S Valmary-Degano1, P Colpart2, L Villeneuve3, F Monnien2, L M'Hamdi4, G Lang Averous5, M Capovilla6, F Bibeau7, M-H Laverriere8, V Verriele-Beurrier9, H Ben Rejeb10, P Dartigues11, J Hommell-Fontaine12, F-N Gilly13, S Isaac12, E Mery4. 1. Department of Pathology, Besançon University Hospital, 3 Boulevard Fleming, F-25030, Besançon, France; University of Bourgogne Franche-Comté, F-25000, Besançon, France. Electronic address: svalmary@univ-fcomte.fr. 2. Department of Pathology, Besançon University Hospital, 3 Boulevard Fleming, F-25030, Besançon, France. 3. Pôle Information Médicale Evaluation Recherche, Unité de Recherche Clinique, Hospices Civils de Lyon, F-69000, Lyon, France. 4. Department of Pathology, Claudius Regaud Institute, IUTC Oncopôle, F-31100, Toulouse, France. 5. Department of Pathology, Hautepierre University Hospital, F-67000, Strasbourg, France. 6. Department of Pathology, Baclesse Institute, F-14000, Caen, France. 7. Department of Pathology, Caen University Hospital, F-14000, Caen, France. 8. Department of Pathology, Grenoble University Hospital, F-38000, Grenoble, France. 9. Department of Pathology, Paul Papin Cancer Center, F-49000, Angers, France. 10. Department of Pathology, Bergonie Institute, F-33000, Bordeaux, France. 11. Department of Pathology, Gustave Roussy Institute, F-94000, Villejuif, France. 12. Department of Pathology, Lyon-Sud University Hospital, F-69310, Pierre-Bénite, France. 13. Department of Digestive Surgery, Lyon-Sud University Hospital, F-69000, Lyon, France.
Abstract
BACKGROUND: Epithelioid peritoneal malignant mesothelioma (EPMM) is the most common subtype of this aggressive tumor. We compared two antibodies against PD-L1, a recent theranostic biomarker, and evaluated the prognostic value of PD-L1 expression by mesothelial and immune cells in EPMM. METHODS: Immunohistochemistry was performed on 45 EPMM. Clinical and pathological data were extracted from the RENAPE database. Using E1L3N and SP142 clones, inter-observer agreement, PD-L1 expression by mesothelial and immune cells and inter-antibody agreement were evaluated. The prognostic relevance of PD-L1 expression was evaluated in 39 EPMM by univariate and multivariate analysis of overall survival (OS) and progression-free survival (PFS). RESULTS: Inter-observer agreement on E1L3N immunostaining was moderate for mesothelial and immune cells, and fair for mesothelial and poor for immune cells using SP142. Using E1L3N, 31.1% of mesothelial and 15.6% of immune cells expressed PD-L1, and 22.2% of mesothelial and 26.7% of immune cells using SP142. Inter-antibody agreement was moderate. In most positive cases, 1-5% of tumor cells were positive. Using E1L3N, PD-L1 expression by lymphocytes was associated with better OS and PFS by both univariate and multivariate analysis. Cytoreductive surgery with hyperthermic intraperitoneal chemotherapy predicted better prognosis than other treatments. Solid subtype was an independent prognostic factor for worse OS. CONCLUSION: E1L3N appeared easier to use than SP142 to evaluate PD-L1 expression. A minority of EPMM expressed PD-L1, and only a few cells were positive. PD-L1 expression by immune cells evaluated with E1L3N was an independent prognostic factor in EPMM.
BACKGROUND: Epithelioid peritoneal malignant mesothelioma (EPMM) is the most common subtype of this aggressive tumor. We compared two antibodies against PD-L1, a recent theranostic biomarker, and evaluated the prognostic value of PD-L1 expression by mesothelial and immune cells in EPMM. METHODS: Immunohistochemistry was performed on 45 EPMM. Clinical and pathological data were extracted from the RENAPE database. Using E1L3N and SP142 clones, inter-observer agreement, PD-L1 expression by mesothelial and immune cells and inter-antibody agreement were evaluated. The prognostic relevance of PD-L1 expression was evaluated in 39 EPMM by univariate and multivariate analysis of overall survival (OS) and progression-free survival (PFS). RESULTS: Inter-observer agreement on E1L3N immunostaining was moderate for mesothelial and immune cells, and fair for mesothelial and poor for immune cells using SP142. Using E1L3N, 31.1% of mesothelial and 15.6% of immune cells expressed PD-L1, and 22.2% of mesothelial and 26.7% of immune cells using SP142. Inter-antibody agreement was moderate. In most positive cases, 1-5% of tumor cells were positive. Using E1L3N, PD-L1 expression by lymphocytes was associated with better OS and PFS by both univariate and multivariate analysis. Cytoreductive surgery with hyperthermic intraperitoneal chemotherapy predicted better prognosis than other treatments. Solid subtype was an independent prognostic factor for worse OS. CONCLUSION: E1L3N appeared easier to use than SP142 to evaluate PD-L1 expression. A minority of EPMM expressed PD-L1, and only a few cells were positive. PD-L1 expression by immune cells evaluated with E1L3N was an independent prognostic factor in EPMM.