Compositions and methods for inhibition of CASP3 and CASP7 for cancer therapy

Abstract

Methods of using inhibition of CASP3 and CASP7 to treat cancer are provided. CASP3 and CASP7 inhibition can be synthetic lethal with lack of expression of functional proteins involved in homologous recombination, DNA damage response pathways, splicing, or mitotic DNA damage checkpoint signaling. CASP3 and CASP7 inhibition can also be lethal to cells when combined with a second therapeutic such as a DNA damaging agent, an autophagy-inducing agent, a proteasome inhibitor, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor. Non-canonical p29 and p30 fragments of CASP7 can be used as a biomarker for prognosis or to determine if a cancer is likely to respond to treatment using inhibition of CASP3 and CASP7. Inhibition of CASP3 and CASP7 can also treat cancer in certain contexts in which resistance to PARP inhibitors has developed by inhibiting PARP1.

Description

COMPOSITIONS AND METHODS FOR INHIBITION OF CASP3 AND CASP7 FOR CANCER THERAPYCROSS-REFRENCE TO RELATED APPLICATIONS

[0001] This application claims priority to, and the benefit of, United States provisional patent application No. 63 / 587254 filed 2 October 2023. The foregoing application is incorporated by reference herein in its entirety for all purposes.SEQUENCE LISTING

[0002] This specification includes a sequence listing in XML format with the file name B537 0210-WO.xml and a size of 30.3 KB created on 1 October 2024 which is filed herewith and which is incorporated by reference herein.TECHNICAL FIELD

[0003] Some embodiments relate to the inhibition of CASP3 and CASP7. Some embodiments relate to methods and compositions for treating cancer using inhibitors of CASP3 and CASP7. Some embodiments relate to methods and compositions for combination therapies for treating cancer using inhibitors of CASP3 and CASP7 in combination with another therapeutic agent.BACKGROUND

[0004] Caspases are activated in response to cell stress, and growing evidence supports their function in non-apoptotic cellular processes. A role for caspases in promoting stress-induced cytoprotective autophagy has been demonstrated in Drosophila.

[0005] Caspases are cysteine dependent aspartic proteases and are traditionally well known for their role in proteolysis during the final stages of apoptosis. It is increasingly appreciated that caspases also play important roles in non-apoptotic cellular processes in normal development and in disease conditions. Given the evolutionarily conserved non-apoptotic caspase roles across phyla, it has been postulated that the primordial function of caspases was to regulate cellular stress adaptations. Further supporting this notion are the distinct CASP3 and CASP7 dependent proteolytic landscapes in cells exposed to non-lethal stress conditionscompared to lethal stress conditions (Conde-Rubio et al., 2021). Such non-apoptotic stress adaptive roles may explain unexpected observations such as the association of high caspase expression with enhanced tumor progression or the lack of correlation between caspase expression and apoptosis in a variety of cancers including breast cancer. Further, caspases are ubiquitously expressed in most cells, even in the absence of cell stress, and the same stress stimuli that activate cell stress response pathways can also activate caspases. However, the processing, regulation, and activity of caspases during non-lethal cellular stress have not been well examined.

[0006] Macroautophagy, hereafter referred to as autophagy, is an evolutionarily conserved intracellular lysosome-mediated degradation and recycling process that plays significant roles in normal development, aging and diseases including cancer. While autophagy commonly occurs at basal levels, it is also one of the major pathways upregulated in response to a variety of stressors such as nutrient deprivation, hypoxia, reactive oxygen species, DNA damage or pathogens. Autophagy supports stress adaptation and cell survival through the degradation and recycling of damaged organelles and macromolecules to facilitate the production of energy or essential cellular components.

[0007] The interplay between the survival-promoting (cyto protective) pathway of autophagy and the cell death pathway of apoptosis ultimately determine cell fate during stress, but the underlying mechanisms are not well understood. It is widely accepted that the interactions between these two processes oppose each other, and hence the final fate of a cell is determined by a tug of war between pathways. In line with this model, caspases were shown to suppress autophagy by direct cleavage of autophagy regulators or core autophagy proteins. Conversely, apoptotic pathway components, including caspases, have been implicated in the positive regulation of cyto protective autophagy in some contexts. However, whether human effector caspases have an evolutionarily conserved function is unknown, as the contexts, molecular mechanisms and pathways involved in non-lethal stress adaptation by human caspases have not been well investigated.

[0008] All caspases exist in the cell as inactive zymogens (pro-caspases) consisting of a pro-domain, a large ~p20 subunit containing the catalytic site, asmall ~pl0 / pl5 subunit, and may contain a linker sequence between subunits. Upon activation in apoptosis, the pro-domain and the linker region are removed by proteolytic cleavage in a temporal order unique to each caspase, ultimately resulting in ~p20 and pl0 / pl5 cleaved caspase fragments. Once activated in the apoptotic pathway, effector caspases are known to cleave a large number of substrate proteins. Therefore, their activity must be tightly regulated, presumably in both apoptotic and non-apoptotic settings. One major outstanding question is how caspases participate in non-apoptotic functions without killing cells. Identified modes of caspase regulation include post-translational modifications (cleavage, phosphorylation, ubiquitylation), subcellular compartmentalization (mitochondria, cytosol, nucleus, stress granules) and modulatory protein interactions such as inhibitor of apoptosis protein (IAP) family members. Allosteric sites in CASP3 and 7 and an exosite (non-active site interaction) in CASP7 can also contribute to regulation and activity. However, caspase processing, activation and regulation in non-apoptotic functions and / or in non-lethal cellular stress conditions have not been well characterized.

[0009] In recent years, a number of natural and synthetic caspase inhibitors have been developed as potential drugs for the treatment of human pathologies related to dysregulation of caspases. A few have advanced to clinical trials, primarily to treat neurodegenerative disorders or in certain viral infections. In pre- clinical studies, caspase inhibition has been tested along with other anticancer therapies to explore the possibility of inducing alternate cell death pathways aimed at increasing cancer cell killing. With this purpose, caspase inhibitors Z-VAD-FMK, M867 and NH-23-C2 were tested in breast or lung cancer cell lines (Kim et al., 2008; Wood and Shillitoe, 2011; Poreba et al., 2019; Dhani et al., 2021). However, these inhibitors have not moved forward due to lack of specificity, development of resistance and / or poor understanding of their effects on caspases or inadequate understanding of caspase functions in general.

[0010] The enzyme poly(ADP-ribose) polymerase 1 (Parpl) functions in multiple DNA repair pathways and has become an attractive therapeutic target for a variety of cancers including BRCA-mutated breast, ovarian, pancreatic and prostate cancers (Slade, 2020; Rose et al, 2020). They are also in clinical trials for the treatment oflung cancer and AML (in Rose et al, 2020) as well as cervical cancer, endometrial cancer, glioma, sarcoma and advanced, refractory or metastatic solid tumours. In some trials, PARP inhibitors are used in combination with a second therapeutic agent. The most common application of Parp inhibitors is the treatment of subsets of these cancers which harbor loss of the homologous recombination (HR) repair proteins BRCA1 or BRCA2. Parp inhibitors are also used for maintenance treatment of BRCA-mutated cancers, and the Parp inhibitor niraparib is also approved for maintenance therapy irrespective of tumor BRCA mutation status (pCODR, 2020; Mirza et al 2016). More recently, the application of Parp inhibitors has expanded to include the treatment of tumors deficient in other components of the HR pathway and DNA damage response, and potentially for tumors with high levels of oxidative and replication stress (see Rose et al, 2020 for a detailed discussion).

[0011] Parp inhibitors are primarily used clinically in the context of cancers that are defective for genes involved in DNA repair pathways. In such contexts, Parp inhibitors result in synthetic lethality, a phenomenon where mutations or disruption in two genes simultaneously result in lethality but mutation or disruption of either single gene is viable. Variations in the concept of synthetic lethality now exist, where chemicals or small molecule inhibitors can be used to effect perturbation in one of the gene or gene products of interest such as Parpl (Zhang et al, Nature Comm, 2021). The synthetic lethal activity of Parp inhibitors enables selective targeting of cancer cells, and "represent a successful example of precision medicine applied in the clinic" (Slade, 2020). However, like other targeted therapies, the development of resistance to Parp inhibitors occurs frequently. There is a need to develop alternate strategies to inhibit Parp activity and / or to overcome resistance to Parp inhibitors.SUMMARY

[0012] One aspect provides a method of killing a cell by inhibiting both CASP3 and CASP7 in the cell having a particular genetic context. In one aspect, the genetic context is that the cell does not express a functional protein or is deficient in or has a loss-of-function mutation in one or more genes involved in in mitotic processes and spindle assembly, mitotic cytokinesis, chromosome localization,regulation of spindle checkpoint, regulation of chromosome separation, RIMA transport, immune response regulation, protein localization, RNA splicing, cell cycle regulation, heterochromatin organization, G2 / M phase transition, regulation of histone acetylation, regulation of histone methylation, homologous recombination and DNA damage response, or mitotic DNA damage checkpoint signaling. In one aspect, the genetic context is that cell does not express functional proteins or is deficient in or has a loss-of-function mutation in one or more genes involved in homologous recombination, DNA damage response pathways, splicing, or mitotic DNA damage checkpoint signaling. In one aspect, the genetic context is that the cell does not express a functional protein corresponding to or is deficient in or has a loss-of-function mutation in one or more of the following genes: BRCA1, BRCA2, SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN 1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18.

[0013] In some aspects, the presence of a genetic context as identified above can be used to determine if a cancer of a patient is likely to respond to treatment using inhibition of both CASP3 and CASP7. In some aspects, expression or relative expression levels of non-canonical p29 or p30 forms of CASP7 in a cancer of a patient can be used as a biomarker to determine if the patient is likely to respond to treatment using anti-cancer therapies, to provide a prognosis for a patient, or todetermine whether a cancer is likely to respond to treatment using inhibition of both CASP3 and CASP7. In some aspects, such methods can be used to select a therapeutic that is an inhibitor of CASP3 and CASP7 for administration to a patient.

[0014] In one aspect, a method of killing a cell is provided by inhibiting both CASP3 and CASP7 in the cell and further exposing the cell to a proteasome inhibitor, a DNA damaging agent, an autophagy-inducing agent, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor.

[0015] In some aspects, the cell is a cancer cell. In some aspects, the cell or cancer has developed resistance to PARP inhibitors. In some aspects, the inhibitor of CASP3 and CASP7 is a small molecule inhibitor of CASP3 and CASP 7 such as M867, Caspase-3 / 7 Inhibitor I or Caspase-3 / 7 Inhibitor II.

[0016] Certain aspects are further described with reference to the drawings and the following detailed description, which are illustrative and not limiting in nature.BRIEF DESCRIPTION OF THE DRAWINGS

[0017] FIGs. 1A-1E show that loss of CASP3 and CASP7 suppresses starvation- induced autophagy.

[0018] FIGs 2A-2G show that loss of CASP3 and CASP7 suppresses proteasome inhibition-induced compensatory autophagy and sensitizes cells to proteasome inhibitors.

[0019] FIG. 3A-3D show that CASP7 is cleaved non-canonically in non-lethal stress conditions, resulting in stable and detectable p29 / p30 fragments.

[0020] FIGs. 4A-4B show the location and sequence of calpain cleavage sites that generate the non-canonical p29 and p30 fragments of CASP7.

[0021] FIGs. 5A-5E show that PARP1 cleavage is increased and that PARylation and LC3B gene expression are reduced in CASP3 / CASP7 double knock-out cells.

[0022] FIGs. 6A-6E show that the CASP7-p29 / p30 fragments promote phosphorylation of the DNA damage response pathway protein H2AX.

[0023] FIGs. 7A-7F shows that inhibition of CASP3 and CASP7 mimic the effects of PARP1 inhibitors and have predicted synthetic lethal interactions with a number of genes.

[0024] FIG. 7G shows the results of in silico analyses to identify CASP3 and CASP7 genetic interactions with the GO terms corresponding to clusters 1-16 listed in Table 5.

[0025] FIGs. 8A-8B show that pharmacological inhibition of CASP3 and CASP7 enhances lethality in BRCA1 deficient SUM149PT breast cancer cell line.

[0026] FIGs. 9A-9B shows that pharmacological inhibition of CASP3 and CASP7 sensitizes cells to proteasome inhibitor MG132.

[0027] FIG. 10 shows that pharmacological inhibition of CASP3 and CASP7 leads to an increase in levels of cleaved Parpl.

[0028] FIGs. 11A-11B show that genetic inhibition of CASP3 and CASP7 sensitizes SKBR3 cells to DNA damage-inducing anticancer drug cisplatin.DESCRIPTION

[0029] The following description contains details which are illustrative and not limiting in nature.

[0030] As used herein, the terms "treat" or "treatment" include therapies administered to achieve any type of beneficial effect, including curing, killing cancer cells, preventing or limiting the spread or multiplication of or otherwise managing a condition including cancer.

[0031] The inventors have now determined that caspase 3 and 7 (CASP3 and CASP7), that normally kill cells, can instead promote adaptation and cell survival in non-lethal stress contexts. Without being bound by theory, the underlying mechanism is believed to involve non-canonical processing of caspase 7 and induction of both autophagy and the DNA damage response. The inventors demonstrate herein that human effector caspases CASP3 and CASP7 have an evolutionary conserved role in the positive regulation of cyto protective autophagy during non-lethal stress. The inventors found that the underlying mechanism involves non-canonical cleavage of CASP7 and modulation of PARP1 activity. The combined loss of CASP3 and 7 mimicked PARP1 inhibition phenotypes, including synthetic lethality with BRCA1.

[0032] The inventors have investigated the effectiveness of an exemplary small molecule inhibitor, caspase 3 / 7 inhibitor 1, in BRCAl-proficient and BRCAl-deficientcell lines, as well as in combination with proteasome inhibition or DNA damageinducing drugs, and have confirmed that inhibition of caspases 3 and 7 can be used to kill cells in the appropriate genetic context or when co-administered together with an appropriate therapeutic. Based on these findings, it can be soundly predicted that cancer treatment strategies can be developed, since inhibiting or knocking out both CASP3 and CASP7 can result in death of certain cells (e.g. cells with an appropriate genetic background including mutations in BRCA1 or other genes involved in DNA damage repair, homologous recombination, mitotic DNA damage checkpoint signaling, splicing, or the like) or can render cells (including cells without mutations in such genes) sensitive to treatment with other active agents such as proteasome inhibitors or DNA damaging agents. In some embodiments, the genetic background is a mutation in one or more genes in one or more of the clusters listed in Table 5, including clusters 15 (homologous recombination and DNA damage response), 16 (mitotic DNA damage checkpoint signaling), 13 (regulation of histone acetylation), 14 (regulation of histone methylation), or 7 (immune response regulation).

[0033] The inventors have further found that dual inhibition of CASP3 and CASP7 unexpectedly leads to an increase in PARP1 cleavage, representing a novel mechanism of inhibition of PARP1 as compared to existing inhibitors.

[0034] More specifically, the inventors have now determined that CASP3 and CASP7 promote cytoprotective autophagy induction in response to non-lethal stress, that CASP7 processing in non-lethal stress is distinct from its processing in apoptosis, that PARP1 activity and DNA Damage Response (DDR) signaling are reduced following CASP3 and 7 inhibition, and that loss of CASP3 and CASP7 is synthetic lethal with loss of BRCA1. The inventors have confirmed that results observed by the genetic manipulation of CASP3 and CASP7 expression can be replicated by using small molecule therapeutic agents to inhibit CASP3 and CASP7.

[0035] The results described in this specification show that effector caspases 3 and 7 (CASP3, CASP7) are required to promote cytoprotective autophagy in response to non-lethal stress in human breast cancer cells. In these conditions, CASP7 underwent non-canonical cleavage at two locations flanking a PARP1 exosite, resulting in stable CASP7-p29 / p30 fragments that could be detected as markersthat non-canonical cleavage is occurring. Loss of CASP3 and 7 led unexpectedly to increased PAR.P1 cleavage and decreased PAR.P1 activity. Also reduced was y-H2AX, and CASP7-p29 / 30 were sufficient to restore robust phosphorylation of this DNA damage response marker. Strikingly, yet consistent with these phenotypes, the loss of CASP3 and 7 exhibited synthetic lethality with loss of BRCA1. These findings support a role for human caspases in stress adaptation through PAR.P1 modulation and reveal new avenues for cancer therapy as described herein.

[0036] In addition to BRCA1, multiple other components in the homologous recombination (HR.) repair pathway, DNA damage response pathways, splicing, and mitotic DNA damage checkpoint signaling, were identified among candidate targets for synthetic lethality with CASP 3+7 inhibition via in siiico analysis, including the clusters listed in Table 5, including clusters 15 (homologous recombination and DNA damage response), 16 (mitotic DNA damage checkpoint signaling), 13 (regulation of histone acetylation), 14 (regulation of histone methylation), or 7 (immune response regulation). The inventors' findings support that contexts in which autophagy is upregulated as an adaptation / resistance mechanism or contexts in which a drug activates the DNA damage response may be sensitive to the dual inhibition of CASP 3+7. Therefore, in addition to BRCA-deficient contexts, cancer patients who are BRCA-proficient may still benefit from treatment strategies where CASP 3+7 inhibition is combined with drugs such as proteasome inhibitors (PI) or DNA damage inducing drugs.

[0037] The inventors' results suggest that combining CASP3 / 7 inhibition with PARP inhibition has an additive effect, and therefore it is soundly predicted that administering CASP3 / 7 inhibitors in combination with PARP inhibitors would have beneficial therapeutic effects for example in the treatment of cancer.

[0038] Additionally, the administration of CASP3 / 7 inhibitors in combination with other types of cancer therapy may be beneficial. For example, it is known that erlotinib leads to resistance development via PARP mediated autophagy upregulation and that combined treatment with erlotinib and PARP inhibitors increase the effectiveness of erlotinib therapy. It is therefore soundly predicted that combining CASP3,7 inhibitors with erlotinib would produce similar effects, andCASP3,7 inhibitors could potentially be useful in additional contexts to help overcome resistance to chemotherapeutic agents.

[0039] ERK inhibitors and MEK inhibitors are known to induce autophagy in pancreatic cancer. Therefore it is soundly predicted that administering inhibitors of CASP3,7 together with inhibitors of ERK or MEK may result in cell death in at least certain contexts. Examples of ERK inhibitors are described for example in Morris et al, 2013, Moschos et al, 2018 and Bhagwat et al, 2020.

[0040] Given the similar phenotypic effects of CASP3 / 7 inhibition and PARP inhibition, it is soundly predicted that inhibition of CASP3 / 7 can be used as a therapeutic strategy in any context in which PARP inhibitors can be used as a therapeutic. See e.g. Hai Nguyen et al, 2023. Further, it is soundly predicted that inhibition of CASP3 / 7 will also provide therapeutic benefit in certain contexts where PARP inhibitor-induced resistance mechanisms are not through restoration of the homologous recombination pathway, e.g. EGFR inhibitors. Specifically, the mode of action on PARP by existing PARP inhibitors is believed to be different than the mode of action of CASP3 / 7 inhibition on PARP, as existing PARP inhibitors have not been reported to cleave PARP, whereas the inventors observed cleavage of PARP when CASP3 and CASP7 were inhibited. Thus, CASP3 and CASP7 inhibitors may be able to overcome certain mechanisms of resistance to PARP inhibitors because they have a different mode of action, and further may be able to provide additional therapeutic effects to further improve the efficacy of PARP inhibitors.

[0041] Further, since CASP3+CASP7 inhibition act like an autophagy inhibitor, then inhibition of CASP3 and CASP7 in combination with an autophagy-inducing agent may be therapeutically effective in contexts where an autophagy-inducing drug and an autophagy inhibitor exhibit therapeutic effects, for example administration of an MEK inhibitor or an ERK inhibitor. See e.g. Kinsey et al, 2019 and Bryant et al, 2019.

[0042] Further, since the p29 and p30 forms of CASP7 are associated with promotion of DNA damage response signaling and promotion of cytoprotective autophagy, it is soundly predicted that patients or cancers expressing the p29 or p30 form of CASP7 (or overexpressing such forms compared to expression levels in healthy tissue) will likely benefit from the inhibition of CASP3 and CASP7.

[0043] It is further soundly predicted that cancers in which a gene that exhibits a synthetic lethal interaction with loss of CASP3 and CASP7 expression, or those in which PAR.P inhibitors are used clinically, will likely benefit from the inhibition of CASP3 and CASP7. Examples of such cancers include breast cancer as well as bladder cancer, lung cancer, pancreatic cancer, liver cancer, brain cancer, esophageal cancer, bone cancer, endometrial cancer, ovarian cancer, prostate cancer, or leukaemia as well as cervical cancer, endometrial cancer, glioma, sarcoma and advanced, refractory or metastatic solid tumours.

[0044] Currently known inhibitors of CASP 3 and CASP 7 include M867, Caspase- 3 / 7 Inhibitor I, and Caspase-3 / 7 Inhibitor II. Some of these inhibitors have different mechanisms of action than Caspase-3 / 7 Inhibitor I (i.e., CASP 3 / 7 inhibitors developed to prevent binding with substrates with DEVD site) (Yoshimori et al., 2004)). CASP 3 / 7 inhibitor 1 is an isatin sulfonamide-based, cell-permeable, reversible small molecule inhibitor. This small molecule binds to the three hydrophobic residues (Tyr204, Trp206, and Phe256) in the S2 pocket surrounding the catalytic cysteine residue of CASP3 and CASP7. The specificity is maintained since only the active sites of CASP3 and CASP7 have these unique hydrophobic residues. However, this inhibitor may still compromise function of other caspases, including caspase 9, at higher concentrations (Lee et al., 2000).

[0045] In some embodiments, inhibitors of CASP3 and CASP7 are used to kill cells, for example cancer cells, having a particular genetic context. In some embodiments, the genetic context is a cell that does not express, or does not express a functional version of, one or more proteins involved in homologous recombination, the DNA damage response, mitotic DNA damage checkpoint signaling, or splicing, or has a deficiency or loss of function in one or more genes involved in such pathways. In some embodiments, the cells are deficient in or have a loss-of-function mutation in one or more genes in one or more of the clusters listed in Table 5, including clusters 15 (homologous recombination and DNA damage response), 16 (mitotic DNA damage checkpoint signaling), 13 (regulation of histone acetylation), 14 (regulation of histone methylation), or 7 (immune response regulation), including genes involved in homologous recombination, the DNA damage response, mitotic DNA damage checkpoint signaling, or splicing. In someembodiments, the genetic context is a cell that does not express, or does not express a functional version of, or that is deficient in or has a loss-of-function mutation in one or more of BRCA (including BRCA1 or BRCA2), SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN 1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18.

[0046] In some embodiments, inhibitors of CASP3 and CASP7 are used in combination with a second therapeutic agent to kill cells, for example cancer cells. In some embodiments, the second therapeutic agent is a proteasome inhibitor, a DNA damaging agent, an autophagy-inducing agent, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor. In some embodiments, the inhibitor of CASP3 is a small molecule inhibitor of CASP3 and the inhibitor of CASP7 is a small molecule inhibitor of CASP7. In some embodiments, the same inhibitor inhibits both CASP3 and CASP7.

[0047] In some embodiments, methods of administering inhibitors of CASP3 and CASP7 to treat cancer are provided. In some embodiments, the inhibitors of CASP3 and CASP7 are co-administered together with a second therapeutic agent to treat cancer. In some embodiments, the second therapeutic agent is a proteasome inhibitor, a DNA damaging agent, an autophagy-inducing agent, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor. In some embodiments, theinhibitor of CASP3 is a small molecule inhibitor of CASP3 and the inhibitor of CASP7 is a small molecule inhibitor of CASP7. In some embodiments, the same inhibitor inhibits both CASP3 and CASP7. In some embodiments, the inhibitor of CASP3 and CASP7 can be administered together with the second therapeutic agent, or can be administered sequentially either before or after the second therapeutic agent. In some embodiments, the inhibitor of CASP3 and CASP7 and the second therapeutic agent can be formulated for administration and administered in any suitable manner, for example orally, intravenously, intramuscularly, intraperitoneally, or the like. In some embodiments, the inhibitor of CASP3 and CASP7 can be administered through a different mode of administration than the second therapeutic agent and optionally on a different schedule, for example one compound may be administered orally on a first dosage schedule and the second compound may be administered intramuscularly or intravenously on a second dosage schedule, or the like.

[0048] In some embodiments, the inhibitor of CASP3 and CASP7 is Caspase-3 / 7Inhibitor I (CAS 220509-74-0) having the following structure:In some embodiments, the inhibitor of CASP3 and CASP7 is M867 (CAS 680999-39-7) having the following structureCaspase- 3 / 7 Inhibitor II(CAS 775289-20-8) having the following structure

[0049] In some embodiments, rather than using inhibitors of CASP3 or CASP7, CASP3 and CASP7 are knocked out or knocked down in the target cells, for example using any gene editing method now known or later developed, such as CRISPR, zinc-finger nucleases (ZFN), transcription activator- 1 ike effector nucleases (TALEN), Sleeping Beauty (SB), meganucleases, or megaTALs, or any suitable construct todecrease expression of CASP3 and CASP7 such as interfering RNAs such as siRNA, miRNA, or the like.

[0050] In some embodiments, the inhibitor of CASP3 is an siRNA or other similar construct such as an miRNA that targets CASP3 and / or the inhibitor of CASP7 is an siRNA or other similar construct such as an miRNA that targets CASP7.

[0051] In some embodiments, the proteasome inhibitor is MG132, Bortezomib, or ixazomib.

[0052] In some embodiments, the DNA damaging agent is doxorubicin, cisplatin, anthracycline, temozolomide, carboplatin, cyclophosphamnide, dacarbazine, oxaliplatin, bendamustine, busulfan, carmustine, chlorambucil, ifosfamide, lomustine, melphalan, streptozocin, trabectedin, CX-5461, istodax or vorinostat.

[0053] In some embodiments, the autophagy-inducing agent is an ERK inhibitor or a MEK inhibitor, metformin, mTOR inhibitors, proteasome inhibitors, sorafenib, sunitinib, apatinib, gefitinib, vorinostat, erlotinib, nilotinib, anti-PD-1 antibodies, anti-PD-Ll antibodies, azacitidine, or the like (see e.g. Lee et al., 2023).

[0054] In some embodiments the PARP inhibitor is Olaparib (AZD2281), ABT-888 (Veliparib), Niraparib, Talazoparib, Rucaparib, Fuzuloparib or Pamiparib.

[0055] In some embodiments, the EGFR inhibitor is erlotinib.

[0056] In some embodiments, the ERK inhibitor is SCH772984 (ERKi), MK- 8353 / SCH900353, or LY3214996.

[0057] In some embodiments, the MEK inhibitor is trametinib.

[0058] In some embodiments, the cells are cancer cells, including breast cancer, bladder cancer, lung cancer, pancreatic cancer, liver cancer, brain cancer, esophageal cancer, bone cancer, endometrial cancer, ovarian cancer, prostate cancer, leukaemia, cervical cancer, endometrial cancer, glioma, sarcoma or advanced, refractory or metastatic solid tumour cells.

[0059] In some embodiments, screening is carried out to determine whether a patient or cancer is likely to respond to treatment using inhibition of CASP3 and CASP7. In some embodiments, such screening includes determining if the patient or cancer has a loss-of-function mutation in or does not express a functional protein corresponding to one or more genes in one or more of the clusters listed in Table 5, including clusters 15 (homologous recombination and DNA damage response), 16(mitotic DNA damage checkpoint signaling), 13 (regulation of histone acetylation), 14 (regulation of histone methylation), or 7 (immune response regulation), including proteins involved in homologous recombination, the DNA damage response, or mitotic DNA damage checkpoint signaling. In some embodiments, such screening includes determining if the patient or cancer has a loss-of-function mutation in or does not express a functional protein corresponding to one or more of BRCA (including BRCA1 or BRCA2), SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN 1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18. If it is determined that the patient or the cancer has a loss-of-function mutation in or does not express a functional protein corresponding to one or more of the foregoing genes, then it is concluded that the patient is likely to respond to treatment using an inhibitor of CASP3 and CASP7.

[0060] In some embodiments, the screening involves determining a homologous recombination repair deficiency (HRD) score in the cancer tissue, for example as described by Li et al., 2023, which is used as a biomarker to predict response to PARP inhibitors. The HRD score takes into account BRCA mutations, and also can include an assessment of mutation signatures, SNPs and other genes. A positive HRD score or a high HRD score can be indicative that a cancer is likely to respond to therapy using PARP inhibitors. The threshold HRD score that would beconsidered a high or positive HRD score may vary depending on the specific type of cancer or the specific PARP inhibitor, for example HRD scores such as greater than 42 may be predictive high HRD scores for breast cancer, while HRD scores greater than or equal to 33 or 42 may be predictive positive HRD scores for advanced ovarian cancer treated with Veliparib versus Olaparib and Niraparib, respectively, and the value of a high or positive HRD score in any particular context could be determined by one skilled in the art.

[0061] In some embodiments, screening is carried out to determine whether a patient is likely to respond to treatment using inhibition of CASP3 and CASP7. In some embodiments, such screening includes assaying the cancer for expression of a p29 form of CASP7, with it being concluded that the patient is more likely to respond to treatment using inhibition of CASP3 and CASP7 if the cancer expresses the p29 form of CASP7 at a higher level than corresponding healthy tissue. In some embodiments, such screening includes assaying the cancer for expression of a p30 form of CASP7, with it being concluded that the patient is more likely to respond to treatment using inhibition of CASP3 and CASP7 if the cancer expresses the p30 form of CASP7 at a higher level than corresponding healthy tissue.

[0062] In some embodiments, a method of selecting a therapeutic wherein the therapeutic is an inhibitor of CASP3 and CASP7 is provided. In some embodiments, such method comprises carrying out one of the above-described screening methods and then, if it is determined that the patient or the cancer is likely to respond to treatment using an inhibitor of CASP3 and CASP7, the method further comprises selecting an inhibitor of CASP3 and CASP7 as the therapeutic.

[0063] In some embodiments, a method of administering a therapeutic wherein the therapeutic is an inhibitor of CASP3 and CASP7 is provided. In some embodiments, such method comprises carrying out one of the above-described screening methods and then, if it is determined that the patient is likely to respond to treatment using an inhibitor of CASP3 and CASP7, the method further comprises administering the inhibitor of CASP3 and CASP7 to the patient. In some embodiments, the inhibitor of CASP3 and CASP7 is a small molecule inhibitor of CASP3 and CASP7, such as M867, Caspase-3 / 7 Inhibitor I or Caspase-3 / 7 Inhibitor II. In some embodiments, the inhibitor of CASP3 and CASP7 is an interfering RNAsuch as an miRNA or an siRNA. In some embodiments rather than administering a therapeutic, CASP3 and CASP7 are knocked out in the cancer cells, for example using any suitable gene editing method now known or later developed, such as CRISPR, zinc-finger nucleases (ZFN), transcription activator-like effector nucleases (TALEN), Sleeping Beauty (SB), meganucleases, or megaTALs.

[0064] In some embodiments, the cancer or cells have developed resistance to PARP inhibitors, and administration of inhibitors of CASP3 and CASP7 may provide an alternative mechanism for decreasing PARP activity via a distinct mechanism of action. In some embodiments even in which cells have not developed resistance to PARP inhibitors, administration of inhibitors of CASP3 and CASP7 together with PARP inhibitors may provide additional therapeutic effects, since the inventors have demonstrated that inhibition of CASP3 and CASP7 is not merely redundant with inhibition of PARP.Examples

[0065] Specific embodiments are further described with reference to the following examples, which are intended to be illustrative and non-limiting in nature.Materials and Methods

[0066] Cell lines and cell culture conditions. Cells were maintained according to accepted conditions and protocols. Cell lines stably transfected with plasmids expressing CASP constructs were maintained using Geneticin (G418) at 1 mg / ml or Hygromycin at 500 pg / ml for selection. The selection media was replaced with standard media two passages before cells were used in experiments.

[0067] Pharmacological treatment. After growing for one day, cells were treated with CASP 3 / 7 inhibitor 1 and / or other drugs used in this study (proteasome inhibitor MG132, staurosporine or cisplatin. Stock solutions were prepared in dimethylsulfoxide (DMSO)). Control cells (non-treated or DMSO) were treated with a comparable concentration of DMSO.

[0068] Crystal violet assay. Cells were plated at 1.5 x 104cells per well in 24-well plates. At 2 - 4 hours after seeding, cells were treated with the drug (CASP 3 / 7 inhibitor 1 and / or MG132 / cisplatin) and allowed to grow in the media containing the drug for 48 hours. For CASP3,7 knockdown, parental cells were transfected with lOnMCASP siRNA or Scramble-siRNA. For Olaparib treatment of SUM149PT, cells were transfected with 5 nM of CASP or Scramble-siRNAs. At 48 hours after transfection or 72 hours after seeding, cells were treated with the drug (PI or Olaparib) and allowed to grow in the media containing the drug for 24 hours. Then the cells were washed twice in PBS and continued to grow in drug-free fresh media for another 3 days. After removing the media and washing with PBS, cells were stained with 0.1% crystal violet for 15 minutes, washed with distilled water and air dried before taking images using the ChemiDoc MP System (BioRad).

[0069] Incucyte system. Viability was tested using the Incucyte system across a range of concentrations of CASP3 / 7 inhibitor 1 : 0 pM, 12.5 pM, 25 pM, 50 pM, 100 pM, 150 pM, and 200 pM. Staurosporine-treated cells were used as a positive control for cell death and DMSO was used as the negative control. After drug treatments, phase confluence values were collected every 2 hours (four images per well) for 72 hours total using the Incucyte system and absolute phase confluence values were used to calculate the Area Under Curve (AUC) at each concentration. IC50 values are based on normalised AUC of each cell line to t=0. For IC50, curve [Inhibitor] vs response -variable slope was used.

[0070] Mice and PDX. To obtain cell lysate from murine organs, tissues from three female CD-I mice, 10 weeks old, were flash-frozen and gifted by Dr. Nancy dos Santos and Nicole Wretham. All experiments were conducted in accordance with the standards and guidelines of the Canadian Council on Animal Care and were approved by the University of British Columbia Animal Care Committee. To obtain cell lysate from PDX, breast tumor tissues originally derived from human were grown as xenografts in mice as detailed in Eirew et al., (2022) harvested, frozen in DMEM / FCS with 6-10% DMSO and provided by Dr.Sam Aparicio and Dr. Peter Eirew. Ethical approval for research using these human breast tumor tissues was obtained from the Research Ethics Board of BC Cancer and Simon Fraser University.

[0071] Starvation, pharmacological or Bafilomycin (BafAl) treatment. After growing for three to four days, cells were subjected to starvation or treated with proteasome inhibitors (or other drug treatments used in this study; rapamycin, staurosporine, olaparib). Cells were subjected to amino acid starvation for two to 24 hours (as indicated) in Earle's Balanced Salt Solution (EBSS). For well-fed cells, themedia was replaced with fresh standard growth media. For the autophagic flux assay with Bafilomycin Al (BafAl), cells were incubated in 50 nM of BafAl to inhibit lysosomal fusion for two hours prior to harvesting cells for subsequent analyses. Staurosporine, Rapamycin, PARP inhibitor, Olaparib, and proteasome inhibitors MG132 and Bortezomib stock solutions were prepared in dimethylsulfoxide (DMSO) and stored in aliquots at -20°C. Drugs were freshly diluted in media to minimize DMSO concentration and cells were treated at indicated concentrations and durations. Control cells (non-treated or DMSO) were treated with a comparable concentration of DMSO.

[0072] Cell lysate extraction for western blotting. Cell lysates for western blotting were obtained according to accepted protocols. Cells were plated in 6-well plates at a density of 2 x 105cells per well for all cell lines except for MDA-MB-231 which was plated at 1.5 x 105cells per well. To prepare lysate from frozen mouse organs or PDXs, samples were cut into small pieces with a single edge blade on a clean petri dish, transferred to lysis Matrix M tubes containing a single zirconium ceramic bead and RIPA lysis buffer (300-500 pl) supplemented with IX complete-protease inhibitor and IX PhosSTOPTMand incubated on ice for 5 minutes. Samples were homogenised using MP Fast Prep Tissue Homogenizer, spun down at 13000 rpm for 2 minutes for debris to form pellets and then the supernatants were transferred to eppendorf tubes before being centrifuged at 13000 rpm for 20 minutes at 4°C. Lysates were transferred to new tubes avoiding pellets and floating lipid / fat layer.

[0073] Western blot analysis. Proteins in cell / tissue lysates were quantitated using the Pierce BCA Protein Assay Kit according to accepted protocols.Protein signals were detected with the Bio-Rad Clarity Western ECL™ Enhanced Western Blotting substrate or with SuperSignal™ West Femto Maximum Sensitivity Substrate using Bio-Rad ChemiDoc XRS (or MP) System. Densitometry was performed using Image Lab 5.1 software (BioRad). Protein levels were normalized to the loading control, either p-actin or vinculin.

[0074] Cell fractionation. Nuclear and cytoplasmic extracts for western blot analyses for H2AX and x-H2AX were prepared using NE-PER™ Nuclear Cytoplasmic Extraction Reagent kit (Pierce) following the manufacturer's instructions.

[0075] Proteasome activity assay. 2000 cells were seeded in a 96-well, white cell culture clear-bottom microplate (Greiner bio-one, 655983) and cultured in appropriate media for up to four days. After treating cells with proteasome inhibitors for 24 hours at indicated concentrations, proteasome activity was measured using Proteasome-Glo Cell-Based Reagents as instructed (Promega). Chymotrypsin-like, caspase-like and trypsin-like activities were individually measured after incubating the corresponding luminogenic substrates with cells for 10 minutes. The degree of proteasomal activity (luminescence) was measured on a Synergy H4 Hybrid (BioTek) plate reader

[0076] siRNA transfection. For siRNA transfections, SKBR3 and SUM149PT cells were plated at 2x l05cells per well and MDA-MB-231 cells were plated at 1.5 x lO5cells per well, in 6-well plates in a total of 2 ml standard media. Twenty-four hours after plating, cells were transfected with 10 nM of CASP siRNA or corresponding Scramble control siRNA (provided by IDT) using 4 pl of Lipofectamine RNAiMAX™ (Invitrogen) as per manufacturer's recommendations. siRNAs are listed in Table 1. 48 hours after transfection, cells were subjected to well-fed / starved or untreated / treated conditions with and / or without 50 nM BafAl treatment for the final 2 hours and harvested for western blot analysis as described above. SUM149PT cell lines were transfected with 5 nM of siRNA in experiments used to detect the combined effect with Olaparib.Table 1. : siRNAs used in knockdown experiments.

[0077] CRISPR-Cas9-based knockout line creation. Single guide RNA sequences targeting the human CASP3 and CASP7 genes were designed using Crispor.org (Concordet and Haeussler, 2018) and cloned into PX458 (Cas9-GFP vector with AMP resistance, Addgene, plasmid # 48138) or PX459 (Cas9- Puromycin vector with AMP resistance, Addgene, plasmid # 62988) backbone vector following the protocol provided by Addgene. For gRNA sequences, see Table 2. Plasmids were then transformed into One Shot Stbl3 chemically competent E. coli cells. The resulting CRISPR plasmids were isolated, sequenced and transfected into the SKBR3 or MDA- MB-231 cell line using Lipofectamine 3000. Cells were selected by FACS sorting (PX458) or puromycin treatment (PX459) andisogenic knockout clones were validated by western blotting and sequencing.Table 2. CRISPR gRNA sequences._

[0078] CASP construct generation and transfection. For CASP3, a clone from Sino Biological (HG10050-CH) was obtained and used directly for transfection. CASP7 plasmids containing CASP7-WT, CASP7-inactive (CASP7-C186A), CASP7-APro or non- cleavable at the prodomain were kindly gifted by Dr. Salvesen (Denault and Salvesen, 2003). The site-directed mutagenesis method was employed to modify nucleotides as needed. CASP7-p30, CASP7-p29, and CASP7 calpain cleavage sites mutated (CASP7-CCM) constructs were generated by PCR using CASP7-WT as the template. The primers and restriction enzymes used are listed below. For CASP7-CCM, the amino acid residues at 1stand 2ndcalpain cleavage sites were changed into 'A' as shown in FIG. 4B. Gene sequencing was performed to verify the sequence integrity of all the constructs.

[0079] To obtain stable cell lines, single or double CASP KO cells (CASP7 KO, CASP3,7 DKO) were plated at 2xl05cells / well and transfected with empty vectors (pcDNA3 for CASP7 constructs and pCMV3-C-His for CASP3 constructs) or plasmids containing the CASP constructs, using Lipofectamine 3000 (Invitrogen) as per manufacturer's instructions. CASP3 and CASP7 construct expressing cells were selected with 500 pg / ml Hygromycin and / or 1 mg / ml G418, respectively. Antibodieswere used to evaluate the expression of CASP constructs in cells that survived drug treatment. The selection agent(s) was removed for two passages prior to experiments. The following primers and restriction sites / enzymes were used in making CASP constructs1. For cloning CASP7-p30 fragment (via Acc65I and Xhol sites)Forward - 5'-TGACCTAGGGTACCATGAGTAAGAAGAAGAAAAATGTC-3' (SEQ ID NO:8)Reverse - 5'-TGACCTAGCTCGAGCTACTTGTCATCGTCGTCCTTGTA-3' (SEQ ID NO:9)2. For cloning CASP7-p29 fragment (via Acc65I and Xhol sites)Forward - 5'-TGACCTAGGGTACCATGCGATCCATCAAGACCACCCGG-3' (SEQ ID NO: 10)Reverse - 5'-TGACCTAGCTCGAGCTACTTGTCATCGTCGTCCTTGTA-3' (SEQ ID NO: 11)3. For site-directed mutagenesis of 1stcalpain cleavage site for CASP7-CCM construct fFVPSLFSKKKKNVTMRSIKTT (SEQ ID NO: 12) to FVPSAAAAKKKNVTAAAIKTT (SEQ ID NO: 13))5'- / 5Phos / CGGTCCTCGTTTGTACCGTCCGCAGCTGCCGCAAAGAAGAAAAAT-3' (SEQ ID NO: 14)4. For site-directed mutagenesis of 2ndcalpain cleavage site for CASP7-CCM construct fFVPSLFSKKKKNVTMRSIKTTT (SEQ ID NO: 12) to FVPSAAAAKKKNVTAAAIKTT (SEQ ID NO: 13))5'- / 5Phos / AAGAAGAAAAATGTCACCGCTGCCGCAATCAAGACCACCCGGGAC-3'(SEQ ID NO: 15)

[0080] Immunoprecipitation of CASP7-p30 / 29 fragments for Edman sequencing. CASP7 or mouse IgG (control) antibody-bound beads were prepared by incubating anti-CASP7 antibody, or mouse-IgG with Dynabeads protein G in IP buffer (25mM Tris-HCI pH 7.5, 150mM NaCI, IX PhosSTOP,l% Np-40, 5% Glycerol and ImM EDTA with IX complete mini protease inhibitor (Roche) at 4 °C overnight. CASP7 KO SKBR3 cells or CASP7 KO SKBR3 cells expressing the CASP7-WT construct were grown in T75 flasks for 3 days, fed with fresh growth media or starved for 4 hours in EBSS, trypsinized, collected and lysed by incubation in cell lysis buffer. Cell lysates wereprocessed according to accepted protocols to capture and release proteins for analysis by western blotting. After PonceauS staining, the band corresponding to CASP7-p30 / 29 that contains ~l-2 ug of proteins on PVDF was cut and sent for Edman degradation sequencing (for the first 10 amino acids at N-terminus).

[0081] Microscopy images of DALGreen-stained cells were obtained from a Zeiss Axio Observer (Zl / 7) inverted fluorescence microscope equipped with an Apotome.2 and an AxioCam MRm R3 camera (Zeiss). Images were obtained at 20X magnification using the Zen software (version 2.5, blue edition; Zeiss) and analyzed using ImageJ64 software.

[0082] For staining with DALGreen, seeded cells were preincubated with either of the dyes in the growth media for 30 minutes at 37 °C, then washed twice with PBS and subjected to autophagy induction by amino acid starvation. Live cells were imaged for fluorescence markers using constant settings. The number and the area of DALGreen positive punctae was measured using ImageJ64 software and normalized to the cell number.

[0083] RIMA isolation and RT-qPCR analysis. Total RNA was isolated and purified using the RNeasy Plus Mini Kit (QIAGEN,) as per manufacturer's instructions. The quantification of RNA was performed using a Nanodrop Spectrophotometer. The quantitative reverse transcription polymerase chain reaction (RT-qPCR) was performed to quantitate selected ATG transcript expression levels using One-Step Plus SYBR Green reagent kit in 15 ul reactions containing 100 ng of total RNA and 0.1 pM of each primer on the 7900 sequence detection system (Applied Biosystems). Manufacturer's recommended cycling conditions were used. The ATG proteins and the sequences of forward and reverse primers designed using Primer-BLAST (Ye et al., 2012) are listed in Table 3. The mRNA expression levels were determined by the Comparative Cycle Threshold (CT) method (2_[delta][delta]ct) normalizing to internal reference gene (lactin). The expression is presented as the fold change relative to the control. RT-qPCR analyses were done in two or three technical replicates and experiments were repeated at least thrice.Table 3. Forward and Reverse primer sequences used for qRT-PCR

[0084] DNA fragmentation / Comet assay. The level of DNA damage (single and double strand breaks) was evaluated by the comet assay (single cell-gel electrophoresis) as previously described (Lu et al., 2017). Images were taken using a fluorescent microscope at 10X objective and DNA damage was quantified as the comet tail moment (product of the tail length x the fraction of total DNA in the tail) using CometScore 2.0 software (Robison et al., 2005). Approximately 200 randomly selected nucleoids were scored per each treatment. Experiments were repeated twice.

[0085] In silico co-essentiality and genetic interaction network mapping. In silico genetic interaction screening was performed using GRETTA (vO.99.2) (Takemon and Marra, 2023) on the R statistical software (v4.2.2) (R Core Team, 2020). The inventors computed on the cancer dependency map (DepMap) public release version 22Q2 (DepMap, 2022) data from FigShare (DepMap, Broad (2022). DepMap 22Q2 Public, figshare. Dataset, doi. org / 10.6084 / m9. figshare.19700056. v2, available online and accessed on 11 August 2022) according to Takemon and Marra (2023). GRETTA was used to identify cell lines with combined low or high CASP3 and CASP7 protein expression. We identified a group of pan-cancer cell lines below the 10th percentile of CASP3 and CASP7 protein expression (16 cell lines; low expressors) and those above the 90th percentile of both CASP3 and CASP7 protein expression (10 cell lines; high expressors) as the control comparator. The normalised (scaled and mean centered) gene and protein expression values for the selected cell lines were extracted from Nusinow et al., (2020) using GRETTA for comparison between the low and high expressor groups.

[0086] Genetic interactors of combined CASP3 and CASP7 low expression were predicted using GRETTA, which performed pairwise Mann-Whitney U tests between the high expressors and low expressors cancer cell line groups for all 17,386 genes to obtain p-values. P-values were adjusted for multiple testing using a permutation (10,000 randomised resampling). Candidate genetic interactors of combined low CASP3 and CASP7 expression were called using a threshold of adjusted Mann- Whitney U test p-value < 0.05 and a median lethality probability > 0.5 in at least one group as previously defined (Takemon and Marra, 2023). To highlight candidate lethal interactors, genetic interaction scores computed by GRETTA were used to generate lethal genetic interaction scores by collapsing genetic interaction scores below zero to zero. A higher interaction score indicates an increased likelihood that a gene knockout is lethal in the test group. Rank is based on lethal GI scores.

[0087] Statistical analysis. In each graph, error bars represent standard error (SE) of n = 3 or 2 independent experiments. As indicated in the legends, statistical significance was calculated by Students T-tests (for comparisons between samples) analysis of variance (two way ANOVA) with appropriate post tests for multiple comparisons (Tukey's, Dunnett, or Sidak) using GraphPad Prism version 8.4.1 for Windows (GraphPad Software, San Diego California USA). Significance (p-values) indicated are relative to the control unless otherwise indicated. P-values less than 0.05 were determined as significant. Significance is indicated as follows unless otherwise noted: *P<0.05, **P<0.01, ***P<0.001, NS., not significant.Example 1.0 - Loss of CASP3 and CASP7 Suppresses Starvation-Induced Autophagy

[0088] The functional conservation of the two-most closely related human caspases, effector caspases 3 and 7 (CASP3, CASP7) was evaluated by the inventors. First, the inventors employed standard LC3B-based autophagy assays as these have been used successfully to monitor autophagy in breast cancer cell lines, including SKBR3, and under similar stress conditions, and found that low levels of cleaved-PARP (at ~89 kDa) were observed in both well fed (non-stressed) control cells and starved (stressed) cells. There were no signs of cell death as indicated by the absence of a marked increase in cleaved-PARPl in the starvation conditions. However, to avoid any potential overlap with cell death pathways, the inventorsselected 8 hours starvation as the optimal time point for detecting non-lethal stress related starvation-induced autophagic flux in SKBR.3 cells in this example.

[0089] Next, single knockdown (KD) and double knockdown (DKD) experiments with CASP-siRNAs were carried out to determine whether amino acid starvation- induced autophagy was CASP3- and / or 7-dependent. The control scramble-siRNA treated cells showed an increase in LC3BII levels in the presence of BafAl indicating an increase in autophagic flux when cells were starved. In CASP3-siRNA or CASP7-siRNA treated cells, the starvation-induced autophagic flux was comparable to that detected in scramble-siRNA treated cells. However, a statistically significant suppression of autophagic flux was observed in cells treated with both CASP3- and CASP7-siRNAs (FIG. 1A). FIG. 1A shows the quantification of LC3BII- based autophagy flux in starved cells relative to the starved scramble-siRNA control as determined by western blot for SKBR3 cells transfected with scramble, CASP3 and / or CASP7 siRNAs (48 hours) and incubated in fed conditions or starved for 8h with BafAl(50 nM) for the final 2 hours.

[0090] Without being bound by theory, in single caspase knockdown experiments, the lack of a significant effect on starvation-induced autophagy upregulation might be due to the presence of residual caspase levels. To address this, the inventors generated CASP3 and CASP7 single knockout (KO) and double knockout (DKO) SKBR3 cell lines. In single caspase KO cells, the inventors found that the effect on starvation-induced autophagy was negligible or undetectable compared to control parental cell lines. Without being bound, this observation suggests that CASP3 and 7 have partially overlapping functions and / or single KO cell lines may have activated compensatory mechanisms to overcome the loss of the other caspase. In support of this possibility, and in agreement with what was observed in DKD cells, there was a significant impairment of autophagic flux in CASP3,7 DKO cell lines (FIG. IB). FIG. IB shows the quantification of LC3BII- based autophagy flux in starved cells relative to the parental control as determined by western blots for CASP3 and CASP7 single or double knockout (DKO) SKBR3 cells in fed or starved (8h) conditions, with BafAl(50 nM) in the final 2h.

[0091] To confirm the specificity of these findings, the inventors generated multiple caspase DKO SKBR3 cell lines. In all five of the CASP3,7 DKO lines tested,it was observed that amino acid starvation-induced autophagy was compromised. To exclude the possibility of cell line-specific effects, experiments were repeated using another breast cancer cell line, MDA-MB-231, and similarly it was found that only when both caspases were knocked-down or knocked-out the induction of autophagic flux was significantly compromised. Lastly, when wild-type constructs of both CASP3 and CASP7 were stably reintroduced into DKO cells, the starvation- induced upregulation of autophagy was partially rescued (FIG. 1C). FIG. 10 shows the quantification of LC3BII-based autophagy flux in starved cells relative to starved parental cells as determined by western blots from SKBR.3 parental, DKO or CASP3,7-WT re-expression in DKO cells in fed or starved (8 hours) conditions, with BafAl (50 nM) in the final 2 hours.

[0092] The processing of LC3B to form LC3BII is a hallmark of autophagy. However, since LC3B is also involved in autophagy-independent processes, the inventors also measured autophagy using an orthogonal method employing the DALGreen (autolysosome marker) fluorescent marker (Iwashita et al., 2018). In accordance with the LC3B-based autophagy assay results, DALGreen (FIGs. 1D-1E) indicated that amino-acid deprivation-induced autophagy is significantly compromised in CASP3,7 DKO cells. CASP3 and CASP7 re-expression fully rescued autophagy. Immunoflurescence images of SKBR.3 parental, DKO or CASP3,7-WT reexpression in DKO cells treated with 0.25 pM DALGreen in fed or starved (8h) conditions were taken, with BafAl (50 nM for 8h) in both fed and starved conditions as controls. FIGs. 1D-1E show quantification of DALGreen immunofluorescence in starved cells. Tukey boxes show number of punctae (FIG. ID) or punctae area (FIG. IE) per cell. n= 600-700 cells. These observations together indicate that CASP3 and CASP7 play a partially redundant positive regulatory role in non-lethal amino acid deprivation-induced autophagy. In graphs in FIG. 1, data are shown as mean ± SEM. n = at least 3 independent experiments except in FIGs. ID and IE.*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS., not significant. In FIGs. 1A, IB, and 10, one-way ANOVA with Dunnett's post-test. In FIGs. ID and IE one-way ANOVA with Tukey's post- test.Example 2.0 - Loss of CASP3 and CASP7 suppresses proteasome inhibition-induced compensatory autophagy and sensitizes cells to proteasome inhibitors

[0093] It is well documented that autophagy is upregulated as a compensatory mechanism to maintain protein homeostasis when proteasome function is compromised. Next, the inventors investigated whether proteasome inhibition (PI)- induced autophagy was also dependent on the effector CASP3 and 7. MG132 at 0.5 pM was chosen as the best concentration of a representative proteasome inhibitor to be used for autophagy monitoring in non-lethal stress conditions (FIGs. 2A-2B). FIG. 2A shows quantification of LC3B-based autophagy flux in MG132 treated cells relative to untreated (0 pM MG132; DMSO vehicle only; NT) SKBR3 shown as determined by western blot for cells treated with MG132 at increasing dosage for 24h, with BafAl (50 nM) in the final 2 hours. The levels of LC3BII were normalized to loading control and shown relative to the untreated control (NT).

[0094] FIG. 2B shows proteasome activity in response to increasing concentrations of MG132 as depicted by chymotrypsin, caspase and trypsin-like activities. The reduction of proteasome activity was confirmed directly by proteasome activity assays (FIG. 2B) and indirectly by the accumulation of ubiquitinated proteins, which were observed to be increased at concentrations of MG132 greater than 0.5 pM.

[0095] In line with what was observed in amino acid starvation conditions, the Pl-induced autophagic flux was significantly inhibited when both CASP3 and 7 were simultaneously knocked-down or knocked-out (FIGs. 2C and 2D). Although the effect was modest, CASP7 KO alone also did result in a statistically significant reduction in Pl-induced autophagy (FIG. 2D). FIG. 2C shows quantification of LC3B-based autophagy flux in MG132 treated cells relative to the MG132 treated scramble-si RNA control as determined by western blots from SKBR.3 cells transfected with scramble, CASP3 and / or CASP7 siRNAs (48 hours) and treated with vehicle DMSO only or MG132 (0.5 pM) for 24 hours, with BafAl (50 nM) in the final 2 hours. FIG. 2D shows quantification of LC3B-based autophagy flux in MG132 treated cells relative to the MG132 treated parental control as determined by western blots from CASP3, 7 single knockout or DKO SKBR3 cells treated with vehicle DMSO or MG132 (0.5 pM) for 24 hours, with BafAl(50 nM) in the final 2 hours.

[0096] Similar results were observed in MDA-MB-231 cell line. The effect of knockdown of CASP3 and / or 7 was also tested in SKBR.3 cells treated with the clinically approved proteasome inhibitor, Bortezomib and results were comparable to what was observed with MG132 treated cells. Pl-induced LC3BII upregulation was partially rescued by re-expression of CASP3 and CASP7, supporting the requirement for these effector caspases in Pl-induced autophagy (FIG. 2E). Collectively, these results indicate that CASP3 and CASP7 have positive regulatory roles in autophagy induction in response to starvation stress or proteasome inhibition stress. FIG. 2E shows quantification of LC3BII- based autophagy flux in MG132 treated cells relative to DMSO treated parental control as determined by western blots from SKBR.3 parental, DKO or CASP3,7-WT reintroduced into DKO cells treated with vehicle DMSO or MG132 (0.5 or 1.0 pM) for 24 hours with BafAl(50 nM) in the final 2 hours.

[0097] Since autophagy has been reported to promote cell death in some contexts the inventors investigated whether the CASP3 and 7-dependent stress- induced autophagy contributes to cell death or cell survival. Crystal violet cell viability assays were carried out following PI treatment (MG132 or Bortezomib) of parental and CASP3,7 knockout or knockdown SKBR3 or MDA-MB-231 cells (FIGs. 2F, 2G). A modest but significant reduction in cell viability in response to PI treatment was observed when both caspases were knocked-down or knocked-out compared to control cells (scramble-siRNA treated or parental). Counter-intuitive to traditional caspase roles, these results suggest that CASP3 and 7 promote cell survival in non-lethal PI stress conditions. FIG. 2G shows representative images of crystal violet assay plates of SKBR3 parental and CASP3, 7 DKO cells treated with indicated concentrations of MG132 for 24 hours and continued to grow in drug free media for another 3 days. FIG. 2F shows quantification of the cell viability shown in FIG. 2G. Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells.

[0098] In graphs in FIG. 2, data are shown as mean ± SEM. n = 3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS., not significant. In FIGs. 2A, 2B, 2C and 2D one-way ANOVA with Dunnett's post-test and I with Tukey's post- test. K with two-way ANOVA with Sidak's post-test.Example 3.0 - CASP7 is cleaved non-canonicallv in non-lethal stress conditions, resulting in stable p29 / p30 fragments

[0099] To help elucidate the molecular mechanisms by which CASP3 and 7 positively regulate cyto protective autophagy, the inventors analyzed their expression and processing pattern following amino acid starvation or treatment with proteasome inhibitors (Pls). CASP3 and 7 processing during apoptosis results in ~p20 and ~pl0 / pl5 cleaved-caspase fragments. Intriguingly, western blot analyses revealed prominent and stable non-canonical CASP7 cleavage fragments at ~29 and 30 kDa (p29 / 30) and a CASP3 cleavage fragment at ~27 kDa (p27) in starvation or PI conditions. In fed or untreated (non-stressed) cells, these non-canonical fragments were either absent or less apparent. The CASP3 fragment at ~27 kDa was inconsistent and / or not readily detectable across all experiments.

[0100] To confirm the observed non-canonical processing, the inventors analysed the processing of CASP7 in different contexts. The inventors directly compared CASP7 processing profiles resulting from starvation stress versus apoptosis by treating cells with Earle's starvation buffer (EBSS) or the apoptosis-inducer staurosporine (FIG. 3A). FIG. 3A shows representative western blots comparing CASP7, CASP3 and PARP1 processing patterns in SKBR3 cells following 4 hr incubation in fed conditions (F), non-lethal stress (starvation) or lethal stress (apoptosis induction by staurosporine; STS, 1 or 2 pM). A marked increase in cleaved-PARP and canonical cleaved-CASP3 (p20) were present only in staurosporine treated cells, confirming that the amino acid starvation conditions employed via EBSS treatment were non-lethal. Only the CASP7-p29 / 30 fragment(s) were produced predominantly in these non-lethal starvation conditions, whereas in staurosporine treated cells we primarily observed the canonical apoptosis- associated CASP7-p20 fragment. Taken together, these results confirm that the CASP7 processing in response to non-lethal stress conditions is distinct from CASP7 processing in apoptosis.

[0101] The inventors next asked if other known inducers of autophagy similarly result in non-canonical CASP7 cleavage. Rapamycin inhibits mTOR and induces autophagy (He and Klionsky, 2009; Sotthibundhu et al., 2016). In both SKBR3 and MDA-MB-231 cell lines, autophagy was induced upon treatment with rapamycin. However, rapamycin did not result in non-canonical CASP7 processing, althoughtreatment with rapamycin under starvation conditions did result in non-canonical CASP7 processing, which is believed to be caused by the starvation conditions. Further, CASP3,7 DKO had no effect on rapamycin-induced autophagy. Collectively, these observations suggest that caspases act upstream or in parallel to mTOR in autophagy regulation.

[0102] Next, the inventors investigated the prevalence and the context dependency of non-canonical CASP7 processing. Analysis of tissues from well-fed mice confirmed the existence of in vivo non-canonical processing of CASP7 in several tissues, including brain, heart, spleen and kidney (FIG. 3B). FIG. 3B shows Representative western blot showing CASP7 immunolabeling in tissues from a female CD-I mouse. Ponceau S labeling serves as the loading control, n = 3 mice. Multiple breast cancer patient-derived xenograft (PDX) specimens also clearly showed the existence of CASP7-p29 / 30 fragments (FIG. 3C). FIG. 3C shows a representative western blot showing CASP7 immunolabeling in human breast cancer PDX specimens from indicated subtypes. Starved SKBR3 serves as the positive control for p29 / p30 fragments. Further, a panel of anthracycline-resistant MDA-MB- 231 cell lines were analyzed (FIG. 3D). FIG. 3D shows representative western blot showing CASP7 immunolabeling in parental MDA-MB-231 cells and derivative epirubicin-resistant (R8a, R8b, R3-1, R3-2) and 5Fu-resistant (R6-1, R6-2) cells in fed conditions. The majority of treatment resistant lines had a clear accumulation of CASP7-p29 / 30 fragments. Taken together, these data confirm the existence of stable non-canonical CASP7 fragments in vitro and in vivo, and in normal and cancer tissues.

[0103] Without being bound by theory, the inventors believe that the existence of these stable non-canonical CASP7 fragments can be used as predictive biomarkers for treatment responses and prognostic biomarkers for patient outcome. Specifically, the observed p29 / p30 fragments of CASP7 are associated with non- lethal stress responses such as increased cytoprotective autophagy and signaling for the DNA damage response, and are also associated with resistance of cancer cells to treatment with chemotherapeutic agents such as anthracyclines (e.g. data presented in FIG. 3D). Thus, the presence of p29 / p30 fragments of CASP7 in cancer cells may be a predictive biomarker to indicate cancer cells that may beresistant to stress-inducing anti-cancer therapies such as chemotherapy, because such cells have mounted a cytoprotective stress response that may suppress the effects of anti-cancer agents. Further, since the presence of p29 / p30 fragments of CASP7 in cancer cells may be associated with treatment resistance, the presence of these fragments may also be used as a prognostic biomarker, since the presence of p29 / p30 fragments of CASP7 in patient tumors may be associated with worse survival outcomes for those patients.

[0104] Further, since certain normal tissues may express certain levels of p29 / p30 fragments of CASP7 under normal conditions (e.g. tissues or organs where stress such as oxidative or nutritional stress is naturally high or in metabolically active organs), in some cases an assessment of the presence of p29 / p30 fragments of CASP7 in a patient cancer may involve a comparison of corresponding healthy tissue from that patient, to enable an evaluation of whether the p29 / p30 fragments of CASP7 are being overexpressed in the cancer cells as compared to normal cells in that specific tissue context.Example 4.0 - Location and sequence of non-canonical cleavage sites in CASP7

[0105] Next, the inventors investigated the identity of the unique CASP7-p29 / 30 fragments observed in non-lethal stress conditions. In apoptosis, the first step in CASP7 activation is removal of the pro-domain. To the best of the inventors' knowledge, no stable accumulation of CASP7 lacking the prodomain has been reported. To determine whether the non-canonical CASP7-p29 / 30 fragment(s) could be the result of pro-domain removal, the inventors stably transfected CASP3,7 DKO cells with a full length CASP7 wild-type construct (CASP7-WT), a pro-domain deletion construct (CASP7-APro) or a pro-domain cleavage site-mutant construct (FIG. 4A). FIG. 4A shows a schematic illustration of subunits, canonical cleavage sites and non-canonical cleavage sites in CASP7 wild-type or in various CASP7 fragments used or observed in this study. Scissors below the WT sequence denote the canonical cleavage sites and the order of processing in apoptosis. Scissors above the WT sequence with black arrows denote the putative calpain cleavage sites (non-canonical).

[0106] In western blot analyses the CASP7-p29 / 30 fragments formed in starvation did not align with the CASP7-APro construct which appeared at ~33.5 kDa. Besides, when starved, the CASP7-APro expressing cells also resulted in CASP7-p29 / 30 fragment(s). Moreover, CASP7-p29 / 30 band(s) were still formed when the cell line expressing CASP7 lacking a pro-domain cleavage site was subjected to starvation. Collectively, these data show that the non-lethal cellular stress-associated cleavage of CASP7 occurs at a non-canonical site(s) downstream of the pro-domain cleavage site.

[0107] To elucidate the precise identity of the non-canonical cleavage sites and fragments, the inventors enriched for CASP7-p29 / 30 fragments by performing CASP7-immunoprecipitation (IP) under starvation conditions. Protein in the 29 / 30 kDa region was extracted and subjected to Edman sequencing which revealed the location of a non-lethal stress-associated cleavage site 10 amino acids downstream of the pro-domain cleavage site of CASP7 (FIGs. 4A and 4B). Sequences on the left of FIG. 4B show the putative calpain cleavage sites in CASP7 WT and the amino acid replacements made in the CASP7 construct with putative calpain-cleavage sites mutated (CASP7-CCM).

[0108] This non-canonical cleavage site and a second site nearby (FIGs. 4A and 4B), corresponding to p30 and p29, respectively, were shown previously to be cleaved in vitro by calpain 1 or 2, but were associated with further processing of CASP7 to yield pl7 and pl8 (Gafni et al., 2009). Using the DeepCalpain algorithm (Liu et al., 2019), the inventors identified predicted calpain cleavage sites conserved in murine CASP7 and confirmed that they corresponded to the size of the CASP7-p30 / p29 bands detected in murine samples. To further verify the cleavage sites, the inventors created a human CASP7 construct containing alanine substitution of residues at the two calpain cleavage sites (CASP7-calpain cleavage mutant, or CASP7-CCM; FIG. 4B). Upon starvation, CASP3,7 DKO SKBR.3 cells stably transfected with CASP7-CCM failed to form abundant CASP7-p29 / p30 as compared to CASP7-WT cells (FIG. 4B). Together, these results validate the non- canonical cleavage of CASP7 and suggest a potential involvement of calpains in the regulation of CASP7 to promote starvation or proteasome-inhibitor (PI)- induced autophagy and / or cell survival in non-lethal stress conditions.Example 5.0 - PARP1 cleavage is increased and PARylation and ATG gene expression are reduced in CASP3.7 DKO cells

[0109] To characterize the underlying molecular mechanisms by which non- canonical processing of CASP7 may contribute to autophagy regulation in stress conditions, the inventors examined the amino acid sequence surrounding the CASP7 calpain cleavage sites. The well-known Poly(ADP ribose) polymerase 1 (PARP1) binding site in CASP7 is located between the two identified calpain cleavage sites. PARP1 and associated PARylation activities were shown to play key roles in stress sensing, stress adaptation and autophagy induction in multiple cell lines and in-vivo mouse models. Further, PARP1 is a substrate of both CASP3 and 7, with the latter showing a higher affinity for PARP1 binding. These factors prompted the inventors to investigate PARP1 levels and PARP1 activity in different CASP backgrounds. Unexpectedly, a higher level of cleaved-PARPl was detected in CASP3,7 DKO cells compared to the parental cells (FIGs. 5A-5B). FIGs. 5A-5B show the quantification of cleaved-PARPl as determined by western blots of indicated proteins from parental and CASP3,7 DKO SKBR3 cells cultured in fed conditions or starved for 8h (FIG. 5A) or treated with MG132 (0.5 pM) for 24 h (FIG. 5B). NT= No Treatment (vehicle DMSO). There was no evidence of cell death in the DKO cells, consistent with previous observations that low levels of cleaved-PARPl (89 kDa) are not always associated with cell death. The cleaved-PARPl level was further increased when cells were stressed with a non-lethal dosage of MG132 (FIG. 5B).

[0110] Next, PAR immunolabeling was performed as a readout for PARylation activity of PARP1 (Garcia-Saura et al., 2021). A significant reduction in PAR labeling was observed in CASP3,7 DKO cells compared to parental cells in both SKBR3 and MDA-MB-231 cell lines (FIGs. 5C-5D). FIGs. 5C-5D show quantification of PAR levels determined by western blots of PAR immunolabeling from parental and CASP3,7 DKO SKBR3 (FIG. 5C) or MDA-MB-231 (FIG. 5D) cells treated with vehicle (DMSO) or MG132 (0.5 pM) for 24 h.

[0111] Although unexpected, these results indicate that in the absence of CASP3 and 7, PARP1 cleavage is enhanced and PARP1 activity (as measured by PAR level) is impaired.

[0112] In cellular stress conditions, PARP1 and PARylation were shown to upregulate cytoprotective autophagy through multiple mechanisms, including thetranscriptional upregulation of autophagy pathway components. Thus, the inventors predicted CASP3,7 DKO cells would result in transcriptional downregulation of autophagy pathway components. Among the components evaluated by RT-qPCR, the inventors found that LC3B and ATG7 transcript levels were reduced in CASP3,7 DKO cells. There was an increase in both transcripts in response to MG132-induced stress (FIG. 5E), with DKO cells showing weaker upregulation compared to parental (Fig. 5E). Re-expression of CASP3,7-WT constructs could partially rescue transcript levels of these genes in DKO cells (FIG. 5E). FIG. 5E shows RT-qPCR analyses of LC3B in SKBR3 parental, CASP3,7 DKO or CASP3,7-WT re-expression in CASP3,7 DKO SKBR3 cells. Cells were treated with vehicle (DMSO) or 0.5pM of MG132 for 24 hours.

[0113] Without being bound by theory, these results are consistent with the possibility that CASP3 and 7 regulate transcription of key autophagy genes, potentially through changes to PARP1 and PARylation. The inventors have conducted further experiments with a non-cleavable PARP1 mutant to confirm without being bound that the mechanism of reduced PARylation in the absence of both CASP3 and CASP7 is related to the increase in PARP1 cleavage observed in this context.

[0114] In graphs in FIG. 5, data are shown as mean ± SEM. n = 3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS., not significant. In FIGs. 5A, 5B, 5C and 5D, one-way ANOVA with Tukey's post-test.Example 6.0 - CASP7-p29 / 30 promote phosphorylation of DDR pathway protein H2AX

[0115] In addition to transcriptional remodeling, PARP1 plays a central role in DNA repair and DNA Damage Response (DDR) pathways in response to cell stress. Both starvation and proteasome inhibition were reported to induce DNA damage and activate autophagy as a part of stress-induced DDR. Western blot analyses by the inventors revealed that MG132 stress-induced phospho-H2AX (y-H2AX) accumulation was significantly reduced in CASP3,7 DKO cells compared to parental cells without any apparent reduction in total H2AX levels (FIG. 6A). FIG. 6A shows quantification of y-H2AX / total H2AX as determined by western blots of proteinsfrom SKBR.3 parental and CASP3,7 DKO cells treated with vehicle DMSO or MG132 (0.5 pM) for 24 hours. Ponceau S staining was used as the loading control.

[0116] In the past, the level of yH2AX was measured to determine the extent of DNA damage. However, more recently, yH2AX levels were shown to represent a readout for the strength of DDR. signaling activity rather than the level of DNA damage. To distinguish between these two possibilities, the inventors used comet assays to directly measure DNA damage (Olive and Banath, 2006). The level of DNA damage / fragmentation in parental versus CASP3,7 DKO cells was not significantly different when treated with the vehicle (DMSO) control (FIG. 6B). FIG. 6B shows quantification of tail moments of comets determined by alkaline comet assay from SKBR3 parental and CASP3,7 DKO cells treated with vehicle DMSO or MG132 (0.5 pM) for 24 h. Cells treated with H2O2 (100 pM for 4h) serve as the positive control. Tukey's box-and whisker plots are based on tail moments determined by CometScore from 200 cells in each of 2 independent assays.

[0117] In response to non-lethal MG132 stress conditions, there was an increase in DNA damage, but it was comparable in both parental and CASP3,7 DKO cells. In contrast, parental, but not CASP3,7 DKO, cells showed an increase in DNA damage in the positive control, H2O2, lethal stress conditions (FIG. 6B). This data indicates without being bound by theory that the reduction of yH2AX observed in CASP3,7 DKO cells in the MG132 non-lethal stress conditions is not due to a reduced level of DNA damage but instead could be due to impaired DDR signaling.

[0118] Next, the inventors investigated the y-H2AX levels in CASP3,7 double and single KO cells (FIG. 6C). FIG. 6C shows quantification of y-H 2 AX / total H2AX determined by western blots of proteins from SKBR3 parental, CASP3,7 DKO or single KO cells treated with vehicle DMSO or MG132 (0.5 pM) for 24 h. Ponceau S staining was used as the loading control.

[0119] Notably, the highest level of y-H2AX was observed in CASP3 KO cells, suggesting that CASP7 plays a prominent role in the H2AX phosphorylation. To further investigate this possibility, and given the proximity of the PARP1 binding site to the CASP7 calpain cleavage sites, the inventors generated CASP3,7 DKO cell lines stably expressing CASP7-CCM, CASP7-p30 or CASP7-p29. In this case, CASP7- CCM, which does not undergo non-canonical cleavage, serves as the full-lengthCASP7 control. When y-H2AX levels were measured in the absence and presence of non-lethal MG132 (FIG. 6D), the inventors observed that the stress-induced y- H2AX accumulation was increased in CASP7-p30 and CASP7-p29 expressing cells compared to CASP7-CCM expressing cells (FIG. 6D). FIG. 6D shows quantification of y-H2AX / total H2AX as determined by western blots of proteins from CASP3,7 DKO SKBR.3 cells stably transfected with indicated CASP7 constructs treated with vehicle DMSO or MG132 (0.5 pM) for 24 h. H3 was used as a loading control.

[0120] Among the three constructs, the greatest level of y-H2AX was observed with CASP7-p29. Comet assays revealed that the degree of DNA damage was not significantly different among the three cell lines in either untreated vehicle (DMSO) control, MG132 treated (non-lethal) or H2O2 treated (lethal) conditions. In fact, the degree of DNA damage appeared slightly reduced in CASP7-p29 expressing cells where the inventors detected the greatest level of y-H2AX (FIG. 6D). Crystal violet cell viability assays following treatment with increasing concentrations of MG132 or Bortezomib, showed no significant differences in cell viability despite the different Y-H2AX levels observed in the three cell lines. Overall, these results support without being bound that non-canonical processing of CASP7 does not lead to different degrees of DNA damage, but rather plays a role in the induction of different levels of y-H2AX signaling.

[0121] To further explore CASP7 in this role, the inventors employed catalytically active wildtype (CASP7-WT) and catalytically inactive (CASP7-inactive, C>A) constructs stably expressed in CASP3,7 DKO cells (FIG. 6E). Expression of the CASP7-WT construct but not the catalytically inactive construct led to a significant increase in the y-H2AX signaling in non-lethal MG132 treated cells, revealing that CASP7 and its catalytic activity are required for the phosphorylation of H2AX or y- H2AX signaling. FIG. 6N shows quantification of y-H2AX / total H2AX determined by western blots of proteins from CASP3,7 DKO SKBR.3 cells re-expressing vector, CASP7-WT (catalytically active) or CASP7-inactive (catalytically inactive) constructs, treated with vehicle DMSO or MG132 (0.5 pM) for 24 h. H3 serves as the loading control.

[0122] In graphs in FIG. 6, data are shown as mean ± SEM. n = 3 independent experiments except for comet assays n = 2. *P<0.05, **P<0.01, ***P<0.001,****p<0.0001, NS., not significant. In FIGs. 6A, 6C, and 6E one-way ANOVA with Dunnett's post-test. In FIGs. 6B and 6D one-way ANOVA with Tukey's post- test. In J two-way ANOVA with Sidak's post-test.Example 7.0 - Inhibition of CASP 3 and 7 mimic effects of PARP1 inhibitors and is synthetic lethal with BRCA1 and other genes involved in DNA processing or repair

[0123] The foregoing data show that inhibition of CASP3 and 7 leads to increased PARP1 cleavage, reduced PARP activity (PARylation), reduced DDR signaling and a suppressed autophagy response. Since these effects phenocopy the genetic or pharmacological inactivation of PARP1, the inventors hypothesized that CASP3,7 inhibition would also phenocopy the synthetic lethal effects of PARP1 in cells that are defective in homologous recombination (Lim et al., 2017; D'Andrea, 2018). To demonstrate this, the inventors knocked down CASP3 and 7 in BRCA1- deficient SUM149PT breast cancer cells and assessed viability by the crystal violet assay. In striking contrast to BRCAl / 2-proficient SKBR3, MDA-MB231 or JIMT1 breast cancer cells, the inhibition of CASP3,7 alone was lethal in SUM149PT cells (FIG. 7A). FIG. 7A shows quantification of cell viability as determined by crystal violet assays for SUM149PT, MDA-MB-231, SKBR3 or JIMT-1 cells transfected with scramble siRNA or CASP3 and 7 siRNAs (lOnM) for 2 days and then cultured in siRNA-free media for another 3 days. Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells. Data show from left to right: scramble siRNA SUM149PT, CASP3+7 siRNA SUM149PT, scramble siRNA MDA-MB-231, CASP3+7 siRNA MDA-MB-231, scramble siRNA SKBR3, CASP3+7 siRNA SKBR3, scramble siRNA JIMT-1 and CASP3+7 siRNA JIMT-1.

[0124] In western blot analyses with all cell lines, a distinct band of cleaved- PARP1 was visible in CASP3,7 knock-down cells compared to the scramble-siRNA control (FIG. 7B). FIG. 7B shows representative western blots showing CASP 3 and 7 levels following siRNA treatment conditions used in crystal violet assays described for FIG. 7A. Immunolabelling of cleaved-PARPl is also shown. n=2 independent experiments.

[0125] Western blot analyses also confirmed that autophagic flux was reduced in the CASP3,7 knock-down SUM149PT cells compared to the scramble-siRNA controlor compared to the knock-down of either CASP3 or CASP7 alone (FIG. 7C). FIG. 7C shows quantification of LC3B-based autophagy flux in proteasome inhibitor (MG132) treated cells determined by western blots of proteins from SUM149PT cells transfected with scramble, CASP3 and / or CASP7 siRNAs (48h) and treated with vehicle DMSO or MG132 (0.5 pM) for 24h, with BafAl (50 nM) in the final 2h. The levels of LC3BII in MG132 treated cells were normalized to loading control and shown relative to the MG132 treated scramble-siRNA control.

[0126] To determine whether CASP3,7 knockdown was redundant or potentially additive with PARP1 inhibition, the inventors treated the SUM149PT cells with the PARP1 inhibitor Olaparib. The combination of CASP3,7 knockdown and olaparib treatment resulted in a further reduction in SUM149PT cell viability compared to Olaparib treatment alone, demonstrating that CASP3 and CASP7 augments PARP inhibitor therapy, rather than being merely duplicative therewith (FIG. 7D). FIG. 7D shows the percentage of stained (viable) cells at each concentration normalized to scramble siRNA treated and Olaparib untreated cells as determined by crystal violet assay. Indicated siRNA transfected cells (scramble, left, and CASP3+7 siRNA right) treated with indicated concentrations of Olaparib for 24h and continued to be cultured in drug free media for another 3 days. These results further confirm that combined inhibition of CASP3 and CASP7 induces synthetic lethality with loss of BRCA1, just as does inhibition of PARP1, and further may indicate that the mode of action of olaparib on PARP is different from the mode of action of CASP3 / CASP7 knockdown on PARP.

[0127] To further assess genetic interactions with CASP3,7, the inventors utilized Genetic inteRaction and EssenTiality neTwork mApper (GRETTA) (Takemon and Marra, 2023) to perform in siiico genetic interaction screens using publicly available pan-cancer cell line data from DepMap (Meyers et al., 2017; Ghandi et al., 2019; Nusinow et al., 2020). Using proteome data from DepMap, the inventors identified cancer cell lines harboring low protein expression of both CASP3 and CASP7 and cancer cell lines with high expression of both CASP 3 and 7 (FIG. 7E and Table 4). FIG. 7E shows a dot plot showing DepMap cancer cell lines with bottom 10% of CASP3, 7 protein expression (n = 16) and top 10% casp3,7 protein expression (n = 10). Normalized CASP3 / 7 protein expression (scaled and mean centered) wereextracted from Nusinow et al. using GRETTA. FIG. 7F shows ranked CASP3,7 genetic interaction scores generated using GRETTA. The cell lines at the bottom 10% of CASP3 & CASP7 protein expression are compared against cells lines at the top 10%. Rank is based on lethal GI scores. The Red points indicate true candidate CASP3,7 lethal Genetic interactions (GIs). The top most lethal genetic interactors are labeled. FIG. 7G is a heatmap showing Jaccard index similarities between GO terms. Based on annotated GO biological processes of genes that were candidate co-essential or genetic interactors (unadjusted p-value < 0.05). GO terms were summarized into 16 distinct clusters as listed in Table 5 using Jaccard index-based hierarchical clustering.

[0128] Comparisons between these two groups resulted in the identification of 142 candidate CASP3,7 synthetic lethal interactors, where perturbation of these genes resulted in significantly higher lethal probabilities in the CASP3,7 low expressor lines compared to the CASP3,7 high expressor lines (FIG. 7G and Table 5). The candidate synthetic lethal interactors included HR pathway genes (FIG. 7G and Table 5), with BRCA1 as the top synthetic lethal hit (FIG. 7F). Collectively, these results show that the combined inhibition of CASP3 and 7 mimics the cellular effects of PARP1 inhibitors, revealing a surprising new avenue for potential therapeutic exploitation. Counter-intuitive to traditional caspase roles, these findings demonstrate that combined inhibition of CASP3 and 7 induce synthetic lethality in BRCAl-defective cells. Further, because CASP3 and CASP7 are at least partially functionally redundant, traditional synthetic lethal screens which examine only pairwise interactions between genes would likely fail to identify this mechanism of synthetic lethality.Table 4. Summary of low and high levels of CASP3 and 7 expressing cancer cell lines identified and analyzed in FIGs. 7E and 7F.Table 5. Summary of clusters (GO BP terms) in FIG. 7G.

[0129] The full list of candidate synthetic lethal interactors identified by this screen are: BRCA1 (and it is expected that other BRCA genes including BRCA2 would behave similarly, and both BRCA1 and BRCA2 function in homologous recombination), SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4,CDC26, N0L8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, N0C4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18.

[0130] In graphs in FIG. 7, unless otherwise noted, data are shown as mean ± SEM. n = 3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****p<0.0001, NS., not significant. In FIGs. 7A and 7C one-way ANOVA with Sidak and Dunnett's post-test, respectively. In FIG. 7D, two-way ANOVA with Sidak's post-test. In I, Mann-Whitney U-test p-value< 0.05. See also FIG. 7G.Example 8.0 - Pharmacological inhibition of CASP3 and CASP7 enhances lethality in the BRCA1 deficient SUM149PT breast cancer cell line

[0131] Due to relatively higher specificity reported in cell-free and cell line-based studies, the inventors chose to investigate the effectiveness of pharmacological inhibition of CASP3 and CASP7 using caspase 3 / 7 inhibitor 1 (CASP 3 / 7 inhibitor 1) (Lee et al., 2000) in the cancer contexts they identified as being likely to benefit from such inhibition of CASP3 and CASP7 activity.

[0132] The inventors' results above establish that genetic knockdown or knockout of CASP 3+7 was lethal in BRCA1 deficient SUM149PT cells but not in BRCAl / 2-proficient SKBR3, MDA-MB-231 or JIMT1 cell lines. To confirm that the pharmacological inhibition of CASP 3+7 was consistent with genetic inhibition, SUM149PT, SKBR3 and MDA-MB-231 cells were treated with the small molecule inhibitor CASP 3 / 7 inhibitor 1 and subjected to crystal violet cell viability assays. Results are shown in FIGs. 8A-1B. FIG. 8A shows the quantification of cell viability determined by crystal violet assay of (from left to right) SKBR3, SUM149PT and MDA-MB-231 cells treated with indicated concentrations of caspase 3 / 7 inhibitor 1 for 48h and then grown in drug free media for another 3 days. Percentage ofstained (viable) cells at each concentration was normalized to untreated cells of the respective cell line. FIG. 8B shows quantification of phase confluence values generated with the Incucyte system over 72h following treatment of indicated cells with caspase 3 / 7 inhibitor 1. Absolute phase confluence values obtained from the Incucyte system were used to calculate the Area Under Curve (AUC) at each concentration. IC50 values are based on normalised AUC of each cell line at each concentration. For IC50, curve [Inhibitor] vs response -variable slope was used. In graphs, data are shown as mean ± SEM. n = 3 independent experiments except n = 2 with MDA 231 cell line. *P<0.05, **P<0.01; In FIG. 8A, two-way ANOVA with Tu key's post- test was used.

[0133] Unlike genetic inhibition, a reduction in cell viability was observed in all three cell lines. However, the viability effect was more pronounced in BRCA1 deficient SUM149PT cells compared to SKBR3 or MDA-MB-231 cells, at least at some concentrations used (FIG. 8A). Further, the difference in cell viability between the BRCA-proficient SKBR.3 and BRCA-deficient SUM149PT cell lines was observed to be greater when viability was determined using cell confluency measures with the more sensitive Incucyte system (FIG. 8B). Similar to genetic inhibition of CASP3+7, these results support that pharmacological inhibition of CASP 3+7 using CASP 3 / 7 inhibitor 1 induces synthetic lethality with loss of BRCA1, supporting the role of such therapeutic agent in the treatment of cancers deficient in BRCA genes including BRCA1 or BRCA2.Example 9.0 - Pharmacological inhibition of CASP3 and CASP7 sensitizes cells to proteasome inhibitor MG132

[0134] The inventors' results above demonstrate that proteasome inhibition- induced compensatory autophagy was CASP3 and CASP7 dependent. Further, a modest but significant reduction in cell viability was observed in response to proteasome inhibitor (PI) treatment in CASP 3+7 double knockdown or double knockout SKBR3 or MDA-MB-231 cells compared to corresponding parental cells. Next, to test whether CASP 3 / 7 inhibitor 1 could also sensitize cells to PI treatment, crystal violet cell viability assays were carried out following PI and CASP 3 / 7 inhibitor 1 treatment alone or in combination. Results are shown in FIGs. 9A-9B. FIG. 9A shows quantification of cell viability determined by crystal violet assays ofSKBR3 cells treated with the indicated concentrations of caspase 3 / 7 inhibitor 1 and / or MG132 for 48h and then grown in drug free media for another 3 days. Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells. FIG. 9B shows quantification of cell viability determined by crystal violet assays of SUM149PT parental cells treated with indicated concentrations of caspase 3 / 7 inhibitor 1 and / or MG132 for 48h and then grown in drug free media for another 3 days. Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells. In graphs, data are shown as mean ± SEM. n = 3 independent experiments except n = l in SUM149PT cell line (panel C). *P<0.05, **P<0.01, ***P<0.001. In FIG. 9A, two-way ANOVA with Tu key's post- test was used.

[0135] Although the effect was modest, combined treatment resulted in a significant reduction in cell viability at some concentrations similar to genetic inhibition of CASP 3+7 (FIG. 9A). The drug combination effect was also tested in BRCA1 deficient SUM149PT cells with similar results observed (FIG. 9B).Example 10.0 - Pharmacological or genetic inhibition of CASP3 and CASP7 have similar effects on PARP1 cleavage

[0136] The increased PARP1 cleavage and the decreased PARylation (PARP activity) in CASP 3+7 double knockout cells were unexpected findings of the inventors' studies. These observations suggest that CASP3 and CASP7 indirectly or directly contribute to PARP1 stability and / or function under normal physiological conditions. The inventors investigated whether CASP 3 / 7 inhibitor 1 also has an effect on PARP1 cleavage. Results are shown in FIG. 10. FIG. 10 shows quantification of cleaved-PARP in caspase 3 / 7 inhibitor 1 treated cells relative to untreated control cells (NT) as determined by western blots of proteins from SKBR3 (left data set) and SUM149 (right data set) parental cells treated with increasing concentration of caspase 3 / 7 inhibitor 1 for 48h. Data are shown as mean ± SEM. n = 3 independent experiments. *P<0.05, **P<0.01, ***P<0.001. In B, two-way ANOVA with Dunnett's post- test.

[0137] An increase in PARP cleavage was detected when SKBR3 or SUM149PT cells were treated with increasing concentration of CASP 3 / 7 inhibitor 1 (FIG. 10). A dose-dependent increase in PARP1 cleavage was evident in the BRCAl-deficientSUM149PT cells, showing the enhanced effects of CASP 3 / 7 inhibition in this genetic context, similar to the enhanced effects observed with genetic inhibition of CASP3 / 7 as described above.Example 11.0 - Genetic or pharmacological inhibition of CASP3 and CASP7 sensitizes cells to DNA damage-inducing anticancer drug, cisplatin

[0138] Based on the wet lab experiments and in-silico data mining, the inventors predicted that combining CASP3+7 inhibition with chemotherapeutic drugs that induce DNA damage (ie., cisplatin, doxorubicin) will increase the sensitivity of cells to the DNA damaging drug. Results are shown in FIGs. 11A-11B. FIG. 11A shows quantification of cell viability determined by crystal violet assays of SKBR3 parental (left bars) and CASP3+7 double knockout cells (right bars) treated with indicated concentrations of cisplatin for 48h and then grown in drug free media for another 3 days. Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells. FIG. 11B shows quantification of cell viability determined by crystal violet assays of SKBR3 parental cells treated with indicated concentrations of caspase 3 / 7 inhibitor 1 and / or cisplatin for 48h and then grown in drug free media for another 3 days (bars from left to right = cisplatin only, CASP3 / 7 inhibitor 1 only, CASP3 / 7 inhibitor + cisplatin). Percentage of stained (viable) cells at each concentration was normalized to respective untreated cells. In graph, data are shown as mean ± SEM. n = 3 or 4 independent experiments. *P<0.05, **P<0.01, ***P<0.001, In FIGs. 11A and 11B, two-way ANOVA with Tukey's posttest was used.

[0139] When tested with crystal violet cell viability assays, a modest but significant reduction in cell viability was detected in response to cisplatin treatment in CASP3+7 double knockout (DKO) SKBR3 cells compared to parental cells (FIG. 11A). Next, the inventors tested whether CASP 3 / 7 inhibitor 1 could also sensitize SKBR3 cells to cisplatin by treating cells with CASP 3 / 7 inhibitor 1 and cisplatin alone or in combination. Unlike with genetic inhibition of CASP3+7, the change in cell viability was not statistically significant with this drug combination, although a slight reduction in cell viability in cells treated with both CASP 3 / 7 inhibitor and cisplatin was still observed. (FIG. 11B).Example 12.0 - Discussion

[0140] Caspases remain categorized primarily as components of the cell death machinery despite studies demonstrating that they have multiple non-apoptotic functions. However, the results described herein shifts the current paradigm of apoptotic caspase- PAR.P1 interactions to one that involves non-canonical caspsase cleavage and the promotion of cell adaptation and survival pathways at the onset of cellular stress or in non-lethal cellular stress conditions. This allows for the inhibition of caspase activity to be used to cause cell death under such non-lethal cellular stress conditions.

[0141] More specifically, the results described herein reveal a novel mechanism of stress adaptation used by cancer cells revealing new vulnerabilities that can potentially be exploited as therapeutic targets or biomarkers. Counter-intuitively, these data support the idea that caspase inhibition, rather than caspase induction, may reduce cell fitness in certain cells and stress contexts. Strikingly, the synthetic lethal phenotype observed in BRCA1 deficient cells points to a context where CASP3,7 inhibition may be most effective. Based on the in siiico findings described herein, the inventors expect that this synthetic lethal relationship will be relevant to multiple cancer cell types. Currently, PARP1 inhibitors are used clinically for the treatment of BRCA-mutated breast, ovarian, pancreatic and prostate cancers (Mateo et al., 2019; Rose et al., 2020). CASP3,7 inhibition could be further explored as a potential alternative or second line treatment in cases of intrinsic or acquired resistance to PARP1 inhibitors. In addition, combining CASP3,7 inhibition with chemotherapeutic drugs that induce DNA damage (ie., cisplatin, doxorubicin) can increase the drug sensitivity. These results also show that CASP3,7 inhibition can be used as an alternate way to inhibit the autophagy pathway which may be beneficial due to clinical challenges associated with existing autophagy inhibitors like hydroxychloroquine, which lack sufficient efficacy due to dose-limiting toxicity or intrinsic treatment resistance.

[0142] An evolutionarily conserved role for caspases in stress adaptation was postulated previously. In support of such a role for human CASP3 and CASP7, the inventors have identified a molecular mechanism that involves modulation of PARP activity, phosphorylation of H2AX for DNA damage signaling, and autophagy. Thenon-lethal stress adaptation role for caspases identified herein also provides a plausible mechanistic explanation for previous reports of associations between increased caspase expression and worse cancer patient outcomes. The inventors propose without being bound by theory this increased caspase expression is used to support tumor stress adaptation, through enhanced PAR.P1 activity, autophagy and DDR. signaling, and may be a targetable vulnerability, particularly in the context of BRCA1 or other HR pathway mutations.

[0143] Inadequate understanding of the functions and underlying molecular mechanisms of caspases have previously contributed to failures in the development of caspase-based therapeutic strategies. However, the studies described herein reveal unexpected new knowledge related to caspase regulation and function, identifying new therapeutic avenues to explore with caspase inhibitors, including CASP 3 / 7 inhibitor 1. The inventors have discovered an autophagy-promoting and cell survival role for the human effector caspases, CASP3 and 7, in human breast cancer cell lines in starvation or proteasome inhibition (Pl)-induced stress conditions. The inventors found that CASP7 is differentially cleaved during non- lethal cellular stress, resulting in stable cleavage fragments (CASP7-p29 / 30) that are distinct from canonical cleaved-CASP7 fragments (CASP7-p20 / pl0) reported in apoptosis. Unexpectedly, the inventors found that PARP activity and DNA damage- induced phosphorylation of H2AX are significantly reduced in CASP3, 7 DKO cells. Strikingly, BRCAl-deficient cells that are known to be highly dependent on PARP activity for DNA repair, exhibited synthetic lethality with combined knockdown of CASP3,7. The results of these examples further support pharmacological inhibition of CASP3+7 as an anticancer approach and support use of the small molecule inhibitor CASP 3 / 7 inhibitor 1 or other inhibitors of CASP 3 and CASP 7 as a candidate for further development as an anticancer drug.

[0144] The results of these examples are in line with the recent work by Conde- Rubio et al., (2021) that compared the proteomic profiles of parental versus CASP3,7 DKO in U2OS osteosarcoma and HCT116 colon carcinoma cells under non- lethal stress or lethal stress conditions. Conde-Rubio et al., (2021) discovered that 92 proteins are cleaved by CASP3 and / or 7 in non-lethal stress conditions compared to hundreds of proteins cleaved in lethal stress conditions. All the proteins found tobe cleaved in non-lethal stress are also cleaved in lethal stress, with the exception of Cullin-5 and CLIP-associating proteinl. Interestingly, 12 proteins, including PARP1, were found to be differentially cleaved by CASP3 and / or 7 in cells exposed to non-lethal stress compared to cells exposed to lethal stress. In lethal stress, PARP1 is cleaved at one site producing two cleavage fragments, and in non-lethal stress PARP1 is cleaved at two sites producing three cleavage fragments (Conde- Rubio et al., 2021). Together, these findings support a role for CASP3 and 7 in stress adaptation by altering the protein cleavage and proteome landscape.References

[0145] The following references are of interest with respect to the subject matter described herein. Each of the following references is incorporated by reference herein in its entirety.• Bhagwat SV, et al. Mol Cancer Ther. 2020 Feb;19(2):325-336.• Bryant KL et al. Nat Med. 2019 Apr;25(4):628-640.• Concordet, J.P., and Haeussler, M. (2018). Nucleic Acids Res. 46, W242- W245.• Conde-Rubio, M. del C. et al. (2021). Cell Death Discov. 7, 1-11.• D'Andrea, A.D. (2018). DNA Repair (Amst). 71, 172-176.• Denault, J.B., and Salvesen, G.S. (2003). J. Biol. Chem. 278, 34042-34050.• Dhani, S., Zhao, Y., and Zhivotovsky, B. (2021). Cell Death Dis. 12, 1-13.• Eirew, P. et al. (2022). Nat. Commun. 13, 1-19.• Garcia-Saura, A.G. et al. (2021). Commun. Biol. 4, 1-8.• Ghandi, M., et al. (2019). Nature 569, 503-508.• Hai Nguyen et al, (2023). Cancer Res 1 April 2023; 83 (7_Supplement): 6183.• Iwashita, H., et al. (2018). FEBS Lett. 592, 559-567.• Kim, K.W., et al. 2008. PLoS One. 3.• Kinsey CG et al. Nat Med. 2019 Apr;25(4):620-627.• Lee, D., S.A. et al. 2000. J. Biol. Chem. 275: 16007-16014.• Lee MJ et al. Biomol Ther (Seoul). 2023 Jan 1;31(1): 1-15.• Li et al. (2023). Genomics, Proteomics, Bioinformatics 21(5):962-975)• Lim, E. et al. (2017). Mol. Cell. Oncol. 4, 1-2.Lu, Y. et al. (2017). J. Vis. Exp. 2017, 2-7.Mateo, J. et al. (2019). Ann. Oncol. 30, 1437-1447.Meyers, R.M. et al. (2017). Nat. Genet. 49, 1779-1784.Mirza MR et al. N Engl J Med 2016;375(22): 2154-2164.Morris, E. J. et al. Cancer Discov. 3, 742-750 (2013).Moschos, S. J. et al. JCI Insight 3, 92352 (2018).Nusinow, D.P., et al. (2020). Cell 180, 387-402. el6.Olive, P.L., and Banath, J.P. (2006). Nat. Protoc. 1, 23-29.Pan-Canadian Oncology Drug Review (pCODR) Expert Review Committee (pERC). Final recommendation for niraparib (Zejula®) for ovarian cancer. 03 September 2020.Poreba, M., et al. 2019. Cell Death Differ. 26: 2695-2709.Rose, M. et al. (2020). Front. Cell Dev. Biol. 8, 1-22.Slade D. Genes Dev. 2020; 34(5-6): 360-394.Takemon, Y, and Marra, M.A. (2023). Bioinformatics 39.Wood, L.W., and E.J. Shillitoe. 2011. Cancer Gene Ther. 18:685-694.Ye, J., et al. (2012). BMC Bioinformatics 13.Yoshimori, A., et al. 2004. BMC Pharmacol. 4: 1-17.Zhang B, et al. (2021). Nat Commun. 2021. 12(1): 1275.

Claims

WHAT IS CLAIMED IS:

1. A method of killing a cell comprising: inhibiting CASP3 in the cell; and inhibiting CASP7 in the cell, wherein the cell does not express functional BRCA1, BRCA2, SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN 1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, P0LE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18; or wherein the cell does not express functional proteins involved in homologous recombination, DNA damage response pathways, splicing, or mitotic DNA damage checkpoint signaling; or wherein the cell does not express functional proteins involved in mitotic processes and spindle assembly, mitotic cytokinesis, chromosome localization, regulation of spindle checkpoint, regulation of chromosome separation, RNA transport, immune response regulation, protein localization, RNA splicing, cell cycle regulation, heterochromatin organization, G2 / M phase transition, regulation ofhistone acetylation, regulation of histone methylation, homologous recombination and DNA damage response, or mitotic DNA damage checkpoint signaling.

2. A method of killing a cell comprising: inhibiting CASP3 in the cell; inhibiting CASP7 in the cell; and exposing the cell to a proteasome inhibitor, a DNA damaging agent, an autophagy-inducing agent, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor.

3. A method as defined in any one of claims 1-2, wherein the step of inhibiting CASP3 comprises exposing the cell to an inhibitor of CASP3, and wherein the step of inhibiting CASP7 comprises exposing the cell to an inhibitor of CASP7.

4. A method as defined in any one of claims 1-3, wherein the cell is a cancer cell.

5. A method as defined in claim 4, wherein the cancer cell is a breast cancer cell, or wherein the cancer cell is a bladder cancer, lung cancer, pancreatic cancer, liver cancer, brain cancer, esophageal cancer, bone cancer, endometrial cancer, ovarian cancer, prostate cancer, leukaemia, cervical cancer, endometrial cancer, glioma, sarcoma or advanced, refractory or metastatic solid tumour cell.

6. A method of treating cancer comprising administering to a subject: an inhibitor of CASP3; and an inhibitor of CASP7; wherein the cancer is deficient in or has a loss-of-function mutation in one or more of BRCA1, BRCA2, SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1,PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIAO3, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, or RPS18; or wherein the cancer does not express functional proteins involved in homologous recombination, DNA damage response pathways, splicing, or mitotic DNA damage checkpoint signaling or wherein the cancer is deficient in or has a loss-of-function mutation in one or more genes involved in mitotic processes and spindle assembly, mitotic cytokinesis, chromosome localization, regulation of spindle checkpoint, regulation of chromosome separation, RNA transport, immune response regulation, protein localization, RNA splicing, cell cycle regulation, heterochromatin organization, G2 / M phase transition, regulation of histone acetylation, regulation of histone methylation, homologous recombination and DNA damage response, or mitotic DNA damage checkpoint signaling.

7. A method of treating cancer comprising administering to a subject: an inhibitor of CASP3; an inhibitor of CASP7; and exposing the cell to a proteasome inhibitor, a DNA damaging agent, an autophagy-inducing agent, a PARP inhibitor, an EGFR inhibitor, an ERK inhibitor, or an MEK inhibitor.

8. A method of treating cancer as defined in any one of claims 6-7, wherein the cancer is breast cancer, or wherein the cancer is bladder cancer, lung cancer, pancreatic cancer, liver cancer, brain cancer, esophageal cancer, bone cancer, endometrial cancer, ovarian cancer, prostate cancer, leukaemia, cervical cancer, endometrial cancer, glioma, sarcoma or advanced, refractory or metastatic solid tumour.

9. A method of determining whether a cancer of a patient is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7, the method comprising : determining if the cancer has a loss-of-function mutation in or does not express a functional protein corresponding to one or more of the following genes: BRCA1, BRCA2, SNRPB2, INTS6, CCDC174, UBA6, C3orf38, YPEL5, UBE2H, CCNL1, TLN1, SLC25A38, RNF168, CHMP7, NUP153, WDR26, ARGFX, IGF1R, BAP1, BARD1, EXOC8, NAA35, VHL, KIF18A, RAC1, NUP98, CRKL, ANLN, FAAP20, SARNP, GLE1, CKS2, TADA2A, CDK2, HAPLN2, PYURF, SUGT1, USP37, DNAJA1, POLR1H, CCNB1, PPP1R2, NSMCE3, CCDC59, USP10, RBBP8, PPP1R12A, TGS1, CHMP3, UBR4, CDC26, NOL8, KANSL3, DHX38, SF3B2, RPAIN, CENPE, BCAS2, EIF3M, DCTN3, TTI2, HAUS4, BMS1, TBCC, UTP11, PRPF3, GEMIN4, PTPN23, HAUS1, KANSL2, ELAC2, TPR, WDHD1, NFYC, SNRNP27, SRP68, WBP11, SMNDC1, NOC4L, DDX42, CPSF6, CHORDCI, PRPF40A, EBNA1BP2, THOC1, TMEM258, SIN3A, PMPCA, NPAT, FNTA, ABCB7, TAF8, DDOST, SLC39A7, PMF1, OTOP1, ELP5, EIF3G, ATP6V1H, TRAPPCI, HAUS8, BRIX1, PDCD7, POLR1F, PSMB6, MED14, EWSR1, RNF113A, CIA03, COPS8, SUPT5H, HAUS7, PSMD6, PSMD3, GGTLC2, CIAO2B, HEATR1, PAF1, CNOT3, POLE2, RBMX, FIP1L1, VMP1, PRPF4, FOXD4, ZPR1, NEDD1, GSPT1, MFAP1, RPA2, CFAP298, MED30, SF3A3, SUPT6H, ZNHIT2, HSPA5, POLR2I, CDC7, NDC80, SNRPF, U2AF2, RPL27, RPS18, a gene involved in homologous recombination, DNA damage response pathways, splicing, or mitotic DNA damage checkpoint signaling, or a gene involved in mitoticprocesses and spindle assembly, mitotic cytokinesis, chromosome localization, regulation of spindle checkpoint, regulation of chromosome separation, RIMA transport, immune response regulation, protein localization, RNA splicing, cell cycle regulation, heterochromatin organization, G2 / M phase transition, regulation of histone acetylation, regulation of histone methylation, homologous recombination and DNA damage response, or mitotic DNA damage checkpoint signaling; and if it is determined that the cancer has a loss-of-function mutation in or does not express a functional protein corresponding to one or more of said genes, concluding that it is likely the patient would respond to treatment using inhibition of CASP3 and inhibition of CASP7.

10. A method of determining whether a cancer of a patient is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7, the method comprising determining a homologous recombination repair deficiency (HRD) score for the cancer and, if the HRD score is positive or high, concluding that the cancer is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7.

11. A method of determining whether a cancer in a tissue of a patient is likely to respond to treatment using stress-inducing anti-cancer therapies, the method comprising: assaying the cancer for expression of a p29 or a p30 form of CASP7; assaying a non-cancerous sample of the tissue for expression of the p29 or the p30 form of CASP7; and if the cancer expresses the p29 or the p30 form of CASP7 at a higher level than the non-cancerous sample of the tissue expresses the p29 or the p30 form of CASP7, concluding the cancer is less likely to respond to treatment using stress-inducing anti-cancer therapies.

12. A method of providing a prognosis for a patient with a cancer in a tissue of the patient, the method comprising:assaying the cancer for expression of a p29 or a p30 form of CASP7; assaying a non-cancerous sample of the tissue for expression of the p29 or the p30 form of CASP7; and if the cancer expresses the p29 or the p30 form of CASP7 at a higher level than the non-cancerous sample of the tissue expresses the p29 or the p30 form of CASP7, concluding the patient has a worse prognosis.

13. A method of determining whether a cancer in a tissue is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7, the method comprising: assaying a sample of the cancer for expression of a p29 form of CASP7; assaying a non-cancerous sample of the tissue for expression of the p29 form of CASP7; and if the cancer expresses the p29 form of CASP7 at a higher level than the non- cancerous sample of the tissue expresses the p29 form of CASP7, concluding the cancer is more likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7.

14. A method of determining whether a cancer in a tissue is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7, the method comprising: assaying a sample of the cancer for expression of a p30 form of CASP7; assaying a non-cancerous sample of the tissue for expression of the p30 form of CASP7; and if the cancer expresses the p30 form of CASP7, concluding the cancer is more likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7.

15. A method of selecting a therapeutic for administration to a patient to treat a cancer of the patient, the method comprising carrying out the method of determining whether the cancer is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7 as defined in any one of claims9-10, or 13-14 and, if it is determined that the cancer is likely to respond to treatment using inhibition of CASP3 and inhibition of CASP7, selecting a therapeutic that is an inhibitor of CASP3 and CASP 7 for administration to the patient.

16. The method as defined in claim 15, further comprising administering the inhibitor of CASP3 and CASP7 to the patient.

17. A method as defined in any one of claims 1-10 or 13-16, wherein the inhibitor of CASP3 comprises a small molecule inhibitor of CASP3 and wherein the inhibitor of CASP7 comprises a small molecule inhibitor of CASP7.

18. A method as defined any one of claims 1-10 or 13-17, wherein the inhibitor of CASP3 is also the inhibitor of CASP7, optionally wherein the inhibitor is M867, Caspase-3 / 7 Inhibitor I or Caspase-3 / 7 Inhibitor II, further optionally wherein the inhibitor is Caspase-3 / 7 Inhibitor I.

19. A method as defined in any one of claims 1-10 or 13-17 comprising, instead of selecting or administering an inhibitor of CASP3 and an inhibitor of CASP7, knocking out CASP3 and CASP7 in the cell or in the cancer, wherein the step of knocking out CASP3 and CASP7 is carried out using any gene editing method now known or later developed, such as CRISPR, zinc-finger nucleases (ZFN), transcription activator- 1 ike effector nucleases (TALEN), Sleeping Beauty (SB), meganucleases, or megaTALs, or knocking down CASP3 and CASP7 in the cell or in the cancer, wherein the step of knocking down CASP3 and CASP7 is carried out by administering an interfering RNA to the cell or to the cancer, optionally wherein the interfering RNA is siRNA or miRNA.

20. A method as defined in any one of claims 2 or 7, wherein the proteasome inhibitor comprises MG132, Bortezomib, or ixazomib.

21. A method as defined in any one of claims 2 or 7, wherein the DNA damaging agent comprises doxorubicin, cisplatin, anthracycline, temozolomide,carboplatin, cyclophosphamnide, dacarbazine, oxaliplatin, bendamustine, busulfan, carmustine, chlorambucil, ifosfamide, lomustine, melphalan, streptozocin, trabectedin, CX-5461, istodax or vorinostat.

22. A method as defined in any one of claims 2 or 7, wherein: the MEK inhibitor comprises trametinib; the EGFR inhibitor comprises erlotinib; the ERK inhibitor comprises SCH772984 (ERKi), MK-8353 / SCH900353, or LY3214996; or the autophagy inducing agent comprises metformin, an mTOR inhibitor, a proteasome inhibitor, sorafenib, sunitinib, apatinib, gefitinib, vorinostat, erlotinib, nilotinib, an anti-PD-1 antibody, an anti-PD-Ll antibody, or azacitidine.

23. A method as defined in any one of claims 2 or 7, wherein the PARP inhibitor comprises Olaparib (AZD2281), ABT-888 (Veliparib), Niraparib, Talazoparib, Rucaparib, Fuzuloparib or Pamiparib.

24. A method as defined in any one of claims 1 to 23, wherein the cell or the cancer has developed resistance to PARP inhibitors.

Images