New platform for collecting real-world data on directing clinical trial optimization, especially for rare mutations in lung cancer.

74 Background: In China, lung cancer pts see different hospitals in various places, but medical info is not shared among different medical centers. It is difficult to obtain complete medical info to evaluate the quality of drug treatment & prognosis of lung cancer pts. Methods: We created a pt’s self-submitted data interactive platform to record and store pt’s info such as locations, gender, diagnosis, treatment etc., & to follow up on pt’s real-world medical info in real time. Based these records, a scientific method was used to establish a real medical treatment data set for lung cancer pts to reflect actual medical treatment process. Results: Between 9/13/2018 - 5/30/2019, 5 rare mutation projects (ROS1, RET, EGFR exon20, HER2, EGFR resistance) were initiated. Data cutoff is 6/15/2019. In all 5 projects, we collected by far the largest number of real-world pt’s self-submitted data reported in China. All cases were confirmed by commercial NGS or PCR. In all 5 projects, there were significant differences in regional distributions of patients. The median age for all 5 projects are: 50, 55, 57, 56 and 55, respectively. A total of 95.6% of pts submitted their 1st-line treatment info & 93.8% received 1st-line treatment. Conclusions: Pts self-submitted interactive platform provides valuable insights for trial design to reconsider site selection, & pts’ criteria selection for the success of the trial in rare mutations lung cancer. [Table: see text]