, LFA-1, ITGB7, TNFR2, DNAM1 (CD226) , SLAMF4 (CD244, 2B4) , CD84, CD96 (Tactile) . CEACAM1, CRTAM, Ly9 (CD229) . CD160 (BY55) , PSGL1, CD100 (SEMA4D) , SLAMF6 (NTB-A, Ly 108) , SLAM (SLAMF1, CD150, IPO-3) , BLAME (SLAMF8) , SELPLG (CD162) , LTBR, and PAG / Cbp. In other embodiments, the transmembrane domain and the linker (such as a spacer) can be derived from different sources, for example, a CD28 transmembrane domain and a CD8 alpha spacer or a CD8 alpha transmembrane domain and a CD28 spacer.

[0100] In some embodiments, the CAR has more than one transmembrane domain, which can be a repeat of the same transmembrane domain, or can be different transmembrane domains.

[0101] In some embodiments, the CAR is a multi-chain CAR, as described in WO2015 / 039523, which is incorporated by reference for this teaching. A multi-chain CAR can comprise separate extracellular ligand binding and signaling domains in different transmembrane polypeptides. The signaling domains can be designed to assemble in juxtamembrane position, which forms flexible architecture closer to natural receptors, that confers optimal signal transduction. For example, the multi-chain CAR can comprise a part of an FCERI alpha chain and a part of an FCERI beta chain such that the FCERI chains spontaneously dimerize together to form a CAR.Pharmaceutical carriers / Delivery of pharmaceutical products

[0102] As described above, the compositions can also be administered in vivo in a pharmaceutically acceptable carrier. By "pharmaceutically acceptable" is meant a materialAttorney Docket No. 11820-039W01YAROl-HPRO that is not biologically or otherwise undesirable, i.e., the material may be administered to a subject, along with the nucleic acid or vector, without causing any undesirable biological effects or interacting in a deleterious manner with any of the other components of the pharmaceutical composition in which it is contained. The carrier would naturally be selected to minimize any degradation of the active ingredient and to minimize any adverse side effects in the subject, as would be well known to one of skill in the art.

[0103] The compositions may be administered orally, parenterally (e.g., intravenously), by intramuscular injection, by intraperitoneal injection, transdermally, extracorporeally, topically or the like, including topical intranasal administration or administration by inhalant. As used herein, "topical intranasal administration" means delivery of the compositions into the nose and nasal passages through one or both of the nares and can comprise delivery by a spraying mechanism or droplet mechanism, or through aerosolization of the nucleic acid or vector. Administration of the compositions by inhalant can be through the nose or mouth via delivery by a spraying or droplet mechanism. Delivery can also be directly to any area of the respiratory system (e.g., lungs) via intubation. The exact amount of the compositions required will vary from subject to subject, depending on the species, age, weight and general condition of the subject, the severity of the allergic disorder being treated, the particular nucleic acid or vector used, its mode of administration and the like. Thus, it is not possible to specify an exact amount for every composition. However, an appropriate amount can be determined by one of ordinary skill in the art using only routine experimentation given the teachings herein.

[0104] Parenteral administration of the composition, if used, is generally characterized by injection. Injectables can be prepared in conventional forms, either as liquid solutions or suspensions, solid forms suitable for solution of suspension in liquid prior to injection, or as emulsions. A more recently revised approach for parenteral administration involves use of a slow release or sustained release system such that a constant dosage is maintained. See, e.g., U.S. Patent No. 3,610,795, which is incorporated by reference herein.

[0105] The materials may be in solution, suspension (for example, incorporated into microparticles, liposomes, or cells). These may be targeted to a particular cell type via antibodies, receptors, or receptor ligands. The following references are examples of the use of this technology to target specific proteins to tumor tissue (Senter, et al., Bioconjugate Chem., 2:447-451, (1991); Bagshawe, K.D., Br. J. Cancer, 60:275-281, (1989); Bagshawe, et al., Br. J. Cancer, 58:700-703, (1988); Senter, et al., Bioconjugate Chem., 4:3-9, (1993); Battelli, et al., Cancer Immunol. Immunother., 35:421-425, (1992); Pietersz and McKenzie, Immunolog. Reviews, 129:57-80, (1992); and Roffler, et al., Biochem. Pharmacol, 42:2062-2065, (1991)).Attorney Docket No. 11820-039WG1YAROl-HPRO Vehicles such as "stealth" and other antibody conjugated liposomes (including lipid mediated drug targeting to colonic carcinoma), receptor mediated targeting of DNA through cell specific ligands, lymphocyte directed tumor targeting, and highly specific therapeutic retroviral targeting of murine glioma cells in vivo. The following references are examples of the use of this technology to target specific proteins to tumor tissue (Hughes et al., Cancer Research, 49:6214-6220, (1989); and Litzinger and Huang, Biochimica et Biophysica Acta, 1104:179-187, (1992)). In general, receptors are involved in pathways of endocytosis, either constitutive or ligand induced. These receptors cluster in clathrin-coated pits, enter the cell via clathrin-coated vesicles, pass through an acidified endosome in which the receptors are sorted, and then either recycle to the cell surface, become stored intracellularly, or are degraded in lysosomes. The internalization pathways serve a variety of functions, such as nutrient uptake, removal of activated proteins, clearance of macromolecules, opportunistic entry of viruses and toxins, dissociation and degradation of ligand, and receptor- level regulation. Many receptors follow more than one intracellular pathway, depending on the cell type, receptor concentration, type of ligand, ligand valency, and ligand concentration.Molecular and cellular mechanisms of receptor-mediated endocytosis has been reviewed (Brown and Greene, DNA and Cell Biology 10:6, 399-409 (1991)).Pharmaceutically Acceptable Carriers

[0106] The compositions, including antibodies, can be used therapeutically in combination with a pharmaceutically acceptable carrier.

[0107] Suitable carriers and their formulations are described in Remington: The Science and Practice of Pharmacy (19th ed.) ed. A.R. Gennaro, Mack Publishing Company, Easton, PA 1995. Typically, an appropriate amount of a pharmaceutically-acceptable salt is used in the formulation to render the formulation isotonic. Examples of the pharmaceutically-acceptable carrier include, but are not limited to, saline, Ringer's solution and dextrose solution. The pH of the solution is preferably from about 5 to about 8, and more preferably from about 7 to about 7.5. Further carriers include sustained release preparations such as semipermeable matrices of solid hydrophobic polymers containing the antibody, which matrices are in the form of shaped articles, e.g., films, liposomes or microparticles. It will be apparent to those persons skilled in the art that certain carriers may be more preferable depending upon, for instance, the route of administration and concentration of composition being administered.

[0108] Pharmaceutical carriers are known to those skilled in the art. These most typically would be standard carriers for administration of drugs to humans, including solutions such asAttorney Docket No. 11820-039W01YAROl-HPRO sterile water, saline, and buffered solutions at physiological pH. The compositions can be administered intramuscularly or subcutaneously. Other compounds will be administered according to standard procedures used by those skilled in the art.

[0109] Pharmaceutical compositions may include carriers, thickeners, diluents, buffers, preservatives, surface active agents and the like in addition to the molecule of choice.Pharmaceutical compositions may also include one or more active ingredients such as antimicrobial agents, anti-inflammatory agents, anesthetics, and the like.

[0110] The pharmaceutical composition may be administered in a number of ways depending on whether local or systemic treatment is desired, and on the area to be treated. Administration may be topically (including ophthalmically, vaginally, rectally, intranasally), orally, by inhalation, or parenterally, for example by intravenous drip, subcutaneous, intraperitoneal or intramuscular injection. The disclosed antibodies can be administered intravenously, intraperitoneally, intramuscularly, subcutaneously, intracavity, or transdcrmally.

[0111] Preparations for parenteral administration include sterile aqueous or non-aqueous solutions, suspensions, and emulsions. Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Aqueous earners include water, alcoholic / aqueous solutions, emulsions or suspensions, including saline and buffered media. Parenteral vehicles include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's, or fixed oils. Intravenous vehicles include fluid and nutrient replenishers, electrolyte replenishers (such as those based on Ringer's dextrose), and the like. Preservatives and other additives may also be present such as, for example, antimicrobials, anti-oxidants, chelating agents, and inert gases and the like.

[0112] Formulations for topical administration may include ointments, lotions, creams, gels, drops, suppositories, sprays, liquids and powders. Conventional pharmaceutical carriers, aqueous, powder or oily bases, thickeners and the like may be necessary or desirable.

[0113] Compositions for oral administration include powders or granules, suspensions or solutions in water or non-aqueous media, capsules, sachets, or tablets. Thickeners, flavorings, diluents, emulsifiers, dispersing aids or binders may be desirable.

[0114] Some of the compositions may potentially be administered as a pharmaceutically acceptable acid- or base- addition salt, formed by reaction with inorganic acids such as hydrochloric acid, hydrobromic acid, perchloric acid, nitric acid, thiocyanic acid, sulfuric acid, and phosphoric acid, and organic acids such as formic acid, acetic acid, propionic acid,Attorney Docket No. 11820-039W01YAROl-HPRO glycolic acid, lactic acid, pyruvic acid, oxalic acid, malonic acid, succinic acid, maleic acid, and fumaric acid, or by reaction with an inorganic base such as sodium hydroxide, ammonium hydroxide, potassium hydroxide, and organic bases such as mono-, di-, trialkyl and aryl amines and substituted ethanolamines.Therapeutic Uses

[0115] Effective dosages and schedules for administering the compositions may be determined empirically, and making such determinations is within the skill in the art. The dosage ranges for the administration of the compositions are those large enough to produce the desired effect in which the symptoms of the disorder are effected. The dosage should not be so large as to cause adverse side effects, such as unwanted cross-reactions, anaphylactic reactions, and the like. Generally, the dosage will vary with the age, condition, sex and extent of the disease in the patient, route of administration, or whether other drugs are included in the regimen, and can be determined by one of skill in the art. The dosage can be adjusted by the individual physician in the event of any counterindications. Dosage can vary, and can be administered in one or more dose administrations daily, for one or several days. Guidance can be found in the literature for appropriate dosages for given classes of pharmaceutical products. For example, guidance in selecting appropriate doses for antibodies can be found in the literature on therapeutic uses of antibodies, e.g., Handbook of Monoclonal Antibodies, Ferrone et al., eds., Noges Publications, Park Ridge, N.J., (1985) ch. 22 and pp. 303-357; Smith et al., Antibodies in Human Diagnosis and Therapy, Haber et al., eds., Raven Press, New York (1977) pp. 365-389. A typical daily dosage of the antibody used alone might range from about 1 pg / kg to up to 100 mg / kg of body weight or more per day, depending on the factors mentioned above.Method of treating cancer

[0116] The disclosed compositions can be used to treat any disease where uncontrolled cellular proliferation occurs such as cancers. A representative but non-limiting list of cancers that the disclosed compositions can be used to treat is the following: lymphomas such as B cell lymphoma and T cell lymphoma; mycosis fungoides; Hodgkin’s Disease; myeloid leukemia (including, but not limited to acute myeloid leukemia (AML) and / or chronic myeloid leukemia (CML)); bladder cancer; brain cancer; nervous system cancer; head and neck cancer; squamous cell carcinoma of head and neck; renal cancer; lung cancers such as small cell lung cancer, non-small cell lung carcinoma (NSCLC), lung squamous cell carcinoma (LUSC), and Lung Adenocarcinomas (LUAD); neuroblastoma / glioblastoma; ovarian cancer; pancreatic cancer; pancreatic adenocarcinoma; prostate cancer; skin cancer;Attorney Docket No. 11820-039W01YAROl-HPRO skin cutaneous melanoma; hepatic cancer; melanoma; squamous cell carcinomas of the mouth, throat, larynx, and lung; cervical cancer; cervical carcinoma; cervical squamous cell carcinoma; breast cancer including, but not limited to triple negative breast cancer; genitourinary cancer; pulmonary cancer; esophageal carcinoma; head and neck carcinoma; large bowel cancer; hematopoietic cancers; testicular cancer; colon adenocarcinoma; and colon and rectal cancers.

[0117] In one aspect, also disclosed herein are methods of treating, decreasing, reducing, inhibiting, ameliorating, and / or preventing a cancer (such as, for example neuroblastoma / glioblastoma), cancer recurrence, and / or metastasis in a subject comprising administering to the subject a cell therapy comprising a chimeric antigen receptor (CAR) (including, but not limited to administration of CAR expressing T cells (CAR T cells), natural killer (NK) cells (CAR NK cells), CAR NK T cells, CAR macrophage (CARMA), and / or peptide-centric CAR) that binds to an HLA-presented peptide.

[0118] It is understood and herein contemplated that the disclosed treatment regimens can used alone or in combination with any anti-cancer therapy known in the art including, but not limited to Abemaciclib, Abiraterone Acetate, AB1TREXATE® (Methotrexate), ABRAXANE® (Paclitaxel Albumin-stabilized Nanoparticle Formulation), ABVD, ABVE, ABVE-PC, AC, AC-T, ADCETRIS® (Brentuximab Vedotin), ADE, Ado-Trastuzumab Emtansine, ADRIAMYCIN® (Doxorubicin Hydrochloride), Afatinib Dimaleate, AFINITOR® (Everolimus), AKYNZEO® (Netupitant and Palonosetron Hydrochloride), ALDARA® (Imiquimod), Aldesleukin, ALECENSA® (Alectinib), Alectinib, Alemtuzumab, ALIMTA® (Pemetrexed Disodium), ALIQOPA® (Copanlisib Hydrochloride), ALKERAN™ for Injection (Melphalan Hydrochloride), ALKERAN™ Tablets (Melphalan), ALOXI® (Palonosetron Hydrochloride), ALUNBRIG® (Brigatinib), AMBOCHLORIN® (Chlorambucil), AMBOCLORIN® (Chlorambucil), Amifostine, Aminolevulinic Acid, Anastrozole, Aprepitant, AREDIA® (Pamidronate Disodium), ARIMIDEX® (Anastrozole), AROMASIN® (Exemestane),ARRANON® (Nelarabine), Arsenic Trioxide, ARZERRA® (Ofatumumab), Asparaginase Erwinia chrysanthemi, Atezolizumab, AVASTIN® (Bevacizumab), Avelumab, Axitinib, Azacitidine, BAVENCIO® (Avelumab), BEACOPP, BECENUM® (Carmustine), BELEODAQ® (Belinostat), Belinostat, Bendamustine Hydrochloride, BEP, BESPONSA® (Inotuzumab Ozogamicin) , Bevacizumab, Bexarotene, BEXXAR® (Tositumomab and Iodine I 131 Tositumomab), Bicalutamide, BICNU® (Carmustine), Bleomycin, Blinatumomab, BLINCYTO® (Blinatumomab), Bortezomib, BOSULIF® (Bosutinib), Bosutinib, Brentuximab Vedotin, Brigatinib, BuMel, Busulfan,Attorney Docket No. 11820-039W01YAROl-HPRO BUSULFFX® (Busulfan), Cabazitaxel, CABOMETYX® (Cabozantinib-S-Malate), Cabozantinib-S-Malate, CAF, CAMPATH® (Alemtuzumab), CAMPTOSAR® (Irinotecan Hydrochloride), Capecitabine, CAPOX, CARAC® (Fluorouracil-Topical), Carboplatin, CARBOPLATIN-TAXOL, Carfilzomib, CARMUBRIS® (Carmustine), Carmustine, Carmustine Implant, CASODEX® (Bicalutamide), CEM, Ceritinib, CERUBIDINE® (Daunorubicin Hydrochloride), CERVARIX® (Recombinant HPV Bivalent Vaccine), Cetuximab, CEV, Chlorambucil, CHLORAMBUCIL-PREDNISONE, CHOP, Cisplatin, Cladribine, CEAFEN® (Cyclophosphamide), Clofarabine, CEOFAREX® (Clofarabine), CLOLAR® (Clofarabine), CMF, Cobimetinib, COMETRIQ® (Cabozantinib-S-Malate), Copanlisib Hydrochloride, COPDAC, COPP, COPP-AB V, COSMEGEN® (Dactinomycin), COTELLIC® (Cobimetinib), Crizotinib, CVP, Cyclophosphamide, CYFOS® (Ifosfamide), CYRAMZA® (Ramucirumab), Cytarabine, Cytarabine Liposome, CYTOSAR-U® (Cytarabine), CYTOXAN® (Cyclophosphamide), Dabrafenib, Dacarbazine, DACOGEN® (Dccitabine), Dactinomycin, Daratumumab, DARZALEX® (Daratumumab), Dasatinib, Daunorubicin Hydrochloride, Daunorubicin Hydrochloride and Cytarabine Liposome, Decitabine, Defibrotide Sodium, DEFITEL1O® (Defibrotide Sodium), Degarelix, Denileukin Diftitox, Denosumab, DEPOCYT® (Cytarabine Liposome), Dexamethasone, Dexrazoxane Hydrochloride, Dinutuximab, Docetaxel, DOXIL® (Doxorubicin Hydrochloride Liposome), Doxorubicin Hydrochloride, Doxorubicin Hydrochloride Liposome, DOX-SL® (Doxorubicin Hydrochloride Liposome), DTIC-DOME® (Dacarbazine), Durvalumab, EFUDEX® (Fluorouracil— Topical), ELITEK® (Rasburicase), ELLENCE® (Epirubicin Hydrochloride), Elotuzumab, ELOXATIN® (Oxaliplatin), Eltrombopag Olamine, EMEND® (Aprepitant), EMPLICITI® (Elotuzumab), Enasidenib Mesylate, Enzalutamide, Epirubicin Hydrochloride , EPOCH, ERBITUX® (Cetuximab), Eribulin Mesylate, ERIVEDGE® (Vismodegib), Erlotinib Hydrochloride, ERWINAZE® (Asparaginase Erwinia chrysanthemi), ETHYOL® (Amifostine), Etopophos ETOPOPHOS® (Etoposide Phosphate), Etoposide, Etoposide Phosphate, EVACET® (Doxorubicin Hydrochloride Liposome), Everolimus, EVISTA® (Raloxifene Hydrochloride), EVOMELA® (Melphalan Hydrochloride), Exemestane, 5-FU® (Fluorouracil Injection), 5-FU® (Fluorouracil-Topical), FARESTON® (Toremifene), FARYDAK® (Panobinostat), FASLODEX® (Fulvestrant), FEC, FEMARA® (Letrozole), Filgrastim, FLUDARA® (Fludarabine Phosphate), Fludarabine Phosphate, FLUOROPLEX® (Fluorouracil— Topical), Fluorouracil Injection, Fluorouracil— Topical, Flutamide, FOLEX® (Methotrexate), FOLEX PFS® (Methotrexate), FOLFIRI, FOLFIRI-BEVACIZUMAB, FOLFIRI-CETUXIMAB, FOLFIRINOX, FOLFOX, FOLOTYN® (Pralatrexate), FU-LV,Attorney Docket No. 11820-039W01YAROl-HPRO Fulvestrant, GARDASIL® (Recombinant HPV Quadrivalent Vaccine), GARDASIL 9® (Recombinant HPV Nonavalent Vaccine), GAZYVA® (Obinutuzumab), Gefitinib, Gemcitabine Hydrochloride, GEMCITABINE-CISPLATIN, GEMCITABINEOXALIPLATIN, Gemtuzumab Ozogamicin, GEMZAR® (Gemcitabine Hydrochloride), GILOTRIF® (Afatinib Dimaleate), GLEEVEC® (Imatinib Mesylate), GLIADEL® (Carmustine Implant), GLIADEL WAFER® (Carmustine Implant), Glucarpidase, Goserelin Acetate, HALAVEN® (Eribulin Mesylate), HEMANGEOL® (Propranolol Hydrochloride), HERCEPTIN® (Trastuzumab), HPV Bivalent Vaccine, Recombinant, HPV Nonavalent Vaccine, Recombinant, HPV Quadrivalent Vaccine, Recombinant, HYCAMTIN® (Topotecan Hydrochloride), HYDREA® (Hydroxyurea), Hydroxyurea, Hyper-CVAD, IBRANCE® (Palbociclib), Ibritumomab Tiuxetan, Ibrutinib, ICE, ICLUSIG® (Ponatinib Hydrochloride), IDAMYCIN® (Idarubicin Hydrochloride), Idarubicin Hydrochloride, Idelalisib, IDHIFA® (Enasidenib Mesylate), IFEX® (Ifosfamide), Ifosfamide, IFOSFAMIDUM® (Ifosfamide), IL-2 (Aldesleukin), Imatinib Mesylate, IMBRUVICA® (Ibrutinib), IMFINZI® (Durvalumab), Imiquimod, IMLYGIC® (Talimogene Laherparepvec), INLYTA® (Axitinib), Inotuzumab Ozogamicin, Interferon Alfa-2b, Recombinant, Interleukin-2 (Aldesleukin), INTRON A® (Recombinant Interferon Alfa-2b), Iodine I 131 Tositumomab and Tositumomab, Ipilimumab, IRESSA® (Gefitinib), Irinotecan Hydrochloride, Irinotecan Hydrochloride Liposome, ISTODAX® (Romidepsin), Ixabepilone, Ixazomib Citrate, IXEMPRA® (Ixabepilone), JAKAFI® (Ruxolitinib Phosphate), JEB, JEVTANA® (Cabazitaxel), KADCYLA® (Ado-Trastuzumab Emtansine), KEOXIFENE® (Raloxifene Hydrochloride), KEPIVANCE® (Palifermin), KEYTRUDA® (Pembrolizumab), KISQALI® (Ribociclib), KYMRIAII® (Tisagenlecleucel), KYPROLIS® (Carfilzomib), Lanreotide Acetate, Lapatinib Di tosylate, LARTRUVO® (Olaratumab), Lenalidomide, Lenvatinib Mesylate, LENVIMA® (Lenvatinib Mesylate), Letrozole, Leucovorin Calcium, LEUKERAN® (Chlorambucil), Leuprolide Acetate, LEUSTATIN® (Cladribine), LEVULAN® (Aminolevulinic Acid), LINFOLIZIN® (Chlorambucil), LIPODOX® (Doxorubicin Hydrochloride Liposome), Lomustine, LONSURF® (Trifluridine and Tipiracil Hydrochloride), LUPRON® (Leuprolide Acetate), LUPRON DEPOT® (Leuprolide Acetate), LUPRON DEPOT-PED® (Leuprolide Acetate), LYNPARZA® (Olaparib), MARQIBO® (Vincristine Sulfate Liposome), MATULANE® (Procarbazine Hydrochloride), Mechlorethamine Hydrochloride, Megestrol Acetate, MEKINIST® (Trametinib), Melphalan, Melphalan Hydrochloride, Mercaptopurine, Mesna, MESNEX® (Mesna), METHAZOLASTONE® (Temozolomide), Methotrexate, METHOTREXATE LPF®Attorney Docket No. 11820-039W01YAROl-HPRO (Methotrexate), Methylnaltrexone Bromide, MEXATE® (Methotrexate), MEXATE-AQ® (Methotrexate), Midostaurin, Mitomycin C, Mitoxantrone Hydrochloride, MITOZYTREX® (Mitomycin C), MOPP, MOZOBIL® (Plerixafor), MUSTARGEN® (Mechlorethamine Hydrochloride) , MUTAMYCIN® (Mitomycin C), MYLERAN® (Busulfan), MYLOSAR® (Azacitidine), MYLOTARG® (Gemtuzumab Ozogamicin), NANOPARTICLE PACLITAXEL® (Paclitaxel Albumin-stabilized Nanoparticle Formulation), NAVELBINE® (Vinorelbine Tartrate), Necitumumab, Nelarabine, NEOSAR® (Cyclophosphamide), Neratinib Maleate, NERLYNX® (Neratinib Maleate), Netupitant and Palonosetron Hydrochloride, NEULASTA® (Pegfilgrastim), NEUPOGEN® (Filgrastim), NEXAVAR® (Sorafenib Tosylate), NILANDRON® (Nilutamide), Nilotinib, Nilutamide, NINLARO® (Ixazomib Citrate), Niraparib Tosylate Monohydrate, Nivolumab, NOLVADEX® (Tamoxifen Citrate), NPLATE® (Romiplostim), Obinutuzumab, ODOMZO® (Sonidegib), OEPA, Ofatumumab, OFF, Olaparib, Olaratumab, Omacetaxine Mepesuccinate, ONCASPAR® (Pcgaspargasc), Ondansetron Hydrochloride, ONIVYDE® (Irinotecan Hydrochloride Liposome), ONTAK® (Denileukin Diftitox), OPDIVO® (Nivolumab), OPPA, Osimertinib, Oxaliplatin, Paclitaxel, Paclitaxel Albumin- stabilized Nanoparticle Formulation, PAD, Palbociclib, Palifermin, Palonosetron Hydrochloride, Palonosetron Hydrochloride and Netupitant, Pamidronate Disodium, Panitumumab, Panobinostat, PARAPLAT® (Carboplatin), PARAPLATIN® (Carboplatin), Pazopanib Hydrochloride, PCV, FEB, Pegaspargase, Pegfilgrastim, Peginterferon Alfa-2b, PEG-INTRON® (Peginterferon Alfa- 2b), Pembrolizumab, Pemetrexed Disodium, PERJETA® (Pertuzumab), Pertuzumab, PLATINOL® (Cisplatin), PLATINOL-AQ® (Cisplatin), Plerixafor, Pomalidomide, POMALYST® (Pomalidomide), Ponatinib Hydrochloride, PORTRAZZA® (Necitumumab), Pralatrexate, Prednisone, Procarbazine Hydrochloride, PROLEUKIN® (Aldesleukin), PROLIA® (Denosumab), PROMACTA® (Eltrombopag Olamine), Propranolol Hydrochloride, PROVENGE® (Sipuleucel-T), PURINETHOL® (Mercaptopurine), PURIXAN® (Mercaptopurine), Radium 223 Dichloride, Raloxifene Hydrochloride, Ramucirumab, Rasburicase, R-CHOP, R-CVP, Recombinant Human Papillomavirus (HPV) Bivalent Vaccine, Recombinant Human Papillomavirus (HPV) Nonavalent Vaccine, Recombinant Human Papillomavirus (HPV) Quadrivalent Vaccine, Recombinant Interferon Alfa-2b, Regorafenib, RELISTOR® (Methylnaltrexone Bromide), R- EPOCH, REVLIMID® (Lenalidomide), RHEUMATREX® (Methotrexate), Ribocichb, R-ICE, RITUXAN® (Rituximab), RITUXAN HYCELA® (Rituximab and Hyaluronidase Human), Rituximab, Rituximab and , Hyaluronidase Human, ,Rolapitant Hydrochloride,Attorney Docket No. 11820-039W01YAROl-HPRO Romidepsin, Romiplostim, RUBIDOMYCIN® (Daunorubicin Hydrochloride), RUBRACA® (Rucaparib Camsylate), Rucaparib Camsylate, Ruxolitinib Phosphate, RYDAPT® (Midostaurin), Sclerosol Intrapleural Aerosol (Talc), Siltuximab, Sipuleucel-T, SOMATULINE DEPOT® (Lanreotide Acetate), Sonidegib, Sorafenib Tosylate, SPRYCEL® (Dasatinib), STANFORD V, Sterile Talc Powder (Talc), STERITALC® (Talc), STIVARGA® (Regorafenib), Sunitinib Malate, SUTENT® (Sunitinib Malate), SYLATRON® (Peginterferon Alfa- 2b), SYLVANT® (Siltuximab), Synribo SYNRIBO® (Omacetaxine Mepesuccinate), TABLOID® (Thioguanine), TAG, TAFINLAR® (Dabrafenib), TAGRISSO® (Osimertinib), Talc, Talimogene Laherparepvec, Tamoxifen Citrate, TARABINE PFS® (Cytarabine), TARCEVA® (Erlotinib Hydrochloride), TARGRETIN® (Bexarotene), TASIGNA® (Nilotinib), TAXOL® (Paclitaxel), TAXOTERE® (Docetaxel), TECENTRIQ® (Atezolizumab), TEMODAR® (Temozolomide), Temozolomide, Temsirolimus, Thalidomide, THALOMID® (Thalidomide), Thioguaninc, Thiotcpa, Tisagcnlccleuccl, TOLAK® (Fluorouracil-Topical), Topotecan Hydrochloride, Toremifene, TORISEL® (Temsirolimus), Tositumomab and Iodine 1 131 Tositumomab, TOTECT® (Dexrazoxane Hydrochloride), TPF, Trabectedin, Trametinib, Trastuzumab, TREANDA® (Bendamustine Hydrochloride), Trifluridine and Tipiracil Hydrochloride, TRISENOX® (Arsenic Trioxide), TYKERB® (Lapatinib Ditosylate) , UNITUXIN® (Dinutuximab), Uridine Triacetate, VAC, Vandetanib, VAMP, VARUBI® (Rolapitant Hydrochloride), VECTIBIX® (Panitumumab), VelP, VELBAN® (Vinblastine Sulfate), VELCADE® (Bortezomib), VELSAR® (Vinblastine Sulfate), Vemurafenib, VENCLEXTA® (Venetoclax), Venetoclax, VERZENIO® (Abemaciclib), VIADUR® (Leuprolide Acetate), VIDAZA® (Azacitidine), Vinblastine Sulfate, VINCASAR PFS® (Vincristine Sulfate), Vincristine Sulfate, Vincristine Sulfate Liposome, Vinorelbine Tartrate, VIP, Vismodegib, VISTOGARD® (Uridine Triacetate), VORAXAZE® (Glucarpidase), Vorinostat, VOTRIENT® (Pazopanib Hydrochloride), VYXEOS® (Daunorubicin Hydrochloride and Cytarabine Liposome), WELLCOVORIN® (Leucovorin Calcium), XALKORI® (Crizotinib), XELODA® (Capecitabine), XELIRI, XELOX, XGEVA® (Denosumab), XOFIGO® (Radium 223 Dichloride), XT ANDI® (Enzalutamide), YERVOY® (Ipilimumab), YONDELIS® (Trabectedin), ZALTRAP® (Ziv-Aflibercept), ZARXIO® (Filgrastim), ZEJULA® (Niraparib Tosylate Monohydrate), ZELBORAF® (Vemurafenib), ZEVALIN® (Ibritumomab Tiuxetan), ZINECARD® (Dexrazoxane Hydrochloride), Ziv-Aflibercept, ZOFRAN® (Ondansetron Hydrochloride), ZOLADEX® (Goserelin Acetate), Zoledronic Acid, ZOLINZA® (Vorinostat), ZOMETA®Attorney Docket No. 11820-039W01YAROl-HPRO (Zoledronic Acid), ZYDELIG® (Idelalisib), ZYKADIA® (Ceritinib), and / or ZYTIGA® (Abiraterone Acetate). The treatment methods can include or further include checkpoint inhibitors including, but are not limited to antibodies that block PD-1 (such as, for example, Nivolumab (BMS-936558 or MDX1106), pembrolizumab, cemiplimab , CT-011, MK-3475), PD-L1 (such as, for example, atezolizumab, avelumab, durvalumab, MDX-1105 (BMS-936559), MPDL3280A, or MSB0010718C), PD-L2 (such as, for example, rHIgM12B7), CTLA-4 (such as, for example, Ipilimumab (MDX-010), Tremelimumab (CP-675,206)), IDO, B7-H3 (such as, for example, MGA271, MGD009, omburtamab), B7-H4, B7-H3, T cell immunoreceptor with Ig and '•TV! domains (TIGIT)(such as, for example BMS-986207, OMP-313M32. MK-7684, AB-154, ASP-8374, MTIG7192A. or PVSRIPO), CD96, B and T-lyraphocyte attenuator (BTLA), V-dornain Ig suppressor of T cell activation (VISTA)(such as, for example, JNJ-61610588, CA-170), TIM3 (such as, for example, TSR-022, MBG453, Sym023, INCAGN2390, LY3321367, BMS-986258, SHR-1702, RO7121661), LAG-3 (such as, for example, BMS-986016, LAG525, MK-4280, REGN3767, TSR-033, BI754111, Sym022, FS118, MGD013, and Immutep).

[0119] The following description of exemplary embodiments provides non-limiting representative examples referencing numerals to particularly describe features and teachings of different exemplary aspects and exemplary embodiments of the present disclosure. The exemplary embodiments described should be recognized as capable of implementation separately, or in combination, with other exemplary embodiments from the description of the exemplary embodiments. A person of ordinary skill in the art reviewing the description of the exemplary embodiments should be able to learn and understand the different described aspects of the present disclosure. The description of the exemplary embodiments should facilitate understanding of the exemplary embodiments of the present disclosure to such an extent that other implementations, not specifically covered but within the knowledge of a person of skill in the art having read the description of embodiments, would be understood to be consistent with an application of the exemplary embodiments of the present disclosure.

[0120] To overcome the challenges detailed above, the exemplary systems, methods, and computer-accessible medium according to the exemplary embodiments of the present disclosure provide an exhaustive search engine to comprehensively profile (e.g., 11) classes of genetic aberrations from RNA-Seq, encompassing all known tumor-specific events. The exemplary systems, methods, and computer-accessible medium according to the exemplary embodiments of the present disclosure can generate a cancer-specific molecular catalogue of (e.g., 11) classes of molecular alterations across (e.g., 21) histologies, and then can utilize thisAttorney Docket No. 11820-039W01YAROl-HPRO catalogue as a search space to interrogate (e.g., 1,564) immunopeptidomics datasets, revealing a multitude of actionable immunotherapy targets across cancers.

[0121] T-cell-based immunotherapies have revolutionized cancer treatment, offering cures to subsets of patients who have historically had limited effective therapeutic options. Treatments such as CAR T cell therapies have shown remarkable efficacy, particularly in hematologic malignancies such as acute lymphoblastic leukemia and non- Hodgkin lymphoma. While targeting lineage specific proteins shared between malignant and healthy B cells has proven clinically manageable in blood cancers as patients can tolerate the loss of healthy B cells, such on-target off-tumor effects are unacceptable in solid tumors, where off-tumor reactivity to essential healthy tissues could be catastrophic. This limitation necessitates the identification of antigens with substantial therapeutic windows. Currently, targeting strategies predominantly focus on membrane proteins, which, while promising, often lack sufficient tumor specificity, contributing to on-target, off-tumor side effects. Indeed, a recent report suggests we may be approaching the saturation point for identifying optimal membrane targets for CAR-T therapy, indicating a critical need for innovative targeting strategies that can distinguish between malignant and healthy tissues more effectively.

[0122] Human Leukocyte Antigen (HLA) molecules present a snapshot of the cellular proteome on the membrane of tumor cells, exposing potential tumor-specific antigens on the cell surface. Multiple therapeutic strategies that exert their therapeutic effects through peptides presented on HLA have delivered curative responses in the clinic, including immune checkpoint inhibitors (ICIs), adaptive transfer of Tumor Infiltrating Lymphocytes (TILs), TCR therapies as well as recent complete response from neoantigen vaccines.Immunotherapies such as ICIs and TILs often rely on high mutational burden, therefore the majority of curative responses to these therapies are applicable only to a limited number of highly mutated tumors. New approaches, such as peptide-centric chimeric antigen receptors (PC-CARs), that can enable the targeting of any peptide on HLA, expanding the targeting of intracellular antigens to low-to-medium mutational tumors, have been developed. Central to the efficacy of such immunotherapies is the identification of tumor-specific HLA-presented peptides (pHLAs), which can be derived from a broad range of genetic aberrations including both tumor dependency genes that underpin the biology of various cancers or accompanying passenger mutations. Elucidating the landscape of pHLAs derived from these various cancer processes could reveal additional tumor vulnerabilities.

[0123] The following description of exemplary embodiments provides non-limiting representative examples referencing numerals to particularly describe features and teachingsAttorney Docket No. 11820-039W01YAROl-HPRO of different aspects of the present disclosure. The exemplary embodiments described should be recognized as being capable of implementation separately, or in combination, with other exemplary embodiments from the description of the exemplary embodiments. A person of ordinary skill in the art reviewing the description of the exemplary embodiments should be able to learn and understand the different described aspects of the present disclosure. The description of the exemplary embodiments should facilitate understanding of the invention to such an extent that other implementations, not specifically covered but within the knowledge of a person of skill in the art having read the description of embodiments, would be understood to be consistent with an application of the exemplary embodiments of the present disclosure.

[0124] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can provide an in silico pipeline to generate and screen AIR-CAR constructs, enabled by combining multiple recent advances in generative Al. For example, a first-generation PC-CAR development process relied on empirical panning for binders and rational selection for those with desired properties, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can provide the rational design of these binders through generative Al tools, followed by in silico cross-reactivity testing which can result in therapies with superior safety and functional profiles. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can utilize a combination of experimental and computational method to transform this manual process into an entirely digital process, thereby realizing new computer functionality that did not previously exist (see Figure IB). The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can run this pipeline to generate an initial pool of AIR-CARs for a validated clinical target PHOX2B.

[0125] To generate AIR-CARs, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can first generate structural scaffolds that conform to the unique structure of a peptide as presented on HLA using a state-of-the-art deep-leaming-empowered structural modeler (e.g., modelling tool), which utilizes a target structure with its hotspots (potential binding residues) to generate an optimal binding interface scaffold. Next, to generate amino acid sequences to interact with the target epitope, these binder backbones can be subjected to a deep-learning-based protein sequence design tool of exemplary systems, methods and computer-accessible mediumAttorney Docket No. 11820-039W01YAROl-HPRO according to the exemplary embodiments of the present disclosure that can generate a diverse set of binder sequences, optimizing for proteins that fold into the scaffold structure and generate strong interactions with the defined epitope. (See, e.g., Ref. 2). To perform orthogonal validation of these computationally generated binders, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can utilize the AlphaFold2 (see, e.g., Ref. 3) to predict protein-protein interactions (see, e.g., Ref. 4). for large-scale multimer structure predictions (see, e.g., Ref.5). The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can perform a high-throughput in silico screening to evaluate each AIRCAR binder for contact with peptide and MHC, calculating a score based on possible molecular interactions within a 6 A distance. Binders with suboptimal contact with PH0X2B or extensive contact with HLA can be removed by the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure, and those with selective contacts to PH0X2B can be selected. Applying this pipeline, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can generate e.g., 100,000 discrete binders potentially able to recognize PHOX2B presented on HLA*A24:02. Additionally, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can generate 100,000 discrete binders potentially able to recognize ORF2 presented on HLA*A24:02. The binders generated by exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can be multi -helix bundles, a structural motif known for its robust folding capability and stability. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can process e.g., 40,000 binders through the pipeline and can identify excellent candidates that meet all stringent orthogonal requirements. All validated binders may exhibit multiple hydrogen bonds and salt bridges the PHOX2B pMHC target. As binder development remains the major bottleneck in PC CAR development, the AIR-CAR approach of exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure has the potential to transform the immunotherapy space, enabling personalized applications of immunotherapies.

[0126] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can provide the first known use of generative Al to design PC-CAR T cells. An exemplary implementation of the exemplaryAttorney Docket No. 11820-039W01YAROl-HPRO systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can transform the development of immunotherapy and pave the way for personalized therapies.

[0127] Example 1 : Select the Top 300 Binders out of 100,000 for Experimental Screening.

[0128] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can build a pipeline that combines state-of-art protein design tools to generate, diversify, and prioritize de novo binders that are predicted to bind with pMHC targets, and can successfully generate e.g., 100,000 binders. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can evaluate promising candidates using a generated pipeline, and binders that meet converge on the same structures by both pipelines can be shortlisted for cross-reactive tests. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can implement a stringent cross-reactive to recapitulate the experimental PC-CAR preclinical screening (see Figures 1 A and IB), using AlphaFold2 and RosettaFold on a panel of peptides predicted by the cross-reactivity algorithm sCRAP (see, e.g., Ref. 6), removing unwanted binding with other peptides presented by HLA to ensure safety against normal tissue.Remarkably, some binders show better binding structure to peptides and less cross-reactivity compared to a lead clinical CAR 10LH. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can select e.g., 300 of the top scoring binders to test using high-throughput CAR library screening technologies for anti-tumor cytotoxicity and cross-reactivity of each clone.

[0129] Example 2: Design a Library of AIR-CARs with Diverse Linkers

[0130] The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can synthesize selected binders from through Twist Bioscience and can be cloned into the CAR vector for lentiviral production. Using this strategy, the binder size can be around 100-250 aa, much smaller than conventional scFvs. As smaller extra-cytoplasmic domains may have different size and geometrical constraints as compared to PC-CARs, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can design a linker library to incorporate different types of linker combinations between the binders and hinge domain of the CAR. This can include various combinations of the flexible GGGGS linker, the rigid EAAAK linker, and multiple domains immunoglobulinAttorney Docket No. 11820-039W01YAROl-HPRO (Ig) domains which could enhance the ability of AIR-CAR to effectively interact with target antigens. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can synthesize a library of linkers and clone these into the library of e.g., 300 AIR-CARs, creating a library in which each AIRCAR is represented with each linker combination. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can screen this pooled library for the optimal combination of AIR-CAR receptor and linker.

[0131] Example 3 Evaluate In Vitro the Binders' Binding Ability and Specificity

[0132] To screen for functional AIR-CARs, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can first transduce the generated library into primary T cells and isolate clones from the pooled AIRCAR library that bind to PHOX2B pMHC tetramer by flow sorting PHOX2B+ cells and those that do not bind predicted cross-reactive pMHC. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can then screen for functionality and cross-reactivity using the high throughput Lightcast system. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can use antigenexpressing target cells (SK-N-AS cells) in the single-cell killing assay, a well-characterized target of the 10LH CAR. (See, e.g., Ref. 6). The most promising AIR-CAR constructs showing the killing ability on SK-N-AS cells can be isolated by exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure for further in vitro validation. To screen for cross-reactivity, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can use a panel of cross-reactive peptides presented on HLA labeled with APC. (See, e.g., Ref. 6). To check the specificity of AIRCAR T, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can conduct a binding specificity test by co-culturing AIR-CAR with HLA-matched cell lines including SW620, HEPG2 and KATOIII. Additionally, specificity can be confirmed by performing alanine scanning of target or decoy peptide and then binding and killing assays can be repeated.

[0133] Exemplary AIR-CARs generation methods

[0134] Artificial Intelligence-generated Receptors (AIR) - Chimeric Antigen Receptors (CARs) are in silico Peptide Centric (PC)-CARs designed entirely in silico and meant toAttorney Docket No. 11820-039W01YAROl-HPRO recognize and bind peptide presented on human leukocyte antigens (HLA) presented on target cells. Upon binding, AIR-CARs are able to activate T cells and redirect their cytotoxic activity against the target-expressing cell, thus killing them. To design AIR-CARs, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can generate structural scaffolds that conform to the unique structure of a peptide as presented on HLA, using a state-of-the-art deep-leaming-empowered structural modelling tool, RFDiffusion. (See, e.g., Ref. 1). RFDiffusion utilizes a target structure with its hotspots (potential binding residues) to generate an optimal binding interface scaffold. Next, to generate amino acid sequences to interact with the target epitope, these binder backbones can be subjected to ProteinMPNN, a deep-leaming-based protein sequence design tool that generates a diverse set of binder sequences. These generated binders are optimized to fold into the scaffold structure created by RFDiffusion, thus allowing the creation of strong interactions with the defined epitope. (See, e.g., Ref. 2). To validate these computationally generated binders, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can utilize the AlphaFold2 (see, e.g., Ref. 3), which can predict protein-protein interactions for large-scale multimer structure predictions (see, e.g., Ref. 7). Application of this pipeline allows exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure to design e.g., 100.000 discrete binders potentially able to recognize the target peptide PHOX2B presented on HLA*A24:02.

[0135] To select potential AIR-CAR that can form specific binding with the target protein from the e.g., 100,000 generated binder structures, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can use three different in parallel high-throughput screening methods. First, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can utilize a built-in scoring system of AlphaFold2, PAE score (Predicted Aligned Error) to measure the confidence in the relative positioning of the binder and target which estimates the binding structure by distance. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can also use a peptide / HLA bond scoring system to evaluate the quality and selectivity of the interface, by quantifying possible bonds between the binder and the target peptide and calculating peptide-contact score and an IILA-contact score for each AIR-CAR binder to prioritize binders that have a peptide-contact score at or above a predetermined cutoff (e.g., > 25) while maintaining limited contact with HLA. TheAttorney Docket No. 11820-039WG1YAROl-HPRO exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can also perform a high-throughput in silico screening to evaluate each AIR-CAR binder for contact with peptide and HLA, calculating the binding area between the target peptide and the binder, thus refining the binding complex to its lowest energy state. Binders with suboptimal contact with PH0X2B or extensive contact with HLA can be removed, and those with selective contacts with PH0X2B can be selected.

[0136] Exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can select the highest-ranked candidates from each method that show specific target-binding interactions. Selected binders can be validated by exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure in vitro by performing binding assay through PHOX2B -tetramer staining and killing assay by cultivating AIR-expressing T cells with the target cells. Of the 16 candidates tested, the binder 637_1 showed highest binding and killing efficiency against target cells. Nonetheless, expression of 637_1 and binding of the target peptide was not optimal, thus exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can further improve 637_1 efficiency, by acting on two fronts: i) modification of the binder through rational mutagenesis; ii) improvement of the CAR domains (hinge, transmembrane and co-stimulatory domains). To modify 637-1, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can design point mutations using the in-silico tool Molecular Operating Environment (MOE). (See, e.g., Ref. 8). These mutations can increase binding affinity towards the PIIOX2B peptide and / or stabilize the protein structure. For the improvement of the CAR portion, the exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can provide different hinges (IgG4, CD28 with or without a further linker, and CD8), transmembrane and costimulatory domains (CD28 and CD8). From preliminary results, the mutation S141Q on 637_ 1 , the hinge IgG4 and the absence of the linker between the hinge and the transmembrane domain bear the most promising results, as they seem to increase both expression and binding of PHOX2B tetramers. The exemplary systems, methods and computer-accessible medium according to the exemplary embodiments of the present disclosure can validate these results by performing a killing assay on target and non-target cells to functionally assess which modification improve 637_1 killing efficiency.Attorney Docket No. 11820-039W01YAROl-HPRO

[0137] Figures 1A and IB are a set of illustrations providing an exemplary AIR-CAR workflow to recapitulate PC-CAR development process according to an exemplary embodiment of the present disclosure.

[0138] Figure 1C shows an example of a model AIR-CAR binder to PHOX2B pMHC. In silico generated model AIR-CAR binder demonstrates superior binding to PHOX2B pMHC than a lead clinical PC-CAR, 10LH.

[0139] Figure ID shows a model of an identified AIR-CAR binder sequence, AIR637, bound to the PHOX2B-A2402 HLA-presented peptide, highlighting the molecular interactions that facilitate binding. Amino acid residues shown in blue are residues in the binder sequence that make contact with the peptide. Amino acid residues in magenta are residues in the HLA-presented peptide that make contact with the binder sequence. Amino acid residues shown in gray are amino acids in the HLA molecule that make contact with the binder sequence. Yellow dashed lines represent hydrogen bonds, green dashed lines represent cation-n interactions and 7t-7t stacking. Red dashed lines represent salt bridges, orange dashed lines represent hydrophobic interactions.

[0140] Example Computing Device

[0141] It should be appreciated that the logical operations described herein with respect to the various figures may be implemented (1) as a sequence of computer-implemented acts or program modules (i.e., software) running on a computing device (e.g., the computing device described in Figure IE), (2) as interconnected machine logic circuits or circuit modules (i.e., hardware) within the computing device and / or (3) a combination of software and hardware of the computing device. Thus, the logical operations discussed herein are not limited to any specific combination of hardware and software. The implementation is a matter of choice dependent on the performance and other requirements of the computing device. Accordingly, the logical operations described herein are referred to variously as operations, structural devices, acts, or modules. These operations, structural devices, acts and modules may be implemented in software, in firmware, in special purpose digital logic, and any combination thereof. It should also be appreciated that more or fewer operations may be performed than shown in the figures and described herein. These operations may also be performed in a different order than those described herein.

[0142] Referring to Figure IE, an example computing device 1000 upon which embodiments of the invention may be implemented is illustrated. This disclosure contemplates that the controller(s) for operating the flexure elements and / or imaging apparatus can be implemented using computing device 1000. It should be understood that theAttorney Docket No. 11820-039W01YAROl-HPRO example computing device 1000 is only one example of a suitable computing environment upon which embodiments of the invention may be implemented. Optionally, the computing device 1000 can be a well-known computing system including, but not limited to, personal computers, servers, handheld or laptop devices, multiprocessor systems, microprocessorbased systems, network personal computers (PCs), minicomputers, mainframe computers, embedded systems, and / or distributed computing environments including a plurality of any of the above systems or devices. Distributed computing environments enable remote computing devices, which are connected to a communication network or other data transmission medium, to perform various tasks. In the distributed computing environment, the program modules, applications, and other data may be stored on local and / or remote computer storage media.

[0143] In its most basic configuration, computing device 1000 typically includes at least one processing unit 1006 and system memory 1004. Depending on the exact configuration and type of computing device, system memory 1004 may be volatile (such as random access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two. This most basic configuration is illustrated in Figure IF by dashed line 1002. The processing unit 1006 may be a standard programmable processor that performs arithmetic and logic operations necessary for operation of the computing device 1000. The computing device 1000 may also include a bus or other communication mechanism for communicating information among various components of the computing device 1000.

[0144] Computing device 1000 may have additional features / functionality. For example, computing device 1000 may include additional storage such as removable storage 1008 and non-removable storage 1010 including, but not limited to, magnetic or optical disks or tapes. Computing device 1000 may also contain network connection(s) 1016 that allow the device to communicate with other devices. Computing device 1000 may also have input device(s) 1014 such as a keyboard, mouse, touch screen, etc. Output device(s) 1012 such as a display, speakers, printer, etc. may also be included. The additional devices may be connected to the bus in order to facilitate communication of data among the components of the computing device 1000. All these devices are well known in the art and need not be discussed at length here.

[0145] The processing unit 1006 may be configured to execute program code encoded in tangible, computer-readable media. Tangible, computer-readable media refers to any media that is capable of providing data that causes the computing device 1000 (i.e., a machine) toAttorney Docket No. 11820-039W01YAROl-HPRO operate in a particular fashion. Various computer- readable media may be utilized to provide instructions to the processing unit 1006 for execution. Example tangible, computer-readable media may include, but is not limited to, volatile media, non-volatile media, removable media and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. System memory 1004, removable storage 1008, and non-removable storage 1010 are all examples of tangible, computer storage media. Example tangible, computer-readable recording media include, but are not limited to, an integrated circuit (e.g., field-programmable gate array or application-specific IC), a hard disk, an optical disk, a magnetooptical disk, a floppy disk, a magnetic tape, a holographic storage medium, a solid-state device, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices.

[0146] In an example implementation, the processing unit 1006 may execute program code stored in the system memory 1004. Eor example, the bus may cany data to the system memory 1004, from which the processing unit 1006 receives and executes instructions. The data received by the system memory 1004 may optionally be stored on the removable storage 1008 or the non-removable storage 1010 before or after execution by the processing unit 1006.

[0147] It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, where appropriate, with a combination thereof. Thus, the methods and apparatuses of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium wherein, when the program code is loaded into and executed by a machine, such as a computing device, the machine becomes an apparatus for practicing the presently disclosed subject matter. In the case of program code execution on programmable computers, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. One or more programs may implement or utilize the processes described in connection with the presently disclosed subject matter, e.g., through the use of an application programming interface (API), reusable controls, or the like. Such programs may be implemented in a high-level proceduralAttorney Docket No. 11820-039W01YAROl-HPRO or object-oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language and it may be combined with hardware implementations.

[0148] Example AIR-CAR Design System

[0149] Figure 2 is a diagram of an example AIR-CAR design system 200 configured to facilitate the development (e.g., design, modification, or the like) of AIR-CARs for therapeutics targeting intracellular tumor driver via pMHC complexes (e.g., class I MHC). The proposed system combines state-of-art protein design tools to generate, diversify, and prioritize de novo binders that are predicted to bind with pMHC targets, and can successfully generate e.g., 100,000 binders. In one implementation, the system 200 can take sequence data 202 as an input and generate a 3D structure of a CAR target and select (e.g., generate, identify) AIR-CAR(s), for example, a subset of binder sequences that can be used to determine one or more immunotherapy targets for treating cancer. As shown the system 200 can obtain data (e.g., data sets) from one or more data entities 221. The one or more data entities 221 can be or comprise databases or datasets associated with research institutions, data management providers and / or the like and can include B-cell antibody libraries, patient data, or the like.

[0150] As depicted in Figure 2, the system 200 comprises one or more machine learning models 205 including one or more deep-learning based protein sequencers 207 (e.g., ProteinMPNN), one or more structural modeling components 209 (e.g., AlphaFold2, RFdiffusion), one or more deep-learning modelers 211 including one or more energy minimization modelers 213 (e.g., Rosetta: fastrelax), and one or more scoring components 215. The machine learning model(s) 205 can be or comprise deep-learning models configured to predict and generate three-dimensional models of protein structures from sequence data.

[0151] The system 200 also includes a modeling component 214, filtering component 216, analyzing component 218, and a visualization engine 220. An example energy minimization modelers 213 can be configured to optimize and / or refine protein structures by identifying low-energy conformations in order to eliminate unrealistic (high-energy) conformations in predicted protein structure.

[0152] This disclosure contemplates that the deep-leaming based protein sequencer(s) 207 can be or comprise ProteinMPNN, Ablang, ThermoMPNN, AbMPNN, SolMPNN, combinations thereof, and / or the like. In various embodiments, the modeling component 214 can be or comprise SimpleFold, AlphaFold2, AlphaFold3, ESMFold, RosettaFold,Attorney Docket No. 11820-039W01YAROl-HPRO RosettaFold2, RosettaFold3, AntiFold, EquiFold, AlphaFlow, ABodyBuilderl, ABodyBuilder2, ABodyBuilder3, combinations thereof, and / or the like. Additionally, the system 200 / machine learning models 205 can be or comprise full pipelines such as Boltz- 1, Boltz-2, BindCraft, AlphaProteo, CHAI-1, CHAI-2, combinations thereof, and / or the like.

[0153] The visualization engine 220 can be configured to generate user interface data for display to an end user (e.g., a report, summary, and / or data object(s) describing genetic aberrations and / or a list of antigens). In some implementations, output data (e.g., immunotherapy targets or additional data) can be used as an input to a search engine 225 to identify tumor-specific antigens. In some embodiments, the analyzing component 218 can process, pre-process and / or transform sequence data 202 for downstream operations and may also select binders in different steps of the method 300 described in connection with Figure 3, for example. The filtering component 216 can perform cross-reactivity tests and eliminate one or more binders from a set of binder sequences based on results of the cross-reactivity tests (e.g., using decoy binders) and the modeling component 214 can supplement or enhance operations of the machine-learning models 205 (see method 300 described in connection with Figure 3). In some implementations, at least a portion of the received / generated data can be transferred to other computing entities or devices as needed to increase processing speed or efficiently handle large amounts of data. Workflow allocation can be optimized based on time and memory constraints while considering internal dependencies.

[0154] Example Method

[0155] Referring now to Figure 3, a flowchart of an example computer-implemented method 300 for modifying, designing, and / or redesigning AIR-CARs is provided. This disclosure contemplates that the method 300 can be independent of HLA allele type.

[0156] In some implementations, the method 300 can be performed by a processing circuitry (for example, but not limited to, an application-specific integrated circuit (ASIC), or a central processing unit (CPU)). In some examples, the processing circuitry may be electrically coupled to and / or in electronic communication with other circuitries of an example computing device, such as, but not limited to, the example computing device 1000 described above in connection with Figure IE. In some examples, embodiments may take the form of a computer program product on a non-transitory computer-readable storage medium storing computer-readable program instruction (e.g., computer software). Any suitable computer-readable storage medium may be utilized, including non-transitory hard disks, CD-ROMs, flash memory, optical storage devices, or magnetic storage devices. This disclosure contemplates that some or all of the steps / operations below can be implemented usingAttorney Docket No. 11820-039W01YAROl-HPRO machine learning models and artificial intelligence-based techniques, such as, but not limited to, deep learning models as described in more detail below.

[0157] At step / operation 302, the method 300 includes generating with a deep-learning-empowered structural modeler, a three-dimensional (3D) structure of a CAR target. In some implementations, the 3D structure of the CAR target comprises at least one of a peptide, an HLA molecule, an HLA-presented peptide, and a beta-2-microglobulin (beta-2-m). The CAR target can comprise a predicted computer structure model or an experimentally solved structure model.

[0158] At step / operation 304, the method 300 includes generating an optimal binding interface scaffold conforming to a unique peptide structure.

[0159] At step / operation 306, the method 300 includes generating, using a deep-leaming-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope.

[0160] At step / operation 308, the method 300 includes predicting, using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets.

[0161] At step / operation 310, the method 300 includes evaluating using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico. In some implementations, evaluating the selectivity of the binder sequences comprises performing, by the at least one processor and using one or more decoy peptides, a cross-reactivity test and eliminating one or more binders from the set of binder sequences based on results of the cross-reactivity test. The one or more decoy peptides can have binding residues within a predefined distance threshold of at least one binder sequence.

[0162] At step / operation 312, the method 300 includes selecting a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target. The interaction score may be or comprise a Predicted Aligned Error (PAE) score. By way of example, a binder that meets an interaction score may have a PAE score within a predefined distance threshold (e.g., between 0 and 6 Angstroms (A)). In some examples, evaluating the selectivity of the binder sequences comprises determining a specificity score for each binder sequence and selecting the subset of binder sequences based on the determined specificity scores.Attorney Docket No. 11820-039W01YAROl-HPRO

[0163] Optionally, at step / operation 314, the method 300 includes performing in vitro testing on the subset of binder sequences to validate each binder’s binding ability and specificity.

[0164] Optionally, at step / operation 316, the method 300 includes treating cancer, such as neuroblastoma or glioblastoma, in a subject using the generated AIR-CAR. For example, step / operation 316 can include administering to the subject a substance, comprising the AIRCAR generating using the proposed method 300. The example substance can be an AIR-CAR immune cell. In some implementations, the immune cell can be or comprise a T cell, B cell, natural killer (NK) cell, NK T cell, or macrophage.

[0165] Figure 4 shows an example AIR-CAR pipeline overview / method 400 for CAR-target, ORF2-HLA-A*24:02, in accordance with certain embodiments of the present disclosure.

[0166] At step / operation 402, the method 400 includes generating structural scaffolds that conform to the unique structure of a peptide as presented on HLA using a dccp-lcarning-empowered structural modeler, which utilizes a target structure with its hotspots (potential binding residues) to generate an optimal binding interface scaffold.

[0167] At step / operation 404, the method 400 includes generating amino acid sequences to interact with the target epitope, these binder backbones can be subjected to a deep-leaming-based protein sequence design tool that can generate a diverse set of binder sequences, optimizing for proteins that fold into the scaffold structure and generate strong interactions with the defined epitope.

[0168] At step / operation 406, the method 400 includes performing orthogonal validation of these computationally generated binders, e.g., using AlphaFold2 to predict protein-protein interactions for large-scale multimer structure predictions.

[0169] At step / operation 408, the method 400 includes performing a first a high-throughput in silico screening to evaluate each AIR-CAR binder for contact with peptide and MHC. Step / operation 408 can include utilizing a built-in scoring system of AlphaFold2, PAE score (Predicted Aligned Error) to measure the confidence in the relative positioning of the binder and target which estimates the binding structure by distance.

[0170] At step / operation 410, the method 400 includes calculating the binding area between the target peptide and the binder and refining the binding complex to its lowest energy state.

[0171] At step / operation 412, the method 400 includes performing a second high-throughput in silico screening to evaluate each AIR-CAR binder for contact with peptide andAttorney Docket No. 11820-039W01YAROl-HPRO HLA, calculating the binding area between the target peptide and the binder, thus refining the binding complex to its lowest energy state. Binders with suboptimal contact with PH0X2B or extensive contact with HLA can be removed, and those with selective contacts with PH0X2B can be selected.

[0172] At step / operation 414, the method 400 optionally includes performing a crossreactivity test with the identified binder sequence and one or more decoy peptides.

[0173] At step / operation 416, the method 400 optionally includes in vitro and in vivo characterization of binder sequences that had low cross-reactivity with decoy peptides.

[0174] Figure 5A is an example of specificity of a representative AIR-CAR. in vitro killing assays of a representative AIR-CAR, AIR-CAR 397 show AIR-CAR 397 kills cells presenting ZNF737, but not ARHGAP18 and MEIS2 peptides. The table shows that all peptides have similar hotspots. Orange-labeled residues from ORF2 are the predicted hotspots bound by 397. The bottom table in Figure 5 A shows the sequence alignment between ORF2 and decoy peptides: in particular, ARHGAP18 shares similar / samc hotspot residues on positions 4, 5, 6, 8; MEIS2 shares similar / same hotspot residues on positions 4, 5, 6; ZNF737 shares similar / same hotspot residues on positions 4, 5, 6, 7, 8; PHOX2B is not a decoy peptide, but a HLA-A2402 presenting peptide set as a control. The left figure in Figure 5A shows a binding assay empty HLA-A2402 tetramer pulsed by decoy peptide. The right figure shows the killing assay against T2-A2402 with the decoy peptide pulsed. Binding and killing efficacy is correlated: better binding (higher MFI on left figure) means better killing efficacy (right figure).

[0175] Figure 5B is an example of the effect of mutations in the CAR target on the ability of a representative AIR-CAR to bind and kill. Figure 5B shows an Ala scan on ORF2 peptide, including K4A, V5A, Y7A, and R8A (e.g. K4A means lysine on position 4 of the ORF2 peptide is mutated to alanine). The left figure in Figure 5B shows binding assay empty HLA-A2402 tetramer pulsed by Ala-mutated peptides. The right figure in Figure 5B shows the killing assay against T2-A2402 with Ala-mutated peptides pulsed.

[0176] Figure 5C is an example of the characterization of two representative AIR-CARs, AIR-CAR 637 and AIR-CAR 397. Each tested AIR-CAR demonstrates on-target binding. Panel (A) in Figure 5C shows Peptide-HLA (pHLA) tetramer staining of non-transduced T-cells (NT), clinical PC-CAR 10LH and AIR-CAR 637 using the target peptide (PHOX2B) and non-target peptides (MY07B and PIIOX2B-R6A). As illustrated, AIR-CAR 637 shows better on-target binding than clinical PC-CAR 10LH. Panel (B) of Figure 5C shows pHLA tetramer staining of NT, clinical PC-CAR 10LH and AIR-CAR 397 using the target peptideAttorney Docket No. 11820-039W01YAROl-HPRO (0RF2) and non-target peptides (CHRNA3 and PH0X2B). AIR-CAR 397 shows on-target binding with only the 0RF2 target peptide. The HLA of all pHLA tetramers are HLA-A*24:02.

[0177] Figure 5D is an example of in vitro validation confirming on-target killing ability of a representative AIR-CAR, AIR-CAR 637. Panel (A) in Figure 5D shows cytotoxicity assay evaluating the killing ability of non-transduced T-cells (NT, open symbols), clinical PC-CAR 10RH (grey circle symbols) and AIR-CAR 637 (black rectangle symbols) on target cells SK-N-AS, which naturally expresses the PHOX2B-HRA-A*24:02. AIR-CAR demonstrates similar on-target PHOX2B killing ability as 10LH. Panel (B) in Figure 5D shows cytotoxicity assay evaluating the killing ability of NT (open symbols) and AIR-CAR 637 (filled symbols) on target cells expressing the PHOX2B-HLA-A*24:02 target (circle symbols) or a single-point mutation PHOX2B-R6A-HLA-A*24:()2 (rectangle symbols). AIRCAR demonstrates on-target PHOX2B but no off-target PHOX2B-R6A killing.Conflucncy(%) in Panels (A) and (B) was normalized on the NT.

[0178] Artificial Intelligence and Machine Teaming

[0179] Embodiments of the present disclosure utilize machine learning models and artificial intelligence for various data processing operations to, for example, refine results to enhance sensitivity and to enhance the accuracy of subsequent searches.

[0180] The term “artificial intelligence” is defined herein to include any technique that enables one or more computing devices or comping systems (i.e., a machine) to mimic human intelligence. Artificial intelligence (Al) includes, but is not limited to, knowledge bases, machine learning, representation learning, and deep learning. The term “machine learning” is defined herein to be a subset of Al that enables a machine to acquire knowledge by extracting patterns from raw data. Machine learning techniques include, but are not limited to, logistic regression, support vector machines (SVMs), decision trees, Naive Bayes classifiers, and artificial neural networks. The term “representation learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, or classification from raw data. Representation learning techniques include, but are not limited to, autoencoders. The term “deep learning” is defined herein to be a subset of machine learning that that enables a machine to automatically discover representations needed for feature detection, prediction, classification, etc. using layers of processing. Deep learning techniques include, but are not limited to, artificial neural network or multilayer perceptron (MLP).Attorney Docket No. 11820-039W01YAROl-HPRO

[0181] Machine learning models include supervised, semi-supervised, and unsupervised learning models. In a supervised learning model, the model learns a function that maps an input (also known as feature or features) to an output (also known as target or targets) during training with a labeled data set (or dataset). In an unsupervised learning model, the model learns patterns (e.g., structure, distribution, etc.) within an unlabeled data set. In a semisupervised model, the model learns a function that maps an input (also known as feature or features) to an output (also known as target or target) during training with both labeled and unlabeled data.

[0182] An artificial neural network (ANN) is a computing system including a plurality of interconnected neurons (e.g., also referred to as “nodes”). This disclosure contemplates that the nodes can be implemented using a computing device (e.g., a processing unit and memory as described herein). The nodes can be arranged in a plurality of layers such as input layer, output layer, and optionally one or more hidden layers. An ANN having hidden layers can be referred to as deep neural network or multilayer perceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, each layer is made of a plurality of nodes, where each node is connected to all nodes in the previous layer. The nodes in a given layer are not interconnected with one another, i.e., the nodes in a given layer function independently of one another. As used herein, nodes in the input layer receive data from outside of the ANN, nodes in the hidden layer(s) modify the data between the input and output layers, and nodes in the output layer provide the results. Each node is configured to receive an input, implement an activation function (e.g., binary step, linear, sigmoid, tanH, or rectified linear unit (ReLU) function), and provide an output in accordance with the activation function. Additionally, each node is associated with a respective weight. ANNs are trained with a dataset to maximize or minimize an objective function. In some implementations, the objective function is a cost function, which is a measure of the ANN’S performance (e.g., error such as LI or L2 loss) during training, and the training algorithm tunes the node weights and / or bias to minimize the cost function. This disclosure contemplates that any algorithm that finds the maximum or minimum of the objective function can be used for training the ANN. Training algorithms for ANNs include, but are not limited to, backpropagation. It should be understood that an artificial neural network is provided only as an example machine learning model. This disclosure contemplates that the machine learning model can be any supervised learning model, semi-supervised learning model, or unsupervised learning model. Optionally, the machine learning model is a deepAttorney Docket No. 11820-039W01YAROl-HPRO learning model. Machine learning models are known in the art and are therefore not described in further detail herein.

[0183] A convolutional neural network ( CNN) is a type of deep neural network that has been applied, for example, to image analysis applications. Unlike a traditional neural networks, each layer in a CNN has a plurality of nodes arranged in three dimensions (width, height, depth). CNNs can include different types of layers, e.g., convolutional, pooling, and fully-connected (also referred to herein as “dense”) layers. A convolutional layer includes a set of filters and performs the bulk of the computations. A pooling layer is optionally inserted between convolutional layers to reduce the computational power and / or control overfitting (e.g., by downsampling). A fully-connected layer includes neurons, where each neuron is connected to all of the neurons in the previous layer. The layers are stacked similar to traditional neural networks.

[0184] Exemplary Aspects

[0185] In view of the described device and processes, herein arc described certain more particularly described aspects of the disclosures. These particularly recited aspects should not, however, be interpreted to have any limiting effect on any different claims containing different or more general teachings described herein, or that the “particular” aspects are somehow limited in some way other than the inherent meanings of the language and formulas literally used therein.

[0186] Further exemplary aspects of the disclosure are provided by one or more of the following examples:

[0187] Example 1. A computer-implemented method for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs), the computer-implemented method comprising: generating, by at least one processor and with a deep-learning-empowered structural modeler, a three-dimensional (3D) structure of a CAR target; generating, by the at least one processor, an optimal binding interface scaffold conforming to a unique peptide structure; generating, by the at least one processor and using a deep-learning-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predicting, by the at least one processor and using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets; evaluating, by the at least one processor and using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivityAttorney Docket No. 11820-039W01YAROl-HPRO of the binder sequences in silico; and selecting, by the at least one processor, a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

[0188] Example 2. The computer-implemented method of Example 1 , wherein the 3D structure of the CAR target comprises at least one of an HLA molecule, an HLA-presented peptide, and a beta-2-microglobulin (beta-2-m).

[0189] Example 3. The computer-implemented method of Example 1 or 2, wherein evaluating the selectivity of the binder sequences comprises: performing, by the at least one processor and using one or more decoy peptides, a cross-reactivity test; and eliminating one or more binders from the set of binder sequences based on results of the cross-reactivity test.

[0190] Example 4. The computer-implemented method of any one of examples 1-3, wherein each interaction score (e.g., Predicted Aligned Error (PAE) score) is within a predefined distance threshold.

[0191] Example 5. The computer-implemented method of any one of examples 1-4, wherein evaluating the selectivity of the binder sequences comprises: determining, by the at least one processor, a specificity score for each binder sequence; and selecting, by the at least one processor, the subset of the binder sequences based on the determined specificity scores.

[0192] Example 6. The computer-implemented method of Example 5, wherein the specificity score for each binder sequence is determined based on possible bonds (e.g., interactions between the CAR target and the binder sequence) that are between 0-6 Angstroms (A).

[0193] Example 7. The computer-implemented method of any one of examples 1-6, wherein the CAR target comprises a predicted computer structure model or an experimentally solved structure model.

[0194] Example 8. The computer-implemented method of any one of examples 1-7, wherein the method of generating the AIR-CAR is independent of HLA allele type.

[0195] Example 9. The computer-implemented method of any one of examples 3-8, wherein the one or more decoy peptides have binding residues within a predefined distance threshold of at least one binder sequence.

[0196] Example 10. The computer-implemented method of any one of examples 1-9, wherein the subset of binder sequences is used for in vitro testing.

[0197] Example 11. A method of treating a cancer in a subject in need thereof, comprising administering to the subject a substance comprising the AIR-CAR generated according to any one of examples 1-10.Attorney Docket No. 11820-039W01YAROl-HPRO

[0198] Example 12. A method of treating a cancer in a subject in need thereof, comprising administering to the subject an AIR-CAR immune cell comprising at least one binding sequence generated according to any one of examples 1-10 to the subject.

[0199] Example 13. The method of Example 12, wherein immune cell is a T cell, B cell, natural killer (NK) cell, NK T cell, or macrophage.

[0200] Example 14. The method of any one of examples 11-13, wherein the cancer comprises neuroblastoma / glioblastoma, colon adenocarcinoma, breast cancer, ovarian cancer, skin cutaneous melanoma, esophageal cancer, lung squamous cell carcinoma, lung adenocarcinoma, pancreatic adenocarcinoma, or cervical squamous cell carcinoma.

[0201] Example 15. A system for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs), comprising: at least one processor; and a memory having instructions thereon, wherein the instractions when executed by the at least one processor, cause the at least one processor to: generate a three-dimensional (3D) structure of a CAR target; generate an optimal binding interface scaffold conforming to a unique peptide structure; generate, using a deep-learning -based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predict, using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets; evaluate, using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; and select a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

[0202] Example 16. A non-transitory computer accessible medium comprising a memory having instructions stored thereon to cause at least one processor to: generate a three-dimensional (3D) structure of a CAR target; generate an optimal binding interface scaffold conforming to a unique peptide structure; generate, using a deep-learning-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope; predict, using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets; evaluate, using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; and select a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.Attorney Docket No. 11820-039W01YAROl-HPRO ReferencesThe following references are hereby incorporated by reference herein in their entireties: 1. Watson, Joseph L., David Juergens, Nathaniel R. Bennett, Brian L. Trippe, Jason Yim, Helen E. Eisenach, Woody Ahern, et al. 2023. “De Novo Design of Protein Structure and Function with RFdiffusion.’’ Nature 620 (7976): 1089-1100.2. Dauparas, J., I. Anishchenko, N. Bennett, H. Bai, R. J. Ragotte, L. F. Milles, B. I. M. Wicky, et al. 2022. “Robust Deep Eearning-based Protein Sequence Design Using ProteinMPNN.” Science, September.3. Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Zfdek, A., Potapenko, A., Bridgland, A., Meyer, C., Kohl, S. A. A., Ballard, A. J., Cowie, A., Romera-Paredes, B., Nikolov, S., Jain, R., Adler, J„ Back, T., ... Hassabis, D. (2021). Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873), 583-589.4. Bryant, P., Pozzati, G. & Elofsson, A. Improved prediction of protein-protein interactions using AlphaFold2. Nat Commun 13, 1265 (2022).5. Bennett, Nathaniel R., Brian Coventry, Inna Goreshnik, Buwei Huang, Aza Allen, Dionne Vafeados, Ying Po Peng, et al. 2023. “Improving de Novo Protein Binder Design with Deep Eearning.’’ Nature Communications 14 (1): 1-9.6.' Yarmarkovich, M., Marshall, Q. F., Warrington, J. M., Premaratne, R., Farrel, A., Groff, D., Li, W., di Marco, M., Runbeck, E., Truong, H., Toor, J. S., Tripathi, S., Nguyen, S., Shen, H., Noel, T., Church, N. L., Weiner, A., Kendsersky, N., Martinez, D., ... Maris, J. M. (2023). Targeting of intracellular oncoproteins with peptidecentric CARs. Nature, 623(7988), 820-8277. Alford, R. F., Eeaver-Fay, A., Jeliazkov, J. R., DiMaio, F. P., Park, H., Shapovalov, M. V., Renfrew, P. D., Mulligan, V. K., Kappel, K., Labonte, J. W., Pacella, M. S., Bonneau, R., Bradley, P., Das, R., Baker, D., Kuhlman, B., Kortemme, T., & Gray, J. J. (2017). The Rosetta all-atom energy function for macromolecular modeling and design. Journal of Chemical Theory and Computation, 13(6), 3031.8. Molecular Operating Environment (MOE), 2024.06 Chemical Computing Group ULC, 910-1010 Sherbrooke St. W„ Montreal, QC H3A 2R7, 2024.

Claims

Attorney Docket No. 11820-039W01YAROl-HPRO What is claimed:

1. A computer-implemented method for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs), the computer-implemented method comprising:generating, by at least one processor and with a deep-learning-empowered structural modeler, a three-dimensional (3D) structure of a CAR target;generating, by the at least one processor, an optimal binding interface scaffold conforming to a unique peptide structure;generating, by the at least one processor and using a deep-learning-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope;predicting, by the at least one processor and using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets; evaluating, by the at least one processor and using a deep-leaming-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; andselecting, by the at least one processor, a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

2. The computer-implemented method of claim 1 , wherein the 3D structure of the CAR target comprises at least one of an HLA molecule, an HLA-presented peptide, and a beta- 2-microglobulin (beta-2-m).

3. The computer-implemented method of claim 1 or 2, wherein evaluating the selectivity of the binder sequences comprises:performing, by the at least one processor and using one or more decoy peptides, a cross-reactivity test; andeliminating one or more binders from the set of binder sequences based on results of the cross-reactivity test.

4. The computer-implemented method of any one of claims 1-3, wherein each interaction score (e.g., Predicted Aligned Error (PAE) score) is within a predefined distance threshold.Attorney Docket No. 11820-039W01YAROl-HPRO5. The computer-implemented method of any one of claims 1-4, wherein evaluating the selectivity of the binder sequences comprises:determining, by the at least one processor, a specificity score for each binder sequence; andselecting, by the at least one processor, the subset of the binder sequences based on the determined specificity scores.

6. The computer-implemented method of claim 5, wherein the specificity score for each binder sequence is determined based on possible bonds (e.g., interactions between the CAR target and the binder sequence) that are between 0-6 Angstroms (A).

7. The computer-implemented method of any one of claims 1-6, wherein the CAR target comprises a predicted computer structure model or an experimentally solved structure model.

8. The computer-implemented method of any one of claims 1-7, wherein the method of generating the AIR-CAR is independent of HLA allele type.

9. The computer-implemented method of any one of claims 3-8, wherein the one or more decoy peptides have binding residues within a predefined distance threshold of at least one binder sequence.

10. The computer-implemented method of any one of claims 1-9, wherein the subset of binder sequences is used for in vitro testing.

11. A method of treating a cancer in a subject in need thereof, comprising administering to the subject a substance comprising the AIR-CAR generated according to any one of claims 1-10.

12. A method of treating a cancer in a subject in need thereof, comprising administering to the subject an AIR-CAR immune cell comprising at least one binding sequence generated according to any one of claims 1-10 to the subject.

13. The method of claim 12, wherein immune cell is a T cell, B cell, natural killer (NK) cell, NK T cell, or macrophage.Attorney Docket No. 11820-039W01YAROl-HPRO 14. The method of any one of claims 11-13, wherein the cancer comprises neuroblastoma / glioblastoma, colon adenocarcinoma, breast cancer, ovarian cancer, skin cutaneous melanoma, esophageal cancer, lung squamous cell carcinoma, lung adenocarcinoma, pancreatic adenocarcinoma, or cervical squamous cell carcinoma.

15. A system for generating Artificial Intelligence-engineered Receptor (AIR) - Chimeric Antigen Receptors (CARs), comprising:at least one processor; anda memory having instructions thereon, wherein the instructions when executed by the at least one processor, cause the at least one processor to:generate a three-dimensional (3D) structure of a CAR target;generate an optimal binding interface scaffold conforming to a unique peptide structure;generate, using a deep-learning -based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope;predict, using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets;evaluate, using a deep-learning-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; andselect a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

16. A non-transitory computer accessible medium comprising a memory having instructions stored thereon to cause at least one processor to:generate a three-dimensional (3D) structure of a CAR target;generate an optimal binding interface scaffold conforming to a unique peptide structure;generate, using a deep-leaming-based protein sequencer, a set of binder sequences optimized for one or more proteins that (i) fold into the optimal binding interface scaffold, and (ii) provide strong interactions with a defined epitope;predict, using an artificial intelligence model, a plurality of target cross-reactivities to generate a panel of putative cross-reactive targets;Attorney Docket No. 11820-039W01YAROl-HPRO evaluate, using a deep-learning-based modeler combined with at least one energy minimization modeler, selectivity of the binder sequences in silico; andselect a subset of binder sequences that meets or exceeds an interaction score with respect to the CAR target.

Images