At this meeting, at the ESMO 2019 meeting, I was presenting the results of the ETOP 9-15 PROMISE-meso trial on behalf of my colleague investigators. This is looking at a group of patients for whom there is no approved standard treatment. These are patients that have failed first line platinum pemetrexed chemotherapy with advanced relapsed pleural mesothelioma.
The hypothesis was that pembrolizumab would potentially improve progression free survival over standard chemotherapy and to this end we enrolled 144 patients with relapsed pleural malignant mesothelioma, all histologies, good performance status, adequate organ function. They were randomised to receive either the control arm of standard chemotherapy, which is either gemcitabine or vinorelbine at investigator discretion, or the investigational arm of pembrolizumab monotherapy. This was given every three weeks as per standard and given until disease progression but could be given beyond progression if patients were deriving ongoing clinical benefit. Treatment discontinued after two years if needed. Importantly, patients in the chemo arm could cross over to receive checkpoint inhibitors at time of progression.
What did you find?
The primary endpoint, unfortunately, was not met. There was no improvement in progression free survival for pembrolizumab over chemotherapy. We looked at a number of other key secondary endpoints and exploratory endpoints. Among secondary endpoints we looked at objective response rate and, in fact, pembrolizumab is an active agent with a response rate four times nearly as high as chemotherapy. The response rate was 22% for pembrolizumab compared to 6% for chemotherapy. Unfortunately duration of response was not different between the two groups. When we look at the waterfall plots we can actually see that pembrolizumab is associated with a greater number of responses and deeper responses than chemo.
We looked at the next key secondary endpoint of overall survival and, unfortunately, overall survival was not different between the two arms either, probably influenced by the crossover that we had because nearly 63% of patients that received chemotherapy crossed over to receive immunotherapy. We tried to account for that by statistical testing but even using those statistical tests we didn’t see a difference.
We did exploratory analyses looking at PD-L1 testing using the E1L3N clone looking at tumour PD-L1 expression and unfortunately there seemed to be no association of PD-L1 positivity and higher outcomes, both in terms of responses, progression free survival or overall survival in that group.
So the key take home message from the PROMISE-meso study is that pembrolizumab is an active agent against standard chemotherapy in malignant pleural mesothelioma but we don’t really know who the patients are that derive the greatest benefit from this agent. Because of the crossover overall survival was very similar between the two arms. To my mind, as long as patients are getting both chemotherapy and a checkpoint inhibitor that really does contribute to the best overall survival for this population.
Taking matters forward we need to do two things. We need to do some clinical and translational work on patients in this and other trials to really work out what are the biological signals that identify those that derive the greatest benefit so that we can perhaps restrict its use to those patients or at least not give it to those that don’t need benefit. Indeed, when we look at the spider plots for the pembrolizumab arm in this trial, some patients are deriving extremely long durable responses. So we definitely need to look biologically to investigate that further.
But, actually, if we take the parallel of lung cancer we’ve moved forward to using immunotherapy together with chemotherapy to warm the tumours up to make them more antigenic, to give greater efficacy to the immunotherapy. Using combination chemotherapy with immunotherapy in the front line setting is the preferred strategy at this moment. In fact, the ETOP group have another study which has just started to recruit taking front line patients, all of whom receive standard carboplatin pemetrexed together with bevacizumab chemotherapy and then are randomised to receive either atezolizumab or placebo on top of that, testing to see whether the four drug combination with atezolizumab is superior to the three drug combination.
