Abstract 655: Development of IHC staining protocols for assessment of PD-L1 expression in cytological samples

Abstract

Abstract Immunohistochemistry (IHC) staining of non-small cell lung cancer (NSCLC) samples for programmed cell death ligand 1 (PD-L1) can help identify patients that may benefit from anti-PDL1 therapy. However, resection or biopsy samples cannot be obtained for some patients with NSCLC. Fine needle aspiration (FNA) is a less invasive method for obtaining samples in such patients, and it is yet to be determined if PD-L1 can be assessed using cytological samples. In this study we had three objectives 1. Develop staining protocols on the VENTANA BenchMark ULTRA automated staining platform for cytology samples fixed with the most common methods using cell lines as a model. 2, Determine the optimal cytology fixation method for VENTANA PD-L1 (SP142) IHC Assay (PD-L1 (SP142)) staining, and 3. Assess PD-L1 expression in tumor cells (TC) and tumor-infiltrating immune cells (IC) in cytological samples prepared from NSCLC patients. KARPAS 299 cell line was used as a model system for the fixation studies. Liquid-based preparations (LBPs) fixed in PreservCyt, conventional smears (unstained and pap-stained) fixed in 95% ethanol, and paraffin embedded cell blocks (CBs) fixed in 95% ethanol, 10% neutral buffered formalin (NBF), PreservCyt, and SurePath preservative were optimized on the BenchMark ULTRA using PD-L1 (SP142) antibody and OptiView DAB Detection and Amplification Kits. NSCLC FNA CBs fixed in 10% NBF (N=69) and a subset of FNAs with matched resections (N=20) were stained with the optimized protocol. FNAs and resections were assessed for percentage of TC and IC with PD-L1 staining. Staining parameters were optimized on the BenchMark ULTRA for all sample types tested, and they all produced a range of moderate to strong PD-L1 expression. CBs fixed in 95% ethanol and 10% NBF produced the highest percentage of staining, with expression in 90% of cells. CBs fixed in PreservCyt and SurePath preservative had a lower percentage of staining, 40% and 75% respectively. LBPs had 25% cells staining and smears ranged from 20%-60%. NSCLC FNA CBs fixed in 10% NBF produced interpretable results when stained with PD-L1 (SP142). IC staining was seen in 8.7% (6/69) of samples, TC staining was seen in 8.7% (6/69) of samples, both IC and TC staining was seen in 14.5% (10/69) of samples. The subset of FNAs with matched resections showed concordance with PD-L1 IC staining in 60% (12/20) of samples and TC staining in 75% (15/20) of samples. The discordant cases showed that FNAs were negative for IC when the resection was negative and positive for TC in 5% of cases when the resection was negative. The VENTANA PD-L1 (SP142) Assay staining parameters were found to be optimal for staining NSCLC FNA CB fixed in 10% NBF. PD-L1 staining was detected in both TC and IC in FNA samples and concordance of FNA and matched resections was high for both IC and TC. A larger study is necessary to validate the use of FNAs for assessment of PD-L1 expression in a clinical setting. Citation Format: Christine Boyiddle, Mark Ruboyianes, Ed Del Valle, Lukas Bubendorf, Kerstin Trunzer, Judith Pugh, Jennifer Boone. Development of IHC staining protocols for assessment of PD-L1 expression in cytological samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 655. doi:10.1158/1538-7445.AM2017-655