Genomic landscape of metastatic clear cell renal cell carcinoma (mccRCC): A retrospective real-world analysis from Flatiron Health (FH)-Foundation Medicine, Inc. (FMI) clinico-genomic database (CGDB).

4552 Background: Although the genomic landscape of mccRCC has been already documented in the literature, previous clinico-genomic studies have been constrained by small sample sizes, or a narrow focus on specific genes. In this large-scale study, we aimed at investigating the real-world frequency and co-occurrence of genomic alterations (GAs) detected in mccRCC samples. Methods: This study used the nationwide (US-based) de-identified FH-FMI CGDB. The de-identified data originated from approximately 280 US cancer clinics (~800 sites of care). Eligible patients (pts) were adults diagnosed with mccRCC, with FH-EHR records between January 1, 2011 and December 31, 2022. GAs were identified in the same time span via comprehensive genomic profiling using FMI tests on tumor tissue and blood samples. Co-occurrence and mutual exclusivity were assessed using Fisher's exact test, with multiple testing correction. Clustering analysis was performed on the gene network graph derived from significant interactions. Results: Out of 1,409 pts with renal cell carcinoma, 923 (65.5%) mccRCC were considered, with a median age at the diagnosis of metastatic disease of 61 years (IQR: 54-69). The majority were male (70.3%) and white (71.1%). Nearly 89% (822) received systemic therapy, with a median number of 2 treatment lines (IQR: 1-4). A total of 958 samples were sequenced, of which 92.2% were performed on tissue, and 7.8% on blood. GAs (81.1% short variant, 17.1% copy number variant, and 1.8% rearrangement) were found in 246 different genes, with a median number of mutated genes per pts of 3 (IQR: 2-5), while in 22 (2.4%) pts, no GAs were detected. VHL (69.7%), PBRM1 (39.9%), SETD2 (24.2%), CDKN2A (19.0%), BAP1 (15.6%), CDKN2B(14.8%), TP53 (14.1%), KDM5C (13.5%), PTEN (11.2%), and TERT (8.7%) were the top-10 mutated genes. Tumor Mutational Burden (TMB) was evaluated in 639 tissue samples, with a median value of 2.6 mut/Mb; 1.6% only of the tumors studied had high TMB. Out of the 796 samples examined for microsatellite instability (MSI), only 0.6% harbored MSI-intermediate, and none showed MSI-high. Co-occurrence analysis revealed 32 significant gene interactions. CDKN2A/B were significantly associated with MTAP, BAP1, and NF2, while PBRM1 was enriched with VHL, SETD2, KDM5C, and MTOR. TSC1co-occurred with TERT, TP53, and VHL.Three clusters of co-occurring genes were identified: (1) VHL, SETD2, PBRM1, KDM5C, MTOR, and NFE2L2; (2) TP53, TSC1, and TERT; (3) CDKN2A, CDKN2B, BAP1, NF2, MTAP, and ARID1A. Mutually exclusive patterns were observed between cluster 1 and cluster 3 genes. Notably, BAP1 was mutually exclusive with PBRM1. Conclusions: This is the largest analysis of the genomic landscape of mccRCC. We identified 3 clusters of co-occurring genes, whose prognostic/predictive value will be object of future research.