Background:From 2010-2019, eoRCC had 1,793 excess cases, the third largest increase after breast and colorectal cancers. However, molecular drivers and survival implications for eoRCC remain unclear. This study uses a multi-omic real-world dataset to compare somatic, germline, immune, transcriptomic, and clinical features of eoRCC vs. typical-onset RCC (toRCC).
Methods:The Tempus Lens Platform (Tempus AI, Inc., Chicago, IL) was used to query the multimodal de-identified database to identify RCC patients (pts) with xT (DNA) and xR (RNA) sequencing, tumor purity ≥30%. Pts with clear cell RCC (n = 1,842) were stratified by age at diagnosis as eoRCC (18 to ≤ 46 yrs., n = 211, 11.5%) and toRCC ( > 46 yrs., n = 1,631, 88.5%). We compared demographic, clinical, somatic and germline genomic features, transcriptomic profiles, and immune infiltration (quanTIseq). Differential expression and pathway enrichment used log2 (TPM+1) normalized RNA-seq data against hallmark and GOBP gene sets. Real-world time to next treatment (rwTTNT) and overall survival (rwOS) were measured from first-line (1L) therapy and assessed with Cox proportional hazards models and Kaplan-Meier methods.
Results:Pts with eoRCC were more often Black/African American (8.7% vs. 3.8%, p = 0.029) and Hispanic/Latino (32% vs. 17%, p < 0.001) compared to pts with toRCC. Germline alterations occurred in 11.3% eoRCC pts and 7.3% in toRCC (q = 0.5), with no age-enriched variants. eoRCC had lower somatic alteration rates in PBRM1 (21% vs. 43%, q < 0.001), SETD2 (13% vs. 25%, q < 0.001), and KDM5C (6% vs. 12%, q = 0.025), while VHL alteration rates were similar (76% in both). Transcriptomic profiling showed eoRCC upregulated pathways related to hypoxia and VEGFR signalling (q < 0.05). In contrast, immune-related pathways (IFNgamma/IFNalpha, IL6/JAK/STAT) and cell cycle regulators (E2F, mTOR) were downregulated (q < 0.05). Immune cell composition analysis indicated reduced infiltration of M1 macrophages in eoRCC vs. toRCC (p = 0.003). In eoRCC, median rwTTNT for 1L treatment trended shorter among pts treated with ipilimumab+nivolumab (n = 13) compared to VEGF+IO (n = 12) (8.8 mos. vs. not achieved; HR 0.23 (95%CI 0.05 – 1.09); p = 0.07). Median rwOS was not significantly different between eoRCC pts treated with ipilimumab+nivolumab (n = 13) vs. VEGF+IO (n = 14) (31.6 vs 25.2 mos.; HR 1.12 (95%CI 0.32- 3.87); p = 0.86).
Conclusions:Age-stratified analysis showed that eoRCC has distinct features with more underrepresented racial/ethnic groups. eoRCC is associated with distinct biological phenotypes, exemplified by hypoxia/VEGF pathway upregulation and downregulation of immune, inflammatory, and cell cycle pathways. These hypothesis-generating findings highlight potential age-specific therapeutic targets and the need to explore age-adapted treatment strategies for this unique and growing population.
VIEW THE PUBLICATION