Dr Pal on Genomic Activity With Adjuvant Atezolizumab vs Placebo in High-Risk Resected RCC

"[Since we have access to patients with tumors that have progressed], I think we can really make some far-reaching conclusions around [the underlying mechanisms of adjuvant atezolizumab in this population]. We, for instance, [identified] that MHC-I pathways appear to be downregulated at the time of recurrence, and that’s a putative mechanism for resistance."

Sumanta Kumar Pal, MD, FASCO, chair of the Kidney and Bladder Cancer Disease Team, co-director of the Kidney Cancer Program, and a professor and vice chair of Academic Affairs in the Department of Medical Oncology & Therapeutics Research at City of Hope, discussed the significance of genomic findings from the phase 3 IMmotion010 trial (NCT03024996), a randomized study evaluating adjuvant atezolizumab (Tecentriq) vs placebo in patients with high-risk resected renal cell carcinoma (RCC).

The trial featured the first large-scale genomic and transcriptomic evaluation of both pre- and post-immunotherapy tumor tissue in the adjuvant RCC setting, Pal began. Among baseline samples, a range of molecular profiles were identified, including angiogenic signatures and a subset of tumors exhibiting high T-effector cell immune gene expression. Within the KIM-1–high subgroup (n = 151 for atezolizumab; n = 149 for placebo), patients with tumors that had a high T-effector cell signature had longer disease-free survival (DFS) with atezolizumab compared with placebo, suggesting the predictive value of this molecular subset, he said.

Within the KIM-1–high population, the median DFS was not estimable (NE) in the atezolizumab arm compared with 21.16 months in the placebo arm. Treatment with atezolizumab resulted in a hazard ratio (HR) for DFS of 0.72 (95% CI, 0.52-0.99), indicating a 28% reduction in the risk of recurrence or death in this biomarker-defined population.

In contrast, among patients in the KIM-1–low subgroup (n = 229 for atezolizumab; n = 223 for placebo), no DFS benefit was observed with atezolizumab vs placebo. The median DFS was 57.23 months for patients receiving atezolizumab and was NE in the placebo arm. The HR was 1.12 (95% CI, 0.88-1.63), indicating no significant difference in DFS between treatment arms in this subgroup.

In patients with available post-progression tissue, recurrent tumors showed increased stromal and proliferation gene signatures regardless of treatment, and tumors treated with atezolizumab showed downregulation of MHC-I expression. According to Pal, these findings suggest a potential immune evasion mechanism via impaired antigen presentation, which could contribute to resistance to PD-L1 blockade in the adjuvant setting.

Serum KIM-1 emerged as the strongest individual biomarker for predicting outcomes with atezolizumab. Although certain molecular subsets, such as high T-effector cell–expressing tumors, may provide context regarding immune responsiveness with atezolizumab, KIM-1 levels consistently correlated with clinical benefit and risk of recurrence across the cohort.