Abstract
Purpose:
Advanced/metastatic forms of clear-cell renal cell carcinomas (ccRCC) have limited therapeutic options. Genome-wide genetic screens have identified cellular dependencies in many cancers. Using the Broad Institute/Novartis combined short hairpin RNA (shRNA) dataset, and cross-validation with the CRISPR/Cas9 DepMap (21Q3) dataset, we sought therapeutically actionable dependencies in kidney lineage cancers.
Experimental design:
We identified preferential genetic dependencies in kidney cancer cells versus other lineages. BCL2L1, which encodes the BCL-XL antiapoptotic protein, scored as the top actionable dependency. We validated this finding using genetic and pharmacologic tools in a panel of ccRCC cell lines. Select BCL-XL-dependent (versus independent) cell lines were then transcriptionally profiled to identify biomarkers and mechanistic drivers of BCL-XL dependence. Cell-based studies (in vitro and in vivo) and clinical validations were used to address physiologic relevance.
Results:
Inactivation of BCL-XL, but not BCL-2, led to fitness defects in renal cancer cells, and sensitized them to chemotherapeutics. Transcriptomic profiling identified a "BCL-XL dependency" signature, including an elevated mesenchymal gene signature. A mesenchymal state was both necessary and sufficient to confer increased BCL-XL dependence. The "BCL-XL dependency" signature was observed in approximately 30% of human ccRCCs, which were also associated with worse clinical outcomes. Finally, an orally bioavailable BCL-XL inhibitor, A-1331852, showed antitumor efficacy in vivo.
Conclusions:
Our studies uncovered an unexpected link between cell state and BCL-XL dependence in ccRCC. Therapeutic agents that specifically target BCL-XL are available. Our work justifies testing the utility of BCL-XL blockade to target, likely, a clinically aggressive subset of human kidney cancers. See related commentary by Wang et al., p. 4600.