Cardiac radiotherapy transiently alters left ventricular electrical properties and induces cardiomyocyte-specific ventricular substrate changes in heart failure

Abstract

Aims

Ongoing clinical trials investigate the therapeutic value of stereotactic cardiac radioablation (cRA) in heart failure patients with ventricular tachycardia. Animal data indicate an effect on local cardiac conduction properties. However, the exact mechanism of cRA in patients remains elusive. Aim of the current study was to investigate in vivo and in vitro myocardial properties in heart failure and ventricular tachycardia upon cRA.

Methods and results

High-density 3D electroanatomic mapping in sinus rhythm was performed in a patient with a left ventricular assist device and repeated ventricular tachycardia episodes upon several catheter-based endocardial radio-frequency ablation attempts. Subsequent to electroanatomic mapping and cRA of the left ventricular septum, two additional high-density electroanatomic maps were obtained at 2- and 4-month post-cRA. Myocardial tissue samples were collected from the left ventricular septum during 4-month post-cRA from the irradiated and borderzone regions. In addition, we performed molecular biology and mitochondrial density measurements of tissue and isolated cardiomyocytes. Local voltage was altered in the irradiated region of the left ventricular septum during follow-up. No change of local voltage was observed in the control (i.e. borderzone) region upon irradiation. Interestingly, local activation time was significantly shortened upon irradiation (2-month post-cRA), a process that was reversible (4-month post-cRA). Molecular biology unveiled an increased expression of voltage-dependent sodium channels in the irradiated region as compared with the borderzone, while Connexin43 and transforming growth factor beta were unchanged (4-month post-cRA). Moreover, mitochondrial density was decreased in the irradiated region as compared with the borderzone.

Conclusion

Our study supports the notion of transiently altered cardiac conduction potentially related to structural and functional cellular changes as an underlying mechanism of cRA in patients with ventricular tachycardia.

Graphical Abstract

Open in new tabDownload slide

Introduction

Stereotactic cardiac radioablation (cRA) has emerged as a novel approach for the treatment of refractory ventricular tachycardia (VT) in selected patients. However, data regarding its efficacy in larger cohorts as well as a solid understanding of the underlying mechanism in human remain elusive.

Ventricular tachycardia ablation through radiation therapy delivers a single-session, highly focused photon radiation (usually 25 Gy) to the arrhythmogenic substrate within the ventricular myocardium.¹ It is suggested as a potentially effective and safe treatment for patients with therapy-refractory VT and has also been associated with favourable outcomes in the setting of electrical storm.^2,3 The technique has been the subject of several recent studies. Smaller trials showed good safety and efficacy profile and the treatment appeared effective especially during the first year after therapy.^2,4–6 Only recently, ‘The Standardized Treatment and Outcome Platform for Stereotactic Therapy Of Re-entrant tachycardia by a Multidisciplinary consortium’ (STOPSTORM) has been established to investigate STereotactic Arrhythmia Radioablation for VT on a multi-centre basis.¹ Mechanistically, ionizing radiation can activate several molecular pathways involved in the pathogenesis of cardiac fibrosis.⁷ However, the proposed mechanism of radiation-induced fibrosis does not explain the rapidity and magnitude with which VT reduction occurs. Only recently, electrophysiological assessment of murine hearts upon cardiac radiotherapy indicated reprogrammed cardiac conduction through increased levels of NaV1.5 and Connexin43. This was in line with shortened QRS durations in the majority of patients who underwent radiotherapy.⁸ However, it remains unclear if these results translate into an actual patient setting and the time dependency of this mechanism is uncertain. We obtained in vivo and in vitro data from a patient with a left ventricular assist device, recurrent VT, and several endocardial radiofrequency ablation attempts following cRA. Our aim was to determine the in vivo effect of cardiac radiotherapy on myocardial tissue molecular biology and conductivity in a time-dependent manner as well as the effectiveness of the treatment.

Methods

Cardiac radioablation procedure

Following informed consent to treatment by cRA, the patient underwent 4D computed tomography for treatment planning performed with a SOMATOM Sensation Open CT scanner (Siemens) as well as a contrast-enhanced electrocardiogram (ECG)-triggered cardiac CT (SOMATOM Definition Flash, Siemens) for co-registration in the treatment planning system (Varian Eclipse version 15.5; Varian Medical Systems Inc. Paolo Alto CA, USA). A pre-specified internal target volume (ITV) for respiratory motion management was enlarged by 4 mm to result in the planning target volume (PTV), which consisted of 104 cm³. The treatment planning procedure and consideration of organs at risk followed the protocol of the RAVENTA study.⁹ Target volume transfer from the electroanatomic mapping (see Supplementary material online, Figure S2) to the treatment planning system was accomplished using the CARDIO-RT software.¹⁰ Dose distribution was calculated using the AAA/Acuros AXB algorithm and cRA was performed on a Varian TrueBeam STX in four non-coplanar volumetric modulated arcs using a 6 mV flattening filter-free photon beams.

In vivo data

The patient received antiarrhythmic therapy as per EP team with a beta-blocker and amiodarone and was switched to mexiletine 12 weeks after cRA. The biopsy procedure was performed under sedation and heparin was administered during all procedures to achieve an acute clotting time of >300 s. Mapping catheters were advanced via femoral access sites through venous trans-septal sheaths into the left ventricle. A 3D electroanatomic high-density mapping was performed using Carto-3 (Biosense Webster, California) and a Pentaray catheter (Biosense Webster, California) at three time points: Before cRA (pre-cRA) as well as 2 and 4 months after cRA (2-month post-cRA and 4-month post-cRA, respectively). The CARTOUNIVU module of Carto-3 (Biosense Webster)¹¹ was used for X-ray-guided left ventricular mapping and regions < 1.5 mV were considered low voltage/scar (i.e. marked in red; Figures 1 and 2). Local activation times were derived from CARTO. Delta local activation time was calculated using local values during pre-cRA and 2- and 4-month post-cRA (46 and 200 and 128 points, respectively) within the control region and the mean local activation time of the approximate region of cRA (51 and 148 and 115 points, respectively). Conduction velocity was calculated using the mean distance between control and cRA regions.¹² Left ventricular biopsies were gathered upon electroanatomic mapping from the septum using an 8.5 French sheath (Vizigo, Biosense Webster) and a cardiac biopsy forceps (H.+H. Maslanka GmbH, Tuttlingen, Germany). Biopsies were immediately used for in vitro experiments and confocal microscopy. For all procedures, the patient provided informed written consent. Data were obtained within the framework of an individualized healing attempt and local ethics committee approval was given (EA2/256/23).

In vitro data

Ventricular septum myocardial biopsies obtained at 4-month post-cRA were stored in calcium-free buffer solution (including 2,3-Butanedione monoxime (BDM)) and single ventricular cardiomyocytes were isolated from tissue using enzymatic digestion.^13–15 In three first step, biopsies were incubated in enzymes for removal of interstitial tissue (Collagenase XI and Protease XXIV for 10 min at 37°C). In the second step, the biopsy was incubated in Collagenase XI to dissociate cardiomyocytes and non-cardiomyocyte tissue. The enzyme reaction was stopped with calcium-free buffer plus 1% BSA (without BDM) after 10 min. Cells were further dissociated mechanically by pipetting and finally placed in 100 µL buffer solution. Calcium was then carefully re-introduced to obtain a final concentration of 0.5 mM. Mitochondrial density was determined upon staining with a mitochondria-specific dye (MitoTracker Orange). 2D confocal image stacks were recorded using a 40/1.3× oil-immersion objective lens (NA = 1.3) with a Zeiss LSM 800 system. Mitochondria were quantified using a custom-made algorithm.^16,17

Western blots

The tissue of ventricular septum myocardial biopsies was homogenized in 60 µL homogenization buffer (1% NP40, 10% Glycerol, 137 mM NaCl, 20 mM Tris HCl pH 7.4, 20 mM NaF, 1 mM Sodium Pyrophosphate, 10 mM EDTA pH 8, 10 mM EGTA pH 7.0, aprotinin 10 M, leupeptin 10 M, pepstatin A, 0.5 mg/mL, and PMSF 0.2 M) and centrifuged in two steps (4000 and 13 000 rpm). The protein concentration was measured by BCA assay and 30 µg protein was loaded per sample (plus LI-COR orange loading buffer and XT reducing agent, Bio-Red) on a 4–12% Criterion XT Bis-Tris acrylamide gel (Bio-Rad). Proteins were separated dependending on their size for 2 h at 100 V in XT MOPS buffer and blotted on nitrocellulose membrane (0.45 µm) in Towbin buffer plus 20% methanol overnight at 80 mA and 4°C. Total protein was detected with Revert700 Total Protein Stain (LI-COR) according to the manufacturer. After blocking in 5% milk powder in TBS-T for 1–2 h, membranes were incubated with primary antibodies overnight to detect the amount of protein of interest [NaV1.5, Thermo Fisher 23016–1-AP, 1:1000: Connexin43, Sigma C6219, 1:2000; transforming growth factor beta (TGF-β), Santa Cruz sc-65378, 1:200]. Incubation with secondary antibody (anti-rabbit 800 and anti-mouse 680; LI-COR 1:10 000) was followed by detection with the Odyssey Detection System. Protein signal was normalized to the total protein stain and calculated relative to the non-radiated sample.

Chemicals and solutions

Chemicals were obtained from Sigma Aldrich (St. Louis, MO, USA) if not noted otherwise. Fluorescent dye Fluo-4 and MitoTracker Orange were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Calcium-free buffer solution consisted of (in mM): 100 NaCl, 10 KCl, 1.2 KH₂PO₄, 5 MgSO₄, 5 MOPS, and 50 Taurin 2 Glucose (for storage and digestion 30 BDM) and was pH adjusted to 7.4 with NaOH.

Statistical analysis

All data are presented as mean ± standard error mean if not indicated otherwise. Analysis was performed in a blinded fashion. GraphPad Prism was used for statistical inference and plotting (GraphPad Software, San Diego, CA, USA). To test for group differences, for data with two groups, student’s t-test or Kruskal–Wallis one-way analysis of variance on ranks in non-normal distributed data was used. A P < 0.05 indicates a significant statistical difference.

Results

A male 69-year-old patient with ischaemic cardiomyopathy presented with recurrent VT upon several endocardial left ventricular ablation attempts. His left ventricular ejection fraction was severely impaired due to ischaemic cardiomyopathy and a left ventricular assist device was implanted 6 years ago (04/2016). Amiodarone and beta-blockers were given throughout the course of the disease as medical antiarrhythmic therapy. The patient was treated with cRA therapy in the context of the RAVENTA trial with a single dose of 25 Gy to a target volume in the left ventricular septum (Figure 2A).

The patient remained in stable condition without any acute procedure-related side effects for the next 8 weeks. During follow-up exams, the patient presented with symptoms of cardiac decompensation and additional episodes of slow VT documented through the internal data storage of the internal cardioverter defibrillator (ICD) (2 and 4 months). We performed repeated invasive 3D electroanatomic mapping and local catheter-based ablation procedures in relation to these episodes; i.e. in addition to obtaining high-density electroanatomic maps, radiofrequency ablation was performed at both 2-month post-cRA and 4-month post-cRA. Ultimately, left ventricular biopsies were gathered at 4-month post-cRA to determine the effect of cardiac radiotherapy in vitro (Figure 3). The patient remained in stable condition with no new antiarrhythmic episodes during follow-up exams for another 4 months after the final ablation. The most recent routine follow-up including an ICD interrogation 7 months upon cRA showed no record of ventricular tachyarrhythmia episodes or ATP/shock delivery since 4-month post-cRA. However, unrelated to the initial cRA procedure, the patient died 5 months after the final radiofrequency ablation related to a left ventricular assist device drive-line infection and a bleeding complication.

In vivo and in vitro findings

Interestingly, cardiac radiotherapy shortened the QRS duration (see Supplementary material online, Figure S1) from 146 ms (baseline) to 131 ms (2-month post-cRA) and 133 ms (4-month post-cRA) in this patient. 3D electroanatomic mapping during 2-month post-cRA upon radiotherapy unveiled unchanged local bipolar voltage in the control area at the junction between healthy and scar tissue. However, local bipolar voltage and unipolar voltage were significantly increased in the irradiated region of the left ventricular septum during the 2-month post-cRA. This was paralleled by a significant reduction of our measure of cardiac conductivity (i.e. delta local activation time) as compared with our baseline measurements before radiotherapy (Figure 1). During the 4-month post-cRA time point, these changes were somewhat mitigated as bipolar voltage but not unipolar voltage was significantly increased as compared with baseline. In addition, cardiac conductivity was significantly prolonged and showed no difference as compared with baseline. Absolute mean local activation time (in relation to the automated annotation reference from the 12-lead ECG within CARTO) for the respective control and cRA regions were 67 ± 4 and 114 ± 5 (pre-cRA), 20 ± 1 and 54 ± 1 (2-month post-cRA) as well as 36 ± 3 and 78 ± 3 (5-month post-cRA). In line with this, the conduction velocity between control and cRA regions was 75, 108, and 85 cm/s. In the next step, we obtained myocardial septal biopsies from control and irradiated regions as indicated in Figure 2B and C during 4-month post-cRA. While the biopsies within the target region (mean dose 25 Gy) were taken from an area irradiated with a maximum dose of 29.7 Gy, the mean dose in the border zone area (i.e. control region) was 2.9 Gy with a maximum dose of 9.2 Gy (Figure 2B, bottom). We then performed molecular biology and isolated cardiomyocyte experiments: as shown in Figure 4, NaV1.5 protein levels were altered in irradiated regions, while Connexin43 and active TGF-β were unchanged. Moreover and in line with the overall notion of altered cardiac electrical properties, mitochondrial density was significantly altered upon radiotherapy (Figure 5).

Discussion

Our study supports the notion of specific (sub)cellular changes previously only seen with pharmacological antiarrhythmic therapy as an underlying mechanism of cardiac radiotherapy in patients with VT.

Only recently, Zhang et al.⁸ found evidence for radiation-induced reprogramming of cardiac conduction. The authors also hypothesized that radiation-induced effects on cardiac conduction may persist over time. Conduction velocity, levels of the cardiac sodium channel (NaV1.5), and levels of Connexin43 remained persistently elevated in an animal model upon radiotherapy. Moreover, post-radiotherapy QRS duration was shortened in a set of patients. These results are in line with findings in human induced pluripotent stem cell (iPSC)-derived cardiomyocytes, which showed altered conduction velocities following high-dose irradiation. Especially in iPSC-derived cardiomyocytes irradiated with higher doses (20 or 25 Gy single dose) conduction velocity increased over time.¹⁸ We observed a comparable behaviour in our patient, i.e. QRS duration as a measure of the time for overall ventricular excitation shortened. However, QRS duration calculation was difficult due to the extensive left ventricle dyssynchrony and scar present (Supplementary material online, Figure S2)¹⁹ and these results warrant further validation in a larger cohort. Delta local activation time as a measure of cardiac conductivity significantly shortened 2-month post-cRA, yet it showed a relative prolongation 4-month post-cRA indicating non-persistent effects of cardiac radiotherapy on cardiac conduction in the present case. Even though our results support the concept of early altered cardiac conductivity upon radiotherapy, we, therefore, observed a marked reversibility of this process already 120 days after the application of 25 Gy. Our patient received radiofrequency ablation upon mapping at 2-month post-cRA. However, this region was not in close vicinity to the designated control or irradiation area (see Supplementary material online, Figure S3) and therefore likely without effect on conduction times and voltage during the subsequent mapping at 4-month post-cRA. Additional radiofrequency ablation was performed at 4-month post-cRA. This was done upon biopsy and electroanatomic mapping and therefore without effect on in vitro or in vivo measurements.

Molecular biology showed altered NaV1.5 levels when comparing irradiated vs. control regions at 4-month post-cRA. In addition, we observed a reduced mitochondrial density and mitochondrial Ca²⁺ uptake has been shown to hamper excitation–contraction coupling.^20,21 Irradiation has been shown to impair mitochondrial respiration, which in turn might also affect subcellular Ca²⁺ propagation.²² Radiation has also been associated with a decreased number of mitochondria and membrane fusion.²³ The reduction of mitochondrial density might aid in subcellular Ca²⁺ propagation and might represent a novel mechanism of cardiac radiotherapy. Of course, it has to be noted that the results presented here display findings from a single patient with only very limited sample numbers and therefore have to be assessed with caution. Bipolar voltage shown in Figures 1 and 2 is partly also determined by the degree of tissue contact, which might be variable between different time points. Even though mapping was performed carefully, this is a major limitation for the interpretation of the local voltage results. Moreover, since tissue-related changes observed might have already been present before cRA, the fact that biopsies were only obtained during the second mapping session upon cRA represents another major limitation of this study.

Overall, even though the patient died of non-treatment-related causes 5 months after the final mapping and ablation, combined treatment with RF ablation and cRA was successful (i.e. no recurrence in the device interrogation) in a short-term follow-up of 3 months upon final ablation and mapping.

Cardiac radiotherapy significantly altered in vivo electrophysiological properties in a time-dependent fashion in our patient with heart failure, VT, and left ventricular assist device. This was in line with an increased expression of voltage-dependent sodium channels and altered mitochondrial density in irradiated areas in vitro as compared with control regions. Reduced mitochondrial density might be a novel underlying mechanism for altered cardiac conductivity observed in these patients.

Supplementary material

Supplementary material is available at Europace online.

Funding

This study was supported by the German Research Foundation (grant number HO 5647/4-1 and SFB1470 B01/A01 to F.H.), the German Centre for Cardiovascular Research (to F.H.) and the European Union Horizon 2020 Framework (Grant Agreement 945119 STOPSTORM to F.M.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References

Author notes