Melanoma intralesional therapy – whether monotherapy or in combination with checkpoint inhibitors – is clearly here to stay, concluded a recent debate at the HemOnc Melanoma and Cutaneous Malignancies Annual Meeting in New York.
In the debate Merrick Ross, from MD Anderson Cancer Center in Houston, Texas, spoke for the motion “Intralesional Monotherapy is here to stay”; while Robert Andtbacka, from the Huntsman Cancer Institute at the University of Utah in Salt Lake City, opposed it, making a case for combination therapy.
Reviewing the spectrum of injectable advanced “unresectable” disease, Ross described how intralesional therapy could be used for stage 3 B/C regionally metastatic in-transit disease with or without nodal disease; stage M1a (distant skin, soft tissue and nodal metastases); and Stage M1a with low-volume visceral disease.
The dual treatment goals, he explained, are locally ablative therapy for local disease control (leading to palliation/symptom control) and the induction of systemic host immune anti-tumor activity. “You can deliver high concentrations of drug very easily, providing very good palliation of symptoms – and durable control may be curative,” said Ross.
Ross outlined the three main intralesional therapies:
- T-VEC (talimogene laherarepvec, Imlygic, OncoVEX): a herpes simplex virus type 1. Phase 3 completed. Already licensed
- PV-10: a 10% solution of the dye Rose Bengal. Phase 2 trial completed, phase 3 trial ongoing
- Coxsackie A21 virus [Cavatak, viralytics]: a naturally occurring “common cold” intracellular adhesion molecules 1(ICAM1) targeted RNA virus. Phase 2 trial completed.
Through a variety of mechanisms, each of these agents is able to selectively invade and lyse tumor cells, leading to the release of tumor-derived antigens that ultimately have the ability to potentiate a systemic T cell-mediated anti-tumor response.
The oncolytic immunotherapy concept, Ross explained, is part of the cancer immune cycle. “You start with ablation of the tumor which would express tumor-derived antigens, then when the tumor is destroyed, you prime dendritic cells to express activated T cells that proliferate and migrate to distant tumors.”
For the remainder of his presentation Ross focused on T-VEC and PV-10, the two agents for which he had the most clinical experience.
In the phase 2 study of PV-10, which took place in 80 patients, a complete response rate of 24% was achieved in both the injected lesions and the uninjected “bystander lesions,” with a disease control rate (DCR) of 71% for injected lesions and 55% for bystander lesions.
To get a good bystander response, patients need to have a good local response, demonstrated by the CR and PR bystander response being 67% for those patients with a strong local response, compared with 5% for those with a negative local response (P<0.0001).
For T-VEC in the phase 3 OPTiM trial, 436 patients with injectable, unresectable stage 3B-4 melanoma were randomized 2:1 to T-VEC intralesional injections (n=295) or GM-CSF (n=141).
Results showed a durable response rate (defined as an objective response lasting for at least 6 months) was achieved in 16.3% of T-VEC patients vs. 2.1% of GM-CSF patients (P<0.0001).
The objective overall response was 26.4% for T-VEC vs. 5.7% for GM-CSF. Furthermore, the risk of developing visceral or bone metastasis was reduced by 59% in patients treated with T-VEC, compared with GM-CSF controls. “This suggests some sort of systemic immune response has developed preventing clinical visceral metastases,” said Ross.
At 12 months, an interim overall survival analysis showed 73.7% of the T-VEC patients had survived versus 69.4% of the GM-CSF patients, indicating a survival trend favoring T-VEC. A subgroup analysis according to disease stage showed Stage 3B/C, 4 M1a patients (HR 0.57) had markedly improved survival in comparison to stage 4 M1b/c patients (HR 1.07)
Summarizing the data, Ross said that studies in stage 3B/C melanoma patients have shown that response rates were higher for intralesional therapies than approved systemic immunotherapies. Grade 3-4 adverse event rates were markedly lower for intralesional therapies (T-VEC<2% in comparison to ipilimumab 19%).
Introducing checkpoint inhibitors, said Ross, would change toxicity to a different category. “I have never seen a pituitary gland disappear or a colon rupture with intralesional therapies, side effects that have been seen with some checkpoint blocking agents,” he said.
Making the case for the superiority of combining intralesional therapies with checkpoint inhibitors – such as ipilimumab, nivolumab and pembrolizumab – Antbacka said that response rates with combinations are better than with either treatment alone and that combinations do not add toxicity.
In the Checkmate 067 trial of nivolumab plus ipilimumab, the best change from baseline in target lesion volume was 51.9% for the combination, versus -34.5% for nivolumab alone and +5.9% for ipilimumab alone. The 40% grade 3-4 adverse event rate seen with the combination, however, was of concern. “There are clearly patients who are not candidates for this combination,” said Andtbacka.
But recent trials combining intralesional therapies with checkpoint inhibitors have revealed enhanced immune responses and no indications of increased toxicity. Notably, a phase 1b trial of ipilimumab plus T-VEC, which enrolled 18 patients, showed no new safety signals or dose limiting toxicities. The DCR was 72 %, with durable responses in 44% and complete regression of uninjected non-visceral and visceral lesions in 39% (with 52% having greater than 50% regression).
In the ongoing phase 1b MASTERKEY-265 trial of T-VEC plus pembrolizumab an interim analysis of 21 patients at 17 weeks showed an ORR of 56% and DCR of 69% for the combination, with no patients having to discontinue treatment due to adverse events.
The phase 3 MASTERKEY-265 study, comparing T-VEC plus pembrolizumab to placebo plus pembrolizumab, has just been initiated. It is hoped that the study, which plans to recruit 660 patients, will provide a definitive answer as to whether combining a viral oncolytic agent and a checkpoint inhibitor improves outcomes and influences side effects.
Studies show the response rates found in advanced unresectable Stage 2/4 melanoma are markedly better for combination therapies (nivolumab plus ipilimumab 52%; T-VEC plus ipilimumab 50%; T-VEC plus pembrolizumab 56%), compared with monotherapies (ipilimumab 6-15%; pembrolizumab 27-38%; nivolumab 34-40%; T-VEC 26%).
“Where checkpoint inhibitors do not work, this can be due to the lack of lymphocytes. Intralesional therapies can boost checkpoint inhibitors through enhancing levels of lymphocytes,” said Antbacka.
Planned trials exploring other combinations include a phase 2 of pembrolizumab plus IL-12 electroporation (a technique using electrical pulses to improve the therapy reach within the lesion); a phase 1b trial of pembrolizumab plus CVA21; and a phase trial of pembrolizumab plus -PV-10.
“Combination therapy is better than monotherapy in advanced melanoma because there is no added toxicity and therapies enhance the effect of checkpoint inhibitors,” said Antbacka. “Monotherapy is clearly only half as good as the combination.”
Sanjiv Agarwala, the meeting’s chairman and moderator from St. Luke’s Cancer Center in Bethlehem, PA, said:
“We have to realize that intralesional therapy is not going anywhere, it is here to stay. It is a new paradigm for potential combinations, and perhaps in the future the ultimate melanoma regimen is going to be with an intralesional therapy with a systemic, checkpoint inhibitor. Monotherapy also is applicable to specific patients.”