Background
Phase III, open-label RCT. 522 patients with advanced or metastatic RCC who had progressed on or after PD-1/PD-L1 checkpoint inhibitor therapy (either as monotherapy or in combination). All histologies eligible (clear cell and non-clear cell). Tested whether adding atezolizumab to cabozantinib could overcome post-IO resistance.
Interventions and follow up
Arm A: Atezolizumab 1200mg IV q3wk + cabozantinib 40mg/day orally
Arm B: Cabozantinib 60mg/day orally (standard post-IO dose)
Primary endpoint: PFS and OS (co-primary)
mFollow up: 15.2 mo
Arm B: Cabozantinib 60mg/day orally (standard post-IO dose)
Primary endpoint: PFS and OS (co-primary)
mFollow up: 15.2 mo
Results
PFS: 10.6 vs 10.8 mo, HR 1.04, P=.72 — no benefit
OS: Not reached vs not reached, HR 1.10, P=.56 — no benefit
ORR: 43% vs 37%
Duration of response: 14.0 vs 14.9 mo — similar
OS: Not reached vs not reached, HR 1.10, P=.56 — no benefit
ORR: 43% vs 37%
Duration of response: 14.0 vs 14.9 mo — similar
Adverse events
Overall grade ≥3: 60.6% (combination) vs 54.1% (cabo alone)
Hepatic: hepatotoxicity grade ≥3 12.5% vs 5.3% — notably elevated with combination
GI: diarrhea grade ≥3 7.3% vs 6.1%
Cardiovascular: hypertension grade ≥3 12.1% vs 13.9%
Immune-mediated (any grade): 39.8% vs 13.3%
Discontinuation: 30.3% vs 19.7%
Hepatic: hepatotoxicity grade ≥3 12.5% vs 5.3% — notably elevated with combination
GI: diarrhea grade ≥3 7.3% vs 6.1%
Cardiovascular: hypertension grade ≥3 12.1% vs 13.9%
Immune-mediated (any grade): 39.8% vs 13.3%
Discontinuation: 30.3% vs 19.7%
Conclusions
Adding atezolizumab to cabozantinib did not improve PFS or OS versus cabozantinib alone in patients with RCC who had progressed on prior immune checkpoint inhibitor therapy. This definitively negative trial demonstrates that re-challenging with a PD-L1 inhibitor post-IO progression adds no benefit and increases toxicity.
Key Limitations
Key Limitations: The negative result is the key finding; no limitations undermine the conclusion. Prior IO treatment was heterogeneous (nivolumab ± ipilimumab, pembrolizumab + axitinib, etc.), but the negative result was consistent across subgroups. The atezolizumab dose used (1200mg q3wk) is standard. The trial provides important and clinically actionable data that IO rechallenge does not rescue post-IO RCC, guiding future trial design. Follow-up was relatively short at 15 months.
Clinical Context
CONTACT-03 definitively closed the door on IO rechallenge as a strategy in post-IO RCC. Cabozantinib monotherapy at 60mg/day remains a standard 2L option after IO-based combinations. Post-IO RCC management should focus on non-IO agents: VEGF-TKIs (cabo, lenvatinib, axitinib), mTOR inhibitors (everolimus), belzutifan (LITESPARK-005), or clinical trial enrollment. This trial is routinely cited to justify avoiding atezolizumab or other PD-1/L1 agents after IO progression.
References
References: Pal SK et al, Lancet 2023