Radiologic, Pathologic, and Deep Learning Predictors of Response to Immune Checkpoint Blockade in Renal Cell
Payal Kapur1, Alana Christie2, Vipul Jarmale3
1Department of Pathology, University of Texas Southwestern Medical Center, Dallas, Texas; Department of Urology, University of Texas Southwestern Medical Center, Dallas, Texas; Department of Kidney Cancer Program at Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, Texas.
Abstract:
Response assessment of primary kidney tumors in the consolidation cytoreductive and neoadjuvant settings offers a unique opportunity to inform postoperative adaptive treatment strategies. Yet, systematic analyses of nephrectomy specimens post immune checkpoint inhibition (ICI) therapies are lacking. We retrospectively identified consecutive renal cell carcinoma (RCC) patients with locoregionally advanced or metastatic RCC who received at least one cycle of ICI-containing doublet therapy prior to nephrectomy at the UTSW 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 (MFS) in neoadjuvant patients. Pathologic response to ICI results in tumor cell death and fibrosis, creating hypocellular areas and increased immune infiltrate, features we utilized 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 infiltrate. Among 99 patients (cytoreductive nephrectomy {CN}, n=66; neoadjuvant nephrectomy {NaN}, n=33), radiologic tumor shrinkage ≥30% (p=0.0036) and the extent of ICI-induced pathologic regression as assessed by central pathology review (HR 0.97; CI 0.95-0.99; p=0.0023), were significantly associated with prolonged FFNT, with similar trends in neoadjuvant cohort. Presence of coagulative tumor necrosis continued to be associated with poor outcomes. These findings were concordant using a DL-based quantitative assessment (HR 0.96; CI 0.93-0.99; p=0.0041). In exploratory multivariable Cox regression, pathologic regression, DL-derived extent of immune infiltrate, 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.
