Genomic alterations in metastatic renal cell carcinoma (RCC): Impact on prognosis and response to frontline therapy.

e16529 Background: Genomic alterations can serve as prognostic and predictive biomarkers for many malignancies. We report a large dataset to describe the impact of genomic alterations on outcomes and response to frontline therapy for metastatic RCC. Methods: Tumor samples from pts with metastatic RCC underwent next-gen sequencing of DNA at centralized CLIA-certified lab (Caris Life Science; Phoenix, AZ). Real-world clinical outcomes were inferred from insurance claims data. As a surrogate for PFS, time-on-treatment (TOT) was evaluated from the start to last date of therapy. OS was calculated from the start date of initial therapy to death or last contact. hazard ratio (HR) calculated using the Cox proportional hazards model, and P values calculated using the log-rank test. Results: Pathogenic mutations in VHL, SETD2, TP53, PBRM1, and BAP1 were evaluated. 185 of 1786 RCC pts (10.4%) had a VHL mutation and TOT for those with a VHL mutation was 11.0 mos vs. 8.5 mos for those without (HR = 0.78, p = 0.042). 68 pts (3.8%) had a SETD2 mutation, and TOT for pts with a mutation was 14.3 mos vs. 9.7 mos for those without (HR = 0.67, p = 0.005). There was no difference in TOT for pts with BAP1 , PBRM1 , or TP53 mutations. For pts with a VHL mutation, OS was 32.0 mos vs. 26.6 mos for those without (HR = 0.67, p = 0.03). There was no statistically significant difference in OS for the other genomic alterations. When treated with frontline IO/IO, pts with a VHL mutation had TOT of 11.3 mos vs. 10.6 mos when treated with frontline IO/TKI (HR = 0.71, p = 0.026). For pts with a BAP1 mutation, TOT was 13.6 mos for those treated with frontline IO/IO vs. 5.1 mos for those treated with frontline IO/TKI (HR = 2.18, p = 0.013). There were no differences in OS for pts treated with IO/IO vs. IO/TKI for any of the genomic alterations. Table 1 outlines TOT and OS by mutation present and therapy received. Conclusions: The presence of a VHL mutation was associated with longer TOT and OS. Pts with VHL and BAP1 mutations treated with frontline IO/IO as opposed to IO/TKI had improved TOT, though no difference in OS. Genomic alterations in RCC may help guide prognosis and choice of frontline therapy. Future work may identify how combinations of these alterations affect prognosis and therapy response. Time on treatment (mos) Treatment to last contact (mos) Mutation Present Mutation Absent HR p-value Mutation Present Mutation Absent HR p-value Genomic mutation:TP53VHLPBRM1SETD2BAP1 9.311.011.314.38.5 9.98.59.99.710.1 1.10.771.10.671.06 0.340.040.470.0050.74 27.132.032.032.030.7 30.126.628.927.328.8 0.800.671.181.211.07 0.370.030.410.360.77 Present mutation: 1L therapy IO/IO 1L therapy IO/TKI HR p-value 1L therapy IO/IO 1L therapy IO/TKI HR p-value TP53VHLPBRM1SETD2BAP1 9.211.312.212.713.6 12.410.69.816.65.1 0.710.711.30.922.18 0.300.030.250.730.01 27.130.732.032.044.1 24.9NANANA24.9 1.661.160.790.731.71 0.310.570.550.420.24