Association of Quality-of-Life Outcomes in Cancer Drug Trials With Survival Outcomes and Drug Class.

IMPORTANCE

Although quality of life (QOL) is an important clinical end point, cancer drugs are often approved based on overall survival (OS) or putative surrogate end points such as progression-free survival (PFS) without QOL data.

OBJECTIVE

To ascertain whether cancer drug trials that show improvement in OS or PFS also improve global QOL of patients with cancer compared with the control treatment, as well as to assess how unchanged or detrimental QOL outcomes are reported in trial publications.

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study included all patients with cancer in the advanced setting who were enrolled into phase 3 randomized clinical trials (RCTs) of cancer drugs reporting QOL data and published in English language in a PubMed-indexed journal in the calendar year 2019. The systematic search of PubMed was conducted in July 2020.

MAIN OUTCOMES AND MEASURES

Association of QOL outcomes with OS and PFS, framing of unchanged QOL outcomes in trial publications, and the association of favorable framing with industry funding of the trials.

RESULTS

A total of 45 phase 3 RCTs enrolling 24 806 participants (13 368 in the experimental arm and 11 438 in the control arm) met the inclusion criteria and were included in the study analyses. Improvement in global QOL with the experimental agent was reported in 11 (24%) RCTs. The RCTs with improved QOL were more likely to also show improved OS vs trials with unimproved QOL (7 of 11 [64%] trials vs 10 of 34 [29%] trials; χ2 = 4.13; P = .04); there was no such association observed for PFS (6 of 11 [55%] trials vs 17 of 34 [50%] trials, χ2 = 0.03; P = .87). Six trials reported worsening QOL, of which 3 (50%) were trials of targeted drugs, and 11 trials reported improvement in QOL, of which 6 (55%) were trials of immunotherapy drugs. Of the 34 trials in which QOL was not improved compared with controls, 16 (47%) reported these results in a positive frame, an observation statistically significantly associated with industry funding (χ2 = 6.35; P = .01).

CONCLUSIONS AND RELEVANCE

In this cohort study, a small proportion of RCTs of cancer drugs showed benefit in global QOL with the experimental agent. These results showed an association between QOL benefit and OS benefit but no such association with PFS benefit. Trials that failed to show improved QOL often reported their QOL outcomes more favorably. Non-immunotherapy-targeted drugs led to worse QOL more often than did cytotoxic agents.