35th ESMO Congress, 8–12 October 2010, Milan

Interview with Professor Timothy Perren (Leeds Teaching Hospitals NHS Trust, UK)

The ICON7 trial

You’re presenting the ICON data tomorrow and we’ve been discussing a lot about that at this conference because the flag was waved at ASCO earlier in the year and people are really quite excited now that we might, after fourteen or fifteen years of not very much happening in advanced ovarian cancer, we might be beginning to see a little crack happening. Can you enlighten us?

Yes, I think that’s right, I think it’s a really exciting time, as you say. This is the first new drug that we’ve looked at in ovarian cancer since the middle of the 1990s that’s shown any effect.

And that’s Bevacizumab?

Yes, and the drug concerned is Bevacizumab or Avastin. So the ICON7 trial was a randomised trial; the control arm of the trial was standard chemotherapy with Carboplatin and Paclitaxel given in the usual way for six cycles, one cycle of treatment every three weeks and then no further treatment. In the research arm of the trial patients were treated with chemotherapy and Bevacizumab for six cycles and then received Bevacizumab for a further twelve cycles of treatment which made eighteen cycles or twelve months of treatment in total. The primary endpoint for our trial was progression free survival and the study showed that progression was delayed and occurred less frequently in patients treated with Bevacizumab than in the control arm.

Now these are patients with advanced ovarian cancer. Had they had debulking?

So these were patients who had got ovarian cancer, primary peritoneal cancer or fallopian tube cancer, so this is a population of patients that would be treated as ovarian cancer in any cancer centre around the world. The patient population that we included in ICON7 was designed to be as representative as possible of the population of women with ovarian cancer who would receive chemotherapy as part of their standard treatment.

So our study included a proportion of patients who’d got early stage disease, so stage 1 or stage 2a disease with high risk factors, so clear cell cancers and grade 3 cancers as well as the patients who had got optimally debulked and sub-optimally debulked more advanced ovarian cancer - stage 3 and 4 disease. The vast majority of the patients had had surgery, about 2% of the patients didn’t have surgery and that was because surgery just was not indicated because of the nature and extent of their disease, it was never going to help those particular patients.

Did it matter which group these were in?

It didn’t appear to matter at all. The effect that we saw for Bevacizumab seemed to be replicated across all of the sub-groups that we looked at.

So volume of disease didn’t seem to be an agent?

So the amount of residual disease didn’t appear to affect what we call the hazard ratio, so the statistical effect of reducing the risk appeared to be constant, whether the patients had got early stage disease or more advanced disease.

But that’s quite new. Most of the studies I’ve seen of new agents in ovarian cancer have been better in the less disease there was…

Well this is interesting of course because, as ever, if you’ve got a statistical effect that’s constant across the groups you see apparently less benefit in numerical terms for the low risk patients and a bigger benefit in numerical terms for the high risk patients because if you reduce the events by 20%, 20% of a lot is a lot and 20% of a little is a little. So it appears as though the effect is bigger although statistically the effect size is much the same.

Side effects of Bevacizumab?

Side effects of Bevacizumab are really very manageable. The things we know about Bevacizumab is that it causes hypertension, raised blood pressure, it can cause proteinuria and we saw some of that in patients and about 18% of patients who were treated with Bevacizumab needed to be treated with additional anti-hypertensive drugs as opposed to about 3 – 4% of patients in the control arm. But that was very easily done, usually with one, sometimes with two anti-hypertensive drugs and usually the effect on blood pressure was reversed when the Bevacizumab was stopped and the patients were able to come off the anti-hypertensive drugs later on.

So there had been concern from some of the earlier studies with Bevacizumab that fistulae would develop, so abnormal connections between internal organs, or abscesses or perforations of the bowel.

Also seen in the GI studies.

As was seen in the GI studies, seen in the advanced ovarian cancer studies and seen, to some extent, in the lung cancer studies as well. And indeed we did see those but they occurred at a very low level - at most 5, 6, 7% of patients developed those sorts of complications. The complication rate was about twice as high in the patients that received Bevacizumab but it was not confined to the group of patients that received Bevacizumab and it occurred in the control group as well. So if you had, say, a 6% chance of developing a fistula in the Bevacizumab arm, about 3% of patients in the control arm developed such complications.

Of course Bevacizumab is given IV, what’s the patient tolerance like over a whole year of therapy?

It doesn’t seem to add substantially to the burden for patients receiving the treatment and when we looked at quality of life, the quality of life was seen to improve as patients went through both chemotherapy alone and chemotherapy plus Bevacizumab. So the addition of Bevacizumab didn’t seem to adversely affect the patients, that improvement as the chemotherapy started to work was seen in both groups. So it did seem to be reasonably well tolerated. It’s a monoclonal antibody as opposed to one of the small molecule targeted drugs and in general the monoclonal antibodies seem to be tolerated much better. And when you get on to the maintenance phase, of course, the chemotherapy is not being given with it, you’re just receiving the Bevacizumab infusion and patients seem to tolerate that part of the treatment really quite well, not dissimilar to the patient experience when they’re receiving a drug like Herceptin of course, which is also a monoclonal antibody.

Do you think you stopped too soon?

That is one of the questions that we will have to answer in future research. I don’t know the answer and clearly we can only comment on the research that we’ve done. I guess what we know from GOG is that continuing Bevacizumab beyond the end of chemotherapy was important because in one of the arms of that study the Bevacizumab was given with the chemotherapy but not continued beyond the end of chemotherapy as maintenance. And although they saw an effect of Bevacizumab it was a very, very much smaller effect than they saw in the maintenance group. The GOG study carried on with Bevacizumab for about three or four more cycles than we did in our study, they also used twice the dose but they treated a much higher risk population of patients. So they had no early stage patients and they had no optimally debulked patients in their study. So whilst about 50% of our patients had had a progression event, about three-quarters of their patients had had a progression event. So it may be that the apparently bigger effect that we see in the GOG study is a reflection of the patient population that was treated and the much higher event rate that follows from that.

So the maintenance question seems to be to be pretty well wrapped up and it’s really very, very encouraging and this is a big step forward. It may just seem to be a few months at the moment but it’s only the progression free survival we’re looking at, we’re not looking at all the curves and we’ve got to wait another few years, maybe a couple more years for survival.

Overall survival which, of course, is absolutely the key endpoint, will be ready in 2012. At the minute we’ve been asked to do an analysis of the overall survival for what’s potentially a regulatory submission by Roche and Genentech to the European Medicines Agency and the FDA. And as part of that we’ve had to do an overall survival analysis, they don’t accept the data unless that’s been done. So we have seen a non-significant trend in favour of improved overall survival but no more than that.

It’s just really too early and you and I both know that.

It’s far too early.

Tell me about the other question which is chemotherapy up front plus or minus Bevacizumab, it’s not the question you asked, I know, but what’s your gut feeling about that?

You’re talking about primary chemotherapy plus or minus Bevacizumab?

Yes, primary chemotherapy plus Bevacizumab. Is the case made now, what do you think?

Well we haven’t tested Bevacizumab in that group and that’s the key thing. But the thing that’s changed in ovarian cancer management in the last five years or so is that we are now using much more chemotherapy as the very first treatment of patients with ovarian cancer, we’re not operating on those patients who have got advanced disease, who have got ascites, who have got poor performance status, who have got low albumin, who have got oedema. Those patients do really very badly indeed with surgery so they’re receiving chemotherapy as the primary treatment up front. Now can you add Bevacizumab into that treatment safely? At the moment we don’t know, there are concerns about Bevacizumab and its influence on wound healing; we’ve got the concerns about Bevacizumab from the advanced recurrent ovarian cancer patients with the fistulae and perforations and what have you. It’s an area that we’re going to have to look at, I’m sure, but one of the questions that will arise is what do we do with the Bevacizumab at the time that the patients are operated on? Because our standard practice in that group at the moment is to give three cycles of chemotherapy, improve the patient, get the disease to respond, then operate and then finish the chemotherapy. And that’s where those studies have taken us at the moment so we’ve got studies looking at that, the EORTC did a very important study, the MRC study will be ready in the next year or eighteen months, that’s finished recruiting now looking at the same question, but neither of those studies have looked at a chemotherapy alone versus surgery followed by chemotherapy. So I think that’s the unanswered question – if you are going to use it, what’s the optimal time to do the surgery? And of course doing the surgery at the end of chemotherapy has never been shown to be useful in any of the studies that we’ve done so far.

And it’s very unpopular with the patient as I recall. I used to do a lot of interbrachial  studies, as you know, it was not easy to sell. Tim, thank you very, very much indeed, I really appreciate you giving us a little bit of glimmer here. Very optimistic data and I think we might be looking at what they call a paradigm shift, are we?

One has to hope so. I think the paradigm will shift in a little while, it’s not quite shifted yet.

Not quite there yet but on the way.

More data.

But clearly anti-angiogenesis treatment, and there are lots of them around, lots of possibilities, does seem to have some sort of impact on ovarian cancer.

I think that’s absolutely right and my final conclusion in the presentation that I will give tomorrow will be that this treatment will influence the discussions that doctors have with their patients but I think more importantly than that it’s going to influence the design of the next generation of trials.

Tim, thank you very much indeed, I appreciate it.