e15175 Background: Given the number of targeted therapies that are approved for non-small lung cancer (NSCLC), timely testing for biomarkers is critical to ensure eligible patients benefit from precision medicines. This study uses real-world data to explore how factors associated with routine and reflex testing in the US are influencing the decision to test for biomarkers. Methods: Ipsos’ MDx Monitor is a multi-stakeholder, physician-reported syndicated patient record database. Cancer-treating physicians in the US (2021 = 56, 2022 = 64, 2023 = 44) were screened for seniority and caseload and submitted data on 414 NSCLC patients (aggregate) who had been tested for biomarkers. Data were collected online in September-October of each year. Results: Comparing sampled physician-cited reasons for testing over time, standard procedure was mentioned more frequently in the reported 2023 cohort (87% in 2023 vs. 62% in 2021) as was lab performs this test as reflex (18% in 2023 vs. 2% in 2021). Testing because of patient histology or to inform treatment decision declined from 2021 to 2023. Reasons that were mentioned consistently more frequently among the reported early-stage patient cohorts vis-à-vis metastatic patients included histology, to inform treatment decision and, notably, patient request. The top cited reasons are shown in Table 1. Conclusions: Results from this study indicate that physicians’ approach to biomarker testing is evolving towards more routine procedures, rather than testing to inform a specific treatment decision, marked by a recorded increase in reflex and broader panel-based testing. Although guidelines focus on assessing metastatic patients’ eligibility for targeted therapies, early-stage patients also stand to benefit from more automated testing processes. Further investigation on time between diagnosis and implementation of biomarker testing, along with subsequent treatment outcomes, is warranted. [Table: see text]