Predictive immunocytochemistry in non-small cell lung cancer cytological samples

Abstract

Non-small cell lung cancer (NSCLC) molecular biomarker testing is obligatory for determining therapy. The aim of this study was to compare immunocytochemistry (ICC) results of NSCLC predictive biomarkers between bronchoscopic and non-bronchoscopic type of cytology samples. This study included archive records of 1109 predictive ICC results (ALK, ROS1, and PD-L1). The ICC was done on bronchoscopic, and non-bronchoscopic NSCLC samples prepared as cytological smears and cytospins, using Dako EnVisionTM FLEX detection visualization system. The ALK, ROS1, and PD-L1 distribution between bronchoscopic and non-bronchoscopic samples was analysed. The future perspective of cytology in precision medicine was reconsidered. The obtained positive results of ALK, ROS1, and PD-L1 ICC were in concordance with the previously observed range. There was no statistically significant difference in ALK, ROS1, and PD-L1 ICC distribution between the bronchoscopic and non-bronchoscopic groups of samples (p=0.730). The comparison of PD-L1 expression, and, separately PD-L1 ≥50% expression, between two groups of samples showed no statistically significant difference (p=0.236; p=0.436). Bronchoscopic and non-bronchoscopic samples prepared as cytological smears and cytospins are a suitable, but underutilized resource for ALK, ROS1, and PD-L1 biomarker analysis. The implementation of optimized predictive immunocytochemistry assays to provide rapid and reliable results for limited tumour samples is necessary.