Efficiency of cytology samples for PD-L1 evaluation and comparison with tissue samples

Abstract

Objective: lung cancer is the leading cause of cancer-related death. PD-L1 blockers have become a first-line option for advanced non-small cell lung cancer (NSCLC) patients. Guidelines require the assessment of PD-L1 expression by immunohistochemistry. Although tissue samples are widely used, cytologic samples could be an alternative. In this study, we compared cytologic samples with tissue samples for PD-L1 evaluation in NSCLC cases. Material and method: Koc University Hospital, Department of Pathology Laboratory Information System was scanned for all PD-L1 tests performed on NSCLC cases, either on tissue samples or cell blocks. The type of the biopsy/aspiration procedure, the tumor type, patient demographics, and the percentage of PD-L1 positive tumor cells were recorded. A total of 73 tissue samples and 49 cell blocks were found to be eligible for the study. Results: the PD-L1 positivity score was at least 1% in 44 of 73 samples of the tissue group and 19 of 49 samples of the cell block group. Tissue samples showed significantly higher positivity compared to the cell blocks (p=0.020). Comparing the frequency of cases with 50% positivity showed no statistically significant difference. A comparison of PD-L1 positivity rates of only the small biopsies and cell blocks also showed no significant difference. Conclusion: although they harbor a limited number of tumor cells, cell blocks prepared from cytologic samples are good alternatives for PD-L1 testing. However, large resections should be used for PD-L1 evaluation whenever possible since even 1% positivity may affect the treatment decision.