Association between severe adverse event management and overall survival in patients treated with immune checkpoint inhibitor with advanced non–small-cell lung cancer.

Abstract

6604 Background: The impact of severe adverse event (sAE) management (mgmt) on clinical outcomes in cancer patients (pts) receiving immune checkpoint inhibitor (ICI) therapy has not been fully examined. We aimed to evaluate the association between mgmt of sAEs and overall survival (OS) in pts receiving ICIs for advanced non-small cell lung cancer (aNSCLC). Methods: Data were drawn from the ConcertAI Patient360 dataset, a curated EMR-based US oncology database representing 19,000+ NSCLC patients. aNSCLC pts who initiated ICI in the first 3 lines of anti-cancer therapy from 1/1/15 - 8/31/21 with no record of clinical trial enrollment or diagnosis of a second primary cancer were included. sAEs (dermatologic, gastrointestinal, endocrine, hepatic, respiratory, rheumatic, renal, and cardiac) and mgmt actions were curated from unstructured data from ICI initiation through the earliest of: 100 days after ICI discontinuation, start of a non-ICI regimen, end of study period, or death. Cox regression was used to evaluate the association between sAE mgmt actions (as time-varying covariate) and OS adjusting for baseline characteristics including line of therapy. The earliest sAE and mgmt action per patient were examined in OS analysis. Results: 3,211 pts were identified (median [IQR] age 67 [13] years, 55.1% male). Median [Q1-Q3] duration of follow-up from ICI initiation was 8.2 [3.3-17.5] months. Most pts had advanced disease at initial diagnosis (82.0%) and started ICI in first line (61.6%). The most common ICI regimens in the first 3 lines were nivolumab (29.5%) and pembrolizumab + carboplatin + pemetrexed (29.1%). 8.6% of pts had at least one sAE, most often diarrhea (3.5%). Median [Q1-Q3] time from ICI initiation to first sAE was 84 [34-211] days. Mgmt actions any time after first sAE included anti-cancer treatment interruptions [dose reduction (4.0%), hold (20.6%), discontinuation (2.2%)], sAE treatment [corticosteroids (71.8%), immunosuppressive drugs (2.5%)], hospital admission [hospitalization (57.4%), emergency department visit (24.6%)], and other/unknown (4.3%). Overall, median [95% CI] OS was 13.6 [12.6-14.6] months. Compared with pts with no sAEs, pts with sAEs whose earliest mgmt action was hospital admission (N=155) or sAE treatment (N=71) had a higher risk of all-cause mortality (adjusted HR [95% CI] 1.61 [1.38-1.88] and 1.53 [1.22-1.91], respectively). Pts with sAEs first managed with anti-cancer treatment interruptions (N=39) had similar risk of all-cause mortality (0.91 [0.66-1.25]) compared with pts with no sAEs. Conclusions: Pts with sAEs first managed with hospital admission or sAE treatment had shorter OS than those with no sAEs. No difference was observed for sAEs first managed with anti-cancer treatment interruptions. Findings suggest early treatment interruption for sAEs does not impact OS.