Comparing manufacturer submitted and pan-canadian oncology drug review reanalysed incremental cost-effectiveness ratios for novel oncology drugs

Abstract

Background: To determine the magnitude of difference between manufacturer-submitted and pan-Canadian Oncology Drug Review (pCODR) calculated incremental cost-effectiveness ratios (ICERs), incremental cost (∆C), and incremental effectiveness (∆E); to examine whether there is a significant difference in the proportion of ICERs deemed cost-effective; to evaluate trends in the ICERs over time; and to identify methodological issues in manufacturer-submitted economic models. Methods: Economic guidance reports for all drug indications submitted from July 2011–November 2018 were extracted from the pCODR database. Cumulative distribution plots were constructed to compare the manufacturer-submitted economic values with both the pCODR lower-and upper-reanalyzed estimates. The proportion of drug reviews considered cost-effective at varying willingness-to-pay (WTP) thresholds by the manufacturer and pCODR were calculated. Manufacturer changes in ICERs over time from 2012 to 2018 were determined. Recurring methodological issues with manufacturer submissions were tallied. Results: There were 73 unique indications that were included. Manufacturer-submitted ICERs were consistently lower than pCODR estimates for most indications. Manufacturer-submitted ICERs were generally more cost-effective over a range of WTP thresholds. From 2012 to 2018, manufacturer and economic guidance panel (EGP) lower limit reanalyzed ICERs did not change significantly over time. However, EGP upper limit re-analyses did show decreasing cost-effectiveness (increasing ICERs). The two most common issues identified in the manufacturer-submitted models were related to survival time horizon and utility estimates. Conclusions: Manufacturers tend to overestimate the cost-effectiveness of their therapies when submitting economic models to pCODR. Although certain methodological issues are still common in manufacturer-submitted models, revision rates are high for most issues raised by pCODR.