You are here

The Which: Clinical Evidence and Ongoing Trials in Anti-Angiogenic Treatments - Gastric Cancer or Gastro-Oesophageal Junction Adenocarcinoma

H. Wilke (Germany)

Hansjochen Wilke covers first-line chemotherapy for gastric and gastro-oesophageal junction adenocarcinoma. The presentation looks at significant clinical trials, their methodologies and results.


Gastric cancer is at the moment one of the tumour entities where anti-angiogenic drugs are in the focus.  Just setting the floor for what I will talk about –

Disclosures statement

Forget about this slide.

Full transcription

Full transcription

Gastric cancer is at the moment one of the tumour entities where anti-angiogenic drugs are in the focus.  Just setting the floor for what I will talk about –

Disclosures statement

Forget about this slide.

First-line chemotherapy for gastric and gastro-oesophageal junction adenocarcinoma

Where are we currently with so-called standard drug therapy in gastric cancer?  That’s where we are in the first-line setting.  Since more than 20 years we are playing around with doublets and triplets which means platinum/fluoropyrimidine compounds either as a doublet or adding anthracycline or a taxane.

But the only thing that we have learned from that is that you can replace this pattern by oxaliplatin, you can replace infusional 5-FU with capecitabine and what’s also clear meanwhile, there are two large randomised trains, you can replace cisplatin with irinotecan.

But overall, the response rate that had been achieved with all those doublets and triplets is usually in the range of 30-40% and if it’s better it depends on the patient selection.

The progression-free survival is in the range of six to seven months and the median overall survival is around 10 months.

We know meanwhile that if you add doxetaxel to one of those modified cisplatin/5-FU or oxaliplatin/5-FU schedules, this appears to improve the results.

What is also clear is that if you have a HER2-positive gastric cancer or adenocarcinoma of the gastro-oesophageal junction, that it clearly improves outcome of these tumours.  That’s where we are. Is there anything new concerning anti-angiogenic drugs in the first-line setting?  No. 

AVAGAST:  overall survival

At the moment, none of those drugs has really improved and the largest trial that has been published in the past was the so-called AVAGAST trial.

This was a trial which compared platinum 5-FU, capecitabine and cisplatin either with bevacizumab or placebo.  This was a relatively large trial with a lot of patients but the problem was that this trial was regarded as a negative trial with a survival difference of two-plus months.

The reason for that was the assumption of this trial and the statistical calculation based on the assumption that there will be a difference of 2.5 months and the patient number was not big enough to reach this goal, but nevertheless, plus two, if you compare two different chemotherapy schedules in gastric cancer and even in such a randomised trial you see a two-plus month survival benefit of  ..., this is a positive result without any doubt.

So for me this is not a negative result, especially focussing in on the data of the Western patient population, the so-called high-risk population with peritoneal carcinomatosis, bone marrow metastases, large tumour burdens, etc, etc and here this is not a pre-planned analysis but it gives you an idea that anti-angiogenesis means something in gastric cancer.  There was a survival benefit of four months and the hazard ratio in this post-hoc analysis was 0.6 with a high p-value, so this is for me a supportive trial confirming what I will talk about now.

Gastric and gastro-oesophageal junction adenocarcinoma:

Phase III studies of second-line treatment

If patients fail the first-line chemotherapy or first-line drug therapy, the next question is what shall we do now?  And you all know that since a couple of years, five, six years ago we really doubted that second-line chemotherapy means anything for the prognosis of those patients and meanwhile we have a number of randomised trials comparing chemotherapy which was usually single agent drug therapy versus best supportive care.

In all of those trials we saw a survival benefit with drug therapy versus best supportive care either when irinotecan was used, when a taxane was used and it was always in the range of about five months versus 3.5 months with best supportive care.  This is not only best supportive care; this was also chemotherapy on demand when the patients failed the best supportive care arm.

And there is another randomised trial with irinotecan versus paclitaxel and it also showed that paclitaxel which is not very frequently used in the Western hemisphere, is as active as irinotecan and the response rates were in the range of 15-20%, clearly in the same range as docetaxel and irinotecan.

REGARD: study design and objectives

What shall we do in this situation?  It’s clear that chemotherapy alone and at the moment the standard of care, if you just think about chemotherapy, a single agent drug therapy such as a taxane or irinotecan, do we have anything more?  Yes, we have meanwhile.

There were two large randomised trials which investigated the role of an antibody against VEGFR-2 and it’s ramucirumab and the first trial was the REGARD trial.

This was a study where just anybody was compared to best supportive care in a two by one randomisation fashion and the overall survival was the primary objective and of course all of the other things that we need in those trials.

REGARD: overall survival

And the outcome of these trials was clear.  There was a survival benefit just with this simple antibody which was well tolerated and many of the investigators said ‘We never knew whether we gave the placebo or the antibody’, so the toxicity profile in this trial was relatively slow but the survival benefit was also clear and the same for the progression-free survival, as you can see here. 

Crossover comparison of salvage therapies for gastric cancer: randomized trials

So for the first time we had an antibody in gastric cancer as a single agent not in combination with chemotherapy for example as trastuzumab was used and this is a retrospective comparison which is normally not allowed, but it gives you an idea that these are the data with chemotherapy where there’s best supportive care, 5.3, 5.2 in the UK trial and 5.3, that was in the other trial, in the Korean trial and four months in the German trial with irinotecan and ramucirumab was very close to all of this data concerning overall survival, so this is a clear alternative for patients if you just think about control of disease for a certain time.

RAINBOW: study design and objectives

The next trial was the so-called RAINBOW trial and this was a trial which compared ramucirumab plus paclitaxel versus paclitaxel alone, a large trial with 660 patients and of course the primary endpoint of this study was overall survival.

Of note in this trial, all patients had to have failed platinum/5-FU and 70% of the patients failed this combination while the patients were still on platinum/5-FU, so really resistant and the other 30% failed within three months after discontinuation of platinum/5-FU, so this was a relatively clearly defined patient population.

RAINBOW: overall survival

Again in this trial there was a clear survival benefit, a plus of 2.2 months for the overall patient population.  What we did in this trial is we restricted the number of Asian patients were recruited into this trial based on the experience that we had with the AVAGAST trial and other trials because the way Japan especially handles their patients is completely different to what we did a couple of years and this might have diluted some of the data of former trials.

Nevertheless, a plus of 2.2 months in the median overall survival and there was also a survival benefit at six and 12 months, so this was a clear result and fulfilled all the statistical things which we needed to say the primary endpoint was reached. 

RAINBOW: progression-free survival and response rates

The progression-free survival was clearly in favour for the combination.  The response rate was 28% versus 16% and the disease control rate 80% versus 64%, so this was the first trial where all efficacy parameters were clearly in favour of this combination.

RAINBOW: efficacy differs according to geographical region

This is an sub-group analysis which will be published probably this year, maybe next year, we will see but it shows something which is really important if we ask the question comparing the AVAGAST trial and these trials. 

There is a difference, yes in the way how patients responded with respect to overall survival especially and in Asia, and most of the patients came from Japan, the overall survival in the combination arm was 12.1 months versus 10.5 months which is far more than we would normally see in the Western patient population. 

But nevertheless the delta in favour of the Western patient population was 2.7 months.  There was a minus in South America.  Forget this data because they end at 40 patients and this was very at the far end of the trial when they started to recruit and I think it was just a slow number of the institutions that caught this data down.

Two-thirds of the patients came from Europe.  That’s of importance.  This is progression-free survival which is very similar as it is in Asia, in Japan especially and the response rate especially in those two, let’s say experienced areas or regions of the world are clear.

So no doubt the Europeans, and especially the Western populations which is Europe, North America and Australia have a clear benefit of this new combination.

Crossover comparison of salvage therapies in gastric cancer: randomized trials

And again this retrospective comparison – you see, that is what has been achieved in single agent therapy.  These were the data and the median overall survival, it was the whole study and it was 8.6 months for the Western patient population.

Without any doubt these were obviously the best results ever been published in a second-line setting of gastric cancer, especially in patients who are really resistant to platinum/5-FU which is the drug as the standard of care in the first-line setting.

So I would say if you need the most active therapy which we have at the moment in our hands, if you see your patients at risk, to the wire within the next eight or ten weeks, then this is probably what we should do and thank you very much.