Abstract
BACKGROUND: Discovery and incorporation of predictive and prognostic biomarkers enhance outcomes for patients with cancer. Clinico-genomic datasets, which retrospectively link real-world clinical data to tumor sequencing data, are important resources for biomarker research, which has historically relied on robust research infrastructures exclusive to large academic centers. The objective was to evaluate the feasibility of a pragmatic, technology-enabled platform at community-based research sites for development of a prospective clinico-genomic database supported by centralized electronic health record (EHR)-based patient ascertainment and data processing. METHODS: Adults with stage IV or recurrent metastatic non-small cell lung cancer or extensive-stage small-cell lung cancer were enrolled at 23 US sites upon initiating a standard line of therapy. Enrollment rates were estimated from eligible populations at individual centers. Clinical data from routinely collected EHR documentation were centrally processed and normalized for quality control. Serial blood samples at pre-specified timepoints (baseline, during treatment and at disease progression/end of therapy) were used for circulating tumor DNA (ctDNA) genomic profiling. RESULTS: Between December 2019 and May 2021, 944 patients enrolled, representing ≈25 % of eligible patients. Eight-hundred seventeen of 944 (87 %), 406 of 606 (67 %) and 398 of 852 (47 %) participants provided qualifying samples for ctDNA testing at baseline, during treatment and at disease progression/end of therapy, respectively. Samples were provided at all three timepoints by 35 % of participants. CONCLUSION: A community-based oncology patient cohort was rapidly enrolled, creating a real-world clinico-genomic dataset. This pragmatic study platform has potential research applications where prospective real-world data may contribute to evidence generation.