Assessing and treating cancer

Abstract

This document provides methods and materials for assessing and / or treating cancer (e.g., a pancreatic cancer such as a pancreatic ductal adenocarcinoma (PDAC)). For example, this document provides methods and materials for using one or more gamma-secretase inhibitors to treat a mammal (e.g., a human) having cancer (e.g., a cancer that includes one or more cancer cells having an increased copy number of one or more gamma-secretase genes). For example, a mammal having cancer (e.g., a pancreatic such as PDAC) can be administered one or more gamma-secretase inhibitors.

Description

[0001] Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0002] ASSESSING AND TREATING CANCER

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] This application claims the benefit of U.S. Patent Application Serial No. 63 / 770,561, filed on March 12, 2025. The disclosure of the prior application is considered part of, and is incorporated by reference in, the disclosure of this application.

[0005] STATEMENT REGARDING FEDERAL FUNDING

[0006] This invention was made with government support under grant CA006973 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0007] TECHNICAL FIELD

[0008] This document relates to methods and materials for assessing and / or treating cancer (e.g., a pancreatic cancer such as pancreatic ductal adenocarcinoma (PDAC)). For example, this document provides methods and materials for using one or more gamma-secretase inhibitors to treat a mammal (e.g., a human) having cancer (e.g., a cancer that includes one or more cancer cells having an increased copy number of one or more gamma-secretase genes). For example, a mammal having cancer (e.g., a pancreatic such as PDAC) cancer can be administered one or more gamma-secretase inhibitors.

[0009] BACKGROUND PDACs occurred in -40,000 patients in 2022, and -88% of these cancers were not curable by surgical or adjuvant therapies (“U.S. Mortality Data, 1969-2022,” Surveillance, Epidemiology, and End Results (SEER) database, available at seer.cancer.gov / mortality / ). These cancers arise in the pancreatic ducts and are generally driven by genetic activations of one oncogene (KRAS) coupled with genetic inactivation of two or three tumor suppressor genes such as CDKN2A, TP53, and SMAD4 (Wood, Cancer J., 18:492 (2012)).

[0010] SUMMARY

[0011] This document relates to methods and materials for assessing and / or treating cancer (e.g., a pancreatic such as PDAC). In some cases, this document provides methods andAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0012] materials for identifying a mammal (e.g., a human) as having cancer. For example, the presence of an increased copy number of one or more gamma-secretase genes in a sample from a mammal can be used to identify that mammal as having cancer. In some cases, this document provides methods and materials for identifying a mammal (e g., a human) as being likely to develop cancer. For example, the presence of an increased copy number of one or more gamma-secretase genes in a sample from a mammal can be used to identify that mammal as being likely to develop cancer (e.g., as having a PDAC precursor legion and as being likely to develop pancreatic cancer). In some cases, this document provides methods and materials for using one or more gamma-secretase inhibitors to treat a mammal (e.g., a human) having cancer (e.g., a pancreatic cancer such as PDAC that includes one or more cancer cells having an increased copy number of one or more gamma-secretase genes). For example, a mammal having cancer (e.g., a pancreatic such as PDAC) can be administered one or more gamma-secretase inhibitors. Gamma-secretase (or y-secretase) is a polypeptide complex that includes (a) a presenilin-1 (PSEN1) polypeptide, (b) a nicastrin polypeptide, (c) an anterior pharynx-defective 1 (APH-1) polypeptide, and (d) a presenilin enhancer 2 (PEN-2) polypeptide. As demonstrated herein, pancreatic cancers have increased copy numbers of genes that encode a polypeptide that is present in a gamma-secretase polypeptide complex. For example, PDACs exhibited increased copy numbers of aNCSTN gene (encoding a nicastrin polypeptide), a APH1 A gene (encoding an APH-1 polypeptide), and / or a PSEN2 gene (encoding a PSEN2 polypeptide) (Fig. 3).

[0013] Having the ability to treat a cancer (e.g., a pancreatic such as PDAC) as described herein (e.g., by administering one or more gamma-secretase inhibitors) provides a unique and unrealized opportunity to treat a cancer type that is typically associated with late-stage detection and low 5-year survival rates (see, e.g., National Center for Health Statistics, U.S. Mortality Data, 1969-2022; Park, JAMA, 326(9):851-862 (2021); and American Cancer Society, Cancer Facts & Figures (2025)).

[0014] In general, one aspect of this document features methods for identifying a mammal as having pancreatic cancer. The methods can include, or consist essentially of, detecting the presence of an increased copy number of one or more of a PSEN1 gene, a PSEN2 gene, a NCSTN gene, an APH1 A gene, and a APH1B gene in a sample obtained from a mammalAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0015] (e g., a mammal suspected of having pancreatic cancer). The mammal can be a human. The pancreatic cancer can be a PDAC.

[0016] In another aspect, this document features methods for identifying a mammal as being likely to develop a pancreatic cancer. The methods can include, or consist essentially of, detecting the presence of an increased copy number of one or more of a PSEN1 gene, a PSEN2 gene, a NCSTN gene, an APH1 A gene, and a APH1B gene in a sample obtained from a mammal (e.g., a mammal at risk of developing a pancreatic cancer). The mammal can be a human. The pancreatic cancer can be a PDAC. The mammal can have a pancreatic intraductal papillary mucinous neoplasm (IPMN) or a mucinous cystic pancreatic neoplasm (MCPN).

[0017] In another aspect, this document features methods for treating a mammal having pancreatic cancer. The methods can include, or consist essentially of, administering a gamma secretase inhibitor and / or an antibody that targets a substrate of a gamma-secretase polypeptide complex to a mammal having pancreatic cancer. The mammal can be a human. The pancreatic cancer can be a PDAC. The method also can include identifying the pancreatic cancer as comprising one or more cancer cells having an increased copy number of one or more of a PSEN1 gene, a PSEN2 gene, a NCSTN gene, an APH1 A gene, and a APH1B gene. The gamma secretase inhibitor can be nirogacestat. The gamma secretase inhibitor can target a PSEN1 polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, and / or an APH-1 polypeptide. The gamma secretase inhibitor can target a PSEN1 polypeptide. The gamma secretase inhibitor can be MRK-560, ELN-318463, ELN-475516, SCH-900229, SCH-1500022, avagacestat, begacestat, DAPT, aDAPT analog, L-685-458, TSAI-1, LY-411575, RO-4929097, semagacestat, Compound 34, and PF-3084014, or a combination thereof. The gamma secretase inhibitor can target a PSEN2 polypeptide. The gamma secretase inhibitor can target a nicastrin polypeptide. The gamma secretase inhibitor can be semagacestat, MK-0752, E2012, BMS-708163, PF-03084014, MK-0725, or DAPT. The gamma secretase inhibitor can target an APH-1 polypeptide. The gamma secretase inhibitor can target a PEN-2 polypeptide. The substrate of the gamma-secretase polypeptide can be a MUC-1 polypeptide, a NOTCH polypeptide, an E-cadherin polypeptide, or an amyloid precursor polypeptide. The antibody that targets a substrate of a gamma-secretaseAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0018] polypeptide complex can be HMFG1, PankoMab, VU-2G7, KL-6, 5E5, 139H2, D1E11, D67C8, D76A6, 8A1, D11B8, L5C5, Val744, DECMA-1, NCH-38, HECD-1, ECCD-1, 67A4, 5H6L18, 5HCLC, PA5-32178, PA5-81393, PA5-19479, Y188, Y188, EPR5119(2), EPR5118-34, mAbP2-l, 22C11, LN27, A8717, MAB348, MOAB-2, or 25524-1-AP.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar or equivalent to those described herein can be used to practice the invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting.

[0020] The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.

[0021] BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 : Chromosome arm gains and losses in PDAC.

[0023] Figures 2A-2B: FISH on PDAC CBPANC18. Figure 2A) A FFPE section hybridized with probes for 1 p36.3 and lq25. Figure 2B) An adjacent section hybridized with probes for 19p 13 and 19q 13. The nuclei are stained with DAPI. The small nuclei at the top right corner of both pictures are non-neoplastic cells (diploid), while most other cells are PDAC ductal structures (aneuploid).

[0024] Figures 3A-3F: Chromosome arm gains and losses in various types of pancreatic tumors. Figure 3A) Chromosome arm gains and losses in Low Grade PanlNs Figure 3B) Chromosome arm gains and losses in High Grade PanlNs Figure 3C) Chromosome arm gains and losses in Low Grade IPMN. Figure 3D) Chromosome arm gains and losses in High Grade IPMN. Figure 3E) Chromosome arm gains and losses in MCN. Figure 3F) Chromosome arm gains and losses in SCA.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0025] Figures 4A-4C: Regions gained in tumors with subchromosomal gains. Figure 4A) Chromosome 8p in PDAC. Figure 4B) Chromosome Iq in PDAC. Figure 4C) Chromomosome Iq in PDAC precursors (PanINs or IPMNs).

[0026] Figures 5A-5B: Visualization of the genes that produce proteins for the two cytobands of interest lq23.2 (Figure 5 A) and lq42.13 (Figure 5B).

[0027] Figure 6. The chromosome arm gains and losses in PDAC were calculated using whole genome sequencing. The y-axis denotes the fraction of tumors with a gain (or loss) for the indicated chromosome arm on the x-axis.

[0028] Figures 7A-7D. PDAC were evaluated using Fluorescence in situ hybridization (FISH). Sections of primary PDAC were evaluated with clinically certified hybridization probes. In each panel, the y-axis denotes the percentage of cells with 3 or more copies of chromosomes Iq (A), Ip (BV), 19q (1c), and 19p (ID). As assessed by WGS, there were 18 tumors in which chromosome Iq was gained (purple bars in middle of panels A, B, C, and D) and another 18 cases in which chromosome Iq was not gained (purple bars at right of panels A, B, C, and D). In all tumors, adjacent sections were used to identify cancer cells (i.e., neoplastic cells) using standard histopathologic criteria, and the cancer cells in these sections were scored by FISH. In a subset of cases, the non-neoplastic cells in the sections of tumors (e.g., normal ductal epithelium or stromal cells) were identically scored by FISH as controls (blue bars at left of panels A B, C and D) as controls.

[0029] Figures 8A-8B. Representative images of the FISH analysis for CBPANC18 (PDAC). A) A FFPE section hybridized with probes for 1 p36.3 (red signal) and lq25 (green signal). B) An adjacent section hybridized with probes for 19p 13 (green signal) and 19ql 3 (red signal). The nuclei are stained with DAPI (blue). The small nuclei at the top right corner of both pictures are non-neoplastic cells (diploid), while most other cells are PDAC ductal structures (aneuploid).

[0030] Figures 9A-9F. The chromosome arm gains and losses in various types of pancreas neoplasia were calculated using whole genome sequencing. In each panel, the y-axis denotes the fraction of tumors with a gain (or loss) for the indicated chromosome arm on the x-axis.

[0031] Figures 10A-10D. Regions gained in tumors with sub-chromosomal gains. A) Chromosome 8q in PDAC (ichorCNA). B) Chromosome Iq in PDAC (ichorCNA). C)Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0032] Chromosome Iq in PDAC precursors (PanINs or IPMNs) (ichorCNA) D) Chromosome Iq in PDAC (QDNAseq). In each panel, the y-axis denotes the fraction of tumors with gains at the indicated chromosomal position on the x-axis.

[0033] Figures 11 A-l IB. Visualizations of the genes that produce proteins for the two cytobands of interest. A) lq23.2 and B) lq42.13. Genes in red have a statistically significant (p<le-5) correlation between copy number status and gene expression.

[0034] DETAILED DESCRIPTION

[0035] This document provides methods and materials for assessing and / or treating cancer (e.g., a pancreatic such as PDAC). In some cases, this document provides methods and materials for identifying a mammal (e.g., a human) as having cancer. For example, the presence of an increased copy number of one or more gamma-secretase genes in a sample from a mammal can be used to identify that mammal as having cancer. In some cases, this document provides methods and materials for identifying a mammal (e.g., a human) as being likely to develop cancer (e.g., a pancreatic such as PDAC). For example, the presence of an increased copy number of one or more gamma-secretase genes in a sample from a mammal can be used to identify that mammal as being likely to develop cancer (e.g., as having a PDAC precursor legion and as being likely to develop pancreatic cancer). In some cases, this document provides methods and materials for using one or more gamma-secretase inhibitors to treat a mammal (e.g., a human) having cancer (e.g., a pancreatic cancer such as PDAC that includes one or more cancer cells having an increased copy number of one or more gamma-secretase genes). For example, a mammal having cancer can be administered one or more gamma-secretase inhibitors. As described herein, one or more genes that encode a polypeptide present in a gamma-secretase complex (e.g., a PSEN1 gene, a PSEN2 gene, a NCSTN gene, a APHlAgene, and a APH1B gene) can have an increased copy number (e.g., can be duplicated) in cancer (e.g., thereby causing increased expression of gamma-secretase).

[0036] In some cases, one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) in need thereof (e.g., a human having cancer (e g., a pancreatic such as PDAC)) to reduce or eliminate the number of cancer cells present within a mammal. For example, the methods and materials described herein canAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0037] be used to reduce the number of cancer cells present within a mammal having cancer by, for example, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95, or more percent. For example, the methods and materials described herein can be used to reduce the size (e.g., volume) of one or more tumors present within a mammal having cancer by, for example, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95, or more percent.

[0038] In some cases, one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) in need thereof (e.g., a human having cancer (e.g., a pancreatic such as PDAC)) to improve survival of the mammal. For example, disease-free survival (e.g., recurrence-free survival) can be improved using the methods and materials described herein. For example, progression-free survival can be improved using the methods and materials described herein. In some cases, the methods and materials described herein can be used to improve the survival of a mammal having cancer by, for example, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95, or more percent.

[0039] In some cases, one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) in need thereof (e.g., a human having cancer (e.g., a pancreatic such as PDAC)) to delay or prevent the onset of one or more symptoms of a cancer (e.g., a pancreatic such as PDAC). Examples of symptoms of a pancreatic cancer (e.g., a PDAC) include, without limitation, belly pain that spreads to the sides or back, loss of appetite, weight loss,jaundice, light-colored stools, floating stools, dark-colored urine, itching, new diagnosis of diabetes or diabetes that's getting harder to control, pain and swelling in an arm or leg, tiredness, and weakness. In some cases, the methods and materials described herein can be used to delay the onset of one or more symptoms of a pancreatic cancer (e.g., a PDAC) by, for example, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95, or more percent.

[0040] Any appropriate mammal having cancer (e.g., a pancreatic such as PDAC) can be treated as described herein (e g., by administering one or more gamma-secretase inhibitors). Examples of mammals that can have cancer and that can be treated as described herein include, without limitation, humans, non-human primates (e.g., monkeys), dogs, cats, horses, cows, pigs, sheep, mice, and rats. In some cases, a mammal having can be a mammal (e.g., a human) that received one or more cancer treatments (e.g., radiation therapies andAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0041] chemotherapies) for a prior cancer. In some cases, a human having cancer to be treated as described herein can be resistant to one or more cancer treatments (e.g., one or more chemotherapies).

[0042] When treating a mammal (e.g., a human) having cancer (e g., a pancreatic such as PDAC) as described herein (e.g., by administering one or more gamma-secretase inhibitors), the cancer can be any type of cancer. In some cases, a cancer can include one or more solid tumors. In some cases, a cancer can be a primary cancer. In some cases, a cancer can be a metastatic cancer. A cancer can be any stage cancer (e.g., stage I, stage II, stage III, or stage IV). In some cases, a cancer that can be treated as described herein can be a pancreatic cancer. In some cases, a cancer that can be treated as described herein is not a breast cancer. Examples of cancers that can be treated as described herein include, without limitation, PDACs, pancreatic neuroendocrine tumors (NETs), pancreatic squamous cell carcinomas, pancreatic adenosquamous carcinomas, and pancreatic colloid carcinomas.

[0043] In some cases, the methods described herein can include identifying a mammal (e.g., a human) as having cancer (e g., a pancreatic such as PDAC). For example, the present of an increased copy number of one or more genes that can encode a polypeptide that is present in a gamma-secretase polypeptide complex can be used to identify a mammal as having pancreatic cancer. In some cases, an increased copy number of one or more genes that can encode a polypeptide that is present in a gamma-secretase polypeptide complex can be detected in a sample (e.g., a biological sample) obtained from a mammal. A sample can be a tissue sample (e.g., a pancreatic tissue sample) or a fluid sample (e.g., a blood sample or a urine sample). Examples of genes that can encode a polypeptide that is present in a gamma-secretase polypeptide complex include, without limitation, a presenilin-1 (PSEN1) gene, a PSEN2 gene, a nicastrin (NCSTN) gene, an anterior pharynx-defective 1 (APH1) A gene, and a APH1B gene. Any appropriate method can be used to identify the presence of an increased copy number of a gene. For example, sequencing (e.g., next generation sequencing such as whole genome sequencing (WGS), exome sequencing, RNA sequencing, and bespoke methods (e.g., realseqs or fastseqs), microarray analyses, flow cytometry, and / or fluorescence in situ hybridization (FISH) can be used to identify the presence of an increased copy number of a gene.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0044] When an increased copy number of one or more genes that can encode a polypeptide that is present in a gamma-secretase polypeptide complex is present, the increased copy number can include any copy number of that particular gene. In a diploid cell, such as a mammalian cell (e.g., human cell), two copies of each autosomal gene are present (i.e., one gene on each of the two alleles). In some cases, a gene having an increased copy number can have more than 2 copies that gene. In some cases, a gene having an increased copy number can have from about 2 to about 10 copies of that gene (e.g., from about 2 to about 9, from about 2 to about 8, from about 2 to about 7, from about 2 to about 6, from about 2 to about 5, from about 2 to about 4, from about 3 to about 10, from about 4 to about 10, from about 5 to about 10, from about 6 to about 10, from about 7 to about 10, from about 8 to about 10, from about 9 to about 10, from about 3 to about 8, from about 4 to about 7, from about 5 to about 6, from about 3 to about 5, from about 4 to about 6, from about 5 to about 7, from about 6 to about 8, or from about 7 to about 9 copies of that gene).

[0045] In some cases, the methods described herein can include identifying a mammal (e.g., a human) as having cancer (e.g., a pancreatic such as PDAC) using one or more additional methods (e.g., one or more methods other than the detection of the presence of an increased copy number of one or more genes that can encode a polypeptide that is present in a gamma-secretase polypeptide complex). Any appropriate method can be used to identify a mammal as having cancer (e.g., a pancreatic such as PDAC). For example, imaging tests (e.g., ultrasound (e.g., endoscopic ultrasound EUS)), CT scans, MRI scans, and positron emission tomography (PET) scans), blood tests (e.g., for pancreatic tumor markers such as CAI 9-9), and / or genetic testing (e.g., to look for inherited DNA changes that increase the risk of pancreatic cancer) can be used to identify mammals (e.g., humans) having cancer (e g., a pancreatic such as PDAC).

[0046] A mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) can be administered or instructed to self-administer one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors. In some cases, gamma-secretase inhibitor can be an agent that can inhibit activity of the gamma-secretase polypeptide complex. A non-limiting example of an agent that can inhibit activity of a gamma-secretase polypeptide complex is nirogacestat (e.g., OGSIVEO®).Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0047] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit a polypeptide present in the gamma-secretase polypeptide complex. Examples of polypeptides that can be targeted by gamma-secretase inhibitor include, without limitation, PSEN1 polypeptides, PSEN2 polypeptides, nicastrin polypeptides, APH-1 polypeptides, and PEN-2 polypeptides. In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit a PSEN1 polypeptide. An inhibitor of a PSEN1 polypeptide can be an inhibitor of PSEN1 polypeptide activity (e g., anti-PSENl antibodies such as neutralizing anti-PSENl antibodies and small molecules that target a PSEN1 polypeptide) or an inhibitor of PSEN1 polypeptide expression (e.g., nucleic acid molecules designed to induce RNA interference (RNAi) of PSEN1 polypeptide expression such as antisense oligonucleotides (ASOs), siRNA molecules, and shRNA molecules). Examples of inhibitors of a PSEN1 polypeptide that can be used as a gamma-secretase inhibitor to treat cancer as described herein include, without limitation, MRK-560 (Chemical Abstracts Service (CAS) No.: 677772-84-8), ELN-318463 (CAS No. 851600-86-7), ELN-475516 (CAS No.: 926658-65-3), SCH-900229 (CAS No.: 1100361-36-1), SCH-1500022, avagacestat, begacestat, DAPT, DAPT analog, L-685-458, TSAI-1, LY-411575, RO-4929097, semagacestat, Compound 34, and PF-3084014.

[0048] Additional inhibitors of aPSENl polypeptide (e g., nucleic acid molecules designed to induce RNAi against PSEN1 polypeptide expression) can be designed based on any appropriate nucleic acid (e.g., a messenger RNA (mRNA)) encoding a PSEN 1 polypeptide sequence. Examples of nucleic acids encoding a PSEN1 polypeptide sequence include, without limitation, those set forth in National Center for Biotechnology Information (NCBI) accession no. NG_007386 (e.g., version NG_007386.2).

[0049] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit a PSEN2 polypeptide. An inhibitor of a PSEN2 polypeptide can be an inhibitor of PSEN2 polypeptide activity (e.g., anti-PSEN2 antibodies such as neutralizing anti-PSEN2 antibodies and small molecules that target a PSEN2 polypeptide) or an inhibitor of PSEN2 polypeptide expression (e.g., nucleic acid molecules designed to induce RNAi of PSEN2 polypeptide expression such as ASOs,Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0050] siRNA molecules, and shRNA molecules).

[0051] Additional inhibitors of a PSEN2 polypeptide (e.g., nucleic acid molecules designed to induce RNAi against PSEN2 polypeptide expression) can be designed based on any appropriate nucleic acid (e.g., a mRNA) encoding a PSEN2 polypeptide sequence. Examples of nucleic acids encoding a PSEN2 polypeptide sequence include, without limitation, those set forth in NCBI accession no. NG_007381 (e.g., version NG_007381.2).

[0052] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit a nicastrin polypeptide. An inhibitor of a nicastrin polypeptide can be an inhibitor of nicastrin polypeptide activity (e.g., anti -nicastrin antibodies such as neutralizing anti-nicastrin antibodies and small molecules that target a nicastrin polypeptide) or an inhibitor of nicastrin polypeptide expression (e.g., nucleic acid molecules designed to induce RNAi of nicastrin polypeptide expression such as ASOs, siRNA molecules, and shRNA molecules). Examples of inhibitors of a nicastrin polypeptide that can be used as a gamma-secretase inhibitor to treat cancer as described herein include, without limitation, semagacestat, MK-0752, E2012, BMS-708163, PF-03084014, MK-0725, DAPT.

[0053] Additional inhibitors of a nicastrin polypeptide (e.g., nucleic acid molecules designed to induce RNAi against nicastrin polypeptide expression) can be designed based on any appropriate nucleic acid (e g., a mRNA) encoding a nicastrin polypeptide sequence.

[0054] Examples of nucleic acids encoding a nicastrin polypeptide sequence include, without limitation, those set forth in NCBI accession no. NG_027935 (e.g., version NG_027935.1).

[0055] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit an APH-1 polypeptide. An inhibitor of a APH-1 polypeptide can be an inhibitor of APH-1 polypeptide activity (e.g., anti-APH-1 antibodies such as neutralizing anti-APH-1 antibodies and small molecules that target a APH-1 polypeptide) or an inhibitor of APH-1 polypeptide expression (e.g., nucleic acid molecules designed to induce RNAi of APH-1 polypeptide expression such as ASOs, siRNA molecules, and shRNA molecules).

[0056] Additional inhibitors of an APH-1 polypeptide (e.g., nucleic acid molecules designed to induce RNAi against APH-1 polypeptide expression) can be designed based on anyAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0057] appropriate nucleic acid (e.g., a mRNA) encoding a APH-1 polypeptide sequence. Examples of nucleic acids encoding a APH-1 polypeptide sequence include, without limitation, those set forth in NCBI accession no. NG_029952 (e.g., version NG_029952.2) and NCBI accession no. NC_000015 (e.g., version NC_000015.10).

[0058] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be an agent that can inhibit a PEN-2 polypeptide. An inhibitor of a PEN-2 polypeptide can be an inhibitor of PEN-2 polypeptide activity (e g., anti-PEN-2 antibodies such as neutralizing anti-PEN-2 antibodies and small molecules that target a PEN-2 polypeptide) or an inhibitor of PEN-2 polypeptide expression (e.g., nucleic acid molecules designed to induce RNAi of PEN-2 polypeptide expression such as ASOs, siRNA molecules, and shRNA molecules).

[0059] Additional inhibitors of a PEN-2 polypeptide (e.g., nucleic acid molecules designed to induce RNAi against PEN-2 polypeptide expression) can be designed based on any appropriate nucleic acid (e.g., a mRNA) encoding a PEN-2 polypeptide sequence. Examples of nucleic acids encoding a PEN-2 polypeptide sequence include, without limitation, those set forth in NCBI accession no. NG_027934 (e.g., version NG_027934.1).

[0060] In some cases, gamma-secretase inhibitor that can be used to treat cancer (e.g., a pancreatic such as PDAC) as described herein can be as described elsewhere (see, e.g., Lee et al., Biochemistry, 50(2011):4973-4980 (2011)).

[0061] When a gamma-secretase inhibitor is an inhibitor of polypeptide expression (e.g., an inhibitor of PSEN1 polypeptide expression, an inhibitor of nicastrin polypeptide expression, an inhibitor of APH-1 polypeptide expression, or an inhibitor of PEN-2 polypeptide expression), one or more nucleic acid molecules designed to induce RNAi of polypeptide expression (e.g., ASOs, siRNA molecules, and shRNA molecules) can be in the form of a vector (e.g., a viral vector or a non-viral vector).

[0062] When a vector used to deliver one or more nucleic acid molecules designed to induce RNAi (e g., ASOs, siRNA molecules, and shRNA molecules) of a polypeptide present in a gamma secretase (e.g., a PSEN1 polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, a APH-1 polypeptide, or a PEN-2 polypeptide) is a viral vector, any appropriate viral vector can be used. A viral vector can be derived from a positive-strand virus or a negative-strandAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0063] virus. A viral vector can be derived from a virus with a DNA genome or a RNA genome. In some cases, a viral vector can be a chimeric viral vector. In some cases, a viral vector used as described herein can be a viral vector that can infect dividing cells. In some cases, a viral vector used as described herein can be a viral vector that can infect non-dividing cells.

[0064] Examples of virus-based vectors that can be used to deliver one or more nucleic acid molecules designed to induce RNAi of a polypeptide present in a gamma-secretase to a mammal (e.g., a human) as described herein include, without limitation, virus-based vectors based on adenoviruses, AAVs, Sendai viruses, retroviruses, or lentiviruses.

[0065] When a vector used to deliver one or more nucleic acid molecules designed to induce RNAi (e.g., ASOs, siRNA molecules, and shRNA molecules) of a polypeptide present in a gamma secretase (e.g., a PSEN1 polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, a APH-1 polypeptide, or a PEN-2 polypeptide) is a non-viral vector, any appropriate non-viral vector can be used. In some cases, a non-viral vector can be an expression plasmid (e.g., a cDNA expression vector).

[0066] In addition to one or more nucleic acid molecules designed to induce RNAi (e.g., ASOs, siRNA molecules, and shRNA molecules) of a polypeptide present in a gamma secretase (e.g., a PSEN1 polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, a APH-1 polypeptide, or a PEN-2 polypeptide), a vector (e.g., a viral vector or a non-viral vector) can contain one or more regulatory elements operably linked to the nucleic acid molecule(s) designed to induce RNAi of a polypeptide present in a gamma secretase. Such regulatory elements can include promoter sequences, enhancer sequences, response elements, signal peptides, internal ribosome entry sequences, polyadenylation signals, terminators, and inducible elements that modulate expression (e.g., transcription or translation) of a nucleic acid. The choice of regulatory element(s) that can be included in a vector can involve consideration of several factors, including, without limitation, inducibility, targeting, and the level of expression desired. For example, a promoter can be included in a vector to facilitate transcription of a nucleic acid molecule designed to induce RNAi of a polypeptide present in a gamma secretase. A promoter can be a naturally occurring promoter or a recombinant promoter. A promoter can be ubiquitous or inducible (e.g., in the presence of tetracycline) and can affect the expression of a nucleic acid encoding a polypeptide in a general or tissue-Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0067] specific manner. As used herein, “operably linked” refers to positioning of a regulatory element relative to a nucleic acid sequence in such a way as to permit or facilitate expression of the nucleic acid sequence. For example, a vector can contain a promoter and nucleic acid sequence designed to induce RNAi of a polypeptide present in a gamma secretase. In this case, the promoter is operably linked to a nucleic acid sequence designed to induce RNAi of a polypeptide present in a gamma secretase such that it drives expression of the nucleic acid sequence to induce RNAi of the polypeptide present in a gamma secretase in cells.

[0068] In some cases, methods for treating a mammal (e.g., a human) having cancer (e.g., a pancreatic cancer such as PDAC that includes one or more cancer cells having an increased copy number of one or more gamma-secretase genes) provided herein can include (e.g., in addition to or in place of one or more gamma-secretase inhibitors) administering to the mammal one or more antibodies (e.g., neutralizing antibodies) that can target (e.g., target and bind) a substrate of a gamma-secretase polypeptide complex. Examples of substrates of a gamma-secretase polypeptide complex include, without limitation, MUC-1 polypeptides, NOTCH polypeptides (e.g., NOTCH1, NOTCH2, NOTCH3, NOTCH4, and cleaved Notchl), E-cadherin polypeptides, and amyloid precursor polypeptides (e.g., APP and APP A4). Examples of antibodies that can target a substrate of a gamma-secretase polypeptide complex include, without limitation, HMFG1, PankoMab, VU-2G7, KL-6, 5E5, 139H2, D1E11, D67C8, D76A6, 8A1, D11B8, L5C5, Val744, DECMA-1, NCH-38, HECD-1, ECCD-1, 67A4, 5H6L18, 5HCLC, PA5-32178, PA5-81393, PA5-19479, Y188, Y188, EPR5119(2), EPR5118-34, mAbP2-l, 2201, LN27, A8717, MAB348, MOAB-2, and 25524-1-AP. In some cases, an antibody that target a substrate of a gamma-secretase polypeptide complex can be as described elsewhere (see, e.g., Pourjafar, Immunotherapy, 12(17): 1269-1286 (2020); Danielczyk, Cancer Immunol Immunother., 55(11): 1337-47 (2006); Thie, PLoS One, 6(l):el5921 (2011); Wu, Nature, 464(7291): 1052-7 (2010);

[0069] Petrova, Mol Biol Cell, 27(21):3233-3244 (2016); and Muller, Nat Rev Neurosci, 18(5):281-298 (2017)).

[0070] In some cases, one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be formulated into a composition (e.g., a pharmaceutically acceptable composition) for administration to a mammal (e.g., a human) having cancer (e.g., aAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0071] pancreatic such as PDAC). For example, one or more gamma-secretase inhibitors can be formulated together with one or more pharmaceutically acceptable carriers (additives), excipients, and / or diluents. Examples of pharmaceutically acceptable carriers, excipients, and diluents that can be used in a composition described herein include, without limitation, cyclodextrins (e.g., beta-cyclodextrins such as KLEPTOSE®), dimethyl sulfoxide (DMSO), sucrose, lactose, starch (e.g., starch glycolate), cellulose, cellulose derivatives (e.g., modified celluloses such as microcrystalline cellulose, and cellulose ethers like hydroxypropyl cellulose (FTPC) and cellulose ether hydroxypropyl methylcellulose (FTPMC)), xylitol, sorbitol, mannitol, gelatin, polymers (e.g., polyvinylpyrrolidone (PVP), polyethylene glycol (PEG), crosslinked polyvinylpyrrolidone (crospovidone), carboxymethyl cellulose, polyethylene-polyoxypropylene-block polymers, and crosslinked sodium carboxymethyl cellulose (croscarmellose sodium)), titanium oxide, azo dyes, silica gel, fumed silica, talc, magnesium carbonate, vegetable stearin, magnesium stearate, aluminum stearate, stearic acid, antioxidants (e.g., vitamin A, vitamin E, vitamin C, retinyl palmitate, and selenium), citric acid, sodium citrate, parabens (e.g., methyl paraben and propyl paraben), petrolatum, dimethyl sulfoxide, mineral oil, serum proteins (e.g., human serum albumin), glycine, sorbic acid, potassium sorbate, water, salts or electrolytes (e.g, saline such as phosphate buffered saline, protamine sulfate, disodium hydrogen phosphate, potassium hydrogen phosphate, sodium chloride, and zinc salts), colloidal silica, magnesium trisilicate, polyacrylates, waxes, wool fat, lecithin, and corn oil.

[0072] In some cases, when a composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors is administered to a mammal (e.g., a human) having cancer (e g., a pancreatic such as PDAC), the composition can be designed for oral or parenteral (including, without limitation, subcutaneous, intramuscular, intravenous, intradermal, intracerebral, intra cisterna magna, or intraventricular injections) administration to the mammal. Compositions suitable for oral administration include, without limitation, liquids, tablets, capsules, pills, powders, gels, and granules. Compositions suitable for parenteral administration include, without limitation, aqueous and non-aqueous sterile injection solutions that can contain anti-oxidants, buffers, bacteriostats, and solutes that render the formulation isotonic with the blood of the intended recipient.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0073] A composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) locally or systemically. For example, a composition containing one or more gamma-secretase inhibitors can be administered locally by direct injection (e.g., an intratumoral injection).

[0074] In some cases, a composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be in the form of a sterile injectable suspension (e.g., a sterile injectable aqueous or oleaginous suspension). This suspension may be formulated using, for example, suitable dispersing or wetting agents (such as, for example, Tween 80) and suspending agents. The sterile injectable preparation can be a sterile injectable solution or suspension in a non-toxic parenterally-acceptable diluent or solvent. Examples of acceptable vehicles and solvents that can be used include, without limitation, saline, mannitol, water, Ringer’s solution, and isotonic sodium chloride solution. In addition, sterile, fixed oils can be used as a solvent or suspending medium.

[0075] A composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase can be administered to a mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) in any appropriate amount (e.g., any appropriate dose). An effective amount of a composition containing one or more gamma-secretase inhibitors can be any amount that can treat a mammal having cancer as described herein without producing significant toxicity to the mammal. The effective amount can remain constant or can be adjusted as a sliding scale or variable dose depending on the mammal’s response to treatment. Various factors can influence the actual effective amount used for a particular application. For example, the frequency of administration, duration of treatment, use of multiple treatment agents, route of administration, and / or severity of the PSP in the mammal being treated may require an increase or decrease in the actual effective amount administered.

[0076] A composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) having cancer (e g., a pancreatic such as PDAC) in any appropriate frequency. The frequency of administration can be any frequency that can treat a mammal having cancer without producing significant toxicity to the mammal. For example, the frequency of administration can be from aboutAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0077] twice a day to about once a month, once a day to about once every two weeks, or from about once every other day to about once a week. The frequency of administration can remain constant or can be variable during the duration of treatment. As with the effective amount, various factors can influence the actual frequency of administration used for a particular application. For example, the effective amount, duration of treatment, use of multiple treatment agents, and / or route of administration may require an increase or decrease in administration frequency.

[0078] A composition containing one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors can be administered to a mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) for any appropriate duration. An effective duration for administering or using a composition containing one or more gamma-secretase inhibitors can be any duration that can treat a mammal having cancer without producing significant toxicity to the mammal. For example, the effective duration can vary from several weeks to several months, from several months to several years, or from several years to a lifetime. Multiple factors can influence the actual effective duration used for a particular treatment. For example, an effective duration can vary with the frequency of administration, effective amount, use of multiple treatment agents, and / or route of administration.

[0079] In some cases, the methods for treating a mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) as described herein (e.g., by administering one or more gamma-secretase inhibitors) can include administering to the mammal one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors as the sole active ingredient to treat the mammal. For example, a composition containing one or more gamma-secretase inhibitors can include the one or more gamma-secretase inhibitors as the sole active ingredient(s) in the composition that is effective to treat a mammal having cancer.

[0080] In some cases, the methods for treating a mammal e.g., a human) having cancer (e.g., a pancreatic such as PDAC) as described herein (e.g., by administering one or more gamma-secretase inhibitors) also can include administering to the mammal one or more (e.g., one, two, three, four, five or more) additional agents used to treat cancer to the mammal. For example, a combination therapy used to treat cancer can include administering to the mammal (e.g., a human) one or more gamma-secretase inhibitors described herein and one orAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0081] more (e.g., one, two, three, four, five or more) anti-cancer agents. In some cases, an anticancer agent can be a chemotherapy. In some cases, an anti-cancer agent can be an immunotherapy. Examples of additional agents that can be used to treat cancer include, without limitation, KRAS inhibitors. In cases where one or more gamma-secretase inhibitors are used in combination with one or more additional agents treat cancer, the one or more additional agents can be administered at the same time (e.g., in a single composition containing both one or more gamma-secretase inhibitors and the one or more additional agents) or independently. For example, one or more gamma-secretase inhibitors described herein can be administered first, and the one or more additional agents administered second, or vice versa.

[0082] In some cases, the methods for treating a mammal (e.g., a human) having cancer (e.g., a pancreatic such as PDAC) as described herein (e.g., by administering one or more gamma-secretase inhibitors) also can include performing one or more (e.g., one, two, three, four, five or more) therapies used to treat cancer on the mammal. For example, a combination therapy used to treat cancer can include administering to the mammal one or more (e.g., one, two, three, four, or more) gamma-secretase inhibitors described herein and performing one or more additional therapies used to treat cancer on the mammal. Examples of therapies that can be used to treat cancer include, without limitation, surgery and / or radiation therapy. In cases where one or more gamma-secretase inhibitors described herein are used in combination with one or more additional therapies used to treat cancer, the one or more additional therapies can be performed at the same time or independently of the administration of one or more gamma-secretase inhibitors described herein. For example, one or more gamma-secretase inhibitors described herein can be administered before, during, or after the one or more additional therapies are performed.

[0083] In some cases, the size of the cancer (e.g., the number of cancer cells and / or the volume of one or more tumors) present within a mammal and / or the severity of one or more symptoms of the cancer (e.g., a pancreatic such as PDAC) being treated can be monitored. Any appropriate method can be used to determine whether or not the size of the cancer present within a mammal is reduced. For example, imaging techniques can be used to assess the size of the cancer present within a mammal (e.g., at different time points).Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0084] The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.

[0085] EXAMPLES

[0086] Example I: Extra copies of gamma-secretase genes play a role in the early phases of pancreatic neoplasia

[0087] This Example describes the discovery that extra copies the gamma secretase gene is found in PDACs.

[0088] Results

[0089] Copy number alterations in PDAC

[0090] Widely varying heterogeneous chromosomal changes have been observed in PDAC (Knouse et al., Annual Review of Cancer Biology, 1 : 335— 54 (2017); Ben-David et al., Nat Rev Genet, 2144-62 (2020); and Douville et al., Proc Natl Acad Sci USA, 115: 1871-6 (2018)). To evaluate pancreatic tumors in a uniform fashion, DNA was purified from the primary tumors of 533 PDAC patients. Primary tumors of the pancreas (as opposed to metastatic lesions in other organs that originated in the pancreas) are often composed of a small number of neoplastic cells embedded within a large number of non-neoplastic cells, such as fibroblasts, inflammatory cells, and endothelial cells. Regions of the primary tumors that had relatively high contents of neoplastic cells were macrodissected and purified DNA from those regions. The DNA was used to make libraries by a technology that allowed high conversion of starting template DNA molecules to library DNA molecules. Furthermore, each DNA molecule was barcoded so that it could be uniquely identified and to minimize sequencing redundancy. The libraries were sequenced to an average coverage depth of 172M (IQR 133M to 184M) and were analyzed by a variation of ichorCNA.

[0091] A summary of chromosome gains and losses in these 533 PDACs is shown in Fig. 1. Sample level calls are provided for all chromosome arms and all autosomal 500kb intervals. The chromosome arms with the greatest fraction of losses were chromosome 9p, 17p, and 18q, which were lost in 60.4%, 69.8%%, and 74.3% of the PDAC respectively.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0092] With respect to gains, two chromosome arms stood out: chromosome Iq and chromosome 8q were gained in 39.8% and 46.3% of the 533 PDACs, respectively. In those cancers with gains of chr Iq, the copy number of chromosome Iq varied from 3 to 7. (Fig. IB). Analogously, in those cancers with gains of chr 8q, the copy number of chromosome 8q varied from 3 to 7 (Fig. 1C).

[0093] Fluorescence in situ hybridization (FISH)

[0094] The WGS data shows that chromosome Iq gains are common among PDAC but does not reveal clonality, i.e., how many cells within the lesion contain gains. To answer this question at the single cell level, FISH was used with a labeled probe from a gene located in the middle of chromosome 1 (ABL2, located at chromosome lq25). In each case, 100 nuclei from the cancer cells within the section were evaluated in a blinded fashion.

[0095] Five PDACs which had gains of the Iq arm as assessed by WGS were evaluated first. In each cancer, FISH showed that >89% of the cells contained had gained a chromosome Iq sequences, with a mean of 3.9 chromosome Iq signals per nucleus (Fig. 2A and 2B).

[0096] As a negative control for the FISH results, sections of non-neoplastic cells from three other pancreata were evaluated. There were no extra chromosome Iq arms in any case, with a mean number of chromosome Iq signals 2.0 per nucleus (Fig. 2C).

[0097] 14 PDACs that appeared to lack clonal gains of chromosome Iq upon WGS analysis were then evaluated. More than 2 chromosome Iq arms in 11 (86%) of these 14 cases were observed (Fig. 2D). In each of these 11 cases, the great majority of nuclei (>99%) harbored 3 or more gains, with a mean number of chromosome Iq signals of 4.1.

[0098] To determine if tetraploidy was the basis for the discordant results between the WGS and in situ hybridization data, the same sections were evaluated with a probe for TP73, located at lp36. With WGS, no chromosome arms were observed to be gained in PDACs in 8 of the 11 cases. In contrast, in situ hybridization revealed that 9 (82%) of the 11 had extra copies of chromosome Ip. To determine whether these extra copies of chromosome Ip were the result of an isolated gain of the entire chromosome 1 (both p and q arms) vs. a more widespread duplication of the genome, hybridization was performed with probes fromAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0099] chromosome 19p and 19q. In all 9 cases (100%) with extra copies of both chromosome Ip and Iq, extra copies of chromosome 19p and 19q were also observed (Fig. 2).

[0100] Given the FISH data on the PDACs without WGS-detectable chromosome Iq gains, the five PDACs with WGS-detectable chromosome Iq gains were re-evaluated. In four of these five cases, FISH revealed extra copies of chromosome Ip, none of these apparent by WGS (Figure 2). And similarly, all five cases had gains of either chromosome 19p or chromosome 19q (four of five with gains of 19p and five of five with gains of 19q) (Figure 2). This provides evidence for genome duplication in those cases with and without chromosome Iq gain. But there was a clear difference between the tumors with and without chromosome Iq gains evident by WGS: those with WGS-detectable Iq gains by WGS had a higher ratio of Iq to Ip signals (median 1.3) than those without WGS-detectable Iq gains (median 1.0).

[0101] Copy number alterations in the precursors to PDAC

[0102] The great majority (-90%) of PDACs develop from PanINs (Pancreatic Intraepithelial Neoplasia). These lesions are located in the ducts and can be classified into Low Grade and High Grade based on the degree of dysplasia upon histopathologic investigation, e.g., their resemblance to normal pancreatic ductal cells, including the size, shape, and position of the cells and the nuclei within the cells. PanINs cannot be detected by imaging because they are <0.5 cm in diameter. Instead, they can only be observed through microscopic evaluation of excised pancreata. There are -1000 low grade PanINs in every individual older than 65 years, and their danger of progressing to malignancy is extremely low (Braxton et al., Nature, 2024;629:679-87 (2024)). In contrast, high grade PanINs are much less common and are thought to have a relatively high chance of progressing to a PDAC (Hosoda et al., J Pathol. 242:16-23 (2017)).

[0103] Because PanINs are so small, microdissection of the lesions was required to a high neoplastic fraction of the tissues used for purifying DNA in purified DNA. Of 67 low grade PanINs, DNA purified from microdissected Low Grade PanINs had relatively few gains or losses compared to PDACs (Fig. 3). Notably, the arms most frequently gained or lost in PDAC (Iq, 8q, 9p, 17p, 18q) were lost in <10% of these lesions.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0104] The evaluation of microdissected High Grade PanINs revealed a picture different from either LG PanINs or PDACs (Fig. 3). The most frequently altered chromosome arm was Iq, gained in 49% of the 37 lesions studied. Chromosome 8q gains were noted only in 27% of the HG PanINs, and chromosome arms 9p, 17p, and 18q were lost in 24%, 41%, and 32% of the lesions. Thus, alterations of chromosomes other than Iq were much less frequent in HG PanINs than in PDACs.

[0105] Most of the PDACs that don’t originate in PanINs are believed to originate from IPMNs (Intraductal Pancreatic Mucinous Neoplasms). These lesions are also located in the pancreatic ducts, and PanINs and IPMNs represent a continuum. Though there are some histopathologic differences noted, it is challenging to distinguish between a small IPMN and a large PanlN (Graham et al., Carcinogenesis, 45:801-16 (2024)). But the lesions are clearly distinguished by clinical imaging: IPMNs can be detected by imaging, such as CT scans, MRI, or sonography, while PanINs cannot be. IPMNs can be histopathologically classified as Low Grade or High Grade in a fashion similar to that of PanINs.

[0106] DNA from 76 IPMNs was evaluated. In LG IPMNs, the most frequent chromosome arm change by far was Iq gain, occurring in 33% of 76 tumors. Other chromosome arms frequently altered in PDACs (8q, 9p, 17p, 18q) were unusual, occurring in <10%. In 71 HG IPMNs, chromosome Iq gain was again gained in a high fraction of tumors (35%), but other gains and losses were also observed (30% had 8q gain, 25% had 9p loss, 27% had 17p loss, 38% had 18q loss). It was notable that LG PanINs had relatively few chromosome Iq gains (7%), while LG IPMNs frequently had gains of chromosome Iq (33%).

[0107] PDACs can also, though more uncommonly, originate from Mucinous Cystic Pancreatic Neoplasms (MCPNs). These are distinguished by epidemiology, histopathology, and mutated genes from PanINs and IPMNs. For example, MCPNs usually occur in middle aged females and often contain mutations in RNF43. 8 microdissected MCPNs were evaluated. None had gains of chromosome Iq (Fig. 3C).

[0108] As an additional comparison, microdissected regions from ten serous cystadenomas (n=l 0), benign tumors that do not progress to malignancy, were evaluated. Few gains or losses of any of the chromosome arms highlighted above were observed in any tumor.

[0109] Instead, the great majority of these tumors had 3p losses occurring in 50% of the lesions.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0110] Localization of the regions on chromosome Iq containing candidate driver genes

[0111] In a heritable disease, linkage analysis can identify candidate genes that may be responsible for the predisposition to the disease. In cancer, which is mainly driven by somatically acquired mutations, the analog of linkage is the identification of gene(s) that are most frequently associated with the genetic alteration of interest in sporadic tumors. This strategy was applies to identify candidate driver gene(s) on chromosome Iq.

[0112] Chromosome 8q was evaluated first. Indeed, chromosome 8q is the most frequently gained chromosome arm in PDAC (Fig. 1), though it is not often gained in PDAC precursors (Fig. 2). The WGS data showed that 50% (n=124) of the PDAC that gained chromosome 8q gained the entire 8q arm, while the remaining 50% (n=123) gained only a subset of chromosome 8q genes. Cancers that gained the entire arm can obviously not be used to identify the presumptive driver gene on that chromosome. However, evaluation of the 123 PDACs that gained only a subset of the genes on the arm can address this question. On chromosome 8q, there was a single broad peak centered at 8q24.21, spanning position 8q: 127,300,001 to 131,500,000 (hgl9) 4,200,000bp— (Fig. 5A). This peak includes CMYC. Moreover, 93 % of the cancers that gained all or any part of chromosome 8q gained this particular region.

[0113] Moving on to chromosome Iq, the WGS data showed that 139 (66%) of the PDACs that gained chromosome Iq gained the entire Iq arm, while the remaining 73 (34%) gained only a subset of chromosome Iq genes. Unlike the single major peak on chromosome 8q, there were two major peaks on Iq, one centered at lq23.2 (chrl: 159, 100,001 to 160,500,000), and the other centered at lq42.13 (chrl: 227,000,001-230,700,000) (Fig. 4B).

[0114] 91% (n=192) of the cancers that gained any part of chromosome Iq gained the lq23.2 locus and 85% (n=180) of the cancers that gained any part of chromosome Iq gained the 1 q42.13. Considering only the 212 PDACs that had gained extra copies of any part of chromosome Iq, 77% (n=164) gained both the lq23.2 and lq42.13 loci, 13% (n=28) gained the lq23.2 locus but not the 1 q42.13 locus, and 8% (n=16) gained the 1 q42.13 locus but not the lq23.2 locus; and 2% (n=4) gained neither. The WGS data on PDAC precursor lesions highlighted the same two regions of chromosome Iq, but the number of tumors was much smaller and the localization therefore of lower resolution (Fig. 4C).Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0115] Identification of candidate oncogenes on chromosome Iq

[0116] According to RefSeq, the critical region on chromosome lq23.2 (chrl:159, 100,001- 160,500,000, Fig. 5A) contained 38 transcribed genes, 32 of which encode proteins (Fig. 5A). To determine the most likely genes responsible for the putative selective growth advantage conferred by this region of chromosome Iq, the correlation between gene gain and gene expression in The Cancer Genome Atlas (TCGA) database was evaluated. There were 9 genes in which the level of expression and copy number was correlated in a statistically significant fashion (p<le-5; Fig. 5A). Notable among these was NCSTN, encoding a component of the y-secretase complex, as explained below.

[0117] The second critical region on chromosome lq42.13 (chrl: 227,000,001-230,700,000, Fig. 5B) contained 61 transcribed genes, 33 of which encode proteins. To determine the most likely genes responsible for the putative selective growth advantage conferred by this region of chromosome Iq, the correlation between gene gain and gene expression was similarly evaluated in The Cancer Genome Atlas (TCGA) database. There were 14 genes in which the level of expression and copy number was correlated in a highly statistically significant fashion (p<le-5, Fig. 5B). Notable among these was PSEN2, encoding another component of the y-secretase complex.

[0118] Relationship of Iqgain to common genetic alterations in PDAC precursor lesions

[0119] The average depth of the WGS was 173 M unique reads per sample, corresponding to ~5.8x coverage, and this was insufficient to reliably evaluate subtle mutations, such as single base substitutions, insertion, or deletions, in the genes driving PDAC development. Primers to evaluate 37 amplicons were therefore designed, considering both the Watson and Crick strands, to detect subtle mutations in commonly altered regions of KRAS, NRAS, GNAS, CDK2NA, SMAD4 and TP53 in a highly specific fashion (Methods). This analysis was performed on 138 DNA samples from carefully dissected precursor lesions, including 31 Low Grade PanINs, 24 High Grade PanINs, 43 Low Grade IPMNs, and 40 high grade IPMNss. Mutations in either KRAS or GNAS were identified in all 138 samples (100%), as expected from prior studies suggesting that mutations in these two genes initiate pancreatic ductal neoplasia. Because all lesions had mutations in these two genes, there could obviouslyAttorney Docket No. 44807-0508W01 / P18672-02

[0120] be no correlation between Iq gain and mutations in KRAS or GNAS. But we could evaluate how often subtle mutations in CDKN2, TP53, or SMAD4 occurred. We found that the majority (76%) of the 38 precursor lesions with Iq gain did not have detectable mutations in CDKN2, TP53, or SMAD4, consistent with the idea that Iq gains can precede these other genetic alterations during tumor progression. Conversely, the majority (65%) of the 26 precursor lesions in which CDKN2, TP53, or SMAD4 mutations were detected did not harbor Iq gains, consistent with the idea that imbalanced Iq gains are not required for tumor progression.

[0121] Correlation of Copy Number vs mRNA Expression for the Cytoband lq23.2

[0122]

[0123] Attorney Docket No. 44807-0508W01 / P18672-02

[0124]

[0125] Correlation of Copy Number vs mRNA Expression for the Cytoband lq42.13

[0126]

[0127] Attorney Docket No. 44807-0508W01 / P18672-02

[0128]

[0129] Together, these results demonstrate that inhibitors or modulators of y-secretase activity can be used for the treatment of PDACs in which extra copies of gamma-secretase genes and / or increased expression of the gamma-secretase protease complex are observed.

[0130] Methods

[0131] Library construction

[0132] A library preparation workflow was developed that can efficiently recover input DNA and simultaneously incorporate double-stranded molecular barcodes (Cohen et al., Nature biotechnology, 39:1220-7 (2021)). In brief, libraries were prepared using an Accel-NGS 2S DNA Library Kit (Swift Biosciences, 21024) with the following critical modifications: 1) DNA was pretreated with 3 U of USER enzyme (New England BioLabs, M5505L) for 15 minutes at 37 °C to excise uracil bases; 2) the SPRI bead / PEGNaCl ratios used after each reaction were 2. Ox, 1.8x, 1.2x and 1.05x for end repair 1, end repair 2, ligation 1 and ligationAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0133] 2, respectively; 3) a custom 50 pM 3' adapter was substituted for reagent Y2 and 4) a custom 42 pM 5' adapter was substituted for reagent B2. Libraries were subsequently PCR amplified in 50-pl reactions using primers targeting the ligated adapters. The following reaction conditions were used: 1 x NEBNext Ultra II Q5 Master Mix (New England BioLabs, M0544L), 2 pM universal forward primer and 2 pM universal reverse primer. Libraries were amplified with 8 or 11 cycles of PCR, depending on how many experiments were planned, according to the following protocol: 98 °C for 30 s, cycles of 98 °C for 10 s, 65 °C for 75 s and hold at 4 °C. If eight cycles were used, the libraries were amplified in single 100-pl reactions. If 11 cycles were used, the libraries were divided into eight aliquots and amplified in eight 50-pl reactions, each supplemented with an additional 0.5 U of Q5 Hot Start High-Fidelity DNA Polymerase (New England BioLabs, M0493L), 1 pl of 10 mM dNTPs (New England BioLabs, N0447L) and 0.4 pl of 25 mM MgC12 solution (New England BioLabs, B9021S). The products were purified with 1.8x SPRI beads (Beckman Coulter, B23317) and eluted in EB buffer (Qiagen).

[0134] Aneuploidy Analysis

[0135] Library DNA was amplified in 50 pl reactions in Ultra Q5 with primers at 2 pM for seven cycles with the following conditions: 98 °C for 30 s, then seven cycles of 98 °C for 10 s to denature, and 65 °C for 75 s to anneal and extend. WGS libraries were sequenced on a NovaSeq 6000 with paired-end 2xl00bpreads. The depth of sequencing averaged 40.0 M reads of 100 bp (IQR 33.8M to 45. IM). Cutadapt was used to trim 27 base pairs from both reads and BWA-MEM was used to align reads to the h l9 genome. Duplicate molecules were marked and removed using samtools. Reads with a quality >10 were binned into 500kb intervals and counted. IchorCNA was then used to perform GC correction and call the estimated tumor fraction using the following parameters: “ — chrs

[0136] “c(l, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,18,19,20,21,22) —normal “c(0.9, 0.95, 0.98)” -estimateScPrevalence FALSE -scStates “c()” -maxCN 3 -ploidy “c(2)” -normalPanel. Two panels were generated — one for fresh frozen samples (n=9) and one for formal-fixed paraffin-embedded (FFPE) samples (n=5). The normal samples used in the panels were not included in the summary statistics described above.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0137] Fluorescence in situ hybridization (FISH):

[0138] A commercially available probe set provided by Abbott Molecular comprised of LSI lp36 / TP73 (SpectrumOrange) and LSI Iq25 / ABL2 (SpectrumGreen) was applied to individual slides, hybridized, and washed according to the TRIS / EDTA FISH protocol. This process was repeated using a commercially available probe set provided by Abbott Molecular comprised of LSI 19ql3 / EHD2 (SpectrumOrange) and 19pl3 / D19S221 (SpectrumGreen). In each case, 100 nuclei from the cancer cells within the section were evaluated in a blinded fashion using two independent analysts.

[0139] Statistical Analysis

[0140] Cohort sample size was not selected for statistical power but rather was based on sample availability.

[0141] Example 2: Evidence that extra copies of chromosome Iq play a role in the early phases of pancreatic neoplasia

[0142] This Example describes the discovery that copy number increases of the y-secretase genes on Iq can be candidate oncogenes in pancreatic neoplasms.

[0143] The results in this Example re-present and expand on at least some of the results provided in other Examples.

[0144] Results

[0145] Copy number alterations in PDAC.

[0146] Previous studies of PDAC, using karyotypic evaluation of DNA sequencing, have documented heterogeneous chromosomal changes, with the occurrence of trisomies of specific chromosomes differing widely among these studies (Knouse et al., Annual Review of Cancer Biology, 1:335-54 (2017); Ben-David et al., Nat Rev Genet, 21:44-62 (2020); and Douville et al., Proc Natl Acad Sci USA, 115:1871-6 (2018)). To evaluate pancreatic tumors in a uniform fashion, we purified DNA from the primary tumors of 535 PDAC (derived from 521 patients). Primary tumors of the pancreas (as opposed to metastatic lesions in other organs that originated in the pancreas) are often composed of a small number of neoplasticAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0147] cells embedded within a large number of non-neoplastic cells, such as fibroblasts, inflammatory cells, and endothelial cells. We therefore carefully dissected regions of the primary tumors that had relatively high contents of neoplastic cells, and purified DNA from those regions. The DNA was used to make libraries by a technology that allowed high conversion of starting template DNA molecules to library DNA molecules. Furthermore, each DNA template molecule was barcoded so that it could be uniquely identified. The libraries were sequenced and had on average 125 M (IQR of 88M to 146M) properly paired unique reads. Segmentation and analysis was performed using ichorCNA (Materials and Methods).

[0148] A summary of chromosome gains and losses in these 535 PDAC is shown in Fig. 6. Higher resolution copy number calls were based on segmentation of 500 kb intervals. Raw ichorCNA outputs and genome wide plots for all samples are available (see, Materials and Methods). The chromosome arms with the highest fraction of losses were chromosome 9p (CDK2NA), 17p (TP53), and 18q (SMAD4), which were lost in 60.4%, 69.9%, and 74.2% of the PDAC, respectively. The predominance of these chromosome arms was expected given that the predominant known PDAC driver genes are CDKN2A, TP53, and SMAD4, located on chromosome 9p, 17p, and 18q, respectively. We then compared our results to The Cancer Genome Atlas which evaluated aneuploidy in 184 pancreatic adenocarcinoma the Affymetrix SNP 6.0 and GISTIC2 algorithm. TCGA reported slightly lower rates of loss on these chromosomes: 42% (9p), 46% (17p), and 62% (18q).

[0149] With respect to gains, two chromosome arms stood out: chromosome Iq and 8q (hereinafter denoted (Iq and 8q) were gained in 39.8% and 46.7% of the 535 PDAC, respectively. TCGA estimates were again slightly lower than ours (Iq — 28% and 8q — 25% ) (Broad Institute TCGA Genome Data Analysis Center, “SNP6 Copy number analysis (GISTIC2)” (Broad Institute of MIT and Harvard, 2016); doi.org / 10.7908 / ClBP0271). In those cancers with gains of Iq, the copy number of Iq varied from 3 to 7 among different cancers. Analogously, in those cancers with gains of 8q, the copy number of 8q also varied from 3 to 7 among different cancers. We asked whether Iq and 8q gains as well as other common chromosomal gains were mutually exclusive or independent events. While someAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0150] deviations from random partitioning were apparent, the effects sizes were small, suggesting minimal if any biological selection.

[0151] We next asked whether the choice of bioinformatic algorithm would change the interpretation of aneuploidy in PDAC. We repeated the analysis using QDNAseq to identify copy number alterations (Materials and Methods). Raw QDNAseq outputs and genome wide plots for all samples are available (see, Materials and Methods). The two most common gains were again Iq (42.2% in QDNAseq vs 39.8% inichorCNA) and 8q (45.8% vs 46.7%), while the most common losses were 9p (69.9% vs 69.9%; 17p (73.6% vs 69.8%), and 18q (81.7% vs 74.2%). The two algorithms therefore produced very similar results, with one notable exception: a subset of the tumors that did not have detectable levels of aneuploidy with ichorCNA had low, but detectable levels of aneuploidy when assessed with QDNAseq.

[0152] Fluorescence in situ hybridization (FISH).

[0153] The WGS data shows that Iq gains are common among PDAC but do not provide data about the distribution of gains among the cells of a tumor, i.e., how many neoplastic cells within the cancer contain gains. The Iq gains reported in Fig. 6 additionally reflect relative gains, i.e., relative to all other chromosomal arms in the same sample, rather than absolute gains. For example, if the cells are uniformly tetrapioid through genome doubling, this would be impossible to discern with the technique we used - all chromosome arms would appear to have a copy number of 2.

[0154] To provide data about cellular distribution and the absolute number of Iq chromosomes per pancreatic cancer cell, we used FISH with a labeled probe from a gene located in the middle of chromosome 1 (ABL2, located at lq25). In each case, 50 to 100 nuclei within formal-fixed and paraffin-embedded (FFPE) sections were evaluated by two individuals in a blinded fashion (Materials and Methods). As controls for these experiments, we evaluated acinar regions of seven PDAC patients, using FFPE sections of the surgically excised normal pancreas distant from the cancer itself. This ensured the absence of cancer cell nuclei in the entire evaluated sample. We additionally evaluated normal acinar regions, or the ducts themselves, of two other PDAC patients. In these two patients, pancreatic cancer cells were observed in the FFPE sections but separated by at least 2 mm from the evaluatedAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0155] non-neoplastic acini and ducts. In the non-neoplastic cells on these sections, the average number of Iq fluorescent signals per non-neoplastic nucleus ranged from 1.8 to 2.1, and only 3.4% of 900 nuclei contained more than three or more fluorescent signals.

[0156] We next evaluated neoplastic cell nuclei in FFPE sections of cancers from 33 PDAC patients, 15 with Iq gains evident in WGS and 18 without Iq gains. FISH confirmed that all 15 patients with WGS-defined chromosome gains in their cancers had extra copies of Iq. The average number of Iq fluorescent signals ranged from 2.9 to 4.9 in these 15 patients, and the majority of nuclei in each patient contained at least three fluorescent signals (Fig. 7 and Fig.

[0157] 8). We found that some of the pancreatic cancer patients in whom WGS did not reveal copy number alterations also had extra copies of Iq. In 11 of these 18 patients, the majority of nuclei contained at least three Iq fluorescent signals per nucleus (Figures 7 and 8). A subset (n=4) of the discordant tumors could be attributed to low tumor cellularity and the technical limitations reliably identifying copy number changes at low admixtures. Most discordances (n=6), however, stem from the limitation of WGS algorithms identifying gains or losses relative to the basal copy number of all changes rather than absolute copy number.

[0158] Our FISH observations are consistent with conclusions from previous studies of cancers, including pancreatic cancers, showing that whole genome duplication is a common event (Carter et al., Nat Biotechnol 30: 413-421 (2012); Notta et al., Nature 538:378-382 (2016); and Mullen et al., Cancer Discovery 15:329-345 (2025)). To provide further evidence for whole genome duplication in the cancers in our cohort, we evaluated three other probes with FISH, from chromosome Ip, 19p, or 19q. These probes were chosen because they have been extensively used and clinically validated in brain tumors for the detection of the 1 p / 19q translocations typical of oligodendrogliomas (32) (33). We found that 72%, 68%, and 72% of the 37 PDAC samples had at least three copies of chromosomes Ip, 19p, or 19q in at least 20% of their neoplastic nuclei. Similarly, 62%, 62%, and 70% of the 37 PDAC tumors contained at least four copies of chromosomes Ip, 19p, or 19q in at least 10% of their neoplastic nuclei. Gains of Ip, Iq, 19p, or 19q of this magnitude (i.e., the majority of cells) were not found in any of the 11 non-neoplastic control samples (Fig. 7).Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0159] Copy number alterations in the precursors to PDAC.

[0160] The great majority (-90%) of PDAC are believed to develop from PanINs (Pancreatic Intraepithelial Neoplasia) lesions (Pittman et al., Arch Pathol Lab Med 141:1606-1614 (2017); and Basturk et al., Am J Surg Pathol 39:1730-1741 (2015)). These lesions are located in the ducts and can be classified into low-grade and high-grade based on the degree of histologic dysplasia, e.g., their resemblance to normal pancreatic ductal cells, including the size, shape, and position of the cells and the nuclei within the cells. PanINs cannot be detected by imaging because they are <0.5 cm in diameter. Instead, they can only be observed through microscopic evaluation of excised pancreata. There is an average of -1000 low-grade PanINs in individuals older than 65 years, and their danger of progressing to malignancy is close to nil (Braxton et al., Nature 629.6rl9-6?>'l (2024)). In contrast, highgrade PanINs are much less common and are thought to have a relatively high chance of progressing to a PDAC. More than 90% of all PanINs contain mutations in KRAS genes, and the genetic and epigenetic alterations responsible for progression from low-grade PanINs to high-grade PanINs are an active area of investigation (Halbrook et al., Cell 186:1729-1754 (2023); and Hosoda et al., J Pathol 242:6-23 (2017)).

[0161] Because PanINs are so small, microdissection of the lesions was required to obtain a high neoplastic fraction in the tissues used for purifying DNA. As controls, we dissected non-neoplastic regions from 45 sections of pancreas containing low or high-grade pre-neoplastic lesions. None of these 45 contained any chromosome gains or losses. In 65 low-grade PanINs derived from 49 patients, aneuploidy was also infrequent: 81% had no chromosome gains or losses. The most frequently altered chromosome arm in LG PanINs was Iq: four (6%) of the 65 had gained Iq (Fig. 9A). In these four, one had no other gains or losses of any chromosome arm, one had a loss of 6q, another had a gain of chromosome 7 (both p and q arms), and the other (PIN 108 S2) was highly aneuploid, with a total of 26 gains or losses, including losses of 9p, 17p, and 18q. This LG PanlN was adjacent to a HG PanlN, with which it shared the gain of Iq and the losses of 9p, 17p, and 18q, suggesting an evolutionary relationship.

[0162] The evaluation of high-grade PanINs revealed a picture different from either LG PanINs or PDAC (Fig. 9B). The most prevalent altered chromosome arm was again Iq,Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0163] gained in 49% of the 37 lesions (derived from 28 patients) studied (Fig. 9B). In the 18 HG PanINs with gain of Iq, two had no gains or losses of any other chromosome arm, and seven (39%) did not gain 8q or lose 9p, 17p, or 18q. In comparison, only three (1.4%) of 212 PDAC with chromosome Iq gain did not gain 8q or lose 9p, 17p, or 18q.

[0164] Most of the PDAC that don’t originate in PanINs are believed to originate from IPMNs (Intraductal Pancreatic Mucinous Neoplasms) (Halbrook et al., Cell 186:1729-1754 (2023); and Hong et al., Clin Cancer Res 18:4303-4312 (2012)). We were able to evaluate 76 LG IPMNs (derived from 44 patients) via WGS. In these lesions, the most frequent chromosome arm change by far was Iq gain, occurring in 25 (33%) of 76 tumors (Fig. 9C) -considerably higher than in LG PanINs (7%; Fig. 9A). In 12 (48%) of the LG IPMNs with Iq gain, no gains or losses of any other chromosome arm were identified. All 25 (100%) of the LG IPMNs with Iq gain did not lose 9p, 17p, or 18q, and only two (8%) gained 8q.

[0165] We also evaluated 70 HG IPMNs (derived from 21 patients) and found that Iq gains were about as common in these lesions (36%, Fig. 9D) as they were in LG IPMNs (33%, Fig.

[0166] 9C). However, other chromosome gains and losses were much more frequently observed in HG IPMNs than in LG IPMNs. Of 25 HG IPMNs with Iq gain, only one (4%) did not have gains or losses of at least one other chromosome. In the 25 HG PanINs, seven (28%) did not gain 8q or lose 9p, 17p, or 18q - far higher than the 1.4% of PDAC which did not gain or lose these chromosome arms.

[0167] PDAC can also, though uncommonly, originate from Mucinous Cystic Pancreatic Neoplasms (MCNs) (Gonda et al., New England Journal of Medicine 391:832-843 (2024)). These are distinguished from PanINs and IPMNs by epidemiology, histopathology, and the genes most commonly mutated. For example, MCPNs usually occur in middle aged females, by histopathologic definition contain ovarian-type stroma and often contain mutations in RNF43. We were able to evaluate nine dissected MCN lesions (derived from nine patients). None had gains of Iq, though 4 (44%) of the nine had gains or losses of other chromosome arms (Fig. 9E).

[0168] Finally, we evaluated carefully dissected regions from ten serous cystadenomas (SC As) (derived from ten patients), benign tumors that do not progress to malignancy (39). None had gains of Iq, though 5 (50%) of the ten had gains or gains or losses of otherAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0169] chromosome arms (Fig. 9F). All five of the SCAs with gains or losses of other chromosomes had lost 3p - the site of the VHL gene known to be a driver gene in this tumor type (Springer et al., Gastroenterology 149:1501-1510 (2015); and Springer et al., Science Translational Medicine ll:eaav4772 (2019)).

[0170] The WGS results of the eight types of pancreatic samples we studied are graphed in. The average number of chromosome arms gained or lost in these lesions varied from zero to 17. No Iq gains were found in normal tissues, MCPNs, or SCAs. Chromosome Iq gains were found in 7%, 49%, 36%, 33%, and 40% of LG PanINs, HG PanINs, LG IPMNs, HG IPMNs, and PDAC, respectively. In those PDACs that did harbor a Iq gain, nearly all cases also had gains of 8q or losses of chromosomes 9p, 17p, or 18q (99% ). But in HG precursor lesions with Iq gain, additional gains or losses of these four chromosome arms were observed in fewer cases (67%, P<0.0001 by 2-tailed Fisher’s Exact test). In LG precursor lesions with Iq gain, additional gains or losses of these other chromosome arms were even less frequent (10%, PO.OOOl by 2-tailed Fisher’s Exact test).

[0171] Localization of the regions on Iq containing candidate driver genes.

[0172] In heritable diseases, linkage analysis can identify candidate genes that may be responsible for the predisposition to the disease. In cancer, which is mainly driven by somatically acquired mutations, the analog of linkage is the identification of gene(s) that are most frequently associated with the genetic alteration of interest in sporadic tumors. For example, this strategy led to the identification of TP53 as the tumor suppressor gene inactivated in most human cancers. We atempted to apply this strategy to identify candidate driver gene(s) on Iq.

[0173] We began with a well-known target of chromosome gain, i.e., 8q. It is believed that CMYC gene is the target of 8q gains in multiple cancer types. Indeed, 8q is the most frequently gained chromosome arm in PDAC (Fig. 6), though it is not as often gained in PDAC precursors (Fig. 9). Our WGS data showed that 50% (n=124) of the PDAC that gained 8q gained the entire 8q arm, while the remaining 50% (n=123) gained only a subset of 8q genes. Cancers that gained the entire arm obviously cannot be used to identify the presumptive driver gene on that chromosome arm. However, evaluation of the 123 PDACAtorney Docket No. 44807-0508W01 / Pl 8672-02

[0174] that gained only a subset of the genes on the arm can address this question. On 8q, there was a single broad peak centered at 8q24.21, spanning positions 127,300,001 to 131,500,000 (Fig. 10A). This peak includes CMYC (128,747,681-128,755,197). Moreover, 93 % of the cancers that gained all or any part of 8q gained CMYC.

[0175] Analogously proceeding to Iq, the WGS data showed that 139 (66%) of the PDAC that gained Iq gained the entire Iq arm, while the remaining 73 (34%) gained only a subset of Iq genes. Unlike the single major peak on 8q, there were two major peaks on Iq, one centered at lq23.2 (159,100,001 to 160,500,000), and the other centered at 1 q42.13 (227,000,001-230,700,000) (Fig. 10B). 91% (n=192) of the cancers that gained any part of Iq (including the entire Iq) gained the lq23.2 locus and 85% (n=180) of the cancers that gained any part of Iq gained the 1 q42.13. Considering only the 212 PDAC that had gained extra copies of any part of Iq, 77% (n=164) gained both the lq23.2 and lq42.13 loci, 13% (n=28) gained the lq23.21ocus but not the lq42.13 locus, 8% (n=16) gained the lq42.13 locus but not the lq23.2 locus, and 2% (n=4) gained neither. The WGS data on PDAC precursor lesions highlighted the same two regions of Iq, but the number of tumors was much smaller and the localization therefore of lower resolution (Fig. 10C).

[0176] Identification of candidate oncogenes on Iq.

[0177] The critical region on lq23.2 contained 38 transcribed genes, 32 of which encode proteins (Fig. 11 A). To determine the most likely genes responsible for the putative selective growth advantage conferred by this region of Iq, we evaluated the correlation between gene gain and gene expression in The Cancer Genome Atlas (TCGA) database. This analysis only examines the impact of copy number alterations without examining the impact of other genetic or epigenetic alterations on transcription levels. There were 9 genes in which the level of expression and copy number was correlated in a statistically significant fashion (p<le-5; red font in Fig. 11A). Notable among these was NCSTN, encoding a component of the y-secretase complex.

[0178] The second critical region on lq42.13 (chrl: 227,000,001-230,700,000, Fig. 1 IB) contained 61 transcribed genes, 33 of which encode proteins. To determine the most likely genes responsible for the putative selective growth advantage conferred by this region of Iq,Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0179] we similarly evaluated the correlation between gene gain and gene expression in The Cancer Genome Atlas (TCGA) database. There were 14 genes in which the level of expression and copy number was correlated in a highly statistically significant fashion (p<le-5, red font in Fig. 1 IB). Notable among these was PSEN2, encoding another component of the y-secretase complex.

[0180] Relationship of Iq gain to common genetic alterations in PDAC precursor lesions.

[0181] The average depth of the WGS was 121 M unique properly paired reads per sample, corresponding to ~5x coverage, and this was insufficient to reliably evaluate subtle mutations, such as single base substitutions, insertion, or deletions, in the genes driving PDAC development. We therefore designed primers to evaluate 37 amplicons, considering both the Watson and Crick strands, to detect subtle mutations in commonly altered regions of KRAS, NRAS, GNAS, CDK2NA, SMAD4 and TP53 in a highly specific fashion (Materials and Methods). This analysis could be performed on 138 DNA samples from carefully dissected precursor lesions, including 31 low grade PanINs, 24 high-grade PanINs, 43 low-grade IPMNs, and 40 high-grade IPMNs. Mutations in either KRAS or GNAS were identified in all 138 samples (100%), suggesting that mutations in these two genes initiate pancreatic ductal neoplasia. Because all lesions had mutations in these two genes, there could be no correlation between Iq gain and mutations in KRAS or GNAS. But we could evaluate how often subtle mutations in CDKN2, TP53, or SMAD4 occurred. We found that the majority (76%) of the 38 precursor lesions with Iq gain did not have detected mutations in CDKN2, TP53, or SMAD4, consistent with the idea that Iq gains can precede these other genetic alterations during tumor progression. Conversely, the majority (65%) of the 26 precursor lesions in which CDKN2, TP53, or SMAD4 mutations were detected did not harbor Iq gains, consistent with the idea that imbalanced Iq gains are not required for tumor progression.Attorney Docket No. 44807-0508W01 / P18672-02

[0182] Summary of mutations and other chromosomal changes and their relationship to chromosome Iq gain (from WGS data)

[0183]

[0184] Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0185]

[0186] Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0187] Materials and Methods

[0188] Patient Samples

[0189] Sections of frozen or paraffin-formalin-fixed and paraffin-embedded (FFPE) tumor tissues were either macro or micro dissected, and DNA was purified. All samples were reviewed and graded by an expert pancreatic pathologist based on histological features.

[0190] Library construction

[0191] We developed a library preparation workflow that can efficiently recover input DNA and simultaneously incorporate double- stranded molecular barcodes. In brief, libraries were prepared using an Accel-NGS 2S DNA Library Kit (Swift Biosciences, 21024) with the following critical modifications: 1) DNA was pretreated with 3 U of USER enzyme (New England BioLabs, M5505L) for 15 min at 37 °C to excise uracil bases; 2) the SPRI bead / PEG NaCl ratios used after each reaction were 2. Ox, 1.8x, 1.2* and 1.05* for end repair 1, end repair 2, ligation 1 and ligation 2, respectively; 3) a custom 50 pM 3' adapter was substituted for reagent Y2 and 4) a custom 42 pM 5’ adapter was substituted for reagent B2. Libraries were subsequently PCR amplified in 50-pl reactions using primers targeting the ligated adapters. The following reaction conditions were used: U NEBNext Ultra II Q5 Master Mix (New England BioLabs, M0544L), 2 pM universal forward primer and 2 pM universal reverse primer. Libraries were amplified with 8 or 11 cycles of PCR, depending on how many experiments were planned, according to the following protocol: 98 °C for 30 s, cycles of 98 °C for 10 s, 65 °C for 75 s and hold at 4 °C. If eight cycles were used, the libraries were amplified in single 100-pl reactions. If 11 cycles were used, the libraries were divided into eight aliquots and amplified in eight 50-pl reactions, each supplemented with an additional 0.5 U of Q5 Hot Start High-Fidelity DNA Polymerase (New England BioLabs, M0493L), 1 pl of 10 mM dNTPs (New England BioLabs, N0447L) and 0.4 pl of 25 mM MgC12 solution (New England BioLabs, B9021 S). The products were purified with 1 8x SPRI beads (Beckman Coulter, B23317) and eluted in EB buffer (Qiagen).Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0192] Aneuploidy Analysis

[0193] Library DNA was amplified in 50 pL reactions in Ultra Q5 with primers at 2 pM for seven cycles with the following conditions: 98 °C for 30 s, then seven cycles of 98 °C for 10 s to denature, and 65 °C for 75 s to anneal and extend. WGS libraries were sequenced on a NovaSeq 6000 with paired-end 2 x 100 bp reads. Cutadapt was used to trim 27 base pairs from both reads and BWA-MEM was used to align reads to the hgl9 genome. Duplicate molecules were marked and removed using samtools. Reads with a bowtie2 alignment quality score >10 were binned into 500kb intervals and counted. IchorCNA was then used to perform GC correction, and segmentation using the following parameters: “ — chrs “c(l, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,18,19,20,21,22) -normalPanel ” Two normal panels were generated — one for fresh frozen samples (n=9) and one for formal-fixed paraffin-embedded (FFPE) samples (n=5). As recommended in the ichorCNA best practices on github, only samples with a GC-Correction MAD <0.15 were used in this study.

[0194] QDNAseq was also used to perform GC correction, segmentation and calling of copy number alterations throughout the genome. Samples that have much larger observed variance than the expected variance may suggest a sample with poor quality. A full set of files (log ratios, segmentation files, plots, .RDS) for both ichorCNA and QDNAseq analyses are publicly available at zenodo.org / records / 17080198 (DOI: 10.5281 / zenodo.17080197). In our analysis, we defined a full arm gain by counting the number of 500kb intervals on a particular chromosome arm with a predicted copy number > 2. If the number of gained intervals was greater than or equal the number of 500kb intervals on the chromosome arm - 3 intervals, it was considered a full arm gain (e.g. no more than 3 non-gained intervals were permitted on the arm to be considered a full arm gain). The procedure was also done for full arm losses using a predicted value of < 2.

[0195] Mutation Analysis

[0196] Following library creation as described above, two separate PCRs were designed to selectively enrich the Watson or Crick strand. Both PCRs used the same gene-specific primer, but each used a different anchoring primer. PCR duplicates derived from each strand could be distinguished by the orientation of the insert relative to the exogenous UID.Atorney Docket No. 44807-0508W01 / Pl 8672-02

[0197] Sequencing reads underwent initial processing by extracting the first 14 nucleotides as the exogenous barcode sequence (UIDs) and masking adapter sequencing Picard’s IlluminaBasecallsToSam (broadinstitute.github.io / picard). Reads were then mapped to the hgl9 reference genome using BWA-MEM and sorted by barcode sequence using Samtools. Duplex mutations were defined as mutations present in >80% of both the Watson and Crick families with the same UID. We used the following metrics for the interpretation of mutations: only samples in which 10 KRAS template molecules were amplified were assessed; only mutant positions with more than two mutant template molecules were considered as bona fide. Only genomic positions with at least 2 or more reported annotations in the Catalogue of Somatic Mutations in Cancer (COSMIC) in genome-wide studies and were confirmed somatic mutations were considered as bona fide; only positions with at least 5x coverage were considered and only positions at least 30 bp away from the end of the molecule were considered

[0198] Fluorescence in situ hybridization (FISH)

[0199] A commercially available probe set purchased from Abbot Molecular, comprised of LSI lp36 / TP73 (SpectrumOrange) and LSI Iq25 / ABL2 (SpectrumGreen), was applied to individual slides, hybridized, and washed according to the TRIS / EDTA FISH protocol. This process was repeated using a commercially available probe set purchased by Abbott Molecular comprised of LSI 19ql3 / EHD2 (SpectrumOrange) and 19pl3 / D19S221 (SpectrumGreen) on adjacent slides. In each case, 50 to 100 nuclei from the cancer cells within the section were evaluated in a blinded fashion using two independent analysts.

[0200] Statistical Analysis

[0201] Cohort samples size was not selected for statistical power but based on sample availability. The two-tailed Fisher’s Exact test (alpha=0.05) was used to test the association with chromosome Iq gain and the mutation status of various genes.Attorney Docket No. 44807-0508W01 / P18672-02

[0202] Example 3: Treating PD AC

[0203] A human identified as having PDAC is administered one or more gamma-secretase inhibitors (e.g., nirogacestat (e.g., OGSIVEO®)) by intratumoral injection. The administered inhibitor(s) can reduce the number of cancer cells within the human.

[0204] Example 4: Treating PDAC

[0205] A human identified as having PDAC is administered one or more inhibitors of a polypeptide present in a gamma-secretase polypeptide complex (e.g., a PSEN1 polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, an APH-1 polypeptide, or a PEN-2 polypeptide) by intratumoral injection. The administered inhibitor(s) can reduce the number of cancer cells within the human.

[0206] OTHER EMBODIMENTS

[0207] It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

Attorney Docket No. 44807-0508W01 / P18672-02WHAT IS CLAIMED IS:

1. A method for identifying a mammal as having pancreatic cancer, wherein said method comprises detecting the presence of an increased copy number of one or more of a presenilin-1 (PSEN1) gene, a PSEN2 gene, a nicastrin (NCSTN) gene, an anterior pharynx-defective 1 (APH1) A gene, and a APH1B gene in a sample obtained from said mammal.

2. The method of claim 2, wherein said mammal is a human.

3. The method of any one of claims 1-2, wherein said pancreatic cancer is a pancreatic ductal adenocarcinoma (PDAC).

4. A method for identifying a mammal as being likely to develop a pancreatic cancer, wherein said method comprises detecting the presence of an increased copy number of one or more of a PSEN1 gene, a PSEN2 gene, a NCSTN gene, an APH1 A gene, and a APH1B gene in a sample obtained from said mammal.

5. The method of claim 4, wherein said mammal is a human.

6. The method of any one of claims 4-5, wherein said pancreatic cancer is a PDAC.

7. The method of any one of claims 4-6, wherein said mammal has a pancreatic intraductal papillary mucinous neoplasm (IPMN) or a mucinous cystic pancreatic neoplasm (MCPN).

8. A method for treating a mammal having pancreatic cancer, wherein said method comprises administering a gamma secretase inhibitor and / or an antibody that targets a substrate of a gamma-secretase polypeptide complex to said mammal.

9. The method of claim 8, wherein said mammal is a human.Attorney Docket No. 44807-0508W01 / P18672-0210. The method of any one of claims 8-9, wherein said pancreatic cancer is a PDAC.

11. The method of any one of claims 8-9, wherein said method comprises identifying said pancreatic cancer as comprising one or more cancer cells having an increased copy number of one or more of a PSEN1 gene, a PSEN2 gene, a NCSTN gene, an APEI1 A gene, and a APH1B gene.

12. The method of any one of claims 8-11, wherein said gamma secretase inhibitor is nirogacestat.

13. The method of any one of claims 8-11, wherein said gamma secretase inhibitor targets a presenilin-1 (PSEN1) polypeptide, a PSEN2 polypeptide, a nicastrin polypeptide, and / or an anterior pharynx-defective 1 (APH-1) polypeptide.

14. The method of claim 13, wherein said gamma secretase inhibitor targets said PSEN1 polypeptide, and wherein said gamma secretase inhibitor is MRK-560, ELN-318463, ELN-475516, SCH-900229, SCH-1500022, avagacestat, begacestat, DAPT, aDAPT analog, L-685-458, TSAI-1, LY-411575, RO-4929097, semagacestat, Compound 34, and PF-3084014, or a combination thereof.

15. The method of claim 13, wherein said gamma secretase inhibitor targets said PSEN2 polypeptide.

16. The method of claim 13, wherein said gamma secretase inhibitor targets said nicastrin polypeptide, and wherein said gamma secretase inhibitor is semagacestat, MK-0752, E2012, BMS-708163, PF-03084014, MK-0725, or DAPT.

17. The method of claim 13, wherein said gamma secretase inhibitor targets said APH-1 polypeptide.Attorney Docket No. 44807-0508W01 / P18672-0218. The method of claim 13, wherein said gamma secretase inhibitor targets said PEN-2 polypeptide.

19. The method of any one of claims 8-11, wherein said substrate of said gamma-secretase polypeptide complex is selected from the group consisting of a MUC-1 polypeptide, a NOTCH polypeptide, an E-cadherin polypeptide, and an amyloid precursor polypeptide.

20. The method of any one of claims 8-11, wherein said antibody is selected from the group consisting of HMFG1, PankoMab, VU-2G7, KL-6, 5E5, 139H2, D1E11, D67C8, D76A6, 8A1, D11B8, L5C5, Val744, DECMA-1, NCH-38, HECD-1, ECCD-1, 67A4, 5H6L18, 5HCLC, PA5-32178, PA5-81393, PA5-19479, Y188, Y188, EPR5119(2), EPR5118-34, mAbP2-l, 22C11, LN27, A8717, MAB348, MOAB-2, and 25524-1-AP.

21. The use of a composition comprising a gamma-secretase inhibitor to treat a mammal having pancreatic cancer.

22. The use of claim 21 , wherein said mammal is a human.

23. The use of any one of claims 21-22, wherein said pancreatic cancer is a PDAC.

24. A composition comprising a gamma-secretase inhibitor for use in the preparation of a medicament to treat a mammal having pancreatic cancer.

25. A composition comprising a gamma-secretase inhibitor for use in the treatment of a mammal having pancreatic cancer.

26. The composition of any one of claims 24-25, wherein said mammal is a human.Attorney Docket No. 44807-0508W01 / P18672-0227. The composition of any one of claims 24-26, wherein said pancreatic cancer is a PDAC.

Images