Time to Entry for New Cancer Medicines: From European Union-Wide Marketing Authorization to Patient Access in Belgium, Estonia, Scotland, and Sweden.

OBJECTIVES

First, to quantify the median time from European Union (EU)-wide approval to first use (launch) for a sample of cancer medicines and number of launches in Belgium, Estonia, Scotland, and Sweden as of June 2015. Second, to assess whether longer times to launch or lack of launches affected medicines with high or low expected additional clinical benefit. Third, to identify possible determinants of the probability of a cancer medicine to be launched.

METHODS

Correlation between time to launch and a set of variables hypothesized to affect launch was tested using a complementary log-log model for a sample of 46 cancer medicines that obtained EU-wide marketing authorization between 2000 and 2014.

RESULTS

In median, for a sample of 24 cancer medicines that obtained marketing authorization between 2010 and 2014, the expected time from EU-wide marketing authorization to first use of a medicine was shortest in Sweden, 3.1 months, followed by Scotland (9.3 months), Belgium (14.8 months), and Estonia (27.8 months). Median times to launch were longer for the entire sample of 46 cancer medicines that obtained marketing authorization between 2000 and 2014. In the all-country model, medicines with shorter times to submission for reimbursement, local manufacturers headquarter (or local sales representative), and a Food and Drugs Administration priority review or a combination of expedited approval programs and medicines launched in Scotland and Sweden were associated with a higher hazard of launch. Longer times since EU-wide approval initially correlate with an increased hazard but as time further elapses they negatively affect the hazard of launch.

CONCLUSIONS

Median times from marketing authorization to first use of cancer medicines were shorter for medicines launched between 2010 and 2014 versus sample-wide (2000-2014). In Estonia, more medicines than in the other countries were not yet launched at the end of the observation period. There was no correlation between Prescrire and European School of Medical Oncology Magnitude of Clinical Benefit Scale ratings of added clinical value and time to launch.