Inflammation and fibrosis characterize different stages of myocardial remodeling in patients after stereotactic body radiotherapy of ventricular myocardium for recurrent ventricular tachycardia

Highlights

• •Stereotactic body radiation therapy is used in ventricular tachycardia
• •Immune cells accumulate in radioablated focus
• •Apoptotic cells in radioablated focus are phagocytosed by macrophages
• •Fibrotic tissue is a final stage of tissue remodeling after radioablation

Abstract

We performed a histological and immunohistochemical analysis of myocardia from 3 patients who underwent radiosurgery and died for various reasons 3 months to 9 months after radiotherapy. In Case 1 (death 3 months after radiotherapy) we observed a sharp transition between relatively intact and irradiated regions. In the myolytic foci, only scattered cardiomyocytes were left and the area was infiltrated by immune cells. Using immunohistochemistry we detected numerous inflammatory cells including CD68+/CD11c+ macrophages, CD4+ and CD8+ T-lymphocytes and some scattered CD20+ B-lymphocytes. Mast cells were diminished in contrast to viable myocardium. In Case 2 and Case 3 (death 6 and 9 months after radiotherapy, respectively) we found mostly fibrosis, infiltration by adipose tissue and foci of calcification. Inflammatory infiltrates were less pronounced. Our observations are in accordance with animal experimental studies and confirm a progress from myolysis to fibrosis. In addition, we demonstrate a role of pro-inflammatory macrophages in the earlier stages of myocardial remodeling after stereotactic radioablation for ventricular tachycardia.

1. Introduction

Ventricular tachycardias (VTs) in structural heart disease impair significantly quality of life and may increase risk of deterioration of heart failure or even sudden cardiac death [1]. Most of the patients with known heart disease at risk are implanted prophylactically with ICD. However, ICDs do not prevent occurrence of VTs [2]. Antiarrhytmics are used to treat VT, but their efficacy is limited and in some cases these drugs can be also pro-arrhythmogenic. Today, the most efficacious strategy of management of VTs is catheter ablation [3]. This method aims at modification of arrhythmogenic substrate. The success rate in high-volume centers is quite high, although recurrences of any form of VT do occur. In a small proportion of cases (up to 10 %), the critical part of the substrate may not be accessible because of inability to reach epicardial surface or deep location within the wall. Various alternative strategies are being studied such as alcohol ablation or bipolar ablation.

Recently, stereotactic body radiation therapy (SBRT) has been used for management of VTs after previously failed catheter ablation. Despite significant progress in the last 15 years, the details of the mechanism of action that would explain both short-term as well as long-term effects of SBRT are largely unknown [4]. Among 3 possible modes of action are induction of fibrosis in the irradiated myocardium, effects on functional conduction block and effects on gap junctions [5]. Histopathological evaluation of irradiated myocardium is only available from hearts collected either post-mortem or from hearts that were explanted during heart transplantation [6], [7], [8]. In the material obtained in this way the authors describe necrotic changes, fibrosis and disruption of intercellular junctions especially in the myocardium collected 2 weeks after radioablation [8]. In our recent case series covering time points of 3, 6, and 9 months after SBRT we have analyzed post-mortem myocardia histologically and immunohistochemically and we have documented apoptotic process in the radioablated myocardium after 3 months, which was visibly attenuated 6-9 months after SBRT, while the radioablated focus turned into a scar tissue mostly [9]. The apoptosis in the radioablated focus coincided with the presence of inflammatory cells. Also, some preclinical studies have shown inflammatory cell infiltration into the radioablated focus [10,11]. The present study aims at identifying immune cell population in the hearts at 3, 6, and 9 months after SBRT in order to establish a role for inflammation in the myocardial response to SBRT.

2. Methods

2.1. Patients

Three patients with a history of cardiovascular disease, repeated catheter ablation for VT, followed by SBRT therapy, who died 3 months to 9 months after SBRT, were included into the study. Project was approved by Institutional Revison Boards IKEM and FN Ostrava (Decision number 324/2018). They all underwent necropsy and the heart specimens were analyzed as described in our previous study [9]. Two patients (Cases 1 and 3) died in the hospital and they underwent routine hospital autopsy and 1 patient (Case 2) died out of the hospital and he underwent forensic autopsy. With an approval of his relatives this patient was further evaluated in the pathology department of Institute for Clinical and Experimental Medicine.

2.2. Histological and immunohistochemical methods

The material was obtained after gross macroscopical evaluation of the heart during autopsy. Two patients (Case 1 and Case 3) underwent autopsy 1 day after death, 1 patient (Case 2) underwent autopsy 3 days after death. The state of organs in all cases corresponded to the time elapsed after death. Even in Case 2, an excessive autolysis was not reported. Some specific findings during autopsies are presented in Table 2. The material was fixed in 10% neutral buffered formalin, thoroughly sampled and then processed to the paraffin blocks. Sections (7µm in thickness) were obtained from radioablated regions and were stained using standard hematoxylin and eosin method for morphological analysis. Additional sections were used for immunohistochemistry (see below).

$$\begin{array}{ccc}& & \mathit{Dete}\mathit{ction}\mathit{of}\mathit{CD}68+\mathit{cells},\mathit{CD}11c+\mathit{cells},\mathit{CD}20\hfill \\ & & +\mathit{cells}\mathit{and}\mathit{mast}\mathit{cell}\mathit{tryp}\mathit{tase}\hfill \end{array}$$

Table 2.

Patient characteristics

Patient	Case 1	Case 2	Case 3
Age/sex (M-male, F-female)	52/M	57/M	67/M
Primary diagnosis	ventricular tachycardia, non-ischemic cardiomyopathy, AV-block	ventricular tachycardia, ischemic cardiomyopathy	ventricular tachycardia, ischemic cardiomyopathy
Time passed between ablation and death	3 months	6 months	9 months
Time spent in ICU	30 days	Out-of -hospital arrest	5 days
Immediate cause of death	Sepsis/renal failure	Ventricular fibrillation	Severe bleeding
Underlying cause of death	Respiratory infection	Acute anterior myocardial infarction	Oesophagopericardial fistula (complication of SBRT)
Mechanism of death	Multiorgan failure	Cardiogenic shock	Hypovolemic shock
State of organs at autopsy and other findings at autopsy	Cardiac hypertrophy, pulmonary edema, chronic systemic congestion, older splenic infarction, ischemic enteritis and colitis with purulent peritonitis, diabetic nephropathy	Acute anteroseptal IM, aneurysm of the lateral wall, cardiac hypertrophy, congestive changes in other organs, nephrosclerosis	Fluidothorax bilateral (500 dx, 700 sin), dilated and hypertrophic left ventricle, diffuse coronary atherosclerosis, arterial bypasses at LAD and RIVP, oesophagus with ulcer 4 × 3 cm communicating with pericardial cavity with adhesions

The sections were deparaffinized and immersed into preheated antigen-retrieval solution - Tris-chelaton III (pH 8.5). Then, the samples were incubated at 98°C for 10 min and allowed to cool to room temperature. The endogenous peroxidase activity and the non-specific antibody binding sites were blocked with 5% goat or bovine serum in Phosphate Buffered Saline (PBS). The primary antibodies were diluted as mentioned in Table 1 in PBS + 1.5% normal goat serum. Visualization was achieved using LSAB+ Dako REAL Detection System, Peroxidase/DAB+, Rabbit/Mouse or using rabbit anti-goat biotinylated secondary antibody followed by Vectastain ABC kit peroxidase. The nuclei were counterstained by Harris's hematoxylin. Negative controls were used for all experiments.

$$\mathit{Dete}\mathit{ction}\mathit{of}\mathit{CD}4+\mathit{cells}\mathit{and}\mathit{CD}8+\mathit{cells}$$

Table 1.

Characterization of antibodies used in the study

Characterization of antibodies used in the study
Antibody	Abbreviation	Catalog ID/Lot ID	Dilution	Retrieval	Clone/Isotype	Produced by
Monoclonal Mouse Anti-Human CD68	CD68	M0814	1:4000	Tris-chelaton III (pH 8.5)	PG-M1	DakoCytomation, Glostrup, Denmark
Monoclonal Mouse Anti-Human CD11c	CD11c	NCL-L-CD11c-563/ 6058708	1:200	Tris-chelaton III (pH 8.5)	5D11	Novocastra™ Leica Biosystems Newcastle Ltd, UK
Monoclonal Mouse Anti-Human Mast cell Tryptase	MCT	M7052	1:6000	Tris-chelaton III (pH 8.5)	D33 IgG1	DakoCytomation, Glostrup, Denmark
Monoclonal Mouse Anti-Human CD20	CD20	M0755	1:400	Tris-chelaton III (pH 8.5)	L26 IgG2a, kappa	DakoCytomation, Glostrup, Denmark
Monoclonal Rabbit Anti-CD4	CD4		R.T.U	CC1 (Tris buffer based)	SP35	Ventana Medical Systems, Inc.; Tuscon, AZ, USA
Monoclonal Mouse Anti-CD8	CD8		1:50	CC1 (Tris buffer based)	C8/144B	DakoCytomation, Glostrup, Denmark

Immunohistochemical detection was performed using antibodies against CD4 (rabbit monoclonal, SP35, Ventana;), CD8 (mouse monoclonal, C8/144B, DAKO, Denmark) on 4 μm thick sections of paraffin embedded tissues, using the Ventana BenchmarkUltra automated stainer (Ventana Medical Systems, Inc.; Tuscon, AZ, USA) with the OptiView Universal DAB Detection Kit (Table 1). This kit is an indirect, biotin-free system for detecting mouse IgG, mouse IgM and rabbit primary antibodies.

2.2.1. Microscopy

Images were obtained using Leica DMLB microscope equipped with CCD camera and LAS software (Leica Microsystems GmbH, Wetzlar, Germany).

3. Results

3.1. Patient history

Patients included for the study were described previously [9]. The basic patients characteristics, primary diagnosis, underlying cause of death, immediate cause of death and mechanism of death as well as other relevant findings during autopsy are presented in Table 2. Associated cardiovascular and systemic diseases as well as pharmacotherapy are presented in Table 3. Here is a summary of patient histories.

Table 3.

Associated diseases and pharmacotherapy

Patient	Case 1	Case 2	Case 3
Associated cardiovascular diseases	non-ischemic cardiomyopathy, AV-block, hypertension (after SBRT - biventricular mechanical support implant)	coronary artery disease, left ventricular dysfunction, inferolateral aneurysm, history of hypertension, pulmonary embolism, caval filter implant	coronary artery disease (previous MI), coronary artery bypass grafting using gastroepiploic artery, hypertension
Associated systemic diseases	Diabetes type II, (after SBRT - postoperative anuria, respiratory infection and sepsis, multiorgan failure)	Not significant	Not significant
Pharmacotherapy	(in the period after SBRT) – furosemide, cerospiron, betablockers, amiodarone, ACE inhibitors, antidiabetics, insulin, combined antibiotic therapy, sedatives, myorelaxants, catecholamines (noradrenaline, empressin)	ACE inhibitors, Ca blockers, verospiron, furosemide, warfarin, statin	Remestyp, antibiotics, antiulcer therapy (proton pump inhibitors), antiarrhythmics, volume expansion, catecholamines

3.1.1. Case 1

A 52-year-old male with nonischemic cardiomyopathy and AV block was implanted with a CRT-D device (Unify Assura, St Jude Medical/Abbott) in 2016. Catheter ablation was performed across the scar area on the lateral wall. For persistent inducibility of 2 VTs from the upper and lower part of the scar, mapping of the aorta and the left main ostium were performed for future co-registration for SBRT as described recently by our group [9]. For early recurrences of VT (160 bpm), the patient was indicated to SBRT. After 2 months without arrhythmias, the patient was re-admitted for a recurrent electrical storm. For recurrences of VT, the patient was implanted with biventricular support (HeartMate 3, Abbott, and Centri-Mag, Levitronics). The postoperative course was complicated by renal failure that required continuous renal replacement therapy. Despite combined antibiotic therapy, the patient became comatose, developed ischemic lesions in the colon, and hepatic failure. He died 3 months after SBRT due to multiorgan failure.

3.1.2. Case 2

A 57-year-old male was resuscitated for out-of-hospital cardiac arrest in 2017, and implanted with a single-chamber ICD (Autogen, Boston Scientific). Detailed examination revealed coronary artery disease and severe left ventricular dysfunction. Three months later, he underwent endocardial catheter ablation for an electrical storm. In January 2018, endo and epicardial ablation of a substrate on the inferolateral wall was performed for recurrences of VT. The patient was admitted 11 months later for recurrences of VT. SBRT was indicated and completed in December 2018 (25Gy delivered to the PTV of 62.1 cm³). He had few recurrences of VT early after. After 6 months of freedom from any arrhythmia, the patient died suddenly out-of-hospital in July 2019.

3.1.3. Case 3

A 67-year-old patient with a history of prior inferior myocardial infarction and coronary artery bypass surgery in 2000. He was implanted with an ICD (Current + VR, St. Jude Medical), and amiodarone was started. In 2016, the patient was referred for catheter ablation due to recurrences of VT. Despite non-inducibility at the end, the patient had recurrences of 2 faster VTs. In June 2018, SBRT was performed. Two early recurrences of VT were noted (25 and 55 days after) but subsequently without arrhythmias. The patient presented with early esophagitis (18 days after) but all complaints resolved on antiulcer therapy. Six months after SBRT, the patient was admitted for dysphagia and a large ulcer in the terminal part of the esophagus was diagnosed. Despite PEG insertion, the patient was readmitted 3 months later for progression of symptoms. Corrective surgery was contraindicated for high risk. A few days later, severe bleeding occurred and the patient died after transient stabilization due to asystole.

3.2. Morphological findings in the ventricular myocardium after SBRT

3.2.1. Case 1

At the irradiated region, we found a remnant of myocardium almost devoid of viable cardiomyocytes. There was almost complete loss of cardiomyocyte stainability corresponding to myolysis. Sharp transition between irradiated region and viable myocardium was detected (Fig. 1 A and B). Cardiomyocytes in the viable region were hypertrophic. In the myolytic region, usually only a network of ECM components was preserved surrounding empty spaces left by cardiomyocytes. Inflammatory cells (arrowheads), presumably macrophages, can be found in the empty spaces (Fig. 1 C). These cells were occasionally multinucleated (Fig. 1 D) In addition, other types of immune cells were scattered in the remaining interstitium (Fig. 1 B-C).

Fig. 1.

Morphological findings in the irradiated heart after 3 months (Case 1). (A) Image shows a transition between irradiated region (*) and relatively preserved neighboring myocardium. There is almost complete loss of cardiomyocyte stainability corresponding to myolysis. (B) A magnified image from the same region shows a sharp transition between myolytic focus (*) and relatively preserved myocardium. A portion of cardiomyocytes from a region next to the myolytic focus is hypertrophic. In the myolytic focus, we can see empty spaces left by cardiomyocytes surrounded by interstitial ECM. Inflammatory cells can be seen. (C) Inflammatory cells (arrowheads), presumably macrophages, can be found in the empty spaces. Many of them contain pigment, which resembles lipofuscin by color. Only several surviving cardiomyocytes are present. (D) Multinucleated cells can occasionally be detected in empty spaces within the myolytic focus. Scale bar in A=100µm, in B-D=50 µm. Hematoxylin-eosin staining.

3.2.2. Case 2

At the irradiated region, a transition between fibrotic area and viable adjacent myocardium was detected. In the fibrotic area, some cardiomyocytes were scattered together with some connective tissue cells and blood vessels (Fig. 2 A and B).

Fig. 2.

Morphological findings in the irradiated heart after 6 and 9 months respectively. (A, B) Case 2 (A) Image shows a transition (dashed line) between irradiated region and viable adjacent myocardium. Irradiated region displays fibrosis. (B) A higher magnification image of the same region. Fibrosis of irradiated region is visible. Some blood vessels and scattered cardiomyocytes are found close to the sharp transition (dashed line). (C, D) Case 3 (C) Image shows a transition between irradiated and viable region. In addition to fibrosis, infiltration with adipose tissue is seen. (D) At the edge of the irradiated region (dashed line), fibrosis and hypertrophic cardiomyocytes are visible. Scale bar in A, C=100µm, in B, D=50 µm. Hematoxylin-eosin staining.

3.2.3. Case 3

Irradiated region was characterized by the edge of hypertrophic myocardium transiting into a fibrotic scar (Fig. 2 C and D) Infiltration with adipose tissue was also detected (Fig. 2 C).

Since the morphological analysis revealed variable level of inflammatory infiltration by different types of immune cells, we decided to perform an immunohistochemical identification of these infiltrating cells in samples from all 3 cases.

3.3. T-lymphocytes in the ventricular myocardium after SBRT

3.3.1. Case 1

Myolytic region was heavily infiltrated by immune cells. CD4+ T-lymphocytes were abundant in this region as well as CD8+ T-lymphocytes (Fig. 3 A and D). In the surrounding viable myocardium both CD4+ as well as CD8+ T-lymphocytes were relatively scarce. The CD4+ lymphocytes were more frequent then CD8+lymphocytes.

Fig. 3.

Immunohistochemical detection of CD4+ and CD8+ cells in the irradiated ventricular myocardium. (A-C) Images showing a result of immunoperoxidase reaction for CD4 marker. (A) Case 1: CD4+ lymphocytes are found at the edge of the myolytic focus. No CD4+ lymphocytes are present in the adjacent viable myocardium. A dashed line marks the approximate border between the viable myocardium and the myolytic region. (B) Case 2: A group of scattered CD4+cells next to the fibrotic lesion. (C) Case 3: CD4+ cells are localized at the interface between fibrotic lesion and hypertrophic myocardium. (D-F) Images showing a result of immunoperoxidase reaction for CD8 marker. (D) Case 1: CD8+ lymphocytes are found at the edge of the myolytic focus. A dashed line marks the approximate border between the viable myocardium and the myolytic region. (E) Case 2: Two isolated CD8+ cells in the fibrotic lesion. (F) Case 3: Isolated CD8+ cells in the fibrotic lesion adjacent to remnants of the myolytic myocardium. Scale bar in A, B, D, E=50μm, in C and F=100 μm.

3.3.2. Case 2

Both CD4+ as well as CD8+ T-lymphocytes were scattered throughout the myocardium. CD4+ cells were more abundant than CD8+ cells. T-lymphocytes were found in smaller groups at the edge of the fibrotic region (Fig. 3 B and E). Inside the fibrotic region, T-lymphocytes were quite rare.

3.3.3. Case 3

Both CD4+ as well as CD8+ T-lymphocytes were scattered throughout the myocardium. CD4+ cells were more abundant than CD8+ cells. T-lymphocytes were often found at the interface between the fibrotic region and remnants of the myolytic myocardium (Fig. 3 C and F).

3.4. Macrophages in the ventricular myocardium after SBRT

3.4.1. Case 1

CD68+ macrophages were also found throughout the sample. Close to the myolytic region and in the region itself, CD68+ cells had rounded shape and voluminous cytoplasm, which contained numerous vesicles especially in the cells found within the myolytic region, where there was several fold higher density of CD68+ cells in contrast to viable myocardium (Fig. 4 A). The cells characterized by this localization and morphology were also immunoreactive for CD11c marker, which is typical for pro-inflammatory macrophage population. They were much more abundant in the myolytic region in contrast to neighboring viable myocardium (Fig. 4 D).

Fig. 4.

Immunohistochemical detection of CD68+cells and CD11c+ cells in the irradiated ventricular myocardium. (A-C) Images showing a result of immunoperoxidase reaction for CD68 marker. (A) Case 1: Numerous rounded and voluminous CD68+ cells are found usually in the empty spaces left by degenerating cardiomyocytes in the myolytic focus. (B) Case 2: Isolated focus of CD68+ cells in the viable myocardium. (C) Case 3: Scattered CD68+ cells are localized at the interface between fibrotic lesion and hypertrophic myocardium. (D-F) Images showing a result of immunoperoxidase reaction for CD11c marker. (D) Case 1: Numerous CD11c+cells are found at the edge of the myolytic focus. Virtually no CD11c+ cells are visible in the adjacent viable myocardium. A dashed line marks the approximate border between viable myocardium and myolytic region. (E) Case 2: Several scattered CD11c+ cells in the fibrotic lesion. (F) Case 3: Accumulation of CD11c+ cells in the hypervascularized region of viable myocardium. Scale bar in A, B, C, D, F=100μm, in E=50 μm.

3.4.2. Case 2

CD68+ cells were detected throughout the viable myocardium, usually having elongated shape and fine processes. Several isolated foci of clustered CD68+ cells were found within the viable myocardium. (Fig. 4 B). CD11c+ cells were very rare (Fig. 4 E).

3.4.3. Case 3

CD68+ cells were detected throughout the viable myocardium, usually having elongated shape and fine processes. The CD68+cells were also detected in hypervascularized foci or at the interface between fibrotic lesion and hypertrophic myocardium (Fig. 4 C). CD11c+ cells were quite rare. Only occasional clusters of CD11c+ cells were found including 1 in the hypervascularized focus (Fig. 4 F).

3.5. B-lymphocytes and mast cells in the ventricular myocardium after SBRT

3.5.1. Case 1

Mast cells immunoreactive for mast cell-tryptase were found scattered throughout viable myocardium, but they were virtually absent in the myolytic region (Fig. 5 A). CD20+ B-lymphocytes were scarce and were mainly found in the myolytic region (Fig. 5 D).

Fig. 5.

Immunohistochemical detection of mast cells and CD20+ cells in the irradiated ventricular myocardium. (A-C) Images showing a result of immunoperoxidase reaction for mast cell marker – mast cell tryptase. (A) Case 1: Mast cell tryptase-positive mast cells are localized in the viable myocardium at the edge of myolytic lesion marked by a dashed line. (B) Case 2: Numerous scattered mast cell tryptase-positive mast cells in the viable myocardium. (C) Case 3: Scattered mast cell tryptase-positive mast cells at the interface between fibrotic region and viable myocardium. (D-F) Images showing a result of immunoperoxidase reaction for CD20 marker. (D) Case 1: CD20+ cell in the myolytic lesion. (E) Case 2: Isolated CD20+cell in the viable myocardium. (F) Case 3: Isolated CD20+cell in the hypervascularized region of viable myocardium. Scale bar in A-C and F=100μm, in D and E=50 μm.

3.5.2. Case 2

Mast cells were relatively frequent and they were scattered throughout the myocardium (Fig. 5 B). Only few isolated CD20+ cells were detected (Fig. 5 E).

3.5.3. Case 3

Mast cells were scattered throughout the sample often in the perivascular regions or in the fibrotic regions (Fig. 5 C). Some CD20+ B-lymphocytes were found in the hypervascularized focus (Fig. 5 F).

4. Discussion

In this case series study, we focused on inflammatory cell populations in 3 patients who underwent SBRT for VT and died 3 months (Case 1), 6 months (Case 2) and 9 months (Case 3) after this procedure. We could observe an active inflammatory process in Case 1 confined to the myolytic region and coinciding with apoptosis described in our previous paper [9]. In patients, who died later than 6 months after the SBRT, the inflammatory process was largely diminished. This observation coincides with low frequency of apoptotic cells described previously [9]. These findings suggest a possible scenario of myocardial remodeling after an irreversible myocardial injury caused by SBRT, culminating in massive apoptosis of cardiomyocytes in ablated regions followed by their clearance by macrophages with concomitant lymphocytic infiltration. This damaged tissue is than replaced by fibrous scar during the following time period of weeks and months.

The histological effects of SBRT were initially described in animal studies [10], [11], [12], [13], [14]. Most of these studies show fibrosis of the irradiated myocardium which was proportional to the radiation dose. Apoptotic cell death followed by myocardial fibrosis was documented in studies performed in rats [13] and pigs [10]. Inflammatory reactions were also reported in some animal studies [10], [11], [12]. Using Hanford-Sinclair mini swine, Sharma et al. performed non-invasive SBRT to create ablation lesions in the atrial myocardium. In addition to loss of myocardial architecture due to vacuolization of cardiomyocytes, fibrosis and calcification, they observed mild occasional monocytic infiltration [12]. Another group used pigs and high-energy heavy ion beams for myocardial ablation and analyzed the ablated regions histologically and immunohistochemically after 6 months of follow-up [11]. Authors observed siderophages in the hemorrhagic regions within the ablated myocardium. In addition, increased infiltration by CD45+ cells was documented immunohistochemically especially in regions irradiated with higher doses.

Information about the inflammatory cells in the irradiated human myocardium are scarce [6]. Available clinical studies either do not report about inflammation [8,15] or they claim that no inflammatory changes were observed [7]. In our recent paper we detected frequent phagocytosing cells morphologically corresponding to macrophages in the myolytic region in myocardium irradiated 3 months prior histological analysis [9]. In the present study, we confirmed immunohistochemically that these cells were of monocyte-macrophage origin as they display macrophage marker CD68 together with CD11c marker indicating pro-inflammatory activation of these macrophages. In addition, these CD68+ cells were accompanied by CD4+ and CD8+ T-lymphocytes and some B-lymphocytes. All these inflammatory cell populations are reduced in myocardia of patients 6 and 9 months after radioablation. These findings suggest that inflammatory cell infiltration is rather transient stage of tissue response to therapeutic myocardial irradiation, which eventually leads to scar formation [8,9]. The radioablation had most likely destroyed resident cardiac mast cells as virtually none was found in the myolytic region. These cells, possibly representing newly recruited bone marrow derived cells, were more common in the fibrotic regions from hearts 3 and 9 months after irradiation, which corresponds with their role in fibrogenesis [16,17].

Based on available histological data, including the present study, it is apparent that the tissue response after SBRT is of a different nature when compared to more commonly used ablation using radiofrequency current. Morphological studies of myocardium after catheter ablation report mostly necrosis as a mechanism of cardiomyocyte cell death [18,19]. In contrast, under therapeutic setting, SBRT with 25 Gy appears to cause sequence of changes leading to apoptosis of cardiomyocytes with subsequent phagocytosis by macrophages as part of the clearing activity, most likely orchestrated by well-represented T-lymphocytes. Importantly, apoptosis induced by radioablation was also documented previously in animal studies [10,13]. Interestingly, both kinds of treatment culminate in formation of the fibrotic scar tissue replacing the targeted myocardial region. What are the upstream events inducing apoptosis of cardiomyocytes several weeks after irradiation remains to be clarified by further experimental and clinical studies. This hypothesis is in line with our clinical experience about delayed effect of SBRT for VT [20,21].

Some research groups did not demonstrate fibrosis when using experimentally radiation doses less than 30 Gy [22]. This opens the hypotheses that the clinical effect is not necessarily related to the development of fibrosis and even deconstruct fibrosis as the main antiarrhythmic mechanism. A study by Zhang DM, et al. demonstrated in experimentally irradiated murine hearts a persistent supraphysiologic electrical phenotype, mediated by increases in components of natrium channel and Cx43 [23]. They suggested that the effect of SBRT is caused by a reprogramming of cardiac conduction, which is increased and this should explain antiarrhythmic effect of SBRT. In a clinical part of the study, they found increased Na_V1.5 expression in 1 explanted heart from the ENCORE study. In addition, retrospective analysis of the ENCORE study data showed that QRS durations after SBRT were shortened in 13 of 19 patients and lengthened in 5 patients. Another study by Cha MJ, et al. evaluated in rat model histologic, ultrastructural and functional changes during 1 month after irradiation of the whole heart by 20-50 Gy dose [24]. Interestingly, despite relatively high radiation doses they found no evidence of myocyte necrosis or apoptosis. Diffuse vacuolization was correlated with interstitial and subsarcolemmal edema. Intercalated discs between myocytes were widened with diverse patterns. They hypothesized that the damage of intercalated discs and gap junctions alters electrophysiological characteristics and explain early effect of SBRT. Other findings were suggestive of microvascular inflammatory responses to radiation, which may later proceed to fibrosis. On contrary to the above observation of increased conduction after SBRT, by Zhang et al., the acute structural changes resulted in intracardiac conduction velocity delay in ECG findings. Another study of 4 human hearts after SBRT explanted at the time of heart transplantation (12 to 250 days) showed in all specimens areas of subendocardial necrosis surrounded by a rim of fibrosis [8]. Electron microscopy demonstrated features consistent with an acute injury such as irregular, convoluted intercalated disc regions with loss of contractile elements. These latter studies suggest that SBRT may have an antiarrhythmic effect before the onset of fibrosis.

4.1. Study limitations

This is a study presenting 3 clinical cases analyzed using histological methods. It is possible that observed changes could be influenced by a specific combination of different treatment methods and comorbidities of patients included in this study. Nevertheless, with those 3 patients it was possible to cover time period between 3 months to 9 months after the SBRT so that we could track the tissue response in at least several time points. Interestingly, our observations are in line with experimental data.

5. Conclusion

This study presents data showing an inflammatory response to treatment of recurrent VT by SBRT. We revealed a high activity of CD68+/CD11c+ macrophages 3 months after therapy characterized by phagocytosing activity. In the irradiated myolytic region, we also found at least 2 populations of T- lymphocytes (CD4+ and CD8+) and B-lymphocytes. This inflammatory response to SBRT is reduced later after irradiation with the exception of mast cells, which most likely repopulated the fibrotic myocardium in the irradiated region. We conclude that this sequence of events represents a specific tissue response to yet poorly defined earlier effects caused by SBRT in the human myocardial tissue.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The authors declare the following financial interests/personal relationships which may be considered as potential competing interests.

Acknowledgments

The work was supported by a project Cooperatio of the Charles University, by a grant project AZV NU20-02-00244 from the Ministry of Health of the Czech Republic and by funding from the European Union's Horizon 2020 research and innovation program under grant agreement No 945119. The authors would like to thank Ing.Lucie Kosová and Ms.Marcela Blažkeová for immunohistochemical methods.