Real-world side effects of targeted therapies: High-throughput association studies leveraging the CancerLinq Discovery lung cancer database

We developed high-precision phenotyping algorithms to identify non-small cell lung cancer (NSCLC) patients receiving targeted therapies in the CLQD database. We then performed phenome-wide association studies (PheWAS) comparing new diagnosis codes in patients receiving each targeted therapy to new codes in patients receiving chemotherapy or immunotherapy. Codes with significant associations were compared to toxicity data reported in clinical trials and the FDA Adverse Event Reporting System (FAERS) database.

We identified 5,278 NSCLC patients who received targeted therapies with the latest CLQD data pull in 2022. For each of the 18 targeted therapies with five or more patients in the database, descriptive statistics and two PheWAS analyses are reported: one for diagnosis codes relative to chemotherapy, and one relative to immunotherapy. These analyses identified significant associations corresponding to known toxicity profiles as well as potentially underreported side effects.

This high-throughput framework augments the characterization of side effect profiles for existing targeted therapies and can proactively monitor for toxicity signals as novel therapies and treatment indications emerge. The importance of collecting real-world data across institutions is highlighted in the ability to find clinically relevant associations even in targeted therapies directed against rare mutations.