During the 2026 North Central Section of the AUA annual meeting, a kidney cancer panel featuring Vidit Sharma, Bradley Leibovich, Lee Ponsky, and Ithaar Derweesh dissected the shifting landscape of renal cell carcinoma management, focusing on adjuvant immunotherapy and emerging biomarkers, UroToday reported.
The Bottom Line
- Adjuvant immunotherapy trials in renal cell carcinoma have yielded variable survival benefits, prompting a shift away from a uniform class effect model toward patient-specific risk stratification.
- New prognostic tools—including circulating biomarkers like KIM-1, RNA expression signatures, digital pathology, and multimodal radiomics models—are expanding precision oncology frameworks.
- Baseline recurrence risk heavily dictates both clinical efficacy and cost-effectiveness, with high-risk patient subgroups deriving the greatest absolute benefit from aggressive systemic therapy.
Biomarker Proliferation Versus Routine Clinical Integration
Interest in renal cell carcinoma biomarkers has expanded substantially over the past several decades. A growing number of publications evaluate these markers because conventional clinicopathologic characteristics alone fail to capture the biologic heterogeneity of the disease. Despite this proliferation of candidate markers, their incorporation into routine clinical practice remains limited.
The clinical rationale for improved biomarker development begins with the variable outcomes observed across adjuvant immunotherapy trials. Dr. Sharma reviewed five major trials during the panel discussion, highlighting important differences in treatment strategies and efficacy outcomes. KEYNOTE-564 evaluated pembrolizumab for one year and demonstrated improvements in both disease-free survival and overall survival. More recently, LITESPARK-022 evaluated pembrolizumab combined with belzutifan for one year and demonstrated a disease-free survival benefit, although overall survival data remain immature.
Conversely, IMmotion010 evaluating atezolizumab, CheckMate-914 evaluating ipilimumab plus nivolumab, and PROSPER evaluating nivolumab in a perioperative strategy did not demonstrate improvements in disease-free survival or overall survival. These divergent findings suggest that therapeutic benefit does not represent a uniform class effect. Differences in treatment regimens, patient populations, and trial designs contribute to observed variability, reinforcing the need to identify patients most likely to benefit.
Weighing Absolute Survival Benefits Against Treatment Toxicities
Efficacy results must be interpreted alongside treatment-related toxicity. In KEYNOTE-564, an absolute disease-free survival benefit of 8.7% and an overall survival benefit of 5.4% at five years were accompanied by a 17.4-percentage-point increase in grade ≥3 adverse events compared with placebo.
Similarly, in LITESPARK-022, a 4.6% absolute disease-free survival improvement at two years and a 0.5% numerical overall survival difference came with a 21.9-percentage-point increase in grade ≥3 adverse events. Because not every patient undergoing nephrectomy experiences recurrence, many receive systemic treatment without deriving meaningful clinical benefit. Dr. Sharma emphasized that while no broadly accepted predictive biomarker currently exists for adjuvant immunotherapy in renal cell carcinoma, accurate prognostication remains vital for individualized risk-benefit discussions.
| Clinical Trial | Regimen | DFS Benefit | OS Benefit | Grade ≥3 AE Increase |
|---|---|---|---|---|
| KEYNOTE-564 | Pembrolizumab (1 Year) | 8.7% (5-Yr) | 5.4% (5-Yr) | +17.4 percentage points |
| LITESPARK-022 | Pembrolizumab + Belzutifan | 4.6% (2-Yr) | 0.5% (Numerical) | +21.9 percentage points |
Baseline Risk Stratification and Economic Modeling
Patients with a higher baseline probability of recurrence derive a greater absolute benefit from adjuvant treatment, even when relative treatment effects remain similar across risk groups. Subgroup analyses from KEYNOTE-564 illustrate this dynamic clearly:
- Intermediate-High Risk: 24-month disease-free survival reached 81.1% with pembrolizumab versus 72.0% with placebo, yielding a hazard ratio of 0.68 (95% CI 0.52–0.89).
- High-Risk Subgroup: 24-month disease-free survival rates were 48.7% and 35.4% respectively, with a hazard ratio of 0.60 (95% CI 0.33–1.10).
- M1 NED Population: 24-month disease-free survival stood at 78.4% with pembrolizumab compared with 37.9% with placebo, corresponding to a hazard ratio of 0.28 (95% CI 0.12–0.66).
Recent frameworks modeled by Das and colleagues demonstrate that absolute improvement decreases as expected disease-free survival with surveillance increases. Furthermore, Markov modeling from Dr. Sharma’s group shows that the cost-effectiveness of adjuvant pembrolizumab depends heavily on baseline risk, remaining economically viable primarily for patients with a progression risk exceeding 59% during observation.
Advancing Multimodal Prognostication Models
To refine risk assessment, researchers are investigating diverse prognostic modalities, including clinical nomograms, urine and blood biomarkers, tissue molecular signatures, digital pathology, radiomics, and the microbiome. Clinical prognostic models remain the most established. The AUA risk classifications—incorporating low, intermediate, high, and very-high-risk categories—provide meaningful survival separation.
In 2024, Mayo Clinic validation data presented by Dr. Sharma and colleagues showed that the Mayo Clinic progression-free survival model achieved an AUC of 0.82, compared with 0.78 for standard AUA classifications. Beyond clinical scores, circulating kidney injury molecule-1 (KIM-1) has emerged as a potential plasma biomarker for distinguishing renal masses and predicting oncologic outcomes, as evaluated by Xu and colleagues in 2024. Tissue-based RNA expression signatures, artificial intelligence-driven digital histopathology, and machine-learning radiomics models are further augmenting predictive accuracy.
Disclaimer: The information provided in this article is for educational and informational purposes only and does not constitute financial advice.
