Homologous recombination repair mutation testing patterns in metastatic castration-resistant prostate cancer: Analysis of real-world data in the United States.

Abstract

102 Background: Homologous recombination repair mutations (HRRms) have predictive/prognostic value in metastatic castration-resistant prostate cancer (mCRPC). While therapies are approved for patients (pts) with HRRm mCRPC, HRRm testing data is limited in the United States (US). This analysis utilized real-world secondary data to assess US HRRm testing patterns in pts with mCRPC. Methods: Data were drawn from the Adelphi Real World Prostate Cancer Disease Specific Programme, a cross-sectional survey with retrospective data collection including medical oncologists (ONC) and urologists (URO) and their mCRPC pts in the US from Nov’22 – Jul'23. ONC/URO reported demographics, disease characteristics, and HRRm testing status for ≥1 of 10 genes of interest ( ATM, ATR, BRCA1, BRCA2, CDK12, CHEK2, FANCA, NBN, PALB2, RAD51C ) for eight consecutively seen adult pts. Analyses were descriptive. Logistic regression assessed variables associated with receiving a HRRm test. Results: Overall, 70 ONC/URO reported on 448 pts, of whom 53% received HRRm testing (ONC 57%, URO 39%). Mean (standard deviation) pt age (years) was 69.2 (8.2) for pts who were tested (vs 70.6 [8.3] not tested). HRRm testing occurred in: 73% with a known family history of HRRm-related (prostate/breast/ovarian/pancreatic) cancer (vs 50% without); 53% with visceral metastases (vs 52% without); 61% with an initial localized diagnosis (Dx; vs 48% regional/47% metastatic); 56% of pts with a Gleason score (GS) ≥8 at initial Dx (vs 43% with <8); and 63% with private insurance (vs 48% with public). Metastatic disease at initial Dx (odds ratio [OR] 0.41), no known family history of HRRm-related cancer (OR 0.40), and a GS <8 at initial Dx (OR 0.51) were significantly associated with lower odds of HRRm testing (all p<0.05; Table). Conclusions: In this analysis, nearly half of pts did not receive a HRRm test. As with prior analyses, disparities in testing were evident for pts with a lower GS at initial Dx, metastatic disease at initial Dx and/or those without a family history of cancer, highlighting a need to increase HRRm testing in the US. Association of covariates with odds ratio for HRRm test. Covariate Odds ratio (95% confidence interval) p value Physician specialty: Urologist vs Medical oncologist 0.49 (0.19–1.30) 0.152 Age (years): ≥65 vs <65 0.91 (0.28–2.90) 0.867 Family history of HRRm-related cancer: No history vs Known history 0.40 (0.18–0.89) 0.025* Ethnicity: Non white vs White 0.94 (0.51–1.73) 0.843 Metastatic sites: Visceral metastases vs No visceral metastases 0.67 (0.35–1.29) 0.237 Gleason score: <8 vs ≥8 0.51 (0.27–0.97) 0.042* Initial diagnosis: Regional vs LocalizedMetastatic vs Localized 0.68 (0.28–1.69)0.41 (0.17–1.00) 0.408 0.049* Insurance status: Private vs Public 1.09 (0.47–2.52) 0.844 *Statistical significance ( P <0.05). HRRm, homologous recombination repair mutation.