Clinicopathological classification of immune checkpoint inhibitor-associated myocarditis: possible refinement by measuring macrophage abundance

Figures

• 

  Fig. 1. Histology and immunohistochemistry (IHC) of endomyocardial biopsies from patients treated with immune checkpoint inhibitors. Representative areas of each respective biopsy sample that underwent staining from serial sequential sections are shown. The hematoxylin and eosin (H&E)-stained slides did not show definite evidence of myocarditis by Dallas criteria in either A Patient 5 or C Patient 20 but did demonstrate cardiomyocyte necrosis and loss with increased inflammatory infiltrates in B Patient 25. The CD68 stain showed an increased number of interstitial macrophages that overlap with the inflammatory cell infiltration and the cardiomyocyte necrosis perceived in H&E-stain in E Patient 25. CD68 macrophage stain also showed an increased number of interstitial macrophages in F Patient 20 but not in D Patient 5. Insets show magnified views with arrows highlighting representative CD68 stained macrophages. Scale bar = 100 µm. ID = identification

• 

  Fig. 2. Interstitial macrophage cell counts in endomyocardial biopsies (EMB). Plot of CD68 positive macrophages per high powered field (HPF) from EMBs of patients with clinical diagnosis of Possible (red circles), Probable (blue triangles), and Definite (green squares) myocarditis. The dotted line at 50 macrophages/HPF demarcates the CD68+ cell count threshold to consider a sample as immunohistochemistry positive and support reclassification of the initial clinical diagnosis (Dx)

• 

  Fig. 3. Correlation of cardiac biomarkers and macrophage cell counts. A Cardiac troponin I and B N-terminal pro-brain natriuretic peptide (NT-pro BNP) levels plotted against corresponding CD68 positive macrophage cells counts per high powered field (HPF). Data were compared using Spearman correlation coefficient analysis with values of P < 0.05 being considered statistically significant

• 

  Fig. 4. Total immune checkpoint inhibitor (ICI) cycles, time interval, and survival probability between first exposure to clinical diagnosis and death in reclassified patients. A Comparison of total ICI cycles. B Percent of patients with clinical diagnosis for myocarditis following cycle 1 of ICI treatment. C Percent survival probability of patients at 15 months following cycle 1 of ICI treatment. The n are defined as Possible (n = 6), Probable (n = 9), and Definite (n = 11) using the reclassified myocarditis patient cohorts. Open circle symbols represent censored patients in Possible (n = 5), Probable (n = 2), and Definite (n = 5) cohorts. Data were compared using the Mantel-Cox log-rank test with P < 0.05 being considered statistically significant

• 

  Fig. 5. Myocarditis classification with and without supplemental immunohistochemistry (IHC) staining. Heatmaps of standard myocarditis classification criteria and selected variables from patients with immune checkpoint inhibitor-associated myocarditis. A Myocarditis classification without IHC and B potential reclassification with CD68 positive IHC. Each column represents characteristics from one patient (n = 26). Blue boxes represent negative/normal data, red boxes represent positive/abnormal data, and gray boxes represent studies not performed. *For steroids, red boxes represent high dose steroids that are defined as the equivalent of intravenous methylprednisolone of 1 gm or higher, blue boxes represent other steroids which are defined as a less than 1 gm, and gray boxes represent no steroids given. CMR = cardiac magnetic resonance imaging; CRP = C-reactive protein; CV = cardiovascular; ECG = electrocardiogram; LGE = late gadolinium enhancement; NT-proBNP = N-terminal pro-brain natriuretic peptide