Key points are not available for this paper at this time.
PRO measures are important complementary endpoints in health technology assessments. However, methodological flaws of PRO documentation and analysis in clinical trials have consequences on their interpretation and subsequent benefit claims. German benefit assessments of abemaciclib (studies MONARCH-2, MONARCH-3 and MONARCH plus), sacituzumab govitecan (studies TROPiCS-02 and ASCENT) and trastuzumab deruxtecan (studies DESTINY-Breast02, DESTINY-Breast03 and DESTINY-Breast04) from 2021 to 2023 were used as examples to illustrate the problems of shortened observation periods for PROs (i.e. European Organisation for Research and Treatment of Cancer Quality of Life Questionnaires EORTC QLQ – Cancer 30 and – Breast Cancer 23) in oncological trials. The pharmaceutical companies presented time-to-event analyses of PRO measures (deterioration by 10 points either with or without subsequent improvement). Statistically significant differences for all analyzed compounds were reported for several EORTC QLQ scales. However, PROs were not recorded beyond end of treatment resulting in a systematically shortened median observation period compared with overall survival, ranging from 25% to 78% in the intervention arm and from 18% to 58% in the comparator arm. Furthermore, the median observation period for PROs was markedly shorter in the comparator arm compared to the intervention arm, ranging from 35% to 73%. Conclusions can therefore only be drawn for the time until end of treatment in the shorter observed arm. Also, due to the shorter observation period, deterioration without subsequent improvement was potentially easier to achieve in the control group. Shortened and different observation periods between treatment arms limit interpretability of study results and provide insufficient data on long-term PRO measures. Trial sponsors should collect PRO data ideally until end of the study and regardless of possible treatment discontinuation. This enables to adequately reflect the patients' perspective when starting a treatment, because drug discontinuation and eventual drug switches are part of a treatment strategy.
Ünal et al. (Wed,) studied this question.