P2.04-08 Preliminary External Validation of the Clinical Definitions of Resistance to Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer

Abstract

Immune checkpoint inhibitors (ICIs) have become an important component of treatment for patients with advanced stage non-small cell lung cancer (NSCLC). Unfortunately, the majority of patients on ICIs will eventually progress. Based on the results from a meta-analysis review of the current ICI literature, Gillaspie et al developed a clinical radiographic response criterion to classify patients as having innate, acquired resistance or durable response to ICIs. The objective of this study was to validate this categorization using prospectively collected clinical data. The study population consisted of Stage IIIB and IV, biopsy-proven NSCLC patients from a single institution treated on or off a clinical trial with single agent ICIs in the first, second line or beyond. De-identified tumor data, stage and treatment data long with response to ICIs were collected prospectively into a database. Patients were censored at last date of follow-up if no progression had occurred. Progression-Free survival (PFS) curves and rates were estimated using the Kaplan-Meier method and then compared to the categorization established by the meta-analysis. From April 2012 to January 2018, 231 patients met criteria for inclusion and analysis. Median age was 65 years, 61% were male and 92% white. Forty-three (18.6%) received ICI in the first line, while the remaining 188 were treated in the second line or above. Analogous to the meta-analysis, our single center data demonstrated three distinct sub-populations of response. PFS curve slopes were evaluated and compared between the proposed classification and our single center experience (Table 1). The slopes of the curves are similar for the Innate, Adaptive or Acquired Resistance and Durable Response categories established in the meta-analysis. Our single-center experience resulted in a slightly steeper slope in the innate response category compared to the clinical trial literature. This may be explained by our cohort including patients who are treated in the third line or beyond whereas most of the meta-analysis trials restricted populations to either first or second line only. A steeper PFS curve would be expected in a more heavily treated population.Table 1PFS Slope Meta-AnalysisPFS Slope ValidationInnate Resistance12.914.0Aquired Resistance3.83.4Durable Response1.21.1 Open table in a new tab This single center assessment of the proposed classification for resistance to immune checkpoint inhibitors in non-small cell lung cancer confirms the three distinct subgroups of response based on a comparison of the PFS curves. A formal statistical validation will be performed and presented. Future studies are planned to include prospectively collected data from additional comprehensive cancer centers within the validation cohort.