The Cancer Drugs Fund (CDF) in England – in its original incarnation between October 2010 and April 2016 – was a ring-fenced fund to provide access to cancer drugs not routinely funded by the NHS. Following reform during 2016, the CDF now offers a managed access option for cancer drugs where there is uncertainty in the cost effectiveness. Drugs funded by the old CDF that had previously been rejected by NICE, have gone through a Rapid Reconsideration process, allowing manufacturers to submit an updated evidence base combined with any appropriate commercial proposal to enable continued cost-effective access.
One of the criticisms levelled at the CDF was the system’s failure to collect consistent high quality outcomes data from the publicly funded use of these drugs. A National Audit Office review of the CDF noted that data collection improved during the lifetime of the Fund, with mandation of data entry and improvements in data sharing arrangements, but that at the time of the report (November 2015) there remained issues of missing data. There has been little use of CDF data in decision-making; for example, we observe no significant use of CDF-generated data in the Rapid Reconsideration submissions for former CDF drugs.
This began to change with a recent appraisal by NICE, of brentuximab vedotin (Adcetris) for treating Hodgkin lymphoma. An important uncertainty in the appraisal related to a particular subgroup of patients – those who cannot have a stem cell transplant. Following treatment with brentuximab vedotin, some of these patients may become eligible for a transplant, which when combined with high-dose chemotherapy has potential to cure the disease. The uncertainty lies in the proportion of patients who subsequently go on to successful transplant treatment – clearly a key component in evaluating this potentially significant benefit provided by the drug.
In a managed access agreement developed within the new CDF, brentuximab vedotin will continue to be made available to this group of patients. In parallel, data on stem cell transplant rates from patients receiving the drug through the old CDF will be analysed retrospectively, by compiling information from the treating clinicians of relevant patients.
To the best of our knowledge, this will be the first example of data from drug usage through the former CDF being used in decision-making, and certainly the first example of its being mandated. The existence of this agreement suggests that the manufacturer is confident that the drug will deliver in the clinic: ie that these retrospective data will validate their estimates in the appraisal submission, and hence support the pricing of the drug. We look forward to the follow-up appraisal with the new data – not only for a decision that benefits NHS patients, but also as a data generation model which others may be able to follow.
Is this the only example of data from CDF usage being used for decision-making? We noticed another example in a recently published draft guidance for sorafenib (Nexavar) in hepatocellular carcinoma. The manufacturer referenced a recently published independent audit using CDF data1, as validation of their extrapolation of trial survival data. In the end the audit study was not used for validation, as its patient population was different from that in the trial; however, the data may have influenced the specification of the eligible patient population in the final appraisal. It may also have influenced the manufacturer’s decision not to collect further data within the new CDF, but to agree a commercial arrangement that would allow recommendation for routine commissioning.
With continued pressure for early access to promising medicines – especially in areas of high unmet need – we can expect to see special funding mechanisms evolving to get these medicines in use before a complete evidence base is available. It is reassuring to see real examples where decision-making benefits from that investment.
Dr Liz Morrell
- King, J et al 2017. Sorafenib for the Treatment of Advanced Hepatocellular Cancer – a UK Audit. Clinical Oncology 29(4), 256–262