Radiologic, Pathologic, and Deep Learning Predictors of Response to Immune Checkpoint Blockade in Renal Cell Carcinoma Patients Undergoing Post-treatment Nephrectomy

Response assessment of primary kidney tumors in the consolidation cytoreductive and neoadjuvant settings offers a unique opportunity to inform postoperative adaptive treatment strategies. However, systematic analyses of nephrectomy specimens after immune checkpoint inhibitor (ICI) therapies are lacking. We retrospectively identified consecutive renal cell carcinoma (RCC) patients with locoregionally advanced or metastatic RCC who received at least 1 cycle of ICI-containing doublet therapy before nephrectomy at the University of Texas Southwestern Medical Center Kidney Cancer Program (2017-2024). Radiologic and pathologic features were centrally reviewed and correlated with clinical outcomes: freedom from start of next systemic therapy (FFNT) in cytoreductive patients and metastasis-free survival in neoadjuvant patients. Pathologic response to ICI as determined by tumor cell death and fibrosis, with hypocellular areas and increased immune infiltrates, features we used to develop deep learning (DL) models. We leveraged these DL models to provide an objective quantitative complement to pathologist-assessed regression and to quantitate immune infiltrates. Among 99 patients (cytoreductive nephrectomy, n = 66; neoadjuvant nephrectomy, n = 33), radiologic tumor shrinkage ≥30% (P = .0036) and the extent of ICI-induced pathologic regression as assessed by central pathology review (hazard ratio, 0.97; 95% CI, 0.95-0.99; P = .0023) were significantly associated with prolonged FFNT, with similar trends in the neoadjuvant nephrectomy cohort. The presence of coagulative tumor necrosis continued to be associated with poor outcomes. These findings were concordant with the DL-based quantitative assessment (hazard ratio, 0.96; 95% CI, 0.93-0.99; P = .0041). In exploratory multivariable Cox regression analyses, pathologic regression, DL-derived extent of immune infiltrates, and largest tumor dimension at nephrectomy remained associated with FFNT. This study provides, to our knowledge, the first integrated, quantitative assessment of radiologic, pathologic, and image-based response metrics following preoperative ICI therapy in RCC. If validated prospectively, these findings may help guide adaptive treatment approaches and clinical trial design.