In the study, the combination of zanzalintinib (XL092), a next-generation TKI, and atezolizumab (Tecentriq) delivered a statistically significant overall survival (OS) advantage compared with regorafenib (Stivarga) in previously treated metastatic CRC. Median OS reached 10.9 months with the combination vs 9.4 months with regorafenib (HR, 0.80; 95% CI, 0.69-0.93; P = .0045). Among patients without liver metastases—a subgroup historically associated with more immune-responsive disease—an interim analysis showed a median OS of 15.9 vs 12.7 months (HR, 0.79; 95% CI, 0.61-1.03; P = .087).
Notably, the OS benefit extended across geographic regions, RAS mutation subtypes, and prior anti-VEGF exposure, suggesting that the combination may offer broad clinical utility within a heterogeneous population. Progression-free survival trends (3.7 months vs 2.0 months; HR, 0.68; 95% CI, 0.59-0.79) and a higher disease control rate (54% vs 41%) further reinforce the potential of this regimen to stabilize disease in a setting where therapeutic durability is often short-lived.
In an interview with OncLive®, Anwar Saeed, MD, discussed the clinical relevance of these findings, what makes zanzalintinib a compelling partner for immune checkpoint inhibition, and how data from STELLAR-303 may influence sequencing decisions for patients with heavily pretreated metastatic CRC.
Saeed is the chief of Gastrointestinal Medical Oncology at UPMC in Pittsburgh, Pennsylvania.
OncLive: What was the rationale for studying zanzalintinib plus atezolizumab in the STELLAR-303 trial?
Saeed: The STELLAR-303 trial is the first immunotherapy-based phase 3 trial that showed a significant survival advantage in patients with chemo-refractory, metastatic, or advanced-stage colorectal cancer that are not MSI-high and not mismatch repair deficient, which represents [approximately] 95% of cases.
Zanzalintinib is a differentiated, next-generation multi-TKI that targets TAM kinases as well as MET and VEGF receptors. In early clinical studies, zanzalintinib [demonstrated] synergy when combined with immune checkpoint inhibitors, and this led to further exploration of the combination in early-phase trials.
We have conducted early-phase trials with an earlier-generation TKI, cabozantinib, which has overlapping targets with zanzalintinib, in combination with immune checkpoint inhibitors. For example, in the [phase 2] CAMILLA trial [NCT03539822], we showed a promising signal in the chemo-refractory colorectal cancer population. Additionally, in the phase 1 STELLAR-001 study [NCT03845166], we also observed a promising signal favoring the combination of zanzalintinib plus atezolizumab vs zanzalintinib monotherapy. [These findings] provided the basis to launch the phase 3 STELLAR-303 study.
What were the key eligibility criteria for patients enrolled in STELLAR-303?
The population enrolled in STELLAR-303 were patients with metastatic colorectal cancer that had progressed on or were intolerant to at least two prior lines of chemotherapy, with or without VEGF-targeted therapy. The tumors were required to be not MSI-high and not mismatch repair deficient. [Approximately] 900 patients globally were randomly assigned in a 1:1 fashion to receive either zanzalintinib plus atezolizumab or regorafenib [at the] regulatory-approved dose. Stratification factors included geographic region, presence or absence of RAS mutations, and presence or absence of liver metastases. The trial had dual primary end points of OS in the intention-to-treat [ITT] population and OS in patients without liver metastasis.
How did the combination of zanzalintinib and atezolizumab perform compared with regorafenib in STELLAR-303?
The results I presented were based on a planned OS analysis in the ITT population, and we also presented an interim analysis of OS in patients without liver metastasis.
STELLAR-303 met its primary end point of OS in patients with chemotherapy-refractory metastatic CRC. We showed a significant advantage using the combination as opposed to regorafenib, with a stratified HR of 0.80, translating into a 20% reduction in the risk of death with a P value of .0045. The median OS with the combination was 10.9 months vs 9.4 months with regorafenib. We observed an early separation of the Kaplan-Meier curves that remained consistently separated favoring the combination. The 24-month OS estimates were 20% with the combination vs 10% with regorafenib.
In the interim analysis of patients without liver metastasis, the HR was 0.79, favoring the combination, with a median OS of 15.9 months vs 12.8 months with regorafenib.
How did treatment outcomes differ across key subgroups in STELLAR-303?
When examining key subgroups, we saw consistent benefit favoring the combination regardless of the presence or absence of liver metastases and regardless of presence or absence of RAS mutations. Importantly, approximately 80% of patients had prior exposure to VEGF-targeted therapy, and in this subgroup, prior exposure did not diminish the benefit of the combination.
Regarding safety, the most common treatment-related adverse effects [TRAEs] of all grades or grade 3 or higher were hypertension, diarrhea, nausea, and decreased appetite. Grade 4 TRAEs were low, seen in 3% of patients receiving the combination versus 4% receiving regorafenib. Palmar-plantar erythrodysesthesia [hand–foot syndrome], a dose-limiting toxicity with regorafenib, was observed at lower rates with the combination, despite the combination arm having longer treatment duration. [No] new safety signals were identified, and the safety profile was consistent with other VEGF TKI plus immune checkpoint inhibitor regimens used in diseases such as renal cell carcinoma.
How should these findings inform treatment selection for patients with metastatic CRC?
As a key message [is that] metastatic CRC remains a major global health challenge, with approximately 150,000 new cases annually in the United States and a 5-year OS rate of [approximately] 15% to 16%. The success of STELLAR-303 with a synergistic combination of a novel agent such as zanzalintinib plus immunotherapy is potentially changing this trajectory.
We are continuing to analyze data from the ongoing STELLAR-303 trial, including the final OS analysis in patients without liver metastasis, and are considering next-generation studies to evaluate this regimen in earlier lines of therapy and in earlier disease settings.
The results of STELLAR-303 clearly send a message of progress and possibility. It opens the door to a new era of immunotherapy-based combination therapy in CRC that is not MSI-high and not mismatch repair deficient.”
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. Lancet. Published online October 20, 2025. doi:10.1016/S0140-6736(25)02025-2
