Early Switch Therapy to Atezolizumab After Vemurafenib/Cobimetinib Run-in Improves OS in BRAF V600–Positive Melanoma

Early switch therapy to atezolizumab (Tecentriq) after run-in with vemurafenib (Zelboraf) plus cobimetinib (Cotellic) led to an improvement in overall survival (OS) rates at 4 and 5 years in patients with BRAF V600–positive melanoma, but was not statistically significant, according to long-term survival data from the phase 2 ImmunoCobiVem trial (NCT02902029).¹

“Yet, a relevant number of patients experienced rapid progression after early switch from targeted therapy to immune checkpoint inhibition [ICI],” Dirk Schadendorf, MD, of the University of Essen and the German Cancer Consortium, Partner Site, in Essen, Germany, said in a presentation of the data at the 2024 ESMO Congress. “The better early tumor control—PFS1—with continuous targeted therapy in arm A and treatment options outside the trial influence the overall effect size for early switch from targeted therapy to ICI within this trial.”

After the 3-month run-in phase of ImmunoCobiVem, patients without progressive disease (PD) were randomly assigned to continue vemurafenib plus cobimetinib (Arm A; n = 69) or switch to atezolizumab (Arm B; n = 66) until first documented PD.2 At a median follow-up of 57 months (IQR, 23-63), final OS data revealed that the median OS was 40.2 months (95% CI, 18.9-not estimable [NE]) in Arm A vs 49.6 months (95% CI, 26.1-NE) in Arm B (HR, 1.17; 95% CI, 0.71-1.91; stratified P = .94). The 60-month OS rates were 40% vs 45%, respectively.¹

Additionally, the final PFS1 data showed that patients in Arm A (n = 68) experienced a median PFS1 of 13.0 months (95% CI, 9.9-15.6) vs 5.9 months (95% CI, 5.5-8.3) in Arm B (n = 65; HR, 0.61; 95% CI, 0.41-0.91; stratified P = .006). The 60-month PFS1 rates were 19% vs 15%, respectively.

When examining patients who crossed over, the time to second progression (PFS3) was 2.8 months (95% CI, 2.1-3.3) among those in arm A who received vemurafenib plus cobimetinib and then atezolizumab (n = 21) compared with 6.0 months (95% CI, 2.1-8.8) among those in arm B who received atezolizumab then vemurafenib plus cobimetinib (n = 35; HR, 1.72; 95% CI, 0.93-3.20). The 48-month PFS3 rates were 10% vs 17%, respectively. Furthermore, the total PFS (PFS2) was 12.6 months (95% CI, 8.3-17.0) in arm A vs 14.6 months (95% CI, 8.6-25.6) in arm B (HR, 1.29; 95% CI, 0.70-2.38; P = .41); 60-month PFS2 rates were 10% vs 21%, respectively.

“No subgroups were identified for which a targeted therapy run in provided clinical benefit,” Schadendorf added.

Detailing the ImmunoCobiVem Trial

“We are trying to answer [the questions of] in those patients where we had tumor control, whether we [can] continue with targeted therapy or are able to maintain the clinical response, that’s the PFS1 primary endpoint. At time of progression, there was crossover possible, PFS3,” Schadendorf said.

ImmunoCobiVem enrolled patients with previously untreated unresectable or metastatic BRAF V600–mutated melanoma from November 23, 2016, to December 27, 2019, and the final database lock occurred on July 8, 2024.^1,2 Patients were randomly assigned 1:1 to arm A and arm B and stratification factors included serum lactate dehydrogenase (LDH; elevated vs normal), metastatic stage of disease (M1c yes vs no), and receival or previous adjuvant therapy (yes vs no).¹

Secondary end points included OS, overall PFS2, PFS3, best overall response rate, duration of response, and safety. The primary end point of PFS1 was defined as PFS from the start of the run-in phase to first PD.

All patients received vemurafenib at 960 mg twice daily on days 1 to 28 in combination with cobimetinib at 60 mg once daily on days 1 to 21 of a 28-day cycle in the 3-month run-in phase. Following the run-in phase, random assignment occurred and patients in arm A continued the combination therapy and those in arm B switched to receive 1200 mg every 3 weeks of atezolizumab. If PD occurred, patients the entered the crossover phase and switched to receive the opposite arm’s regimen. If a second PD event occurred, patients were only followed-up.

“The most important thing here is that these are traditional patients [and arms were] well balanced. There is nothing special to notice [on baseline patient characteristics],” Schadendorf said.

Among all patients in the trial (n = 185), the median age was 58 years (range, 19-91) and most patients were male (63%) and had an ECOG performance score of 0 (85%). Additionally, most patients had stage IV disease by AJCC 7th edition staging (93%) which included M1a (22%), m1b (22%), M1c (49%), and M1c with brain metastases (1%). Patients also had normal LDH levels (51%), received prior adjuvant therapy (13%), and had at least 3 disease sites (42%).

Additional Efficacy Findings

When examining objective response rate by phase of treatment, findings showed that at the end of the run-in phase, patients experienced a complete response (CR; n = 8), a partial response (PR; n = 92), stable disease (SD; n = 33), PD (n = 14), and other reason (n = 36).

In the randomized phase, patients in arm A experienced a CR (n = 14), PR (n = 32), SD (n = 4), and PD (n = 12); 37 patients had eventual PD, 27 discontinued treatment for reasons other than PD, and 5 patients are still on treatment. In the crossover phase, among the 21 patients in arm A, CR (n = 3), SD (n = 1), and PD (n = 15) occurred; 2 patients remain on treatment.

Patients in arm B experienced a CR (n = 12), PR (n = 12), SD (n = 4), and PD (n = 33) in the randomized phase; 50 patients had eventual PD, 12 discontinued treatment for reasons other than PD, and 4 patients remain on treatment. In the crossover phase, among the 35 patients in arm B, CR (n = 4), PR, (n = 11), SD (n = 4), and PD (n = 10) occurred; 2 patients remain on treatment and 10 discontinued therapy for reasons others than PD.

Disclosures: Dr Schadendorf reported receiving partial financial support from Roche Pharma for the conduct of this study and drug supply. He also reported receiving support, grants, or contracts from Bristol-Myers Squibb, MSD, Novartis, Pfizer, and Roche; receiving consulting fees or honoraria from Agenus, AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo, Formycon, Immunocore, InFlarX, Merck Serono, MSD, NeraCare, Novartis, NoviGenix, Pfizer, Philogen, Pierre Fabre, Replimune, Sanofi, Regeneron, and Sun Pharma; receiving support for attending meetings or travel support from Pierre Fabre; and has participated on drug safety monitoring or advisory boards for AstraZeneca, BioNTech, Bristol-Myers Squibb, CureVac, Daiichi Sankyo, Erasca, Immatics, Immunocore, InFlarX, IOBioTech, Merck Serono, MSD, NeraCare, Novartis, Pfizer, Philogen, Pierre Fabre, Replimune, Sanofi/Regeneron, and Sun Pharma. He also has leadership roles for EORTC-MG, DeCOG, German Cancer Society, Fiege Stiftung, Deutsche Hautkrebsstiftung, NVKH e.V. and EuMelaREg.

References

1. Schadendorf D, Gogas H, Kandolf L, et al. Early switch from targeted to immunotherapy in advanced BRAFV600-positive melanoma: Long-term OS and final PFS results of the randomized phase II ImmunoCobiVem trial. Presented at: 2024 ESMO Congress; September 13-17, 2024; Barcelona, Spain. Abstract LBA45.
2. Livingstone E, Gogas H, Kandolf-Sekulovic L, et al. Early switch from run-in treatment with vemurafenib plus cobimetinib to atezolizumab after 3 months leads to rapid loss of tumour control in patients with advanced BRAFV600-positive melanoma: The ImmunoCobiVem phase 2 randomised trial. Eur J Cancer. 2023;190:112941. doi:10.1016/j.ejca.2023.112941