Health‐related quality of life in oncology drug reimbursement submissions in Canada: A review of submissions to the pan‐Canadian Oncology Drug Review

Abstract

Background In Canada, the Canadian Agency for Drugs and Technologies in Health (CADTH) evaluates and makes recommendations for the reimbursement of cancer drugs. One component of its recommendation is based on an economic evaluation, which typically takes the form of a cost‐utility analysis. A cost‐utility analysis measures the effects of competing therapies with quality‐adjusted life‐years (QALYs). The data for this calculation typically come from generic, preference‐based measures of health‐related quality of life (HRQOL). The objective of this review is to determine the frequency at which HRQOL data are collected alongside cancer drug trials and used in the cost‐utility analysis submitted to the CADTH pan‐Canadian Oncology Drug Review (pCODR). Methods Submissions between 2015 and 2018 to pCODR, the group charged with evaluating cancer drug submissions at CADTH, were reviewed. All pCODR submissions, either in progress or completed, were publicly available online. The search was restricted to completed evaluations. Results Forty‐three submissions met the inclusion criteria. The incremental gain in QALYs in most submissions from the new technology was small (median incremental gain, 0.86; interquartile range, 0.6‐1.39). More than half of the submissions (56%) did not include original data on HRQOL, with most relying on previous studies of variable relevance and quality. Re‐analyses by pCODR based on concerns over HRQOL data used in the submitted model were common (52%). Conclusions Drug manufacturers do not consistently collect data on HRQOL alongside clinical trials and instead rely on evidence generated in previous studies to inform cost‐utility analyses. These findings should induce manufacturers to collect original HRQOL data that are simultaneously relevant to patients and decision makers.

Comparing Manufacturer Submitted and Pan-Canadian Oncology Drug Review Reanalysed Incremental Cost-Effectiveness Ratios for Novel Oncology Drugs

“…Manufacturer-submitted models also considered a higher proportion of cancer drugs to be cost-effective based on a variety of WTP thresholds when compared to the EGP reanalyzed estimates. This incongruity between manufacturer-submitted models and the EGP reanalysis was also highlighted by Raymakers et al, who found only 11 of the included 43 (26%) pCODR submissions had manufacturer-submitted ICERs that fell within the EGP calculated range [17]. While this study included a limited sample size, our analysis of 73 submissions further confirms this discrepancy between the manufacturer-submitted and EGP reanalyzed ICERs.…”

“…Again, prior research has similarly documented concerns that manufacturers tend to make more optimistic assumptions in their model inputs with respect to costing and utility estimates [11]. Raymakers et al reported that manufacturers do not consistently collect health-related quality of life data in their clinical trials in order to calculate QALYs, but tend to rely on lower quality evidence from previous studies [17]. Thus, concerns regarding utility estimates or data used to calculate QALYs in the economic models submitted is an ongoing concern.…”

Describing Sources of Uncertainty in Cancer Drug Formulary Priority Setting across Canada

“…Furthermore, Raymakers et al (2021) found that nearly a quarter (24%) of submissions to pCODR were made on the basis of early phase clinical trials (Phase I or II) [16]. Other pCODR analyses have demonstrated that the majority of submissions between 2015 and 2018 lacked any data on quality of life endpoints [12].…”

“…One report per tumor type was included in this analysis to ensure a manageable qualitative sample. This type of inclusion and exclusion criteria has been applied in prior studies that have analyzed pCODR submission documents [ 12 ]. Any inconsistencies in the pERC documents were resolved with further analysis of the Final Clinical Guidance Report.…”

Cost-effectiveness of brentuximab vedotin in advanced stage Hodgkin’s lymphoma: a probabilistic analysis

“…Utility values, for example, used to calculate QALYs were not original data collected alongside the ECHELON-1 trial and were taken from a previous study [17]. To the best of our knowledge, the ECHELON-1 trial did not report on HRQoL data (if collected) that could be used for economic analysis, which is common for oncology clinical trials [27]. Similarly, our analysis assumed that treatment costs were based on Canadian list prices and did not include additional associated costs.…”