Patient-Reported Side Effect Bother: Understanding the Value of the Baseline Report

Patient-perceived treatment tolerability can affect patient ability and willingness to remain on therapy. We sought to examine completion rates for a single item of overall side effect bother at baseline and at the first on-treatment assessment, the association between this item with other patient-reported outcomes (PROs) and the odds of early discontinuation due to clinician-assessed adverse events or reasons other than disease progression. Data were from three commercial cancer trials in solid tumours, focusing on the safety population. The GP5 item from the Functional Assessment of Cancer Therapy (FACT) was used for side effect bother. Other PROs included items on specific symptoms, functional impacts and global health status, all drawn from validated measures. Descriptive statistics were used for completion rates, and correlation and logistic regression analyses were used to examine associations. GP5 was dichotomised as 0-1 ('low') versus 2-4 ('high'). Completion rates were at or above 90% at baseline for all items. GP5 completion rates were 5% lower than completion rates for other items (89.8% versus 94.9%) at baseline, but this was not seen after baseline. Among patients with non-missing baseline GP5, 11.8-15.7% of cancer treatment-naïve patients reported high bother, compared with 23.9% of treatment-experienced patients. Patients with high bother at baseline had higher odds of early discontinuation compared with those with low bother, but this was not statistically significant after covariate adjustment. Continued collection of the GP5 item and concomitant work aiming to understand reasons for missingness as well as interpretation is important for evaluating tolerability in cancer trials.