Background
Phase III, randomized, open-label, N=901 (451 zanzalintinib-atezolizumab, 450 regorafenib). Previously treated metastatic colorectal adenocarcinoma (MSI-H/dMMR excluded), ECOG 0-2, median age ~62 years. MSS mCRC is largely immunotherapy-refractory, motivating a chemotherapy-free TKI plus checkpoint-inhibitor combination.
Interventions and follow up
Arm A: Zanzalintinib 100 mg oral daily + atezolizumab 1200 mg IV Q3W
Arm B: Regorafenib 160 mg oral days 1-21 of 28-day cycle
Primary endpoint: Overall survival (ITT and subset without liver metastases)
Median follow up: 18.0 months at OS analysis
Arm B: Regorafenib 160 mg oral days 1-21 of 28-day cycle
Primary endpoint: Overall survival (ITT and subset without liver metastases)
Median follow up: 18.0 months at OS analysis
Results
Overall survival (ITT): Median 10.9 vs 9.4 months, stratified HR 0.80, 95% CI 0.69-0.93, P=.0045
OS without liver metastases (interim): Median 15.9 vs 12.7 months, HR 0.79, 95% CI 0.61-1.03, P=.087
OS without liver metastases (interim): Median 15.9 vs 12.7 months, HR 0.79, 95% CI 0.61-1.03, P=.087
Adverse events
Overall: Grade ≥3 treatment-related AE 60% (zanzalintinib-atezolizumab) vs 37% (regorafenib); 5 treatment-related deaths (1%) vs 1 (<1%)
Most common grade ≥3 (experimental vs regorafenib): hypertension 15% vs 9%, proteinuria 6% vs 2%, fatigue 6% vs 2%, diarrhea 6% vs 2%; palmar-plantar erythrodysesthesia less frequent (any grade 16% vs 50%)
Most common grade ≥3 (experimental vs regorafenib): hypertension 15% vs 9%, proteinuria 6% vs 2%, fatigue 6% vs 2%, diarrhea 6% vs 2%; palmar-plantar erythrodysesthesia less frequent (any grade 16% vs 50%)
Conclusions
STELLAR-303 is the first phase III trial to demonstrate an OS benefit with an immunotherapy-based regimen (zanzalintinib-atezolizumab) in relapsed/refractory MSS mCRC, offering a chemotherapy-free option with a novel mechanism for heavily pretreated patients.
Key Limitations
Open-label design; absolute OS gain modest (1.5 months in ITT); the liver-metastasis-free subgroup analysis did not reach significance (P=.087); higher grade ≥3 toxicity and treatment-related deaths versus regorafenib.
Clinical Context
Refractory MSS mCRC currently relies on regorafenib or trifluridine-tipiracil ± bevacizumab. STELLAR-303 positions zanzalintinib-atezolizumab as a potential later-line alternative; regulatory and ESMO guidance positioning will depend on the totality of efficacy and toxicity data.