Integrated molecular analysis of Asian ovarian cancer: Gene expression and whole exome sequencing analyses from the iPocc Translational Research study (TriPocc).

Abstract

5562 Background: iPocc is a randomized, multicenter international study comparing the efficacy of first line intravenous (IV) dose-dense paclitaxel and intraperitoneal (IP) carboplatin with standard IV dose-dense paclitaxel and carboplatin in women with epithelial ovarian, fallopian tube or primary peritoneal cancer. TriPocc is the translational research protocol evaluating the molecular and predictive biomarkers for differences in treatment outcome. Methods: Fresh frozen (FF) biopsies (N = 123) collected during surgeries from participating institutions in Japan and Singapore were included into TriPocc. Briefly, DNA and RNA were extracted from FF biopsies and paired germline DNA were extracted from the peripheral blood mononuclear cells (PBMC). Whole-transcriptome profiling (WTP) was done by Affymetrix Human Transcriptome Array (HTA) 2.0 (N = 120) for gene expression molecular subtypes (GEMS). Whole-exome sequencing (WES) was done by Nextera Rapid Capture Custom Enrichment Kit for library preparation followed by Illumina NextSeq 500 for paired somatic and germline sequencing (N = 116). The Broad Institute Genome Analysis Tool Kit (GATK) V3.7 was used for variant calling. Results: High grade serous carcinoma (HGSC) was the most frequent histotype (N = 72; 60%), followed by endometrioid (N = 18; 15%) and clear cell carcinoma (N = 16; 13.3%). Mes was the dominant GEMS (N = 43; 35.8%), followed by Epi-B (N = 29; 24.2%), Stem-A (N = 27; 22.5%), Stem-B (N = 17; 14.2%), and Epi-A (N = 4; 3.3%). Unique mutations were discovered in HGSC. Top significant mutations identified (q-value lower than 1xE-9) included ITGA4, UCN3, COL4A3, ACP2, SEH1L, AGL, FAM26F, OR5H15, EXO1, TEKT3, ZNF720, IFI44L, MERTK, TP53, DHX57, TTC31, which were not reported previously in the TCGA cohort. BRCA1/2 mutations (26.1% of all samples) were found in 35.2% of HGSC and 13.5% of non-HGSC, which is higher than previously reported. Conclusions: The integrated genomic analysis from TriPocc has enabled the discovery of potentially novel molecular alterations in Asian ovarian cancers. The predictive value of these molecular alterations will be assessed once the iPocc outcome data are mature.