Analysis of characteristics and trends of randomized controlled trials of gastric cancer between 2000 and 2019 / 中华外科杂志

ABSTRACT

Objectives: To analyze the current development status of gastric cancer (GC) randomized controlled trials (RCT) between 2000 and 2019, and to review the basic characteristics of published RCT. Methods: ClinicalTrials.gov was searched for phase 3 or 4 RCT conducted between January 2000 and December 2019 with the keyword "gastric cancer", and the development trend of different types of RCT during different time periods was described. Basic features of registered RCT such as intervention, study area, single-center or multicenter, sample size, and funding were presented. PubMed and Scopus databases were searched to judge the publication status of studies completed until June 2016. The adequacy of the report was estimated by the Consolidated Standards of Reporting Trials (CONSORT) checklist. Design flaws were evaluated by Cochrane tool and/or whether a systematic literature review was cited. The data was analyzed by χ2 test or Fisher exact test. Results: There were 262 RCT including in the present study. The number of GC-RCT registered on ClinicalTrials.gov had been on the rise from 1 case in 2000 to 30 cases in 2015. The proportion of RCT associated with targeted therapy or immunotherapy increased from 0 during 2000-2004 to 37.1% (36/97) during2015-2019. The RCT registered in Asia was 191 cases, while that in non-Asia region was 71 cases. The proportion of multi-center RCT from non-Asia was higher than that from Asia (70.4% (50/71) vs. 50.3% (96/191), χ²=8.527, P=0.003). The proportion of RCT published was 59.1% (81/137). Among the published RCT, 65 (80.2%) studies were reported adequately, but 63 (77.8%) studies had avoidable design limitations. Conclusions: Targeted therapy and immunotherapy have become research hotspots in the treatment of GC. At present, there are inadequate multicenter RCT in Asia, and the publication rate of RCT is low. A considerable number of published RCT are reported inadequately and have avoidable design flaws.