(87) Use of Immunosuppressive Drugs (ISD) and The Effectiveness of Immune Checkpoint Inhibitor (ICI) Therapy Among Patients with Advanced/Metastatic Non-Small Cell Lung Cancer (NSCLC) in the United States (US)

Publication Number: 761

Background: The use of corticosteroids (CS) and other ISD may alter the effectiveness of ICIs among patients with advanced/metastatic NSCLC.

Objectives: This study aims to examine the impact of CS and ISD use and occurrence of mild or moderate immune-related adverse events (irAEs) on the treatment outcomes of ICI therapy among patients with advanced/metastatic NSCLC.

Methods: This was a retrospective cohort study using US electronic medical record data from the ConcertAI Patient360 NSCLC dataset. A sample of 400 patients who received ICIs between 1/1/2015 and 10/31/2019 after advanced/metastatic NSCLC diagnosis were selected using stratified random sampling. The initiation of ICI therapy was set as the index date. Mild/moderate AEs were identified based on documented grade or severity. The treatment outcomes were assessed using Kaplan-Meier analysis by baseline CS/ISD use (within 180 days prior to index date) and using 3-month landmark analysis by irAE status for duration of therapy (DOT), time to next treatment (TTNT), progression-free survival (PFS), and overall survival (OS). OS was also analyzed using Cox proportional hazards (PH) regression analysis.

Results: Among 400 study patients, 131 (32.8%) had no baseline CS/ISD use (Cohort 1) and 269 (67.2%) had baseline CS/ISD use (Cohort 2). Overall, the mean (standard deviation) age was 67.1 (10.0) years with the majority being White (71.0%) and male (51.3%). About half of the study sample had an ECOG score < 2 (49.3%). Most patients (75.0%) received their index ICI treatment in combination with chemotherapy. Over half of patients were Stage IV at initial diagnosis (61.8%). The most common irAEs during ICI therapy were fatigue (55.3%) and nausea (28.0%). Cohort 1 was generally older than Cohort 2 (mean age 69.4 ± 10.1 vs. 66.1 ± 9.8 years). The proportion of patients with brain metastasis in Cohort 2 (42.0% vs. 20.6% in Cohort 1) and the proportion of patients with pleural lung effusion metastasis in Cohort 1 (14.5% vs. 6.7% in Cohort 2) was higher than their counterparts.
Cohort 1 had longer median DOT (9.6 [6.9, 12.5] vs. 5.9 [5.3, 6.7] months), TTNT (8.8 [6.9, 12.5] vs. 5.0 [4.2, 5.9] months), PFS (8.6 [5.5, 17.0] vs. 4.5 [3.5, 5.9] months), and OS (17.4 [12.6, 23.9] vs. 9.2 [5.9, 11.7] months) than Cohort 2. Adjusted Cox PH analysis of OS showed baseline CS/ISD use was associated with higher risk of death when compared to no baseline exposure (HR 1.38 [1.02, 1.86]). There were no differences observed in treatment outcomes by occurrence of irAEs by 3 months.

Conclusions: This study presents preliminary results indicating that use of CS/ISD before the initiation of ICI therapy, although common, is associated with decreased clinical effectiveness of ICI therapy among patients with advanced/metastatic NSCLC diagnosis in the US.