Abstract B046: Homologous recombination DNA repair dependent survival associations in high-grade serous ovarian carcinoma: an Ovarian Tumor Tissue Analysis consortium study
Bibliographic record
Abstract
Abstract Despite homologous recombination (HR) deficiency being an important predictor of survival in high-grade serous ovarian carcinoma (HGSC), its status is rarely considered in large-scale studies of prognostic signatures and biomarkers. We sought to develop a gene expression-based classifier of HR deficient (HRD) HGSCs, suitable for archival FFPE tumor samples, to determine whether established prognostic markers were dependent on HR status. Using 361 HGSCs from The Cancer Genome Atlas (TCGA) where HR status was defined by the genomic gold standard, a HRD score, we trained and tested a ridge penalized logistic regression model. This 218-gene mRNA expression signature was then applied in 6002 HGSCs from the Ovarian Tumor Tissue Analysis (OTTA) consortium to predict the probability of a sample being HRD. In this large cohort of FFPE samples, we sought to understand the relationship between HR and overall survival. Applying Kaplan-Meier analyses and multivariate Cox proportional hazard regression modelling, adjusting for age and stratified by stage and study, we examined the relationship between HR status and the PrOTYPE subtype, SPOT prognostic score and other known prognostic markers in HGSC. Our machine learning model was 84% accurate in a held-out subset of the TCGA (n=73). Of the 6002 OTTA HGSCs, 40.4% were predicted to be HRD. These cases were significantly younger, were enriched for early stage (FIGO I/II) and had improved outcomes when compared to HR proficient (HRP) cases (hazard ratio [HZR]=0.71, 95% confidence interval [CI]=0.66-0.76, p<2x10-16). A linear increase in survival time was observed as the probability of a sample being HRD increased. HGSCs with >75% likelihood of being HRD had a median 5-year survival of 49.3% (95%CI=46.2-52.6%) which was a 17.7% increase from those with <25% likelihood. Both PrOTYPE and SPOT were significantly associated with survival, independent of HR status (p<2x10-16 and p<2x10-16 respectively). Notably the probability of being HRD was negatively correlated with an unfavorable SPOT score (R=-0.32, p<2x10-16), suggesting that SPOT partially captures the survival advantage of HR deficiency. Several individual biomarkers, assessed in 1248 to 2858 cases, showed HR dependent survival associations. HRD HGSCs with high GMNN expression had significantly improved, long-term outcomes (4-10years, HZR=0.39, 95%CI=0.20-0.78, p=0.01), and may be a marker of exceptional survival, while AR negative HRD cases had poorer outcomes, earlier in disease progression (0-4years, HZR=1.43, 95%CI=1.03-1.98, p=0.03). Significantly poorer prognoses were observed in HRP cases expressing PTEN (HZR=1.39, 95%CI=1.04-1.87, p=0.03), and in HRD cases with high expression of MyD88 (HZR=1.21, 95%CI=1.01-1.45, p=0.04). RB1 and FOXJ1 protein expression, and CD8+ T cell abundance were prognostic irrespective of HR status. Our study contributes to understanding the association between HR and HGSC survival in a statistically powered cohort and demonstrates heterogeneity in outcomes through clinically relevant prognostic signatures and biomarkers. Citation Format: Ashley L. Weir, Derek Chiu, Aline Talhouk, Mike S. Anglesio, Dale W. Garsed, Anna DeFazio, Holly H. Harris, Joellen M. Schildkraut, Andrew Berchuck, Ellen L. Goode, Paul D.P. Pharoah, David Huntsman, David Bowtell, Jen A. Doherty, Nadia M. Davidson, Martin Köbel, Susan J. Ramus. Homologous recombination DNA repair dependent survival associations in high-grade serous ovarian carcinoma: an Ovarian Tumor Tissue Analysis consortium study [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(18_Suppl):Abstract nr B046.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".