An innovative team approach to improve SCLC care across an integrated health system.

Abstract

8111 Background: Small cell lung cancer (SCLC) is a highly aggressive cancer, with improved overall survival due to recent advancements in therapy. However, as many as 40% of patients (pts) may not receive guideline-concordant care for SCLC due to barriers including delays in diagnosis and referrals to oncology. We developed and implemented an electronic medical record (EMR) alert system based on ICD codes for early identification of pts with suspected SCLC and early intervention by thoracic medical oncology at University Hospitals Seidman Cancer Center. We hypothesized our intervention would increase rates of guideline concordant care and ultimately improve pt outcomes, compared to historical control. Methods: An informatics approach was used to identify pts with possible SCLC from 5/5/2024-5/17/2025 by incorporating Structure Query Language, Server Reporting Services, and text search of the EMR and pathology systems. Daily reports were reviewed by the thoracic medical oncology team, who intervened when needed to ensure appropriate staging and referrals. Clinicodemographic data was collected of pts with confirmed SCLC compared to a historical control of patients diagnosed between 2/5/2019-11/29/2021. Chi-square tests assessed statistical differences between the groups, and Kaplan-Meier (KM) survival analysis estimated overall survival (OS) and progression free survival (PFS). Results: 101 pts with SCLC were identified in the current group (C), and 130 pts in the historical group (H). Age, sex, smoking history, and baseline ECOG status, were similar between groups. A numerically higher percentage of pts in C were diagnosed with limited-stage SCLC (38.6%) compared to H (27.7%), p = 0.106). A numerically higher number of pts in C received guideline-concordant baseline imaging (93.1%) compared to H (83.8%, p = 0.05). Among pts who received treatment (tx), median time to tx was longer in C (19 days, interquartile range [IQR] 9-31) than H (13 days, IQR 5-22, p = 0.001). Pts in C were significantly more likely to receive 1st-line guideline-concordant tx than H (94.0% vs 83.3%, p = 0.02), although similar likelihood of enrolling in hospice before any therapy (12.9% vs. 11.5%, p = 0.917). Pts in C with progression were numerically more likely to receive 2 nd -line and 3 rd -line tx than H (61.3% vs. 57.1%; 78% vs. 53.1%, respectively). Pts in C had statistically significantly improved PFS compared to H (p = 0.008). Conclusions: Successful creation of an EMR alert system to identify SCLC significantly increased rates of guideline-concordant tx of SCLC. Although not statistically significant, pts were also more likely to receive guideline-concordant imaging and treatment in the 2 nd and 3 rd line settings compared to historical control.