Moderated Poster
Presentation Authors: Renzo DiNatale*, Gowtham Jayakumaran, Mazyar Ghanaat, Julian Marcon, Rose Brannon, Hikmat Al-Ahmadie, Samson Fine, Anuradha Gopalan, Joseph Sirintrapun, Satish Tickoo, Maria Arcila, Robert Motzer, Jonathan Coleman, Paul Russo, Timothy Chan, Victor Reuter, Ari Hakimi, Yingbei Chen, New York, NY
Introduction: Unclassified renal cell carcinoma (uRCC) constitutes a large portion of aggressive kidney tumors that shows limited response to standard therapies. Clinicopathologic parameters or biomarkers to stratify this neoplasia are lacking. Our group recently reported a molecular analysis on 62 uRCCs (discovery cohort – [DC]) and identified distinct molecular subsets. We aim to validate this molecular stratification in an independent clinical cohort and further investigate molecular features that could help refine prognostication and guide management.
Methods: The diagnosis of uRCC in all cases was rendered at our institution by an experienced GU pathologist, based on the current WHO criteria. Tumor samples from 75 patients (validation cohort) with uRCC were profiled with our institutional and previously-validated next-generation sequencing panel (MSK-IMPACT®). Germline testing results were available in 37 cases. Allele-specific copy number analysis was used to characterize putative driver copy-number alterations. Finally, clonality estimates computed using the allelic frequencies, purity and ploidy estimates were used to further refine the molecular groupings. Survival analysis was performed using the Kaplan-Meier method and Cox regression models.
Results: The frequency of somatic mutations in the validation set was highly consistent with findings in the DC. Germline alterations were detected in previously unsuspected cases. The integrative analysis supported the presence of 5 major molecular subsets. Univariate cox regression analysis showed a significantly worse prognosis associated with the NF2 group. Clonality analysis confirmed the previously-described putative drivers as early events during tumor development. Rare cases with alterations indicating sensitivity or resistance to immunotherapy were also identified.
Conclusions: Molecular stratification of high-grade uRCC helps improve prognostication and provides rationale for different management strategies.
Source of Funding: NIH/NCI Cancer Center Support Grant P30 CA008748
