Genetic Analysis Reveals the Prognostic Significance of the DNA Mismatch Repair Gene MSH2 in Advanced Prostate Cancer

DNA damage repair is frequently dysregulated in advanced prostate cancer and has been linked to cancer susceptibility and survival outcomes. The aim of this study is to assess the influence of genetic variants in DNA damage repair pathways on the prognosis of prostate cancer. Specifically, 167 single nucleotide polymorphisms (SNPs) in 18 DNA damage repair pathway genes were assessed for association with cancer-specific survival (CSS), overall survival (OS), and progression-free survival (PFS) in a cohort of 630 patients with advanced prostate cancer receiving androgen deprivation therapy. Univariate analysis identified four SNPs associated with CSS, four with OS, and two with PFS. However, only MSH2 rs1400633 C > G showed a significant association upon multivariate analysis and multiple testing adjustments (hazard ratio = 0.75, 95% confidence interval = 0.63–0.90, p = 0.002). Furthermore, rs1400633 risk allele C increased MSH2 expression in the prostate and other tissues, which correlated with more aggressive prostate cancer characteristics. A meta-analysis of 31 gene expression datasets revealed significantly higher MSH2 expression in prostate cancer than in normal tissues (p < 0.001), and this high expression was associated with a poor prognosis of prostate cancer (p = 0.002). In summary, we identified MSH2 rs1400633 as an independent prognostic biomarker for prostate cancer survival, and the association of MSH2 with cancer progression lends relevance to our findings.

A calibrated cell-based functional assay to aide classification of MLH1 DNA mismatch repair gene variants

PURPOSE: Functional assays provide important evidence for classifying the disease significance of germline variants in the DNA mismatch repair genes. We sought to develop a cell-based approach for testing the function of variants of uncertain significance (VUS) in the MLH1 gene. METHODS: Using CRISPR gene editing, we knocked-in MLH1 VUS into the endogenous MLH1 loci in human embryonic stem cells. We examined their impact at the RNA and protein level, including their ability to maintain stability of microsatellite sequences and instigate a DNA damage response. We calibrated these assays by testing well-established pathogenic and benign control variants. RESULTS: Five VUS resulted in functionally abnormal protein, 15 VUS resulted in functionally normal protein, and one VUS showed mixed results. Furthermore, we converted the functional outputs into a single odds in favor of pathogenicity score for each VUS. CONCLUSION: Our CRISPR-based functional assay successfully models phenotypes observed in patients in a cellular context. Using this approach, we generated evidence for or against pathogenicity for utilization by variant classification expert panels. Ultimately, this information will assist in proper diagnosis and disease management for suspected Lynch syndrome patients.

TMOD-08. DEVELOPING DERIVATIVE GBM PDX, IN VIVO, FROM TREATMENT NAÏVE SOURCES

PURPOSE Post-therapy recurrent glioblastoma (GBM) patient-derived xenografts (PDX), developed from corresponding treatment-naïve PDX, could serve as useful resources for identifying therapeutics with activity against recurrent GBM. The goal of this study was to determine whether treatment-naïve intracranial GBM PDX, in mice receiving radiotherapy (RT) and/or temozolomide (TMZ), acquire the same mutations that occur in post-RT+TMZ GBMs from patients. METHODS Luciferase-modified, treatment-naïve GBM PDX were engrafted in the brains of athymic nude mice, followed by treatment with RT only (2 Gy/day x 5), TMZ only (10 mg/kg/day x 5), or RT+TMZ. Bioluminescence imaging was used to monitor intracranial tumor growth, response to treatment, and recurrence from treatment. Some mice with recurrent tumors received additional TMZ treatment. When mice became symptomatic, intracranial tumors were resected and engrafted subcutaneously in a new mouse host, then sequentially propagated subcutaneously into additional host mice. After the third passage, whole-exome sequencing (WES) was done, comparing post-therapy with treatment-naïve PDX sequence variants. RESULTS Analysis of PDX WES showed the following: 1) TMZ consistently caused more genes to incur coding sequence mutations than RT, as much as 13x more; 2) TMZ-treated tumor mutations were mostly G-C to A-T transitions (71-92%), consistent with the known mutagenic effect of TMZ; and 3) post-therapy PDX acquire similar mutations as do recurrent GBMs in patients, for example involving DNA mismatch repair gene MSH6. One of the derivative PDX with MSH6 mutation has been retested for response to RT and TMZ, with results showing its having become TMZ, but not RT resistant. CONCLUSIONS The mutation profiles of RT+TMZ-treated PDX are similar to those reported for GBMs that recur after RT+TMZ in patients. The new PDX resources described here may prove useful for identifying effective treatments against recurrent GBM.

302 Characterization of TIGIT and PVR expression in colorectal liver metastases

BackgroundMetastatic colorectal cancer (CRC) is common and lethal and generally not responsive to current immunotherapies. We hypothesize that efficacious T cell-based immunotherapy can be developed for this malignancy, provided that immune checkpoints relevant to liver metastasis, the first site of disease progression, are targeted. Here, we characterized CRC liver metastases by RNAseq, FACS and in vitro functional assays to identify candidate immune checkpoints.MethodsWe performed deep RNAseq clustering and differential gene expression analysis on bulk RNA extracted from 52 mismatch repair gene proficient CRC liver metastases. By multiparameter FACS, we analyzed the expression of candidate immune checkpoints in cell suspensions derived from 18 liver metastases, matched non-tumoral livers, and pre-operative PBMCs. We evaluated IFN-γ (ELISA) secretion and tumor lysis (Incucyte) of tumor-infiltrating T lymphocytes (TILs) expanded from liver metastases stimulated by autologous cancer cells with or without monoclonal antibodies blocking candidate immune checkpoints.ResultsOut of 52 metastases, 21 (40.3%) clustered as immune reactive (IR) defined by concurrent high expression of transcripts related to antigen processing, immune cell lineage, immune checkpoints, interferon-gamma response, cytokines, and chemokines, whereas 25 (48.1%) were classified as non-IR. Of all inhibitory ligands assessed, PVR and PVRL2 had the highest expression, both in IR and non-IR metastases, and higher than PD-L1 and PD-L2 expression. The expression of corresponding receptors TIGIT and CD226 was significantly higher in IR compared to non-IR metastases, at absolute levels higher than PD-1. By FACS analysis, PVR and PVRL2 expression by tumor-infiltrating myeloid and tumor cells was higher than PD-L1 and PD-L2 expression. High PVR expression was also found in hepatocytes, liver macrophages and circulating monocytes in the same patients. In TILs, TIGIT was significantly overexpressed in activated CD4+CD25+ (74.8 ±3.0%) and CD8+CD25+ (68.7 ± 8.4%) compared to resting CD25neg T cells, an expression pattern that was not seen for PD-1 or in T cells infiltrating the liver or circulating in the blood. The majority of cancer cell lines derived from liver metastases expressed PVR, but low levels of PD-L1. TIL clones expanded from liver metastases expressed TIGIT at various levels inducible by TCR stimulation. Upon co-culture with autologous cancer cell lines, TIL clones were more lytic and secreted more IFN-γ in presence of anti-TIGIT blocking antibody.ConclusionsBy expression and functional data, the TIGIT/PVR immune suppressive axis appears as a biologically promising target for the development of immunotherapy in patients with CRC metastatic to the liver.AcknowledgementsThis work is supported by Bristol Myers Squibb and by the Quebec Cancer Consortium. A.B. holds a postdoctoral scholarship award from the Institut du cancer de Montréal. S.T. holds a Junior 2 clinical-scientist salary award from the Fond de recherche Santé-Québec. The University of Montreal Roger des Groseillers Research Chair in hepatopancreatobiliary surgical oncology supports the biobanking and clinicopathological database associated with this project.Ethics ApprovalInstitutional review board approvals were obtain to conduct this project (16.262) and all patients provided informed consent to contribute to this project with biospecimens and clinicopathological data (09.237).

Prevalence and spectrum of DNA mismatch repair gene variation in the general Chinese population

BackgroundIdentifying genetic disease-susceptible individuals through population screening is considered as a promising approach for disease prevention. DNA mismatch repair (MMR) genes including MLH1, MSH2, MSH6 and PMS2 play essential roles in maintaining microsatellite stability through DNA mismatch repair, and pathogenic variation in MMR genes causes microsatellite instability and is the genetic predisposition for cancer as represented by the Lynch syndrome. While the prevalence and spectrum of MMR variation has been extensively studied in cancer, it remains largely elusive in the general population. Lack of the knowledge prevents effective prevention for MMR variation–caused cancer. In the current study, we addressed the issue by using the Chinese population as a model.MethodsWe performed extensive data mining to collect MMR variant data from 18 844 ethnic Chinese individuals and comprehensive analyses for the collected MMR variants to determine its prevalence, spectrum and features of the MMR data in the Chinese population.ResultsWe identified 17 687 distinct MMR variants. We observed substantial differences of MMR variation between the general Chinese population and Chinese patients with cancer, identified highly Chinese-specific MMR variation through comparing MMR data between Chinese and non-Chinese populations, predicted the enrichment of deleterious variants in the unclassified Chinese-specific MMR variants, determined MMR pathogenic prevalence of 0.18% in the general Chinese population and determined that MMR variation in the general Chinese population is evolutionarily neutral.ConclusionOur study provides a comprehensive view of MMR variation in the general Chinese population, a resource for biological study of human MMR variation, and a reference for MMR-related cancer applications.

DNA Repair Gene Polymorphisms and Chromosomal Aberrations in Exposed Populations

DNA damage and unrepaired or insufficiently repaired DNA double-strand breaks as well as telomere shortening contribute to the formation of structural chromosomal aberrations (CAs). Non-specific CAs have been used in the monitoring of individuals exposed to potential carcinogenic chemicals and radiation. The frequency of CAs in peripheral blood lymphocytes (PBLs) has been associated with cancer risk and the association has also been found in incident cancer patients. CAs include chromosome-type aberrations (CSAs) and chromatid-type aberrations (CTAs) and their sum CAtot. In the present study, we used data from our published genome-wide association studies (GWASs) and extracted the results for 153 DNA repair genes for 607 persons who had occupational exposure to diverse harmful substances/radiation and/or personal exposure to tobacco smoking. The analyses were conducted using linear and logistic regression models to study the association of DNA repair gene polymorphisms with CAs. Considering an arbitrary cutoff level of 5 × 10–3, 14 loci passed the threshold, and included 7 repair pathways for CTA, 4 for CSA, and 3 for CAtot; 10 SNPs were eQTLs influencing the expression of the target repair gene. For the base excision repair pathway, the implicated genes PARP1 and PARP2 encode poly(ADP-ribosyl) transferases with multiple regulatory functions. PARP1 and PARP2 have an important role in maintaining genome stability through diverse mechanisms. Other candidate genes with known roles for CSAs included GTF2H (general transcription factor IIH subunits 4 and 5), Fanconi anemia pathway genes, and PMS2, a mismatch repair gene. The present results suggest pathways with mechanistic rationale for the formation of CAs and emphasize the need to further develop techniques for measuring individual sensitivity to genotoxic exposure.