Integrative molecular characterization of sarcomatoid and rhabdoid renal cell carcinoma

Abstract

Sarcomatoid and rhabdoid (S/R) renal cell carcinoma (RCC) are highly aggressive tumors with limited molecular and clinical characterization. Emerging evidence suggests immune checkpoint inhibitors (ICI) are particularly effective for these tumors, although the biological basis for this property is largely unknown. Here, we evaluate multiple clinical trial and real-world cohorts of S/R RCC to characterize their molecular features, clinical outcomes, and immunologic characteristics. We find that S/R RCC tumors harbor distinctive molecular features that may account for their aggressive behavior, including BAP1 mutations, CDKN2A deletions, and increased expression of MYC transcriptional programs. We show that these tumors are highly responsive to ICI and that they exhibit an immune-inflamed phenotype characterized by immune activation, increased cytotoxic immune infiltration, upregulation of antigen presentation machinery genes, and PD-L1 expression. Our findings build on prior work and shed light on the molecular drivers of aggressivity and responsiveness to ICI of S/R RCC.

Fig. 1. Genomic characterization of S/R RCC reveals distinctive genomic features.

a Overview of the clinical, molecular, and cell line data. (b) Comparison of S/R vs. non-S/R RCC by mutations & indels, deletions, and amplifications in the CheckMate, OncoPanel, and TCGA cohorts. *q < 0.05 (Fisher’s method meta-analysis of Fisher’s two-sided exact tests); ICI: Immune Checkpoint Inhibitor; IF: Immunofluorescence; IHC: Immunohistochemistry; RNA-seq: RNA-sequencing; S/R: Sarcomatoid/Rhabdoid; TCGA: The Cancer Genome Atlas; WES: Whole Exome Sequencing.

Fig. 2. Transcriptional profiling of S/R RCC reveals the molecular correlates of its poor prognosis and identifies subsets of non-S/R tumors associated with a poor prognosis.

a Heatmap and bar plots of the ssGSEA scores and GSEA normalized enrichment scores for the non-immune “Hallmark” gene sets that were found to be significantly enriched (q < 0.25) in S/R compared to non-S/R RCC in both the TCGA and CheckMate cohorts independently. P-value calculated using a phenotype permutation-based two-sided test with 1000 permutations. Adjustments for multiple testing (50 “Hallmark” gene sets) were made using the false discovery rate (FDR) method. b Kaplan–Meier curves for OS by MYC v1 score within the S/R group of the CheckMate (anti-PD-1 arm) and TCGA (stage IV) cohorts; MYC v1 score dichotomized at the median. Log-rank test two-sided p-value reported without adjustment for multiple testing. c Kaplan–Meier curves for OS by MYC v1 score within the non-S/R group of the CheckMate (anti-PD-1 arm) and TCGA (stage IV) cohorts; MYC v1 score dichotomized at the median of the S/R group. Log-rank test two-sided p-value reported without adjustment for multiple testing. EMT: Epithelial Mesenchymal Transition; MYC v1: MYC Targets Version 1; S/R: Sarcomatoid/Rhabdoid; TCGA: The Cancer Genome Atlas.

Fig. 3. Improved clinical outcomes of S/R RCC tumors on immune checkpoint inhibitors across clinical trial and real-word cohorts.

OS on ICI compared to non-ICI in the a Harvard, b IMDC and c CheckMate S/R RCC cohorts. TTF on ICI compared to non-ICI in the d Harvard and e IMDC S/R RCC cohorts and f PFS in the CheckMate S/R RCC cohort. g Summary table of overall response rate (among evaluable patients) on ICI compared to non-ICI in patients with S/R RCC across the Harvard, IMDC, and CheckMate cohorts. 95% CI: 95% Confidence Interval; Adj. Adjusted; Ever: Everolimus; HR: Hazard Ratio; ICI: Immune Checkpoint Inhibitor; IMDC: International Metastatic Renal Cell Carcinoma Database Consortium; Nivo: Nivolumab; NE: Not Evaluable; OS: Overall Survival; S/R: Sarcomatoid/Rhabdoid. *Adjusted for IMDC (International Metastatic Renal Cell Carcinoma Database Consortium) risk groups, line of therapy, and background histology. **Adjusted for MSKCC (Memorial Sloan Kettering Cancer Center) risk groups.

Fig. 4. The immune-inflamed phenotype of S/R RCC tumors.

a Heatmap and bar plots of the ssGSEA scores and GSEA normalized enrichment scores for the immune “Hallmark” gene sets that were found to be significantly enriched (q < 0.25) in S/R compared to non-S/R RCC in the TCGA and CheckMate cohorts independently. P-value calculated using a phenotype permutation-based two-sided test with 1000 permutations. Adjustments for multiple testing (50 “Hallmark” gene sets) were made using the false discovery rate (FDR) method. b Boxplots of the comparison of CIBERSORTx and Th immune cell populations between S/R and non-S/R RCC, with two-sided Mann-Whitney U test comparisons corrected for multiple comparison testing (q value reported). Only variables which were significant (q < 0.05) in both the CheckMate and TCGA cohorts independently are shown. The CheckMate results are displayed in this figure (N = 39 S/R and N = 247 non-S/R RCC). The center of the box represents the median. The upper and lower hinges represent the 75^th and 25^th percentiles, respectively. The whiskers extend in both directions until the largest or lowest value not further than 1.5 times the interquartile range from the corresponding hinge. Outliers (beyond 1.5 times the interquartile range) are plotted individually. c Bar plot of the comparison of the proportions of tumors that were PD-L1 positive (≥1% on tumor cells) in S/R compared to non-S/R RCC. Two-sided Fisher’s exact test p-value reported (p = 7.4×10⁻⁰⁷). TCGA: The Cancer Genome Atlas.