Safety of particle radiotherapy in patients with cardiac implantable electronic devices: Review of literature

Particle radiotherapy primarily includes proton-beam therapy (PBT) and carbon-ion therapy (CIT) to target tumors. PBT/CIT can be delivered precisely within the body (Bragg peak) and is, thus better than conventional radiotherapy with photons.1 Generation of secondary neutrons and electromagnetic interference (EMI) during particle therapy creates a risk of cardiac implantable electronic device (CIED) malfunction.2, 3 Particle beam can be targeted to the tumor using passive scattering or pencil beam scanning. Passive scattering is characterized by a broader spread of the proton beam and increased secondary neutron exposure. Given the rising use of particle therapy, clinicians are increasingly faced with the challenge of irradiating patients with CIEDs.4 This review aims to investigate the safety of particle therapy in the presence of CIED. We searched the literature and found 10 in vivo studies (including one case report) on particle therapy in patients with CIEDs (see Supplement for search strategy). Table 1 summarizes the treatment/CIED type, treated location, patient demographics, and the reported CIED malfunctions. Fisher's exact two-tailed p-values were used for all comparisons. Mean: 78 Range: 75−85 F: 1 M: 7 Thorax: 1 Abdomen: 7 PPM: 28 ICD: 14 CNS: 1 Thorax: 25 Abdomen: 2 Pelvis: 14 Passive scatter Pelvis: passive scatter/pencil beam Mean: 80 Range: 72−90 Head/neck: 1 Thorax: 3 Abdomen: 1 Pelvis: 2 Median: 79 Range: 43−89 F: 12 M: 19 PPM: 28 ICD: 3 PBT: 10 CIT: 21+1a Head/neck: 14 CNS: 8 Thorax: 4 Abdomen: 2−4 Pelvis: 2−4 Median: 72 Range 61−94 F: 7 M: 15 PPM: 21 ICD: 1 Head/neck: 3 Thorax: 2 Abdomen: 5 Pelvis: 9 Bone & soft t: 3 Median: 81 Range: 60−97 F: 15 M: 54 PPM: 64 ICD: 5b PBT: 46 CIT: 22 PBT+CIT: 1+1a Head/neck: 4 Abdomen: 27 Thorax: 20 Pelvis: 15 Bone & soft t: 4 PBT: 115−235 CIT: 140−400 Mean: 75 SD: 12 F: 14 M: 37 PPM: 39 ICD: 12 Head/neck: 28 CNS: 6 Thorax: 5 Abdomen: 6 Pelvis: 6 Mean: 73 SD: 11 F: 14 M: 45 PPM: 42 ICD: 17 Head/neck: 16 Thorax: 28 Abdomen: 4 Pelvis: 11 Mean: 77 Range: 77 Mean: 71 Range: 64−78 Case series: 9 Case report:1 N = 294 Weighted meand: 77 Range: 25−97 F: 63 (25%), 189 (75%) Unknown: 42 PPM: 241 (82%) ICD: 53 (18%) PBT: 229 (77%) CIT: 67 (23%) N = 296a Thorax: 89 (30%) Abdomen: 55 (19%) Pelvis: 64 (22%) Other: 88 (30%) Passive scatter: 131 (47%) Pencil beam: 91 (31%) Unclear: 66 (22%) Weighted mean: 262 Range: 140−400 PBT: 2/8 PPM: 2/8 MDT: 0/3 Abbott/SJM: 1/2 BTK: 1/1 Guidant: 0/1 Intermedics: 0/1 PBT: PPM: 2 -Abdomen, Abbott (46) -Thorax, BTK x2, (23/73)e PBT: 4/42 PPM: 2/28 ICD: 2/14 ICD: 0 (1 event of expected ERI, abdomen) PBT: 4 PPM: 2 -Thorax: 2 x1 (16) x1 (4) ICD: 2 -Thorax: 2 x1 (40) x2 (33 & 48) PBT: 2/7 PPM: 2/7 MDT: 1/3 Abbott/SJM: 1/2 Boston: 0/2 PBT: 2 PPM: 2 -Medtronic, thorax x1 (53) -Abbott, abdomen x1 (26) Seidensaal, 2019 PBT: 0/10 CIT: 0/22 PPM: 0/28 ICD: 0/3 CIT: 3/22 PPM: 2/21 ICD: 1/1 CIT: 3 PM: 2 (29%, 31%) ICD: 1 (51%) PBT: 6/47f CIT: 0/23 PPM: 4/64 ICD: 1/4 CRT-D: 1/1 MDT: 1/26 Abbott/SJM: 2/11 Boston: 0/7 BTK: 2/5 Other/unknown: 1/20 PBT: 5 CRT-D: 1 -MDT, abdomen x1, N/R PPM: 4 -Abbott, abdomen x2 (19, 27) x1 (40) -Biotronik, H/N x1 (27) -Unknown, thorax x2 (24, 66) PBT: 1/51 PPM/ICD: N/Rg PBT: 3/59 PPM: 3/42 ICD: 0/17 MDT: 0/28 Abbott/SJM: 0/10 Boston: 0/17 BTK: 3/3 Sorin: 0/1 PBT: 3 PPM: 3 -Biotronik, thorax x3 x8 x1 PBT: 0/1 ICD: 0/1 PBT: 2/4 PPM: 2/4 MDT: 1 /2 (50%) Abbott/SJM: 0/1 (0%) BTK: 1/1 PBT: 1 PPM, Medtronic, pelvis x1 PBT: 1 PPM:1 -Biotronik, pelvis x1 PBT: 20/229 (9%) CIT: 3/67 (4%) PPM: 17/202 (8%) ICD: 5/41 (12%) PPM/ICD: 1/51(2%)g MDT: 3/78 (4%) Abbott/SJM: 4/29 (14%) Boston: 0/29 (0%) BTK: 7/10 (70%) PBT: 2/229 (1%) CIT: 0/67 PPM: 1/ (Pelvis, MDT) ICD: 1 × 3 (Thorax, BTK) 4 (does not incl. 1 expected ERI) PBT: 17/229 (7%) CIT: 0/67 PPM: 14, x26/241 (6%) ICD: 3, x4/53 (6%) Thorax: 10/89 (11%) Abdomen: 5/55 (9%) Pelvis: 1/64 (2%) Others: 1/88 (1%) In total, there were 294 patients [males 189 (64%), females 63 (21%), unknown 42 (14%); average age 77 (range 25−97) years; pacemakers 241 (82%), implantable cardioverter-defibrillators/ICDs 53 (18%, including one cardiac resynchronization therapy-defibrillator/CRT-D)], who received 296 courses of particle therapy treatments [PBT 229 (77%), CIT 67 (23%); thorax 89 (30%), abdomen 55 (19%), pelvis 64 (22%), other 88 (30%)]. Overall, CIED interactions occurred in 20/229 (9%) PBT and 3/67 (4%) CIT treatments (p = .3). EMI oversensing was rare and occurred in one thoracic PBT treatment (three episodes in a Biotronik ICD) and one pelvis PBT treatment (Medtronic pacemaker). An unexpected decline in battery longevity was reported in four treatments [PBT 1 (0.4%), CIT 3 (4%), (p = .04)]. There was no unexpected premature battery depletion that required replacement during/immediately after particle treatment. Factory reset was the most common CIED interaction but occurred exclusively with PBT and not CIT [PBT 17/229 (7%), CIT 0/67, (p = .03); pacemakers 14/241 (6%), ICDs 3/53 (6%); thorax 10/89 (11%), abdomen 5/55 (9%), pelvis 1/64 (2%), others 1/88(1%), (thorax/abdomen vs. pelvis/other p = .0007)]. There were no irreversible non-reprogrammable malfunctions or serious adverse patient events. These results are summarized in Figure 1. Table S1 shows a summary of in-vitro studies. Hashimoto et al. studied ICDs by delivering 40 sessions of PBT with the device placed 0.3 cm lateral and 3 cm distal to the radiation field, with power-on resets in eight (20%) and partial resets in five (12.5%) sessions. In a separate work, Hashimoto et al. studied four pacemakers and four ICDs with both PT and CIT, finding that the soft error rate in PBT is twice as that of CIT per unit of radiation delivered. Bjerre et al. studied pacemakers and ICDs at distances of 0.5 , 5.0, and 10.0 cm from spread-out Bragg peak (SOBP) of proton therapy, with no difference in resets to back-up mode [6/24 (25%), 4/13 (31%), and 4/12 (33%) respectively].5 Between baseline and day 36, there were no statistical differences in decrease in expected battery life [10.8 ± 3.9, 18.3 ± 5.4, and 21.9 ± 5.0 months]. One pacemaker (Boston Scientific Corporation, Watertown, Massachusetts) underwent a non-reprogrammable reset. We found several heterogeneous in vitro and in vivo studies reporting experience with particle therapy in patients with CIEDs. The rate of any CIED interactions in vivo was low (pacemaker 8%, ICD 12%, p = .6) with no serious adverse events reported. Reprogrammable resets were the most common malfunctions, followed by battery depletions, and rarely, oversensing. Although there was one non-reprogrammable reset reported from an in vitro study, no irreversible malfunctions were reported in the in vivo studies. We noted CIED malfunction rates varied based on the type of particle therapy, device manufacturer, and among different centers. Interestingly, Biotronik devices had a higher rate of back-up resets than other manufacturers in both in vitro and in vivo studies. Further, PBT had a higher rate of device resets compared to CIT (7% vs. 0%, p = .03). This may be due to a higher secondary neutron exposure with proton beams compared to carbon ions.6 Particle beams tend to produce secondary particles after encountering matter, including neutrons, gamma rays and secondary charged particles.7 Secondary neutrons can create ions in the semiconductor microchip in the CIED by undergoing nuclear reaction with boron nuclei, potentially leading to single event software errors.8 Studies simulating neutron exposure to CIEDs suggest that this is the primary mechanism for device malfunction.9 While simulation studies indicate that the neutron exposure to CIED is inversely dependent on distance from the irradiated region in pencil beam proton therapy but is independent of the distance in passively scattering therapy, we found an overall decrease in-vivo malfunctions as the distance increases.10 Newer CIEDs with complementary metal-oxide-semiconductor (CMOS) circuits offer smaller size, longer battery life, and greater reliability but make the device more susceptible to radiation-induced malfunctions. The oxide component of CMOS can accumulate positive charge when subjected to ionizing radiation, leading to phantom electrical connections between parts of the circuit.11 This can explain battery loss observed in some patients (more commonly with CIT).12 Unless directly irradiated, the current generation of pacemakers is generally protected from EMI due to shielding and insulative coating around the circuit. CIED function, however, can exhibit various malfunctions due to EMI oversensing, including pacing inhibition, rapid ventricular pacing (in response to atrial oversensing), false tachyarrhythmia detection, and transient reversion to asynchronous pacing.13 Nevertheless, the reported incidence of noise oversensing is less than 1% (2/229). Our study is subject to several limitations. First, the included studies exhibited heterogeneous methodologies, varying reporting styles, and disparate statistical approaches. The small numbers and event rates, and the heterogeneity among the included studies made it not conducive to conduct a statistical meta-analysis. Consequently, our synthesis of data lacks rigorous statistical validation. Additionally, the incorporation of case reports and case series, without well-defined inclusion criteria, introduces a limitation in assessing prevalence or incidence. Further, the inclusion of an abstract, offering limited information and undergoing limited pre-publication evaluation, further contributes to the study's limitations. Multiple guidelines have discussed concerns regarding the use of radiotherapy in patients with CIEDs, latest being the 2022 EHRA consensus statement.14 Citing an in vitro study by Bjerre et al.,5 this statement suggests that proton therapy should be treated as "high-risk" for device malfunction. While particle therapy can potentially affect CIED function in a minority of patients, our review shows that these effects are generally minimally deleterious esp. when treatment is individualized and delivered with appropriate level of monitoring. In most cases, the benefits of particle therapy may outweigh the risks involved. Nevertheless, caution needs to be taken with close monitoring of CIED function to prevent adverse outcomes if device resets, battery depletion, or oversensing occurs, especially in pacing dependent and ICD patients. Our simplified institutional protocol for patients with CIEDs undergoing particle therapy is shown in Figure 2 (detailed version available as Figure S1). There is a need for ongoing surveillance and reporting of effects of particle therapy on CIED function. No funding was received for this manuscript. The authors declare no conflicts of interest. Author disclosures are available in the Supplementary Information. Data used in this study is publicly available in the referenced studies. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.