Comprehensive immunohistochemical analysis of PD-L1 shows scarce expression in castration-resistant prostate cancer

Christian D. Fankhauser, Peter J. Schüffler, Silke Gillessen, Aurelius Omlin, Niels J. Rupp, Jan H. Rueschoff, Thomas Hermanns, Cédric Poyet, Tullio Sulser, Holger Moch,

Oncotarget · 2017 · 57 citations · 35 references

DOIFull text

Open access

Concepts

Abstract

// Christian D. Fankhauser 1 , Peter J. Schüffler 2, 3 , Silke Gillessen 4 , Aurelius Omlin 4 , Niels J. Rupp 5 , Jan H. Rueschoff 5 , Thomas Hermanns 1 , Cedric Poyet 1 , Tullio Sulser 1 , Holger Moch 5 and Peter J. Wild 5 1 Department of Urology, University Hospital Zurich, University of Zurich, Zurich, Switzerland 2 Department of Computer Science, ETH Zurich, Zurich, Switzerland 3 Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA 4 Department of Medical Oncology and Hematology, Cantonal Hospital, St. Gallen, Switzerland 5 Department of Pathology and Molecular Pathology, University Hospital Zurich, University of Zurich, Zurich, Switzerland Correspondence to: Peter J. Wild, email: peter.wild@usz.ch Keywords: prostate cancer, PD-L1, immunotherapy, immune response Received: September 30, 2016      Accepted: November 15, 2017      Published: December 04, 2017 ABSTRACT Background: We aimed to analyze the frequency and distribution of PD-L1 expression in specimens from prostate cancer (PC) patients using two different anti-PD-L1 antibodies (E1L3N, SP263). Materials and Methods: PD-L1 immunohistochemistry was performed in a tissue microarray consisting of 82 castration-resistant prostate cancer (CRPC) specimens, 70 benign prostate hyperplasia (BPH) specimens, 96 localized PC cases, and 3 PC cell lines, using two different antibodies (clones E1L3N, and SP263). Staining images for CD4, CD8, PD-L1, and PanCK of a single PD-L1 positive case were compared, using a newly developed dot-wise correlation method for digital images to objectively test for co-expression. Results: Depending on the antibody used, in tumor cells (TC) only five (E1L3N: 6%) and three (SP263: 3.7%) samples were positive. In infiltrating immune cells (IC) 12 (SP263: 14.6%) and 8 (E1L3N: 9.9%) specimens showed PD-L1 expression. Two PC cell lines (PC3, LnCaP) also displayed membranous immunoreactivity. All localized PCs or BPH samples tested were negative. Dot-wise digital correlation of expression patterns revealed a moderate positive correlation between PD-L1 and PanCK expression, whereas both PanCK and PD-L1 showed a weak negative Pearson correlation coefficient between CD4 and CD8. Conclusions: PD-L1 was not expressed in localized PC or BPH, and was only found in a minority of CRPC tumors and infiltrating immune cells. Protein expression maps and systematic dot-wise comparison could be a useful objective way to describe the relationship between immuno- and tumor-related proteins in the future, without the need to develop multiplex staining methods.

References

35