Diagnostic Utility of the PD-L1 Immunostaining in Biopsy Specimens of Patients with Biliary Tract Neoplasms

Background　Anti-programmed death 1/programmed death ligand 1 (PD1/PD-L1) antibodies have been successfully used as treatment agents for several solid tumors; however, it is difficult to predict their effectiveness. We evaluated whether biopsy specimens could predict the positive status of PD-L1 in surgically resected tissue.
Methods　Among 91 patients who underwent tissue sampling with endoscopic or liver biopsy before surgery for biliary tract neoplasms in an academic center, 45 (49%) patients were selected for retrospective analysis because the quality and quantity of their biopsy specimens were adequate for histologic evaluation. We performed immunohistochemical staining to investigate the PD-L1 expression in both resected and biopsy specimens. The percentage of the positively stained cells was calculated for subsequent use in the correlation investigation.
Results　The biopsy methods were endoscopic retrograde cholangiopancreatography (ERCP) in 28 cases, percutaneous liver biopsy in 10 cases, and endoscopic ultrasound fine-needle aspiration in 7 cases. Among the 45 patients, when patients with > 10% positive tumor cells in surgically resected tissues were regarded as truly positive PD-L1, the positive and negative concordance rates between surgically resected tissues and biopsy samples were 56% (5/9) and 100% (36/36), respectively. With regard to the use of preoperative biopsy as a diagnostic tool, all (5/5) PD-L1-positive patients had a positive resected specimen. The accuracy of each biopsy method was as follows: ERCP, 89% (25/28); fine-needle aspiration, 86% (6/7); and liver biopsy, 100% (10/10).
Conclusions　Biopsy samples could be a surrogate material for the assessment of the PD-L1 expression with substantial positive and high negative concordance rates.

The version of record of this article, first published in Journal of Gastrointestinal Surgery, is available online at Publisher’s website: http://dx.doi.org/10.1007/s11605-021-05197-6