Data sharing in the era of precision medicine: a scientometric analysis

Background: Multiple precision medicine programs in oncology have been launched, leading to the collection of large amount of clinical and genomic data. Tumor heterogeneity and the accumulation of rare and of unknown significance genomic alterations require to study thousands of individuals to identify clinically relevant genomic drivers. Better the scale is or will be, better our understanding of a disease is or would be. In this context, data sharing appears as a precondition of the success of precision medicine in oncology. The work we present here attempts to describe the current stage of data sharing in precision medicine with a focus on oncology. Methods: A scientometric study of the publications indexed in the Web of Science (WoS) database was conducted by applying quantitative methods. A search string was defined by selecting relevant keywords, and specific metrics such as the research area, publication year, funding organization, and geographical localization were studied. A third-party software (VOSViewer) was used for analyzing and visualizing bibliometric networks. Results: A set of 672 documents were obtained between 1900 and 2019, year 2005 was a turning point, and the trend reached 86–113 publications per year over the last three years. Western Europe and Northern America accounted for 80% of the whole world production. From the 672 publications, diverse research areas were identified (i.e., computer science and medical informatics), as well as specific medical specialties (i.e., medical genetics and oncology). The term co-occurrences map identified the main challenges associated with data sharing. Conclusions: This area of research is relatively new with an unequal quantitative production of scientific literature across countries and institutions. The presence of non-medical scientific disciplines such as computer science was not that surprising as data sharing had to face major technical challenges. The results of term occurrences reflected the main parameters that govern data sharing in precision medicine but also its obstacles. Our study provided a picture of an emerging and interdisciplinary field that could be of interest to all stakeholders facing common challenges to promote data sharing in precision medicine.