Myocardial PD-L1 Expression in Patients With Ischemic and Non-ischemic Heart Failure

Abstract

Objective: Immune checkpoints inhibitors are promising and wide-spread agents in anti-cancer therapy. However, despite their efficacy, these agents could cause cardiotoxicity, a rare but life-threatening event. In addition, there are still no well-described predictive factors for the development of immune-related adverse events and information on high risk groups. According to known experimental studies we hypothesized that cardiovascular diseases may increase myocardial PD-L1 expression, which could be an extra target for Checkpoint inhibitors and a potential basis for complications development. Methods: We studied patterns of myocardial PD-L1 expression in non-cancer-related cardiovascular diseases, particularly ischemic heart disease (n = 12) and dilated cardiomyopathy (n = 7), compared to patients without known cardiovascular diseases (n = 10) using mouse monoclonal anti-PD-L1 antibody (clone 22C3, 1:50, Dako). Correlation between immunohistochemical data and echocardiographic parameters was assessed. Statistical analyses were performed using R Statistical Software-R studio version 1.3.1093. Results: In the myocardium of cardiac patients, we found membranous, cytoplasmic, and endothelial expression of PD-L1 compared to control group. In samples from patients with a history of myocardial infarction, PD-L1 membrane and endothelial expression was more prominent and frequent, and cytoplasmic and intercalated discs staining was more localized. In contrast, samples from patients with dilated cardiomyopathy displayed very faint endothelial staining, negative membrane staining, and more diffuse PD-L1 expression in the cytoplasm and intercalated discs. In samples from the non-cardiac patients, no convincing PD-L1 expression was observed. Moreover, we discovered a significant negative correlation between PD-L1 expression level and left ventricular ejection fraction and a positive correlation between PD-L1 expression level and left ventricular end-diastolic volume. Conclusions: The present findings lay the groundwork for future experimental and clinical studies of the role of the PD-1/PD-L1 pathway in cardiovascular diseases. Further studies are required to find patients at potentially high risk of cardiovascular adverse events associated with immune checkpoint inhibitors therapy.

Keywords: PD-L1; cardio-oncology; cardiotoxicity; checkpoint; dilated cardiomyopathy; ischemic heart disease; myocardial infarction.

Figure 1

Histological and immunohistological examination of the myocardium samples from the patient with ischemic heart disease (A) hematoxylin and eosin staining; (B,C) PD-L1 staining with CMPS 5% and 20% respectively. Blue arrows indicate positive endothelial PD-L1 expression, black arrows indicate positive membrane PD-L1 expression; from the patient with dilated cardiomyopathy (D) hematoxylin and eosin staining; (E,F) PD-L1 expression in cytoplasm and intercalated discs, without endothelial, perivascular, and membrane patterns with ICDPS 70% and 100% respectively; from the control without ICI treatment (G) hematoxylin and eosin staining; (H) absence of PD-L1 expression; (I) extremely poor cytoplasmic PD-L1 expression.

Figure 2

Immunohistological examination of the myocardium samples from patients treated with ICI. With pre-existing CVD—left column (A) PD-L1 expression in ICD with CMPS 50%, ICDPS 40%; (B) CD-3 infiltration 7 cells per mm²; (C) CD-68 infiltration 142 cells per mm²; Without pre-existing CVD—right column (D) absence of PD-L1 expression; (E) CD-3 infiltration 17 cells per mm²; (F) CD-68 infiltration 92 cells per mm².

Figure 3

Correlations between PD-L1 expression (CMPS, ICDPS) and echocardiographic parameters (LVEF and LVEDV). (A) Significant moderate negative correlation between CMPS and LVEF; (B) significant moderate positive correlation between CMPS and LVEDV; (C) significant moderate negative correlation between ICDPS and LVEF (D) significant moderate positive correlation between ICDPS and LVEDV.