ASCO 2011 Annual Meeting, 3—7 June 2011, Chicago

Treatment options for ovarian cancer

Professor Martin Gore – The Royal Marsden Hospital, London, UK

What’s so interesting in the last two years is that the whole definition of standard of care has been called into question, so that we know that patients need a platinum and a taxane but really there the recommendation stops because we have a wealth of data about other treatments that can be reasonably thought to be a standard of care rather than one single gold standard. And what has just been published by the GCIG are their recommendations not just about the appropriateness of different end points in clinical trials but also about what are acceptable and reasonable control arms, in other words standards of care in ovarian cancer. And they are quite varied and they vary from weekly taxol with carboplatin to IP therapy for optimally debulked patients, and of course the emerging bevacizumab update in first line and second line, which is very intriguing.

Any arguments about surgery?

Yes, there are lots of arguments about surgery. I think that some of those arguments are pretty sterile. You cannot name a randomised trial of all comers in first line of advanced ovarian cancer that does not show that residual disease is the most powerful prognostic factor after stage; such a randomised trial does not exist in the literature. So I think performing randomised trials to try and answer the question as to whether or not surgery is important really are a waste of time, because we know that if you can debulk a patient properly…

It’s worth it.

It is worthwhile so why are we turning somersaults over this? If a patient is not fit, or if you don’t think you can debulk them, it’s neoadjuvant chemotherapy; it’s as simple as that. In the age of targeted therapies, in the age of bevacizumab, in the age of PARP inhibitors, in the age of TKIs, we have much more important things to question and to spend our money on in clinical trials.

So post debulking what’s it going to look like based on the ICON data, the GOG data? Not all of the data’s mature yet.

No, it’s not mature but I think we are looking at maintenance therapies, I think we are looking at targeted maintenance therapies, I think we are looking at molecular selection of patients for those targeted therapies.

How do you molecularly target for bevacizumab?

Well I think that predictive factor is not known, but if we look at PARP inhibitors I think we are moving towards a place where if a patient has got a BRCA mutation, and maybe we should be looking for BRCA mutations in all the high grade serous tumours.

And doing it up front?

And doing it up front; I mean that’s the direction we’re heading, so that’s an immediate molecular selection. I’m not saying that is going to be the standard of care but that’s the direction. When we get molecular selection for bevacizumab, when we get molecular selection for TKIs in ovarian cancer, that’s where we are heading, and in five years’ time.

It will be like breast cancer?

It will be like breast cancer, we will have up front surgery, there will be plenty of tissue to do that molecular selection with, and away we go.

What are the candidates for the angiogenesis inhibitors, because of course this is powerful, powerful stuff and we’ve only got the mature data from one big trial on bevacizumab but there is a pipeline of other possible inhibitors coming along, and some of them oral , some can be given different ways.

Yes.

Where would you put your money on?

I am a bit loathe to kind of name a drug because I know from my experience...

No, I don’t want you to name to drug; I’m looking at name a selection technique, but maybe it doesn’t matter, maybe everyone should have clean-up treatment with an antiangiogenesis inhibitor post-surgery and post chemotherapy.

So my experience in renal cell cancer is that there is a tumour type where quite a lot is known about the biology and we have all these new drugs, and we are having a great deal of difficulty with the molecular selection at the moment in that tumour type. And I am sure we’ll get there in ovarian cancer but I don’t currently see which is the molecular selection that is going to help us. If we look at melanoma, for instance, we already know that a couple of molecular selections might well be helpful: c-KIT, BRAF for instance, and I think it will be like that for ovarian cancer and indeed all solid tumours, so if you take a pie chart for every solid tumour it will not matter that a mutation exists only in 2% of patients because the whole of that pie chart will be taken up by many such 2%s and you will have a range of drugs to treat them and that’s the way we’re going. I think the question in ovarian cancer as to which is the right antiangiogenic is complicated by the fact that there may be some specific biology in ovarian cancer relating to antiangiogenesis that we are not quite sure of yet.

A couple of things came out of a discussion I mediated at ESMO and Nicoletta Colombo picked up on two things: one is the residual tumour mass didn’t seem to be such a big deal in terms of getting response you got good response from the antiangiogenesis inhibitor, bevacizumab it was, irrespective of bulk being left behind. And the other issue was looking at the data of both the transatlantic trials, she felt that we might be stopping the bevacizumab too early at twelve months because she seemed to have the feeling that there was still an effect carrying on. What do you feel about that? Because this is such an important finding, that there is an overall survival difference, we have not enjoyed that in our time.

So that’s a very easy question for me to answer because the pre-clinical biology is very clear. If you stop an antiangiogenic all the blood vessels grow back very quickly exactly the way they were before, so to stop it in the maintenance setting makes no biological sense at all.

That’s what we did in both trials.

Yeah, which I think was a mistake but of course in the OCEANS trial they didn’t, they treated until PD. And then we have the colorectal data that suggests that even treating after progression might be important, and of course that’s what happens in prostate cancer with the hormone therapy, people carry on to hold the disease. So this is not really a new concept treating people certainly up to PD, and past PD, with targeted agents because after all hormone therapy in prostate cancer is a targeted therapy.

So BRCA1 mutations, OK that’s straightforward, and the time to get an answer is shortening. BRCAness, how do we get a handle on that?

I think the handle on that… we’ve currently got more of a handle on that than we think we have.

What, morphology?

Simple morphology and the grade of the tumours; that’s pretty good and I don’t think we should…

You’re happy with that?

I’m not delirious, but I think we do have something there. And that’s going to be a straight biological learning, I think, for targeting those patients.

And lastly, TKIs, what’s the selection going to be there?

I think that’s difficult because some of the emerging data suggests that combinatorial therapy there might be quite important, in other words combinations of targeted agents might be more important. Now, in other tumour types, of course, combining targeted agents is not a great idea.

It produces combined toxicity?

Combined toxicity and actually maybe you can do better with single agent sequential therapies in terms of having repeated prolongations of PFS. But some of the data does look like that and Debbie Armstrong showed some of those data this morning at ASCO. Yes, we don’t yet have a feel for what’s the right one, there is the pazopanib maintenance trial, of course, that’s out there, but I don’t think we are quite ready yet to be sure, and the 1120 trial as well which will be presented which will be of interest.

Lots of parallels with renal carcinoma aren’t there?

Well I think so because I’m in both, but once you learn how to handle these drugs in one tumour type you start to think about how you use them in another tumour type and there are lots of similarities, and actually lots of learnings, between tumour types as well with this class of agent.

Next year’s ASCO in ovarian cancer, what do you think the headlines will be? What are you sitting on the edge of your seat waiting for, mature data?

I think probably some more mature data from some of the maintenance studies is really what I am interested in. Since the introduction of taxol, which took response rates from 60% to 75-80% and took CR rates from 15-20% to 40%, what we have been waiting for is maintenance therapy. We are unlikely to get better than response rates at 75-80% so all the action is in maintenance therapy and that’s what excites me and to see it coming to fruition is wonderful to see, and I think that’s what we are going to be talking about for the next five years.

Martin, thanks very much indeed.

Thank you.

Back to the meeting.

Thank you.