Relapsed / Refractory · Phase III global open-label randomized; N=901 refractory metastatic colorectal adenocarcinoma excluding MSI-H/dMMR. Zanzalintinib+atezolizumab vs regorafenib. Dual primary OS endpoints in ITT and no-liver-metastases subgroup. · ITT median OS 10.9 vs 9.4 months

STELLAR-303

Clinical question: Does zanzalintinib plus atezolizumab improve OS over regorafenib in refractory metastatic colorectal cancer?

ITT median OS 10.9 vs 9.4 months; HR 0.80 (95% CI 0.69–0.93; P=0.0045). No-liver-metastases interim HR 0.79 (95% CI 0.61–1.03; P=0.087). · Registry:

Design, population and treatment

Phase III global open-label randomized; N=901 refractory metastatic colorectal adenocarcinoma excluding MSI-H/dMMR. Zanzalintinib+atezolizumab vs regorafenib. Dual primary OS endpoints in ITT and no-liver-metastases subgroup.

Key efficacy outcome

10.9 · intervention

9.4

10.9 · comparator

Not calculable from median OS; approximate NNH 5 for grade ≥3 treatment-related AEs (23-point increase).

Grade ≥3 treatment-related AEs 60% vs 37%; treatment-related deaths 1% vs <1%.

Safety and absolute effect

Median OS (months)

https://pubmed.ncbi.nlm.nih.gov/41130252/

Interpretation and limitations

Current-practice context · 14 November 2026: Primary publication/abstract and trial identifier checked. Guideline-level current-practice mapping was not established in this publication-led batch; no NCCN citation provenance is claimed. Where detailed QoL, follow-up or toxicity data were not present in the retrieved abstract, they are not inferred.

Primary source

Primary publication · DOI: 10.1016/S0140-6736(25)02025-2