Improving the care of patients with lung cancer hospitalized with immune checkpoint inhibitor-related pneumonitis: Implementation of a clinical pathway algorithm

e18730 Background: Immune checkpoint inhibitor related pneumonitis (ICI-P) is one of the most frequent reasons for discontinuing immunotherapy. High mortality rates have been reported among those with grades ≥3 ICI-P. Early recognition and timely management are key to better patient outcomes. We conducted a quality improvement project that involved developing and implementing a clinical pathway algorithm to standardize the care of patients admitted to our Onco-Hospital Medicine (OHM) service. Methods: We included patients with lung cancer admitted to the OHM service with suspicion of ICI-P from January 1, 2020 to December 31, 2022. We developed a multi-modal educational intervention targeted to providers of the OHM service to promote the use of the developed pathway. Interventions were rolled out in 2 phases and included educational sessions (phase 1), distribution of flashcards, informational notepads, monthly emailers, and presentation of an animation of the clinical pathway (phase 2). Our primary outcome was the time to first ICI-P-directed treatment, i.e. time to treatment (TTT). We used the Kruskal-Wallis test to compare TTT between the pre-intervention and post-intervention phases. Results: Out of 82 patients admitted with a suspicion of ICI-P, 60 were confirmed to have ICI-P based on multi-disciplinary consensus. 19 patients were included in the pre-intervention group, 30 in post-intervention phase 1, and 11 in post-intervention phase 2. 92% of patients in our cohort had NSCLC, and 95% had stage 4 disease. Pembrolizumab was the most common ICI used (55%), followed by combination Ipilimumab+Nivolumab (15%). Pulmonology was consulted in 98% of patients. 98% of patients were treated with steroids. Median TTT(range) was 1 (0-12) day in the pre-intervention group, and 2 (0-17) days and 1 (0-6) day after post-intervention phases 1 and 2, respectively (p = 0.3). Utilization of immunotoxicity order set increased from 0% during pre-intervention to 20% and 27% after phase 1 and phase 2, respectively (p = 0.03). Prescription of Pneumocystis jirovecii pneumonia (PJP) prophylaxis among patients discharged on high-dose steroids increased from baseline but was not statistically significant (pre-intervention = 71%, phase 1 = 95%, phase 2 = 100%, p = 0.09). There was no significant change in ICU admissions or inpatient mortality from pre-intervention to post-intervention phases 1 and 2 (p = 0.5 and 0.6, respectively). Conclusions: We showed a decrease in time to intervention from baseline to post-intervention phase 2 after our educational interventions. While this change was not statistically significant, our multi-level approach was able to help increase the utilization of immunotoxicity tools on the OHM service. Continued education on the OHM service is needed to promote adherence to the developed pathway and help improve patient outcomes.