Autore: Marta Ponzano, Department of Health Sciences – Section of Biostatistics, University of Genoa, Genoa, Italy
ponzano.marta@gmail.com
Alessio Signori, University of Genoa Andrea Bellavia, Harvard T.H. Chan School of Public Health Alessio Carbone, University of Genoa Maria Pia Sormani, University of Genoa
Categoria Primaria: Salute dei migranti, richiedenti asilo e rifugiati
Categoria Secondaria: Epidemiologia clinica
Introduction Distinctive differences in multiple sclerosis (MS) have been observed by race and ethnicity (e.g. age of onset, environmental risk factors, disease severity, progression, mortality, therapy response and tolerability) and social determinants of health (SDoH) contribute to these differences. In this context, it is thus imperative that clinical research on MS can be generalized to the target population, including disadvantaged groups based on race/ethnicity and other SDoH. Recent years have seen an increasing awareness on the relevance of diversifying target populations, and based on a regulation effective January 18, 2017, it is mandatory for sponsors and investigators to report in ClinicalTrials.gov trial participants’ race/ethnicity (if collected).
Objective In the context of MS phase 3 clinical trials, we performed a systematic review with the aim to 1) assess how often race or ethnicity was reported in ClinicalTrials.gov, 2) evaluate whether the population was representative enough, 3) make comparisons with the corresponding published material.
Methods We included all the MS phase 3 clinical trials registered on ClinicalTrials.gov between September 2007 and December 2023. When race or ethnicity were reported, we searched for the corresponding published results using PubMed as well as a machine learning-based web tool.
Results Out of the 99 included studies, 56% reported race or ethnicity, of which only 26% pre-requirement and almost all (40/41) after requirement. Studies reporting race or ethnicity contributed to a total of 33,891 participants, mainly enrolled in Eastern Europe (60%). Most were white (93%) and the median percentage of White participants in the studies was 93% (IQR 86-98% across 54 studies), compared to 3% for Black (IQR 1-12% across 54 studies) and 0.2% for Asian (IQR 0-1% across 50 studies). Four trials with White percentages ranging between 94.9% and 99.8% completely omitted race/ethnicity in the corresponding publication and for other 4 trials the percentage of white participants was reported only in two pooled analyses, combining participants of two trials. Additionally, even when information was reported, some differences in terminology between ClinicalTrials.gov and publications were identified and categories with fewer participants were often collapsed.
Conclusions Collecting and accurately reporting race/ethnicity is an important first step to provide transparency to the representativeness of data. However, the observed failure to enroll enough participants from historically disadvantaged groups in clinical trials results in limited generalizability of the results and can further contribute to health inequalities, with potential clinical implications. More efforts are thus necessary at design phase, by addressing SDoH that historically limit the enrollment of underrepresented population.