Oesophageal cancer carries a very heavy global cancer burden accounting for more than half a million deaths per year. The treatment of this disease still remains very suboptimal – patients receiving first line therapy for advanced oesophageal squamous cell cancer had a median survival of around 10 months or less. So we conducted our study, it was called the CheckMate 648 study, which is the largest randomised controlled trial conducted in this disease setting where we evaluate a dual checkpoint inhibition using ipilimumab and nivolumab as well as checkpoint inhibition plus chemotherapy, nivolumab plus chemo, versus chemotherapy alone in our study.
What was the methodology used in this study?
In this study patients with unresectable metastatic squamous cell cancer of the oesophagus were recruited irrespective of the tumour PD-L1 expression. Patients were untreated for advanced disease. Altogether 970 patients were randomly allocated into three different treatment groups. The first group was nivolumab plus chemotherapy, the chemotherapy used is cisplatin and 5FU given every four weeks. The second experimental arm is the combination of nivolumab plus ipilimumab, here the ipilimumab is given every six weeks. The third arm, which is the control arm, is chemotherapy alone using cisplatin and 5FU given every four weeks. The primary endpoint of this study was using the biomarker specified tumour PD-L1 expression of 1% or more and is used in both overall survival and progression free survival as the primary endpoints. The secondary endpoints included overall and progression free survival in all randomised populations, objective response rate and safety.
What were your findings?
As we could expect in a large randomised controlled trial like this, the baseline characteristics of patients were well-balanced across the three treatment arms. Just under half of our patients were PD-L1 expression positive, i.e. 1% or more. Around 70% of the patients came from Asian countries.
In our first treatment comparison, nivolumab plus chemotherapy versus chemotherapy alone, and in our primary patient population of PD-L1 expression 1% or more we found a highly statistically significant improvement in overall survival in favour of nivolumab plus chemotherapy. The hazard ratio was 0.54 which is a 46% relative reduction in the risk of death in favour of nivolumab plus chemotherapy. This translated into more than 6 months’ improvement in median overall survival and also a very significant difference in terms of one year overall survival rates.
In the all randomised patient population we again observed a statistically significant improvement in overall survival in favour of nivolumab plus chemotherapy. Again here we observed an improvement of median survival of 2.5 months. In terms of progression free survival we observed a statistically significant improvement of PFS in our PD-L1 positive expression patients in favour of nivolumab plus chemo. However, in our all randomised patient population there was a trend towards improvement but that did not pass the statistical boundary.
But what we observed in CheckMate 648 in terms of PFS, because the progression free survival was assessed by central review, there were in fact a lot of patients who move on to subsequent treatment therapy before the CT scan showed disease progression. So as a pre-planned analysis we also looked at investigator assessed progression free survival and here we observed a superior PFS in favour of nivolumab plus chemotherapy, even in the all randomised patient population.
So in our second treatment comparison looking at a chemotherapy free option of ipilimumab plus nivolumab versus chemotherapy, again in our primary patient population of PD-L1 positive expression we observed a significant overall survival benefit in favour of ipilimumab plus nivolumab compared to chemotherapy alone.
In our all randomised patient population we also observed a statistically significant improvement in overall survival in favour of ipilimumab plus nivolumab compared to chemotherapy alone. For progression free survival, as we would expect from other randomised controlled trials for IO or checkpoint inhibitors versus chemotherapy, there were no differences in terms of PFS between the ipilimumab plus nivolumab compared to chemotherapy. However, there was a trend towards a better progression free survival especially in the latter half of the PFS curve in our PD-L1 expression positive patient population.
The other secondary endpoints were objective response rate and here in both the treatment comparisons i.e. nivolumab plus chemotherapy versus chemotherapy alone as well as ipilimumab plus nivolumab versus chemotherapy alone, again in our primary patient population of PD-L1 positive patients we observed a much higher response rate in the IO containing arms compared to the chemotherapy arm. The complete response rate was more than three times higher for the nivolumab containing arm compared to the chemotherapy alone. For the response that was achieved by nivolumab either with chemotherapy or nivolumab plus ipilimumab they were much more prolonged compared to the objective response that was achieved by chemotherapy alone.
Lastly, for safety what we observed across all the patients we treated across the three arms, the safety profiles are different between the chemotherapy containing arm compared to the IO containing or the ipilimumab plus nivolumab arms. Importantly, the treatment related mortality was very low across all three arms and very similar across the three arms.
In conclusion, with the overall survival benefit seen in both our primary patient population of PD-L1 expression positive of 1% or more as well as in the all randomised patient population we now have two new potential standards of care with nivolumab plus chemotherapy or nivolumab plus ipilimumab in patients with advanced oesophageal squamous cell cancer based on our CheckMate 648 study which is the largest randomised controlled trial conducted to date in this disease setting.
How can these results impact the future treatment of ESCC?
From this significant overall survival benefit that we’ve seen we now have two new treatment options. Now when we have patients with newly diagnosed advanced oesophageal squamous cell cancer the decision is to give an IO drug, i.e. nivolumab in this situation, and the decision really for the clinicians is whether they want to add chemotherapy to nivolumab or whether we should add ipilimumab to nivolumab. One, of course, is still a chemotherapy containing option, the second option with ipilimumab and nivolumab is a chemotherapy free option.
