Compositions and treatment of her2 heterogenous, her2-low, or her2 therapy-resistant cancers

Abstract

Described are methods and compositions for treating a cancer in a subject having a tumor that is HER2-low, HER2 heterogenous, or wherein the subject has a tumor has resistance to anti- HER2 therapy. In the method for tumor treatment an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, and a HER2 binding agent or HER2 inhibitor, are administered to the subject.

Description

MRG: 0680.003419W001 COMPOSITIONS AND TREATMENT OF HER2 HETEROGENOUS, HER2-LOW, OR HER2 THERAPY-RESISTANT CANCERSSTATEMENT OF FEDERALLY SPONSORED RESEARCH

[0001] This invention was made with government support under CAI 97623 awarded by the National Institute of Health. The Government has certain rights in the invention.BACKGROUND

[0002] The receptor tyrosine kinase HER2 ( human epidermal growth factor receptor 2; ErbB2), a plasma membrane receptor, is a member of the human epidermal growth factor receptor (ERBB) family which includes Herl (EGFR, ErbBl), Her3 (ErbB3), and Her4 (ErbB4). HER2 is activated by homodimerization or heterodimerization with another ERBB member. This leads to trans autophosphorylation and initiates different downstream signaling pathways (e.g., JAK / STAT, Ras / MEK / ERK, PI3K / AKT, and PLCy / PKC) through substrate protein phosphorylation, that in turn affects different cellular functions such as proliferation, differentiation, motility, survival, migration, and adhesion.

[0003] HER2 over-expression and / or amplification is associated with aggressive forms of cancer. For example, HER2-positive (HER+) cancer is one subtype of breast cancer which can be aggressive. Other subtypes of breast cancer include basal-like, luminal A / B, claudin-low, and normal breast-like, based on the expression levels of estrogen and progesterone receptors (ER and PR), cytokeratins 5 / 6, and claudins 3 / 4 / 7.

[0004] Within HER2 breast cancers there are further subtypes based on the level of expression of HER2 in the tumor, and / or further subtypes based on the spatial distribution of tumor cell populations that express different levels of HER2 (HER2 heterogeneity).

[0005] According to 2018 ASCO / CAP guidelines the evaluation of HER2 status is evaluated and defined by an immunohistochemistry (IHC) score as follows: 0 or 1+ for HER2-negative, 2+ for HER2-equivocal and 3+ for HER-positive. HER2-equivocals are further examined by in situ hybridization (ISH) assay to categorize tumors as either HER2-negative or HER2-positive. The HER2 profile of cells in a tumor (e.g., the presence of HER2 / low cells) dictate what type of anti-tumor therapy should be used in patient having such a profile. For example, the anti-HER2 antibody-drug conjugate (ADC) trastuzumab-deruxtecan (T-DXd) has been used to deliver cytotoxic deruxtecan to cells with low HER2 levels.MRG: 0680.003419W001

[0006] While ADCs that target HER2 are available, different mechanisms of resistance to anti-HER2 targeted therapies are known. For example, for cells that have low levels of HER2 expression there are small amounts of HER2 targets that the ADC can bind to, which in turn limits the amount of cytotoxic drug that can be delivered to the cell. Other forms of resistance include alterations in the PI3K pathway which can increase estrogen receptor (ER) expression and stimulates cell growth, reducing inhibition by antibodies. HER2 mutations that reduce the binding of inhibitors, such as tyrosine kinase inhibitors (TKIs) and anti-HER2 antibodies also contribute to resistance. Many patients treated for metastatic breast with ADCs such trastuzumab or with TKIs such as lapatinib develop resistance over time, and continued treatment with these therapeutics becomes ineffective long term.

[0007] Therapies that use a combination of an ADCs such trastuzumab with other chemotherapeutic agents, such as paclitaxel or docetaxel, or with other agents that target HER2, such as lapatinib, are known. E.g., see Pinto, et al., The Breast, 22 (Supp. 2), S152-S155, 2013. However, these approaches can increase the chemotherapeutic agent load in the patient which may not be tolerated well. Further, combination treatments that use agents directed against the same target, HER2, may not be particularly effective against tumor cells that have low levels of HER2 expression, such as HER2 / low cells.

[0008] Treatment of HER2 tumors remain significant challenges, especially in tumors that are HER2-low or are HER2 heterogenous, or in patients that are resistant to known anti-HER2 targeting therapies.SUMMARY

[0009] The disclosure relates to methods for treating a subject having a HER2+tumor, such as subject having breast cancer.

[0010] The disclosure provides a method of treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy. The method includes administering to the subject an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. The method also includes administering the HER2 binding agent or HER2 inhibitor to the subject.

[0011] In aspects, the inhibitor is directed against a target protein or gene expressing the target protein. In aspects , the target protein is ATP binding cassette subfamily C member 1MRG: 0680.003419W001 (ABCC1); a deubiquitinase that is ubiquitin specific peptidase 9 X-linked or ubiquitin carboxyl-terminal hydrolase 17-like protein 19, a nucleic acid-associated protein that is DNA-dependent protein kinase catalytic subunit, exosome RNA helicase MTR4, zinc finger protein 717, non- structural maintenance of chromosomes element 3 homolog, and isoleucine— tRNA Ligase, cytoplasmic; Bcl-2-like protein 1 ; cytohesin-interacting protein; vesicle transport protein GOT1 A; moesin-ezrin-radixin like tumor suppressor; guanine nucleotide exchange protein SMCR8; ERBB receptor feedback inhibitor 1; tyrosine-protein phosphatase that is tyrosineprotein phosphatase non-receptor type 12 or tyrosine-protein phosphatase non-receptor type 11; or dolichyl-diphosphooligo-saccharide— protein glycosyltransferase subunit 2 or dolichyl-phosphate beta-glucosyltransferase, or the target gene is ABCC1, USP9X, USP17L19, PRKDC, SKIV2L2, ZNF717, NDNL2, IARS, BCL2L1, CYTIP, GOLT1A, NF2, SMCR8, ERRFI1, PTPN12, PTPN11,RPN2, or ALG5.

[0012] In aspects, the cancer comprising the HER2 tumor is breast cancer.

[0013] In aspects, the HER2 binding agent is an anti-HER2 antibody or fragment thereof, optionally linked to a drug such as a cytotoxic agent.

[0014] In aspects, the HER2 inhibitor is a tyrosine kinase inhibitor (TKI) directed against HER2.

[0015] In aspects, the pharmaceutical composition includes an anti-HER2 antibody or fragment thereof, or a tyrosine kinase inhibitor (TKI) directed against HER2.

[0016] The disclosure also provides use of pharmaceutical composition(s) for treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy. The one or more pharmaceutical composition(s) includes (a) an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, and (b) a HER2 binding agent or HER2 inhibitor.

[0017] The administration of the inhibitor of a target protein or gene encoding the target protein increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. This is beneficial in tumors that are HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy, as the HER2 binding agent or HER2 inhibitor is able to more readily affect cells in the tumor that would otherwise be non-responsive to anti-HER2 treatment or require high doses of an anti-HER2 treatment to affect the tumor. The methods andMRG: 0680.003419W001 compositions of the disclosure also allow for uses of lower concentrations of anti-HER2 antibody drug conjugates, which in turn decreases possible side effects and makes anti-tumor therapy more tolerable to the patient.

[0018] This summary is not intended to describe each disclosed embodiment or every implementation. The description that follows more particularly exemplifies illustrative embodiments. In several places throughout the disclosure, guidance is provided through lists of examples, which examples may be used in various combinations. In each instance, the recited list serves only as a representative group and should not be interpreted as an exclusive list. Thus, the scope of the present disclosure should not be limited to the specific illustrative embodiments described herein, but rather extends at least to the embodiments described by the language of the claims, and the equivalents of those embodiments. Any of the elements that are positively recited in this specification as alternatives may be explicitly included in the claims or excluded from the claims, in any combination as desired. Although various theories and possible mechanisms may have been discussed herein, in no event should such discussions serve to limit the claimable subject matter.BRIEF DESCRIPTION OF THE FIGURES

[0019] Embodiments will be described, by way of example, with reference to the following drawings.

[0020] Figure 1A shows graphs of HER2 levels HER2+ breast cancer cells from flow cytometry.

[0021] Figure IB shows graphs of flow cytometry results of HER2 levels of HER2hland HER210cells.

[0022] Figure 1C is a Western blot image of proteins in HER2hland HER210cells.

[0023] Figure ID shows microscope images of FISH analysis from HER2hland HER210cells.

[0024] Figure IE shows graphs of copy number variations from WES of paired HERZ111and HER210cells.

[0025] Figure IF is a panel of mutations in paired HERZ111and HER210cells.

[0026] Figure 1G shows a graph of PAM50 subtype probability scores of paired HERZ111and HER210cells.MRG: 0680.003419W001

[0027] Figure 1H shows graphs of basal, luminal and mesenchymal markers in HER2hland HER210cells.

[0028] Figure II shows a graph and reactome pathways enriched in genes uniquely mutated in HER210cells.

[0029] Figure 1 J shows UMAP visualization plots of marker expression in HER2hland HER210cells.

[0030] Figure IK shows volcano plots of differentially expressed genes in paired HERZ111and HER210cells.

[0031] Figure IL shows a heatmap panel of enriched gene sets in HER111and HER210cells.

[0032] Figure IM shows box plots of enrichment scores of different ERBB / PI3K pathway signatures in HER2hland HER210cells.

[0033] Figure 2A shows graphs of dose response curves to indicated compounds in HER21" and HER210cells.

[0034] Figure 2B shows an immunoblot analysis of EGFR, HER2, and downstream signaling pathway components in treated HERZ111and HER210cells.

[0035] Figure 2C shows charts of flow cytometry data analysis of HERZlo subpopulation of patient derived organoids from HERZ heterogeneous tumor T537 treated with T-DXd or neratinib.

[0036] Figure 2D shows micrograph images of FISH analysis in control and resistant cells and a bar plot depicts quantification of the FISH signal.

[0037] Figure 2E shows a component analysis plot of RNA-seq profiles of control cells, HER2hland HER210subpopulations, and resistant cells.

[0038] Figure 2F shows bar graphs of Euclidean distances between PCI and PC2 of the indicated RNA-seq samples from Figure 2D.

[0039] Figure 2G shows graphs of dose-response curves to the indicated compounds in control and resistant cells.

[0040] Figures 2H and 21 show graphs of dose-response curves to the indicated compounds in HERZ111, HER210, and parental cell lines.

[0041] Figures 2 J and 2K show graphs of response curves of the specified compounds tested at increasing concentrations in paired HER2hi and HER21o cells in the indicated cell lines.MRG: 0680.003419W001

[0042] Figures 2L and 2M show bar plots depicting relative cell counts normalized to vehicle-treated controls after T-DXd or T-DM1 treatment.

[0043] Figure 2N shows graphs of dose-response curves to the indicated compounds in HER2hiand HER210cells from HCC1954, 21NT, and 21PT cell lines.

[0044] Figure 20 shows stacked bar plots of relative proportions of HER21" and HER210cells after treatment with the indicated agents.

[0045] Figure 2P shows graphs of flow cytometry data from HER210subpopulation in patient derived organoids from T565 and Metsl5 treated with T-DXd or neratinib.

[0046] Figure 3 A shows graphs of cell numbers of HER2hland HER210cells grown in monoculture or in co-culture in the presence of the indicated compounds.

[0047] Figure 3B shows a heatmap plot panel depicting the relative bystander effect of T-DXd on cells.

[0048] Figure 3C shows a graph of dose-response curves to T-DXd in HER210and TNBC cells.

[0049] Figure 3D shows bar plots of days per passage in the indicated conditions and HER2hland HER210cells ratios.

[0050] Figure 3E shows graphs of Shannon entropy measuring barcode diversity over HER2hland HER210cell passages in the indicated conditions.

[0051] Figure 3F shows bar plots of barcodes numbers present in HER2hland HER210cells after culturing.

[0052] Figure 3G shows bar plots showing quantification of ERBB2 amplified cells in resistant HCC1954 and 21PT cells.

[0053] Figure 3H shows a heatmap panel of Euclidian distances between PCI and PC2 of the indicated RNA-seq samples from Figure 2D.

[0054] Figure 31 shows a heatmap panel of enriched gene sets in HER210and resistant cells to control cells.

[0055] Figure 3J shows a heatmap panel of enriched gene sets in resistant and control cells.

[0056] Figure S3E shows an immunoblot image of EGFR, HER2, and downstream signaling pathway components in control and resistant cells.MRG: 0680.003419W001

[0057] Figure 3L shows graphs of dose-response curves to the indicated compounds in control cells and resistant derivatives.

[0058] Figure 3M shows graphs of IC50 correlation of the indicated compounds and the percentage of HER210cells based on Fig. 1A.

[0059] Figure 3N shows bar graphs of relative HER2hi and HER21o cells fractions and normalized cell count after compound treatment.

[0060] Figure 4A is a graph of tumor volume over time following mammary fat pad injection of HER2hlor HER21o cells.

[0061] Figure 4B shows immunohistochemistry micrographs of HER2hland HER210having different HER2 levels.

[0062] Figure 4C is a bar plot graph of HER2hland HER210cells in tumors.

[0063] Figure 4D is a bar plot graph of HER2hland HER210cells in tumors.

[0064] Figure 4E is a bar graph of PCNA-positive HER2hland HER210cells in heterogenous tumors.

[0065] Figure 4F is a bar graph of PCNA-positive HER2hland HER210cells in tumors.

[0066] Figure 4G is a graph of data from CELLCHAT analysis of scRNA-seq data depicting incoming and outgoing interactions of cell types.

[0067] Figure 4H is a plot graph presenting the pathways of the overall signals comparing HER21o and HER2hi cell populations.

[0068] Figure 41 is a graph illustrating the presence of PDPN+ stromal cells in the neighborhood of HER2hi and HER21o cells.

[0069] Figure 4 J is a schematic illustration for tumor treatment with neratinib or T-DXd or neratinib.

[0070] Figure 4K is a graph of tumor size following neratinib or T-DXd treatment.

[0071] Figure 4L are fluorescence micrographs of HER2hlcells and HER210cells in tumors after treatment.

[0072] Figure 4M is quantification HER2hland HER210cells in heterogeneous tumors after drug treatment.

[0073] Figure 4N is a schematic illustration for tumors treatment with neratinib or T- DXd or neratinib.MRG: 0680.003419W001

[0074] Figure 40 is a graph of tumor growth and recurrence of homogeneous and heterogeneous tumors upon T-DXd or neratinib treatment.

[0075] Figure 4P is a graph of survival curves in treated homogeneous and heterogeneous tumor-bearing mice.

[0076] Figure 4Q is quantification HER2hland HER210cells in heterogeneous tumors after drug treatment.

[0077] Figure 4R show graphs depicting HER2hi and HER21o cells growth in culture in the presence of various compounds.

[0078] Figure 4S show graphs reflecting HER2hi and HER21o cells mixed at different ratios and grown in culture in the presence of various compounds.

[0079]

[0080] Figure 4T is a heatmap panel showing gene sets significantly enriched in the RNA-seq data.

[0081] Figure 4U is a heatmap panel of protein expression in monoculture and co-culture showing expression differences.

[0082] Figure 4V is a heatmap panel depicting the most expressed cytokines secreted in monoculture or co-culture.

[0083] Figure 5 A is a schematic illustration showing an experimental process using HER210cells infected with Cas9 and whole genome CRISPR KO library and co-cultured with HER2hlcells in the presence of T-DXd.

[0084] Figure 5B is a bubble plot showing enrichment in KEGG pathways of the CRISPR hits linked to resistance and synthetic lethality.

[0085] Figure 5C is a scatter plot showing hits from the CRISPR screen linked to resistance or synthetic lethality.

[0086] Figure 5D shows histograms of cell counts from T-DXd treatment and inhibition of ABCC1 or USP9X in 21NT HER210cells.

[0087] Figure 5E shows bar graphs of cell counts following treatments in 21 NT TDR.

[0088] Figure 5F shows a bar graph of cell counts upon inhibition of ABCC1 in 21 NT HER210cells.

[0089] Figure 5G is a graph of the change in bystander effect in ABCC1 KO 21NT HER210cells.MRG: 0680.003419W001

[0090] Figure 5H is a schematic illustration of a survival assay experiment.

[0091] Figure 51 is graph of tumor volume during the indicated treatment.

[0092] Figure 5 J is a Kaplan-Meier plot showing time to tumor volume endpoint in mice with the indicated treatments.

[0093] Figure 5K shows stacked bar plots of days per passage in various conditions, and the ratio of HER2hi and HER21o in co-culture.

[0094] Figure 5L show graphs of Shannon entropy over different passages of treated 21PT cells.

[0095] Figure 5M shows histograms of remaining barcodes in 21 PT cells in various conditions.

[0096] Figure 6A shows immunoblot images illustrating an increase of HER2 ubiquitination upon USP9X inhibition in the presence of T-DXd in 21NT HER210cells.

[0097] Figure 6B shows immunoblot images illustrating interaction between USP9X and HER2.

[0098] Figure 6C is a bar graph of Proximity Ligation Assay (PLA) between HER2 and USP9X in 21NT HER210cells treated with vehicle or T-DXd.

[0099] Figure 6D shows immunoblot images illustrating HER2 levels of 21 NT HER210cells after treatment with T-DXd upon USP9X inhibition.

[0100] Figure 6E is graph showing HER2 levels after cycloheximide treatment in 21NT HER210cells upon T-DXd treatment with USP9X inhibition.

[0101] Figure 6F shows immunoblot images illustrating HER2 and calnexin levels from cells of Figure 6E.

[0102] Figure 6G is a graph of quantification of PLA intensity per cell.

[0103] Figure 6H shows immunoblot images illustrating proteins of total cell lysate (TCL) and lysosome enriched fraction of 21NT HER210cells treated as in Figure 6G.

[0104] Figure 61 is a graph of USP9X mRNA expression in human tumors before and after T765 DMl / pertuzumab neoadjuvant treatment.

[0105] Figure 6J is a schematic illustration of an experimental procedure where 21NT HER2 heterogeneous tumors are treated with one dose of T-DXd, G9, or both.

[0106] Figure 6K is a graph of tumor growth of individual tumors following indicated treatment, n = 8-10 tumors, two-way ANOVA.MRG: 0680.003419W001

[0107] Figure 6L is a graph of survival curves showing time to tumor volume endpoint in treated mice.

[0108] Figure 6M shows heatmap plots of p-values from tests for differential barcode selection among the top 30 barcodes across various conditions.

[0109] Figure 7A shows plot graphs depicting tumor growth of xenografts derived from 21 NT HERZ111or HER210used alone or in mixture.

[0110] Figure 7B shows a graph of the number of HER2hi and HER21o relative to tumor growth over time.

[0111] Figure 7C show a plot showing pathways of the overall signals from HER21o and HER2hi cells defined by Cellchat analysis of scRNA-seq data.

[0112] Figure 7D shows bar graphs of stromal cells in the neighborhood of HER2hi and HER21o cells.

[0113] Figure 7E shows waterfall plots graphs of change in volume of HER2 homogeneous and heterogeneous tumors following treatment with neratinib or T-DXd.

[0114] Figure 7F shows a graph of growth curve of tumor size under treatment with T-DXd.

[0115] Figure 7G shows a waterfall plot graph of change in tumor volume of HER2 homogeneous and heterogeneous tumors following T-DXd treatment.

[0116] Figure 7H shows a bar graph of quantification of ERSBZ-amplified and nonamplified cells on cell cultures derived from vehicle or T-DXd heterogeneous recurrent tumors from Fig. 4R.

[0117] Figure 71 is a graph showing response to increasing doses of T-DXd of cell cultures derived from vehicle or T-DXd-treated heterogeneous recurrent tumors.

[0118] Figure 8A shows immunoblot images illustrating ABCC1 and USP9X protein levels from ABCC1 KO (sgABCCY), USP9X KO (sgDSPP ) 21NT, and HCC1954 HER210cells.

[0119] Figure 8B shows bar graphs of cell counts with or without T-DXd treatment upon genetic or pharmacologic inhibition of ABCC1 or USP9X in 21 NT and HCC1954 HERZ111and HER210cells.

[0120] Figure 8C shows immunoblot images illustrating expression of ABCC1 in 21NT and HCC1954 resistant derivates.MRG: 0680.003419W001

[0121] Figure 8D shows bar graphs of relative T-DXd-treated to untreated cell counts of control and ABCC1 or USP9X inhibitor-treated HCC1954 TDR cells.

[0122] Figure 8E shows bar graphs of relative T-DXd-treated to untreated cell counts cell counts upon genetic or pharmacologic inhibition of ABCC1 or USP9X in 21NT and HCC1954 HER21o cells.

[0123] Figure 8F is a bar graph showing ABCC1 KO in HCC1954 HER21o cells increases the bystander effect of TDXd in co-culture with HCC1954 HER2hi cells.

[0124] Figure 8G is a bar graph reflecting ABCC1 mRNA expression in cell lines.

[0125] Figure 8H is a bar plot of the relative fraction of HER2hi and HER21o cells in recurrent tumors.

[0126] Figure 9A shows bar graphs of T-DM1 -treated to untreated cell counts of USP9X, G9, and FT709 inhibited 21NT HER21o or 21NT and HCC1954 TMR cells.

[0127] Figure 9B shows immunoblot images of EGFR / HER2 signaling pathway components upon USP9X inhibition with or without T-DXd treatment.

[0128] Figure 9C is a flow cytometry histogram of HER2 upon USP9X genetic or pharmacologic inhibition with or without T-DXd.

[0129] Figure 9D shows bar graphs of HER2 protein levels in 21NT and HCC1943 HER21o cells.

[0130] Figure 9E shows immunoblot images of HER2 immunoprecipitants (IP) for ubiquitin (Ub) upon USP9X inhibition in the presence of T-DXd in HCC1954 HER21o cells.

[0131] Figure 9F shows immunoblot images of USP9X and HER2 in control IgG, USP9X or HER2 immunoprecipitants from 21NT HER2hi and HCC1954 HER21o cells.

[0132] Figure 9G is a bar plot graph illustrating quantification of PLA intensity per cell.

[0133] Figure 9H shows immunoblot images of HER2 from 21NT HER2hi and HCC1954 HER21o cells after T-DXd treatment and upon USP9X inhibition.

[0134] Figure 91 is a bar graph illustrating quantification of PLA intensity per cell.

[0135] Figure 9J shows immunoblot images of proteins from total cell lysate and lysosome enriched fraction of HCC1954 HER21o cells treated with T-DXd and FT709.

[0136] Figure 9K is a bar plot graph of proportion of HER2hi and HER21o cells in the indicated recurrent tumors.MRG: 0680.003419W001DETAILED DESCRIPTION

[0137] All publications, patents and patent applications mentioned in the specification are indicative of the level of skill of those skilled in the art to which this invention pertains. All publications, patents and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

[0138] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0139] A “subject” to which administration is contemplated includes, but is not limited to, a human (i.e., a male or female of any age group, e.g., a pediatric subject (e.g., infant, child, adolescent) or adult subject (e.g., young adult, middle-aged adult, or senior adult)) and / or a nonhuman animal, for example, a mammal (e.g., a primate (e.g., cynomolgus monkey, rhesus monkey); or a domestic animal. The term “subject” is used interchangeably with the term “patient” according to the disclosure.

[0140] The terms “administer,” “administering,” or “administration” refers ingesting, injecting, inhaling, or otherwise introducing a pharmaceutical composition of the disclosure to the subject in need of treatment.

[0141] The terms “treatment,” “treat,” and “treating” refer to reversing, alleviating, delaying the onset of, or inhibiting the progress of a “pathological condition” (e.g., a disease, disorder, or condition, including one or more signs or symptoms thereof) described herein. In some embodiments, treatment may be administered after one or more signs or symptoms have developed or have been observed. Treatment may also be continued after symptoms have resolved, for example, to delay or prevent recurrence and / or spread.

[0142] The terms “condition,” “disease,” and “disorder” are used interchangeably.MRG: 0680.003419W001

[0143] A “therapeutically effective amount” or “therapeutically effective amounts” is or are an amount(s) sufficient to provide a therapeutic benefit in the treatment of a condition, which therapeutic benefit may be or comprise, for example, reduction in frequency and / or severity, and / or delay of onset of one or more features or symptoms associated with the condition. A therapeutically effective amount means an amount of therapeutic agent(s), alone or in combination with other therapies, that provides a therapeutic benefit in the treatment of the condition. The term “therapeutically effective amount” can encompass an amount that improves overall therapy, reduces or avoids symptoms or causes of the condition, or enhances the therapeutic efficacy of another therapeutic agent. Therapeutically effective amounts can involve administration of a plurality of doses, potentially over time (e.g., according to a dosing regimen). The prescribed administration of therapeutic agent and the doses over time constitute a “treatment regimen” that can be given to a subject in need. In some modes of practice, therapeutically effective amounts can be provided in single dosage forms.

[0144] The term “therapeutic agent” (e.g., an inhibitor of a target protein or gene encoding the target protein, or a HER2 binding agent or HER2 inhibitor) refers to an agent having one or more therapeutic properties that produce a desired, usually beneficial, effect. For example, a therapeutic agent may treat, ameliorate, and / or prevent disease.

[0145] The term “chemotherapeutic agent” refers to a therapeutic agent known to be of use in chemotherapy for cancer.

[0146] The term “inhibitor” refers to an agent that (i) decreases or suppresses one or more effects of another agent; and / or (ii) decreases or suppresses one or more biological events. An inhibitor may reduce level and / or activity or one or more agents that it targets. An inhibitor may be or include agents of various chemical class including, for example, small molecules, polypeptides, nucleic acids, carbohydrates, lipids, metals, and / or other entity that shows the relevant inhibitory activity. An inhibitor may be direct (in which case it exerts its influence directly upon its target) or indirect (in which case it exerts its influence by other than binding to its target; e.g., by interacting with a regulator of the target, for example so that level or activity of the target is altered). An inhibitor may be a receptor antagonist, e.g., a receptor ligand or drug that blocks or dampens a biological response by binding to and blocking a receptor rather than activating it like an agonist. The term “antagonist” may be used interchangeably with “inhibitor” according to the disclosure.MRG: 0680.003419W001

[0147] The term “inhibit” or “inhibition” in the context of modulating level (e.g., expression and / or activity) of a target is not limited to only total inhibition. Thus, in some embodiments, partial inhibition or relative reduction is included within the scope of the term “inhibition.” In some embodiments, the term refers to a reduction of the level (e.g., expression, and / or activity) of a target to a level that is reproducibly and / or statistically significantly lower than an initial or other appropriate reference level, which may, for example, be a baseline level of a target. In some embodiments, the term refers to a reduction of the level (e.g., expression and / or activity) of a target to a level that is less than 75%, less than 50%, less than 40%, less than 30%, less than 25%, less than 20%, less than 10%, less than 9%, less than 8%, less than 7%, less than 6%, less than 5%, less than 4%, less than 3%, less than 2%, less than 1%, less than 0.5%, less than 0.1%, less than 0.01%, less than 0.001%, or less than 0.0001% of an initial level, which may, for example, be a baseline level of a target.

[0148] In this disclosure, terms such as “a,” “an,” and “the” are not intended to refer to only a singular entity, but include the general class of which a specific example may be used for illustration. The terms “a,” “an,” and “the” are used interchangeably with the term “at least one.” The phrases “at least one of’ and “comprises at least one of’ followed by a list refers to any one of the items in the list and any combination of two or more items in the list.

[0149] As used herein, the term “or” is generally employed in its usual sense including “and / or” unless the content clearly dictates otherwise. The term “and / or” means one or all of the listed elements or a combination of any two or more of the listed elements. Thus, the term “and / or” as used in a phrase such as “A, B, and / or C” is intended to include “A and / or B and / or C” and to thus encompass each of the following embodiments: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A (alone); B (alone); and C (alone).

[0150] The words “preferred” and “preferably” refer to embodiments of the disclosure that may afford certain benefits, under certain circumstances. However, other claims may also be preferred, under the same or other circumstances. Furthermore, the recitation of one or more preferred claims does not imply that other claims are not useful, and is not intended to exclude such other claims from the scope of the disclosure.

[0151] Herein, any of the elements or combinations of elements that are recited in this specification in open-ended language (e.g., comprise and derivatives thereof), are considered to additionally be recited in closed-ended language (e.g., consist and derivatives thereof) and inMRG: 0680.003419W001 partially closed-ended language (e.g., consist essentially, and derivatives thereof). That is, it is understood that wherever aspects are described herein with the language “comprising,” otherwise analogous aspects described in terms of “consisting of’ and / or “consisting essentially of’ are also contemplated. In other words, if an embodiment comprises A, B and C, embodiments consisting essentially of A, B and C are also contemplated, as are embodiments consisting of A, B and C. As used herein, the term “comprising” is intended to mean that the compositions or methods include the recited steps or elements, but do not exclude others. “Consisting essentially of’ shall mean rendering the claims open only for the inclusion of steps or elements, which do not materially affect the basic and novel characteristics of the claimed compositions and methods. “Consisting of’ shall mean excluding any element or step not specified in the claim.

[0152] Groupings of alternative elements or embodiments disclosed herein are not to be construed as limitations. Each group member may be referred to and claimed individually or in any combination with other members of the group or other elements found therein. It is anticipated that one or more members of a group may be included in, or deleted from, a group for reasons of convenience and / or patentability. When any such inclusion or deletion occurs, the specification is herein deemed to contain the group as modified thus fulfilling the written description of all Markush groups used in the appended claims.

[0153] Unless otherwise indicated, all numbers expressing quantities of components, molecular weights, and so forth used in the specification and claims are to be understood as being modified in all instances by the term “about.” Accordingly, unless otherwise indicated to the contrary, the numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the present invention. At the very least, and not as an attempt to limit the doctrine of equivalents to the scope of the claims, each numerical parameter should at least be construed in light of the number of reported significant digits and by applying ordinary rounding techniques. “About” may be understood as within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the stated value. All numerical values provided herein are modified by the term about.

[0154] Ranges provided herein are understood to be shorthand for all values within the range, including fractions / decimals. For example, a range of 1 to 50 is understood to include any number, combination of numbers, or sub-range from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8,MRG: 0680.003419W001 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50.

[0155] Parts of the disclosure may optionally refer to “one embodiment,” “an embodiment,” “certain embodiments,” or “some embodiments,” etc., which means that a particular feature, configuration, composition, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of such phrases in various places throughout this specification are not necessarily referring to the same embodiment of the invention. Furthermore, the particular features, configurations, compositions, or characteristics may be combined in any suitable manner in one or more embodiments. The term “embodiment” and “embodiments” may be used interchangeable with “aspect” and “aspects”, respectively.

[0156] The disclosure provides a method of treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy. The method includes administering to the subject an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. The method also includes administering the HER2 binding agent or HER2 inhibitor to the subject.

[0157] HER2 can be overexpressed in various cancer types, including breast cancer, gastric cancer, lung cancer, salivary cancer, vaginal cancer, bladder cancer, endometrial cancer, cervical cancer, and colorectal cancer. E.g., Tapia, C., etal. Modern Pathology, 20(2): 192-198, 2007; Nelson, E.L., Clin. Invest. (Lond.) (2014) 4(8), 705-728. HER2 overexpression can result in the formation of aggressive tumors that are resistant to therapy, which are associated with a poor prognosis.

[0158] As used herein, “breast cancer” refers to a heterogeneous disease characterized by the uncontrolled growth of breast tissue cells. Breast cancer can have multiple disease types, each with each disease type having unique characteristics and treatment responses. Genetic factors, such as mutations in the BRCA1 and BRCA2 genes are associated with high risk factor of developing breast cancer. Women with BRCA1 and BRCA2 mutations have up to a 70-80% lifetime risk of developing breast cancer.MRG: 0680.003419W001

[0159] Early-stage primary breast tumors are referred to as Ductal Carcinoma In Situ (DCIS), which is confined to the milk ducts and have not spread to surrounding tissues and which is considered a non-invasive type of breast cancer.

[0160] The most common subtypes of cancer that has spread from the ducts or lobules into surrounding breast tissue is Invasive Ductal Carcinoma (IDC) which accounts for the majority of all breast cancers. Invasive Lobular Carcinoma (ILC) originates in the milkproducing lobules and is responsible for a minority of (about 10%) breast cancer cases. TripleNegative Breast Cancer (TNBC) is one of the most aggressive forms of breast cancer, and is characterized by lacking estrogen, progesterone, and HER2 receptors. Due to the lack of these receptors, targeting TNBC with directed therapies is challenging. Other breast cancer types include HER2-positive and Hormone Receptor-Positive Breast Cancer (HR+) cancers.

[0161] According to gene expression profiles, there are four main biological forms of breast cancer which are basal-like, luminal A, luminal B, and HER2-enriched (Nature, 490, 61-70, 2012). Clinically, patient breast cancer classification is evaluated by overexpression and / or amplification of the following biomarkers: estrogen receptor (ER) and progesterone receptor (PgR), which are hormone receptors, and the receptor HER2 ( human epidermal growth factor receptor 2; ErbB2) which is a plasma membrane receptor and a member of the human epidermal growth factor receptor (ERBB) family which includes Herl (EGFR, ErbBl), Her3 (ErbB3), and Her4 (ErbB4). Accordingly, three different breast cancer subtypes are defined: luminal-like (ER-positive and / or PgR-positive and HER2-negative), HER2-positive (HER2 overexpression / amplification, with any ER and PgR expression), and triple negative (ERnegative, PgR-negative and HER2-negative) tumors.

[0162] In about 15% of breast cancers show increased expression / amplification of the ERBB2 gene, which encodes human epidermal growth factor receptor 2 (HER2) (Schalper, K.A., etal., Arch. Pathol. Lab. Med. 138:213-219, 2014). Within HER2 breast cancers there are further subtypes based on the level of expression of HER2 in the tumor, and / or further subtypes based on the spatial distribution of tumor cell populations that express different levels of HER2 (HER2 heterogeneity). ASCO / CAP guidelines (2018) are used to evaluate HER2 status (Wolff, A.C., et al., Arch. Pathol. Lab. Med. 142:1364-1382, 2018) and these ASCO / CAP guidelines have been updated in 2023 (Wolff, A.C., etal., r ch Pathol Lab Med 147(9):993-1000, 2023). According to the ASCO / CAP guidelines the evaluation of HER2 status is evaluated and definedMRG: 0680.003419W001 by an immunohistochemistry (IHC) score as follows: 0 or 1+ for HER2-negative, 2+ for HER2-equivocal and 3+ for HER-positive. HER2-equivocals are further examined by in situ hybridization (ISH) assay to categorize tumors as either HER2-negative or HER2-positive (Marchid, C., et al., Semin. Cancer Biol. 72, 123-135, 2020). The following algorithm for evaluation of HER2 expression according to the 2018 ASCO / CAP guidelines, taken from Table 1 from Giugliano, F., et al., Cancers 15, 1385, 2023, is reproduced below:

[0163] The following abbreviations in Table 1 from Giugliano, F., et al. ibid. , are used: TC: tumor cells; HER2: human epidermal growth factor receptor 2; *: if ISH positive; #: if ISH negative; °: Note: in clinical practice, the evaluation of the status of HER2 by ISH is performed only when IHC score is 2+; if ISH result is between 4 and 6, breast cancer is classified as HER2-negative.

[0164] The ASCO / CAP guidelines convey the predictive role of HER2, which can facilitate the identification of patients with HER2-positive tumors that would be prescribed an anti-HER2 therapy, versus patients that have HER2-negative BC (Slamon, D. J., et al., N. Engl. J. Med. 344:783-792, 2001). Anti-HER2 therapies prescribed include using compounds that bind HER2 and / or block downstream signaling, such as pertuzumab, trastuzumab, deruxtecan (T-DXd), trastuzumab emtansine (T-DM1), trastuzumab deruxtecan (T-DXd), lapatinib, neratinib, and tucatinib.

[0165] Ivanova, M., et al. (Virchows Archiv 484:3-14, 2024), addressed the categorization of “HER2-low” cells, with reference to DESTINY-BreastO4 (DB-04) trial findings in June 2022, which focused on patients with metastatic breast cancer who were classified as “HER2-low,” meaning those with immunohistochemistry (IHC) HER2 1+ or 2+ andMRG: 0680.003419W001 negative in situ hybridization (ISH) results. As noted in Ivanova etal. (ibid.) the DB-04 trial excluded patients with HER2 IHC score 0 status but noted patients with this score could benefit from T-DXd therapy. Ivanova et al. (ibid.) further notes that providing the percentage (10% or less) of immunostained cells in samples with an IHC score of 0 could be useful for investigating the potential advantages of new ADCs in the context of HER2-ultra low BC (defined as a score of 0 with incomplete and faint staining in >0 and < 10% of tumor cells) in (see Sajjadi, E., et al., Front Mol Biosci 10:1176309, 2023). Ivanova etal. (ibid.) also refers to the DES TINY-BreastO6 (DB-06) trial (NCT04494425) investigating T-DXd in a patient cohort with HER2 score 0, as defined by ASCO / CAP, or HER2-ultra low, including IHC values >0 and < 10%, to establish the clinical validity and utility of the “HER2-zero > 0” biomarker, and to evaluate the potential clinical benefits of the drug by assessing progressively lower levels of HER2 IHC. Additionally, the DESTINY-BreastO6 trial (NCT04494425) again showed improved PFS with T-DXd versus chemotherapy in both HER2-low and HER2-ultralow (tumors with an IHC score > no staining < 1+, which include tumors with incomplete or faint / barely perceptible staining in < 10% of tumor cells) cohorts. Curigliano, G., et al. (Journal of Clinical Oncology 42(17)_suppl, 2024) reports the DB-06 (NCT04494425) evaluated T-DXd in patients with HER2-low or HER2-ultralow (IHC 0 with membrane staining) and reported that DXd significantly improved PFS vs TPC in HER2-low, and that ITT and HER2-ultralow results were consistent with HER2-low.

[0166] Accordingly, as used herein, “HER2-low” refers to cells having an immunohistochemistry (IHC) HER2 1 + or 2 + and negative in situ hybridization (ISH) results, and “HER2-low” also includes “HER2-ultralow” which are from tumors with an IHC score > no staining < 1+, which include tumors with incomplete or faint / barely perceptible staining in < 10% of tumor cells.

[0167] As reported by Giugliano, etal. (ibid.) up to 30% of HER2-positive breast cancers show heterogeneity in HER2 expression and different patterns of spatial distribution, which means that “heterogeneity” is the variability in the distribution and expression of the HER2 protein within a single tumor. HER2 genetic heterogeneity is defined as subclonal diversity within the tumor. Giugliano also reports that regarding the HER2 spatial distribution, three distinct patterns of distribution of cells with heterogeneous HER2 expression have been described: “clustered type”, “mosaic type” and “scattered type”. In the clustered type two different tumor clones (one with HER2 amplification and the other with normal HER2 status) areMRG: 0680.003419W001 identified. The mosaic type displays a diffuse intermingling of cells with different HER2 expression. The scattered type is characterized by isolated HER2-positive cells in a HER2-negative field. According to ASCO / CAP (2013) guidelines heterogeneity is defined as the presence of >10% to <50% tumor cells with a ratio > 2.0 when using dual probes or >6 HER2 signals / cell when using single probes, selecting 2-4 representative invasive tumor areas (Wolff, A.C., et al., J. Clin. Oncol. 31, 3997-4013, 2013). Accordingly, as used herein, “HER2 heterogenous” refers to tumors as defined ASCO / CAP (2013) guidelines, with tumors displaying any of the clustered type, mosaic type, or scattered type patterns of heterogeneity.

[0168] Aspects of the disclosure provide a method of treating a subject, wherein the subject has a tumor that is resistant to anti-HER2 therapy. A tumor that is resistant to anti-HER2 therapy displays a sub-optimal or no response to an anti-HER2 therapeutic agent such as an anti-HER2 antibody, an anti-HER2 ADC, or a small molecule HER2 inhibitor, such as a TKI. The resistance may be de novo or be from acquired resistance. It is known that a high percentage of patients receiving anti-HER2 therapy for metastatic HER2-positive breast cancer eventually see the disease progress because of de novo or acquired resistance to the anti-HER2 therapy. E.g., Pernas S., and Tolaney S.M., Therapeutic Advances in Medical Oncology. 11:1-16, 2019.Forms of resistance are believed to be due to activating mutations in the PI3K pathway, genomic alterations in the cell regulators p2721, cyclin E22, and cyclin D / CDK423, Smith, A.E., et al. Nat Commun 12, 6667 (2021). Exemplary forms of resistant to anti-HER2 therapy include trastuzumab resistant breast cancer, or breast cancer demonstrating primary or acquired resistance to HER2 inhibition.

[0169] Methods of the disclosure include administering to the subject an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. The method also includes administering the HER2 binding agent or HER2 inhibitor to the subject.

[0170] As used herein, a “target protein” refers to a protein that is present in a cell of a tumor that is HER2 heterogeneous, HER2-low, or a protein that is present in a cell of a tumor from a patient is resistant to anti-HER2 therapy, that when inhibited increases sensitivity of the cells of a tumor to a HER2 binding agent or HER2 inhibitor. By contacting cells of a tumor with the inhibitor of the target protein, the cells become more sensitive and therefore more susceptibleMRG: 0680.003419W001 to the cytotoxic activity of the HER2 binding agent or HER2 inhibitor, which in turn increases death in the cells of the tumor and improves treatment.

[0171] Correspondingly, a “gene encoding a target protein” refers to a gene that encodes a protein that is present in a cell of a tumor that is HER2 heterogeneous or HER2-low, or a gene that encodes a protein that is present in a cell of a tumor from a patient is resistant to anti-HER2 therapy, that when inhibited, such as by reducing expression of the gene, and therefore production of the target protein, increases sensitivity of the cells of a tumor to a HER2 binding agent or HER2 inhibitor.

[0172] Target proteins and genes that encode these proteins, that when inhibited, increase sensitivity of the cells of a tumor to a HER2 binding agent or HER2 inhibitor, can be understood from the disclosure of the application, for example, target proteins and genes that encode that were identified using a method of the disclosure. For example, protein and gene targets were identified by first culturing HER2-low cells in the presence of a HER2 therapeutic agent (e.g., T-DXd or neratinib) to generate cells that become less responsive to the HER2 therapeutic agent, reducing or eliminating individual gene activity, such as by performing knockouts of individual genes in the cell (e.g., using CRISPR knock out), and then treating the gene-deleted cells with a HER2 binding agent or HER2 inhibitor to determine what deleted genes caused increased sensitivity HER2 binding agent or HER2 inhibitor. These target genes that, when deleted, cause reduced or eliminated expression of the target protein, and in turn increase sensitivity of the HER2-low cells to the HER2 binding agent or HER2 inhibitor.

[0173] Methods and compositions of the disclosure for treating a tumor that is HER2 heterogeneous or HER2 / low, or wherein the tumor is resistant to anti-HER2 therapy, involve administration of an inhibitor of a target protein or gene that encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. That is, when the inhibitor is given to the subject, the inhibitor affects cells in the HER2 / low or HER2 het tumor, or tumor that is resistant to HER2 therapy, in one or more ways that makes the cells more responsive or more sensitive to, the cytotoxic effects of the HER2 binding agent or HER2 inhibitor. The inhibitor can act in a direct or indirect matter to increase the responsiveness or sensitize cells in the HER2 / low or HER2 het tumor. Accordingly, the inhibitor may be referred to as an “agent for HER2 sensitization” or an “agent for increasing HER2 responsiveness”.MRG: 0680.003419W001

[0174] The inhibitor that increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor refers to an agent (e.g., a small molecules like organic chemicals, or large molecules like proteins and nucleic acids) that interfere, block, or inhibit a target protein or gene that encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. The inhibitor can be a natural or synthetic compound that has a biological effect to inhibit or significantly reduce the activity or expression of the target protein or a transcript that encodes the target protein. In some modes of treatment or compositions of the disclosure, the inhibitor is a small organic molecule, an antibody, a polypeptide, an aptamer, an antisense oligonucleotide, or a siRNA.

[0175] In some modes of treatment and compositions of the disclosure the inhibitor is a small organic molecule. A “small organic molecule” as used herein, is a molecule having a size in the range of organic molecules generally used in pharmaceuticals and as known in the art. small organic molecules do not include biological macromolecules such as antibodies and nucleic acids which are otherwise known as “large molecules”. In aspects of the disclosure, the inhibitor is a small organic molecule having a molecular weight in the range of 100 g / mol to 1200 g / mol, in the range of 100 g / mol to 1000 g / mol, or in the range of 100 g / mol to 800 g / mol.

[0176] The inhibitor of the target protein that increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor can be an antibody or fragment thereof. An antibody that is an inhibitor can bind to a target protein to reduce its function, activity, or amount.Antibodies and or fragments thereof include any antibody-like molecule that has an antigen binding region and includes antibody fragments that have an antigen binding domain such as Fab', Fab, F(ab')2, single domain antibodies (DABs or VHH), Fv, scFv (single chain Fv), diabodies, bispecific antibody fragments, among other protein-based molecules that have an antigen binding domain. Methods of treatment and / or compositions of the disclosure can include a monoclonal antibody that binds to the target protein and inhibits its function. The antibody or fragment thereof can be internalizing or non-internalizing. The term “non-internalizing antibody” refers to an antibody or fragment thereof that binds to the target protein, which in some cases may be present on a cell surface, and that, when bound, does not enter the cell and become degraded in the lysosome.

[0177] Antibodies or antibody fragments which can function as inhibitors and bind to the target proteins of the disclosure are known in the art or can be prepared using all or a portion ofMRG: 0680.003419W001 the target protein sequence. Techniques for preparing and using various antibody-based constructs and fragments are well known in the art.

[0178] Polypeptides that interact and inhibit the target protein can also be used. An inhibitor polypeptide can specifically bind to the target protein and inhibits its function, such as by binding to and sequestering the target protein, thereby preventing it from performing its biological function. Polypeptides include those having at least two amino acid residues, such as polypeptides having 2-10 amino acid residues (often referred to as “peptides”), oligopeptides having 11-100 amino acid residues, and longer peptides, such as those having more than 100 amino acid residues in length, including as well as proteins.

[0179] The inhibitor of the target protein that increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor can be an aptamer. Aptamers include DNA or RNA oligonucleotides and synthetic versions thereof, that can bind to the target protein with high affinity and specificity and exhibit an inhibitory effect similar to small molecule inhibitors or inhibitor antibodies and fragments thereof.

[0180] The inhibitor can be a compound that inhibits expression of the gene that encodes the target protein. By reducing expression of the gene, less protein is produced, which acts similarly to an antagonist that may bind directly to the target protein and inhibit its function. In turn reducing expression of the target protein, this increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor. The inhibitor of the target gene expression can be a natural or synthetic compound that has a biological effect to inhibit or significantly reduce the expression of the gene encoding the target protein. The inhibitor of gene expression can inhibit target protein expression in one or more ways, such as by reducing production of an RNA template from a DNA sequence, such as by inhibiting transcription; reducing processing of an RNA transcript, such as by reducing transcript splicing, editing, 5' cap formation, and / or 3' end formation; by reducing translation of the RNA transcript into a polypeptide or protein; and / or, in some cases, by reducing post-translational modification of the target protein.

[0181] The inhibitor of the expression of the gene encoding the target protein can be performed using an antisense oligonucleotide. Antisense oligonucleotides include antisense RNA oligonucleotides and antisense DNA oligonucleotides, and synthetic analogues of these chemistries, can interfere with the translation of mRNA encoding the target protein, by binding thereto and thus preventing protein translation or increasing mRNA degradation. In turn thisMRG: 0680.003419W001 decreases the level of the target proteins and its activity in the tumor cell, making it more sensitive to the HER2 binding agent or HER2 inhibitor. For example, one or more antisense oligonucleotide(s), such as having a length in the range of 15 to 30 nucleotides and complementary to a unique region of the mRNA transcript of the sequence encoding the target protein can be synthesized.

[0182] The oligonucleotide can be composed of nucleotides that can include ribonucleotides, deoxyribonucleotides and modified nucleotides. Modified nucleotides can include modified nucleobases, sugar moieties and / or phosphate-binding regions. Modified nucleotides with a bases include, but are not limited to pseudouracil, 3 -methyluracil, 5-methylcytosine, 6-azapyrimidine, 4-acetylcytosine, 5 -(carboxy hydroxymethyl) uracil, 1-methylhypoxanthine, 2,2-dimethylguanine, 3 -methylcytosine, 2-methyladenine, 2-methylguanine, N6-methyladenine, 5 -methyloxyuracil, purine, 2-aminopurine, isoguanine, and the like. Chemical modifications at the 2'-position of ribose, such as replacement of the 2' OH of ribose with an oxygen, sulfur, nitrogen, or halogen atom containing group can be made.Modification at other positions of the sugar includes, for example, replacement of O at the 4' position of ribose or deoxyribose with S, bridging between 2' and 4' positions of the sugar, e.g., LNA (Locked Nucleic Acid) or ENA (2'-O,4'.C-Ethylene-bridged Nucleic Acids).Exemplary modification of the phosphate-binding region includes replacement of the phosphodiester bond with a phosphoroamidate bond, a phosphorothioate bond, a phosphorodithioate bond, an alkyl phosphonate bond, or a boranophosphate bond.

[0183] The oligonucleotide directed to reduce expression of the gene encoding the target protein can be done by intravenous injection or infusion. Methods for using antisense techniques for specifically alleviating gene expression of genes whose sequence is known are well known in the art (e.g. see U.S. Pat. Nos. 6,566,131; 6,410,323; 6,046,321; and 5,981,732).

[0184] Small inhibitory RNAs (siRNAs) can also be used to reduce gene expression for the target protein. A small double stranded RNA (dsRNA) can be delivered to the tumor cell alone or using a vector or construct that produces the dsRNA in order to reduce gene expression by RNA interference (RNAi). In turn this decreases the level of the target proteins and its activity in the tumor cell, making it more sensitive to the HER2 binding agent or HER2 inhibitor. The use of siRNA and dsRNA for inhibiting gene expression is known, see for example, WO 01 / 36646, WO 99 / 32619, and WO 01 / 68836.MRG: 0680.003419W001

[0185] Other methods that can be used for reducing expression of the target protein include ribozyme and endonuclease technologies. A ribozyme molecule can hybridize to a complementary target RNA, followed by endonucleolytic cleavage. A ribozyme cleavage site can be identified within the RNA target and engineered hairpin or hammerhead motif ribozyme molecules can be designed and used to reduce the level of target mRNA encoding the target protein. In some cases, a CRISPR-cas endonuclease system is used to reduce gene expression of the target protein. See for example, U.S. Pat. No. 8,697,359 Bl and US 2014 / 0068797.

[0186] In yet other methods, the target protein can be targeted for degradation by a small molecule induces degradation of the target protein. For example, the small molecule that induces degradation of the target protein can be a molecular glue degrader or a proteolysis targeting chimera (PROTAC). Small molecule degraders are described in Tsai, J.M., et al. (Nature Reviews Molecular Cell Biology, 25:740-757, 2024). Molecular glue degraders are monovalent small molecules that bind to either the ligase or target (without appreciable affinity for the other) and facilitate a neo-interface between the E3 ubiquitin ligase, the target and the molecular glue. PROTACs contain two distinct binding moieties (often referred to as warheads), one binding the E3 ligase and one binding the target, which are connected by a linker to induce proximity.

[0187] According to an aspect of the disclosure, the target protein is ATP binding cassette subfamily C member 1 (ABCC1), or the target gene is ABCC1. ABCC1 has a sequence according to UniProt Accession No: P33527 and is a 1531 amino acid protein. ABCC1 is described in NCBI Gene ID: 4363. ABCC1 Mediates export of organic anions and drugs from the cytoplasm and Mediates ATP-dependent transport of glutathione and glutathione conjugates, leukotriene C4, estradiol- 17-beta-o-glucuroni de, methotrexate, antiviral drugs and other xenobiotics. ABCC1 also hydrolyzes ATP with low efficiency.

[0188] In an aspect, the inhibitor of ABCC1 is selected from biricodar, MK571, probenecide, and reversan, which have structures as follows:MRG: 0680.003419W001 biricodar MK571probenecide reversan

[0189] According to an aspect of the disclosure, the inhibitor of ABCC1 is a nucleic acid molecule that hybridizes to and at least reduces the expression of ABCC1 or an RNA transcript ofABCCl. For example, wherein the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of ABCC1.

[0190] According to an aspect of the disclosure, the target protein is a deubiquitinase selected from ubiquitin specific peptidase 9 X-linked and ubiquitin carboxyl-terminal hydrolase 17-like protein 19, or the target gene are USP9X or USP17L19. USP9X has a sequence according to UniProt Accession No: Q93008-1 and is a 2,554 amino acid protein. USP9X is described in NCBI Gene ID: 8239. Ubiquitin carboxyl-terminal hydrolase 17-like protein 19 has a sequence according to UniProt Accession No: D6RCP7 and is a 530 amino acid protein.USP17L19 is described in NCBI Gene ID: 100287404.

[0191] Ubiquitin specific peptidase 9 X-linked is a deubiquitinase involved both in the processing of ubiquitin precursors and of ubiquitinated proteins and may play a regulatory role at the level of protein turnover by preventing degradation of proteins through the removal of conjugated ubiquitin. Ubiquitin carboxyl-terminal hydrolase 17-like protein 19 is a deubiquitinating enzyme that removes conjugated ubiquitin from specific proteins to regulate cellular processes such as cell proliferation, progression through the cell cycle, apoptosis, cell migration, and the cellular response to viral infection.

[0192] In an aspect, the inhibitor of ubiquitin specific peptidase 9 X-linked is selected from WP1130, FT709, and EOAI3402143 (G9), which have structures as follows:MRG: 0680.003419W001FT709

[0193] According to an aspect of the disclosure, the inhibitor of USP9X or USP17L19 is a nucleic acid molecule that hybridizes to and at least reduces the expression of USP9X or USP17L19 or an RNA transcript of USP9X or USP17L19. For example, wherein the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of USP9X or USP17L19.

[0194] According to an aspect of the disclosure, the target protein is a nucleic acid-associated protein that is DNA-dependent protein kinase catalytic subunit, exosome RNA helicase MTR4, zinc finger protein 717, non-structural maintenance of chromosomes element 3 homolog, and isoleucine— tRNA Ligase, cytoplasmic, or the target gene is PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS.

[0195] DNA-dependent protein kinase catalytic subunit, encoded by PRKDC, has a sequence according to UniProt Accession No: P78527 and is a 4128 amino acid protein. DNA-dependent protein kinase catalytic subunit is a serine / threonine-protein kinase that acts as a molecular sensor for DNA damage. PRKDC is described in GenBank: U47077.5.MRG: 0680.003419W001

[0196] Exosome RNA helicase MTR4, encoded by SKIV2L2, has a sequence according to UniProt Accession No: P42285 and is a 1042 amino acid protein. Exosome RNA helicase MTR4 catalyzes the ATP-dependent unwinding of RNA duplexes with a single-stranded 3' RNA extension. SKIV2L2 is described in NCBI Gene ID: 23517.

[0197] Zinc finger protein 717, encoded by ZNF717, has a sequence according to UniProt Accession No: Q9BY31 and is a 914 amino acid protein. ZNF717 is believed to be a transcriptional regulator. ZNF717 is described in NCBI Gene ID: 100131827.

[0198] Non-structural maintenance of chromosomes element 3 homolog, encoded by NDNL2, has a sequence according to UniProt Accession No: Q96MG7 and is a 304 amino acid protein. NDNL2 is described in NCBI Gene ID: 56160. NDNL2 is a component of the SMC5-SMC6 complex, which is involved in repair of DNA double-strand breaks by homologous recombination.

[0199] Isoleucine-tRNA ligase, cytoplasmic, encoded by IARS1, has a sequence according to UniProt Accession No: P41252 and is a 1262 amino acid protein. Isoleucine-tRNA ligase, cytoplasmic catalyzes the specific attachment of an amino acid to its cognate tRNA in a 2-step reaction: the amino acid (AA) is first activated by ATP to form AA-AMP and then transferred to the acceptor end of the tRNA. IARS1 is described at NCBI Gene ID: 3376.

[0200] In an aspect, the inhibitor of DNA-dependent protein kinase catalytic subunit, encoded by PRKDC, is AZD7648, CC-115, M3814, KU-0060648, NU7026, NU7441 (KU-57788), PI-103, PIK-75 HC1, PP121, SF2523, SU11752, or VX-984, which have structures as follows:MRG: 0680.003419W001SU11752 VX-984

[0201] According to an aspect of the disclosure, the inhibitor of PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS is a nucleic acid molecule that hybridizes to and at least reduces the expression of is PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS or an RNA transcript of PRKDC, SKIV2L2, ZNF717, NDNL2, ox IARS. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS.MRG: 0680.003419W001

[0202] According to an aspect of the disclosure, the target protein is Bcl-2-like protein 1 encoded by BCL2L1. Bcl-2-like protein 1 has a sequence according to UniProt Accession No: Q07817and is a 233 amino acid protein. Bcl-2-like protein 1 is an inhibitor of cell death and inhibits activation of caspases. BCL2L1 is described in NCBI Gene ID: 598.

[0203] In an aspect, the inhibitor of Bcl-2-like protein 1 is A-l 155463, A-1331852, ABT263 (navitoclax), or ABT199 (Venetoclax), which have structures as follows:ABT263 (navitoclax)

[0204] According to an aspect of the disclosure, the inhibitor of BCL2L1, is a nucleic acid molecule that hybridizes to and at least reduces the expression of BCL2L1 or an RNA transcript of BCL2L1. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of BCL2L1.

[0205] According to an aspect of the disclosure, the target protein is cytohesin-interacting protein, encoded by CYTIP, or the target gene is CYTIP. Cytohesin-interacting protein has a sequence according to UniProt Accession No: 060759 and is a 359 amino acid protein.Cytohesin-interacting protein binds to cytohesin-1 (CYTH1) and modifies activation of ADP ribosylation factors (ARFs) by CYTH1 and thought to sequester CYTH1 in the cytoplasm.CYTIP is described in NCBI Gene ID: 9595.MRG: 0680.003419W001

[0206] According to an aspect of the disclosure, the inhibitor of CYTIP, is a nucleic acid molecule that hybridizes to and at least reduces the expression of CYTIP or an RNA transcript of BCL2L1. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of CYTIP.

[0207] According to an aspect of the disclosure, the target protein is Vesicle transport protein GOT1 A, encoded by GOLT1A, or the target gene is GOLT1A. GOLT1 A has a sequence according to UniProt Accession No: Q6ZVE7 and is a 132 amino acid protein. GOLT1A is involved in fusion of ER-derived transport vesicles with the Golgi complex. GOLT1A is described in NCBI Gene ID: 127845.

[0208] According to an aspect of the disclosure, the inhibitor of GOLT1A, is a nucleic acid molecule that hybridizes to and at least reduces the expression of GOLT 1 A or an RNA transcript of GOLT1A. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of GOLT1A.

[0209] According to an aspect of the disclosure, the target protein is moesin-ezrin-radixin like tumor suppressor, encoded by NF2, or the target gene is NF2. Moesin-ezrin-radixin like tumor suppressor has a sequence according to UniProt Accession No: P35240 and is a 595 amino acid protein. Moesin-ezrin-radixin like tumor suppressor is a putative regulator of the Hippo / SWH (Sav / Wts / Hpo) signaling pathway. NF2 is described in NCBI Gene ID: 4771.

[0210] According to an aspect of the disclosure, the inhibitor of NF2 is a nucleic acid molecule that hybridizes to and at least reduces the expression of NF2, or an RNA transcript of NF2. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of NF2.

[0211] According to an aspect of the disclosure, the target protein is guanine nucleotide exchange protein SMCR8, encoded by SMCR8, or the target gene is SMCR8. SMCR8 has a sequence according to UniProt Accession No: Q8TEV9 and is a 937 amino acid protein.SMCR8 is a component of the C9orf72-SMCR8 complex, a complex that has guanine nucleotide exchange factor (GEF) activity and regulates autophagy. SMCR8 is described in NCBI Gene ID: 203228.

[0212] According to an aspect of the disclosure, the inhibitor of SMCR8 is a nucleic acid molecule that hybridizes to and at least reduces the expression of SMCR8 or an RNA transcriptMRG: 0680.003419W001 of SMCR8. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of SMCR8.

[0213] According to an aspect of the disclosure the target protein is ERBB receptor feedback inhibitor 1, encoded by ERRFH, or the target gene is ERRFI1. ERBB receptor feedback inhibitor 1 has a sequence according to UniProt Accession No: Q9UJM3 and is a 462 amino acid protein. ERBB receptor feedback inhibitor 1 is a negative regulator of EGFR signaling for several EGFR family members, including ERBB2, ERBB3 and ERBB4, and inhibits EGFR catalytic activity by interfering with its dimerization. ERRFH is described in NCBI Gene ID: 54206.

[0214] According to an aspect of the disclosure, the inhibitor of ERRFH is a nucleic acid molecule that hybridizes to and at least reduces the expression of ERRFH or an RNA transcript ERRFH. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of ERRFH.

[0215] According to an aspect of the disclosure, the target protein is a tyrosine-protein phosphatase selected from tyrosine-protein phosphatase non- receptor type 12, encoded by PTPN12, or tyrosine- protein phosphatase non- receptor type 11, encoded by PTPN11, or the target gene is PTPN12 or PTPN1L

[0216] Tyrosine-protein phosphatase non-receptor type 11 has a sequence according to UniProt Accession No: Q06124 and is a 593 amino acid protein. Tyrosine-protein phosphatase non-receptor type 11 acts downstream of various receptor and cytoplasmic protein tyrosine kinases to participate in the signal transduction from the cell surface to the nucleus. PTPN11 is described in NCBI Gene ID: 5781.

[0217] Tyrosine-protein phosphatase non-receptor type 12 has a sequence according to UniProt Accession No: Q05209 and is a 780 amino acid protein. Tyrosine-protein phosphatase non-receptor type 12 dephosphorylates cellular tyrosine kinases, such as ERBB2 and PTK2B / PYK2, and thereby regulates signaling via ERBB2 and PTK2B / PYK2. PTPN12 is described in NCBI Gene ID: 5782.

[0218] According to an aspect of the disclosure, the inhibitor of PTPN11 or PTPN12, is a nucleic acid molecule that hybridizes to and at least reduces the expression of PTPN11 or PTPN12 or an RNA transcript of PTPN11 or PTPN12. For example, the nucleic acid moleculeMRG: 0680.003419W001 is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript of PTPN11 or PTPN12.

[0219] According to an aspect of the disclosure, the target protein is a glycosyltransferase selected from dolichyl-diphosphooligo-saccharide— protein glycosyltransferase subunit 2, encoded by RPN2, or dolichyl-phosphate beta-glucosyltransferase, encoded by ALG5, or the target gene is RPN2 or ALG5.

[0220] Dolichyl-diphosphooligosaccharide— protein glycosyltransferase subunit 2 has a sequence according to UniProt Accession No: P04844 and is a 631 amino acid protein.Dolichyl-diphosphooligosaccharide— protein glycosyltransferase subunit 2 is a subunit of the oligosaccharyl transferase (OST) complex that catalyzes the initial transfer of a glycan from the lipid carrier dolichol-pyrophosphate to an asparagine residue within an Asn-X-Ser / Thr consensus motif in nascent polypeptide chains. RPN2 is described in NCBI Gene ID: 6185.

[0221] Tyrosine-protein phosphatase non-receptor type 12 has a sequence according to UniProt Accession No: Q05209 and is a 780 amino acid protein. Tyrosine-protein phosphatase non-receptor type 12 dephosphorylates cellular tyrosine kinases, such as ERBB2 and PTK2B / PYK2, and thereby regulates signaling via ERBB2 and PTK2B / PYK2. ALG5 is described in NCBI Gene ID: 5782.

[0222] According to an aspect of the disclosure, the inhibitor of RPN2 or ALG5, is a nucleic acid molecule that hybridizes to and at least reduces the expression of RPN2 or ALG5or an RNA transcript of RPN2 or ALG5. For example, the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA) directed against an RNA transcript (ARPN2 or ALG5.

[0223] In aspects, methods and compositions of the disclosure administer a HER2 binding agent to the subject for treating the tumor that is HER2 heterogeneous or HER2-low, or a tumor that is resistant to HER2 therapy. The HER2 binding agent can be an anti-HER2 antibody or fragment thereof.

[0224] An anti-HER2 antibody refers to any antibody or antigen-binding fragment thereof that specifically binds to HER2. Anti-HER2 antibodies include monoclonal antibodies, polyclonal antibodies, and antigen-binding fragment that specifically bind to HER2.Exemplary HER2-binding sequences as well as exemplary anti-HER2 antibody sequences are known in the art (see, for example, U.S. Pat. Nos. 5,821,337 and 6,870,034).MRG: 0680.003419W001

[0225] Exemplary HER2 antibodies include those antibodies that are approved for treatment of a HER2 expressing cancer, such as HER2 expressing breast cancer. The approved antibody Trastuzumab (HERCEPTIN) binds to HER2 to cause a signal transduction blockade and prevention of HER2 cleavage. The approved antibody Pertuzumab (PERJET A) blocks HER2 dimerization with other EGF receptors, thereby blocking ligand-activated signaling.Trastuzumab is described in U.S. Pat. Nos. 5,821,337 and 6,870,034. See also and Molina etal., Cancer Res. 61(12):4744-9, 2001). Other HER2 antibodies include disitamab, ABP 980, and DX-CH09.

[0226] In aspects of the disclosure, the anti-HER2 antibody or fragment thereof is conjugated to a drug, such as a cytotoxic agent. Exemplary drugs that can be linked to an anti-HER2 antibody or fragment thereof, include, but are not limited to DM1 (maytansine), DXd (topoisomerase I inhibitor), MMAE (auristatin), DUBA (duocarmycin), TLR7 / 8 agonist, DP104n, tubulysin, and SLTA.

[0227] Various anti-HER2 antibody-drug conjugates are known and include trastuzumab emtansine (T-DM1), trastuzumab deruxtecan (T-DXd), disitamab vedotin (RC48), trastuzumab duocarmazine, TAA013, ZRC-3256, MRG002, ARX788, BDC-1001, A166, FS-1502, SHR-A1201, DP303c, BI-CON-02, ALT-P7, DX126-262, ZW49, HS630, B003, SBT6050, SHR-A1811, MT-5111, and GQ1001. See, for example, Zhang, X., etal. (Antibody Therapeutics, 5(1) 18-29, 2022).

[0228] In aspects, methods and compositions of the disclosure administer a HER2 inhibitor to the subject for treating the tumor that is HER2 heterogeneous or HER2-low. The HERZ inhibitor can be a tyrosine kinase inhibitor (TKI) directed against HER2.

[0229] Exemplary TKIs directed against HER2 include, but are not limited to lapatinib, canertinib, neratinib, tucatinib (or irbinitinib), CP-724714, tarloxitinib, mubritinib, afatinib, varlitinib, and dacomitinib.

[0230] As used herein, a “treatment period” refers to the period of time starting with the initial administration of the inhibitor of a target protein or gene encoding the target protein, increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, and / or HER2 binding agent or HERZ inhibitor.

[0231] The treatment period can be monitored to assess the patient response, including the reduction of HER2 tumor size and / or reduction in number of malignant HER2 breast cancerMRG: 0680.003419W001 cells over course of the treatment period. After treatment is commenced cells can optionally be taken from the patent or the breast cancer tumor can be monitored to determine if the inhibitor of the target protein or target gene, and HER2 binding agent or HER2 inhibitor are providing one or more expected biological responses.

[0232] The treatment period can be adjusted, if necessary, to achieve the therapeutic goal of reduction of HER2 tumor size or elimination of the HER2 tumor in the subject. In some modes of treatment, the treatment period can be at least 7 days, at least 10 days, at least 14 days, at least 21 days, or at least 28 days, or at least one or more months (two, three, four, five, six, etc. months).

[0233] In some modes of treatment, the administration of the inhibitor of a target protein or gene encoding the target protein does not completely coincide with the administration of the HER2 binding agent or HER2 inhibitor. For example, the inhibitor of a target protein or gene encoding the target protein can be administered on one or more days that are different from the one or more days of administration of the HER2 binding agent or HER2 inhibitor. In other modes of treatment for at least a portion of the treatment period there may be coinciding administration of the inhibitor of a target protein or gene encoding the target protein and the HER2 binding agent or HER2 inhibitor.

[0234] In aspects of the disclosure, administration of the inhibitor of the target protein or gene is performed during a period that at least partially overlaps with a period of administration of the HER2 binding agent or HER2 inhibitor. In aspects, administration of the inhibitor of the target protein or gene occurs at least in part prior to administration of the HER2 binding agent or HER2 inhibitor. For example, administration of the inhibitor of the target protein or gene is carried out for a period of time at least sufficient for the inhibitor to increase sensitivity of the tumor to the HER2 binding agent or HER2 inhibitor. For example, the initial administration of the inhibitor of a target protein or gene encoding the target protein is performed at least 12 hours, at least 24 hours, at least 36 hours, at least 48 hours before administering the HER2 binding agent or HER2 inhibitor. Prior to the administration of the HER2 binding agent or HER2 inhibitor, HER2 cells can be sensitized towards by the inhibitor of a target protein or gene encoding the target protein, making them more responsive to the cytotoxic action of HER2 binding agent or HER2 inhibitor when it is administered.MRG: 0680.003419W001

[0235] The treatment period can be monitored to assess the patient response, including the reduction of HER2 tumor cells over course of the treatment period. After treatment commences the tumor(s) can be monitored to determine if the inhibitor of a target protein or gene encoding the target protein and / or HER2 binding agent or HER2 inhibitor are providing one or more expected biological responses. The treatment period can be adjusted, if necessary, to achieve the therapeutic goal of reduction of malignant myeloid cells in the subject.

[0236] Administration can provide a therapeutically active amount of the inhibitor of the target protein or gene in the body over a first period of time, and administration provides a therapeutically active amount of the HER2 binding agent or HER2 inhibitor over a second period of time, and the first period of time overlaps with the second period of time.

[0237] Dosages of the inhibitor of a target protein or gene encoding the target protein and / or HER2 binding agent or HER2 inhibitor used in methods of the disclosure can vary depending on one or more factors such as the type of drug, the condition of the subject, the disease state, the administration route chosen, etc. The inhibitor of a target protein or gene encoding the target protein and / or HER2 binding agent or HER2 inhibitor used in methods of the disclosure can be administered in “therapeutically effective amount” or “therapeutically effective amounts” which can vary from subject to subject.

[0238] Aspect 1 is a method of treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy, the method comprising administering to the subject an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 antagonist, and administering the HER2 binding agent or HER2 antagonist to the subject.

[0239] Aspect 2A is the method of aspect 1 wherein the target protein ATP binding cassette subfamily C member 1 (ABCC1), or the target gene is ABCC1. Aspect 2B is the method of aspect 1 wherein the target protein is encoded by a deubiquitinase that is ubiquitin specific peptidase 9 X-linked or ubiquitin carboxyl-terminal hydrolase 17-like protein 19, or the target gene is USP9X or USP17L19. Aspect 2C is the method of aspect 1 wherein the target protein is a nucleic acid-associated protein that is DNA-dependent protein kinase catalytic subunit, exosome RNA helicase MTR4, zinc finger protein 717, non-structural maintenance of chromosomes element 3 homolog, and isoleucine— tRNA Ligase, cytoplasmic, or the target geneMRG: 0680.003419W001 is PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS. Aspect 2D is the method of aspect 1 wherein the target protein is Bcl-2-like protein 1, or the target gene is BCL2L1. Aspect 2E is the method of aspect 1 wherein the target protein is cytohesin-interacting protein, or the target gene is CYTIP. Aspect 2F is the method of aspect 1 wherein the target protein is vesicle transport protein GOT1 A, or the target gene is GOLT 1 A. Aspect 2G is the method of aspect 1 wherein the target protein is moesin-ezrin-radixin like tumor suppressor, or the target gene is NF2. Aspect 2H is the method of aspect 1 wherein the target protein is guanine nucleotide exchange protein SMCR8, or the target gene is SMCR8. Aspect 21 is the method of aspect 1 wherein the target protein is ERBB receptor feedback inhibitor 1, or the target gene is ERRFIL Aspect 2 J is the method of aspect 1 wherein the target protein is a tyrosine-protein phosphatase that is tyrosineprotein phosphatase non-receptor type 12 or tyrosine-protein phosphatase non-receptor type 11, or the target gene is PTPN12 or PTPN11. Aspect 2K is the method of aspect 1 wherein the target protein is a glycosyltransferase that is dolichyl-diphosphooligo-saccharide-protein glycosyltransferase subunit 2 or dolichyl-phosphate beta-glucosyltransferase, or the target gene is RPN2 or ALG5.

[0240] Aspect 3 A is the method of any one of aspects 1-2K, wherein the tumor that is HER2-low, HER2 heterogeneous, or wherein the tumor is resistant to anti-HER2 therapy, is either hormone receptor positive (HR+) or hormone receptor negative (HR). Aspect 3B is the method of any one of aspects 1-3 A, wherein the subject having the tumor that is HER2 heterogeneous or HER2-low, or resistant to HER2 therapy, was diagnosed for HER2 expression using fluorescence in situ hybridization (FISH) and / or immunohistochemistry (IHC).

[0241] Aspect 4A is the method of any one of aspects 1-2K, wherein the tumor that is resistant to anti-HER2 therapy is resistant to anti-HER2 antibody therapy or small molecule inhibitor of HER2 therapy. Aspect 4B is the method of aspect 4A, where the tumor has resistance to trastuzumab, trastuzumab-deruxtecan (T-DXd), trastuzumab emtansine (T-DM1), disitamab, pertuzumab, DX-CH09, lapatinib, neratinib, tucatinib, afatinib (BIBW2992), CUDC-101, or canertinib therapy. Aspect 4C is the method of aspect 4A or 4B, wherein the tumor has developed resistance to the anti-HER2 therapy or small molecule inhibitor of HER2 therapy which was used in combination with a chemotherapeutic agent.

[0242] Aspect 5A is the method of any one of aspects 1-4C, wherein the subject that comprises the tumor has cancer that is breast cancer, gastric cancer, lung cancer, salivary cancer,MRG: 0680.003419W001 vaginal cancer, bladder cancer, endometrial cancer, cervical cancer, or colorectal cancer. Aspect 5B is the method of aspect 5A, wherein the cancer is breast cancer. Aspect 5C is the method of aspect 5B, wherein the subject that has a tumor characterized as HER2-low has triple-negative breast cancer (TNBC).

[0243] Aspect 6A is the method of any one of aspects 1-5C, wherein the inhibitor of a target protein (a) is not a polypeptide, (b) is a small molecule, or both (a) and (b). Aspect 6B is the method of aspect 6A, wherein the small molecule has a molecular weight in the range of 100 g / mol to 1200 g / mol, in the range of 100 g / mol to 1000 g / mol, or in the range of 100 g / mol to 800 g / mol.

[0244] Aspect 7A is the method of any one of aspects 2A or 3A-6B, wherein the inhibitor of ABCC1 is biricodar, MK571, probenecide, or reversan. Aspect 7B is the method of any one of aspects 2B or 3A-6B, wherein the inhibitor of USP9X is WP1130, G9, FT709, or EOAI3402143. Aspect 7C is the method of any one of aspects 2C or 3A-6B, wherein the inhibitor ofPRKDC is AZD7648, CC-115, M3814, KU-0060648, NU7026, NU7441 (KU-57788), PI-103, PIK-75 HC1, PP121, SF2523, SU11752, or VX-984. Aspect 7D is the method of any one of aspects 2C or 3A-6B, wherein the inhibitor of isoleucine— tRN A Ligaseis reveromycin. Aspect 7E is the method of any one of aspects 2D or 3A-6B, wherein the inhibitor Bcl-2-hke protein 1 is A-l 155463, A-1331852, ABT263, or ABT199.

[0245] Aspect 8A is the method of aspect 6A or 6B, wherein the small molecule induces degradation of the target protein. Aspect 8B is the method of aspect 8A, wherein the small molecule that induces degradation of the target protein is a molecular glue degrader or a proteolysis targeting chimera (PROTAC).

[0246] Aspect 9A is the method of any one of aspects 1-5C, wherein the inhibitor of the target gene is a nucleic acid molecule that hybridizes to and at least reduces the expression of a target gene or an RNA transcript of the target gene. Aspect 9B is the method of aspect 9A, wherein the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA).

[0247] Aspect 10A is the method of any one of aspects 1-9B, wherein the HER2 binding agent is an anti-HER2 antibody or an antigen-binding fragment thereof. Aspect 10B is the method of aspect 10A wherein the anti-HER2 antibody is trastuzumab, disitamab, pertuzumab, or DX-CHO9, or an antigen binding fragment thereof. Aspect 10C is the method of aspect 10AMRG: 0680.003419W001 or 10B, wherein the anti-HER2 antibody or fragment thereof is linked to a drug. Aspect 10D is the method of aspect 10C, wherein the drug is a cytotoxic agent. Aspect 10E is the method of aspect 10D, wherein the drug is DM1 (maytansine), DXd (topoisomerase I inhibitor), MMAE (auristatin), DUBA (duocarmycin), TLR7 / 8 agonist, DP104n, tubulysin, or SLTA. Aspect 10E is the method of aspect 10C or 10D, wherein the anti-HER2 antibody or fragment thereof that is linked to a drug is selected from trastuzumab emtansine (T-DM1), trastuzumab deruxtecan (T-DXd), disitamab vedotin (RC48), trastuzumab duocarmazine, TAA013, ZRC-3256, MRG002, ARX788, BDC-1001, A166, FS-1502, SHR-A1201, DP303c, BI-CON-02, ALT-P7, DX126-262, ZW49, HS630, B003, SBT6050, SHR-A1811, MT-5111, and GQ1001.

[0248] Aspect 11 A is the method of any one of aspects 1 -9B, wherein the HER2 antagonist is a tyrosine kinase inhibitor (TKI) directed against HERZ. Aspect 1 IB is the method of aspect 11 A, wherein the TKI directed against HERZ is lapatinib, neratinib, or tucatinib.

[0249] Aspect 12A is the method of any one of aspects 1-1 IB, wherein administration of the inhibitor of the target protein or gene is performed during a period that at least partially overlaps with a period of administration of the HERZ binding agent or HER2 antagonist. Aspect 12B is the method of any one of aspects 1-12A, wherein administration of the inhibitor of the target protein or gene occurs at least in part prior to administration of the HERZ binding agent or HERZ antagonist. Aspect 12C is the method of any one of aspects 12A or 12B, wherein administration of the inhibitor of the target protein or gene is carried out for a period of time at least sufficient for the inhibitor to increase sensitivity of the tumor to the HERZ binding agent or HERZ antagonist. Aspect 12D is the method of any one of aspects 1-12C, wherein administration provides a therapeutically active amount of the inhibitor of the target protein or gene in the body over a first period of time, and administration provides a therapeutically active amount of the HERZ binding agent or HERZ antagonist over a second period of time, and the first period of time overlaps with the second period of time. Aspect 12E is the method of aspect 12D, wherein the first period of time starts before the second period of time.

[0250] Aspect 13A is the method of any one of aspects 1-12E, wherein the inhibitor is administered from a composition that is different from a composition used to administer the HERZ binding agent or HER2 antagonist.

[0251] Aspect 14 is a pharmaceutical composition comprising an inhibitor of a target protein or gene encoding the target protein, wherein the inhibitor increases sensitivity of theMRG: 0680.003419W001 tumor to a HER2 binding agent or HER2 antagonist, and a HER2 binding agent or HER2 antagonist.

[0252] Aspect 15A is use of pharmaceutical compositions for treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy, wherein the one or more pharmaceutical composition(s) comprise (a) an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 antagonist, and (b) a HER2 binding agent or HER2 antagonist. Aspect 15B is use of pharmaceutical compositions according to Aspect 14 A, wherein (a) the subject has any one more of feature(s) of Aspects 3A-5C, (b) the compositions has any one or more compound(s) of Aspects 6A-1 IB, (c) use is carried out using any feature of Aspects 11 A-12E, or any combination of (a)-(c)EXAMPLES

[0253] Methods

[0254] Cell Culture

[0255] Human breast cancer cells lines MDA-MB-231 (RRID: CVCL 0062), HCC 1954 (RRID: CVCL 1259), BT-474 (RRID :CVCL_0179), ZR-75-30 (RRID :CVCL_1661), SK-BR-3 (RRID: CVCL 0033), MDA-MB-361 (RRID:CVCL_0620) and MDA-MB-453 (RRID:CVCL_0418) were purchased from ATCC and 21NT (RRID:CVCL_7933) and 21PT (RRID:CVCL_7934) cells were provided by Arthur Pardee, Dana-Farber Cancer Institute. Cells were cultured in conditions recommended by the supplier and routinely tested for mycoplasma and murine pathogens. 21NT and 21PT cells were cultured with DMEM / F12, 5% Horse Serum, 10 pg / mL insulin, 20 ng / mL EGF and 0.5 pg / mL hydrocortisone. All cell lines were supplemented with 50 U / mL penicillin.

[0256] Viral Infection and constructs

[0257] For virus production, HEK293FT cells (RRID:CVCL_6911) were transfected using Lipofectamine 3000 (Thermo Fisher Scientific) with lentiviral packaging plasmids pMD2.G (RRID:Addgene_12259) and psPAX2 (RRID :Addgene_l 2260) and indicated plasmids. The viral supernatant was collected 48h hours later, filtered through a 0.45 pm syringe filter and added to cells with 8 pg / mL Polybrene (Millipore). For expression vectors H2B-mCherry (pLV-mpgk-H2B-mCherry-Luc2, designed and made using VectorBuilder) and H2B-GFP (pLV-MRG: 0680.003419W001 mpgk-H2B-GFP-Luc2, VectorBuilder), cells were sorted for GFP or mCherry expression 4 days after infection. For CRISPR KO pools, cells were selected 72h later using 2 pg / mL puromycin. To validate the CRISPR screen hits, 2 most depleted guides were selected: ABCC1 (1) AATGTTCAGGGGAAACCGG and (2) TGGGTCCAGGTTCATTCGG, USP9X (1) CTCCTGGCAAACTCACAGG and (2) CAAGTGCTATATCTAACAG, ROSA26 control GGTGATCTAGTATTTCTTG Guides were cloned in lentiCRISPRv2 (Addgene #52961) following the recommended protocol (Sanjana, N.E., et al. Nature Methods;! l(8):783-4, 2014).

[0258] Flow Cytometry and Cell Sorting (FACS)

[0259] The panel of HER2+ breast cancer cell lines and MDA-MB-231 were stained using anti-HER2 PE or isotype control PE for 20 min at room temperature. Flow cytometry was performed using LSRFORTESSA (BD Bioscience, RRID:SCR_018655), analysis was done with FLOWJO (version 10, RRID:SCR_008520). To generate stable HER21" and HER210cell lines, HCC1954, 21NT and 21PT expressing H2B-mCherry or H2B-GFP were stained with anti-HER2 APC and sorted for either high or low HER2 expression levels using the BD FACS ARIA II cell sorter (RRID:SCR_018934) as depicted in Figure 1A. After sorting, the cells were grown and resorted for HER2 levels 4 other times.

[0260] For quantification of the tumor composition by flow cytometry, tumors were homogenized using micro pestles immediately after collection and digested for 1 hour with digestion media (2% wt / vol collagenase IV, 2% wt / vol hyaluronidase, and 2% wt / vol bovine serum albumin in DMEM / F12) at 37° C on a shaker. The solutions were filtered through a 500 pm mesh (Tetko, 03-500-47), washed with PBS, and frozen in 10% dimethyl sulfoxide / fetal bovine serum at -80°C to preserve before flow cytometry of all samples at once. For the acquisition, cells were thawed, washed with PBS and passed through a 35 pm cell strainer tube (Falcon) and analyzed using an LSRFORTESSA (BD 800 Biosciences, RRID:SCR_018655). All flow cytometry experiments were performed at the DFCI Flow Cytometry Core (RRID:SCR_009751).

[0261] Fluorescent in situ hybridization (FISH)

[0262] Cells were resuspended in 75 mM KC1, incubated for 10-20 minutes at room temperature and washed in fixative (3:1 methanol-to-acetic acid). The cells were then spread onto a slide, allowing the fixative to evaporate. Slides were washed with water, incubated with proteinase K solution in PBS at 37°C for 2 minutes, followed by post-fixation in ice-coldMRG: 0680.003419W001 Carnoy’s solution (3:1 ethanol-to acetic acid). The slides were then subjected to ethanol washes and air-dried. Probe mix from PATHVYSION HER2 DNA Probe Kit (Abbott) was applied to the slides, covered with a coverslip, and sealed with rubber cement. The slides were incubated at 75°C for 7 minutes and then at 37°C overnight in a humidified chamber. After hybridization, coverslips and rubber cement were removed, and slides were placed in a 0.4x SSC / 0.3% NP-40 wash solution at room temperature for 2 minutes, followed by incubation for 2 minutes in preheated at 74°C 0.4x SSC / 0.3% NP-40 wash solution. Next, slides were washed 2 times for 2 minutes in SSC / 0.1% NP-40 at room temperature, followed by a 3-minute wash in 2x SSC at room temperature. Slides were then mounted using DAPI mounting medium (Vector Laboratories) and imaged using Nikon ECLIPSE Ti2-E fluorescence microscope for quantification and confocal Zeiss 980 (RRID:SCR_025048) from the Molecular Cancer Imaging Facility in Dana-Farber Cancer Institute for figure image.

[0263] Western blot / Immunoblot analyses

[0264] Cells were lysed with RIPA lysis buffer (50 mM Tris pH 7.6, 0.1% SDS, 1% NP-40, 150 mM NaCl, 5 nM EDTA), supplemented with phosphatase and protease inhibitors (Fisher Scientific). Protein concentration was determined using the Pierce 660 nm Protein Assay 824 Reagent (Thermo Fisher), and proteins were heat-denatured in the presence of 0-mercaptoethanol. Equal amounts of protein were loaded onto a NuPAGE Novex 8% or 4- 12% gradient Bis-Tris gel (Fisher Scientific) and separated by electrophoresis. Proteins were transferred to PVDF membranes (Bio-Rad) using a wet NuPAGE transfer system, with transfer buffer containing 20% methanol, for 1 hour 30 minutes at 90 V. The membranes were then blocked with blocking buffer (5% BSA in 0.1% Tween20 TBS (TBST)) for 1 hour at room temperature, followed by overnight incubation with primary antibodies in blocking buffer.Membranes were washed and incubated for 30 minutes at room temperature with the appropriate secondary antibodies. After additional washing, the membranes were developed using CLARITY Western ECL substrate (Bio-Rad) and imaged with the CHEMIDOC MP imaging system (BioRad, RRID:SCR_ 019037). Quantification of immunoblots were performed using the Gels function of FIR (RRID:SCR_002285).

[0265] RNA-seq

[0266] RNA was isolated using the RNeasy Mini Kit (QIAGEN) and libraries were prepared using the KAPA mRNA HyperPrep (Roche) strand-specific sample preparation kitsMRG: 0680.003419W001 from 100 ng of purified total RNA, following the manufacturer’s protocol on a BIOMEK i7 (Beckman Coulter). The resulting double stranded DNA libraries were quantified using a QUBIT fluorometer (Thermo Fisher Scientific) and the 4200 TAPESTATION (Agilent, RRID:SCR_018435). The final library pool was sequenced on an Illumina NOVASEQ 6000 (Illumina, RRID:SCR_016387), targeting 40 million 150-bp read pairs per library at the Dana-Farber Cancer Institute Molecular Biology Core Facility (RRID:SCR_009754). RNA-seq data were processed using the VIPER pipeline (Cornwell, M. et al., BMC Bioinformatics 19(1): 135, 2018). Fastq files were aligned to the human reference GRCh37 / hgl9 genome with the STAR RNA-Seq aligner (version STAR_2.5.1b, RRID:SCR_004463) (Dobin, A., etal., Bioinformatics 29(1):15-21, 2013), followed by transcript assembly with Cufflinks v2.2.1 (RRID:SCR_014597) and RSeQC v2.6.2 (RRID:SCR_005275) (Wang, L., et al., Bioinformatics 28(16):2184-5, 2012).

[0267] Differential gene expression analyses were conducted on absolute gene counts 901 for RNA-Seq data and raw read counts for transcriptomic profiling data using DESeq2 vl.18.1 (RRID:SCR_000154) (Love, M.I., et al., Genome Biol 15(12):550, 2014). Gene set enrichment analysis (GSEA, RRID:SCR_003199) was performed using the Broad GSEA Application (GSEA Java; v4.1.0) with Hallmark gene sets v7.4 (Dobin, A., et al., Bioinformatics 2013;29(l): 15-21; Wang., L., etal., Bioinformatics 2012;28(16):2184-5). PAM50 subtype prediction was carried out using the GENEFU package, and the resulting subtype probabilities were used to estimate the likelihood of each molecular subtype for each sample.

[0268] Drug dose response and growth assay, treatment

[0269] 1000 cells were plated in 96-well plates (6 days experiments) or in 48-well plates (9-12 days growth assays). Trastuzumab deruxtecan (T-DXd, MedChemExpres HY-138298) and trastuzumab emtansine (T-DM1, MedChemExpress HY-P9921) were in dissolved PBS.Neratinib (MedChemExpress HKI-272), irbinitinib (tucatinib, Selleckchem S8362), deruxtecan (DXd, MedChemExpress HY-13631E), paclitaxel (Sigma T7191), G9 (EOAI3402143, MedChemExpress HY-111408), FT709 (MedChemExpress HY-145967), reversan (Selleckchem E1742) and biricodar (Selleckchem E1781) were dissolved in dimethyl sulfoxide. G9, FT709, reversan, and biricodar were used at a concentration of 1 pM. Plates were imaged with Celigo Image Cytometer (Nexcelom Bioscience, RRID:SCR_018808) to count cells based on the nuclear H2B-mCherry or H2B-GFP signal from the cells, or Hoechst 33342 (Life Technologies)MRG: 0680.003419W001 staining. For pulse-chase experiments, cycloheximide (Sigma, 100 pg / mL) was added in 6 well plates for the indicated times.

[0270] Generation of resistant cell lines

[0271] HCC1954, 21NT, and 21PT cells were culture for over 3 months in total starting with DMSO (Ctl) or doses close to IC50 (tucatinib 1 pM; neratinib 100 nM; T-DM1 10 ng / mL for HCC1954 and 100 ng / mL for 21NT and 21PT; T-DXd 100 ng / mL for HCC1954 and 1 pg / mL for 21NT and 21PT) and then increased after 2-3 passages when cells were growing (tucatinib 3 pM; neratinib 300 nM; T-DM1 30 ng / mL for HCC1954 and 300 ng / mL for 21NT and 21PT; T-DXd 300 ng / mL for HCC1954 and 3 pg / mL for 21NT and 21PT). For maintenance, cells were kept in their initial doses close to IC50.

[0272] Mouse model and tumorigenesis assay

[0273] Animal experiments were conducted in accordance with protocol #11-023, approved by the DFCI Institutional Animal Care and Use Committee, and in compliance with NIH guidelines. NOD.Cg-Prkdcscid I12rgtmlWjl / SzJ (NSG) mice, aged 5-6 weeks, were purchased from the Jackson Laboratory (RRID:IMSR_GPT:T005557). For mammary fat pad injection, mice were anesthetized with continuous isoflurane inhalation. IM cells in 50 pL of PBS 50% Matrigel (Corning) were injected in both abdominal mammary glands using insulin syringes and the incision was closed with surgical clips. Post-operative analgesia was provided through the application of topical ropivacaine. Tumor sizes where measure using a caliper. At end point, tumor fluorescence where visualized using CHEMIDOC MP imaging system (BioRad).

[0274] In vivo treatment

[0275] Mice bearing tumors were treated with neratinib (MedChemExpress HKI-272) at 40 mg / kg, administered daily by oral gavage in 200 pL of 0.5% methylcellulose or vehicle control. Trastuzumab deruxtecan (T-DXd , MedChemExpress HY-138298) was administered once weekly via tail vein injection at a dose of 10 mg / kg in 100 pL PBS. For combination experiments, a reduced dose of 5 mg / kg T-DXd was used. G9 (EOAI3402143;MedChemExpress, HY-111408) was administered intraperitoneally at 15 mg / kg in 200 pL every other day. Reversan (MedChemExpress, HY-107643) was administered daily at 10 mg / kg. BothMRG: 0680.003419W001 G9 and Reversan were prepared in a vehicle consisting of 5% DMSO and 40% PEG400 in PBS and sonicated to ensure proper dissolution.

[0276] IHC / Immunohistostaining

[0277] Tissue sections were deparaffinized using xylene and rehydrated in ethanol baths, and antigen retrieval was carried out using Target Retrieval Solution, pH 6 (Agilent) for 20 minutes in a steamer. Slides were incubated in blocking buffer (5% normal goat serum in TBST) for Ih and incubated overnight with indicated primary antibodies in blocking buffer. Secondary antibodies were applied in blocking buffer for 2 hours. Endogenous fluorescence was quenched using a TRUEVIEW Autofluorescence Quenching Kit (Vector Laboratories) for 5 minutes. Images were captured using a Nikon ECLIPSE Ti2-E fluorescence microscope. Quantification of PCNA intensity and positivity in the GFP+ or mCherry+ nuclei were performed using Image!

[0278] CRISPR Screen

[0279] For the pooled genome-wide CRISPR screen, the human CRISPR knockout library H3 (Addgene #133914) was used following the provided protocol. Briefly, 120 million 21NT HER210cells were infected with the H3 library at a MOI of 0.3, ensuring a high probability that most cells received only one sgRNA. After puromycin selection, the cells were cultured for one month in the presence of with vehicle or T-DXd (500 ng / mL). This experiment was run in biological duplicate, then genomic DNA was extracted using phenol / chloroform / isoamyl alcohol. Library preparation PCR was performed on the genomic DNA to construct the sequencing library as recommended in the provided protocol, increasing the number of cycles for the first PCR to 24. NextSeq 500 SE 75 sequencing was performed by the Molecular Biology Core Facilitiy at Dana-Farber Cancer Institute (RRID:SCR_009754). CRISPR data were analyzed using MAGeCK pipeline (RRID:SCR_025016) (Li, W., et al. Genome Biology 15(12):554, 2014). In brief, raw sequencing data were processed to obtain read counts for each sgRNA after normalization using control sgRNAs of the library. Maximum-likelihood analysis of gene essentialities test (MAGeCK MLE) and sgRNA enriched only in T-DXd (bottom center of the MLE scatter plot) or only in control (top center) were considered. Gene set enrichment analysis of associated hits was performed using MAGeCK GSEA tool. The MAGeCK Flute package was used for data visualization.

[0280] Immunoprecipitation and Co-ImmunoprecipitationMRG: 0680.003419W001

[0281] For immunoprecipitation, cells were lysed using Co-IP buffer (50 mM NaCl, 10 mM Tris pH8, 1% NP-40, 10% glycerol) including phosphatase and protease inhibitors. Lysates were incubated for Ih with 10 pL of anti-USP9X (Santa Cruz sc-365353, RRID:AB_10846088), anti-mouse IgG antibody (Santa Cruz sc-2025, RRID: AB 737182), anti-Human IgG antibody [IG266] (Abeam ab200699, RRID: AB_3698401) and for HER2 IP, cells were treated for 4h with T-DXd (10 pg / mL). Thereafter, lysates were incubated 30 min with Pierce™ Protein A / G Magnetic Beads and washed were made with Co-IP buffer using a magnetic rack. Proteins were then heat-denatured in the presence of P-mercaptoethanol and ran on gels, see Immunoblot section for more information.

[0282] Lysosome Enrichment

[0283] The lysosome enrichment kit (Thermo Scientific) was used following the company recommendations. Briefly, 8-12M cells were lysed by sonication (Qsonica Q125 with cooled bath) using 2s pulse at 20% amplitude with 3 s pause between each pulse for 1 min. Cell lysates were put on the top of the generated gradient and centrifuge in Beckman ultracentrifuge using SW 55 Ti rotor at 34600 rpm for 2h at 4°C. Lysosome pellets and total cells were lysed using RIPA buffer and ran on gels as described in the Immunoblot section.

[0284] Short tandem repeat (STR) and Whole Exome Sequencing (WES)

[0285] DNA isolation from cells was performed using DNEASY Blood & Tissue Kit (Qiagen). gDNA was fragmented to 200bp on a Covaris R230 instrument according to manufacturer’s protocol. Libraries were prepared using IDT xGEN 2S Plus DNA reagents on a Beckman Coulter Biomek i7 liquid handling platform from approximately 200 ng of DNA according to manufacturer’s protocol with 14 cycles of PCR amplification. Finished libraries were quantified by Qubit fluorometer and fragment size distribution was evaluated by Agilent TAPESTATION 4200 (RRID:SCR_019547). Libraries were pooled together for target enrichment using TWIST Exome v2.0 reagents and hybrid capture was performed with a 16-hour hybridization incubation using a custom probe panel according to manufacturer’s protocol. Postcapture library pools were quantified by Qubit fluorometer and Agilent TapeStation 4200 (RRID: SCR 019547). Library pools were further evaluated for quality and pool balance with shallow sequencing on an Illumina MISEQ (RRID:SCR_016379). Subsequently, libraries were sequenced on an Illumina NOVASEQX+ with paired-end 150 bp reads by the Molecular Biology Core Facilities 850 at Dana-Farber Cancer Institute (RRID:SCR_009754).MRG: 0680.003419W001

[0286] Sequenced reads (FASTQ files) were aligned to hg38 version of human genome with BWA853 MEM v.0.7.15 (RRID:SCR_022192) and preprocessed following GATK best practices (DePristo, M.A., et al., Nature Genetics 43(5):491-8, 2011). The quality of alignment and potential sample contamination was evaluated with the following software tools FastQC (RRID:SCR_014583), Picard (RRID:SCR_006525), and FastQ-Screen (RRID:SCR_000141). Short nucleotide variants (point mutations and indels) were called using GATK Mutect2 (RRID:SCR_001876 ) using a “panel of normals” based on samples from the 1000 Genomes Project (RRID:SCR_008801) and provided in the Broad Institute’s public data repository.Identified variants were annotated with VEP (McLaren, W, et al. Genome Biology2016; 17(1 ): 122) and transformed into MAF files with the vcf2maf script. The variant calls were further annotated with ONCOKB (RRID:SCR_014782 ) using the oncokb-annotator python package. Copy number segment files were produced by CNVkit (Talevich, E., et al., PLoS Comput Biol 12(4):el004873, 2016) utilizing a (RRID:SCR_021917) ‘flat’ genome reference file to reduce background noise. The list of the most mutated genes from patients' samples was generated using cBioPortal platform (www.cbioportal.org / , RRID:SCR_014555) selecting for breast cancer patients in TCGA cohort. Finally, the STR profiling was performed by the Dana-Farber Cancer Institute Molecular Diagnostics Labs Research Services.

[0287] CyTOF

[0288] Antibodies used for CyTOF were obtained in carrier-free solutions and conjugated to lanthanide metals by the CyTOF Antibody Resource and Core Facility at Brigham and Women’s Hospital. Cells were cultured in 10 cm dish and gently detached using detach cells with Cell dissociation buffer (Gibson) to preserve surface proteins. Approximately 1 x 106cells per condition were incubated with 100 pmol / L Rhodium intercalator (A103Rh; Fluidigm) for 30 minutes at 37°C in complete media. Samples were then barcoded using the Cell-ID 20-Plex Pd Barcoding Kit (Fluidigm) per manufacturer’s instructions and barcoded samples were pooled and processed together. Cells were fixed with 1.6% paraformaldehyde (Electron Microscopy 875 Sciences) for 10 minutes, blocked with Human TruStain FcX (BioLegend) for 10 minutes, and stained with surface antibodies for 30 minutes at room temperature. Following surface staining, cells were permeabilized in methanol on ice for 10 minutes and incubated with intracellular antibody cocktail for 30 minutes. Cells were then fixed overnight at 4°C in Fix and Perm Buffer (Fluidigm) containing 62.5 nmol / L Intercalator-IR (Fluidigm). Prior to acquisition, samples wereMRG: 0680.003419W001 washed in Cell Acquisition Solution (CAS; Fluidigm), resuspended in CAS with EQ Four Element Calibration Beads (1:10 dilution; Fluidigm), and filtered through a 35 pm strainer. Data were acquired on a HELIOS CyTOF mass cytometer (Fluidigm, RRID:SCR_019916), normalized as previously described, and analyzed using Cytobank (RRID:SCR_014043). All staining steps and washes were performed using Cell Staining Media (PBS with 0.5% BSA and 0.02% sodium azide). For analysis, cells were gated for intact singlets, viable (RhAl 03-negative), and non-apoptotic (cleaved PARP-negative) populations. viSNE clustering was performed on the c-PARP-negative population.

[0289] CycIF and spatial analysis

[0290] For the heterogeneous tumor images, cells were annotated according to the nuclei staining of mCherry (HER2hl), GFP (HER210) or only positive for Dapi (Other). Cell nuclei images were segmented and labeled using CELLPOSE 4.0.4 pretrained model with flow threshold set to 0.4 and cellprob threshold to 0. Centroid were computed using scikit- image 0.24.0. For the CycIF cohort, data annotations and coordinates from the original paper were used (Janiszewska, M., et al., JCI Insight, 6(11) doi 10.1172, 2021). For both sets, spatial analysis was performed using scimap 2.3.4 in python 3.9 (RRID:SCR_008394). Preprocessing of the data sets was done with custom scripts using pandas 2.2.3. Preprocessing involved filtering irrelevant data, removing scenes that contained too few cells of interest. Spatial analysis was accomplished using scimap spatial count function with the KNN method, K=11. The method determines the 10 nearest neighbors for each cell, irrespectively of physical distances. Composition of neighborhoods is then averaged out for each cell type.

[0291] Cell Barcoding and sequencing preparation

[0292] The high-complexity CLONMAPPER library (Gutierrez, C., et al., Nat Cancer 2(7):758-72, 2021) was used. 5e5 cells were infected at a multiplicity of infection (MOI) of 0.1, corresponding to 10% BFP+ cells, to ensure integration of only one barcode per cell. The gating strategies are shown in Fig. 3H. BFP+ sorted cells were minimally expanded to allow the growth of enough cells to start the experiment. 5e5 cells were plated at the start of the experiments and the same amount was kept at each passage. Cells were then cultured for up to 10 passages in DMSO, neratinib (200 nM for HCC1954, 100 nM for 21PT) or T-DXd (50 ng / mL for HCC1954, 1 ug / mL for 21PT). HCC1954 T-DXd treated cells were stopped at 4 passages since HER210cells already took over the co-culture condition. Cells from every passage wereMRG: 0680.003419W001 collected, and for the selected ones, genomic DNA was extracted using phenol / chloroform / isoamyl alcohol, and library preparation PCR was performed on the genomic DNA. The PCR conditions were changed from the previous publication, improving its reliability. A PCR step was added where the barcoding construct is amplified first with smaller primers (actcggtgccactttttcaagttg and gagggcctatttcccatgattccttc). Then a second PCR was performed on the first reaction to add the adapters and barcodes for sequencing as in the original protocol (Gutierrez, C., ibid.).

[0293] Barcoding Analysis

[0294] Constant flanking sequences were trimmed from 151 base pair barcode sequencing reads using CUTADAPT version 4.2 (Martin, M., EMBnet. journal 17(1), 10-12, 20111). The CUTADAPT output was then filtered using seqtk version 1.3 to only retain reads of 20 base pairs, consistent with the length of the barcode sequence. Additionally, barcodes that only had one read across all samples were filtered out. Barcodes were categorized as belonging to the HER21" group if their relative frequency averaged over the pre-treatment HER2hlmonoculture samples was larger than their relative frequency averaged over the pre-treatment HER210monoculture samples, and vice-versa for the HER210group. Plots pertaining to the barcode data were generated in R version 4.4.1, using the packages ggplot2(RRID:SCR_014601) and fishplot (Miller, C.A., etal., BMC Genomics 17(l):880, 2016).

[0295] Diversity was quantified in two ways. Firstly, the minimum number of unique barcodes to jointly constitute at least 90% of the sample (Figs. 4F, 4V) was calculated. Secondly, the Shannon Entropy (Figs. 4F, S4F) was calculated:where pi denotes the relative frequency of the i-th barcode. For co-culture samples relative frequencies and the corresponding Shannon Entropy within the HER21" group and the HER210group were calculated separately.

[0296] Testing for differential barcode selection across conditions. A centered log ratio transform to the barcode frequencies of the top 30 barcodes was applied. For each pair of conditions, these top 30 barcodes and perform Welch’s t-tests comparing the mean prevalence of a given barcode between the two conditions were gone over. All p-values per pairwiseMRG: 0680.003419W001 comparison were then ranked, a Bonferroni correction was applied, and the smallest p-value was reported. P-values below the significance threshold of 5% thus indicate that at least one barcode within the top 30 barcodes was differentially selected between conditions.

[0297] Proximity Ligation Assay (PLA)

[0298] Cells were fixed with 4% paraformaldehyde (PF A), thoroughly washed with PBS, and then permeabilized using TBST (TBS with 0.2% Triton X-100). After blocking in TBST containing 5% goat serum for 1 hour, cells were incubated with primary antibodies overnight at 4°C. Thereafter, probe hybridization, ligation, and amplification steps were carried out strictly following the manufacturer's instructions (DUOLINK In Situ, Sigma). For the negative control, only the HER2 antibody was added, and the rest of the protocol was performed as the other samples. Slides were then mounted with SLOWFADE containing DAPI (Life Technologies). Imaging was performed using Nikon ECLIPSE Ti2-E fluorescence microscope. Quantifications of the PLA intensity were done with ImageJ using a phalloidin staining (Invitrogen) to delimit the cell edges.

[0299] Patient derived organoids (PDOs)

[0300] Primary tumors collected fresh from Brigham and Women’s Hospital 1348 and the patient-derived PDX were dissociated using 2 mg / mL BSA, 2 mg / mL collagenase type IV (Worthington Biochemical Corporation, Cat# LS004189), 2 mg / mL hyaluronidase (Sigma-Aldrich, Cat# H3506) in DMEM / F12 media, filtered through a 500 pm strainer and frozen viable in FBS 10% DMSO. T537 is a grade 3, HER2+ IDC, ER low-positive, untreated. T565 is ER+, PR+ and HER2+ IDC. The patient-derived xenograft Metsl5 is derived from ascites of a tumor classified as ER+, PR+, HER2- treated with chemo, Al and fulvestrant, but shows positive heterogeneous IHC staining for HER2, also confirmed by flow cytometry during this experiment (Aouad, P., et al., Nature Communications 2022;13(l):4975). After recovery, organoids were generated as described (Li, C.M.-C., et al., Cell Reports 2020;33(13)). Brefly, cells were resuspended in Matrigel and 50 pL droplets where place into 24-well plate. 700 pL of organoid culture medium (advanced DMEM / F12 supplemented with 1% GlutaMax (Gibco, Cat# 35050061), 1% HEPES (1 M), 1% penicillin / streptomycin, 0.1% amphotericin B (Gibco, CAS# 1397-89-3), 1XB27 supplement (Gibco, Cat# 17504044), 500 ng / mL recombinant R-spondin (Peprotech, Cat# 120-38), 1.25 mM N-acetylcystine (Sigma-Aldrich, CAS# 616-91-1), 10 mM Nicotinamide (Sigma-Aldrich, CAS# 98-92-0), 1 pM SB202190 (Sigma-Aldrich, CAS# 152121-MRG: 0680.003419W001 30-7), 5 ng / mL FGF7 (Peprotech, Cat# 100-19), 20 ng / mL FGF10 (Peprotech, Cat# 100-26), 5 nM heregulin beta-1 (Peprotech, Cat# 100-03), 5 ng / mL EGF (Peprotech, Cat# AF-100-15)) was added to cover the matrix. Subsequently, the medium was changed every 3 days. For flow cytometry, organoids were dissociated in single cell suspensions using TrypLE Express Enzyme (Gibco) and HER2 staining was performed as described previously. For Metsl5, RPF+ cancer cells were gated to eliminate normal murine cells from the analysis.

[0301] scRNA-seq library preparation

[0302] Tumor samples were digested as described in the PDOs section. To remove debris and non-viable cells, a Percoll density gradient (50% / 40% / 20% layers) was utilized. After a 30-minute centrifugation at 2000xg at 4°C without brakes, cells were recovered from the 20 / 40% interface. The cells were washed, filtered 100 pm, and resuspended in PBS + 0.04% BSA.Around 20,000 cells were processed using the lOx Genomics ChromiumTM instrument (lOx Genomics) according to the manufacturer’s recommendations and the library were generated using Chromium Next GEM Single Cell 5' HT Kit v2 (lOx Genomics). Library quality was verified using Bioanalyzer High Sensitivity DNA Kit (Agilent). Sequencing was performed at the Dana-Farber Molecular Biology Core (RRID:SCR_009754) on an Illumina NovaSeq 6000 platform, generating 150bp paired-end reads at a density of 40 million reads per library. EMBO cohort data were downloaded from Pal, B. et al. (EMBO J., 2021;40(l l):el07333), including only HER2+ and treatment naive tumors.

[0303] scRNA-seq data analysis

[0304] Copy number variation analysis

[0305] Copy number variations (CNVs) were inferred from single-cell RNA sequencing data using the inferCNV R package v 1.24.0 (RRID:SCR_021140). Tumor cells were defined based on the expression of epithelial markers, including EPC AM and keratins (KRT), and were selected from the integrated single-cell dataset. Immune and stromal cell populations, identified by the expression of lineage-specific markers such as PTPRC and COL1A1, were used as reference (normal) cells for CNV inference. Raw gene expression counts were used as input, and genes were ordered according to their chromosomal positions based on the human genome annotation. InferCNV was run with a cutoff value of 0.1, appropriate for lOx Genomics dropletbased scRNA-seq data. Expression values were normalized relative to the reference cell populations to estimate large-scale chromosomal copy number alterations in tumor cells. ToMRG: 0680.003419W001 reduce noise and identify coherent CNV patterns, inferCNV applied smoothing across adjacent genes along each chromosome, followed by unsupervised hierarchical clustering of tumor cells based on their inferred CNV profiles. In addition, a Hidden Markov Model (HMM) was used to segment the genome and classify chromosomal regions into discrete CNV states, including losses, neutral regions, and gains. CNV-1397 defined tumor subclones were subsequently identified and used for downstream analyses and visualization.

[0306] CellChat

[0307] Processed count matrices from each cohort were first merged using Seurat (RRID:SCR_007322) (Satija, R., et al., Nature Biotechnology 2015;33(5):495-502). Low-quality cells were filtered using the following criteria: nCount_RNA > 1,000, nFeature_RNA < 1,000, mitochondrial gene percentage < 20%, ribosomal gene percentage < 40%, and logl0(genes per UMI) > 0.8. Data normalization was performed using SCTransform (RRID:SCR_022146), regressing out ribosomal percentage, mitochondrial percentage, nCount RNA, and nFeature RNA. Dimensionality reduction and clustering were performed using the top 30 principal components. Cell clusters were manually annotated using gene module scores based on curated marker gene sets. UMAP visualization was generated using the scCustomize package (RRID:SCR_024675). Cell-cell communication analysis was performed using CellChat (RRID:SCR_021946) (Jin, S., et al., Nature Communications 2021 ; 12(1): 1088) to infer ligand receptor interactions between cell types. Overall information flow for each signaling pathway was extracted from the CellChat object, and differential pathway activity was assessed between HER2hi and HER21o epithelial cells.

[0308] Reverse Phase Protein Array (RPPA)

[0309] HER2hi and HER21o cells were place in mono or co-culture (1 : 1) for 8 days before collecting and sending cells to the Functional Proteomics RPPA Core Facility at MD Anderson Cancer Center (RRID:SCR_016649). Details of the RPPA platform as performed by the RPPA Core are described in (Siwak, D.R., et al., Adv Exp Med Biol 2019; 1188:113-47). In brief, proteins were denatured using SDS and 2-mercaptoethanol and diluted lysates were arrayed on nitrocellulose-coated slides (Grace Bio-Labs) by the Quanterix 2470 Arrayer (RRID:SCR_027241). Slides were processed using validated primary antibodies and biotin-conjugated secondary antibodies. The signal was amplified via the GenPoint platform (AgilentMRG: 0680.003419W001

[0310] Technologies) and visualized through DAB colorimetric staining. Following imaging with a TissueScope (Huron Digital Pathology), spot intensities were quantified 1478 using customized Array-Pro Analyzer software (Media Cybernetics). To determine relative protein levels, RPPA SPACE pipeline (Shehwana, H., et al., Bioinformatics 2022;38(22):5131-3) was used. This method fits a logistic model to a single curve generated from all slide samples, treating signal intensity as the response variable and dilution steps as the independent variable. For quality control, the resulting curves plot signal intensity against log2 protein concentration. Final protein concentrations were adjusted for loading using a two-step median- centering process (across both samples and antibodies).

[0311] Cytokine array

[0312] HER2hi and HER21o cells were place in mono or co-culture (1:1) for 8 days. Cells were then counted and replated to culture in Opti-MEM (Gibco) without FBS for 24h. Collected supernatant from 3 biological replicates were pooled and analyzed using Proteome Profiler Human XL Cytokine Array (R&D Systems). Imaging was done using a Chemidoc MP device (BioRad) and quantification was performed using ImageJ (RRID:SCR_003070).

[0313] Statistical Analyses

[0314] All statistical analyses were performed using GRAPHPAD PRISM (RRID:SCR_002798). For single comparisons of normally distributed data, unpaired t-test or Welch’s t-test (when the assumption of equal variance was implausible) were used. For comparisons involving more than two groups, one way or two-way ANOVA was performed with correction for multiple comparisons. A Chi-square test without Yates’ correction was used to assess the association between treatment conditions and ERBB2 amplification status across 675-1,421 cells per resistant cell lines. Furthermore, Pearson’s correlation coefficient (two-tailed) was calculated to evaluate the relationship between drug IC50 values and the proportion of HER210cells. Survival analyses were assessed using the log-rank (Mantel -Cox) test. All statistical analyses were conducted with a 95% confidence interval, and corresponding P-values (P < 0.05 considered significant) were reported for each experiment.

[0315] Data availability

[0316] All raw and processed RNA-seq data were deposited to NCBI GEO database under accession number GSE300628.MRG: 0680.003419W001

[0317] Example 1: Generation and characterization of HER2 heterogeneous breast cancer models

[0318] To assess intratumoral heterogeneity for HER2, flow cytometry for cell surface HER2 in a panel of human HER2+ breast cancer cell lines (Fig. 1A) was performed. The highest fraction of HER210cells in the 21 PT, 21 NT, and HCC1954 cell lines at 29.4%, 10.9% and 14.3%, respectively (Fig. 1A) was observed. 21NT and 21PT are two cell lines with distinct phenotypes and genotypes derived from the same primary tumor of a patient with an aggressive HER2+ breast cancer resistant to chemotherapy (Band, V., etal., Cancer Res 50(22): 7351-7, 1990), while the HCC1954 line was derived from a high-grade HER2+ breast cancer following radiation treatment (Gazdar, A.F., et al., Int J Cancer 78(6):766-74, 1998). To isolate HER21" and HER210subpopulations, the cells were sorted and then regrown, repeating the process a total of five times. The isolated HER2hland HER210populations maintained their HER2 expression levels even after more than 10 passages (Fig. IB and 1C), and their identity was confirmed by short tandem repeat (STR) profiling.

[0319] Next, FISH was performed to assess ERBB2 copy number and it was observed ERBB2 amplification in HER2hlbut not in HER210cells, consistent with the clinical definition of HER2 heterogeneous tumors (Fig. ID). Additional genetic differences between paired HER2hland HER210subsets from the same cell line by whole exome sequencing (WES), which revealed largely overlapping copy number variation (CNV) except for the ERBB2 amplicon (Fig. IE) was investigated. Additionally, analysis of the single nucleotide variants (SNV) revealed both shared and distinct mutations between HER21" and HER210cells in tumor suppressor and oncogenes that are mutated in breast cancer according to the TCGA cohort (Fig. IF). The p.R816H ERBB2 mutation detected in the HER2hi population of the 2 INT and 21 PT cell lines is not known to be oncogenic and it is potentially neutral. Overall, multiple SNVs were shared between the matched populations at comparable allelic frequencies suggesting that HER21" and HER210cell populations within the same tumor have a common clonal origin. However, no mutations were shared between all HER210cells from the different cell lines indicating that HER2 heterogeneous tumors are not driven by recurrent mutations. Conversely, genes specifically mutated in HER210cells included pathways linked with growth factor (e.g. IGF and PDGF) and PI3K signaling that drive cell proliferation, survival, and oncogenic transformation in cancer (Fig. II).MRG: 0680.003419W001

[0320] To characterize differences between HER2hland HER210cells at the single cell level, including signaling pathway activity, cytometry by time of flight (CYTOF) analysis was performed. HER2 was the most consistently differentially expressed protein with high expression in HER2hlcells in both models, while markers of cell proliferation (e.g., CDK1) and cell state (e.g., CD49F, CD24) showed more subtle and less consistent differences ( Fig. 1 J). HER2 interacting proteins (HER3 and EGFR) did not show clear differences between HERZ111and HER210cells while HER210cells had slightly lower levels of downstream components of the HER2 signaling pathway (phospho- AKT and phospho-S6).

[0321] We next utilized RNA-sequencing to further characterize these cells and determined that the main difference between HER2hland HER210cells was the expression of genes located in the ERBB2 amplicon (Fig. IK) was explored. There was a small overlap in the differentially expressed genes between HER2hland HER210cells from different lines, and no clear clustering of the samples based on the differentially expressed genes, suggesting little in common between HER2hland HER210cells from different tumors have little in common (Data not shown). Analysis of signaling pathway activity using Gene Set Enrichment Analysis (GSEA) also did not show consistent pathway differences between HER2hland HER210cells across cell lines (Fig. IL and IM). However, PAM50 subtype (Perou, C.M., etal., Nature, 406(6797): 747-52, 2000) probability predictions consistently showed that HER210cells are less HER2-like and more basal compared to their HER2hlcounterparts (Fig. 1G). This observation is in agreement with data from the NCT02326974 clinical trial demonstrating that treatment-naive HER2 heterogeneous tumors are transcriptionally more basal than HER2 non-heterogeneous tumors (Li, Z., ibid. . To extend these findings, cyclic immunofluorescence (CycIF) data obtained from a subset of treatment-naive HER2+ tumors in the NCT02326974 trial (Janiszewska, M., et al., JCI Insight 2021 ;6(11)) was re-analyzed, focusing on cells with high versus low HER2 expression. Consistent with the PAM50 prediction scores above, HER210cells from those patient tumors had higher levels of basal (CK5 and CK14) and mesenchymal (vimentin), but lower expression of luminal (C 135 K19, CK7 or CK8) markers in the patient tumors (Fig. 1H).Overall, these data demonstrate that the HER2 heterogeneous cell line models accurately reflect the genetic and transcriptomic features of HER2 heterogeneous breast cancer were generated.

[0322] Generation and characterization of HER2 heterogeneous models are shown in Figures 1A-1H. Figure 1 A, Flow cytometry histograms (left) and dot plots (right) showing cellMRG: 0680.003419W001 surface HER2 protein levels across HER2+ breast cancer cell lines. Figure IB, Flow cytometry histograms showing cell surface HER2 protein levels in sorted HER2hland HER210subpopulations within the indicated cell lines. Figure 1C, Immunoblot analysis of HER2 protein levels in HERZ111and HER210cells. Tubulin was used as loading control. Figure ID, FISH analysis of ERBB2 and CEP 17 copy numbers in HER111and HER210cells derived from the indicated cell lines. Scale bar, 25 pm. Figure IE, Copy number variation (CNV) plots of paired HER2hland HER210cells. There were CNV detected in both HER21" and HER210cells, and also CNV unique to HER2hland HER210cells. Figure IF, Plot depicting the presence of point mutations in genes commonly mutated in HER2+ breast cancer in paired HERZ111and HER210cells. Figure 1G, Plots depicting PAM50 subtype probability scores of paired HER2hland HER210cells assessed by RNA-seq. Figure 1H, Graphs showing the fraction of HERZ111and HER210cancer cells positive for the indicated subtype-specific markers in HER2+ breast tumors (n=20 patients) based on cyclic immunofluorescence (Janiszewska M., et al. JCI Insight 2021;6(ll)), t-test.

[0323] Analysis of HERZ111and HER210paired cells differences are shown in Figs. SI A -IM. Figure S1A, Reactome pathways enriched in genes uniquely mutated in HER210cells combining all cell lines. Figure SIB, UMAP visualization of the expression of the indicated markers by CyTOF in HERZ111and HER210cells in the 21PT and HCC1954 cell lines. Figure IK, Volcano plot of differentially expressed genes in paired HER2hland HER210cells. Figure IL, Heatmap of the Normalized Enrichment Score (NES) from Hallmark GSEA comparing HERZ111and HERZ10cells within each cell line, displaying gene sets significantly enriched in at least one comparison and indicating only the significant values (p<0.05). Figure IM, Box plots showing enrichment scores of different ERBB / PI3K pathway signatures in HER2hland HER210cells.

[0324] Example 2: HER210cells are resistant to HER2-targeting ADCs, but respond to HER2 kinase inhibitors

[0325] To facilitate the tracking of HERZ111and HERZ10cells in live cultures, the cells were marked with H2B 142 mcherry (HER2hlcells) and H2B-GFP (HER210cells) using lentiviral transduction. The response of these HERZ111and HERZ10subpopulations as well as parental cell lines to HERZ- targeting agents including ADCs (trastuzumab deruxtecan, T-DXd and trastuzumab emtansine, T DM1) and HERZ tyrosine kinase inhibitors (TKIs) (neratinib and tucatinib) was then assessed. HER210cells showed reduced sensitivity to ADCs but not to HERZMRG: 0680.003419W001 TKIs compared to corresponding HER2hland parental cells (Fig. 2A, Fig. 2H and 21). Tucatinib, a HER2-specific TKI, was slightly less effective than neratinib in 21PT cells, thus, we also tested two additional HER2-specific inhibitors (zongertinib and CP-724714) but did not observe significant differences between HER21" and HER210cells in either model (Fig. 2B). The difference in response between HER2hland HER210cells to T-DXd was less pronounced in the 21NT line but this was not the case in the HCC1954 model. The response of 21NT and 21PT HER210cells to HER2 -targeting ADCs was comparable to that of MDA-MB-468 basal and MDA-MB-231 mesenchymal triple negative breast cancer (TNBC) cell lines lacking HER2, while HCC1954 HER210cells were more sensitive than TNBC cells, consistent with their higher level on HER2 (Fig. 2L and 2M). Sensitivity to chemotherapeutic agents like paclitaxel and 5-fluorouracil (5-FU), and deruxtecan (DXd), the payload of T-DXd, did not show significant differences indicating that the reduced response of HER210cells to T-DXd is not due to the payload or general chemoresistance (Fig. 2N). Furthermore, immunoblot analysis of downstream components of the EGFR / HER2 signaling pathway demonstrated baseline differences between HER2hi and HER21o populations, but both showed a pronounced decrease following neratinib treatment, in line with observations (Fig. 2B).

[0326] Cellular competition experiments by treating 1 : 1 mixture of HER2hland HER210cell cultures with HER2- targeting ADCs or TKIs were also performed. Using these assays it was found that HER210cells dominated the population after 10 passages of ADC but not TKI treatment (Fig. 20). This finding was observed in the HCC1954 and 21NT but not in the 21PT model, due to the higher sensitivity of 21 PT HER2hlcells to TKIs compared to their HER210counterparts (Fig. 21). To validate our results in clinical samples, single-cell RNA-seq (scRNA-seq) was performed on two HER2+ tumors (T537 and T565) and patient-derived organoids (PDOs) were also generated from them. Both the scRNA-seq data on the clinical samples and flow cytometry on the PDOs confirmed the presence of HER2hi and HER21o cancer cell subpopulations (Fig. 2C Fig. 2P). We treated these PDOs with T-DXd or neratinib for 2 weeks and observed an increase in the relative fraction of the HER21o subpopulation, confirming that HER21o cells are less sensitive to HER2-targeting agents (Fig. 2C and Fig. 20). This observation was also validated in a patient-derived xenograft (PDX, Metsl5) cultured as organoids (Fig. 2P).MRG: 0680.003419W001

[0327] To further investigate the dynamics of HER21" and HER210cell populations and mechanisms of resistance to HER2-targeting agents in the breast cancer models, resistant derivatives of each of the three parental cell lines to all four agents by prolonged culture in the presence of the drug were generated. ERBB2 copy numbers by FISH was first assessed, which revealed selection for ERBB2 non-amplified cells in T-DXd (TDR) and T-DM1 (TMR) resistant derivates, whereas neratinib (NR) and tucatinib (TR) resistant cells still showed the same level of ERBB2 amplification as cells passaged in vehicle for the same time (Ctl) (Fig. 2D and Fig. 3G). Accordingly, analysis of RNA-seq data demonstrated that TDR and TMR cells clustered more closely with corresponding HER210cells from the same cell line, based on Euclidean distance in a principal component analysis (PC A), while HER2hlcells showed similar distance from all resistant derivates (Fig. 2E and 2F, Fig. 3H). Apoptosis and TGF- signaling were specifically enriched in TDR and TMR, respectively, compared to HER210cells, suggesting that selection for ERBB2 non-amplified cells is not the only feature of resistance to these ADCs ( Fig. 31).Furthermore, GSEA across all three models showed differences between ADC and TKI resistant derivates in E2F and MYC targets (Fig. 3J). Immunoblot analysis of parental and resistant cells at baseline for HER2 / EGFR signaling pathway components identified a decrease in HER2 itself as the main distinguishing feature of ADC-resistant cells (Fig. S3E). Cross-resistance across the four agents was also investigated and it was found that TDR and TMR cells were resistant to both ADCs but still sensitive to TKIs whereas NR and TR cells were resistant to both TKIs but sensitive to T-DXd and T-DM1 (Fig. 2G, Fig. 3L).

[0328] To investigate potential drivers of resistance in these lines, correlations between the half maximal inhibitory concentration (IC50) values of the four HER2- targeting agents and the proportion of HER210cells across a panel of HER2+ breast cancer cell lines, finding significant positive correlation between HER210cells proportions and the IC50 to T-DXd and T-DM1 ( Fig. 3M) was found. Tucatinib, which exhibits greater specificity for HER2, demonstrated a trend similar to that of the ADCs, but not the pan-HER inhibitor neratinib. Given these observations, whether combining ADCs with neratinib would more effectively inhibit HER2 heterogeneous cultures was next tested. It was found that the combination of neratinib and T-DXd was more effective than either agent alone in all three parental cell lines ( Fig. 20), in line with prior cell culture and clinical data (Freedman, R.A., etal., Ann Oncol 35(ll):993-1002,MRG: 0680.003419W001 2024; Li, B.T., etal., Cancer Discovery 10(5):674-87, 2020), supporting this approach as a potential strategy for HER2 heterogeneous tumors.

[0329] The effect of HER2-targeting agents on HER21" and HER210population is shown in Figures 2A-2F. Figure 2A, Dose response of the indicated compounds tested at various concentrations in parental 21 NT and HER2hland HER210derivates. Comparison to HER2hlcells, p-values for HER21o and parental. Figure 2B, Immunoblot analysis of EGFR, HER2, and downstream signaling pathway components in HER21" and HER210cells treated with vehicle or neratinib. Figure 2C, FISH analysis of ERBB2 and CEP17 copy numbers in control 21NT cells and derivates resistant to T-DXd (TDR), T-DM1 (TMR), neratinib (NR), or tucatinib (TR). Bar plot depicts quantification of FISH signal in 675-1421 cells per condition. Chi square without Yates correction. Scale bar 25 pm. Figure 2C: Flow cytometry analysis of HER21o subpopulation of patient derived organoids from HER2 heterogeneous tumor T537 treated with T1008 DXd or neratinib. Figure 2D: FISH analysis of ERBB2 and CEP17 copy numbers in control 21NT cells and derivates resistant to T-DXd (TDR), T-DM1 (TMR), neratinib (NR), or tucatinib (TR). Bar plot depicts quantification of FISH signal in 675-1421 cells per condition. Chi-square without Yates correction. Scale bar 25 pm. Figure 2E: Principal component analysis plot of RNA-seq profiles of HCC1954 control cells, HER2hland HER210subpopulations, and resistant derivates. Figure 2F, Bar graph showing Euclidean distances between PCI and PC2 of the indicated RNA-seq samples from panel E. Figure 2G, Dose-response of the indicated compounds tested at various concentrations in control 21NT cells and resistant derivates.Comparison to vehicle control cells. Data are presented as mean ±SEM, n = 3, two-way ANOVA (panels A and G).

[0330] HER2 -targeted therapy response in HER2 heterogeneous cell models is shown in Figures. 2H - 2P. Figure 2H, Figure 21, Dose response curves of the indicated compounds tested at various concentrations in HER21", HER210, and corresponding parental HCC1954 (A) and 21PT (B) cell lines. Comparison to HER21" cells, p-values in for HER210and parental. Figure 2J, Figure 2K, dose response curves of the specified compounds tested at increasing concentrations in paired HER2hi and HER21o cells in the indicated cell lines. Figure 2L, Figure 2M, Bar plots depicting relative cell counts normalized to vehicle-treated controls following 6 days of treatment of the indicated cell lines with T-DXd or T-DM1. Figure 2N, Dose response curves of the indicated compounds tested at various concentrations in HER2hland HER210cells fromMRG: 0680.003419W001 HCC1954, 21NT, and 21PT cell lines. Figure 20, Stacked bar plots showing the relative proportions of HER2hland HER210cells after ten passages in co-cultures treated with the indicated agents. HER21" and HER210cells were mixed at 1 : 1 ratio at the start of the experiment. Figure 2P, Analysis of HER21o subpopulation by flow cytometry in patient derived organoids from T565 and Metsl 5 treated with T-DXd or neratinib for two weeks. Data are presented as mean ± SEM, n = 3, one- war ANOVA (panels L, M, O) or two-way ANOVA (panels H, I, N).

[0331] Example 3: HER2hland HER210cells cooperate in co-culture

[0332] HER210cells were a minor subpopulation and slower growing compared to HER2hlcells in each of the three parental lines (Fig. 1 A); thus, their persistence raised the possibility of subclonal cooperation between HER2hland HER210cells. To test this hypothesis, the growth of HER21" and HER210cells when cultured alone (monoculture) and in 1 : 1 co-culture was compared. It was observed that in control conditions, both HER2hland HER210cells grew faster in co-culture compared to monoculture, (Fig. 3A, Fig. 4R and 4S). To further study the interaction between HER2hland HER210cells in vitro, we conducted RNA-seq analysis on sorted cells from monoculture and co-culture. GSEA revealed that genes related to epithelial-to-mesenchymal transition (EMT) were commonly upregulated in the co-cultures across cell lines (Fig. 4T). Additionally, Reverse Phase Protein Array (RPPA) revealed a decrease of E-cadherin in co-culture of 21PT as one of the most differentially present proteins in this condition (Fig. 4U). In HCC1954, co-culture increased the expression of immune checkpoint proteins B7-H3 and PD-L1. A cytokine array also showed an increase of angiogenesis related factors in coculture of HCC1954 (Fig. 4V). Even though the two cell lines show differences the pathways regulated, these data suggest enhanced plasticity and adaptation when HER2hland HER210cells are interacting in co-culture.

[0333] Interestingly, the response of HER2hlor HER210cells to HER2-targeting agents was the same in monoculture and co-culture (Fig. 3A, Fig. 4R and 4S). T-DXd has been proposed to have a bystander effect (Ogitani. Y., etal., Cancer Sci 107(7):1039-46, 2016), however, in the breast cancer models, HER210cells did not show increased sensitivity to T-DXd when co-cultured with HER21" cells (Fig. 3A, Fig. 3G). To further investigate the bystander effect of T-DXd, the same cell lines (MDA-MB-468 basal TNBC and KPL4 HER2+ inflammatory breast cancer cells) used in the original paper describing the T-DXd bystanderMRG: 0680.003419W001 effect, together with the panel of HER210and HER2hlcells, and the MDA-MB-231 mesenchymal TNBC line were tested. The bystander effect was quantified by comparing the response to T-DXd of various cells with low expression of HER2 in monoculture versus co-culture with HER21" cells.

[0334] While the previously reported bystander effect with the MDA-MB-468 TNBC cell line was confirmed, in the other cell line co-cultures minimal bystander activity (Fig. 3B) was observed. Next, it was tested whether the difference between MDA-MB-468 cells and the other HER2 low lines was due to their sensitivity to the DXd payload alone and found that MDA-MB-468 cells were significantly more sensitive to DXd than the other cell lines (Fig. 3C). These results indicate that the T-DXd bystander effect is determined by the inherent sensitivity of the target cells to DXd.

[0335] To investigate the influence of co-culture on the clonal dynamics of HER2hi and HER21o cells, we employed a DNA barcoding strategy and uniquely labelled cells in each population using the ClonMapper library (Gutierrez, C., et al., Nat Cancer 2021;2(7):758-72). We compared clonal selection of HER2hi and HER21o populations in the HCC1954 and 21PT models grown either in monoculture or in a 1:1 co-culture, under treatment with DMSO, neratinib or T-DXd for up to 10 passages. The ratio of HER2hi to HER21o cells and the time per passage were monitored throughout the experiment and we again observed that treatment with T-DXd led to a rapid outgrowth of HER21o cells, but not treatment with neratinib, in HCC1954 cells (Fig. 3D, Fig. 5K). In this model, HER21o completely outcompeted HER2hi cells within just four passages, marking the endpoint of this treatment condition (Fig. 3D). Barcode sequencing revealed a decrease in clonal diversity over time across all groups, as evidenced by a decrease in Shannon entropy and a reduction in the number of most frequent barcodes needed to cover 90% of the final population (Fig. 3E and 3F, Fig. 5L and 5M). Both changes were more pronounced in the presence of HER2-targeting agents compared to DMSO, indicating selection for treatment-resistant subclones. Additionally, for cells treated with T-DXd, we observed significant differences when comparing co-372 culture to monoculture conditions. In HER2hi cells treated with T-DXd, the reduction in clonal diversity was less pronounced in co-culture conditions, indicating reduced selection (Fig. 3E and 3F, Fig. 5L and 5M). In contrast, HER21o cells experienced stronger selection when co-cultured with HER2hi cells, suggesting the presence of a bystander effect.MRG: 0680.003419W001

[0336] Assessing the identity of the top 30 most abundant barcodes revealed that different clones were selected by neratinib and T-DXd treatment indicating different mechanisms of resistance. FISH plots tracking the frequencies of the top 30 barcodes in HCC1954 and 21PT across HER2hi and HER210cells, respectively, in the treatment conditions, were generated (Data not shown). Stacked barplots were generated comparing the relative proportion of barcodes at end point of 21PT HER21" and HER210cells (Data not shown). The high similarity in clonal selection patterns between replicates and across monoculture and co-culture conditions in most conditions suggests that resistance is largely driven by pre-existing cellular traits. One exception to this pattern were HCC1954 HER21" cells under T-DXd treatment which exhibited greater variability in barcode selection across replicates. While shared barcodes were also observed between those replicates, this greater variability suggests that stochastic or acquired mechanisms may also contribute to resistance in this condition. Further, differential selection of barcodes between conditions by testing for differences in the relative frequencies of the top 30 barcodes and correcting for multiple hypothesis testing (Methods; Fig. 4T) was assessed.

[0337] Based on these data, it is concluded that HER2hland HER210cells influence each other’s phenotype in co-culture, thereby impacting therapeutic responses.

[0338] Subcl onal dynamics of HER21" and HER210cells during evolution to resistance in mono- or in co-cultures is shown in Figures 3A-3F. Figure 3A, Plots depicting viable cell numbers of HCC1954 HER2hland HER210cells grown separately in monoculture or together in co-culture in the presence of DMSO, T-DXd, TDM1, neratinib or tucatinib. Figure 3B, Heatmap depicting the relative bystander effect of T-DXd in a panel of HER210or TNBC cells in coculture with different HER21" cells (mean of n = 3). Figure 3C, Dose-response of DXd tested at various concentrations in different HER210and TNBC cells. Data are presented as mean ± SEM, n = 3, two-way ANOVA (Panels A and C). Figure 3D, Bar plot depicting the number of days per passage in the indicated conditions, including the ratio of HER21" and HER210cells in initially 1 : 1 co-culture. Figure 3E, Shannon entropy measuring barcode diversity over different passages of HER21" and HER210cells growing in mono- or co-cultures in the presence of DMSO, neratinib or T-DXd. Figure 3F, Bar plot illustrating the number of barcodes remaining in the top 90% of HER2hland HER210cells after 10 passages (DMSO and neratinib) or 4 passages (T-DXd). Oneway ANOVA comparing all conditions to DMSO neratinib (HER21", p<0.0001, HER210,MRG: 0680.003419W001 p<0.0001) and T-DXd, (HER2hl, p<0.0001, HER210monoculture, p=0.0070, and co-culture pO.OOOl).

[0339] Characterization of HER2-targeting therapy resistant cells is shown in Figs. 3G- 3N. Fig. 3G, Bar plots showing quantification of ERBB2 amplified cells determined by FISH in HCC1954 and 21PT cells resistant to the indicated HER2-targeting agents. FISH signal was assessed in 539-1,312 cells per condition. Chi-square without Yates correction. Fig. 3H, Heatmap showing the Euclidian distances between PCI and PC2 of the indicated RNA-seq samples from in Fig. 2D. Fig. 31, Heatmap of the Normalized Enrichment Score (NES) from Hallmark GSEA comparing HER21o, TDR and TMR to control cells from HCC1954, showing gene sets significantly enriched in at least one condition and indicating only the significant values (p<0.05). Fig. 3 J, NES from Hallmark GSEA comparing resistant and control cells within each cell line, showing gene sets significantly enriched in at least one condition and indicating only the significant values (p<0.05). Fig. S3E, Immunoblot analysis of EGFR, HER2, and downstream signaling pathway components in control and resistant cells. Fig. 3L, Dose response curves of the indicated compounds tested at various concentrations in control HCC1954 and 21PT cells and their resistant derivatives. Comparison of resistant to control cells, p-values are indicated for each comparison. Fig. 3M, Correlation of IC50 of the indicated compounds and the percentage of HER21o cells determined by flow cytometry in Fig. 1 A. Pearson r correlation test (two-tailed). Fig. 3N, Stacked bar blots showing the relative fraction of HER2hi and HER21o cells and the cell count normalized to untreated condition after treatment with neratinib, T-DXd, or their combination for 10 days. Data are presented as mean ± SEM, n = 3, one-way ANOVA (panels G and N) or two-way ANOVA (panel L).

[0340] Example 4: The impact of HER2 heterogeneity on tumorigenesis and treatment outcomes in vivo

[0341] Subsequently, xenograft assays were performed to assess the impact of HER2 heterogeneity on tumor growth and response to HER2-targeted therapies in vivo. We injected HER2hlor HER210cells alone (homogeneous) or as a 1 : 1 mix (heterogeneous) into the mammary fat pad of immunodeficient NSG mice. It was found that HER210cells were unable to form tumors on their own, but both HER2hlhomogeneous and heterogeneous tumors grew well in all three models tested (Fig. 4A and Fig. 7A). The presence of 263 HER210H2B-GFP+ and HER21"MRG: 0680.003419W001 H2B264 mCherry+ cells within tumors were verified by fluorescence imaging and confirmed by immunofluorescence for HER2 that they retained their low and high levels of HER2 expression, respectively (Fig. 4B). Intriguingly, HER210cells were maintained in heterogeneous tumors despite their lack of tumor-initiating capacity, again implying subclonal interactions between the two cell populations. By investigating the spatial distribution of the cells within tumors, it was observed that HER2hland HER210cells did not intermingle extensively but rather grew in clusters, with HER2hlcells more frequently neighboring other HER21" cells, and similarly, HER210cells being more commonly surrounded by HER210cells (Fig. 4B and 4C). The same pattern was detected in human HER2- positive tumors based on spatial analysis using CycIF in the NCT02326974 clinical trial cohort (Janiszewska, M., etal., JCI Insight 6(11), 2021) (Fig. 4D).

[0342] To better understand the cell population dynamics during tumor growth, we performed a time course experiment, monitoring the ratios of HER2hi and HER21o cells in the tumors by flow cytometry ( Fig. 7B). HER21o cells initially make up a substantial proportion of the tumors, but their ratio decreases during the exponential growth phase as the HER2hi population increases. This observation suggests that HER21o cells can persist after injection but are less proliferative than HER2hi cells. Quantitative assessment of cellular proliferation by immunofluorescence for the S-phase marker PCNA demonstrated that PCNA+ cells were significantly more commonly HER2hi than HER21o, which was also observed in the CycIF data of the human cohort (Janiszewska, Ibid. (Fig. 4E and 4F).

[0343] To study the interaction of the HER2hi and HER21o cells with their environment, we focused on human tumors. We performed scRNA-seq of our two HER2 heterogeneous tumors (T537, T565) and analyzed these data with CellChat, which revealed that tumor cells with low levels of HER2 had greater amount of interaction with stromal cells than HER2hi tumor cells (Fig. 4G). Dominating the signal affecting HER21o cells were extracellular matrix (ECM) 456 proteins like collagen and laminin (Fig. 4H). This observation was also consistent with analyses of a publicly available data set of six HER2+ breast tumors (Pal, Ibid. , which also showed collagen as the most enriched signal in HER21o cells (Supplemental 7C). Furthermore, we investigated whether HER21o cells are in close proximity to different stromal cells with the CycIF data set (Janiszewska, Ibid}, and we observed that their neighborhood are enriched in Podoplanin+ stromal cells (Fig. 4Land Supplemental 7H).MRG: 0680.003419W001

[0344] Next, the response of HER2 homogeneous and heterogeneous tumors to HER2-targeting therapies was evaluated. To this end, mice with palpable tumors were randomly assigned to vehicle, neratinib (40 mg / kg), or T-DXd (10 mg / kg) treatment groups and treated for two weeks (Fig. 4J). Both neratinib and T-DXd treatment reduced the size of both HER2 homogeneous and heterogeneous tumors (Fig. 4K and 4L, Fig. 7E). However, T-DXd was significantly more effective in HERZ111homogeneous compared to heterogeneous tumors, leading to complete regression in all cases, while the efficacy of neratinib was not affected by HER2 heterogeneity. Residual HER2 heterogeneous tumors following T-DXd treatment were predominantly composed of H2B-GFP+ HER210cells based on green fluorescence of the tumors, which was confirmed by immunofluorescence and flow cytometry (Fig. 4L and 4M). In contrast, this shift in HER2hland HER210cell populations was not observed in tumors from neratinib-treated mice. To better reflect the clinical setting, 8 continuous rounds of T-DXd (1 mg / kg every week) were administered. This regimen also led to a greater response in homogeneous tumors than heterogeneous tumors (Fig. 7F and 7G).

[0345] To assess the impact of these treatments on recurrence and long-term survival, the experiments and followed mice bearing HER2 homogeneous or heterogeneous tumors after stopping T-DXd and neratinib treatment (Fig. 4N) were repeated. It was observed that neratinib treated tumors recurred quickly after stopping treatment and that there was no difference in time to recurrence between HER2 homogeneous and heterogeneous tumors (Fig. 4L and 4M). In contrast, tumors in T-DXd treated mice took more time to regrow with HER2 heterogeneous tumors recurring earlier than homogeneous ones (Fig. 40 and 4P). A subset of mice (2 out of 5) with HER2 homogeneous tumors did not have recurrence more than 140 days after stopping T-DXd treatment, whereas all heterogeneous tumors recurred within 92 days (Fig. 40 and 4P). Moreover, recurrent HER2 heterogeneous tumors were predominantly composed of HERZ111cells in contrast to the tumors collected at the end of short-term T-DXd treatment, (Fig. 4Q). These HERZhi cells regained HERZ expression and were still amplified at the ERBB2 locus ( Fig. 7H). This observation recapitulates the experience in the NCT02326974 clinical trial, where non-heterogeneous residual tumors after T-DM1 and pertuzumab treatment had a decreased HER2 immunohistochemistry score, but did not have changes in their ERBB2 FISH status (Li, Z., et al., J Clin Invest 2024; 134(7)). Therefore, when we culture the cells from the recurrent tumors, we observed that they are still sensitive to T-DXd ( Fig. 71). Thus, while rare HERZhi cells mayMRG: 0680.003419W001 resist T-DXd and recover after treatment, these findings suggest that in heterogeneous tumors, HER210cells resistant to T-DXd may help sustain the survival of some HERZ111cells, as they might then have a greater capacity to recover after discontinuation of treatment.

[0346] Overall, xenograft experiments confirm that the presence of HER210cells drives therapeutic resistance to T-DXd in HER2 heterogeneous tumors.

[0347] The impact of HER2 heterogeneity on tumorigenesis and treatment responses in vivo are shown in Figures 4A-4Q. Figure 4A, Tumor growth of mammary fat pad injection of 21NT homogeneous HER2hi or HER21o, and heterogeneous 1:1 mix (n=5). Figure 4B, Immunohistochemistry for mCherry (HER2hi), GFP (HER21o) and HER2 reveals that HER2hi, HER21o retain their different HER2 levels in heterogeneous tumors in vivo. Scale bar, 50 pm. Figures 4C, 4D, Stacked barplot representing the neighborhood composition of HER2hi and HER21o cells in 21 NT HER2 heterogeneous tumors (C, n=5 tumors) or human tumors from the CycIF of (Janiszewska, Ibid) (D, n=20 patients), t-test (p-value for HER2hi and for HER21o). Figures 4F, 4G, Quantification of staining for PCNA staining proliferation marker in HER2hi and HER21o cells in HER2 heterogenous 21 NT (F, n=5) or human tumors (20) (G, n=20 patients), t-test. Figure 4G, CellChat analysis of scRNA-seq data depicting the amount of incoming and outgoing interactions of every cell type. Figure 4H, Plot presenting the pathways of the overall signals comparing HER21o and HER2hi cell populations. Figure 41, Presence of PDPN+ stromal cells in the neighborhood of HER2hi and HER21o cells (n=20 patients, t-test). Figure 4J, Schematic depicting the in vivo experiment set up where 21NT homogeneous and heterogeneous tumors are treated with neratinib or T-DXd. Figure 4K, Curve of changes in tumor size under treatment with neratinib or T-DXd. n = 9-10, two-way ANOVA. Figure 4L, Fluorescence pictures of tumors after 2 weeks treatment showing HER2hi cells and HER21o cells. Scale bar, 1 cm. Figure 4M, Quantification by flow cytometry of the composition of heterogeneous tumors in HER2hi and HER21o cells after 2 weeks of treatment. n=5, One-way ANOVA. Figure 4N, Schematic depicting the survival assay where tumors are treated for 2 weeks with T-DXd or neratinib and mice are followed until tumor recurrence. Figure 40, Tumor growth and recurrence of homogeneous and heterogeneous tumors upon T-DXd or neratinib treatment. n=10-12 tumors, two-way ANOVA. Figure 4P, Survival curves showing time to tumor volume endpoint in mice treated with indicated treatments comparing homogeneous and heterogeneous tumor-bearing mice. n=5-6 mice, Long-rank (Mantel-Cox) test. Figure 4Q, Quantification by flow cytometry ofMRG: 0680.003419W001 the composition of heterogeneous tumors in HER2hi and HER21o cells after recurrence. n=5, One-way ANOVA. Data are presented as mean ± SEM (panels C-F, I, K, M, and Q).

[0348] The effect of co-culture on HER2hi and HER21o cells is illustrated in Figures 4R-4V. Figure 4R, Graphs depicting the relative growth of HER2hi and HER21o cells in monoculture or in co-culture during treatment with the indicated compounds. Data are presented as mean ± SEM, n = 3, two-way ANOVA. Figure 4S, Relative growth of HCC1954 HER2hi and HER21o cells mixed at different ratios after nine days with the indicated treatment. Data are presented as mean ± SEM, n = 3, one-way ANOVA. Figure 4T, NES from Hallmark GSEA comparing sorted cells from monoculture and co-culture, showing gene sets significantly enriched in the RNA-seq data in at least one condition and indicating only the significant values (p<0.05). Figure 4U, Quantification of Reverse Phase Protein Array (RPPA) comparing monoculture and co-culture showing only proteins with at least 1.5-fold difference between conditions. Figure 4V, Heatmap depicting the quantification of secreted proteins in the media of cells growing as monoculture or co-culture measured by a cytokine array, showing only the most expressed cytokines.

[0349] Example 5: Identification of sensitizers to T-DXd

[0350] Based on this data, increasing the efficacy of T-DXd in HER2 heterogeneous tumors would require the targeting of HER210cells either by drugs that specifically inhibit them or by enhancing their sensitivity T-DXd via synthetic lethal interactions. To identify such targets, a whole genome CRISPR-Cas9 knock out (KO) viability screen in HER210cells in co-culture with HERZ111cells treated with vehicle control or T-DXd for one month (Fig. 5 A) was performed. sgRNAs preferentially depleted (synthetic lethal hits) or enriched (resistance hits) in the presence of T-DXd but not in vehicle control using the maximum-likelihood analysis of gene essentialities (mle) from the MAGeCK pipeline (Li W, etal., Genome Biology 15(12):554, 2014) were identified. GSEA analysis of the hits revealed resistance associated with mTOR signaling, basal transcription factors, and cellular senescence, while synthetic lethality involved aminoacyl-tRNA biosynthesis, nucleotide excision repair, and ubiquitin-mediated proteolysis (Fig. 5B).

[0351] Because a goal was to improve sensitivity to T-DXd, the synthetic lethal hits and genes with pharmacological inhibitors available were prioritized. Thus, two of the top hits with preferential drop out in the T-DXd treated group, ABCC1 (ATP binding cassette subfamily C member 1) transporter and USP9X (Ubiquitin Specific Peptidase 9 X-linked), which haveMRG: 0680.003419W001 available pharmacological inhibitors (Fig. 5C). Furthermore, USP9X was selected because it offered the potential for a novel vulnerability, as it was not directly linked to established DXd efficacy pathways like other hits including PRKDC (DNA repair) and BCL2L1 (apoptosis). First, derivates of HER210cells in the 21NT and HCC1954 models with two individual sgRNAs for both genes to validate the CRISPR screen results ( Fig. 8A) were generated.

[0352] Next, the sensitivity of these KO cells to T-DXd together with pharmacologic inhibitors of ABCC1 (reversan and biricodar) and USP9X (G9 and FT709) were tested and it was confirmed that the inhibition of ABCC1 or USP9X sensitizes cells to T-DXd (Fig. 5D and Fig. 8B). The expression of ABCC1 was increased in HCC1954 TDR cells ( Fig. 8C) and pharmacologic inhibition of both ABCC1 and USP9X enhanced the growth inhibitory effects of T-DXd in TDR resistant lines (Fig. 5E and Fig. 8D). ABCC1 (also known as Multidrug resistance associated protein 1, MRP1), promotes therapeutic resistance by transporting multiple compounds including chemotherapeutic agents out of the cells (Borst, P., etal., JNCI: Journal of the National Cancer Institute 92(16): 1295-302, 2000; Loganzo, F., etal., Mol Cancer Ther 14(4):952-63, 2015). Hence, it was tested whether its inhibition sensitizes the cells to T-DXd by increasing the efficacy of its payload, DXd. It was found that both genetic and pharmacologic inhibition of ABCC1 potentiated the growth suppressing effect of DXd while loss of USP9X activity had no effect (Fig. 5F and Fig. 8E). Because the bystander effect of T-DXd is defined by the sensitivity of the neighboring cells to DXd, it was observed that ABCC1 inhibition increased the bystander effect of T-340 DXd in HER21" and HER210cell co-cultures (Fig. 5G and Fig. 8F). Notably, ABCC1 expression was the lowest in the cell line most sensitive to DXd, MDA-MB-468, compared to the others (Fig. 3C and Fig. 8G).

[0353] The effect of combining the ABCC1 inhibitor reversan with T-DXd on tumor growth and recurrence-free survival of mice with HER2 heterogeneous tumors (Fig. 5H) was then tested. Reversan alone had no impact on tumor growth and the recurrence rate after discontinuing treatment (Fig. 51 and 5J). However, its combination with T-DXd led to more pronounced suppression of tumor growth and prolonged recurrence-free survival (Fig. 51 and 5J). Flow cytometry analysis of the cellular composition of recurrent tumors showed no significant changes in the relative fraction of HER2hland HER210cells, although there were few HER210cells, thus, the potentiation of the bystander effects of T-DXd cannot be excluded. Therefore, inhibiting ABCC1 presents a promising therapeutic strategy to enhance the efficacy of T-DXd,MRG: 0680.003419W001 which is interesting in a HER2 heterogeneous context that can benefit from an increased bystander effect.

[0354] To gain mechanistic insight into the mechanism by which USP9X sensitized HER210cells T353 DXd, it was first evaluated if inhibition of USP9X also increases the efficacy of other HER2-targeting ADCs like T-DM1. It was found that both genetic deletion and pharmacologic inhibition of USP9X sensitized both HER210cells and TMR cells to T-DM1 (Fig.9 A), suggesting that its mechanism of action is not specific to T-DXd. Thus, how USP9X inhibition affected EGFR / HER2 signaling by immunoblot analysis for components of the signaling pathway we explored, but no significant effects were observed (Fig. 9B). USP9X inhibition also did not change the cell surface levels of HER2 at baseline or after T-DXd treatment as assessed by flow cytometry in HER210cells (Fig. 9C and 9D). However, a pronounced increase of ubiquitinated HER2 in the presence of T-DXd upon USP9X inhibition was detected (Fig. 6A and Fig. 9E). To test if USP9X directly interacts with HER2, HER2 or USP9X were immunoprecipitated from both HERZ111and HER210cells and performed immunoblot for both proteins. HER2 was detected in USP9X immunoprecipitates, and conversely, USP9X was present in HER2 immunoprecipitates (Fig. 6B and Fig. 9F). Proximity ligation assay (PLA) experiments were also performed that confirmed an interaction between HER2 and USP9X 366 in both HER111and HER210cells (Fig. 6C and Fig. 9G) also showing that the interaction was further increased following T-DXd treatment. Subsequently, it was investigated if USP9X deletion or inhibition leads to enhanced degradation of HER2 in the presence of T-DXd and found that the reduction of HER2 protein levels following 48h of treatment with T-DXd was potentiated by USP9X inhibition in both HER210and HERZ111cells (Fig. 6D and Fig. 9H). To assess HERZ protein stability in these treatment conditions, we performed cycloheximide pulse chase experiments. These analyses revealed a decrease in HERZ stability upon T-DXd treatment combined with USP9X inhibition (Fig. 6E and 6F).

[0355] Since HERZ is internalized and transported to the lysosomes after T-DXd treatment, it was next assessed if USP9X inhibition increases the fraction of HERZ in the lysosomes by performing PLA experiments between HERZ and LAMPE A significant increase of HERZ in the lysosomes after T-DXd treatment was detected which was further potentiated with USP9X inhibition (Fig. 6G, Fig. 91). This finding was also validated by immunoblot for HERZ using lysosome-enriched cell fraction of HER210cells (Fig. 6H and Fig. 9J).MRG: 0680.003419W001

[0356] Therefore, HER2 is more efficiently targeted to lysosomes upon T-DXd treatment when USP9X is inhibited, which is shown to lead to an increase of the ADC payload release and cell death (Khongorzul, P., et al. , Molecular Cancer Research 18(1):3-19, 2020).

[0357] To validate the clinical relevance of these findings, the expression of USP9X in RNA-seq data comparing HER2+ tumors before and after T-DM1 + pertuzumab neoadjuvant treatment from the NCT02326974 clinical trial was investigated (Filho, O.M., etal., Cancer Discov ll(10):2474-87, 2021; Li Z, etal., J Clin Invest 134(7), 2024). The expression of USP9X was found to be higher in post- compared to pre-treatment tumors in patients who did not achieve a pCR, in both non-heterogeneous (no HET) or heterogeneous (HET) tumors (Fig. 61), confirming that increased expression of USP9X is likely a clinically relevant mechanism of resistance to HER2-targeting ADCs.

[0358] Lastly, the therapeutic potential of combining T-DXd and USP9X inhibitors for the treatment HERZ heterogeneous tumors (Fig. 6 J) was tested. USP9X inhibition alone had no effect on tumor growth and recurrence-free survival (Fig. 6K and 6L). However, its combination with T-DXd led to more pronounced tumor suppression and it significantly increased recurrence-free survival with a few mice remaining tumor-free in the time frame of the experiment (Fig. 6K and 6L). USP9X inhibition did not affect the relative fraction of HER2hland HER210cells in residual tumors, although again there were very few HER210cells detected, thus, differences in the effect of USP9X inhibition on cells with different HER2 levels cannot be excluded ( Fig. 9K).

[0359] Overall, these data show that targeting USP9X or ABCC1 can offer an effective therapeutic strategy to improve the efficacy of T-DXd in HERZ heterogeneous tumors.

[0360] CRISPRKO cellular viability screen with T-DXd in co-cultures is shown in Figures 5A-5J. 733 Figure 5A, Schematic showing 21NT HER210cells infected with Cas9 and whole genome CRISPR KO library and put in co-culture with HER2hlcells in the presence or not of T-DXd for 1 month. Figure 5B, Bubble plot illustrating the enrichment in KEGG pathways of the CRISPR hits linked to resistance and synthetic lethality. Figure 5C, Scatter plot from mle analysis depicting hits from the CRISPR screen only changed in the T-DXd condition linked to resistance or synthetic lethality. Figure 5D, Histogram showing cell count normalized to the untreated control with or without T-DXd for 10 days upon genetic (sgABCCl, sgUSP9X) or pharmacologic (reversan or biricodar, G9 or FT709) inhibition of ABCC1 or USP9X in 21 NTMRG: 0680.003419W001 HER210cells. Figure 5E, Treatment with G9, FT709, reversan or biricodar increased T-DXd effect in 21NT TDR after 10 days. Figure 5F, Bar graph showing cell counts normalized to the untreated control with or without DXd upon genetic (sgA8CC7) or pharmacologic (reversan or biricodar) inhibition of ABCC1 in 21NT HER210cells. Figure 5G, ABCC1 KO in 21NT HER210cells increases the bystander effect of T-DXd in co-culture with HCC1954 HER21" cells. Figure 5H, Schematic outline of the survival assay experiment. Figure 51, Graph illustrating the volume of individual tumors during the indicated treatment, n = 8-10 tumors, two-way ANOVA. Figure 5 J, Kaplan-Meier plot showing time to tumor volume endpoint in mice with the indicated treatments (n=4-5 mice, Long-rank (Mantel-Cox) test). Data are presented as mean ± SEM, n = 3, One-way ANOVA (D-G).

[0361] Characterization of HER2 homogeneous and heterogeneous tumors is shown in Figures 5K- 5M. Figure 5K, Plot depicting the growth of xenografts derived from 21 NT HER2hlor HER210cells injected individually (homogenous) or as a 1 : 1 mix (heterogeneous) (n=5).HER2hland HER210cells are marked with mCherry and GFP fluorescent markers, respectively. Figure 5L, Graph illustrating the weight of homogeneous and heterogeneous tumors at end point following two weeks treatment of mice with vehicle, neratinib or T-DXd. P-values are calculated by t-test. Figure 5M, Waterfall plots showing change in tumor volume of HER2 homogeneous and heterogeneous tumors following treatment with neratinib or T-DXd (n=9- 10 tumors).

[0362] The effect of USP9X inhibition on lysosomal targeting of HER2 and response to T752 DXd is shown in Figures 6A-6L. Figure 6 A, Immunoblot analysis of HER2 immunoprecipitation (IP) probed for ubiquitin (Ub) shows an increase of HER2 ubiquitination upon USP9X inhibition in the presence of T-DXd in 21 NT HER210cells. Figure 6B, Coimmunoprecipitation experiments show an interaction between USP9X and HER2. Figure 6C, Proximity Ligation Assay (PLA) between HER2 and USP9X in 21NT HER210cells treated with vehicle or T-DXd. Scale: 20 pm. Quantification of PLA intensity per cell. 15 cells from each replicate. Figure 6D, Immunoblot analysis of HER2 levels of 21NT HER210cells after treatment with T-DXd upon USP9X inhibition. Figure 6E and Figure 6F, Cycloheximide chase for HER2 in 21NT HER210cells upon T-DXd treatment with USP9X inhibition, including the quantification of HER2 levels. n=3, 759 Two-way ANOVA. Figure 6G, Quantification of PLA intensity per cell. 15 cells from each replicate. Figure 6H, Immunoblot of total cell lysate (TCL) and lysosome enriched fraction of 21 NT HER210cells treated as in panel G. LAMP1 was used asMRG: 0680.003419W001 a lysosome marker, CTCF as nuclear marker and TUFM as a mitochondria marker, confirming lysosome enrichment. Figure 61, USP9X mRNA expression in human tumors before and after T765 DMl / pertuzumab neoadjuvant treatment from NCT02326974 clinical trial (Filho, ibid., ' Li, Z., ibid. ), separated by pCR or no pCR, and No HET or HET. Figure 6 J, Schematic depicting the survival experiment where 21 NT HERZ heterogeneous tumors are treated with one dose of T-DXd, G9 or both. Figure 6K, Tumor growth of individual tumors following indicated treatment, n = 8-10 tumors, two-way ANOVA. Figure 6L, Survival curves showing time to tumor volume endpoint in mice treated with indicated treatments. n=4-5 mice, Long-rank (Mantel-Cox) test.

[0363] Fig. 6M illustrates clonal evolution of HER2hi and HER21o cells upon treatment with neratinib or T-DXd. The frequencies of the top 30 barcodes in HCC1954 and 21 PT across HER2hi and HER21o cells, respectively, in various treatment conditions was obtained from culturing studies (data not shown). Fig. 6M is a heatmap of p- values from tests for differential barcode selection among the top 30 barcodes across conditions.

[0364] Characterization of HER2 homogeneous and heterogeneous tumors is shown in Figures 7A-7I. Figure 7A, Plot depicting the growth of xenografts derived from 21NT HER2hi or HER21o cells injected individually (homogenous) or as a 1:1 mix (heterogeneous) (n=5). HER2hi and HER21o cells are marked with mCherry and GFP fluorescent markers, respectively. Figure 7B, Graph depicting a time course experiment showing the ratio of HER2hi and HER21o cells quantified by flow cytometry along with the tumor growth over time. Figure 7C, Plot showing pathways of the overall signals from HER21o and HER2hi cells defined by Cellchat analysis of scRNA-seq data. Figure 7D, Presence of indicated stromal cells in the neighborhood of HER2hi and HER21o cells (n=20 patients, t-test). Figure 7E, Waterfall plots showing change in tumor volume of HER2 homogeneous and heterogeneous tumors following 2- weeks treatment with neratinib or T-DXd from Fig. 4K (n=9-10 tumors). Figure 7F, Growth curve of tumor size under treatment with T-DXd. n = 8, two-way ANOVA. Figure 7G, Waterfall plot showing change in tumor volume of HER2 homogeneous and heterogeneous tumors following 8 rounds of T-DXd (n=8 tumors). Figure 7H, Quantification of ERSBJ-amplified and non-amplified cells by FISH on cell cultures derived from vehicle or T-DXd heterogeneous recurrent tumors from Fig. 40. Figure 71, Response to increasing doses of T-DXd of cell cultures derived from vehicle or T-DXd-treated heterogeneous recurrent tumors.MRG: 0680.003419W001

[0365] The effect of ABCC1 inhibition on response to T-DXd efficacy is shown in Figures 8A-8G. Figure 8A, Immunoblot analysis of ABCC1 or USP9X protein levels using lysates from control (sgCtrl) and ABCC1 KO (sgABCCl) or USP9X KO (sgUSP9X) 21NT and HCC1954 HER210cells. Calnexin or tubulin was used as loading control. Figure 8B, Bar graphs showing cell counts normalized to the untreated control with or without T-DXd upon genetic (sgABCCl, sgUSP9X) or pharmacologic (reversan or biricodar, G9 or FT709) inhibition of ABCC1 or USP9X in 21NT and HCC1954 HER2hiand HER210cells. Figure 8C, Immunoblot analysis showing the expression of ABCC1 in 21NT and HCC1954 resistant derivates. Calnexin was used as loading control. Figure 8D, Bar graph depicting relative T-DXd-treated to untreated cell counts of control and ABCC1 or USP9X inhibitor-treated HCC1954 TDR cells. Figure 8E, Bar graph showing cell counts normalized to the untreated control with or without DXd upon genetic (sgABCCl, sgUSP9X) or pharmacologic (reversan or biricodar, G9 or FT709) inhibition of ABCC1 or USP9X in 21NT and HCC1954 HER210cells. Figure 8F, ABCC1 KO in HCC1954 HER210cells increases the bystander effect of TDXd in co-culture with HCC1954 HER2hlcells. Figure 8G, ABCC1 mRNA expression in the indicated cell lines, data from the Depmap database. Figure 8H, Bar plot depicting the relative fraction of mCherry (HER2hl) and GFP (HER210) cells in recurrent tumors assessed by flow cytometry. Data are presented as mean SEM, n = 5. Data are presented as mean ± SEM, n = 3, one-way ANOVA (panels B-F).

[0366] The effect of USP9X on HER2 protein levels and response to T-DXd response is illustrated in Figures 9A-9K. Figure 9A, Bar graph depicting relative T-DM1 -treated to untreated cell counts of control (sgCtrl) or USP9X genetic (sgUSP9X) or pharmacologic (G9 and FT709) inhibited 21NT HER210or 21NT and HCC1954 TMR cells. Figure 9B, Immunoblot analysis of EGFR / HER2 signaling pathway components upon USP9X inhibition with or without T-DXd treatment. Figure 9C, Representative histogram of flow cytometry for cell surface HER2 upon USP9X genetic or pharmacologic inhibition with or without T-DXd. Figure 9D, Bar graph depicting cell surface HER2 protein levels normalized to the untreated control in 21 NT and HCC1943 HER210cells (n=3). Figure 9E, Immunoblot analysis of HER2 immunoprecipitants (IP) for ubiquitin (Ub) in control and upon USP9X inhibition in the presence of T-DXd in HCC1954 HER210cells. Figure 9F, Immunoblot analysis of USP9X and HER2 in control IgG, USP9X or HER2 immunoprecipitants from 21NT HER2hland HCC1954 HER210cells. Figure 9G, Bar plot illustrating quantification of PLA intensity per cell. 15 cells from each replicate.MRG: 0680.003419W001 Figure 9H, Immunoblot for HER2 using lysates from 21NT HER2hland HCC1954 HER210cells after T-DXd treatment and upon USP9X inhibition. Figure 91, Bar graph depicting quantification of PLA intensity per cell. 15 cells from each replicate. Figure 9 J, Immunoblot analysis of total cell lysate (TCL) and lysosome enriched fraction of HCC1954 HER210cells treated as in with T-DXd and FT709. LAMP1, CTCF, and TUFM were used as lysosomal, nuclear, and mitochondrial marker, respectively. Tubulin was used as loading control. Figure 9K, Bar plot showing the relative fraction of mCherry (HER2hl) and GFP (HER210) cells in the indicated recurrent tumors quantified by flow cytometry. Data are presented as mean ± SEM, n = 5). Data are presented as mean ± SEM, n = 3-4, one-way ANOVA (panels A, G and I).

Claims

MRG: 0680.003419W001 What is claimed is:

1. A method of treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti-HER2 therapy, the method comprisingadministering to the subject an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, andadministering the HER2 binding agent or HER2 inhibitor to the subject.

2. The method of claim 1 wherein the target protein ATP binding cassette subfamily C member 1 (ABCC1), or the target gene is ABCC1.

3. The method of claim 1 wherein the target protein is encoded by a deubiquitinase that is ubiquitin specific peptidase 9 X-linked or ubiquitin carboxyl-terminal hydrolase 17-like protein 19, or the target gene is USP9X or USP17L19.

4. The method of claim 1 wherein the target protein is a nucleic acid-associated protein that is DNA-dependent protein kinase catalytic subunit, exosome RNA helicase MTR4, zinc finger protein 717, non-structural maintenance of chromosomes element 3 homolog, and isoleucine- -tRNA Ligase, cytoplasmic, or the target gene is PRKDC, SKIV2L2, ZNF717, NDNL2, or IARS.

5. The method of claim 1 wherein the target protein is Bcl-2-like protein 1, or the target gene is BCL2LP6. The method of claim 1 wherein the target protein is cytohesin-interacting protein, or the target gene is CYTIP.

7. The method of claim 1 wherein the target protein is vesicle transport protein GOT1 A, or the target gene is GOLT 1 A.MRG: 0680.003419W0018. The method of claim 1 wherein the target protein is moesin-ezrin-radixin like tumor suppressor, or the target gene is NF2.

9. The method of claim 1 wherein the target protein is guanine nucleotide exchange protein SMCR8, or the target gene is SMCR8.

10. The method of claim 1 wherein the target protein is ERBB receptor feedback inhibitor 1 , or the target gene is ERRFH.

11. The method of claim 1 wherein the target protein is a tyrosine- protein phosphatase that is tyrosine-protein phosphatase non-receptor type 12 or tyrosine-protein phosphatase nonreceptor type 11 , or the target gene is PTPN12 or PTPN1P12. The method of claim 1 wherein the target protein is a glycosyltransferase that is dolichyl- diphosphooligo-saccharide-protein glycosyltransferase subunit 2 or dolichyl-phosphate beta- glucosyltransferase, or the target gene is RPN2 or ALG5.

13. The method of any one of the previous claims, wherein the tumor that is HER2-low, HER2 heterogeneous, or wherein the tumor is resistant to anti-HER2 therapy, is either hormone receptor positive (HR+) or hormone receptor negative (HR).

14. The method of any one of the previous claims wherein the subject having the tumor that is HER2 heterogeneous or HER2-low was diagnosed for HER2 expression using fluorescence in situ hybridization (FISH) and / or immunohistochemistry (IHC).

15. The method of any one of claims 1-12, wherein the tumor that is resistant to anti-HER2 therapy is resistant to anti-HER2 antibody therapy or small molecule inhibitor of HER2 therapy.MRG: 0680.003419W001 16. The method of claim 15 where the tumor has resistance to trastuzumab, trastuzumab- deruxtecan (T-DXd), trastuzumab emtansine (T-DM1), disitamab, pertuzumab, DX-CH09, lapatinib, neratinib, tucatinib, afatinib (BIBW2992), CUDC-101, or canertinib therapy.

17. The method of claim 15 or 16, wherein the tumor has developed resistance to the anti-HER2 therapy or small molecule inhibitor of HER2 therapy which was used in combination with a chemotherapeutic agent.

18. The method of any of the previous claims wherein the subject that comprises the tumor has cancer that is breast cancer, gastric cancer, lung cancer, salivary cancer, vaginal cancer, bladder cancer, endometrial cancer, cervical cancer, or colorectal cancer.

19. The method of claim 18 wherein the cancer is breast cancer.

20. The method of claim 19, wherein the subject that has a tumor characterized as HER2-low has triple-negative breast cancer (TNBC).

21. The method of any one of the previous claims wherein the inhibitor of a target protein (a) is not a polypeptide, (b) is a small molecule, or both (a) and (b).

22. The method of claim 21 wherein the small molecule has a molecular weight in the range of 100 g / mol to 1200 g / mol, in the range of 100 g / mol to 1000 g / mol, or in the range of 100 g / mol to 800 g / mol.

23. The method of any one of claims 2 or 13-22, wherein the inhibitor of ABCC1 is biricodar, MK571, probenecide, or reversan.

24. The method of any one of claims 3 or 13-22, wherein the inhibitor of USP9X is WP1130, G9, FT709, or EOAI3402143.MRG: 0680.003419W001 25. The method of any one of claims 4 or 13-22, wherein the inhibitor of PRKDC is AZD7648, CC-115, M3814, KU-0060648, NU7026, NU7441 (KU-57788), PI-103, PIK-75 HC1, PP121, SF2523, SU11752, or VX-984.

26. The method of any one of claims 4 or 13-22, wherein the inhibitor of isoleucine— tRN A Ligase is reveromycin.

27. The method of any one of claims 5 or 13-22, wherein the inhibitor Bcl-2-like protein 1 is A- 1155463, A-1331852, ABT263, orABT199.

28. The method of claim 21 or 22 wherein the small molecule induces degradation of the target protein.

29. The method of claim 28 wherein the small molecule that induces degradation of the target protein is a molecular glue degrader or a proteolysis targeting chimera (PROTAC).

30. The method of any one of claims 1-20, wherein the inhibitor of the target gene is a nucleic acid molecule that hybridizes to and at least reduces the expression of a target gene or an RNA transcript of the target gene.

31. The method of claim 30 wherein the nucleic acid molecule is an antisense oligonucleotide (ASO) or a small interfering RNA (siRNA).

32. The method of any of the previous claims wherein the HER2 binding agent is an anti-HER2 antibody or an antigen-binding fragment thereof.

33. The method of claim 32 wherein the anti-HER2 antibody is trastuzumab, disitamab, pertuzumab, or DX-CHO9, or an antigen binding fragment thereof.

34. The method of claim 32 or 33 wherein the anti-HER2 antibody or fragment thereof is linked to a drug.MRG: 0680.003419W00135. The method of claim 34 wherein the drug is a cytotoxic agent.

36. The method of claim 35 wherein the drug is DM1 (maytansine), DXd (topoisomerase I inhibitor), MMAE (auristatin), DUBA (duocarmycin), TLR7 / 8 agonist, DP104n, tubulysin, or SLTA.

37. The method of any one of claims 34-36 wherein the anti-HER2 antibody or fragment thereof that is linked to a drug is selected from trastuzumab emtansine (T-DM1), trastuzumab deruxtecan (T-DXd), disitamab vedotin (RC48), trastuzumab duocarmazine, TAA013, ZRC- 3256, MRG002, ARX788, BDC-1001, A166, FS-1502, SHR-A1201, DP303c, BI-CON-02, ALT-P7, DX126-262, ZW49, HS630, B003, SBT6050, SHR-A1811, MT-5111, and GQ1001.

38. The method of any one of claims 1-31, wherein the HER2 inhibitor is a tyrosine kinase inhibitor (TKI) directed against HER2.

39. The method of claim 38 wherein the TKI directed against HERZ is lapatinib, neratinib, or tucatinib.

40. The method of any of the previous claims wherein administration of the inhibitor of the target protein or gene is performed during a period that at least partially overlaps with a period of administration of the HER2 binding agent or HER2 inhibitor.

41. The method of any of the previous claims wherein administration of the inhibitor of the target protein or gene occurs at least in part prior to administration of the HER2 binding agent or HERZ inhibitor.

42. The method of claim 40 or 41, wherein administration of the inhibitor of the target protein or gene is carried out for a period of time at least sufficient for the inhibitor to increases sensitivity of the tumor to the HERZ binding agent or HER2 inhibitor.MRG: 0680.003419W001 43. The method of any of the previous claims wherein administration provides a therapeutically active amount of the inhibitor of the target protein or gene in the body over a period of time that overlaps with a period of time where there is a therapeutically active amount of the HER2 binding agent or HER2 inhibitor in the body.

44. The method of claim 43 wherein the therapeutically active amount of the inhibitor of the target protein or gene in the body occurs at least prior to when there is a therapeutically active amount of the HER2 binding agent or HER2 inhibitor in the body.

45. The method of any of the previous claims, wherein the inhibitor is administered from a composition that is different from a composition used to administer the HER2 binding agent or HER2 inhibitor.

46. A pharmaceutical composition comprising an inhibitor of a target protein or gene encoding the target protein, wherein the inhibitor increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, and a HER2 binding agent or HER2 inhibitor.

47. The use of pharmaceutical compositions for treating a subject, wherein the subject comprises a tumor that is HER2 heterogeneous or HER2-low, or wherein the tumor is resistant to anti- HER2 therapy, wherein the one or more pharmaceutical composition(s) comprise (a) an inhibitor of a target protein or gene encoding the target protein, wherein inhibition of the target protein or gene increases sensitivity of the tumor to a HER2 binding agent or HER2 inhibitor, and (b) a HER2 binding agent or HER2 inhibitor.

48. The use of pharmaceutical compositions according to claim 47, wherein (a) the subject has any one more of feature(s) of claims 13-20, (b) the compositions have any one or more compound(s) of claims 21-39, (c) use is carried out using any feature of claims 40-45, or any combination of (a)-(c).

Images