Adjuvant pembrolizumab after nephrectomy is now the standard of care for patients with clear cell renal cell carcinoma (ccRCC) at increased risk of recurrence. Yet it remains the only approved adjuvant immunotherapy in this disease, and the field still lacks a validated biomarker to distinguish patients who need it from those already cured by surgery alone.
How to stratify recurrence risk after nephrectomy, interpret the divergent results across five phase III adjuvant trials, weigh toxicity in a potentially cured population, and navigate the absence of a predictive selection biomarker will be among the topics discussed at the Community Oncology Global Congress 2026, an international virtual congress taking place 28–30 August 2026 dedicated to advancing cancer care in community settings worldwide.
The Scale of the Problem
An estimated 80,450 new kidney cancer diagnoses and 15,160 deaths are expected in the United States in 2026, with renal cell carcinoma accounting for roughly 90% of cases (American Cancer Society, 2026). Excess body weight is implicated in 19% of kidney cancer deaths globally, and incidental radiologic detection raises concerns about overdiagnosis (WHO 2026 Global Cancer Report). Surgery is curative for the majority with localised disease, but recurrence occurs in up to 40% of patients with higher-stage or higher-grade tumours (Furukawa et al., 2025).
Recurrence Risk Stratification After Nephrectomy
Several clinicopathological models guide risk assessment. The Leibovich score incorporates tumour stage, lymph node status, tumour size, nuclear grade and necrosis, yielding five-year metastasis-free survival rates of approximately 97% for low-risk, 74% for intermediate-risk and 31% for high-risk patients (Leibovich et al., Cancer, 2003). The UISS and SSIGN scores offer complementary prognostic information (Patard et al., JCO, 2004).
The KEYNOTE-564 TNM-based risk strata are now the preferred classification for operable ccRCC: intermediate-high risk is defined as pT2 grade 4 or sarcomatoid N0M0 or pT3 any grade N0M0; high risk as pT4 any grade N0M0 or any pT N+M0; and M1 NED after complete metastasectomy (Powles et al., Annals of Oncology, 2024; ESMO). These models are clinically useful but imperfect — intermediate-high risk patients represent a heterogeneous population in which a substantial proportion will never recur without adjuvant treatment.
Trial Evidence for Adjuvant Checkpoint Inhibition
Five phase III trials have evaluated adjuvant immune checkpoint inhibitors in RCC. Only KEYNOTE-564 has demonstrated both DFS and OS benefit.
KEYNOTE-564
KEYNOTE-564 is a phase 3, double-blind, randomised trial that evaluated adjuvant pembrolizumab 200 mg every three weeks for up to 17 cycles (approximately one year) versus placebo in 994 patients with ccRCC at intermediate-high, high or M1 NED risk after nephrectomy. At 57.2 months median follow-up, 48-month OS was 91.2% versus 86.0% (HR 0.62; 95% CI 0.44–0.87; P = 0.005), with a DFS HR of 0.72 (95% CI 0.59–0.87). Five-year follow-up confirmed durable OS and DFS benefit (Haas et al., ASCO 2025). Grade 3–4 treatment-related adverse events occurred in 18.6% versus 1.2%, immune-mediated adverse events in 36.5% versus 7.3%, and 21.1% discontinued pembrolizumab due to adverse events (Choueiri et al., NEJM, 2024).
IMmotion010
IMmotion010 is a phase 3, multicentre, double-blind, randomised trial conducted across 215 centres in 28 countries that evaluated adjuvant atezolizumab 1200 mg every three weeks for 16 cycles (approximately one year) versus placebo in 778 patients with clear cell or sarcomatoid RCC at increased risk of recurrence after nephrectomy. The trial failed its primary endpoint: median DFS was 57.2 months with atezolizumab versus 49.5 months with placebo (HR 0.93; 95% CI 0.75–1.15; P = 0.50). There were no treatment-related deaths (Pal et al., Lancet, 2022).
CheckMate 914
CheckMate 914 is a phase 3, double-blind, randomised trial conducted across 145 hospitals and cancer centres in 20 countries that evaluated adjuvant nivolumab plus ipilimumab (Part A) and nivolumab monotherapy (Part B) versus placebo in patients with localised high-risk ccRCC after radical or partial nephrectomy. In Part A, at median follow-up of 37.0 months, nivolumab plus ipilimumab did not improve DFS versus placebo (HR 0.92; 95% CI 0.71–1.19; P = 0.53).
The expected treatment period was 24 weeks (approximately 6 months). All-cause adverse events led to discontinuation of nivolumab plus ipilimumab in 32% of treated patients (Motzer et al., Lancet, 2023). Part B similarly showed no DFS benefit for nivolumab monotherapy (Motzer et al., JCO, 2025).
PROSPER
PROSPER is a phase 3, open-label, randomised trial that tested a perioperative approach — neoadjuvant nivolumab before nephrectomy followed by adjuvant nivolumab — versus surgery alone in 819 patients with high-risk RCC of any histology across 183 community and academic sites in the United States and Canada. The trial was stopped for futility: recurrence-free survival HR was 0.94 (95% CI 0.74–1.21; one-sided P = 0.32). Only 44% of patients completed the full nivolumab course, and treatment-related deaths occurred in 9 patients (2%) in the nivolumab arm versus 3 (<1%) in the surgery-only arm (Allaf et al., Lancet Oncology, 2024).
RAMPART
RAMPART is a phase 3, international, randomised trial that used the Leibovich score for risk stratification, randomising patients across three arms: active monitoring, durvalumab monotherapy for one year and durvalumab plus tremelimumab for one year. At the 2026 ASCO Annual Meeting, James M. Larkin presented results showing that durvalumab monotherapy did not significantly improve DFS compared with active monitoring (3-year DFS 78% versus 72%; HR 0.74; 95% CI 0.53–1.04; one-sided P = 0.041).
In the higher-risk subgroup, durvalumab plus tremelimumab improved DFS versus active monitoring (HR 0.52; 95% CI 0.34–0.80), with no corresponding benefit in intermediate-risk patients (interaction P = 0.019 for the combination) (Larkin et al., ASCO 2026, Abstract LBA4511; Oncodaily).
LITESPARK-022
LITESPARK-022 is a phase 3, double-blind, multinational trial that evaluated adjuvant pembrolizumab plus belzutifan, a HIF-2α inhibitor, versus pembrolizumab plus placebo in 1,841 patients with ccRCC at increased risk of recurrence. The combination significantly improved DFS (HR 0.72; 95% CI 0.59–0.87; P < 0.001), with 24-month DFS of 80.7% versus 73.7%. At this interim analysis, with only 29% of the planned overall survival events observed, OS did not differ significantly between the groups (HR 0.78; 95% CI 0.51–1.19; P = 0.24). Grade 3 or higher adverse events were substantially more frequent with the combination (52.1% versus 30.2%), driven primarily by anemia (84.0% any grade) and hypoxia (7.0%) (Choueiri et al., NEJM, 2026).
Absence of a Validated Selection Biomarker
No predictive biomarker currently guides adjuvant pembrolizumab use (Buti et al., Crit Rev Oncol Hematol, 2025). PD-L1 expression is not predictive in this setting, and PD-L1 testing is not recommended to inform adjuvant RCC treatment decisions (Powles et al., Ann Oncol, 2024; ESMO).
A ctDNA analysis from KEYNOTE-564 presented at ASCO 2026 showed that baseline ctDNA was detectable in only 5.4% of patients with the 16-plex assay and 8.2% with the 64-plex assay. Sensitivity for predicting DFS events was just 10–15%, despite specificity of 96–99%. Pembrolizumab improved DFS regardless of baseline ctDNA status, and on-treatment ctDNA clearance by cycle 5 was numerically higher with pembrolizumab than placebo (Choueiri et al., ASCO 2026; Oncodaily).
These findings highlight both the prognostic potential and current limitations of exome-based ctDNA: low positivity rates preclude its use for patient selection today.
Toxicity in a Potentially Cured Population
The toxicity calculus differs fundamentally in the adjuvant setting. Many patients are already cured by surgery. Immune-related adverse events in these patients represent harm without compensating benefit.
In KEYNOTE-564, immune-mediated adverse events occurred in 36.5% versus 7.3%, and 21.1% of patients discontinued pembrolizumab due to adverse events (Choueiri et al., NEJM, 2024). Endocrinopathies such as hypothyroidism and adrenal insufficiency may be permanent, requiring lifelong hormone replacement in patients who may never have needed treatment at all.
The updated EAU guidelines recommend discussing the risk of overtreatment and side effects with every patient, and determining the adjuvant strategy on a case-by-case basis (EAU Guidelines, 2025).
When Patients Recur After Adjuvant Therapy
A retrospective study of 70 patients with high-risk RCC treated with adjuvant immunotherapy reported a 21% recurrence rate after a median follow-up of 30.2 months. Two-thirds of recurrences presented as oligometastatic disease, most commonly in lung (60%), lymph nodes (33%) and the renal bed (13%). Patients who received metastasis-directed local therapy had a 24-month post-progression survival of 100%, compared with 86% for those who received systemic therapy.
All patients who started systemic treatment received VEGFR-TKI-based regimens, consistent with the current consensus that ICI rechallenge during or shortly after adjuvant immunotherapy is not recommended (Ciccarese et al., Cancer Immunol Immunother, 2026).
The Questions Every Oncologist Will Face
The evidence base has grown rapidly, but several critical questions remain unanswered. Whether intermediate-high risk patients at the lower end of that category should routinely receive adjuvant pembrolizumab remains unclear, given that RAMPART found no benefit for immunotherapy in intermediate-risk disease. Whether ctDNA-guided strategies can spare lower-risk patients from adjuvant treatment is equally uncertain, as current assays lack the sensitivity required for clinical decision-making. The addition of belzutifan to pembrolizumab improves DFS but has not demonstrated an OS benefit at this interim analysis, while substantially increasing toxicity — a trade-off that remains difficult to justify for every patient.
When recurrence occurs after adjuvant immunotherapy, the appropriate default remains VEGFR-TKI monotherapy, but whether ICI rechallenge can be considered after a prolonged treatment-free interval is an open question without prospective data. Finally, permanent endocrinopathies — hypothyroidism, adrenal insufficiency, hypophysitis — may affect patients who were surgically cured and never needed treatment at all, and the community oncologist must be equipped to manage them for the long term.
These are decisions facing oncologists in daily practice, and the evidence to guide them remains incomplete. How to navigate these decisions in real-world community practice will be a focus of the Community Oncology Global Congress 2026, taking place virtually 28–30 August 2026. Register now.
Find more videos on OncoDaily YouTube TV.