Background Programmed death-ligand 1 (PD-L1) expression, as assessed by immunohistochemistry (IHC), is used to select patients with non-small cell lung cancer (NSCLC) for anti-programmed cell death protein 1 (PD-1)/PD-L1 therapy. The current study evaluated the feasibility and efficacy of PD-L1 immunostaining and quantitation on direct Papanicolaou-stained cytological smears compared with formalin-fixed paraffin-embedded samples (cytological cell blocks and surgical resection specimens) in NSCLC cases using 2 commercially available assays: the PD-L1 IHC 22C3 pharmDx assay (Agilent Technologies/Dako, Carpinteria, CA, USA) and the Ventana SP263 Assay (Ventana Medical Systems Inc, Tucson, Arizona). Methods PD-L1 immunostaining using either both or one of the assays was tested in 117 sets of paired samples obtained from 62 NSCLC cases. The tumor proportion score was reported in every case following the recommendations of the International Association for the Study of Lung Cancer (IASLC). Results In 57 sets of samples, both PD-L1 assays were used. Due to the availability of samples, only 1 assay was performed in 3 sets of samples and in 2 cases, only cytology smears were used and tested for both assays. A total of 113 sets of paired samples finally were evaluated; 4 cases could not be studied due to intense nonspecific background staining. A significant concordance between the 2 assays on cytological smears was found. Concordance between paired cytological smears and formalin-fixed paraffin-embedded samples was observed in 97.3% of the cases. Conclusions The quantification of PD-L1 expression on direct Papanicolaou-stained cytology smears is feasible and reliable for both PD-L1 assays.