I’m thrilled to present the results of the STELLAR-303 trial which we presented at the ESMO Congress and we published simultaneously in The Lancet. The STELLAR-303 trial was the first immunotherapy-based phase III 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 95% of cases.
Zanzalintinib is a differentiated next-generation multi-tyrosine kinase inhibitor that targets the TAM kinases as well as MET and VEGF receptor. In pre-clinical studies zanzalintinib has shown synergy when combined with immune checkpoint inhibitors, this has led to further exploring the regimen in early phase clinical trials. We have explored and conducted early phase clinical trials with an early generation multi-tyrosine kinase inhibitor called cabozantinib, which has overlapping tyrosine kinase targets with zanzalintinib, in combination with immune checkpoint inhibitor in a trial called the CAMILLA trial. In that study we have shown a promising and encouraging signal for the combination in chemo-refractory colorectal cancer.
Additionally, we have demonstrated in the phase I STELLAR-001 study a meaningful benefit with the combination of zanzalintinib and atezolizumab versus zanzalintinib with a survival benefit favouring the combination in this population.
So this data has provided the needed ground to launch the phase III STELLAR-303 study. The population that went on the STELLAR-303 trial were metastatic chemo-refractory colorectal cancer patients that progressed or were intolerant to at least two prior lines of chemotherapy with or without VEGF targeted therapy. The patients who were enrolled had tumours that are not MSI high and not mismatch repair deficient. Patients were randomised 1:1 to receive either zanzalintinib once daily in combination with atezolizumab infusion once every three weeks versus regorafenib dosed at regulatory approved labelling. The trial has stratification factors and that’s based on geographic location, presence or absence of RAS mutations and presence or absence of liver metastases. The trial has dual primary endpoints of overall survival in the intention to treat population as well as overall survival in patients without liver metastases.
The results I presented are based on a planned overall survival in the intention to treat population and I also presented an interim analysis of overall survival in patients without liver metastases. The study continues to final overall survival in patients without liver metastasis.
As far as the efficacy results, the STELLAR-303 trial met its primary endpoint of overall survival in the intention to treat population. We have shown a significant survival advantage with the combination versus regorafenib with a stratified hazard ratio of 0.80, translating into a 20% reduction in the risk of death in patients receiving the combination with a p-value of 0.0045. The median overall survival was 10.9 months with the combination versus 9.4 months with regorafenib. We have shown an early separation of the Kaplan-Meier curve favouring the combination that consistently favoured the combination thereafter. The 24-month overall survival estimates were 20% in patients receiving the combination versus 10% in patients receiving regorafenib.
As far as interim analysis of overall survival in patients without liver metastasis we’ve shown a stratified hazard ratio of 0.79 with a median overall survival of 15.9 months in patients receiving the combination as opposed to 12.8 months in patients receiving regorafenib.
We have also looked at subgroup analysis, looking also at the Forest plot from the study we have seen consistent relative benefit seen with the combination as opposed to regorafenib in all key subgroups, including benefit seen regardless of presence or absence of liver metastases, regardless of presence or absence of RAS mutations. Importantly, I would like to highlight that prior exposure to VEGF-targeted therapy did not dimmish the benefit of the combination of atezolizumab and zanzalintinib, noting that 80% of the population that went on the trial had prior exposure to VEGF-targeted therapy which represents the Western population and what we see in real-world practice.
As far as the safety profile for the combination, most of the treatment-related adverse events were grade 1-3 and mostly were fatigue, hypertension, diarrhoea, nausea and loss of appetite. As far as the grade 4 treatment-related adverse events, those were low, seen in 3% of patients treated with the combination as opposed to 4% in patients treated with regorafenib. One important thing to note as far as the safety is that all grade and grade 3 and above PPE, or hand foot syndrome, which is a dose-limiting toxicity for the class of agents with VEGF tyrosine kinase inhibitor were much lower in patients receiving the combination as opposed to patients receiving regorafenib.
Overall, there were no new emerging safety signals and the safety signal was consistent with the safety profile seen with similar VEGF tyrosine kinase inhibitors in combination with immune checkpoint inhibitors that are available in other disease types like renal cell carcinoma.
As key messages I would like to say that metastatic colorectal cancer represents a major global health challenge with 150,000 new cases expected annually in the United States alone and 5-year overall survival of 15-16%. The success of the STELLAR-303 trial with synergistic combinations combining targeted therapy zanzalintinib with immune checkpoint inhibitor is potentially changing this trajectory.
We are now looking at analysing further data from the STELLAR-303 study, specifically looking at the final analysis of overall survival in the subset of patients without liver metastases. We’re also looking at the next generation of trials that could be developed to capitalise on the results seen in STELLAR-303, potentially exploring the combination in earlier lines in patients with metastatic colorectal cancer as well as potentially exploring it in earlier stages of the disease, in the neoadjuvant or adjuvant setting. Another avenue that we could explore is combining new immune modulators that could be added to the doublet of zanzalintinib and atezolizumab.
A take home message I would like to say that STELLAR-303 results send a clear message of progress and possibility which opens the door for a new era of immunotherapy-based chemotherapy-free combinations in colorectal cancer patients that are not MSI high and not mismatch repair deficient.
