Cost regulators for NHS therapies in England and Wales say they are minded not to recommend funding for MSD’s Keytruda as a treatment for classical Hodgkin lymphoma, because its cost-effectiveness is uncertain.

Having rejected the drug’s for relapsed or refractory classical Hodgkin lymphoma in adults unable to have an autologous stem cell transplant who failed treatment with brentuximab vedotin, the National Institute for Health and Care Excellence is still considering whether it can be used in patients who have had both the transplant and prior treatment.

However, the appraisal committee has requested that the company provides a cost-comparison with Bristol Myers Squibb’s immunotherapy Opdivo (nivolumab), which is approved by NICE in this setting, to help better inform its decision.

Analyses suggest that having Keytruda (pembrolizumab) after brentuximab vedotin may lead to longer progression-free survival than current treatment, which would increase the number of people who can have curative allogeneic stem cell transplant.

But while there are models that predict how many people treated with the drug will be able to have allogeneic stem cell transplant and their long-term outcomes, compared with those having standard care, they can’t be used to determine cost-effectiveness, the Institute noted.

This is because of uncertainties in the assumptions that have not been fully explored, and also because the modelled survival estimates for standard care do not match either the clinical evidence or the claim that end-of-life criteria are met, it explained.

The anti-PD-1 therapy won European clearance in the Hodgkin lymphoma setting in May last year, on data from the KEYNOTE-087 and KEYNOTE-013 trials, which included patients regardless of the PD-L1 expression.

KEYNOTE-087 showed an objective response rate of 69 percent in the Keytruda (pembrolizumab) arm with a complete response rate of 22 percent and a partial remission rate of 47 percent, while KEYNOTE-013’s data showed 58 percent, 19 percent and 39 percent, respectively.