Real-world demographics, clinical characteristics, and testing and treatment patterns in patients with non-small-cell lung cancer and extensive-stage small-cell lung cancer in Belgium: the AIBED study

Background Real-world evidence is essential to understand the implementation of lung cancer treatments. This study describes the characteristics, testing patterns, and treatment of lung cancer patients in Belgium. Methods This multicenter study used routinely collected longitudinal electronic health record data from five Belgian hospitals (2019-2021). Natural language processing was used to process unstructured data, which was combined with structured data to generate an Observational Medical Outcomes Partnership (OMOP) common data model (OMOP-CDM) data warehouse. We generated a lung cancer cohort and subcohorts of special interest: stage IB-IIIB resected non-small-cell lung cancer (NSCLC), unresected stage III NSCLC, (EGFR-mutated) stage IV NSCLC, and extensive-stage small-cell lung cancer (ES-SCLC). Results We included 1952 patients, of whom 87% (1699) had NSCLC and 13% (253) SCLC (86.6% ES-SCLC). Stage IB-IIIB resected NSCLC included 17.3%, unresected stage III NSCLC 12.3%, and stage IV NSCLC 45.0% of NSCLC patients. Brain metastasis around diagnosis was reported in 16.4% of stage IV NSCLC (22.3% in EGFR mutated) and 15.1% of ES-SCLC patients. Across NSCLC subcohorts, 53.9% of patients had a programmed death-ligand 1 (PD-L1)- positive tumor (≥1%) and 12.7% an EGFR mutation-positive tumor. In stage IB-IIIB resected NSCLC patients, resection status was retrieved for 83.0% of patients, of whom 93.0% had complete resection. Among 327 stage III NSCLC patients, 63.9% ( n = 209) were unresected, 36.7% ( n = 120) had a PD-L1-positive tumor, 18.3% ( n = 60) completed chemoradiotherapy, and 11.3% ( n = 37) initiated durvalumab. Conclusions This is the first Belgian study to provide real-world insights into patient trajectories in lung cancer patients leveraging artificial intelligence technologies.