Abstract B020: Unveiling the tumor-associated microbiome in ovarian cancer: Correlations with clinical features, outcomes, and treatment strategies

Abstract Ovarian cancer (OC) is the leading cause of gynecologic cancer-related mortality in women. Most patients are diagnosed with advanced-stage disease and develop chemotherapy resistance, underscoring the need for biomarkers of early detection and prognostic assessment. Emerging research highlights the tumor-associated microbiome as a promising yet underexplored prognostic biomarker in many non-gynecologic cancers. We aimed to investigate the tumor-associated microbiome in OC, focusing on its differential composition across disease stages and its association with patient survival using bulk tumor RNA sequencing data. Utilizing data from the Oncology Research Information Exchange Network (ORIEN) collected from consenting patients under the Total Cancer Care Protocol (NCT03977402), we analyzed 758 ovarian tumor RNA-seq samples from 21 cancer centers across the United States. The Microbiome Research Interest Group processed RNAseq data through exotic v2. 1 and normalized the counts based on total sequencer output. We performed DESeq2 analysis to explore the differential microbial abundance across tumor subtypes. Kaplan–Meier survival curves with log-rank tests (mt. surv R package) evaluated associations between microbe abundance and overall survival (OS). Multivariate Cox proportional hazards models were adjusted for confounding clinical variables. P-values were corrected for multiple comparisons using the Benjamini–Hochberg method. The final data set included 182 early-stage and 502 advanced-stage tumors; of those, the majority were high-grade serous (n = 474) and 38 low-grade serous cases. Microbial abundance profiling revealed distinct tumor subtypes’ microbial signatures. In advanced-stage tumors, Pseudomonas (log2FC = 1. 64, q = 0. 025) and Bacteroides (log2FC = 2. 84, q = 0. 026) were significantly enriched, whereas no microbes were significantly enriched in early-stage tumors. High-grade serous ovarian tumors showed many significantly enriched microbes, but none were enriched in low-grade serous. In particular, Bacillus demonstrated the most substantial increase in abundance (log2FC = 12. 04, q = 0. 001), followed by Escherichia (log2FC = 9. 67, q = 0. 001), Candida (log2FC = 8. 90, q = 0. 001), Pseudomonas (log2FC = 3. 47, q = 0. 001) and Fusarium (log2FC = 3. 51, q = 0. 001). High Candida abundance was significantly associated with poorer overall survival in OC patients (HR = 1. 77; 95% CI: 1. 35–2. 33; p 0. 001), independent of age, disease stage, and menopausal status. Our findings reveal tumor subtype-specific microbial profiles in OC, with Pseudomonas strongly linked to advanced-stage disease and Candida abundance associated with poorer survival outcomes. Pseudomonas has been associated with obesity-related inflammation in tissues. Candida has been associated with treatment resistance and hypoxia and has been linked with worse survival in colorectal cancer patients. These findings underscore the potential influence of the tumor-associated microbiome on patient outcomes and highlight its emerging significance as a predictive factor in OC. Citation Format: Yogita Mehra, Rebecca Hoyd, Daniel Spakowicz, Aik Choon Tan, Therese Bocklage, Ahmad A. Tarhini, Bodour Salhia, Melanie Rutkowski, Greg Riedlingeer, Song Yao, Ashiq Masood, Craig Shriver, Debra L. Richardson, Dinesh Pal Mudaranthakam, Carlos Chan, Jesus Gonzalez Bosquet, Michelle Churchman, Nicole Marjon, Robert J. Rounbehler, Julia Chalif, David M. O’Malley, Laura Chambers. Unveiling the tumor-associated microbiome in ovarian cancer: Correlations with clinical features, outcomes, and treatment strategies abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Ovarian Cancer Research; 2025 Sep 19-21; Denver, CO. Philadelphia (PA): AACR; Cancer Res 2025;85 (18Suppl): Abstract nr B020.