EADO Congress 2015

Vemurafenib in patients with BRAF600 mutation–positive metastatic melanoma

Prof Axel Hauschild - University of Kiel, Kiel, Germany

You’ve been presenting a study on vemurafenib, so that’s a RAF inhibitor, and the analysis of long-term response. This, of course, is in patients with BRAF, mutated BRAF, V600 mutated disease. It was an open label study, can you tell me about the study? It’s quite big, over 3,000 patients, so what did you do and what did you find?

That’s an early access programme so it’s an EAP and therefore the number of patients treated in various countries is very high. We call this study a real life study because patients were not treated under the conditions of a clinical trial but in a real world setting. The findings are quite interesting because so many patients have been treated there are certain subgroup analyses feasible. We found out that patients with a long duration of response, as defined as more than twelve months, have the following prognostic factors. It’s typically patients with low LDH measured in the serum of the patients; it’s typically patients with no brain metastases and patients who were belonging to a M1a stage of disease which means this is patients typically with relatively low tumour load who are doing well. These patients can have very long responses, over years.

And, of course, they were all treatment naïve, they had not a RAF inhibitor before?

That’s true.

But they weren’t all second line, first line, what were these patients?

These were first line patients, you are absolutely right. But I was presenting at this meeting data on second line as well. So we had four clinical trials to report on vemurafenib. The first one was a phase I which was presented already in the year 2011. The second one was the so-called BRIM-2 and then BRIM-3 which was bringing the drug to the market, and this is the early access programme. So there are four clinical trials available and we have a meta-analysis of all the clinical trials which show exactly the same prognostic parameters.

Significantly you’ve been talking here about duration of response and you’ve got a big increase in duration of response. What did you get and what’s the significance of this? It’s a big increase but still the numbers of patients are quite small at one year, aren’t they?

You are right, I agree. But on the other hand we need to say the question was if there is long-term survival on the horizon. We figured out that the long-term survival, and the longest data which is available is the three year survival, it’s in the range of 30%, three zero, not thirteen, one three, which means one third of the patients are still alive after three years. Some other colleagues were saying, ‘Well, with targeted therapies everybody is dying quickly because there is resistance,’ but obviously there are people who don’t have resistance.

So you’re beginning to show there is a tail of survivors with these targeted therapies?

If we are using the word tail we need to be sure that this is a plateau. The tail, yes it’s true, but we don’t know if this is a plateau. I would love to see the four and five year survival data. If four and five year survival is in the same range we can talk about the plateau and then the tail would have a clinical impact.

In this study with vemurafenib, initially then, what are the figures, very briefly, that have come out of this that you would pass on to doctors?

There are 10-12% of patients still on the treatment after three years which means that it’s a reasonable number of patients who don’t have a progressive disease. There’s another 10% of patients still under treatment who had before progressive disease. So it’s treatment beyond progression. That was a scene which was never touched in the past but we know from tyrosine kinase inhibitors and also from the competitor, the dabrafenib, that this matters and these patients still have a benefit. So there might be minor progressive disease, just one new lesion, so it’s not clinically meaningful if they progress so we decided to continue. I think this is the main finding but still we need to say longer duration of more than twelve months is only a small subset of patients, not more than one third of the patients.

And your study with dabrafenib, how did that go?

We evaluated the BREAK-3 trial and I was the global PI of this clinical trial for long-term responses. 31% of the patients are still alive after three years so the figures for vemurafenib and for dabrafenib are almost the same. You cannot compare the trials directly because for vemurafenib it was phase I and phase II where we can report the three year survival and for dabrafenib it was a phase III trial. But it’s almost the same and this looks quite good.

So what’s the take-home message on these RAF inhibitors for doctors?

The take-home message is the lower the tumour load is the better the outcome of the patients is. So this is against the dogma which is in the room if you treat melanoma patients because typically we are arguing that patients with high tumour load are the best candidates for BRAF inhibitor treatment which is true because they might have a nice response. But the best outcome is always in the same group of patients, those with low tumour load and favourable prognostic factors which means for first line treatment you can either use the BRAF inhibitors or you can use immunotherapy. It’s useful in every group of patients.

But in the sequencing of therapy, because of these longer term survivors, would you start with a targeted therapy now, do you think possibly?

No, I would say we simply don’t know it. There is a consensus among most but not all key opinion leaders in melanoma that there is no data available at this point of time which points to one direction or the other one which means BRAF inhibitors first or the immunotherapeutic agents. But I need to say something which is very important because now we are facing single agent treatment with dabrafenib and vemurafenib and single agent treatment is no longer a standard of care. It is combo of BRAF and MEK inhibition and in our new German guideline which will be released shortly it’s very clear that only the combo is used.

And of course we’ve got quite a bit of data on combinations so that’s the way to go, you reckon, at the moment?

Absolutely.