Dosimetric Comparison of Particle Radiotherapy and Photon VMAT for Left-Sided Breast Cancer Post Breast-Conserving Surgery

Abstract

Introduction

Carbon ion radiotherapy shows potential for left-sided breast cancer, yet comparative dosimetric data remains limited. This study retrospectively compared the dosimetric characteristics of proton and carbon ion radiotherapies with conformal radiation therapy, intensity-modulated radiation therapy, and volumetric-modulated arc therapy (VMAT) for left-sided breast cancer patients after breast-conserving surgery, exploring heavy ion radiotherapy’s potential as promising treatment.

Materials and Methods

Eighteen patients staged I/II were enrolled. Treatment plans were created using three treatment planning systems: Oncentra® for photon beams, RayStation® for proton, and CiPlan® for carbon ion. The primary objective was to deliver 100% of prescribed dose to 95% of the PTV and PTVtb volumes. Dosimetric parameters for the target and organs at risk (OARs), including conformity indices (CI) and homogeneity indices (HI), were compared using the Friedman M-test.

Results

All treatment methods met the target coverage planning objective and the prescribed dose, without any statistically significant differences. Carbon ion therapy (CI: 0.69±0) exhibited improved conformality compared with IMRT (CI: 0.43±0.01, P < 0.01) and CRT (CI: 0.3±0.01, P < 0.01), without significant advantage over VMAT. The carbon ion method (HI: 1.08±0.00) didn’t exhibit any significant improvement in homogeneity compared with other treatments. Carbon ion therapy generally reduced the Dmean or V5 of OARs, particularly providing superior protection than VMAT for organs including contralateral breast (Dmean: 0.19±0.06 Gy (RBE)), heart (Dmean: 1.23±0.35 Gy (RBE), V5: 3.92±4.55 %), and contralateral lung (V5: 0.01±0.00 %). The carbon method also demonstrated significant dosimetric superiority over proton in V₅ of ipsilateral lung (Carbon: 18.89±7.79 % vs proton: 26.29±9.02%, P = 0.011).

Conclusion

The study suggested heavy ion therapies could provide dosimetric advantages over photon and proton therapies in planning conformity and OAR protection. These advantages could bring potential clinical benefits, reduced side effects, and improved therapeutic outcomes.

Keywords

dosimetric analysis left-sided breast cancer heavy ion radiotherapy carbon ion therapy organs at risk

Introduction

Breast cancer, a common malignant tumor, significantly impacts patients’ lives. With advancements in medical technology, post-surgical radiotherapy following breast-conserving surgery has become a crucial treatment approach.¹ Its primary objectives are to eliminate residual cancer cells, lower the risk of recurrence, and preserve breast shape and function.^1,2 Optimizing radiotherapy involves multiple factors, and radiation type plays a critical role in enhancing treatment efficacy. With the evolution of computer and imaging technologies, radiotherapy has progressed from traditional two-dimensional wedge board radiotherapy and three-dimensional conformal radiation therapy (3D-CRT) to more advanced techniques, such as intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT).^3,4 To minimize dose-related toxicity to organs at risk (OARs) and reduce the risk of secondary malignancies, the development of innovative radiotherapy techniques remains essential.⁵

Compared with conventional photon radiotherapy, particle radiotherapy is characterized by the “Bragg peak” and the high linear energy transfer (LET).⁶ While various ions, including helium and argon, can be utilized for radiation, carbon ions are the most frequently recommended and utilized in contemporary heavy ion radiotherapy. The usage of particle therapy for breast cancer has expanded rapidly due to increased accessibility to proton therapy centers worldwide, primarily involving proton radiotherapy (PT) and carbon ion radiotherapy (CIRT). Heavy ion therapy is a subset of particle therapy and shares similarities with proton therapy, although it has been less extensively studied. Studies have demonstrated that, compared with conventional photon-based techniques, particle therapy, including proton and heavy ion radiation, delivers lower radiation doses to non-target tissues while maintaining or even enhancing target coverage.^7-9 As the most significant adverse effects of breast radiotherapy arise from radiation exposure to nearby organs, such as the heart and lungs, this advantage has the potential to improve the therapeutic efficacy of radiotherapy. However, the physical and dosimetric differences between proton therapy and carbon ion therapy remain underexplored. Previous research demonstrated that, compared with all photon-based radiotherapy techniques, proton therapy possesses dosimetric benefits across all stages of breast cancer treatment, even outperforming IMRT.¹⁰ Compared with well-designed photon-based three-dimensional conformal treatment plans, proton therapy can reduce heart irradiation doses by 2-3 times while improving target coverage.^11-13 In contrast, while several physical and biological advantages of carbon ion therapy have been demonstrated, there is still a lack of comprehensive dosimetric analyses and large-scale clinical trials for breast cancer patients.^9,14-16

Currently, deep inspiration breath hold (DIBH) is recognized as the state-of-the-art technique for heart-sparing in left-sided breast cancer radiotherapy. However, a notable proportion of patients cannot tolerate reproducible DIBH due to advanced age, compromised pulmonary function, or psychological distress. For this vulnerable subset, evaluating optimal treatments under free-breathing (FB) conditions is clinically essential. Therefore, this dosimetric analysis aimed to comparatively evaluate carbon ion, proton, and photon therapies under FB conditions following breast-conserving surgery. This approach serves as a rigorous dosimetric ‘stress test’ to assess the intrinsic physical superiority of the heavy ion Bragg peak without the anatomical benefits of DIBH. Ultimately, these findings may provide a basis for developing standardized technical guidelines and advancing carbon ion radiotherapy for early-stage breast cancer.

Methods

Patient Selection and Setup

A total of 18 patients with left-sided breast cancer who underwent breast-conserving surgery at our center between September 2021 and January 2022 were retrospectively included in this study. To simulate the clinical scenario for patients who are not candidates for DIBH, all treatment plans were generated based on free-breathing (FB) simulation CT scans. The specific inclusion criteria for selecting these 18 cases were as follows: (1) pathologically confirmed early-stage breast cancer and were staged as I/II (T1-2N0M0) according to the AJCC 8th edition staging system¹⁷; (2) left-sided tumors (to specifically evaluate the dosimetric impact on the heart and coronary arteries); (3) a history of receiving postoperative radiotherapy following breast-conserving surgery (BCS); (4) availability of complete, high-quality simulation CT imaging acquired in the supine position under free-breathing conditions; and (5) complete availability of original radiotherapy planning and dosimetric data for retrospective extraction and recalculation. This retrospective study was approved by the Institutional Review Board of Peking University Third Hospital Medical Science Research Ethics Committee (No. IRB00006761-M20260215). The study was conducted in accordance with the Declaration of Helsinki of 1975, as revised in 2024. As this was a retrospective analysis, the requirement for informed consent was waived. All patient details have been strictly de-identified such that the identity of any person may not be ascertained in any way. Patients were positioned supine with their left arms raised above their heads and immobilized using an extended wing board with a T-bar handgrip device. CT images were obtained using a Philips Big Bore 16-slice scanner (Philips Medical Systems, Best, The Netherlands) with a 5 mm slice thickness. The scanning range extended from the temporomandibular joints to the first lumbar vertebral body. Patients were instructed to breathe freely during the scan.

Structure Delineation

For each patient, the clinical target volume (CTV) was delineated on planning CT scan by a specialized radiation oncologist following the Radiation Therapy Oncology Group (RTOG) recommendations published in the RTOG breast contouring atlas. The CTV included the excision cavity volume, architectural distortion, lumpectomy scar, seroma, and surgical clips, with a uniform 1 cm expansion in all directions. The planning target volume (PTV) was generated by adding a 5 mm margin around the CTV in all directions. The tumor bed planning target volume (PTVtb) was defined by cropping out regions less than 3 mm from the surface. OARs, including the spinal cord, heart (outlined to the pulmonary trunk, involving the pericardium while excluding major vessels), bilateral lungs, contralateral breast, and skin on the affected side, were carefully contoured. Additionally, cardiac substructures, such as the left atrium, left ventricle, right ventricle, right atrium, left anterior descending coronary artery, and right coronary artery were delineated for dose evaluation. A representative CTV and OAR delineation is illustrated in the transversal views in Figure 1.

Figure 1.

Example of typical delineations of CTV and OAR. Abbreviations: RCA: right coronary artery; LAD: left anterior descending coronary artery; RA: right atrium; RV: right ventricle; LA: left atrium; LV: left ventricle. CTV: clinical target volume.

Dose Prescription and Treatment Planning

A hypofractionated regimen of 42 Gy in 15 fractions was administered to the PTV, with an additional elevation of 11.2 Gy in 4 fractions delivered to the PTVtb. The primary objective of treatment planning was to ensure that 95% of the PTV and PTVtb received 100% of the prescribed dose while limiting the volume receiving more than 110% of the prescribed dose to less than 10%. Dose constraints for OARs were optimized based on the thresholds outlined in the Guideline of Target Delineation and Treatment Planning of Adjuvant Radiotherapy for Breast Cancer and the Guidelines for Radiotherapy of Breast Cancer (Chinese Medical Doctor Association, 2020 edition). Photon beam plans, including CRT, IMRT, and VMAT, were created using Oncentra® TPS 4.3.0.410 software (Elekta, Stockholm, Sweden), with dose calculations performed using the collapsed cone convolution algorithm and a 3.0 mm dose grid. A 6 MV X-ray beam was utilized. Detailed beam arrangements were established for each photon technique. For 3DCRT, a classic tangential technique was employed with a single isocenter, a couch angle of 0°, and a collimator angle of 0°. The beam arrangement consisted of two opposing coplanar tangential fields aligned with the chest wall, utilizing physical or dynamic wedges to improve dose homogeneity within the target volume. For IMRT, a 4-to-6 coplanar tangential field technique was utilized with a single isocenter. The beam directions primarily included bilateral tangential main fields, supplemented by 1 to 2 anterior or posterior fields. All fields were coplanar with a couch angle of 0° and a collimator angle of 0°. The plans were generated using inverse-optimized IMRT with simultaneous optimization of multi-leaf collimator (MLC) leaf positions and dose rates to ensure adequate target coverage while maximally sparing the heart and lungs. For the VMAT plans, a dual-arc technique was employed using a single isocenter. The first arc consisted of a clockwise gantry rotation from 330° to 178°, while the second arc rotated counterclockwise from 178° to 330°, both with a couch angle of 0° and a collimator angle of 0°. The MLC leaf positions, dose rate, and gantry rotation speed were optimized simultaneously. Proton beam planning was conducted using RayStation® TPS. For proton plans, a 3-field pencil beam scanning (PBS) technique was utilized, with the isocenter placed at the geometric center of the breast target. Based on a standard 360° gantry convention, the arrangement included a left tangential field, a second field with a gantry angle selected between 40° and 50°, and a third field between 355° and 340° (with a couch angle of 0°). For proton planning, the energy layer spacing and spot spacing were set to ‘Automatic with scale’ using scaling factors of 0.8 and 0.5, respectively. Based on the treatment planning system’s algorithm, this corresponds to an energy layer spacing of 0.8 times the 80% distal width of the Bragg peak, and a spot spacing of 0.5 times 1.06 of the average projected sigma (σ). The snout clearance, defined as the air gap in the system, was adjusted between 7 and 12 cm depending on specific patient positioning. To account for clinical uncertainties, different optimization strategies were employed based on the capabilities of the respective treatment planning systems. For the proton plans, inverse robust optimization was strictly applied, incorporating a setup uncertainty of 5 mm in all orthogonal directions and a beam range uncertainty of ± 3%. Target coverage was prioritized by setting V_95%≥100% of the prescribed dose as a hard constraint in the objective function. Carbon ion beam planning was performed in ciPlan® TPS (version 1.0, Lanzhou Ion Therapy, Gansu, China). For carbon ion plans, a spot scanning technique with a single isocenter was employed. Due to hardware configurations, a combined-room irradiation approach was utilized: a horizontal beam (gantry 90°, couch 180°) and a vertical beam (gantry 0°, couch 0°). For the carbon ion plans, which currently lack automated inverse robust optimization, a conventional margin-based forward planning approach was utilized. Uncertainties were managed by applying a 5 mm geometric setup margin (forming the PTV) and accounting for a 3% range uncertainty during the beam path and distal/proximal margin design. Beam weights and trajectories were iteratively optimized to ensure PTV coverage while driving down the doses to surrounding OARs.

Plans across the different modalities (photon, proton, and carbon ion) were generated by a team of senior medical physicists, each with over 5 years of experience in radiotherapy planning and being highly proficient with their respective treatment planning systems. The average time spent optimizing each plan was approximately 2 to 3 hours. Regarding the planning objectives for the organs at risk (OARs), a strict ALARA (As Low As Reasonably Achievable) principle was applied. Rather than merely meeting the standard dose constraints, the planners consistently pushed the optimization process to minimize the dose to the heart, lungs, and other OARs as much as physically possible, without compromising the prescribed target coverage and conformity.

To establish radiobiologic equivalency and allow for direct dosimetric comparisons among photon, proton, and carbon ion beams, the doses for particle therapies were expressed as RBE-weighted doses in units of Gy (RBE), strictly following the International Commission on Radiation Units and Measurements (ICRU) guidelines. For proton beam planning, a generic, constant RBE value of 1.1 was applied. For carbon ion therapy, the RBE is variable and depends on the dose and linear energy transfer (LET). In the ciPlan® TPS, the dose-averaged LET is first computed, and the RBE-weighted dose is subsequently derived based on the modified Microdosimetric Kinetic Model (mMKM) combined with the linear-quadratic (LQ) formalism. The human salivary gland (HSG) tumor cell line was utilized as the reference cell line. Specifically, the LQ parameters for the carbon ion RBE calculation used a representative dataset with a fixed β₀ of 0.0615 Gy⁻², while α₀ was approximately 0.117 Gy⁻¹, which was adjusted and calibrated based on specific in vitro cell survival experimental data.¹⁸

The Plan Assessment and Statistics

The following parameters were recorded and compared between IMRT and VMAT plans. For PTV, D_max (2%), D_min (98%), D_mean, V₁₀₅, V₁₁₀, Vtp (PTV volume within the prescribed isodose surface), Vt (PTV volume), D_5% and D_95% were collected. The conformity index (CI) and the dose homogeneity index (HI) were calculated based on the definitions provided below. CI and HI values closer to 1 indicate better conformal coverage.

C I = \frac{V_{t p}^{2}}{V_{t} \times V_{p}}

(Vtp: PTV volume within the prescribed isodose surface, Vt: PTV volume, Vp: Prescription volume in the body.)

H I = \frac{D_{5 %}}{D_{95 %}}

(D_5%: = minimum dose to 5% of the PTV, D_95% = minimum dose to 95% of the PTV)

To evaluate the irradiated dose to OARs, the analysis included the mean dose (D_mean) and V_x (the volume of OAR receiving at least x Gy), depending upon the organs. The collected dosimetric parameters for the left anterior descending artery (LAD) and right coronary artery (RCA) were D_mean and D_max (2%), for the heart were D_mean, V₅, V₁₀ and V₂₀, for the left lung were D_mean, V₅ and V₂₀, for the right lung and breast were D_mean, D_max (2%) and V₅. Examples of typical dose distributions for the different treatment techniques are illustrated in Figure 2. Finally, Friedman M-test was used to compare each parameter. Statistical significance was set at P < 0.05. The reporting of this study conforms to STROBE guidelines.¹⁹

Figure 2.

Dose distributions for the VMAT, IMRT, and carbon ion plan. Abbreviations: VMAT = volumetric modulated arc therapy; IMRT = intensity-modulated radiation therapy

Results

Patient Characteristics

The detailed clinical characteristics of the 18 patients included in this study are summarized in Table 1. All patients presented with early-stage (I/II) left-sided breast cancer. The mean age of the cohort was 47.7 ± 11.5 years, and the mean body mass index (BMI) was 23.0 ± 3.0 kg/m². The majority of the tumors (50.0%) were located in the upper-outer quadrant. The mean planning target volume (PTV) was 735.65 ± 265.22 cc.

Table 1.

Summary of Patient Characteristics

Characteristic	Value
No. of patients	18
Age (years)
Mean ± SD	47.7 ± 11.5
Range	24 – 67
Body Mass Index (kg/m²)
Mean ± SD	23.0 ± 3.0
Clinical stage, n (%)
Stage I (T1N0M0)	8 (44%)
Stage II (T2N0M0)	10 (56%)
Tumor location, n (%)
Upper-outer quadrant	9 (50.0%)
Lower-outer quadrant	1 (5.6%)
12 o’clock position	1 (5.6%)
9 o’clock position	3 (16.7%)
6 o’clock position	2 (11.1%)
Not recorded	2 (11.1%)
PTV Volume (cc)
Mean ± SD	735.65 ± 265.22

Target Dose Comparison

The results of target dose comparison are presented in Table 2 and Figure 3. For the PTV and PTVtb, all methods successfully met the target coverage objectives, and no significant differences were found between them. Regarding hot spots, the proton method significantly outperformed CRT (V107%: 0.53% ± 1.03% vs 51.12% ± 17.75%, P < 0.01) and carbon ion therapy (V107%: 0.53% ± 1.03% vs 45.39% ± 33.60%, P < 0.01) in reducing hot spots inside the PTVtb. However, the carbon ion method performed worse than VMAT (45.39% ± 33.60% vs 6.62% ± 9.44%, P < 0.01). In terms of homogeneity and conformity, the proton method demonstrated better homogeneity compared with CRT (HI: 1.05 ± 0.01 vs 1.11 ± 0.03, P < 0.01) and carbon ion therapy (HI: 1.05 ± 0.01 vs 1.08 ± 0.03, P < 0.01), as well as superior conformality compared with IMRT (CI: 0.59 ± 0.08 vs 0.43 ± 0.12, P < 0.01) and CRT (CI: 0.59 ± 0.08 vs 0.30 ± 0.08, P < 0.01), although it exhibited no significant superiority over VMAT. The carbon ion method demonstrated advantages only in conformality compared with IMRT (CI: 0.69 ± 0.05 vs 0.43 ± 0.12, P < 0.01) and CRT (CI: 0.69 ± 0.05 vs 0.30 ± 0.08, P < 0.01), while performed worse than VMAT in terms of homogeneity (1.08±0.03 vs 1.05±0.02, P < 0.01).

Table 2.

Comparison of Target Dosimetry Among Five Methods (Mean ± SD and P-Value)

	Mean±SD					P-value
	VMAT	IMRT	CRT	Carbon	Proton	ARC vs IMRT	ARC vs CRT	ARC VS carbon	ARC vs proton	IMRT vs CRT	IMRT vs carbon	IMRT vs proton	CRT vs carbon	CRT vs proton	Carbon vs proton
V107%/%	6.62±9.44	21.00±30.88	51.12±17.75	45.39±33.60	0.53±1.03	1.000	0.001	0.001	1.000	0.061	0.061	0.072	1.000	0.000	0.000
V100%/%	98.05±1.33	97.92±2.11	98.46±1.79	96.96±2.06	96.64±3.37	1.000	1.000	0.114	0.731	1.000	0.132	0.820	0.072	0.512	1.000
CI	0.57±0.08	0.43±0.12	0.30±0.08	0.69±0.05	0.59±0.08	0.072	0.000	0.052	1.000	0.578	0.000	0.013	0.000	0.000	0.234
HI	1.05±0.01	1.05±0.02	1.11±0.03	1.08±0.03	1.05±0.01	1.000	0.000	0.009	1.000	0.000	0.009	1.000	1.000	0.000	0.000

Abbreviations: ARC = volumetric modulated arc therapy; IMRT = intensity-modulated radiation therapy; CRT = three-dimensional conformal radiation therapy; SD = standard deviation.

Figure 3.

Plots comparing V100%, V107%, HI, and CI among different treatment methods.P < 0.05 is marked by **, while P < 0.01 is marked by ***. Abbreviations: ARC = volumetric modulated arc therapy; IMRT = intensity-modulated radiation therapy; CRT = three-dimensional conformal radiation therapy; PTVtb = planning target volume (tumor bed); HI = homogeneity index; CI = conformity index; ns = not significant difference

Comparison of Organs at Risk

The results are presented in Table 3. The Dmean of the contralateral breast was significantly lower for the proton method compared to CRT (0.36±0.31 Gy (RBE) vs 1.21±0.41 Gy, P < 0.01) and VMAT (0.36±0.31 Gy (RBE) vs 2.19±0.38 Gy, P < 0.01). The Dmean for the carbon plan was also lower compared to IMRT (0.19±0.24 Gy (RBE) vs 0.85±0.21 Gy, P < 0.05), CRT (P < 0.01), and VMAT (P < 0.01). Regarding the heart, the Dmean in the proton methods (0.81±0.14 Gy (RBE)) was significantly lower than in CRT (10.76±6.28 Gy, P < 0.01), VMAT (6.49±8.96 Gy, P < 0.01), and IMRT (5.40±1.10 Gy, P < 0.01), while the carbon ion plans (1.23±0.59 Gy (RBE)) achieved a lower Dmean than CRT (P < 0.01) and IMRT (P < 0.01). Additionally, both particle methods reduced the Dmean of the LAD compared to CRT (33.73±7.77 Gy; proton: 4.64±3.15 Gy (RBE), P < 0.01; Carbon: 4.75±3.09 Gy (RBE), P < 0.01) and IMRT (19.74±10.25 Gy; proton: P < 0.01; Carbon: P < 0.01), as well as the Dmean of the RCA compared to CRT (3.62±4.36 Gy; proton: 0.15±0.18 Gy (RBE), P < 0.01; Carbon: 0.36±0.33 Gy (RBE), P < 0.01) and VMAT (4.12±2.92 Gy; proton: P < 0.01; Carbon: P < 0.01). V5 of the contralateral lung was also reduced in the particle methods compared to VMAT (3.03±3.14 %; proton: 0.12±0.38 %, P < 0.01; Carbon: 0.01±0.03 %, P < 0.01), though it was not significantly lower than in IMRT or CRT. Both proton and carbon methods also significantly reduced the Dmax of the spinal cord compared to VMAT and CRT. No significant difference was found between the proton and Carbon methods for the dosimetric parameters discussed above. Proton and carbon ion methods, however, did not demonstrate any significantly lower dose on ipsilateral lung (assessed by D_mean and V₂₀) or skin (assessed by D_mean, V₄₀, V₅₀) than VMAT or IMRT. Regarding the proton and carbon ion comparison, the carbon method demonstrated significant dosimetric superiority over proton in V₅ of ipsilateral lung (Carbon: 18.89±7.79 % vs proton: 26.29±9.02%, P = 0.011) and statistically insignificant dosimetric trend in other metrics of OARs including contralateral breast (Dmean), ipsilateral lung (Dmean, V₂₀), contralateral lung (V₅), skin (Dmean) and spinal cord (Dmax).

Table 3.

Comparison of Dosimetry of OARs Among Five Methods

OAR	Parameters	Limits	VMAT	IMRT	CRT	Carbon	Proton
Right Breast	Dmean/Gy	4	2.19±0.38	0.85±0.21	1.21±0.41	0.19±0.24	0.36±0.31
Left Lung	Dmean/Gy	10	7.25±0.89	7.97±1.48	14.13±2.33	4.92±2.37	5.86±2.16
	V5/%	40	31.27±2.14	28.47±3.90	40.29±7.17	18.89±7.79	26.29±9.02
	V20/%	20	12.06±2.64	15.60±3.49	30.63±5.16	9.18±5.65	12.42±5.14
Right Lung	V5/%	20	3.03±3.14	0.03±0.13	0.05±0.12	0.01±0.03	0.12±0.38
Heart	Dmean/Gy	5	6.49±8.96	5.40±1.10	10.76±2.51	1.23±0.59	0.81±0.37
Heart	V5/%	40	19.34±4.54	20.02±5.59	30.27±7.18	3.92±2.13	4.27±1.78
LAD	Dmean/Gy	25	13.00±7.73	19.74±10.25	33.73±7.77	4.75±3.09	4.64±3.15
RCA	Dmean/Gy	25	4.12±2.92	3.84±10.78	3.62±4.36	0.36±0.33	0.15±0.18
Skin	Dmean/Gy	-	40.25±5.42	41.12±5.67	44.68±6.62	34.52±2.76	43.64±2.43
Spinal Cord	D2%/Gy	-	3.70±1.23	0.58±0.07	0.77±0.09	0.00±0.00	0.02±0.02

Abbreviations: OAR = organs at risk; ARC = volumetric modulated arc therapy; IMRT = intensity-modulated radiation therapy; CRT = three-dimensional conformal radiation therapy; RCA = right coronary artery; LAD = left anterior descending coronary artery. Dmean = mean dose; V5 = the volume of OAR receiving at least 5 Gy irradiation; V20 = the volume of OAR receiving at least 20 Gy irradiation; D2% = the minimum dose received by the hottest 2% of the target volume (near-maximum dose).

Discussion

Over the past few decades, advancements in radiotherapy of breast cancer have significantly improved treatment outcomes. However, therapy-related complications remain a concern. Common adverse effects include dose-related coronary toxicity, radiation pneumonitis, and an increased risk of secondary malignancies, particularly in left-sided breast cancer. A study of 35,000 women diagnosed with breast cancer between 1976 and 2006 found that radiation to the left breast increased the risk of acute myocardial infarction, angina, pericarditis, and valvular heart disease.²⁰ Cardiovascular events occurring decades later have been attributed to excess and unintended radiation exposure to OARs.^21,22 This dosimetric comparison analysis aimed to evaluate the advantages and limitations of carbon ion therapy in a specific clinical setting, left-sided breast cancer following breast-conserving surgery, compared with conventional photon and proton therapies. The proton method demonstrated superior dose homogeneity and conformity relative to photon techniques, whereas carbon ion therapy exhibited less homogeneity than proton therapy and VMAT. Additionally, carbon ion therapy was less effective in reducing hot spots within the PTVtb compared to other planning methods. In terms of OAR sparing, particle therapies, particularly carbon ions, demonstrated dosimetric advantages by potentially reducing the radiation dose to the contralateral breast, heart, LAD, RCA, contralateral lung, and spinal cord. Given that long-term cardiovascular complications from breast cancer radiotherapy contribute to approximately 1% excess mortality,²¹ the observed reduction in OAR radiation exposure suggests a meaningful clinical benefit, particularly for patients with pre-existing cardiac or pulmonary conditions.

Studies on proton and carbon ion radiotherapy for breast cancer, particularly carbon ion therapy, are mainly limited by single-center, retrospective designs with small sample sizes. Research on PT in early and locally advanced breast cancer has primarily consisted of exploratory, single-center studies. By 2022, only approximately 34 clinical cases of CIRT for breast cancer had been reported.²² These studies have primarily concentrated on target coverage, tumor local control rates, acute and late adverse events, and the risks of radiation-induced cardiotoxicity and pneumonitis. Clinical trials involving uniform scanning or pencil beam scanning (PBS) PT have reported a higher incidence of complications, including radiation dermatitis, rib fractures, and esophagitis.^23,24 Careful consideration of chest wall toxicity is essential when designing PT treatment plans. Skin-sparing PBS technology may improve the feasibility of PT for whole-breast irradiation.⁸ In this study, 18 patients with left-sided breast cancer who underwent breast-conserving surgery were analyzed. Proton beam planning was performed using RayStation® TPS, leveraging prior clinical experience. However, data on carbon ion beam planning for whole-breast irradiation remain scarce. A study comparing passive and scanning irradiation methods for CIRT in breast cancer found no significant differences in dose distribution between the two approaches.²⁵ Carbon ion beam planning was conducted in ciPlan® TPS in this study, utilizing a forward design principle and dual-field interpenetrating irradiation. Larger clinical studies will be necessary to determine whether the improved CTV coverage observed with CIRT can contribute to lower breast cancer recurrence rates compared to modern CIRT-based techniques.

The distinctive physical properties of carbon ions have long attracted interest, leading to investigations into their clinical potential. One key property is the Bragg peak effect, which describes a sharp increase in dose deposition at a specific depth, followed by a rapid decline as the beam travels through tissue. This characteristic enables the Spread-Out Bragg Peak (SOBP) technique, allowing particle beams to deliver highly concentrated radiation doses to tumors while minimizing exposure to surrounding healthy tissues. This precision may help reduce both acute and late adverse effects of radiotherapy.²⁶ Additionally, carbon ion beams exhibit a steeper lateral dose fall-off, a smaller penumbra, compared to protons and photons, offering a potential advantage in improving radiotherapy precision.^26,27 Moreover, carbon ion radiation is characterized by high linear energy transfer (LET), meaning it deposits a significantly greater amount of energy per unit of track length as the particle traverses the tissue, causing clustered and severe DNA damage. While both carbon ions and protons share similar physical properties, carbon ions require more energy to reach the same tumor depth. This results in greater ionization density within the target, leading to more extensive irreparable DNA damage and potentially enhancing tumor cell death. Because of these properties, heavy ions, particularly carbon ions, exhibit a higher RBE compared to conventional photon and proton radiation. RBE is a critical measure when assessing the effectiveness of tumor cell eradication, further highlighting the therapeutic potential of carbon ion radiotherapy.^9,26

In terms of homogeneity and conformality, several key points merit discussion. In conventional carbon-ion therapy, a ridge filter is used to generate a SOBP that covers the PTV. However, because the SOBP has a fixed width, its proximal peak cannot be precisely conformed to the PTV, necessitating the combination of multiple fields to achieve the desired conformity. One approach to improve conformality is the layer-stacking method, where a multi-leaf collimator dynamically shapes the beam for each target layer, thereby enhancing the sparing of adjacent OARs, typically the skin.²⁸ Particle radiotherapy is generally delivered via either passive scattering or active scanning methods. Passive scattering employs compensators to broaden and shape the beam for a more conformal dose distribution, whereas active scanning uses a narrow beam that delivers the dose in a series of finely modulated beamlets via magnetic fields.⁹ Although active scanning is often regarded as superior in dose conformality, as evidenced in studies involving cancers such as pancreatic cancer,^28,29 it may not offer the same dosimetric advantage over passive scattering in the context of thoracic tumors or breast cancer.^25,30 For breast cancer, the shallow location of the target might contribute to this outcome, given that the beam spot size tends to increase when beam energy decreases to reduce the depth of penetration.³¹ It should also be noted that many current carbon-ion treatment systems lack active scanning capabilities. Furthermore, while active scanning techniques, such as PBS can support both forward and inverse planning depending on the treatment planning system’s design and algorithm, inverse planning in carbon-ion therapy remains challenging due to the complexity of carbon-ion beam physics and control. This limitation may restrict improvements in the homogeneity and conformality of carbon-ion treatment plans and help explain the issues mentioned above. These challenges underscore the need for further technological advancements.

This study has certain limitations. The sample size was limited to 18 patients, which, while providing statistical power, would benefit from a larger cohort to generate more reliable dose statistics and risk assessments. Additionally, this study is a comparative dosimetric analysis based on simulations, highlighting the need for further rigorous research extending from physical dose calculations to biological dose modeling and ultimately to biological and clinical effects. While favorable dose distribution suggests potential clinical benefits, it does not inherently confirm a therapeutic advantage. Another complexity arises from the variation in RBE with tissue penetration depth in particle therapy. More specifically, RBE fluctuates with particle energy, and due to the non-uniform energy distribution of carbon ions across the treatment field, accurately predicting RBE at different dose depths is critical for clinical application.^32-34 This also raises concerns regarding the uncertainty of range and RBE at the distal end of carbon-ion tracks, which may lead to unintended high biological doses in critical structures near the target, such as the heart and coronary arteries in left-sided breast cancer. Furthermore, the biological effects of carbon-ion radiotherapy extend beyond the target area. Studies have reported that carbon-ion irradiation of mammary tissue can induce non-targeted, out-of-field upregulation of COX-2 and DNA damage in tissues outside the treatment field, presenting additional challenges in evaluating its biological impact.³⁵ Advancing research in microdosimetry and radiation biology will be essential for refining biological dose calculations and improving the accuracy of biological effect assessments.

Another notable limitation of this study is the utilization of free-breathing (FB) CT scans for treatment planning. We fully acknowledge that deep inspiration breath hold (DIBH) is currently the state-of-the-art technique for left-sided breast cancer especially in IMRT, as it anatomically displaces the heart away from the chest wall, thereby reducing cardiac dose.^36,37 However, a significant proportion of patients in real-world clinical practice cannot tolerate or achieve reproducible DIBH due to poor pulmonary function, advanced age, or anxiety.³⁸ Therefore, dosimetric evaluations under FB conditions remain highly clinically relevant for this specific patient subset. Furthermore, utilizing FB scans in this study serves as a rigorous dosimetric “stress test.” Evaluating these modalities under FB conditions demonstrates that even without the anatomical heart-sparing advantage provided by DIBH, particle therapy—particularly carbon ions—can still significantly reduce the dose to the heart and LAD. This underscores the intrinsic physical superiority of the heavy ion beam’s Bragg peak and steep dose fall-off. Future studies combining particle therapy with DIBH and advanced respiratory motion management are warranted to explore the ultimate limits of OAR sparing.

Finally, regarding the treatment planning methodology, a 3.0 mm dose calculation grid was utilized for the photon plans. While the AAPM TG-218 guidelines recommend a finer grid (≤2.0 mm) for optimal IMRT/VMAT accuracy,³⁹ our grid size was dictated by hard technical constraints: the original simulation CTs were acquired with a 5.0 mm slice thickness, and the clinical TPS fundamentally locked the calculation grid to match this image resolution limit. Utilizing a finer calculation grid on 5.0 mm thick slices would largely yield interpolated pseudo-resolution rather than genuine dosimetric improvement. Most importantly, because this identical calculation grid was applied uniformly across all photon modalities, the internal consistency of the study is maintained, ensuring that the relative dosimetric comparisons drawn remain robust and valid.

Conclusions

The dosimetric analysis of heavy ion radiotherapy for left-sided breast cancer following breast-conserving surgery showed that carbon-ion radiation offered superior conformality for PTV and PTVtb compared to certain photon-based methods. Additionally, carbon-ion radiotherapy demonstrated dosimetric advantages over some CRT, IMRT, VMAT, and proton therapies in reducing radiation exposure to critical organs, including the contralateral breast, heart, LAD, RCA, contralateral lung, and spinal cord. This suggests a potential benefit in protecting vital organs during treatment. These findings highlight the promising role of heavy ion radiotherapy in managing early-stage breast cancer, especially for left-sided cases where heart and lung protection is crucial. Further research is needed to refine and standardize carbon-ion radiotherapy protocols for breast cancer treatment following breast-conserving surgery.

Footnotes

Acknowledgements

The authors used an artificial intelligence tool (ChatGPT, OpenAI, San Francisco, CA, USA) to assist with language editing of the manuscript. The authors reviewed and edited the content as needed and take full responsibility for the final manuscript.

ORCID iDs

Wenlingjie Huang

Jiaming Wu

Ethical Considerations

This retrospective study was approved by the Peking University Third Hospital Medical Science Research Ethics Committee (Project No. IRB00006761-M20260215; Approval No. (2026) Medical Ethics Review No. (202-02); Date of approval: March 25, 2026).

Consent to Participate

Patient consent was waived by the Peking University Third Hospital Medical Science Research Ethics Committee due to the retrospective nature of this study. All patient data were completely de-identified prior to analysis, and no patient privacy was compromised.

Author Contributions

Conceptualization: Ping Jiang and Wenlingjie Huang. Methodology: Runhong Lei, Jiaming Wu, Tianye Niu. Data curation and investigation: Ping Jiang, Xiaoyun Ma, Wanbin Meng, Xiuwen Deng and Yancheng Ye. Formal analysis: Wenlingjie Huang and Runhong Le. Writing (original draft preparation): Wenlingjie Huang and Runhong Lei. Writing (review and editing): Ping Jiang and Junjie Wang. Supervision: Junjie Wang and Yancheng Ye. All authors read and approved the final manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the following funds: Key Specialty program of Natural Science Foundation of Beijing Municipality (Z20008 to P.Jiang), Beijing, China. National Key Research and Development Program (2023YFC2413905), Beijing, China. National Key Research and Development Program (2022YFC2404606), Beijing, China.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.*

Appendix

References

Speers

Pierce

. Postoperative radiotherapy after breast-conserving surgery for early-stage breast cancer: A review. JAMA Oncol. 2016;2:1075-1082.

De La Cruz Ku

Karamchandani

Chambergo-Michilot

, et al. Does breast-conserving surgery with radiotherapy have a better survival than mastectomy? A meta-analysis of more than 1,500,000 patients. Annals of Surgical Oncology. 2022;29:6163-6188.

Haffty

Buchholz

Mccormick

. Should intensity-modulated radiation therapy be the standard of care in the conservatively managed breast cancer patient? Journal of Clinical Oncology. 2008;26:2072-2074.

Jagsi

. Progress and controversies: Radiation therapy for invasive breast cancer. CA: A Cancer Journal for Clinicians. 2014;64:135-152.

Burt

Ying

Poppe

Suneja

Gaffney

. Risk of secondary malignancies after radiation therapy for breast cancer: Comprehensive results. Breast. 2017;35:122-129.

Pompos

Durante

Choy

. Heavy ions in cancer therapy. JAMA Oncology. 2016;2:1539-1540.

Ruan

Okamoto

Ohno

Zhou

. Particle radiotherapy for breast cancer. Frontiers in Oncology. 2023;13:1107703.

Mutter

Choi

Jimenez

, et al. Proton therapy for breast cancer: A consensus statement from the particle therapy cooperative group breast cancer subcommittee. Int J Radiat Oncol Biol Phys. 2021;111:337-359.

Malouff

Mahajan

Mutter

, et al. Carbon ion radiation therapy in breast cancer: A new frontier. Breast Cancer Research and Treatment. 2020;181:291-296.

10.

Kammerer

Guevelou

Chaikh

, et al. Proton therapy for locally advanced breast cancer: A systematic review of the literature. Cancer Treat Rev. 2018;63:19-27.

11.

Patel

Nyamwanda

, et al. Postmastectomy radiation therapy technique and cardiopulmonary sparing: A dosimetric comparative analysis between photons and protons with free breathing versus deep inspiration breath hold. Pract Radiat Oncol. 2017;7:e377-e384.

12.

Kozak

Katz

Adams

, et al. Dosimetric comparison of proton and photon three-dimensional, conformal, external beam accelerated partial breast irradiation techniques. Int J Radiat Oncol Biol Phys. 2006;65:1572-1578.

13.

Hernandez

Zhang

Sanders

Newhauser

. A treatment planning comparison of volumetric modulated arc therapy and proton therapy for a sample of breast cancer patients treated with post-mastectomy radiotherapy. Journal of Proton Therapy. 2016;1:119.

14.

Akamatsu

Karasawa

Omatsu

Isobe

Ogata

Koba

. First experience of carbon-ion radiotherapy for early breast cancer. Japanese Journal of Radiology. 2014;32:288-295.

15.

Karasawa

Omatsu

Shiba

Irie

Wakatsuki

Fukuda

. A clinical study of curative partial breast irradiation for stage i breast cancer using carbon ion radiotherapy. Radiation Oncology. 2020;15:265.

16.

Karasawa

Omatsu

Arakawa

, et al. A phase I clinical trial of carbon ion radiotherapy for stage i breast cancer: Clinical and pathological evaluation. J Radiat Res. 2019;60:342-347.

17.

Giuliano

Edge

Hortobagyi

. Eighth edition of the ajcc cancer staging manual: Breast cancer. Ann Surg Oncol. 2018;25:1783-1785.

18.

Fuchs

Parisi

Furutani

Georg

Beltran

. Influence of nozzle configurations on radiobiological effects in carbon ion radiation therapy estimated using GATE/Geant4. Z Med Phys. 2025;36:238-246. doi:10.1016/j.zemedi.2025.04.004. Published online May 8.

19.

von

Altman

Egger

Pocock

Gøtzsche

Vandenbroucke

. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147:573-577.

20.

McGale

Darby

Hall

, et al. Incidence of heart disease in 35,000 women treated with radiotherapy for breast cancer in denmark and sweden. Radiother Oncol. 2011;100:167-175.

21.

Ebctc

. Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: An overview of the randomised trials. Early breast cancer trialists' collaborative group. Lancet. 2000;355:1757-1770.

22.

Fan

Pei

. Value of carbon-ion radiation therapy for breast cancer. Int J Part Ther. 2024;14:100629.

23.

Luo

Cuaron

Braunstein

, et al. Early outcomes of breast cancer patients treated with post-mastectomy uniform scanning proton therapy. Radiother Oncol. 2019;132:250-256.

24.

Verma

Iftekaruddin

Badar

, et al. Proton beam radiotherapy as part of comprehensive regional nodal irradiation for locally advanced breast cancer. Radiother Oncol. 2017;123:294-298.

25.

Matsubara

Karasawa

Furuichi

, et al. Comparison of passive and scanning irradiation methods for carbon-ion radiotherapy for breast cancer. J Radiat Res. 2018;59:625-631.

26.

Mohamad

Sishc

Saha

, et al. Carbon ion radiotherapy: A review of clinical experiences and preclinical research, with an emphasis on DNA damage/repair. Cancers (Basel). 2017;9:66.

27.

Rackwitz

Debus

. Clinical applications of proton and carbon ion therapy. Semin Oncol. 2019;46:226-232.

28.

Kanematsu

Inaniwa

. Treatment planning of carbon-ion radiotherapy. In: Tsujii

Kamada

Shirai

Noda

Tsuji

Karasawa

, eds. Carbon-ion radiotherapy: Principles, practices, and treatment planning. Tokyo: Springer Japan; 2014:87-97.

29.

Shiomi

Mori

Shinoto

Nakayama

Kamada

Yamada

. Comparison of carbon-ion passive and scanning irradiation for pancreatic cancer. Radiother Oncol. 2016;119:326-330.

30.

Karube

Mori

Tsuji

, et al. Carbon-ion pencil beam scanning for thoracic treatment - initiation report and dose metrics evaluation. J Radiat Res. 2016;57:576-581.

31.

Inaniwa

Furukawa

Kanematsu

, et al. Evaluation of hybrid depth scanning for carbon-ion radiotherapy. Med Phys. 2012;39:2820-2825.

32.

Karger

Peschke

. Rbe and related modeling in carbon-ion therapy. Phys Med Biol. 2017;63:01TR02.

33.

Schlaff

Krauze

Belard

O'Connell

Camphausen

. Bringing the heavy: Carbon ion therapy in the radiobiological and clinical context. Radiat Oncol. 2014;9:88.

34.

Friedrich

Weyrather

Elsasser

Durante

Scholz

. Accuracy of rbe: Experimental and theoretical considerations. Radiat Environ Biophys. 2010;49:345-349.

35.

Wang

Chai

, et al. Induction of non-targeted stress responses in mammary tissues by heavy ions. PLoS One. 2015;10:e0136307.

36.

Lai

Luo

, et al. Meta-analysis of deep inspiration breath hold (DIBH) versus free breathing (FB) in postoperative radiotherapy for left-side breast cancer. Breast Cancer. 2020;27(2):299-307.

37.

Yang

Zhang

, et al. Comparison of Deep Inspiration Breath Hold Versus Free Breathing in Radiotherapy for Left Sided Breast Cancer. Front Oncol. 2022;12:845037.

38.

Gaál

Kahán

Paczona

, et al. Deep-inspirational breath-hold (DIBH) technique in left-sided breast cancer: various aspects of clinical utility. Radiat Oncol. 2021;16(1):89.

39.

Miften

Olch

Mihailidis

, et al. Tolerance limits and methodologies for IMRT measurement-based verification QA: Recommendations of AAPM Task Group No. 218. Med Phys. 2018;45(4):e53-e83.