Frontline Cabozantinib Plus Atezolizumab Shows Favorable Survival in Advanced HCC

The combination of cabozantinib (Cabometyx) and atezolizumab (Tecentriq) led to a statistically significant improvement in progression-free survival (PFS) and a numerical improvement in overall survival (OS) vs sorafenib (Nexavar) as frontline treatment in patients with advanced hepatocellular carcinoma (HCC), according to findings from the first planned analysis of the phase 3 COSMIC-312 trial (NCT03755791) that were presented during the European Society for Medical Oncology (ESMO) Asia Virtual Oncology Week 2021.¹

At a median follow-up of 15.8 months, the results showed a median PFS of 6.8 months with the combination (n = 250) vs 4.2 months with sorafenib (n = 122) in the PFS intention-to-treat (ITT) population, meeting the co-primary end point of the study (HR, 0.63; 99% CI, 0.44-0.91; P = .0012).

“We are encouraged by the significant improvement in PFS observed in COSMIC-312, suggesting cabozantinib in combination with atezolizumab holds potential as a treatment to reduce the risk of disease progression or death for patients with advanced liver cancer,” R. Kate Kelley, MD, lead investigator and professor of clinical medicine in the Division of Hematology/Oncology at the University of California, San Francisco, said in a press release. “Patients with this aggressive form of cancer, who may also have other comorbid conditions due to liver disease, face a poor prognosis and are in need of additional approaches to treatment.”

To be eligible for enrollment into the global, multicenter, randomized phase 3 trial, patients had to have a histological or cytological diagnosis of HCC, disease that is not amenable to a curative approach, measurable disease per RECIST v1.1 criteria, Barcelona Clinic Liver Cancer stage category B or C disease, and an ECOG performance status of 0 to 1.²

Patients were randomized 2:1:1 to 1 of 3 arms: 40 mg of cabozantinib daily plus 1200 mg of atezolizumab every 3 weeks; 400 mg of sorafenib twice daily; or 60 mg of cabozantinib daily.

PFS and OS served as co-primary end points, with a secondary end point of PFS comparing cabozantinib with sorafenib. All PFS analyses were performed by a blinded independent review committee (BIRC) per RECIST version 1.1 criteria.

Additional results showed that at a median follow-up of 13.6 months, the median OS was 15.4 months with atezolizumab/cabozantinib (n = 432) vs 15.5 months with sorafenib (n = 217) in the ITT population, reflecting a numerical but not statistically significant improvement with the combination (HR, 0.90; 96% CI, 0.69-1.18; P = .438).

According to subgroup analyses performed by disease etiology, the median PFS in patients with hepatitis B (n = 109) was 6.7 months with the combination vs 2.7 months with sorafenib (HR, 0.46; 95% CI, 0.29-0.73). The median OS in patients with hepatitis B (n = 191) was 18.2 months vs 14.9 months, respectively (HR, 0.53; 95% CI, 0.33-0.87).

The median PFS in patients with hepatitis C (n = 105) was 7.9 months with the combination vs 5.6 months with sorafenib (HR, 0.64; 95% CI, 0.38-1.09). The median OS in patients with hepatitis C (n = 203) was 13.6 months vs 14.0 months, respectively (HR, 1.10; 95% CI, 0.72-1.68).

The median PFS in non-viral patients (n = 158) was 5.8 months with the combination vs 7.0 months with sorafenib (HR, 0.92; 95% CI, 0.60-1.41). The median OS in non-viral patients (n = 255) was 15.2 months and was not reached, respectively (HR, 1.18; 95% CI, 0.78-1.79).

When assessed by blinded independent central review (BICR), the median PFS was 5.8 months with cabozantinib monotherapy (n = 188) vs 4.3 months with sorafenib (n = 217) in the ITT population, reflecting a 29% reduction in the risk of disease progression or death (HR, 0.71; 99% CI, 0.51-1.01; P = .0107).

Further findings indicated that the objective response rate by BICR in the ITT population was 11% with the combination vs 3.7% with sorafenib and 6.4% with cabozantinib monotherapy. The disease control rates were 78%, 65%, and 84%, respectively.

Regarding safety, grade 3 or higher adverse effects (AEs) with the combination, sorafenib, and cabozantinib monotherapy, respectively, were palmar-plantar erythrodysesthesia (7.9%, 8.2%, and 8.5%), hypertension (7.0%, 6.3%, and 11.0%), aspartate aminotransferase increase (6.5%, 2.4%, and 5.3%) and alanine aminotransferase increase (6.3%, 1.9%, and 5.9%).

Grade 3/4 treatment-related AEs (TRAEs) occurred in 51% of patients in the combination arm, 30% of patients in the sorafenib arm, and 52% of patients in the single-agent cabozantinib arm. Grade 5 TRAEs occurred in 1.9%, 0.5%, and 0.5% of patients, respectively.

Treatment discontinuations due to TRAEs occurred in 6.1% of patients in the combination arm, 7.7% of patients in the sorafenib arm, 8.5% of patients in the cabozantinib monotherapy arm, and 14.0% of patients with cabozantinib and/or atezolizumab.

“Exelixis has a longstanding commitment to patients with liver cancer, and we are pleased to present more detailed efficacy and safety data during this ESMO Virtual Plenary Session reinforcing the potential of cabozantinib in combination with atezolizumab for patients with this disease who are in need of additional first-line treatment options,” Michael M. Morrissey, PhD, president and chief executive officer of Exelixis, said. “We look forward to the final COSMIC-312 OS analysis in early 2022 and to submitting a supplemental new drug application to the FDA at that time.”

References

1. Exelixis Announces Detailed Results from Phase 3 COSMIC-312 Pivotal Trial of Cabozantinib in Combination with an Immune Checkpoint Inhibitor in Patients with Previously Untreated Advanced Liver Cancer at ESMO Asia Virtual Oncology Week 2021. News release. Exelixis, Inc. November 20, 2021. Accessed November 23, 2021. https://yhoo.it/3FFlkDc
2. Study of cabozantinib in combination with atezolizumab versus sorafenib in subjects with advanced HCC who have not received previous anticancer therapy (COSMIC-312). ClinicalTrials.gov. Updated June 4, 2021. Accessed June 28, 2021. https://clinicaltrials.gov/ct2/show/NCT03755791