How quickly are novel cancer therapies adopted after FDA approval and incorporation into EHR-embedded clinical decision support tools? A description of ordering patterns between users and non-users of an EHR-embedded oncology CDS tool.

508 Background: Oncology clinical decision support (CDS) tools provide up-to-date access to changes in treatment guidelines and notification of newly approved therapies to keep providers informed of rapid changes in the oncology environment. Our goal is to describe ordering patterns of oncologists who use an EHR-embedded CDS tool and those who do not after FDA approval for a new regimen combination and subsequent availability in the CDS tool. Methods: Flatiron Assist is a customizable CDS tool embedded in the electronic health record (EHR) to facilitate selection and documentation of National Comprehensive Cancer Network (NCCN) guideline concordant cancer care. This tool is updated on a monthly basis whereby updates to the NCCN guidelines are incorporated with subsequent communication delivered to users. For this retrospective description, we compared time in days to the first regimen order placed at 13 community oncology practices (348 MDs) using the CDS tool to 30 community oncology practices (294 MDs) not using the CDS tool. We focused on regimens ordered for the combination of nivolumab + platinum doublet (FDA approved on 3/4/22, incorporated into the NCCN guidelines on 3/16/22 and released in the CDS tool on 4/12/22) for neoadjuvant therapy in non-small cell lung cancer (NSCLC). Time to first order placed in practices that did and did not use the CDS tool was described as mean days with 95% confidence intervals (CI) and compared using linear regression with adjustment for number of MDs within a practice and robust standard errors from 4/12/2022 to 5/19/2023 which included 211 unique orders. Results: Of the 211 orders assessed during the observation window, the mean time to first order was 150 days (95% CI 118-182) for CDS tool users and 209 days (95% CI 177-242) for non-CDS tool users, favoring CDS tool users by 59 days (P=0.009, Table 1). Conclusions: Users of an EHR-embedded CDS tool ordered a new regimen sooner after FDA approval than non-users of the CDS tool. Limitations include the narrow sample size and evaluation of only one FDA approval. Future research will investigate whether the same benefit is seen for label expansions compared to net new regimen approvals as well as the impacts on outcomes and indicators of improved quality of care.[Table: see text]