Durable T-VEC Responses Fail to Prolong Survival in Updated Analysis

Michael A. Postow, MD

Talimogene laherparepvec (T-VEC) significantly improved durable response rates (DRR) but failed to extend overall survival (OS) in patients with advanced melanoma, according to results from an analysis of the phase III OPTiM study announced by Amgen, the company developing the vaccine.

“I think what we’ll see with treatments like T-VEC, or intralesional injectable treatments, are using these treatments in combinations with other immunotherapeutics,” Michael A. Postow, MD, an attending physician in the Melanoma and Immunotherapeutics Service at Memorial Sloan Kettering Cancer Center, said in an interview. “I expect, in the future, we’ll be partnering T-VEC with other drugs that target other immune strategies, such as the PD-1 treatments.”

T-VEC is engineered through the genetic alteration of the herpes simplex 1 virus to secrete the cytokine GM-CSF within the tumor, causing cell lysis. Based on this mechanism of action, T-VEC was compared with GM-CSF alone in the OPTiM study. GM-CSF is not currently an approved treatment for patients with melanoma but studies have shown that it may enhance the body’s immune response. A phase II study presented at the 2013 ASCO Annual Meeting showed that adding GM-CSF to treatment with the CTLA-4 inhibitor ipilimumab improved OS by 4.8 months versus ipilimumab alone for patients with metastatic melanoma.

Ongoing clinical trials are planned that will investigate T-VEC in combination with immune checkpoint inhibiting antibodies. One such study, recently announced by Merck, will examine the combination of T-VEC with the PD-1 inhibitor MK-3475 in previously untreated advanced melanoma. Additionally, a phase I/II trial is examining the combination of T-VEC and ipilimumab in patients with untreated stage IIIB-IV melanoma (NCT01740297).

“We remain encouraged that the study met its primary endpoint of achieving durable responses in patients with metastatic melanoma," Sean E. Harper, MD, executive vice president of Research and Development at Amgen, said in a release. "We missed statistical significance on the secondary endpoint of overall survival but the strong trend in survival benefit supports further research of talimogene laherparepvec to better understand its role in melanoma, both as a single-agent and in combination with other therapies."

In the OPTiM study, 436 patients with unresected stage IIIB/C and IV melanoma were randomized in a 2:1 ratio to receive intralesional T-VEC (n = 295) or subcutaneous GM-CSF (n = 141). T-VEC was administered at ≤4 mL x10⁶ pfu/mL for 3 weeks followed by ≤4 mL x10⁸ pfu/mL every 2 weeks. GM-CSF was administered at 125 μg/m² for 14 days every 28 days. The primary endpoint of the study was DRR with OS representing a secondary outcome measure.

Data presented at the 2013 ASCO Annual Meeting showed that treatment with T-VEC resulted in a DRR of 16.3% versus 2.1% with GM-CSF. The objective response rate was 26.4% compared with 5.7% and the complete response rate was 10.8% versus 0.7% for T-VEC and GM-CSF, respectively.

An interim survival analysis was presented in November at the 2013 Society for Melanoma Research (SMR) meeting demonstrating a median OS with T-VEC of 23.3 months compared with 19.0 months with GM-CSF (HR = 0.79, 95% CI: 0.61-1.02; P = 0.0746). In the updated analysis, Amgen noted that the HR for OS and the median improvement in months for T-VEC versus GM-CSF was nearly the same as reported at the interim analysis. However, the updated P value from more mature data failed to reach statistical significance (P = .051).

A retrospective analysis of the OPTiM study presented at the Society of Surgical Oncology Symposium in March 2014 showed that T-VEC injections promoted tumor shrinkage in 64% of patients with advanced melanoma. Additionally, shrinkage occurred in 32% of uninjected nonvisceral metastatic lesions and in 16% of uninjected visceral metastatic lesions.

A total of six patients converted from unresectable to resectable melanoma, following the administration of T-VEC. In total, 37 surgeries were performed during the course of the trial: 15 resulted in no evidence of disease and three showed a pathologic complete response.

The most frequently reported all grade adverse events for the T-VEC arm were fatigue (50.3%), chills (48.6%), and pyrexia (42.8%). Grade 3/4 adverse events were limited, with cellulitis affecting 2.1% of patients in the T-VEC arm.