Advertisement

Predictors of a positive Cancer Drug fund decision

      OBJECTIVES

      Since 2011, the United Kingdom has set aside £200 million per year through the Cancer Drug Fund (CDF) to pay for oncology treatments not reviewed or approved by NICE. The CDF scores drugs on progression-free survival (PFS), overall survival (OS), quality of life (QoL), safety, unmet need, and strength of evidence (SE). The scores determine if the drug will be included on the CDF priority list. This analysis attempts to determine the weight each score has on the reimbursement decision.

      METHODS

      All available CDF decision summaries post April 2013 were analyzed. Scores for PFS, OS, QoL, safety, unmet need and SE were extracted from each decision summary. The CDF decision was classified as positive (recommended) or negative (do not recommend). Deferred decisions or drugs not scored were excluded. A probit model was used to estimate the probability of a positive decision based on the scores.

      RESULTS

      Drugs filling an unmet need, or drugs with the similar/improved toxicity predicted a positive reimbursement decision perfectly. Drugs with significantly worse toxicity predicted a negative decision perfectly. Because of perfect prediction, these variables (including SE) were excluded from the model. Of the remaining variables in the model (PFS, OS, and QoL), only OS was significant. An increase in OS was related to a higher probability of getting a positive reimbursement decision (p=.017). If OS was less than two months, the probability of a positive decision was 41%, but the probability of a positive decision increases to 99% for 6-7 months OS.

      CONCLUSIONS

      Unmet need and similar/improved toxicity are perfect predictors of a positive CDF decision. If a drug do not fill an unmet need or has worse toxicity, improvements in OS increase the probability of a positive decision. If the drug improves OS by 6-7 months there is a 99% probability of a positive decision.