ON4 REVIEW OF THE APPLICATION OF HEALTH-RELATED QUALITY OF LIFE (HRQOL) DATA IN RECENT HEALTH TECHNOLOGY ASSESSMENTS (HTAS) OF CANCER IMMUNOTHERAPIES (IOS) BY THE NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE (NICE)

Abstract

The estimation of quality-adjusted life years (QALYs) for a given intervention necessitates the analysis of HRQOL data. For cancer treatments, this typically comprises an assessment of utility according to progression status; however, recent literature advocates an alternative approach for patients treated with IOs—using the time until a patient’s death (time to death [TTD]). We reviewed NICE assessments of IOs to understand which methods of analyzing HRQOL data have been previously used and accepted. Completed assessments of IOs published prior to December 2018 were identified via the NICE website. Manufacturer submissions and final appraisal determination documents were downloaded and searched. We extracted the preferred method submitted by each company and the corresponding committee discussion, including whether the method was ultimately accepted or rejected. A total of 21 assessments were identified, of which 20 received positive NICE recommendations. In 10 assessments, a TTD-based approach was presented (using the TTD as the only explanatory variable for utility [n=7] or using TTD in combination with progression status or treatment arm [n=3]). The remaining 11 assessments presented a progression-based approach. In 2 cases, NICE committees ultimately rejected the TTD-based method presented by the manufacturer, favoring a progression-based approach. Key committee criticisms across methods included selection bias (for progressed utilities or those near the end of life) and ‘implausibly high’ values for health states far from death (e.g. >360 days) versus comparable general population estimates. Alternative approaches to analyzing HRQOL data in submissions of IOs to NICE have been accepted other than the typical progression-based approach. Because of reporting limitations, it remains to be seen whether the use of alternative approaches for utility data may lead to markedly different cost-effectiveness results. Consequently, further research is required to establish the relative performance of each method and their appropriateness for decision making.