LAUSR

their eligibility and predict responsiveness for immune-checkpoint inhibitors. Lung cancer is frequently diagnosed on cytology specimens. Although PD-L1 IHC assays are validated on formalin fixed paraffin embedded (FFPE) tissue, evidences are present in literature which showed high concordance of cytology samples with paired biopsies. However liquid based cytology (LBC) processed specimens have not been tested for PD-L1 expression. Method: Immunohistochemistry using the anti-PD-L1 clone SP263 was performed on BD SurePath LBC processed bronchial washing and brushing smears and paired FFPE endobronchial biopsy specimens. The presence of 100 viable tumor cells was considered adequate for PD-L1 testing. Staining was interpreted positive if membranous and/or cytoplasmic protein expression at any intensity greater than background staining was detected in at least 25% of tumor cells. Result: There were 26 patients with 13 adenocarcinomas and 13 squamous cell carcinomas which were diagnosed by bronchial brushings, washings and concurrently obtained endobronchial biopsy. Twelve out of total 26 biopsies showed cytoplasmic and membranous PD-L1 positivity in >25% tumor cells (46%). Corresponding LBC smear showed PDL1 positivity in 9 cases establishing concordance of 88.4%. No negative case was positive on cytology. However, there were following challenges while interpreting PDL1 IHC on LBC processed smears: 1. Thick clusters of cells: LBC can cause rounding up of epithelial cells and create thick clusters of tumor cells. Therefore, it was necessary to see under higher magnification (40x) for proper interpretation of membrane staining. 2. Identification of tumor cells: Although identification of tumor cells was done before PD-L1 staining, after staining they were interpreted only on higher magnification to better differentiate them from alveolar macrophages. 3. PD-L1 staining in macrophages: Five cases showed very strong cytoplasmic positivity in macrophages. Since tumor cells showed cytoplasmic and membranous positivity it was easy to differentiate tumor cell positivity from macrophage positivity. Necrotic background in one case stained strongly with PD-L1 immunocytochemistry. 4. Nuclear positivity of PD-L1 positivity: One case, where corresponding biopsy was positive for PD-L1 IHC, showed aberrant nuclear positivity. Conclusion: PD-L1 IHC can be performed on LBC processed smears. Though there are challenges in interpretation of immunoexpression inherent to the LBC smears, they can be used in situations when histology material is not available. Future studies are needed to determine their ability to predict response to immunotherapy.