The KEYNOTE-040 study is a phase III trial in patients with recurrent or metastatic squamous cell carcinoma of the head and neck who had failed a platinum-containing regimen. This was predicated on the promising data that we saw with single agent pembrolizumab and, in fact, with single agent anti-PD1, anti-PD-L1 in similar patients. Now the formal test of pembrolizumab versus standard of care in essentially second line recurrent or metastatic head and neck cancer.

The trial did not meet its primary endpoint and really based on fairly stringent statistical design. However, when we look at the data a little bit more closely what we see is an efficacy portfolio that is completely in conjunction and reminiscent of what we know about anti-PD1 in this space. The final hazard ratio is 0.81; the hazard ratio that would have been positive is 0.8 so very close to being positive. The response rate was 14%, exactly what we would expect with pembrolizumab, and when we look at the subsets that expressed PD-L1 in the tumour we see even more exaggerated benefit with in the low expressers a hazard ratio of about 0.74 and in the very high expressing tumours a hazard ratio of 0.54. So clearly pembrolizumab is active, it’s effective and it remains an important part of our treatment in patients with recurrent metastatic head and neck cancer.

So will there be follow-up trials looking, like you say, more closely at these subsets or adding in something to just try and bump that response up to a significance? Because it would be a shame to see it left by the wayside.

Exactly, and I don’t think it will be left by the wayside at all. Pembrolizumab as a single agent still has a role and it still has utility. I don’t believe that practices will change based on these results, in fact if anything it augments what we already know. However, looking forward one can imagine an increased benefit with combinations, perhaps with chemotherapy, perhaps with other immunotherapies and looking further and further at subsets that may have an especial benefit to these types of agents.

What we saw from the toxicity profile between pembrolizumab and standard of care again is exactly what we expected. Pembrolizumab is a better tolerated agent, at least when we look at the rates of adverse events overall, and a much lower adverse event profile when we look at grade 3-5. In fact, a third of the grade 3-5 adverse events as compared to standard of care chemotherapy. So overall this seems to be an effective agent that is better tolerated than standard of care chemotherapy.