JAMA Netw Open 2021; 4: e210380, ISBN 2574-3805 (Journal)
Herbrand A. K., Schmitt A. M., Briel M., Ewald H., Goldkuhle M., Diem S., Hoogkamer A., Joerger M., Moffa G., Novak U., Hemkens L. G., Kasenda B.
IMPORTANCE: In many health systems, access to off-label drug use is controlled through reimbursement restrictions by health insurers, especially for expensive cancer drugs. OBJECTIVE: To determine whether evidence from randomized clinical trials is associated with reimbursement decisions for requested off-label use of anticancer drugs in the Swiss health system. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study used reimbursement requests from routinely collected health records of 5809 patients with drug treatment for cancer between January 2015 and July 2018 in 3 major cancer centers, covering cancer care of approximately 5% of the Swiss population, to identify off-label drug use. For each off-label use indication with 3 or more requests, randomized clinical trial evidence on treatment benefits was systematically identified for overall survival (OS) or progression-free survival (PFS). Data were analyzed from August 2018 to December 2020. EXPOSURES: Available randomized clinical trial evidence on benefits for OS or PFS for requested off-label use indications. MAIN OUTCOMES AND MEASURES: The main outcome was the association between evidence for treatment benefit (expressed as improved OS or PFS) and reimbursement in multivariable regression models. RESULTS: Among 3046 patients with cancer, 695 off-label use reimbursement requests in 303 different indications were made for 598 patients (median [interquartile range] age, 64 [53-73] years; 420 [60%] men). Off-label use was intended as first-line treatment in 311 requests (45%). Reimbursement was accepted in 446 requests (64%). For 71 indications, including 431 requests for 376 patients, there were 3 or more requests. Of these, 246 requests (57%) had no supporting evidence for OS or PFS benefit. Reimbursement was granted in 162 of 246 requests without supporting evidence (66%). Of 117 requests supported by OS benefit, 79 (67%) were reimbursed, and of 68 requests supported by PFS benefit alone, 54 (79%) were reimbursed. Evidence of OS benefit from randomized clinical trials was not associated with a higher chance of reimbursement (odds ratio, 0.76, 95% CI, 0.45-1.27). CONCLUSIONS AND RELEVANCE: These findings suggest that in a health care system enabling access to off-label use, it was frequently intended as a first-line treatment in cancer care. Availability of randomized clinical trial evidence showing survival benefit was not associated with reimbursement decisions for off-label anticancer drug treatment in Switzerland. A transparent process with criteria considering clinical evidence is needed for evidence-based reimbursement decisions to ensure fair access to cancer treatments.