The combination of zanzalintinib (XL092) and atezolizumab (Tecentriq) improved overall survival (OS) compared with regorafenib (Stivarga) in patients with previously treated metastatic colorectal cancer (CRC), according to findings from the phase 3 STELLAR-303 trial (NCT05425940) presented at the 2025 ESMO Congress and published concurrently in The Lancet.
Across the intention-to-treat (ITT) population, data showed a median OS of 10.9 months with the zanzalintinib combination vs 9.4 months with regorafenib (HR, 0.80; 95% CI, 0.69-0.93; P = .0045). An interim analysis of findings in a subset of patients without liver metastases demonstrated a median OS of 15.9 months vs 12.7 months in each respective arm (stratified HR, 0.79; 95% CI, 0.61-1.03; P = .087).
Investigators noted an OS improvement with zanzalintinib/atezolizumab vs regorafenib across most key subgroups, which included patients from Asia (HR, 0.77; 95% CI, 0.59-1.00) and the rest of the world (HR, 0.82; 95% CI, 0.68-0.99), as well as those with RAS wild-type (HR, 0.79; 95% CI, 0.61-1.01) and RAS-mutated disease (HR, 0.80; 95% CI, 0.66-0.98). Additionally, the experimental combination improved OS for patients with (HR, 0.78; 95% CI, 0.65-0.94) and without liver metastases (HR, 0.77; 95% CI, 0.59-1.01); this benefit extended to those with (HR, 0.80; 95% CI, 0.68-0.95) and without prior anti-VEGF antibody therapy (HR, 0.80; 95% CI, 0.56-1.15).
The median progression-free survival (PFS) across the ITT population was 3.7 months with zanzalintinib/atezolizumab and 2.0 months with regorafenib; statistical significance for PFS could not be claimed at the time of analysis per prespecified hierarchical testing strategy (HR, 0.68; 95% CI, 0.59-0.79). The experimental combination generally conferred PFS improvements across patient subgroups.
Data showed an objective response rate (ORR) of 4% (95% CI, 2%-6%) in the zanzalintinib/atezolizumab arm, which consisted entirely of partial responses (PRs). Additionally, the ORR was 1% (95% CI, 0%-3%) in the regorafenib arm, which consisted entirely of PRs. The disease control rate (DCR) in the experimental and control arms was 54% (95% CI, 49%-58%) vs 41% (95% CI, 36%-45%), respectively.
“STELLAR-303 is the first phase 3 trial to show improved OS with an immune checkpoint inhibitor [ICI]–based combination in metastatic CRC that is not microsatellite instability–high [MSI-H] or mismatch repair deficient [dMMR]. The combination significantly prolonged OS compared to regorafenib in patients with previously treated metastatic CRC,” presenting author Anwaar Saeed, MD, a professor in the Department of Medicine of the Division of Hematology and Oncology at the University of Pittsburgh Medical Center and UPMC Hillman Cancer Center, stated in the presentation.1 “The combination of zanzalintinib and atezolizumab represents a potential novel chemotherapy-free for heavily pretreated [patients with] metastatic CRC in need of more improved therapy. The trial continues to the planned final OS in the subset of patients without liver metastases.”
In the open-label phase 3 STELLAR-303 trial, 901 patients with metastatic CRC documented to not have MSI-H or dMMR status were randomly assigned 1:1 to receive zanzalintinib at 100 mg orally each day plus atezolizumab at 1200 mg intravenously every 3 weeks (n = 451) or regorafenib at 160 mg orally once day on days 1 to 21 of each 28-day cycle (n = 450).
The trial’s dual primary end points were OS in the ITT population and among patients without liver metastases. Key secondary end points included PFS, ORR, and safety.
Patients 18 years and older with metastatic disease that had radiographically progressed on, was refractory to, or showed intolerance to prior standard-of-care treatment including fluoropyrimidine plus irinotecan and oxaliplatin with or without an anti-VEGF antibody, an anti-EGFR antibody, and a BRAF inhibitor were eligible for enrollment on the trial. Investigators stratified patients based on geographic region, RAS mutation status, and presence of liver metastases.
The median age was 60 years (range, 29-84) in the zanzalintinib/atezolizumab arm and 60 years (range, 29-85) in the regorafenib arm, with most patients from each being male (58% vs 60%), White (55% vs 53%), and from regions outside of Asia (65% vs 65%). Most patients from each respective arm had an ECOG performance status of 1 (53% vs 55%), a presence of liver metastases (59% vs 56%), BRAF wild-type status (75% vs 78%), and RAS-mutated disease (59% vs 60%). Nearly all patients across both arms received prior treatment with fluoropyrimidine plus irinotecan and oxaliplatin (>99% vs 100%).
The median time from random assignment to first subsequent line of anticancer therapy was 4.5 months (range, 0.2-20.4) in the zanzalintinib/atezolizumab arm and 3.7 months (range, 0.1-17.6) in the regorafenib arm; 44% and 42% of patients in each arm received subsequent treatment. Most patients in each respective arm received subsequent therapy with TAS-102 (27% vs 25%), bevacizumab (Avastin; 20% vs 20%), and TAS-102 in combination with bevacizumab (16% vs 16%).
Any-grade treatment-related adverse effects (TRAEs) occurred in 95% of the zanzalintinib/atezolizumab arm and 92% of the regorafenib arm; 57% and 42% from each experienced serious AEs, and 26% and 10% had serious TRAEs. Overall, 18% and 15% of patients in each arm had AEs resulting in treatment discontinuation, with 61% vs 40% requiring dose reduction of zanzalintinib and regorafenib, respectively, due to toxicity. Investigators noted treatment-related deaths due to intestinal perforation (n = 2) with zanzalintinib, pneumonitis (n = 1) and renal failure (n = 1) with atezolizumab, altered state of consciousness (n = 1) with zanzalintinib plus atezolizumab, and jejunal perforation (n = 1) with regorafenib.
In the zanzalintinib/atezolizumab and regorafenib arms, respectively, the most common AEs included diarrhea (50% vs 24%), hypertension (34% vs 26%), fatigue (33% vs 20%), nausea (31% vs 13%), and decreased appetite (30% vs 20%).
References
- Saeed A, Park YS, Tabernero J, et al. Zanzalintinib plus atezolizumab (zanza + atezo) vs regorafenib (rego) in patients (pts) with previously treated metastatic colorectal cancer (mCRC): primary overall survival (OS) analysis from the randomized, open-label, phase 3 STELLAR-303 study. Presented at the European Society for Medical Oncology (ESMO) Congress 2025; October 17-21, 2025; Berlin, Germany. LBA30.
- Hecht JR, Park YS, Tabernero J, et al. Zanzalintinib plus atezolizumab versus regorafenib in refractory colorectal cancer (STELLAR-303): a randomised, open-label, phase 3 trial. The Lancet. Published online October 19, 2025. doi:10.1016/S0140-6736(25)02025-2