T-VEC injections into melanoma lesions show significant response rate

Bookmark and Share
Published: 2 Jun 2014
Views: 4472
Prof Kevin Harrington - Institute of Cancer Research, London, UK

Prof Harrington talks to ecancertv at ASCO 2014 about a randomised phase III trial of talimogene laherparepvec (T-VEC) versus subcutaneous granulocyte-macrophage colony-stimulating factor (GM-CSF) for the treatment of unresected stage IIIB/C and IV melanoma.

ecancer's filming at ASCO has been kindly supported by Amgen through the ECMS Foundation. ecancer is editorially independent and there is no influence over content.

This study was using a replication competent herpes simplex virus which is genetically modified. It’s genetically modified in a number of very important ways which makes it a cancer selective virus which has the potential to activate the immune system. So effectively it’s what called an oncolytic immunotherapy.

How do you make your herpes virus?

The Herpes virus was isolated from a clinical isolate from a cold sore; it was genetically engineered by deleting the ICP34.5 gene, two copies of the ICP34.5 gene, from the virus and replacing that with a cassette in which the human granulocyte macrophage colony stimulating factor was driven by a CMV promoter. In addition, the ICP47 gene is deleted from the virus and that has the role of robbing the virus of the ability to be invisible to the immune system. So normally ICP47 inhibits antigen presentation in infected cells so by deleting that gene antigen presentation proceeds and the virus then becomes visible. There’s an additional benefit, one other benefit from that ICP47 deletion, in that it unleashes the activity of an earlier promoter that drives replication of the virus more actively. So we have a virus that grows in cancer cells selectively but makes human GM-CSF as a potent immunostimulatory protein.

You compared this with injecting GM-CSF subQ?

The study randomised patients with stage 3B through to stage 4 melanoma, unresectable disease. Patients were randomised in a two to one basis to T-VEC or to subcutaneous GM-CSF. The T-VEC was administered as an intratumoral injection, initially at a low dose of 106 so-called plaque-forming units, and then escalating at the second dose to 108 PFU/ml, maximum dose of 4ml of injectate into deposits of melanoma. The control arm received subcutaneous GM-CSF 14 days out of a 28 day cycle.

Was there a rational basis for using subQ GM-CSF?

There are data to show that GM-CSF has some activity in this disease and certainly in discussion, when planning this study, which was in the era before the emergence of some of the more novel immunotherapeutics such as anti-CTLA4 and anti-PD1 blockade, this was a discussion that took place with the FDA and this was encouraged as being a reasonable choice of control arm in this study.

What happened?

The study shows that the primary endpoint, and I’ll need to explain what the primary endpoint was, the primary endpoint was an event called durable response rate. Durable response rate was a response that evolved at any time during the first year of therapy and which lasted for six continuous months. The patients were allowed to show progression before then going on to achieve response, because we know that often patients receiving immunotherapies will show that pattern, but once the patient began to respond, in order to be called a durable responder they had to stay in response for at least six months. So that was the primary endpoint of the study; the primary endpoint of the study showed that there was a very significant difference between the GM-CSF control arm and the T-VEC active therapy arm, such that the odds ratio for that was nearly 9.

What was the duration of response rates?

The response, by definition, had to be at least six months. The relative rates were around 17% versus 2%. The overall response rates between the two arms were around 26% in the T-VEC arm versus about 6% in the GM-CSF arm.

So these were significant responses?

These responses, and it’s important to stress this, these were assessed by an independent review committee that reviewed radiology and photography and clinical data. This was the entirety of the patients’ disease, not just the injected lesions.

What should doctors make of this in terms of progression-free survival?

That’s difficult to talk about in terms of progression-free survival because, of course, we know a number of these patients will show an interval progression and then will subsequently respond. So this is why this durable response rate primary endpoint was used in preference to progression free survival to take account of this and to allow us to capture the patients who were showing a response. What we do know is those patients who did respond to the T-VEC therapy frequently had very durable responses to that treatment.

What did the phase III trial show in terms of efficacy?

The data clearly demonstrate that injection of T-VEC into melanoma deposits yields response of those injected deposits. So we know from the analysis that about two-thirds of injected lesions will show a response and a significant number of those responses will be complete remissions of the individual lesions. We also know that about a third of uninjected lesions will also show a response and that’s indicative of a systemic immune response being generated in situ in the injected lesions. Furthermore, about 15% of uninjected visceral lesions showed a response to the T-VEC, again indicative of a systemic immune response. So what I would say to clinicians is that this looks like a very interesting treatment paradigm for melanoma, the notion that you can deliver local injections into deposits of disease, yielding both local responses but, actually more importantly, triggering systemic anti-melanoma responses throughout the body.

How does your research relate to what’s being done with checkpoint inhibitors at the moment?

This is really something that we need to work out. The data for T-VEC suggests that the most appropriate use of this agent might be relatively early in the course of the disease when the patient has in transit disease or nodal relapse or subcutaneous metastatic disease, in which case the agent appears particularly effective. There is also a drive towards assessing this agent in combination with checkpoint blockade because clearly the notion that we could drive local melanoma cytotoxicity with the virus, releasing a potently immunostimulatory cytokine such as GM-CSF, releasing tumour-associated antigens into that inflammatory microenvironment and, at the same time, use either anti-CTLA4 or anti-PD1, anti-PDL1 blockade to essentially take the brakes off the immune system looks like a very, very nice mechanism of combination therapy and certainly that is a direction of travel for future research.

Should doctors be encouraged by the progress that’s being made in melanoma?

Melanoma is absolutely the classical example now of a disease that has gone from being something that we regarded very nihilistically as something that we almost couldn’t treat or very rarely would see a response, to a disease now where many of our patients can realistically expect good responses to treatment. Indeed, in those patients who achieve very good responses we’re really seeing a tail on the curve such that some of these patients we can begin to think are cured of their disease.

What’s the take-home message?

This is a completely novel sort of therapy, so-called oncolytic immunotherapy. It’s a sort of therapy that is designed to deliver local therapy by local injection but a systemic outcome by activating systemic immune responses.