Abstract
Disclosure: I. Tessler: None. N.A. Gecel: None. G. Avior: None. E. Klang: None. E. Alon: None. T. Kolitz: None. Aim: Thyroid cancer prevalence rate continues to increase, being the most common endocrine cancer. During the years, significant progress in methods for diagnosis, risk-stratification and management were developed, with the recent advance of molecular testing. Here we explore the evolution of thyroid cancer management over the past five decades through a text-mining analysis. Materials and methods: We have queried PubMed for all available 'thyroid cancer' related entries published during 1970-2022. The following data were extracted for each entry: year of publication, publishing journal, title, keywords, and abstract text. Search terms belonged to demographics, histology, diagnosis methods, treatment, and follow-up. Annual trends of publications were plotted. The slopes of publication trends were calculated by fitting regression lines to the yearly number of publications. We also retrieved the number of citations per publication and analyzed the most impactful papers on each topic.Results: Our search yielded 95,931 papers published during the study period. Publications regarding diagnosis (39,490) outnumbered all other topics, in which molecular testing has significantly sharpest raise (slope p=34.1±3.3). Most publication focused on females (n=56,198) and differentiated thyroid cancer (n=45,953). Surgery remains the most studied treatment method during the whole study period, with recent raise in studies regarding active-surveillance (slope 14.2±1.3). Studies concerning shared decision-making and patient experience emerged in 2000, with a sharp increase in recent years (slope 3.8±0.3, p < .001).Conclusions: Our study provides an overview of the trends in thyroid cancer research over the past five decades. Publications regarding molecular testing outnumbered all other diagnosis methods, identified as the primary trend of the current era. Publications on active surveillance have rapidly risen in recent years, highlighting this method's role in thyroid cancer. Keywords: Machine learning; artificial intelligence; Thyroid cancer; Molecular-tests Presentation Date: Saturday, June 17, 2023