C31-10 Analysis of Patient Reported Outcomes and Home Spirometry Ability to Detect Pneumonitis

Abstract Rationale Pneumonitis is an immune-related adverse event from checkpoint inhibitors that increases morbidity and mortality. Pneumonitis occurs in about 33% of patients receiving concurrent chemotherapy/radiation followed by immunotherapy (CRT-ICI). We investigated whether patient-reported outcomes (PROs) such as the MD Anderson Symptom Inventory (MDASI) or home spirometry (HS) forced vital capacity (FVC) could identify pneumonitis to facilitate clinical evaluation and treatment. Methods Patients with newly-diagnosed locally advanced non-small cell lung cancer (NSCLC) undergoing CRT-ICI were enrolled into a prospective observational study. Participants were followed for 62 weeks from CRT-ICI initiation but were taken off study if immunotherapy was stopped. Participants completed the “your shortness of breath at its worst” (MDASI-SOB) and “your coughing at its worst” (MDASI-cough) components of MDASI weekly for 10 weeks and then every 2 weeks until study completion. Missed MDASI timepoints were imputed using a “last value carried forward” approach. Participants performed HS twice weekly for 3 weeks and then weekly until study completion. A 1-point increase in response to either MDASI-SOB or MDASI-cough, or a 10% or greater decrease in FVC from baseline were considered significant. Grade 2 or higher pneumonitis was considered a significant event. A “true positive” result was if a significant change in MDASI or HS was detected from two weeks prior to onset of pneumonitis to one week after onset of pneumonitis. Results 84 patients were enrolled. Patients were excluded if they never received immunotherapy, withdrew consent within 2 months of enrollment, or provided less than 1 month of data. 59 patients were analyzed for MDASI and 26 developed pneumonitis. MDASI detected 18 and had a sensitivity of 0.69 and specificity of 0.75. Pneumonitis was detected by MDASI-SOB alone in 3 instances, MDASI-cough alone in 5 instances, and by both in 10 instances. MDASI detected pneumonitis a median of 3.5 days earlier than clinical recognition. 47 patients were analyzed for HS and 23 developed pneumonitis. 9 events did not have corresponding HS data. HS detected 9 of the remaining 14 with a sensitivity of 0.64 and specificity of 0.84. The median time from immunotherapy initiation to pneumonitis onset was 57 days. Conclusion Monitoring for pneumonitis with PROs such as MDASI is feasible and has modest sensitivity. HS performs similarly but the real world sensitivity may be hindered by non-adherence and logistical challenges. Further longitudinal analyses are planned for accurate positive and negative predictive values for HS and PROs. This abstract is funded by: None