Amgen’s T-VEC Gets Negative FDA Report Card; Treatment Subgroup Could Be Salvation

Advisory Committee will weigh concerns about study design and results of Amgen’s metastatic melanoma immunotherapy.

Given FDA’s extensive concerns about the pivotal trial for Amgen Inc.’s metastatic melanoma immunotherapy talimogene laherparepvec (T-VEC), the company may have a tough time winning approval for its originally proposed indication. However, FDA suggested the drug might be appropriate for a subgroup of patients and is even planning to seek formal advisory committee endorsement of the idea.

FDA’s briefing document for the April 29 advisory committee raised a wide range of concerns, including the use of granulocyte-macrophage colony-stimulating factor (GM-CSF) as a control in Amgen’s pivotal Phase III study, the meaningfulness of the primary endpoint of durable response rate, and the “absence of a clear effect on overall survival.”

FDA’s Oncologic Drugs and Cellular, Tissue, and Gene Therapy advisory committees will review T-VEC for treatment of patients with regionally or distantly metastatic melanoma as a direct injection into skin lesions at a joint meeting.

Amgen acknowledged the possibility that it wouldn’t get that full indication during its earnings call last week (Also see "Amgen Hopes For Broad T-VEC Label, But Is Prepared To Discuss Narrower Approach" - Pink Sheet, 22 Apr, 2015.).

FDA’s briefing document and committee questions, though, suggest it has moved well down the path towards such an approval – even as agency concerns about study design and execution could sink the entire application.

Questions For Advisory Committees

• Is there a clinical benefit associated with T-VEC?
• What are the safety issues associated with the treatment?
• Does T-VEC have an overall favorable benefit-risk profile for some population other than the proposed indicated population?
• Are the dosing instructions sufficient?
• Do you have any recommendations to modify the proposed postmarketing protocol?
• Does T-VEC have an overall favorable benefit-risk profile for the treatment of injectable regionally or distantly metastatic melanoma?

FDA Identifies Possible Treatment Subgroup

“Considering that melanoma patients now have multiple treatment options, it is unclear whether talimogene laherparepvec offers an acceptable benefit-risk profile for the proposed indicated population,” FDA stated. “However, there may be melanoma patients for whom talimogene laherparepvec would be an appropriate alternative to the currently approved therapies.”

“For example, 16.3% of subjects in the talimogene laherparepvec group had a durable response, but subgroup analyses showed a durable response in 33.0% of subjects with Stage IIIB or IIIC melanoma who received talimogene laherparepvec, and a durable response in 23.9% of subjects who received talimogene laherparepvec as first-line therapy,” the agency added.

“Talimogene laherparepvec’s overall benefit-risk profile might be more favorable in such patients, or patients with fewer treatment options, than in the proposed indicated population.” FDA said those with fewer treatment options may include patients whose tumors do not have a BRAF mutation.

FDA asked the advisory committee to discuss whether the drug has an overall favorable benefit-risk profile for some population other than the proposed indicated population (see box with questions for panel).

Amgen Says FDA Previously OKed Comparator

T-VEC is a herpes simplex-1 virus that has been genetically modified to include a gene required for human GM-CSF production and to prevent replication in normal tissue. The genetically modified virus selectively infects tumor cells while sparing normal cells.

The first-in-class oncolytic immunotherapy has a unique mechanism of action and the review could set a precedent for other intralesional drugs (Also see "Amgen’s T-VEC AdCom May Shed Light On How FDA Reviews Intralesional Immunotherapies" - Pink Sheet, 13 Feb, 2015.).

Amgen’s biologics license application is based on a Phase III study, OPTiM, of 436 subjects with unresectable stage IIIB, IIIC, and IV melanoma who were randomized to receive T-VEC or GM-CSF. The primary efficacy endpoint was durable response rate, defined as the percentage of subjects with complete response or partial response maintained continuously for six or more months and beginning at any point within 12 months. Secondary efficacy endpoints were overall survival, overall response rate, time to response and duration of response.

T-VEC demonstrated a statistically significant higher durable response rate, including complete or partial response maintained for at least six months, compared with subjects who received GM-CSF (15.6% v. 1.4%). FDA said it was unclear whether T-VEC administration was also associated with improvement in overall survival.

But Amgen’s briefing document states that at the time of the primary analysis, T-VEC “showed a positive trend in survival” as the median overall survival was 4.4 months longer in the T-VEC arm than in the GM-CSF arm. The company also said T-VEC improved the overall response rate compared with GM-CSF (26.4% v. 5.7%) and had a higher complete response rate (10.8% v. 0.7%).

With regard to the selection of GMC-CSF as an appropriate comparator, Amgen said this was discussed and agreed upon with FDA as part of the special protocol assessment.

“A placebo control was rejected since subjects would derive no potential treatment benefit and its use would be a barrier to enrollment,” Amgen said.

FDA Cites Possible Investigator Bias

In addition to the choice of control, FDA questioned other aspects of the study design. It noted that the protocol stipulated that subjects were to receive treatment until week 24 or achievement of a complete response.

Four (1.4%) of subjects randomized to the T-VEC arm never received the drug and 172 (58.3%) of subjects in the T-VEC arm withdrew from the study before 24 weeks. In contrast, 14 (9.9%) of those in the control arm never received the drug and 106 (75.1%) withdrew before 24 weeks.

“Subject or investigator bias regarding the relative benefit of talimogene laherparepvec and the control may have influenced the determination that it was in the best interest of the subject to stop treatment or to be given other therapy for melanoma,” FDA stated.

Subjects that were assessed by investigators to have reached nine months of therapy or complete or partial response were referred to the Endpoint Assessment Committee (EAC) for a determination of durable response. FDA said there was discordance between the EAC and the investigators with regard to durable response in 21 subjects. Compared to the investigator assessment, the EAC assessed 14 subjects as not durable responders and 7 subjects as durable responders.

In addition, FDA said some of the assessments of durable response rate were subjective, susceptible to investigator bias, and could ultimately influence the determination of stable disease, complete response and partial response.

The agency also questioned the enrollment of subjects who had only small or very small lesions, expressing concerns about the reliability of injection and measurement.

FDA also said the evidence T-VEC had a systemic effect was limited and difficult to quantitate.

Other Treatments Have ‘Inherent Limitations,’ Amgen Says

Amgen is emphasizing T-VEC’s potential as an alternative treatment option to other melanoma treatments. And while the company appears to have accepted the possibility of a subpopulation approval, the briefing documents do not propose limiting use to a subpopulation.

“Even with the remarkable recent advances in the melanoma field, there is still a need for additional treatment options,” Sean Harper, Amgen’s VP of R&D, said in a statement. “If approved talimogene laherparepvec would represent an entirely new class of agent for the treatment of metastatic melanoma.”

In its briefing document, the company noted other melanoma treatments approved in the past few years and cited their “inherent limitations.”

Until 2011, only dacarbazine and high-dose interleukin-2 were available for treatment of unresectable or advanced melanoma. That year, FDA approved Bristol-Myers Squibb Co.’s Yervoy (ipilimumab), an immune checkpoint inhibitor, and Roche’s Zelboraf (vemurafenib), a v-raf murine sarcoma viral oncogene homolog B1 (BRAF) inhibitor. Amgen said overall response rates were 10.9% and 48.4% for Yervoy and Zelboraf, respectively, and complete response rates were less than 2% for either treatment.

In 2013, FDA approved GlaxoSmithKline PLC’s BRAF inhibitor Tafinlar (dabrafenib) and MEK inhibitor Mekinist (trametinib). Amgen said objective response rates were 52% and 22% for the two treatments, respectively, and complete response rates were less than or equal to 3% for either treatment. The combination of the two, also approved by FDA, had a complete response rate of 9%.

And in 2014, FDA approved Merck & Co. Inc.’s Keytruda (pembrolizumab) and Bristol-Myers Squibb’s Opdivo (nivolumab) programmed death receptor-1 (PD-1) blocking antibodies.

Approved Therapies for Advanced Melanoma
Drug	Mechanism of Action	Approval
Dacarbazine	Antineoplastic chemotherapy	1975
Proleukin (interleukin-2)	Cytokine signaling molecule	1998
Bristol-Myers Squibb’s Yervoy (ipilimumab)	Immune checkpoint inhibitor	2011
Roche’s Zelboraf (vemurafenib)	BRAF signal transduction inhibitor	2011
GlaxoSmithKline’s Tafinlar (dabrafenib)	BRAF signal transduction inhibitor	2013
GlaxoSmithKline’s Mekinist (trametinib)	MEK inhibitor	2013
GSF’s Tafinlar/Mekinist	BRAF/MEK inhibitors	2014
Merck’s Keytruda (pembrolizumab)	Programmed death receptor-1 (PD-1) blocking antibodies	2014
Bristol-Myers Squibb’s Opdivo (nivolumab)	Programmed death receptor-1 (PD-1) blocking antibodies	2014
Sources: FDA, Amgen

As for their limitations, Amgen said the rate of complete response is still low for immunotherapies, and with the targeted agents (e.g. vemurafenib, dabrafenib, trametinib) the duration of response can be limited due to innate or acquired resistance. In addition, Amgen said vemurafenib, dabrafenib and trametinib are indicated only for patients with BRAF mutant tumors, which comprise 40% to 50% of melanomas.

With regard to their safety profile, Amgen said each class of agent is associated with specific toxicities: the checkpoint inhibitors and anti-PD1 agents are associated with serious and sometimes fatal immune-related adverse events, and the targeted agents can be associated with severe skin toxicity, secondary skin cancers, and serious febrile reactions.

The most common adverse events with T-VEC were fatigue, chills, pyrexia, nausea, influenza-like illness and injection site pain. Serious adverse events included cellulitis, impaired wound healing and immune-mediated disease (e.g. glomerulonephritis). FDA said the incidence of treatment-emergent serious adverse events was 25.7% in the T-VEC arm and 13.4% in the control arm. FDA noted that one subject required a below-the-knee amputation for an infection that became resistant to medical interventions after T-VEC administration, but said the relationship to T-VEC is unclear.

Herpetic events, primarily oral herpes, were reported in 5.5% of subjects treated with T-VEC.

Amgen is conducting a postmarketing, prospective, observational, cohort study to evaluate the incidence of suspected herpetic infection, risk of secondary transmission, and risk of herpetic infections in immunocompromised individuals.

The company is also conducting an ongoing, multicenter, observational registry study to evaluate long-term safety of subjects who received T-VEC in clinical trials, and an on-going, single-arm Phase 2 study to evaluate clinical biodistribution and shedding in treated patents at the end of treatment.

Amgen is also proposing a Risk Evaluation and Mitigation Strategy with a communication plan to inform health care providers and patients about the risks of herpetic infections and accidental exposure associated with T-VEC.