Biomarker for determining prognosis of ovarian cancer, determination method, and therapeutic agent

Abstract

To provide a biomarker for determining the prognosis of ovarian cancer, a method for determination, and a therapeutic agent capable of improving the prognosis of refractory ovarian cancer.SOLUTION: The present invention provides a biomarker for determining the prognosis of ovarian cancer, which comprises at least one of p62 protein and p62 mRNA.SELECTED DRAWING: Figure 7E

Description

[Technical field]

[0001] The present invention relates to a biomarker, a method for determining the prognosis of ovarian cancer, and a therapeutic agent. [Background technology]

[0002] High-grade serous carcinoma (HGSC) is the most common histological type of epithelial ovarian cancer. It is difficult to detect early and is often diagnosed as advanced cancer, and has a poor prognosis due to its high recurrence rate. For a long time, the standard treatment for patients with advanced epithelial ovarian cancer was a combination of tumor-reducing surgery and chemotherapy with carboplatin (CBDCA) and paclitaxel (PTX), but optimal chemotherapy according to the genetic profile of each patient has gradually developed. BRCA1 / 2 gene mutations contribute to the oncogenesis of ovarian cancer by causing genomic instability through their involvement in DNA double-strand break repair, but chemotherapy with platinum agents and PARP inhibitors shows strong antitumor effects in HGSC patients with BRCA1 / 2 gene mutations by converting genomic instability into a weakness through a synthetic lethal mechanism.

[0003] In fact, the Cancer Genome Atlas (TCGA) cohort study showed that BRCA1 / 2 pathogenic mutations in HGSC patients result in better overall survival (OS) (see, for example, Non-Patent Document 1), and it has been reported that the PARP inhibitor olaparib extended progression-free survival (PFS) and overall survival (OS) the most in HGSC patients with BRCA1 / 2 gene mutations, who are highly sensitive to platinum agents (see, for example, Non-Patent Document 2).

[0004] However, there are no promising treatment options for "refractory" HGSC patients, whose tumors continue to progress despite chemotherapy with platinum agents or poly(ADP-ribose) polymerase (PARP) inhibitors, and the prognosis of these patients is extremely poor. HGSC patients without BRCA1 / 2 gene mutations (BRCA1 / 2 wild-type) may be refractory cases, and in fact, a whole-genome analysis of HGSC patients showed that all refractory HGSC cases in a cohort did not have BRCA1 / 2 pathogenic mutations. Some reports have pointed out that the causes of refractory HGSC in BRCA1 / 2 wild-type HGSC include the absence of DNA double-strand break repair deficiency, low immunogenicity, and high intratumor heterogeneity. However, no new treatments or biomarkers have been developed to improve the prognosis of BRCA1 / 2 wild-type HGSC, and the development of new treatments is urgently needed. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Bell D et al (2011) Nature 474:609-615 [Non-Patent Document 2] Moore K et al (2018) N Engl J Med 379:2495-2505 Summary of the Invention [Problem to be solved by the invention]

[0006] If we could detect poor prognostic factors for ovarian cancer and develop some kind of targeted therapy, this would have a significant impact on the survival of ovarian cancer patients, including those with refractory HGSC.

[0007] The present invention aims to solve the above-mentioned conventional problems and achieve the following objectives: That is, the present invention aims to provide a biomarker for determining the prognosis of ovarian cancer, a method for determining the prognosis, and a therapeutic drug that can improve the prognosis of refractory ovarian cancer. [Means for solving the problem]

[0008] The means for solving the above problems are as follows. <1> A biomarker for determining the prognosis of ovarian cancer, comprising: The biomarker is characterized by being at least one of p62 protein and p62 mRNA. <2> A method for determining the prognosis of ovarian cancer, comprising: Detecting the presence or amount of a biomarker, p62 protein and / or p62 mRNA, in a sample obtained from the subject; determining that there is a risk of ovarian cancer if the biomarker is present; The method is characterized by comprising: <3> The specimen is at least one of a pathological tissue specimen of ovarian cancer and a blood specimen. <2> This is the determination method described in. <4> the presence or amount of said biomarker in said sample from a particular subject; comparing the presence or amount of the biomarker in the sample obtained from the subject after a period of time; From the above comparison, If the biomarker is no longer detectable or the amount is reduced after the passage of the certain period of time, the prognosis is good. If there is no change in the amount of the biomarker after the period of time, there is no change in prognosis; or determining that the prognosis is poor when the biomarker becomes detectable or increases in amount after the specified time has elapsed; The above further comprises <2> or <3> This is the determination method described in. <5> A therapeutic agent for treating ovarian cancer, comprising: This is a therapeutic drug characterized by containing an mTOR inhibitor. <6> The method according to claim 1, wherein the mTOR inhibitor is everolimus. <5> It is a therapeutic drug described in the above. <7> The ovarian cancer is an ovarian cancer in which the biomarker according to claim 1 is present. <5> or <6> It is a therapeutic drug described in the above. <8> The method according to any one of claims 1 to 4, wherein the ovarian cancer is platinum-insensitive ovarian cancer. <5> from <7> The therapeutic agent according to any one of the above claims. <9> The method according to claim 1, wherein the ovarian cancer is a wild-type ovarian cancer of the BRCA1 / 2 gene. <5> from <8> The therapeutic agent according to any one of the above claims. <10> The above in combination with paclitaxel, carboplatin and / or a PARP inhibitor. <5> from <9> The therapeutic agent according to any one of the above claims. Effect of the Invention

[0009] According to the present invention, it is possible to solve the above-mentioned problems in the conventional art and achieve the above-mentioned object, thereby providing a biomarker for determining the prognosis of ovarian cancer, a method for determining the prognosis, and an ovarian cancer therapeutic agent that can improve the prognosis of difficult-to-treat ovarian cancer. [Brief description of the drawings]

[0010] [Figure 1A] FIG. 1A is a schematic illustrating the computational approach using the R package “GSVA” (gene set variation analysis) and the transcriptome dataset of TCGA advanced ovarian cancer cases. [Figure 1B] FIG. 1B is a heat map showing the correlation between GSVA score and prognosis in all advanced ovarian cancer patients. [Figure 1C] FIG. 1C is a heat map showing the correlation between GSVA score and prognosis in patients with advanced ovarian cancer without BRCA1 / 2 mutations. [Figure 1D] Figure 1D is a Kaplan-Meier survival curve showing the prognosis of patients with advanced ovarian cancer without BRCA 1 / 2 mutations according to the difference in "RAS protein signaling" score. [Figure 2A] FIG. 2A is an immunofluorescence image of normal oviductal epithelium of a nude mouse. [Figure 2B]FIG. 2B shows bright field images of normal fallopian tube epithelial organoids (cas-nFTE) and Trp53 knockout organoids (casP). [Figure 2C] Figure 2C shows RNA-seq analysis of cellular components of casP organoids compared to cas-nFTE organoids. [Figure 2D] FIG. 2D is a schematic diagram showing the procedure for establishing mouse HGSC model organoids. [Figure 2E] FIG. 2E shows a bright field image of the established organoid spheres and a pathological image of the tumor formed by subcutaneously implanting RPM cells into nude mice. [Figure 2F] FIG. 2F shows the results of Western blotting of RPM, RPMN, RPMP and RPMNP cells. [Figure 2G] FIG. 2G shows unsupervised hierarchical clustering of RNA-seq expression data from nFTE, RPM, RPMN, RPMP, and RPMNP organoids. [Figure 2H] FIG. 2H is a bar graph showing mutational signatures from SBS96 of RPM, RPMN, RPMP, and RPMNP cells. [Figure 3A] FIG. 3A is a Kaplan-Meier survival curve of recipient mice inoculated intraperitoneally with RPM, RPMN, RPMP, or RPMNP cells. [Figure 3B] FIG. 3B shows dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of carboplatin (CBDCA). [Figure 3C] FIG. 3C shows dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of paclitaxel (PTX). [Figure 3D] FIG. 3D shows dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of olaparib. [Figure 3E] FIG. 3E is a graph showing the results of RNA-seq analysis of the MSigDB hallmark gene set that is enriched in RPMNP cells compared to RPM cells. [Figure 3F] FIG. 3F is a graph showing the enrichment plot of mTORC1 signaling in the MSigDB hallmark gene set. [Figure 3G] FIG. 3G is a graph showing the enrichment plot of endoplasmic reticulum stress response in the MSigDB hallmark gene set. [Figure 4A] FIG. 4A shows viable cell staining images of RPMNP cells treated with CBDCA+0.1% DMSO and RPMNP cells treated with CBDCA+100 nM everolimus. [Figure 4B] FIG. 4B is a graph showing dose-response curves of RPMNP cells treated with CBDCA+0.1% DMSO and RPMNP cells treated with CBDCA+100 nM everolimus (EVL). [Figure 4C] FIG. 4C is a graph showing dose-response quantification of the BRCA1 / 2 wild-type human ovarian cancer cell line, SKOV3. [Figure 4D] FIG. 4D is a graph showing dose-response quantification of the BRCA1 / 2 wild-type human ovarian cancer cell line, Caov3. [Figure 4E] FIG. 4E is a graph showing a proteomic expression comparison of RPM incubated with DMSO and RPMNP incubated with DMSO. [Figure 4F] Figure 4F is a dot plot of proteins whose expression was decreased by administration of an mTOR inhibitor among the proteins whose expression was increased in RPMNPs by comparing the proteome expression of RPM cultured in DMSO and RPMNPs cultured in DMSO. [Figure 4G] FIG. 4G shows Western blotting images detecting p62, S6, and pS6 proteins in RPM, RPMN, RPMP, and RPMNP cells. [Figure 4H] FIG. 4H is a diagram showing viable cell staining images of p62 knockout RPMNP cells (RPMNP-p62KO) treated with CBDCA and RPMNP cells treated with CBDCA. [Figure 4I]FIG. 4I is a graph showing dose-response curves of p62 knockout RPMNP cells treated with CBDCA (RPMNP-p62KO) and RPMNP cells treated with CBDCA. [Figure 4J] FIG. 4J is a graph showing dose-response quantification of RPM cells overexpressing p62 (RPM-p62OE) and RPM cells overexpressing EGFP (EGFP) as a control when treated with CBDCA. [Figure 4K] FIG. 4K shows Western blotting images detecting p62, LC-3B, and pS6 proteins in RPMNP cells treated with vehicle alone, everolimus alone, and both everolimus and the autophagy inhibitor hydroxychloroquine. [Figure 5A] FIG. 5A is a schematic diagram showing the procedure for obtaining and following up 15 pairs of human HGSC clinical samples before (PreNAC) and after (PostNAC) chemotherapy. [Figure 5B] FIG. 5B shows multiplex immunofluorescence staining of a BRCA mutant case before (left) and after (right) chemotherapy. [Figure 5C] FIG. 5C shows multiplex immunofluorescence staining of a BRCA wild-type case before (left) and after (right) chemotherapy. [Figure 5D] FIG. 5D is a graph showing the rate of increase in p62 fluorescence intensity per area in tumor tissues after chemotherapy. [Figure 6A] FIG. 6A is a graph showing the RNA expression levels of Ho-1 in RPM, RPM-p62OE, RPMNP, and RMNP-p62KO cells. [Figure 6B] FIG. 6B is a graph showing the RNA expression levels of Nqo1 in RPM, RPM-p62OE, RPMNP, and RMNP-p62KO cells. [Figure 6C] FIG. 6C is a graph showing the RNA expression levels of Ho-1 and Nqo1 in RPMNP cells administered with solvent or 1 μM everolimus. [Figure 6D]FIG. 6D is a Kaplan-Meier curve showing Ho-1(bK286B10) RNA expression and overall survival (OS) of ovarian cancer patients. [Figure 6E] FIG. 6E is a Kaplan-Meier curve showing progression-free survival (PFS) of Ho-1(bK286B10) RNA expression and prognosis of ovarian cancer patients. [Figure 6F] FIG. 6F is a Kaplan-Meier curve showing NQO1 RNA expression and overall survival (OS) of ovarian cancer patients. [Figure 6G] FIG. 6G is a Kaplan-Meier curve showing progression-free survival (PFS) of NQO1 RNA expression and prognosis of ovarian cancer patients. [Figure 7A] FIG. 7A shows viable cell staining images of RPMNP cells treated with TC treatment+0.1% DMSO and RPMNP cells treated with TC treatment+100 nM everolimus. [Figure 7B] FIG. 7B is a graph showing dose-response curves of RPMNP cells treated with TC+0.1% DMSO and RPMNP cells treated with TC+100 nM everolimus (EVL). [Figure 7C] FIG. 7C shows the treatment regimens of the four groups. [Figure 7D] FIG. 7D is a graph showing tumor growth curves following administration of vehicle alone, TC, everolimus (EVL), and TC plus everolimus (EVL). [Figure 7E] FIG. 7E is a model diagram showing the mechanism of action of the p62 biomarker and an mTOR inhibitor in refractory ovarian cancer. [Figure 8A] FIG. 8A is a Kaplan-Meier survival curve showing the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 gene mutation according to differences in the "PI3K-associated signal" score (PI3K score). [Figure 8B] FIG. 8B is a scatter plot showing the correlation between RAS and PI3K scores in patients with advanced ovarian cancer without BRCA1 / 2 mutations. [Figure 9A] FIG. 9A shows immunohistochemical images of normal mouse fallopian tube epithelium stained with PAX8 antibody and α-tubulin antibody. [Figure 9B] FIG. 9B shows Sanger sequencing of the Trp53 locus in casP organoids. [Figure 9C] FIG. 9C is a graph showing RNA-seq analysis of cellular components of casP organoids compared to cas-nFTE organoids. [Figure 9D] FIG. 9D shows whole mount immunofluorescence images of nFTE organoids stained with PAX8, α-tubulin, and DAPI. [Figure 9E] FIG. 9E shows representative immunohistochemical images of nFTE and mFTE organoids stained with MKI67 antibody. [Figure 9F] Figure 9F is a schematic diagram showing the mutation information of Nf1 and Pten genes in RPM, RPMN, RPMP, and RPMNP organoids. [Figure 10] FIG. 10 is a graph showing enrichment plots of the autophagosome maturation pathway analyzed by Preranked GSEA examining enrichment of Biological Process Ontology gene sets in RPMNP cells compared to RPM cells. [Figure 11A] FIG. 11A is a graph showing dose-response curves of RPM, RPMN, RPMP, and RPMNP cells with EVL treatment. [Figure 11B] FIG. 11B shows a stained well image of RPMNP cells treated with EVL for 7 days. [Figure 11C] FIG. 11C shows Western blotting images detecting p62 and EGFP proteins in RPM, RPM-p62OE, RPMNP, and RMNP-p62KO cells. [Figure 11D] FIG. 11D shows viable cell staining images of RMNP-p62KO cells and RPMNP cells treated with CBDCA + PTX (TC). [Figure 11E] FIG. 11E is a graph showing the dose-response curves of RMNP-p62KO cells and RPMNP cells treated with CBDCA + PTX (TC). [Figure 12A] FIG. 12A is a Kaplan-Meier survival curve showing the “cellular response to reactive oxygen species” score calculated using GSVA and the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 mutations. [Figure 12B] FIG. 12B is a scatter plot showing the correlation between ROS scores and RAS scores in patients with advanced ovarian cancer without BRCA1 / 2 mutations. [Figure 12C] FIG. 12C is a scatter plot showing the correlation between ROS and PI3K scores in patients with advanced ovarian cancer without BRCA1 / 2 mutations. [Figure 12D] FIG. 12D is a Kaplan-Meier curve showing GPX4 RNA expression and overall survival (OS) of ovarian cancer patients. [Figure 12E] FIG. 12E is a Kaplan-Meier curve showing progression-free survival (PFS) of GPX4 RNA expression and prognosis of ovarian cancer patients. [Figure 12F] FIG. 12F is a Kaplan-Meier curve showing CAT RNA expression and overall survival (OS) of ovarian cancer patients. [Figure 12G] FIG. 12G is a Kaplan-Meier curve showing progression-free survival (PFS) of CAT RNA expression and prognosis of ovarian cancer patients. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] (Biomarkers) The biomarker of the present invention is a biomarker for determining the prognosis of ovarian cancer, and is at least one of p62 protein and p62 mRNA.

[0012] (Judgment method) The determination method of the present invention is a method for determining the prognosis of ovarian cancer, and includes the steps of detecting the presence or amount of a biomarker, which is at least one of p62 protein and p62 mRNA, in a sample obtained from a subject, and determining that there is a risk of ovarian cancer if the biomarker is present.

[0013] In order to solve the above object, the inventors conducted intensive studies. As a result, from the results of basic research using clinical specimens and ovarian cancer mouse models, in ovarian cancer with wild-type BRCA1 / 2 genes, after administration of platinum-based anticancer drugs, the mTOR pathway is activated, and p62, an autophagy marker, is highly expressed, resulting in acquisition of resistance to chemotherapy. Further, using a mouse model, it was found that when an mTOR inhibitor is co-administered with the current standard treatment, TC therapy, the tumors of ovarian cancer with wild-type BRCA1 / 2 genes are significantly reduced, and the combined effect with the anticancer drug is high. Therefore, it was found that the expression of p62 protein or mRNA can be evaluated for advanced ovarian cancer with poor prognosis, and it can be used as a biomarker for predicting the usefulness and therapeutic effect of the combined therapy of TC therapy and an mTOR inhibitor, leading to the completion of the present invention.

[0014] According to the biomarker, determination method, and therapeutic agent of the present invention, it is possible to provide a biomarker for determining the prognosis of ovarian cancer, a determination method, and an ovarian cancer therapeutic agent capable of improving the prognosis of refractory ovarian cancer.

[0015] For ovarian cancer, in addition to TC therapy that combines paclitaxel and carboplatin, in recent years, as molecular target drugs, PARP inhibitors or angiogenesis inhibitors have been used. However, among ovarian cancers, ovarian cancers with low platinum sensitivity including wild-type BRCA1 / 2 genes are known as ovarian cancers with poor prognosis. These advanced ovarian cancers with poor prognosis have limited effects of conventional TC therapy and the like, still have a high recurrence rate and are lethal diseases, have high unmet medical needs, and there is a demand for the development of new treatment methods. Therefore, a biomarker capable of determining the prognosis of ovarian cancer and a new treatment method for suppressing recurrence have great clinical significance.

[0016] <p62 protein and p62 mRNA> The p62 protein is also called ubiquitin-binding protein p62 or sequestosome-1, and is a protein encoded by the SQSTM1 gene in humans. The p62 protein is an autophagosome cargo protein that binds other proteins targeted for selective autophagy, and by binding to GATA4 and targeting it for degradation, it can inhibit GATA-4-associated senescence and senescence-associated secretory events. mTOR is involved in stabilizing p62 by suppressing autophagy.

[0017] In one embodiment, p62 has the base sequence shown in SEQ ID NO: 1 (Genbank accession ID: NM_003900) and the amino acid sequence shown in SEQ ID NO: 2 (Genbank accession ID: NP_003891), as registered in Genbank accession ID: NG_011342. In addition, multiple p62 protein mutants have been reported. For example, mutants registered as the same protein in UniProt (https: / / www.uniprot.org / ), such as Genbank accession ID: NP_001135770, and genes and mRNAs encoding these are also included in p62 of this embodiment.

[0018] The p62 protein, which is a biomarker in this embodiment, is not particularly limited as long as it is a protein that has been transcribed from the locus of the SQSTM1 gene, translated, and optionally post-translationally modified, and can be appropriately selected depending on the purpose. However, the amino acid sequence of p62 is preferably at least 90% identical to the amino acid sequence of SEQ ID NO: 2 (NP_003891), and more preferably at least 95% identical. The antibody capable of detecting the p62 protein is not particularly limited and can be appropriately selected depending on the purpose. For example, anti-p62 antibodies include PM045 (Sigma-Aldrich) and ab56416 (Abcam).

[0019] The p62 mRNA, which is the biomarker of this embodiment, is not particularly limited as long as it is a protein transcribed and spliced ​​from the locus of the SQSTM1 gene and can be appropriately selected depending on the purpose. However, the base sequence of p62 is preferably 90% or more identical to the base sequence of SEQ ID NO: 1 (Genbank accession ID: NM_003900) (or the sequence of its CDR region), and more preferably 95% or more identical. The hybridization probe and PCR primer capable of detecting the p62 mRNA are not particularly limited and can be appropriately selected depending on the purpose. The PCR primers are not particularly limited and can be appropriately selected depending on the purpose. For example, a primer set of p62 / SQSTM1 forward primer: AGGATGGGGACTTGGTTGC (SEQ ID NO: 3) and p62 / SQSTM1 reverse primer: TCACAGATCACATTGGGGTGC (SEQ ID NO: 4) can be mentioned.

[0020] As described in the Examples below, it has been found that the presence and amount of p62 protein or p62 mRNA in a specimen correlates with the prognosis of progressive ovarian cancer with poor prognosis, and therefore the prognosis of ovarian cancer can be determined by detecting the presence and amount of p62 protein or p62 mRNA in a specimen. Therefore, p62 can be used as a biomarker for determining the prognosis of ovarian cancer.

[0021] <Ovarian cancer> The ovarian cancer is not particularly limited and can be appropriately selected depending on the purpose, but is preferably at least one of ovarian cancer with low platinum sensitivity and BRCA1 / 2 gene wild-type ovarian cancer. Here, "platinum-sensitive" ovarian cancer refers to ovarian cancer in which the recurrence period is 6 months or more after the end of platinum-based anticancer drug treatment, which is the standard treatment usually consisting of a combination of carboplatin and paclitaxel, whereas "platinum-insensitive ovarian cancer" refers to ovarian cancer in which the recurrence period is less than 6 months.

[0022] -BRCA1 gene- BRCA1 (breast cancer susceptibility gene I) is a tumor suppressor gene that encodes a nuclear phosphorylated protein that maintains genome stability. It is known that mutations in the BRCA1 gene cause genetic instability, ultimately resulting in breast cancer or ovarian cancer (hereditary breast and ovarian cancer syndrome). Cases involving BRCA1 gene mutations are involved in approximately 40% of hereditary breast cancers and more than 80% of hereditary ovarian and breast cancers. The BRCA1 protein, a transcription product of BRCA1, forms a large complex in the nucleus with numerous other tumor suppressor factors and is involved in the homologous recombination repair mechanism.

[0023] One embodiment of wild-type BRCA1 has the base sequence shown in SEQ ID NO: 5 (Genbank accession ID: NM_007294) and the amino acid sequence shown in SEQ ID NO: 6 (Genbank accession ID: NP_009225), as registered in Genbank accession ID: NG_005905. In addition, multiple BRCA1 protein mutants have been reported, and for example, mutants registered as the same protein in UniProt, such as Genbank accession ID: NM_007300, and the genes encoding these are also included in the BRCA1 of this embodiment.

[0024] -BRCA2 gene- BRCA2 (breast cancer susceptibility gene II) is a type of tumor suppressor gene, and the gene product (protein) of BRCA2 is involved in DNA repair of double-strand breaks via the homologous recombination pathway. It is known that mutations in the BRCA2 gene cause genetic instability, making breast cancer and ovarian cancer more likely to occur, and that it accounts for a significant proportion of early-onset prostate cancer in men.

[0025] One embodiment of wild-type BRCA2 has the base sequence shown in SEQ ID NO: 7 (Genbank accession ID: NM_000059) and the amino acid sequence shown in SEQ ID NO: 8 (Genbank accession ID: NP_000050), as registered in Genbank accession ID: NG_012772. In addition, multiple BRCA2 protein mutants have been reported, and for example, mutants registered as the same protein in UniProt, such as Genbank accession ID: NM_001406720, and the genes encoding these are also included in BRCA2 in this embodiment.

[0026] Each step in the determination method of this embodiment will be described below. <Detection process> The detection step is a step of detecting the presence or amount of a biomarker, which is at least one of p62 protein and p62 mRNA, in a sample obtained from a subject.

[0027] The subject is generally a mammal, such as a human, a non-human primate, a dog, a cat, a mouse, a rat, a cow, a horse, a pig, etc. Among these, a human is preferable. The subject may be any of adults, infants, children, and elderly subjects, and is preferably a subject suspected of having ovarian cancer, a subject diagnosed or confirmed to have a disease associated with ovarian cancer, a subject receiving therapeutic intervention for a disease associated with ovarian cancer, etc.

[0028] The specimen is not particularly limited and can be appropriately selected depending on the purpose, but is preferably at least one of a pathological tissue specimen of ovarian cancer and a blood specimen.

[0029] The method for the detection is not particularly limited, and any known method can be appropriately selected depending on the purpose. When the biomarker is p62 protein, for example, Western blotting, immunohistochemistry (IHC), immunofluorescence (IF), immunochromatography, enzyme-linked immunosorbent assay (ELISA), chemiluminescence enzyme immunoassay (CLEIA), chemiluminescence immunoassay (CLIA), fluorescent enzyme immunoassay (e.g., Luminex TM Examples of such methods include the xMAPTM system (Medical and Biological Laboratories Co., Ltd.), the aptamer method, immunoturbidimetric method, dye colorimetric method, immunonephelometric method, latex agglutination method, the Jaffe method, and gold colloid colorimetric method. When the biomarker is p62 mRNA, examples of the biomarker include in situ hybridization, quantitative reverse transcription polymerase chain reaction (RT-qPCR), and spatial transcriptome analysis.

[0030] The detection can detect or measure the presence, absence, amount, or effective amount, eg, the concentration level, of the biomarker in a sample from a subject.

[0031] <Judgment process> The determining step is a step of determining that there is a risk of ovarian cancer when the biomarker is present. The detection limit of the presence or amount of the biomarker can be set based on the detection limit of the detection method used, and a reference value or cutoff value may be set as necessary. The reference value for the amount of the biomarker can be appropriately selected depending on the purpose, for example, 1.5 times or more, 2 times or more, based on the amount of the biomarker in a non-ovarian cancer subject. "Risk of ovarian cancer" means, for example, the risk of ovarian cancer with low platinum sensitivity, the risk of ovarian cancer with poor prognosis, and the like.

[0032] [How to determine prognosis] In one embodiment of the method, prognosis in a specific subject can be determined. In this embodiment, the determination method further includes a comparison step and a prognosis determination step in addition to the detection step and determination step.

[0033] <Comparison process> The comparing step is a step of comparing the presence or amount of the biomarker in a sample obtained from a particular subject with the presence or amount of the biomarker in a sample obtained from the subject after a certain period of time has passed. The amount of the biomarker may be an absolute value or a relative value, as long as it is a comparable value. The certain period of time is not particularly limited and can be appropriately selected depending on the purpose, and may be, for example, on the order of seconds, minutes, hours, days, months, or years. For example, for follow-up observation after therapeutic intervention, a detection result before treatment may be compared with a detection result after treatment after a certain period of time has elapsed from before treatment, or a determination method may be performed periodically, such as in a regular medical checkup.

[0034] <Prognosis assessment step> The prognosis determining step is carried out by the comparison, (1) If the biomarker is no longer detectable or the amount is reduced after the passage of the certain period of time, the prognosis is good; (2) if there is no change in the amount of the biomarker after the given period of time, there is no change in the prognosis; or (3) determining that the prognosis is poor if the biomarker becomes detectable or the amount thereof increases after the lapse of the certain period of time. This makes it possible to determine whether or not ovarian cancer has low platinum sensitivity without the need to monitor the therapeutic effects of chemotherapy using platinum agents over the long term, and to assess the therapeutic effects and monitor the progress of the disease.

[0035] (Therapeutic drug) The therapeutic agent of the present invention is a therapeutic agent for treating ovarian cancer and contains an mTOR inhibitor.

[0036] (Treatment method) The treatment method of this embodiment is a method for treating ovarian cancer, and includes administering the therapeutic agent to a subject.

[0037] The mTOR inhibitor is not particularly limited and can be appropriately selected depending on the purpose, and examples thereof include everolimus, sunitinib, sirolimus, temsirolimus, vistusertib, etc. These may be used alone or in combination of two or more. Among these, everolimus is preferred.

[0038] The therapeutic agent may consist of an mTOR inhibitor, may contain other drugs, and may contain a pharma- ceutically acceptable carrier. The therapeutic agent may be used in combination with other drugs, and from the viewpoint of preventing recurrence of ovarian cancer with low platinum sensitivity or improving prognosis, it is preferable to use at least one of paclitaxel and carboplatin, and more preferable to use paclitaxel and carboplatin.It may also be used in combination with a PARP inhibitor.The PARP inhibitor is not particularly limited and can be appropriately selected according to purpose, and examples thereof include olaparib, niraparib, etc.

[0039] "Pharmaceutically acceptable" means a non-toxic ingredient, composition, etc. that is physiologically acceptable and does not normally cause gastrointestinal disorders, allergic reactions such as dizziness, or similar reactions when administered to humans. Examples of the carrier include solvents, dispersion media, oil-in-water or water-in-oil emulsions, aqueous compositions, liposomes, microbeads and microsomes, biodegradable nanoparticles, etc.

[0040] The therapeutic agent may be formulated with an appropriate carrier depending on the route of administration. The administration route of the therapeutic agent is not particularly limited and can be appropriately selected depending on the purpose, and may be, for example, administered orally or parenterally. Parenteral administration routes include, for example, percutaneous, nasal, abdominal, intramuscular, subcutaneous, or intravenous.

[0041] When the therapeutic agent is orally administered, the pharmaceutical composition of the present invention may be formulated together with a suitable carrier for oral administration by methods known in the art in the form of, but not limited to, powder, granules, tablets, pills, sugar-coated tablets, capsules, liquids, gels, syrups, suspensions, wafers, and the like.

[0042] The therapeutic agents may be formulated using methods known in the art so as to provide quick, sustained or delayed release of the active ingredient after administration to a mammal.

[0043] The therapeutic agent formulated in the above manner may be administered in an effective amount through various routes including oral, transdermal, subcutaneous, intravenous or intramuscular, where "effective amount" refers to an amount of a substance that allows the diagnosis or tracking of a therapeutic effect when administered to a patient.

[0044] The dosage of the therapeutic agent can be appropriately selected depending on the administration route, the subject, the target disease and its severity, age, sex, body weight, individual differences, and disease state. The therapeutic agent may have a different content of active ingredient depending on the severity of the disease, but typically, based on an adult, the effective amount per administration is 0.1 mg to 50 mg (e.g., 5 mg), and may be administered several times a day.

[0045] When the therapeutic agent is used in combination with another drug such as paclitaxel, carboplatin, or a PARP inhibitor, the therapeutic agent and the other drug may be contained in a single formulation, or each may be contained in a different formulation. When the formulations are separate, the formulations may be administered simultaneously or separately at different times. EXAMPLES

[0046] The present invention will be described in more detail below with reference to examples, but the present invention is not limited to the following examples.

[0047] Materials and Methods

[0048] <<Analysis of TCGA ovarian cancer dataset>> RNA sequence gene expression data and survival data of TCGA ovarian cancer patients (422 cases, 429 files) were downloaded from the National Cancer Institute GDC Data Portal (https: / / portal.gdc.cancer.gov / ). Cases with FIGO stage classification of IIIa or above were extracted and defined as "WHOLE" (391 cases). In the cBioPortal for cancer genomics (https: / / www.cbioportal.org / ), cases with BRCA1 / 2 mutations were excluded from the genetically profiled WHOLE cases, and the remaining patients were defined as "BRCA-WT" (202 cases).

[0049] <<Drugs>> Carboplatin (CBDCA), paclitaxel (PTX), and olaparib are anti-cancer drugs used in the standard treatment of HGSC. Everolimus (EVL) is an allosteric mTORC1 inhibitor. Hydroxychloroquine is an autophagy inhibitor that inhibits the fusion of phagosomes and lysosomes. For in vitro cell culture tests, carboplatin (manufactured by Sigma-Aldrich) and hydroxychloroquine (manufactured by Sigma-Aldrich) were diluted in distilled water, and paclitaxel (manufactured by Sigma-Aldrich), olaparib (manufactured by CEM), and everolimus (manufactured by CEM) were diluted to the recommended concentration of DMSO (manufactured by Sigma-Aldrich). In animal experiments, paclitaxel (manufactured by Sigma-Aldrich) was diluted in 5% DMSO + 40% polyethylene glycol #300 + 5% Tween 80 + 50% distilled water and stored in DMSO. Everolimus was diluted in 30% propylene glycol + 5% Tween 80 + 65% distilled water and stored in propylene glycol. Carboplatin was diluted in distilled water as in the in vitro cell culture study.

[0050] <<Animal experiments>> Mice were housed in an environmentally controlled room and followed the protocol approved by the Keio University Institutional Animal Experimentation Committee (20064) in accordance with the "Institutional Guidelines for Animal Experiments at Keio University". Normal oviduct epithelial organoids were established from the oviduct epithelium of Rb1fl / flTrp53fl / flMycLSL / LSL mice (nFTE, JAX strain number: 029971) and B6J.129(B6N)-Gt(ROSA)26Sortm1(CAG-cas9*,-EGFP)Fezh / J mice (cas-nFTE, JAX strain number: 026175). BALB / cA nu / nu mice were used as recipients to transplant tumor cells derived from FTE organoids.

[0051] Tumor cells were harvested using TrypLE and counted using trypan blue and a cell counter. To investigate the in vivo properties of FTE-derived tumor cells, 1 × 10 6 Cells were suspended in 150 μL of PBS and administered intraperitoneally (day 1). For in vivo treatment assays, 2 × 10 cells suspended in 50 μL of medium / Matrigel (1:2 v / v) were administered intraperitoneally. 5 Cells were inoculated subcutaneously into each flank (day 1). Tumor volumes were measured every 3 days from day 7 onwards according to the following formula: V = (maximum tumor dimension) x (minimum tumor dimension) 2 x π / 6. In principle, the administration methods were as follows: CBDCA: 30 mg / kg intraperitoneal administration (ip), PTX: 8 mg / kg ip, and EVL: 5 mg / kg oral administration (po).

[0052] <<Organoid culture and viral cell engineering>> To establish fallopian tube epithelial (cas-nFTE, nFTE) organoids, fallopian tubes were removed under a microscope, cut into pieces, digested with collagenase type I at 37° C., and then incubated in TrypLE at 37° C. Dispersed fallopian tube epithelial cells were mixed with Matrigel (Corning) and solidified on a 48-well plate at 37° C. When the Matrigel was stable, organoid medium was added. Organoid medium was based on Advanced DMEM / F12 (Thermo Fisher Scientific) supplemented with 10 mM HEPES (Thermo Fisher Scientific), 2 mM GlutaMAX-I (Thermo Fisher Scientific), 50 × B27 (Gibco), 1 mM N-acetyl-L-cysteine ​​(Sigma-Aldrich), R-spondin CM (JSR Life Science or in-house), Noggin CM (JSR Life Science or in-house), Afamin / Wnt3A CM (JSR Life Science), 50 ng / mL human EGF (PeproTech), and 500 nM A83-01 (Tocris Bioscience). At the first addition of organoid medium at the time of establishment or passage, 10 μM Y-27632 (Sigma-Aldrich) was also added to the medium. To establish HGSC model organoids, nFTE organoids were transfected with pCMV-cre plasmid (VectorBuilder, VB190701-1025ges) or Cre lentiviral vector (pLenti-III) (Applied Biological Materials) transfection or infection, and the medium was replaced with medium containing 10 μM Nutlin-3 (Cayman Chemical) approximately 2 days after transfection or infection.cDNA encoding EGFP and human p62 (sequestosome-1, SQSTM1) were obtained from the DNASU Plasmid Repository (Arizona State University) and individually subcloned into the pLEX_307 lentiviral vector (Addgene plasmid #41392, Addgene).

[0053] <<Knockout of Trp53, Pten, Nf1, and Sqstm1 via CRISPR / Cas9>> Cas9 mouse oviduct organoids were trypsinized, and 1 x 10 6 cells were collected into an electroporation cuvette (NepaGene) containing OPTI-MEM, the appropriate single guide RNA (sgRNA), and the pCMV-Cre plasmid (VectorBuilder, VB190701-1025ges). Electroporation was performed using NEPA21 (NepaGene). After electroporation, the cells were incubated for 30 minutes, mixed with Matrigel (Corning), and covered with organoid medium containing 10 μM Y-27632 (Sigma-Aldrich). Around 2 days after transfection, the medium was replaced with medium containing 10 μM Nutlin-3 (Cayman Chemical) to select for Trp53 KO organoids.

[0054] sgTrp53 RNA was synthesized by Integrated DNA Technologies (IDT): Alt-R® CRISPR-Cas9 sgRNA Trp53 AAGTCACAGCACATGACGG Here, the Trp53 sgRNA is AAGTCACAGCACATGACGG (SEQ ID NO: 9).

[0055] sgPten, sgNf1, and sgSqstm1 RNAs were synthesized at FASMAC and cloned into the lentiCRISPR v2 vector (Addgene plasmid #52961, Addgene). For Pten knockout, the blasticidin resistance gene was subcloned into the lentiCRISPR v2 vector using In-Fusion (Takara Bio Inc.). For Nf1 knockout, the zeocin resistance gene was subcloned into the lentiCRISPR v2 vector. Pten knockout RPM (RPMP) cells, Nf1 knockout RPM (RPMN) cells, Pten- and Nf1 knockout RPM (RPMNP) cells, and p62 knockout RPMNP cells were established by lentivirus infection.

[0056] The sequences of the CRISPR oligos are as follows: Pten CRISPR Oligo1: CACCGAGATCGTTAGCAGAAACAAA (SEQ ID NO: 10) Pten CRISPR Oligo2: AAACTTTGTTTCTGCTAACGATCTC (SEQ ID NO: 11) Nf1 CRISPR Oligo1: CACCGCTCGTCGAAGCGGCTGACCA (SEQ ID NO: 12) Nf1 CRISPR Oligo2:AAACTGGTCAGCCGCTTCGACGAGC (SEQ ID NO: 13) Sqstm1 CRISPR Oligo1: CACCGATGGTGGGCGATGTTCCCGC (SEQ ID NO: 14) Sqstm1 CRISPR Oligo2: AAACGCGGGAACATCGCCCACCATC (SEQ ID NO: 15)

[0057] For single organoid cloning to establish RPM, RPMN, RPMP, and RPMNP organoids, organoids were trypsinized and dissociated. Dissociated cells were diluted with culture medium, mixed with Matrigel, and covered with organoid medium containing 10 μM Y-27632. After 7–14 days, single clone organoids were excised, expanded, and screened for mutations in the target genes according to the reported protocol. Confirmation of CRISPR / Cas9 knockout cell lines was performed by Sanger DNA sequencing, WES, or RNA sequencing. For Sanger sequencing, genomic DNA was extracted from cells using the DNeasy blood & Tissue Kit (QIAGEN). The deleted DNA region was then PCR amplified using primers spanning the potential deletion site using Prime STAR GXL Premix (Takara Bio Inc.). After separation on an agarose gel, the DNA fragments were excised from the agarose gel and purified using a QIAquick Gel Extraction Kit (QIAGEN). The DNA fragments were then PCR amplified using In-Fusion primers. After separation on an agarose gel, the DNA fragments were excised from the agarose gel and purified. The DNA fragments were cloned into BamHI-digested pUC19 using In-Fusion (Takara Bio). The transformation products using DH5α were plated on LB agar carbenicillin plates and cultured overnight at 37°C. The base sequences of the colonies were sequenced and the data were analyzed.

[0058] < <qpcr>> Total RNA was isolated from organoids or 2D cultured cells using the RNeasy Mini Kit (QIAGEN), genomic DNA was removed, and then reverse transcription was performed using the PrimeScript RT reagent Kit with gDNA Eraser (Perfect Real Time) (Takara Bio Inc.). RT-qPCR was performed using StepOnePlus Real-Time PCR Systems (Applied Biosystems) or Thermal Cycler Dice Real Time System TP800 (Takara Bio Inc.). The sequence information of the primer set used is as follows. p62 / SQSTM1 forward primer: AGGATGGGGACTTGGTTGC (SEQ ID NO: 3) p62 / SQSTM1 reverse primer: TCACAGATCACATTGGGGTGC (SEQ ID NO: 4) Actin forward primer: GGCATAGAGGTCTTTACGGATGTC (SEQ ID NO: 16) Actin reverse primer: TATTGGCAACGAGCGGTTCC (SEQ ID NO: 17) NQO1 forward primer: AGGATGGGAGGTACTCGAATC (SEQ ID NO: 18) NQO1 reverse primer: AGGCGTCCTTCCTTATATGCTA (SEQ ID NO: 19) HMOX1 forward primer: AAGCCGAGAATGCTGAGTTCA (SEQ ID NO: 20) HMOX1 reverse primer: GCCGTGTAGATATGGTACAAGGA (SEQ ID NO: 21)

[0059] <<Western Blotting>> Cell pellets were lysed in 2x sodium dodecyl sulfate (SDS) buffer or RIPA lysis buffer containing 0.1 M dithiothreitol (DTT). Protein concentrations were measured by Pierce BCA Protein Assay Kits (Thermo Fisher Scientific). Equal amounts of lysed protein extracts were loaded, separated by SDS-PAGE, and transferred to PVDF membranes (IPFL00010, Millipore). Membranes were blocked with 5% nonfat milk in TBS plus 0.05% Tween 20 for 40 min at room temperature, followed by incubation with primary antibodies. Blots were washed and incubated with fluorescently labeled anti-mouse IgG (#NA931V, Cytiva) or anti-rabbit IgG (#NA9340V, Cytiva) at room temperature. Membranes were analyzed by enhanced chemiluminescence (PerkinElmer) using a LAS-3000 (Fujifilm Corporation). Information regarding the antibodies used is listed in Table 1 below.

[0060] <<Drug susceptibility assay>> Tumor cells harvested using TrypLE Express (Gibco) were filtered through a 100 μm nylon cell strainer (Falcon® 100 μm cell strainer, Corning) and counted with trypan blue using a TC20 automated cell counter (Bio-Rad). Cells were seeded into 96-well plates on day 1. CBDCA (Sigma-Aldrich), PTX (Sigma-Aldrich), Olaparib (CEM), EVL (CEM), and combinations of these drugs or solvents were added at the indicated concentrations when medium was refreshed on day 2. ATP concentrations were measured using CellTiter-Glo 2.0 (Promega, G9243) according to the manufacturer's instructions, and luminescence was measured on day 4 using an EnVision multimode plate reader (PerkinElmer). Results were normalized to control samples and analyzed. The sum of squares F test was used to analyze the difference in IC50 of each compound.

[0061] <<Cell lines and cell culture>> All cell lines were cultured under conditions of 5% CO2 and 37°C. SKOV3 was cultured in RPMI Medium 1640 (manufactured by Gibco) supplemented with 10% fetal bovine serum. Caov3 was cultured in a medium (D-MEM / Ham's F-12 (containing L-glutamine and phenol red), manufactured by Fujifilm Wako Pure Chemical Corporation) supplemented with 10% fetal bovine serum. After establishing HGSC modeling organoids from nFTE organoids, each organoid could be cultured in a medium as a cell line.

[0062] <<Colony formation assay>> Tumor cells collected using TrypLE Express (manufactured by Gibco) were filtered using a 100-μm nylon cell strainer (manufactured by Falcon or greiner) and counted using trypan blue with a TC20 automatic cell counter (manufactured by Bio-Rad) or a hemocytometer Burker-turk board (manufactured by ERMA). RPM-EGFP, RPMNP, RPM-p62OE, RPMNP-p62KO, SKOV3, and Caov3 cells were seeded in 6-well plates on day 1. The indicated concentrations of CBDCA (manufactured by Sigma-Aldrich), PTX (manufactured by Sigma-Aldrich), EVL (manufactured by CEM), and combinations of these drugs or solvents were added for 7 days. The total culture days and the number of primary seeded cells were optimized for each cell line. On the final day, after washing the wells with ice-cold PBS, the cells were fixed with ice-cold methanol for 10 minutes. As a result, the cells were stained with crystal violet solution and washed with distilled water. After drying overnight at room temperature, the stained cells were scanned with white paper and analyzed using the open-source image processing software "ImageJ" (National Institutes of Health, NIH). To evaluate the efficacy of each compound, an F-test of the sum of squares or a student t-test was used.

[0063] <<RNA sequence analysis>> Cultured organoids were subjected to RNA extraction using the RNeasy Mini kit (Qiagen). Sequence libraries were prepared using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (E7760, New England BioLabs). All libraries were evaluated using TapeStation4200 (Agilent) and quantified using Qubit4.0 (Thermo Fisher Scientific). Library pooling was sequenced using 50bp PE reads on the NextSeq2000 platform (Illumina). Sequence reads obtained from the fastq files were trimmed, mapped to the reference mouse genome "GRCm39", and read counts were obtained using CLC Genomics Workbench (Qiagen). Heatmap analysis was performed using TPM values. Euclidean distance was used for distance measurement and complete linkage was used for the clustering criterion. Gene set enrichment analysis was performed using the software "GSEA". Preranked GSEA was used for comparison of gene expression, and gene expression was compared using log2(normalized TPM ratio), excluding genes with extremely low expression. Hallmark gene sets of mouse orthologs and gene sets derived from Biological Process Ontology were referenced.

[0064] <<DIA Proteome Analysis>> Proteins in cell samples were extracted with 100 mM Tris-HCl (pH 8.5) containing 2% sodium dodecyl sulfate (SDS) using a sonicator. Protein extracts were reduced with 20 mM tris(2-carboxyethyl)phosphine (TCEP) at 80°C for 10 min and alkylated with 35 mM iodoacetamide at room temperature for 30 min. Protein purification and digestion were performed using a magnetic bead-based proteome sample preparation method (Single-Pot Solid-Phase-enhanced Sample Preparation, SP3). Trypsin digestion was performed overnight at 37°C using 500 ng Trypsin / Lys-C Mix (Promega). Cell digests were purified using a GL-Tip SDD (GL Sciences) according to the manufacturer's protocol. Peptides were redissolved in 2% acetonitrile (ACN) containing 0.1% trifluoroacetic acid (TFA) and quantified for LC-MS analysis. Two mobile phases, A and B, containing H2O and 80% acetonitrile in water, were prepared with 0.1% formic acid (FA). Then, 300 ng of digested peptides were loaded onto an UltiMate 3000 RSLC nano system (Thermo Fisher Scientific). The eluted peptides were analyzed by overlapping window DIA using a Q Exactive HF-X (Thermo Fisher Scientific). The LC-MS / MS data were analyzed by Scaffold DIA (Proteome Software), a qualitative / quantitative analysis software, using a mouse spectral library constructed from the mouse UniProtKB / Swiss-Prot database to identify and quantitate peptides and proteins with FDR values ​​below 1%. Furthermore, the expression values ​​were analyzed by grouping relative proteins, normalizing them by median, and excluding proteins counted with unique peptides.To explore significant factors related to mTORC1 signaling, we listed and analyzed proteins included in the “mouse-ortholog hallmark mTORC1 signal transduction” by excluding proteins that were not expressed in one or more samples.

[0065] <<Gene sequencing and data analysis of mouse organoids>> DNA was isolated from cultured organoids using the QIAamp DNA Mini Kit (Qiagen). DNA purity and concentration were determined using NanoDrop. TM The genomic DNA samples were examined using a 2000 spectrophotometer (Thermo Fisher Scientific). Genomic DNA samples were fragmented using a sonicator. The fragments were treated with End Prep Enzyme Mix (Thermo Fisher Scientific) to repair the ends, and the A-Tailing Mix was used to add a Poly-A tail to the 3' end, followed by purification. Adapters were then ligated to both ends of the purified fragments. After further size selection, the fragments were amplified using Pre-Capture PCR primers. To capture the exon regions, Block Mix, Hybridization Buffer, and Capture Library were injected into 750 ng of library for 24 hours. The captured products were eluted using magnetic beads (Dynabeads MyOne Streptavidin T1, Thermo Fisher Scientific). Each sample was then amplified with PCR P5 and P7 primers. After library preparation, the size and concentration of each sample were measured. Each capture library was loaded onto an Illumina Novaseq for 2 × 150 paired-end sequencing according to the manufacturer's instructions (Illumina). Raw image files and base calling were performed with NovaSeq Control Software (NCS) + OLB + GAPipeline-1.6 (Illumina) on the NovaSeq instrument. After adapter reads were removed with Cutadapt, clean data were aligned to the reference genome with BWA, and results were de-duplicated with Picard. SNV / InDel calling was performed with GATK or Samtools, and variants were annotated with Annovar. Somatic mutation analysis was performed with Mutect2 as needed.The somatic mutation vcf file obtained from WES analysis was analyzed using the R package "musicatk" based on COSMIC signatures, and the exposure fractions were decomposed into representative ovarian cancer mutation signatures including signature 1 (age), signatures 2 and 13 (APOBEC), signature 3 (homologous recombination (HR) deficiency), signature 4 (smoking), signature 6 (mismatch repair (MMR) deficiency), signature 7 (ultraviolet exposure), and signature 10 (POLE).

[0066] <<Human sample>> The collection of patient data and ovarian cancer tissues was conducted at Keio University Hospital with the approval of the institutional ethics committee (approval numbers 20070081, 20210111). This study was carried out in accordance with all relevant guidelines and regulations. All individuals who participated in this study received appropriate informed consent and gave their consent. HGSC resected tumor samples were fixed with formalin and paraffin-embedded (FFPE).

[0067] <<IHC and IF, multiplex immunofluorescence>> For immunohistochemistry (IHC) and immunofluorescence (IF), organoids in Matrigel were collected from culture plates, fixed with formalin, and paraffin-embedded. FFPE specimen slides were deparaffinized using ClearPlus (or xylene) and ethanol, and endogenous peroxidase was inactivated with methanol and H2O2. Subsequently, H&E staining, IHC, IF, or multiplex immunofluorescence (multiplex IF) was performed. For multiplex IF, the PerkinElmer Opal (registered trademark) Multiplex IHC Kit (manufactured by PerkinElmer) was used, and each protein was stained according to the following procedure. AR6 Buffer or AR9 Buffer was diluted with distilled water and antigens were activated at 95°C or 120°C for the optimal time. After quenching, slides were placed in TBS-T. Primary antibodies diluted in antibody diluent / block were added to slides blocked with antibody diluent / block at room temperature. After washing with TBS-T, Opal Polymer HRP MS + Rb was added at room temperature. After washing again with TBS-T, Opal Fluorophore of each color diluted in 1× Plus Amplification Diluent was added for 10 minutes at room temperature. Finally, specimens were stained with Spectral DAPI at room temperature and washed. Slides were scanned using Vectra Polaris and image quantitative analysis was performed using the software "inForm". Information regarding the antibodies used is listed in Table 1 below.

[0068] [Table 1]

[0069] <<Statistical analysis>> Statistical analysis was performed using medical statistics software GraphPad Prism 9 software (GraphPad System, Inc.), R language, or Python. A p value of <0.05 was considered significant. Data are expressed as mean ± standard error of the mean (SEM).

[0070] <Exploring biological pathways related to the prognosis of patients with advanced serous ovarian cancer> To explore the important biological processes involved in the prognosis of advanced serous ovarian cancer patients, we performed gene set variation analysis (GSVA) ​​on 391 advanced serous ovarian cancer cases (WHOLE) from the TCGA ovarian cancer dataset to examine the relationship between activation of pathways related to biological processes and prognosis (Figure 1A). GSVA is a method to score the activity of specific pathways in each sample, and we used it to divide ovarian cancer cases into two groups based on the score of all pathways.

[0071] Here, Figure 1A is a schematic diagram showing a computational approach using the R package "GSVA" (gene set variation analysis) and the transcriptome dataset of TCGA advanced ovarian cancer cases. This computational approach explored the association between the prognosis of TCGA advanced ovarian cancer cases and pathways in the biological process ontology associated with this.

[0072] For transcriptome analysis, the log-scaled TPM value of each gene in primary tumor cases was used as gene expression data, and the chrY gene and genes with extremely low expression were excluded. Next, the R package "GSVA" (Humans; C5; BP; gene set size 50-500) was run on WHOLE and BRCA-WT, respectively, to score the activity of each pathway, and cases were divided into positive (or equal to 0) and negative groups according to the score of each pathway. As a result, for all analyzed pathways, the p-value between the two groups was calculated by log-rank test, and the pathways were ranked by p-value.

[0073] The results are shown in Figure 1B. Figure 1B is a heat map showing the correlation between GSVA score and prognosis in all advanced ovarian cancer patients. Specifically, the GSVA scores of Biological Process Ontology pathways with p-values ​​<0. 05 (only pathways with p-values ​​<0.01 are shown) were ranked by log-scale p-values ​​using the log-rank test, which examined the correlation between GSVA score and prognosis in all advanced ovarian cancer patients (n=391). Each column shows the GSVA score of the same case. The cases were arranged according to the GSVA score of the most significant pathway, "sterol transport" (in Figure 1B). In addition, in Figures 1B and 1C, each pathway described after "GOBD" indicates each pathway classified according to the Gene Ontology Data Base.

[0074] As a result, the activation of 23 pathways out of 1590 signals was significantly correlated with poor or good prognosis in WHOLE patients (p-value < 0.01), and it was found that the activation of signals such as "sterol transport" and "inorganic anion transport" (inorganic anion transport in Fig. 1B) was significantly correlated with poor prognosis (Fig. 1B).

[0075] Interestingly, recent studies have reported that cholesterol membrane transport induces reprogramming of tumor-associated macrophages and promotes tumor growth. It has also been reported that phosphate homeostasis via SLC34A2-XPR1 is important for the survival of ovarian cancer cells. Therefore, the validity of this computational approach to discover important pathways that determine the prognosis of serous ovarian cancer cases was demonstrated.

[0076] <Search for biological pathways related to the prognosis of patients with advanced serous ovarian cancer with wild-type BRCA1 / 2> Next, to search for biological pathways involved in the prognosis of patients with advanced serous ovarian cancer with wild-type BRCA1 / 2, gene set variation analysis (GSVA) was performed on 202 cases of BRCA1 / 2 wild-type advanced serous ovarian cancer (BRCA-WT) from the ovarian cancer dataset of The Cancer Genome Atlas (TCGA), and the relationship between the prognosis of BRCA-WT patients and the activation of each pathway was similarly examined.

[0077] The results are shown in Figs. 1C to 1E. Figure 1C is a heatmap showing the correlation between GSVA score and prognosis in patients with advanced ovarian cancer without BRCA1 / 2 mutations. Specifically, the GSVA scores of Biological Process Ontology pathways with p-values ​​<0. 05 (only pathways with p-values ​​<0.01 are shown) were ranked by log-scale p-values ​​using the log-rank test to examine the correlation with prognosis only for patients with advanced ovarian cancer without BRCA1 / 2 mutations (n=202), not for all patients with advanced ovarian cancer (n=391). Each column shows the GSVA score of the same case. Cases were arranged according to the GSVA score of the most important pathway, “regulation of small GTPase mediated signal transduction” (Figure 1C).

[0078] Figure 1D shows Kaplan-Meier survival curves showing the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 mutations according to the difference in "RAS protein signaling" (Ras-related signaling transduction in Figure 1C) score (RAS score). Specifically, the RAS score was calculated using GSVA, and Kaplan-Meier survival curves were shown for the prognosis of patients with advanced ovarian cancer without BRCA 1 / 2 mutations, for the "LOW" group including patients with a RAS score of less than 0, and the "HIGH" group including patients with a RAS score of 0 or more. In Figure 1D, the p-value from the log-rank test is shown.

[0079] [Table 2]

[0080] Activation of GTPase-related signals, such as "control of small GTPase-mediated signal transduction," was identified as the most significant poor prognostic factor specifically in BRCA-WT cases compared to WHOLE cases (Figure 1C). In particular, in the GTPase family signal transduction pathway, activation of "RAS protein signal transduction" correlated with poor prognosis in BRCA-WT patients (Figure 1D). Next, to visualize the connections of significant pathways specific to BRCA-WT (p value < 0.05), we grouped the pathways by GSVA score using the Variational Bayesian Gaussian Mixture Model (VBGMM). Only seven pathways (Table 2) corresponded to GTPase-related pathways, including the RAS-crosstalking pathway "phosphatidylinositol 3 kinase signaling" (hereinafter sometimes referred to as "PI3K-related signaling" in Table 2).

[0081] In addition, Figures 8A-B are shown as supplementary data. Figure 8A shows Kaplan-Meier survival curves showing the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 gene mutations according to the difference in the "PI3K-associated signal" score (PI3K score). Kaplan-Meier survival curves are shown for each group: the "LOW" group including patients with a PI3K score of less than 0, and the "HIGH" group including patients with a PI3K score of 0 or more. The p-values ​​from the log-rank test are shown. Figure 8B is a scatter plot showing the correlation between RAS scores calculated using GSVA and PI3K scores in patients with advanced ovarian cancer without BRCA1 / 2 mutations. The Pearson correlation coefficient r was calculated and shown in the figure.

[0082] The results in Figures 1A-E and 8A-B show that a computational approach using the TCGA dataset to search for biological process ontologies that are strongly associated with the progression of TCGA advanced ovarian cancer cases indicates that abnormalities in RAS / PI3K signaling are associated with poor prognosis in patients with BRCA1 / 2 wild-type serous ovarian cancer.

[0083] To investigate how abnormalities in RAS / PI3K signaling affect the phenotype of BRCA1 / 2 wild-type HGSC, we generated HGSC model organoids from mouse fallopian tube epithelium using the following procedure. Mouse fallopian tube epithelial organoids are excellent for analyzing phenotypes caused by signal abnormalities, since they can be used to create ovarian cancer models with minimal genetic mutations without the need to immortalize normal cells. Therefore, we used a stepwise genetic engineering technique to create a BRCA1 / 2 wild-type (BRCA-WT) HGSC model with abnormalities in RAS / PI3K signaling.

[0084] The results are shown in Figures 2A-I and Figures 9A-F as supplementary data. FIG. 2A shows immunofluorescence images of normal fallopian tube epithelium excised from a nude mouse stained with PAX8 (red), α-tubulin (green), and DAPI (blue). FIG. 2B shows bright field images of normal fallopian tube epithelial organoids (cas-nFTE) and Trp53 knockout organoids (casP).

[0085] FIG. 9A shows immunohistochemical images of normal mouse fallopian tube epithelium stained with PAX8 antibody and α-tubulin antibody. Figure 9B is a diagram showing the Sanger sequence of the Trp53 locus in casP organoids. "mFTE" indicates the base deletion site of Trp53 in casP organoids, and "nFTE" indicates the normal sequence of the corresponding Trp53 locus.

[0086] First, we confirmed that the oviduct epithelium excised from nude mice was composed only of secretory cells and ciliated cells, similar to human oviduct epithelium (Figure 2A and Figure 9A). Next, we established mouse oviduct organoids (hereinafter referred to as "cas-nFTE") from B6J.129(B6N)-Gt(ROSA)26Sortm1(CAG-cas9,-EGFP)Fezh / J mice (see public literature: Platt RJ et al(2014)Cell 159:440-455). We introduced sgRNA against Trp53 into the cas-nFTE organoid to knock out Trp53, performed nutlin-3 selection and single organoid cloning, and named the isolated organoid "casP" (casP in Figure 2B; and Figure 9B).

[0087] Figure 2C shows RNA-seq results by Preranked Gene Set Enrichment Analysis (GSEA) showing enrichment of MSigDB Gene Ontologies in cellular components of casP organoids compared to cas-nFTE organoids. The y-axis represents gene sets and the x-axis represents normalized enrichment scores (NES). Figure 9C is a graph showing RNA-seq analysis results by Preranked Gene Set Enrichment Analysis (GSEA) showing enrichment of MSigDB Gene Ontologies in biological processes in casP organoids compared to cas-nFTE organoids. The y-axis represents gene sets and the x-axis represents normalized enrichment scores (NES).

[0088] After confirming the Trp53 frameshift mutation and clonality by whole exome sequencing (WES), we analyzed gene expression differences due to Trp53 knockout by RNA sequencing and gene set enrichment analysis (GSEA). As a result, the expression of cilia-related gene sets was reduced in casP organoids (Figure 2C and Figure 9C). It has been reported that differentiation into ciliated cells is reduced in serous intraepithelial carcinoma (STIC), a precancerous lesion of human HGSC (see public literature: Abdelhamed ZA et al (2018) Int J Gynecol Cancer 28:1535-1544). Therefore, the impairment of p53 function in cas-nFTE was thought to reflect the oncogenic process of HGSC. Therefore, casP cells were subcutaneously transplanted into nude mice to evaluate their in vivo carcinogenicity. As a result, casP cells did not form subcutaneous tumors, suggesting that Trp53 knockout alone is insufficient for the transformation of fallopian tube epithelium.

[0089] FIG. 2D is a schematic diagram showing the procedure for establishing a mouse HGSC model organoid from Rb1fl / flTrp53fl / flMycLSL / LSL mice. It has already been reported that gene alterations related to Tp53, RB1, and MYC are frequently observed in HGSC (see public literature: Bell D et al (2011) Nature 474:609-615). fl / fl Trp53 fl / fl Myc LSL / LSL Normal fallopian tube organoids (nFTEs) established from mice (see public literature: Mollaoglu G et al (2017) Cancer Cell 31:270-285) were used as the source of organoids modeling the fallopian tube epithelium (Figure 2D).

[0090] FIG. 9D shows whole mount immunofluorescence images of nFTE organoids stained with PAX8, α-tubulin, and DAPI. FIG. 9E shows representative immunohistochemical images of nFTE and mFTE organoids stained with MKI67 antibody.

[0091] Figure 2E shows a representative bright field image of the established organoid sphere (lower row) and a pathological image of a tumor formed by subcutaneously transplanting RPM cells into a nude mouse (upper row). RPM: Rb1 - / - Trp53 - / - MycOE, RPMN:Rb1 - / - Trp53 - / - MycOE+Nf1KO, RPMP:Rb1 - / - Trp53 - / - MycOE+PtenKO, RPMNP:Rb1 - / - Trp53 - / - MycOE+Nf1KO+PtenKO. Normal fallopian tube epithelium (nFTE) organoids were established from the fallopian tubes of Rb1fl / flTrp53fl / flMycLSL / LSL mice.

[0092] After confirming that nFTE organoids were composed of secretory and ciliated cells (Figure 9D), we then performed the following procedure to knock out Rb1 and Trp53 and overexpress Myc in nFTE organoids by Cre transduction, and found that MKI67-expressing cells increased (Figure 9E). Furthermore, the genetically engineered organoids (mFTE) were able to develop subcutaneous tumors in nude mice that were well stained with human HGSC marker proteins such as WT1 and PAX8 (Figure 2E). Since the subcutaneous tumors mimicked human HGSC pathology, we were able to detect Rb1 in the subcutaneous tumors. - / - Trp53 - / - We established MycOE HGSC model organoids (hereinafter referred to as "RPM").

[0093] Figure 2F shows the results of Western blotting of RPM, RPMN, RPMP and RPMNP cells, detecting GAPDH, NF1, PTEN, MEK, pMEK, AKT and pAKT proteins. Figure 9F is a schematic diagram showing the mutation information of Nf1 and Pten genes in RPM, RPMN, RPMP, and RPMNP organoids.

[0094] Next, to generate poor prognosis models with abnormalities in RAS / PI3K signaling from RPM organoids, we knocked out Nf1, Pten, or both using the CRISPR / Cas9 system (see Ran FA et al. (2013) Nat Protoc 8:2281-2308) and successfully generated various HGSC model organoids (referred to as "RPMN", "RPMP", and "RPMNP", respectively). Clonality and activation of RAS / PI3K signaling were confirmed by WES and Western blotting (WB) (Figures 2F and 9F).

[0095] FIG. 2G shows unsupervised hierarchical clustering of RNA-seq expression data from nFTE, RPM, RPMN, RPMP, and RPMNP organoids.

[0096] We performed RNA sequencing and unsupervised hierarchical clustering based on the transcriptome profiles of nFTE, RPM, RPMN, RPMP, and RPMNP, and the results showed that the four HGSC model organoids were genetically distant from nFTE (Figure 2G).

[0097] Figure 2H is a bar graph showing mutation signatures from 96 Single Base Substitution (SBS) Signatures of RPM, RPMN, RPMP, and RPMNP cells analyzed using the R package "musicatk" based on the COSMIC signature. The Y-axis shows the exposure fraction, which means the proportion of each oncogenic exposure. Signature 1: age, Signatures 2 and 13: APOBEC, Signature 3: homologous recombination (HR) deficiency, Signature 4: smoking, Signature 6: mismatch repair (MMR) deficiency, Signature 7: UV exposure, Signature 10: POLE.

[0098] Next, we characterized the exome profiles of RPM, RPMN, RPMP, and RPMNP organoids by SBS96 mutation signatures. Previous studies have shown that the five subtypes of human epithelial ovarian cancer have completely different mutation signature profiles; in particular, the dominant contributor in HGSC is signature 3, which is associated with homologous recombination deficiency (Figure 2H). Thus, these results suggested that nFTE-derived HGSC modeling organoids are similar to human HGSC not only morphologically but also genetically.

[0099] <HGSC with abnormal activation of RAS / PI3K signaling has a worse prognosis and reduced autophagy function> To analyze the in vivo characteristics of the developed HGSC model, RPM, RPMN, RPMP, and RPMNP cells were inoculated intraperitoneally into nude mice as recipient mice, and the survival rates of the recipient mice were compared. Figure 3A shows the results of 1×10 6 13 shows Kaplan-Meier survival curves of nude mice (n=5 for each point) intraperitoneally implanted with RPM cells, RPMN cells, RPMP cells, or RPMNP cells. All cells were able to form tumors, and among them, the survival rate of recipient mice of RPMNP cells was found to be the poorest (Figure 3A).

[0100] Next, chemotherapy sensitivity testing was performed. FIG. 3B shows the dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of carboplatin (CBDCA). FIG. 3C shows the dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of paclitaxel (PTX). Figure 3D shows the dose-response curves of RPM, RPMN, RPMP, and RPMNP cells following administration of olaparib. In Figures 3B to 3D, cell viability was measured using CellTiter-Glo (registered trademark) 2.0 (Promega) after 1 day of culture and after 3 days of treatment, and normalized to vehicle-treated cells.

[0101] As a result, RPMN cells were more resistant to CBDCA and PTX than RPM cells, and RPMP cells were more resistant to CBDCA. Of note, RPMNP cells were resistant to CBDCA, PTX, and olaparib (Figure 3B-D). Therefore, these results suggest that aberrant RAS / PI3K crosstalk due to Nf1- or Pten-deficiency contributes to the poor progression of BRCA1 / 2 wild-type HGSC, which was consistent with the computational analysis of TCGA.

[0102] Next, to investigate transcriptomic changes following genetic manipulation in organoid lines, we performed RNA-seq analysis by GSEA of Molecular Signatures Database (MSigDB) hallmark gene sets between RPM and RPMNP organoids. Figure 3E is a graph showing the results of RNA-seq analysis of MSigDB hallmark gene sets enriched in RPMNP cells compared to RPM cells. In Figure 3E, the y-axis shows gene sets with FDR-q values ​​less than 0.05, and the x-axis shows normalized enrichment scores (NES). FIG. 3F shows an enrichment plot of mTORC1 signaling in the MSigDB hallmark gene set. FIG. 3G shows an enrichment plot of unfolded protein response in the MSigDB hallmark gene set. In Figure 3F-G, the y-axis indicates the enrichment score (ES), and the x-axis indicates the list of MSigDB hallmark genes correlated with "mTORC1 signaling" or "endoplasmic reticulum stress response" in order of correlation degree, from positive to negative correlation from the left. The set FDR (False Discovery Rate)-q value is shown in the figure. Figure 10 is a graph showing an enrichment plot of the autophagosome maturation pathway analyzed by Preranked GSEA, which examines the enrichment of gene sets in the biological process ontology in RPMNP cells compared to RPM cells. The p-value is shown in the figure.

[0103] As a result of RNA-seq analysis, in RPMNP organoids, cell proliferation-related signal gene sets such as "MYC targets v1 / 2", "E2F targets", and "G2M checkpoint" were activated, and the gene set "apoptosis" was suppressed (Figure 3E). Furthermore, in RPMNP organoids, "mTORC1 signaling" and the unfolded protein response, which is known to inversely regulate autophagy, were activated (Figure 3F, Figure 3G). In addition, GSEA between RPM organoids and RPMNP organoids for the gene set "autophagosome maturation" showed that the early stage of autophagy was downregulated in RPMNP organoids (Figure 10A).

[0104] From the results of Figures 3A - G and Figure 10A as supplementary data, it was suggested that the genotype activating RAS / PI3K crosstalk leads to poor prognosis and chemotherapy resistance and inhibits autophagy via the mTORC1 signal.

[0105] <SQSTM1 (p62), which escaped autophagic degradation by mTORC1 signal activation, confers platinum agent resistance to BRCA1 / 2 wild-type HGSC> To examine whether the activation of the mTORC1 signal affects tumor progression, RPM, RPMN, RPMP, and RPMNP cells were treated with everolimus (EVL), an allosteric mTORC1 inhibitor, and cell proliferation was examined.

[0106] The results are shown in Figures 4A - G and Figures 11A - B as supplementary data. Figures 4A-B show viable cell staining images (Figure 4A) and dose-response curves (Figure 4B) of RPMNP cells treated with CBDCA + 0.1% DMSO and CBDCA + 100 nM everolimus. In Figure 4A, a representative stained well of viable cell staining images is shown. In Figure 4B, cell viability was measured by crystal violet staining and normalized to cells cultured without CBDCA, respectively. p-values ​​by sum-of-squares F-test are shown. Figure 4C-D shows dose-response quantification of BRCA1 / 2 wild-type human ovarian cancer cell lines, SKOV3 (Figure 4C) and Caov3 (Figure 4D), which were treated with DMSO, everolimus, 2 μM CBDCA, 2 μM CBDCA + everolimus, 10 μM CBDCA, and 10 μM CBDCA + everolimus, respectively. Figure 4E is a graph showing the proteome expression comparison of RPM incubated in DMSO and RPMNP incubated in DMSO. Proteins involved in mTORC1 signaling, a hallmark of MSigDB, are listed on the y-axis, and proteins upregulated in RPMNP are boxed. Figure 4F is a dot plot of proteins whose expression was decreased by administration of an mTOR inhibitor among proteins whose expression was increased in RPMNPs by comparing the proteome expression of RPMNPs cultured with DMSO and RPMNPs cultured with everolimus. The proteins included in the box in Figure 4E are plotted, and the proteins whose expression was decreased in RPMNPs treated with everolimus are shown in a box. 4G is a diagram showing Western blotting images of p62, S6, and pS6 proteins detected in RPM, RPMN, RPMP, and RPMNP cells, and a detection image of GAPDH protein is shown as a control.

[0107] Dose-response curves of EVL-treated RPM, RPMN, RPMP, and RPMNP cells are shown in Figure 11A. Cell viability was measured after 1 day of culture and 3 days of treatment using CellTiter-Glo 2.0 (Promega) and normalized to vehicle-treated cells. Figure 11B shows representative images of stained wells of RPMNP cells treated with EVL for 7 days.

[0108] The proliferation of RPM, RPMN, and RPMP cells was inhibited to some extent by EVL, but EVL had no effect on the proliferation of RPMNP cells, a model cell line of mTORC1 signal-activated HGSC (Figures 11A-B). However, surprisingly, when administered with CBDCA, EVL enhanced the antitumor effect of CBDCA in RPMNP cells (Figure 4A-B). Furthermore, the combined effect of EVL was also demonstrated against representative human BRCA1 / 2 wild-type ovarian cancer cell lines such as SKOV3 and Caov3 (Figure 4C-D).

[0109] To investigate the molecular mechanism of the combined effect, we performed proteomic analysis of proteins included in the gene set "mTORC1 signaling" in RPM cells, RPMNP cells, and EVL-treated RPMNP cells. As a result, SQSTM1 (p62) was detected as a protein that was significantly increased by abnormalities in RAS / PI3K crosstalk and decreased by EVL (Figure 4E-F). Next, we verified that genetic manipulation that gradually activates RAS / PI3K increases the expression of p62 and activation of mTORC1 signaling (Figure 4G).

[0110] To investigate the effect of p62 on chemotherapy sensitivity, we created p62 knockout RPMNP cells (RPMNP-p62KO) and p62 overexpressing RPM cells, named "RPM-p62OE" (Figure 11C). Chemosensitivity tests were performed using the created RPM-p62OE.

[0111] The results are shown in Figures 4H-K and supplementary data in Figure 11C. Figures 4H-I show viable cell staining images (Figure 4H) and dose-response curves (Figure 4I) of p62 knockout RPMNP cells treated with CBDCA (RPMNP-p62KO) and RPMNP cells treated with CBDCA. In Figure 4H, a representative stained well of the viable cell staining image is shown. In Figure 4I, the cell viability was normalized to the solvent-treated cells. The p-values ​​by the sum of squares F test are shown. Dose-response quantification of RPM cells overexpressing p62 (RPM-p62OE) treated with CBDCA is compared to the dose-response quantification of RPM cells expressing EGFP as a control (Figure 4J). Results are shown for vehicle alone, 5 μM, and 10 μM CBDCA. p values ​​from Student's t-test are shown; *p value < 0.05, **p value < 0.01, ***p value < 0.001, ****p value < 0.0001. Figure 4K shows Western blotting images of p62, LC-3B, and pS6 proteins in RPMNP cells treated with solvent alone, everolimus alone, and both everolimus and the autophagy inhibitor hydroxychloroquine. Actin protein detection is shown as a control.

[0112] Figure 11C shows Western blotting images of p62 and EGFP proteins detected in RPM, RPM-p62OE, RPMNP, and RMNP-p62KO cells. Actin protein detection images are shown as a control. Figures 11D-E show viable cell staining images (Figure 11D) and dose-response curves (Figure 11E) of RMNP-p62KO cells and RPMNP cells treated with CBDCA + PTX (TC). p values ​​from the sum of squares F test are shown; *p value < 0.05, **p value < 0.01, ***p value < 0.001, ****p value < 0.0001.

[0113] Chemotherapy sensitivity tests showed that RPMNP-p62KO cells were more sensitive to CBDCA than RPMNP cells, and also to TC treatment (CBDCA+PTX), the standard treatment for HGSC patients (Figure 4H-I, Figure 11D-E). On the other hand, RPM-p62OE cells were more resistant to CBDCA than RPM cells expressing EGFP as a control (Figure 4J). These results indicated the possibility of p62 as a therapeutic target for BRCA1 / 2 wild-type HGSC with poor prognosis.

[0114] Next, we investigated how p62 is regulated in HGSC. p62 is known to induce the aggregation of unnecessary proteins as a ubiquitin-binding protein and to be degraded by autophagy. Therefore, we quantified intracellular p62 when EVL or both EVL and hydroxychloroquine, an autophagy inhibitor, were added. As a result, EVL-induced degradation of p62 was rescued by inhibiting autophagy (Fig. 4K).

[0115] The results in Figure 4A-K and supplementary Figure 11A-E suggest that EVL enhances the efficacy of platinum-based therapy against BRCA1 / 2 wild-type HGSC via autophagy-mediated p62 degradation regulated by mTORC1.

[0116] <Activation of mTORC1-SQSTM1 by chemotherapy> Although HGSC patients are usually initially sensitive to chemotherapy, they experience a high rate of relapse, which is thought to be due to minimal residual disease after chemotherapy, although the mechanism behind this remains unclear. We hypothesized that the mTORC1-p62 interaction described above may be involved in the acquisition of chemotherapy resistance in residual tumors.To investigate the dynamics of mTORC1-SQSTM1 activation in human HGSCs, we analyzed 15 pairs of HGSC tumor samples resected before and after neoadjuvant chemotherapy (NAC) from each patient (PreNAC and PostNAC, Fig. 5A).

[0117] Here, FIG. 5A is a schematic diagram showing the procedure for collecting and following up 15 pairs of human HGSC clinical samples before chemotherapy (PreNAC) and after chemotherapy (PostNAC).

[0118] The results are shown in Figures 5B to 5D. FIG. 5B shows Opal Multiplex IHC multiplex immunofluorescence staining of a BRCA mutant case before (left) and after (right) chemotherapy. Figure 5C is a diagram showing Opal Multiplex IHC multiplex immunofluorescence staining before (left) and after (right) chemotherapy in BRCA wild-type cases. In the color drawings of FIGS. 5B-C, orange: pS6, cyan: CK, red: p62, and blue: DAPI are shown respectively. Figure 5D is a graph showing the post-chemotherapy increase rate of fluorescence intensity per area of p62 in tumor tissues. The black bar graph indicates BRCA wild-type, the white bar graph indicates BRCA mutant type, and the dark gray bar graph indicates the type with unknown mutation.

[0119] Analysis of the changes in fluorescence intensity of pS6 and p62 per tumor area before and after chemotherapy showed that there did not seem to be a dramatic change in the distribution of p62 in BRCA1 / 2 mutant cases with good prognosis (Figure 5B). On the other hand, pS6 and p62 increased in BRCA1 / 2 wild-type cases with poor progression (Figures 5C-D). These results suggested that the activation of mTORC1-p62 is not only regulated by the respective properties of HGSC such as abnormal RAS / PI3K signals, but also promoted by chemotherapy.

[0120] <mTORC1-SQSTM1 activity controls the NRF2 signal, which is important for poor prognosis in ovarian cancer> p62 has been reported to have several domains that activate pathways such as the NF-κB signal and the NRF2 signal, but the role of p62 in the chemotherapy resistance of ovarian cancer was unclear. To investigate how p62 induces chemotherapy resistance to platinum agents, the expression of pathway genes such as Nqo1 and Ho-1, which are controlled downstream by p62, was analyzed by qPCR using RPM, RPM-p62OE, RPMNP, and RPMNP-p62KO cells. Next, to investigate the effect of everolimus (EVL) on the p62 downstream pathway, the expression of Nqo1 and Ho-1 genes when control or EVL was added to RPMNP cells was analyzed by qPCR.

[0121] The results are shown in Figures 6A to 6C. Figures 6A and B are graphs showing the RNA expression levels of Ho-1 (Figure 6A) and Nqo1 (Figure 6B) in RPM, RPM-p62OE, RPMNP, and RMNP-p62KO cells. FIG. 6C is a graph showing the RNA expression levels of Ho-1 and Nqo1 in RPMNP cells administered with solvent or 1 μM everolimus. In Figures 6A-C, RNA expression was quantified by qPCR.

[0122] As a result, the expression of Nqo1 and Ho-1 was elevated in RPMNP compared to RPM and was regulated in response to the presence of p62 ( Figures 6A,B ). Furthermore, the expression of both Nqo1 and Ho-1 was decreased under EVL treatment (Fig. 6C ).

[0123] Furthermore, to evaluate the impact of the NRF2-ROS response pathway on the progression of HGSC, the correlation between the expression of these genes and patient prognosis was analyzed using the Kaplan Meier Plotter (public literature: Lanczky A, Gyorffy B (2021) J Med Internet Res. https: / / doi.org / 10.2196 / 27633).

[0124] The results are shown in Figures 6D-G and supplementary data Figures 12A-G. 6D and E are Kaplan-Meier curves showing Ho-1(bK286B10) RNA expression and prognosis of ovarian cancer patients. 6F and G are Kaplan-Meier curves showing NQO1 RNA expression and prognosis of ovarian cancer patients. In Figure 6D-G, cases of stage 2+3+4 serous ovarian cancer in which optimal debulking surgery was performed were selected and divided by the lower quartile of RNA expression levels of each gene (lower 25% vs. middle 50% + upper 25%). Kaplan-Meier curves were plotted using a Kaplan-Meier plotter. Figure 6D and F show overall survival (OS), and Figure 6E and G show progression-free survival (PFS). p-values ​​from the Log-rank test are shown.

[0125] Figure 12A shows Kaplan-Meier survival curves showing the "cellular response to reactive oxygen species" score (ROS score) calculated using GSVA and the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 mutation. Specifically, the ROS score was calculated using GSVA, and Kaplan-Meier survival curves were shown for the prognosis of patients with advanced ovarian cancer without BRCA1 / 2 mutation, for the "LOW" group including patients with a ROS score of less than 0 and the "HIGH" group including patients with a ROS score of 0 or more. 12B and C are scatter plots showing the correlation between the ROS score calculated using GSVA and the RAS score (FIG. 12B) or PI3K score (FIG. 12C) in patients with advanced ovarian cancer without BRCA1 / 2 mutations.

[0126] 12D and E are Kaplan-Meier curves showing GPX4 RNA expression and OS (FIG. 12D) or PFS (FIG. 12E) of ovarian cancer patients' prognosis. Figures 12F and G are Kaplan-Meier curves showing CAT RNA expression and prognosis of ovarian cancer patients. In Figure 12D-G, cases of stage 2+3+4 serous ovarian cancer in which optimal debulking surgery was performed were selected and divided by the lower quartile of RNA expression levels of each gene (lower 25% vs. middle 50% + upper 25%). Kaplan-Meier curves were plotted using a Kaplan-Meier plotter. Figure 12D and F show overall survival (OS), and Figure 12E and G show progression-free survival (PFS). p-values ​​from the Log-rank test are shown.

[0127] As a result, high expression of NRF2 downstream genes, such as NQO1 and HO-1, was strongly correlated with poor overall survival (OS) and progression-free survival (PFS) (Figures 6D,E,F,G). Furthermore, when we examined important biological processes in BRCA-WT HGSC patients analyzed by GSVA (Figure 1A), we found that activation of the "cellular response to reactive oxygen species" correlated with poor progression in BRCA-WT HGSC patients, and that there was also an association between RAS / PI3K signaling activation and activation of the "cellular response to reactive oxygen species" (Figures 12A, B, C). Furthermore, analysis using Kaplan Meier Plotter suggested that increased expression of antioxidant genes such as GPX4 was correlated with poor progression in patients with serous ovarian cancer (Figures 12D, E). These results suggested that the mTORC1-p62 linkage activates the NRF2-ROS response pathway, which is important for the poor prognosis of patients with serous ovarian cancer.

[0128] <mTORC1-SQSTM1 activation is a therapeutic target for refractory BRCA1 / 2 wild-type HGSC> As mentioned above, aberrant RAS / PI3K crosstalk in BRCA1 / 2 wild-type HGSC worsens prognosis via the mTORC1-autophagy-SQSTM1-NRF2 pathway. To evaluate the potential of EVL as a combination therapy, we performed both in vitro and in vivo studies. In the in vitro study, the sensitivity of RPMNP cells to TC treatment was compared with or without EVL by colony formation assay.

[0129] The results are shown in Figures 7A to 7D. Figures 7A-B show viable cell staining images (Figure 7A) and their dose-response curves (Figure 7B) of RPMNP cells treated with TC treatment + 0.1% DMSO and RPMNP cells treated with TC treatment + 100 nM everolimus. In Figure 7A, a representative stained well of viable cell staining images is shown. In Figure 7B, cell viability was measured by crystal violet staining and normalized to cells cultured without TC treatment, respectively. p-values ​​by sum-of-squares F-test are shown. Figure 7C shows the treatment regimen for the four groups. Each group received 2 × 10 5 Starting with five nude mice inoculated with RPMNP cells subcutaneously on both sides of each flank, tumor burden was evaluated every three days from day 7. The administration methods were, in principle, CBDCA: 30 mg / kg intraperitoneal administration (ip), PTX: 8 mg / kg ip, and EVL: 5 mg / kg oral administration (po). Figure 7D is a graph showing the growth curve of subcutaneous tumors administered with vehicle alone (Vehicle), CBDCA+PTX (TC), everolimus (EVL), and TC+everolimus (TCEVL). # indicates mouse death. Tumor volumes on days 31 and 40 were statistically analyzed by one-way ANOVA test and Student t-test, respectively; *p value<0.05, **p value<0.01, ***p value<0.001, ****p value<0.0001.

[0130] Similar to the combined effect of EVL and CBDCA, EVL increased the sensitivity of RPMNP cells to TC treatment ( Figures 7A,B ). For in vivo validation, RPMNP cells were subcutaneously inoculated into nude mice and treated with control, TC treatment, EVL, and TC treatment + EVL (Figure 7C). The total dose of CBDCA and PTX administered per 3 weeks was not significantly different from that of TC treatment for human HGSC patients based on the calculated human equivalent volume, but subcutaneous tumors derived from RPMNP cells showed strong resistance to TC treatment on day 31 (Figure 7D). Therefore, RPMNP cells were considered to reflect the characteristics of human refractory HGSC. Consistent with previous in vitro assays, the combination of TC treatment and EVL showed a much higher antitumor effect than TC treatment on day 31 (Figure 7D). Interestingly, EVL alone inhibited tumor growth, and its antitumor effect persisted until day 31. However, subcutaneous tumors treated with EVL alone began to grow gradually thereafter. On day 40, the tumor volume of subcutaneous tumors treated with the TC treatment + EVL regimen was the smallest.

[0131] These in vitro and in vivo evaluations suggested that the combination of TC treatment and mTORC1 inhibitors could be a potent chemotherapy for BRCA1 / 2 wild-type HGSC. FIG. 7E shows a model diagram illustrating the mechanism of action of the p62 biomarker and an mTOR inhibitor in difficult-to-treat ovarian cancer, as predicted from this example.< / qpcr>

Claims

1. A biomarker for determining the prognosis of ovarian cancer, comprising: A biomarker characterized by being at least one of p62 protein and p62 mRNA.

2. A method for determining the prognosis of ovarian cancer, comprising: Detecting the presence or amount of a biomarker, p62 protein and / or p62 mRNA, in a sample obtained from the subject; determining that there is a risk of ovarian cancer if the biomarker is present; A method for determining whether or not a particular

3. The method according to claim 2 , wherein the specimen is at least one of a pathological tissue specimen of ovarian cancer and a blood specimen.

4. the presence or amount of said biomarker in said sample from a particular subject; comparing the presence or amount of the biomarker in the sample obtained from the subject after a period of time; From the above comparison, If the biomarker is no longer detectable or the amount of the biomarker is reduced after the passage of the certain period of time, the prognosis is good. if there is no change in the amount of the biomarker after the period of time, there is no change in prognosis; or determining that the prognosis is poor when the biomarker becomes detectable or increases in amount after the specified time has elapsed; The method according to claim 2 or 3, further comprising:

5. A therapeutic agent for treating ovarian cancer, comprising: A therapeutic agent comprising an mTOR inhibitor.

6. The therapeutic agent according to claim 5, wherein the mTOR inhibitor is everolimus.

7. The therapeutic agent according to claim 5 or 6, wherein the ovarian cancer is an ovarian cancer in which the biomarker according to claim 1 is present.

8. The therapeutic agent according to claim 5 or 6, wherein the ovarian cancer is ovarian cancer with low platinum sensitivity.

9. The therapeutic agent according to claim 5 or 6, wherein the ovarian cancer is a wild-type ovarian cancer of the BRCA1 / 2 gene.

10. The therapeutic agent according to claim 5 or 6, which is used in combination with at least one of paclitaxel, carboplatin and a PARP inhibitor.

Images