Patient-reported outcomes are under-utilised in evaluating supportive therapies in paediatric oncology – A systematic review of clinical trial registries

Background: Children with cancer suffer from numerous symptoms and side-effects, making supportive interventions indispensable to improve their quality of life. The gold standard for evaluating the latter is patient-reported outcome (PRO) assessment. This systematic review investigates the current practice of clinical outcome assessment (COA) in clinical trials on supportive interventions. Methods: ClinicalTrials.gov and EudraCT were searched for trials including children and adolescents (≤21 years) with cancer receiving supportive care registered 2007–2020. The use of different types of COAs was analysed, focusing on PRO assessment and the domains measured with PRO measures (PROMs). Associations with trial characteristics were investigated using univariate and multivariable analyses. Results: Of 4789 identified trials, 229 were included. Among them, 44.1 % relied on PROMs, the most commonly used COA. The proportion of trials using PROMs did not significantly differ over time. In the multivariable analysis, intervention type (higher PROM use in behavioural vs. medical interventional trials) and cancer type (higher PROM use in mixed and solid tumour samples vs. haematological samples) were significant predictors of PROM use. The majority of trials using PROMs (59.6 %) measured more than one health domain. ‘Physical health’ was the most frequently assessed domain (92.6 %). Conclusion: Less than half of registered clinical trials investigating supportive interventions for children with cancer used PROMs. This result is striking since supportive care explicitly focuses on patients’ quality of life, which is best assessed using PROMs. Our systematic review underlines the need to identify barriers for PROM implementation and to improve PRO research in paediatric oncology.