A prospective phase II trial of 10-fraction whole-breast radiotherapy following breast-conserving surgery

Highlights

• •A 10-fraction whole-breast radiotherapy with sequential tumor-bed boost was prospectively evaluated.
• •Grade 2 acute dermatitis occurred in 17.2% of patients, with no grade ≥3 events.
• •The most frequently affected site was the nipple–areola complex.
• •Prior chemotherapy and low CD4 + T-cell counts predicted higher dermatitis risk.

Abstract

Background

Moderate hypofractionation is established as the standard of care for adjuvant whole-breast radiotherapy after breast-conserving surgery. Ultra-hypofractionated regimens further shorten treatment but have raised concerns about toxicity and cosmetic outcomes. In younger Asian patients frequently requiring a boost, we evaluated a 10-fraction whole-breast radiotherapy schedule as a practical and tolerable choice.

Methods

This single-arm, prospective phase II trial enrolled 64 patients with early-stage breast cancer at Peking University Cancer Hospital between November 2023 and March 2025. All patients received intensity-modulated radiotherapy (IMRT) with a prescribed dose of 37 Gy in 10 fractions, followed by a sequential tumor-bed boost of 7.4 Gy in 2 fractions according to clinical indications. The primary endpoint was the incidence of grade ≥2 acute radiation dermatitis; secondary endpoints included patient-reported outcomes, cosmetic assessment, oncologic outcomes and exploratory immune profiling. Clinical Trial registration: ChiCTR2300075391.

Results

All patients completed radiotherapy without interruption. Tumor-bed boost was delivered to 58 patients (90.6%). Grade 2 acute dermatitis was observed in 17.2% of patients, with no grade ≥3 events, and all reactions resolved to grade 0–1 within three months. Among the 11 cases with grade 2 dermatitis, the most frequently affected site was the nipple–areola complex (6 cases, 54.5%). Patient-reported cosmetic and breast symptom scores transiently worsened at two weeks after radiotherapy but recovered thereafter. No significant decline occurred in functional or global quality-of-life domains. Exploratory analyses suggested that prior chemotherapy and lower baseline CD4⁺ T-cell counts were associated with a higher risk of grade ≥2 dermatitis.

Conclusion

This 10-fraction whole-breast radiotherapy regimen was acceptable. These findings support this regimen as a practical alternative for adjuvant breast radiotherapy.

Introduction

Breast-conserving surgery followed by adjuvant whole-breast radiotherapy is a standard component of care in the management of early-stage breast cancer [1]. For many years, it was commonly delivered with conventional fractionation (50 Gy in 25 fractions over 5 weeks). Over the last two decades, however, clinical practice has shifted toward moderate hypofractionation (e.g., 40–42.5 Gy in 15–16 fractions), supported by randomized trial evidence demonstrating comparable local control with acceptable toxicity [2], [3], [4]. Importantly, shorter schedules lessen the treatment burden and can reduce healthcare resource utilization under real-world clinical practice [5].

Recent trials [6], [7] have evaluated ultra-hypofractionated whole breast radiotherapy delivered in five fractions, allowing treatment completion within one week. However, the logistical time advantage may be offset when a sequential boost is delivered using conventional fractionation. In addition, increasing the dose per fraction has raised concerns to normal tissue toxicity and cosmetic outcomes [8], [9]. These issues are particularly relevant when extending these regimens to broader and diverse clinical settings in Asian populations, where a younger age at diagnosis often necessitates a tumor bed boost and differences in anatomical features may influence toxicity profiles [10]. Therefore, a regimen that balances efficiency with a conservative dose-per-fraction is a clinically need.

A 10-fraction regimen offers a pragmatic compromise between conventional and ultra-hypofractionated schedules. Several prior studies have evaluated 10–11-fraction schedules completed with 2–2.5 weeks, yet the available data remain limited by modest sample sizes, heterogeneity in boost delivery, and a scarcity of data in Asian populations [11], [12], [13]. On this basis, we conducted a prospective phase II study to assess the safety and tolerability of a 10-fraction regimen. Additionally, we performed an exploratory analysis of risk factors for acute dermatitis and characterized lymphocyte kinetics.

Method

Study design and participants

This prospective, single-arm phase II study was conducted at Peking University Cancer Hospital. The protocol was approved by the institutional review board and ethics committee of Peking University Cancer Hospital. The study was performed in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Written informed consent was obtained from all participants before enrollment. The trial was registered at the Chinese Clinical Trial Registry (ChiCTR2300075391).

Patients were prospectively enrolled between November 2023 and March 2025 at Peking University Cancer Hospital. Eligible patients were females aged 18–70 years with histologically confirmed invasive breast cancer or ductal carcinoma in situ (DCIS) who had undergone breast-conserving surgery with negative margins. Pathological eligibility criteria were included yp/pTis-T2M0, and yp/pN0 or pN1 without high-risk features (defined as young age, extensive lymphovascular invasion, hormone receptor negativity, or high-grade histology). An Eastern Cooperative Oncology Group (ECOG) performance status of 0–1 was required. Key exclusion criteria included bilateral breast cancer, previous radiotherapy to the breast or chest wall, pregnancy or lactation, and active collagen vascular disease.

Neoadjuvant and/or adjuvant systemic therapy was permitted according to routine clinical practice. Radiotherapy was initiated within 12 weeks after surgery in patients not receiving adjuvant chemotherapy, and within 8 weeks after adjuvant chemotherapy in those who did. Concurrent endocrine therapy and anti-HER2 agents were permitted. Concurrent intravenous cytotoxic chemotherapy was prohibited; however, concurrent oral capecitabine was permitted according to clinical indications. In practice, only one patient (1.6%) received capecitabine during radiotherapy.

For exploratory immune profiling, peripheral venous blood was collected at two prespecified time points: baseline (within one week before radiotherapy) and at the end of radiotherapy (on the day of or one day before the final fraction). Samples were processed in the clinical laboratory of Peking University Cancer Hospital using standard flow cytometry to quantify absolute counts of lymphocyte subsets, including total T cells (CD3 + ), helper T cells (CD3 + CD4 + ), cytotoxic T cells (CD3 + CD8 + ), B cells (CD3-CD19 + ), and natural killer cells (CD3-CD16 + CD56 + ).

Radiotherapy procedures

Patients were immobilized in the supine position using a breast board in combination with a thermoplastic device. Computed tomography (CT) simulation was performed with a slice thickness of 5 mm, covering from the mandible to the inferior margin of the liver.

The clinical target volume for the whole breast (CTVbr) was contoured on the planning CT to encompass all visible ipsilateral breast tissue, excluding the most superficial 3–5 mm of skin and the underlying pectoralis muscle, in accordance with institutional practice based on RTOG/ESTRO recommendations. The tumor bed clinical target volume (CTVtb) was delineated using surgical clips, postoperative seroma, preoperative imaging operative reports, and the location of the surgical incision. A 5-mm isotropic margin was then added to generate the planning target volumes for the breast and tumor bed (PTVbr and PTVtb), which were cropped 3–5 mm from the skin surface.

All patients received whole-breast radiotherapy using intensity-modulated radiotherapy (IMRT) with 6-MV photons. The prescribed dose to the whole breast was 37 Gy in 10 fractions (3.7 Gy/fraction) delivered over 2 weeks. A sequential tumor-bed boost was recommended for patients with predefined high-risk features, including age ≤50 years, age >50 years with high-grade histology, or close surgical margins (<2 mm for DCIS). In addition, other adverse pathological features (e.g., high-risk histologic subtype, lymphovascular invasion, perineural invasion, or extensive/high-grade DCIS) could be taken into account following multidisciplinary discussion in accordance with institutional practice. Patients meeting these criteria received a tumor bed boost of 7.4 Gy in 2 fractions (3.7 Gy/fraction), resulting in a total treatment duration of 2.4 weeks.

According to START [3], [14] and FAST-Forward [7] trials, the whole-breast dose of 37 Gy in 10 fractions (3.7 Gy per fraction) was chosen based on radiobiological modelling using an α/β value of 4 Gy for tumor control (EQD2 ≈ 47.5 Gy), 3 Gy for late normal-tissue effects (EQD2 ≈ 49.6 Gy), and 10 Gy for acute skin reactions (EQD2 ≈ 42.2 Gy). A detailed EQD2 comparison of this schedule with standard moderate and ultra-hypofractionated whole-breast regimens is provided in Supplementary Methods S1 and Supplementary Table 1.

Treatment plans were optimised with the goal that at least 95% of the PTVbr received the prescription dose while keeping the maximum dose within the PTV ≤ 110% of the prescription and respecting institutional dose–volume constraints for the ipsilateral lung, heart/left anterior descending artery (for left-sided cases) and contralateral breast (detailed in Supplementary Tables 2–3). Image-guided radiotherapy (IGRT) using cone-beam CT (CBCT) was used for setup verification. CBCT frequency was platform-specific: daily CBCT on one linear accelerator, and CBCT during the first three fractions followed by periodic verification (fractions 6 and 11) on the other. A setup deviation threshold of 5 mm was applied. Patient-specific quality assurance was conducted using EPID-based portal dosimetry (Varian Eclipse Portal Dosimetry) with gamma analysis to verify field-level dose delivery. Detailed specifications for target delineation, dosimetric constraints, and verification protocols are provided in the Supplementary Methods.

Outcome assessments

The primary endpoint was the incidence of grade ≥2 acute radiation dermatitis. Secondary endpoints included cosmetic outcomes, patient-reported quality of life (QoL), late radiation-induced toxicity, and oncologic outcomes.

Acute skin toxicity was graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. Acute radiation dermatitis was prospectively assessed at baseline, weekly during treatment, at the completion, 2–3 weeks, and 3 months after radiotherapy.

Cosmetic outcomes from physician reports were evaluated using the Harvard cosmetic scale, while patient-reported QoL was assessed using the Breast Cancer Treatment Outcome Scale (BCTOS) and the European Organisation for Research and Treatment of Cancer quality-of-life questionnaires (EORTC QLQ-C30 and QLQ-BR23). Assessments were performed at baseline, the completion of radiotherapy, 2 weeks, 6, 12, 18, and 24 months after radiotherapy, and annually thereafter. Late radiation-induced toxicity was evaluated at predefined visits at 6, 12, 18, and 24 months after radiotherapy, and annually thereafter. Evaluation used the Radiation Therapy Oncology Group (RTOG) late radiation morbidity criteria and the Late Effects in Normal Tissues–Subjective, Objective, Management, and Analytic (LENT-SOMA) scale. Oncologic outcomes, including ipsilateral breast tumor recurrence, regional recurrence, distant metastasis, and survival events, were recorded every 3 months during the first 2 years after treatment, every 6 months during years 3–5, and annually thereafter. Further details on endpoint definitions and scoring criteria are provided in the Supplementary Methods.

This report focuses on acute toxicity and associated exploratory endpoints. Data on late toxicity and oncologic outcomes are still maturing and will be reported separately upon longer follow-up.

Statistical analysis

The sample size was determined based on the primary endpoint of acute grade ≥2 radiation dermatitis. The expected incidence of grade ≥2 acute radiation dermatitis was assumed to be 20%, while 35% was considered as the upper limit of acceptable toxicity. With a one-sided significance level (α) of 0.10 and a statistical power of 80%, 50 patients was required. Allowing for an anticipated dropout or non-evaluable rate of 20%, the planned sample size was 63. Sample size calculations were performed using PASS software (NCSS, LLC, Kaysville, UT, USA). Continuous variables were reported as mean ± standard deviation (SD) or median with interquartile range (IQR), as appropriate. Categorical variables were summarized as frequencies and percentages. Longitudinal changes in patient-reported outcomes and QoL scores were analyzed using paired t test or Wilcoxon signed-rank test, as appropriate. Exploratory analyses were performed to evaluate potential predictors of grade ≥ 2 acute radiation dermatitis. Because of the limited number of events, Firth’s penalized likelihood regression was applied to reduce small-sample bias. Differences in lymphocyte subsets between patients with and without grade ≥ 2 dermatitis were compared using the Wilcoxon rank-sum test. All statistical analyses were performed using R software (version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria). All statistical tests were two-sided, and a P < 0.05 was considered statistically significant.

Results

Patient characteristics and treatment delivery

A total of 69 patients were assessed for eligibility, and 64 were enrolled and completed radiotherapy (Fig. 1). All enrolled patients were included in the acute toxicity analysis.

Fig. 1.

Study flow diagram.

Baseline demographic and clinicopathological characteristics are summarized in Table 1. The mean age was 47.3 years (SD, 9.2). Patients were characterized by a relatively low mean body mass index (24.5 kg/m², SD 3.3) and generally small breast volumes (95.3% classified as small or medium). Most patients had hormone receptor–positive disease (ER+, 90.6%; PR+, 87.5%). Most patients had early-stage disease (86% pathological stage 0–I; 95.3% pN0).

Table 1.

Baseline clinicopathological characteristics of the study population.

Characteristic	Level	N = 64
Age, years, mean (SD)		47.3 (9.2)
Menopausal status, n (%)	Premenopausal	20 (31.3)
	Postmenopausal	28 (43.8)
	Uncertain	16 (25.0)
Family history of malignancy, n (%)	None	34 (53.1)
	Gynecologic malignancies*	10 (15.6)
	Other malignancies	20 (31.3)
BMI, kg/m2, mean (SD)		24.5 (3.3)
Tumor size, cm, mean (SD)		2.0 (0.78)
Tumor location, n (%)	Upper outer quadrant	32 (50.0)
	Upper inner quadrant	12 (18.8)
	Lower outer quadrant	4 (6.3)
	Lower inner quadrant	2 (3.1)
	Central/subareolar	10 (15.6)
	Unknown	4 (6.2)
ER, n (%)	Negative	6 (9.4)
	Positive	58 (90.6)
PR, n (%)	Negative	6 (9.4)
	Positive	56 (87.5)
	Unknown	2 (3.1)
HER2, n (%)	Negative	45 (70.3)
	Positive	16 (25.0)
	Unknown	3 (4.7)
Molecular subtype, n (%)	Luminal A	27 (42.2)
	Lunimal B (HER2-positive)	13 (20.3)
	Luminal B (HER2-negative)	16 (25.0)
	HER-2 enriched	3 (4.7)
	TNBC	2 (3.1)
	Unknown	3 (4.7)
Histological grade, n (%)	Low	7 (10.9)
	Intermediate	34 (53.1)
	High	19 (29.7)
	Unknown	4 (6.3)
Histological subtype, n (%)	IDC	50 (78.1)
	DCIS	5 (7.8)
	Others	9 (14.1)
Neoadjuvant therapy, n (%)	Yes	15 (23.4)
	No	49 (76.6)
Pathological T stage, n (%)	T0	11 (17.2)
	Tis	6 (9.4)
	T1	38 (59.4)
	T2	9 (14.1)
Pathological N stage, n (%)	N0	61 (95.3)
	N1mi	2 (3.1)
	Unknown	1 (1.6)
Pathological stage, n (%)	0	17 (26.6)
	I	38 (59.4)
	II	8 (12.5)
	Unknown	1 (1.6)
Lymphovascular invasion, n (%)	Present	6 (9.4)
	Absent	58 (90.6)
Adjuvant therapy, n (%)	None	2 (3.1)
	Chemotherapy	10 (15.6)
	Anti-HER2 therapy	15 (23.4)
	Endocrine therapy	57 (89.1)
Breast volume**, n (%)	Small (<500 cm3)	29 (45.3)
	Medium (500–1000 cm3)	32 (50.0)
	Large (>1000 cm3)	3 (4.7%)
Breast defect status***, n (%)	None or minimal defect	60 (93.8)
	Mild defect	4 (6.3)
	Moderate defect	0
	Severe defect	0

*Gynecologic malignancies include breast, cervical, endometrial, and ovarian cancers.

^**Breast volume was categorized based on whole-breast CTV volume.

^*** Breast defect status was graded into four categories (none or minimal, mild, moderate, severe) with reference to the four-point Harvard cosmetic scale.

ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; IDC, invasive ductal carcinoma; DCIS, ductal carcinoma in situ; Tis, carcinoma in situ; N1mi, micrometastasis in regional lymph nodes; CTV, clinical target volume.

A sequential tumor bed boost was delivered to 58 patients (90.6%), while 6 patients (9.4%) received whole-breast radiotherapy alone. Concurrent systemic therapy during radiotherapy included endocrine therapy in 82.8% and anti-HER2 therapy in 23.4% of patients; one patient received concurrent oral capecitabine. Dosimetric parameters indicated favorable organ sparing and are presented in Supplementary Table 4.

Acute radiation dermatitis

Table 2 shows that grade 2 acute dermatitis occurs in 11 of 64 patients (17.2%), typically within 1–2 weeks after completion of radiotherapy. By 3 months post-treatment, all cases had recovered to grade 0–1. No grade ≥ 3 events were observed. The nipple–areola complex was the most frequently affected subsite (6/11, 54.5%), with the remaining events involving the inframammary fold (n = 2), axilla (n = 2), and surgical scar (n = 1). No treatment interruptions or discontinuations due to acute skin toxicity.

Table 2.

Acute skin toxicity at different time points.

Skin toxicity grade	During radiotherapy N* = 64, n (%)	2 weeks after radiotherapy N*=64, n (%)	3 months after radiotherapy N* = 46, n (%)
Grade 0	14 (21.9)	13 (20.3)	32 (69.6)
Grade 1	50 (78.1)	40 (62.5)	14 (30.4)
Grade 2	0	11 (17.2)	0
Grade 3	0	0	0
Grade 4	0	0	0

^*N indicates the number of patients with available assessments at each time point.

Patient-reported outcomes and cosmetic assessment

Patient-reported outcomes (PROs) are summarized in Table 3. At 2 weeks after radiotherapy, significant increases were observed in BCTOS scores for cosmetic appearance and breast pain (both P < 0.001), whereas functional scores did not change significantly (P = 0.254). Similarly, scores for breast symptoms on the EORTC QLQ-BR23 scale increased significantly (P < 0.001). However, global health status and fatigue on the EORTC QLQ-C30 remained stable (P = 0.889 and P = 0.286, respectively), although a modest increase in general pain scores was observed (P = 0.012). Physician-assessed cosmetic outcomes were evaluated using the Harvard cosmetic scale (Supplementary Table 5). A transient decline was observed at 2 weeks after radiotherapy, with recovery by 3 months.

Table 3.

Changes in patient-reported outcomes from baseline to 2 weeks after radiotherapy.

Measure	Domain, n (paired)	Before radiotherapy	2 weeks after radiotherapy	P value, r/|d|
BCTOS*, median (IQR)	Cosmetic, n = 57	1.18 (1.00–––1.36)	1.27 (1.18–––1.61)	<0.001, 0.56
	Function, n = 56	1.00 (1.00–––1.00)	1.00 (1.00–––1.15)	0.254, 0.15
	Pain, n = 57	1.17 (1.00–––1.17)	1.33 (1.17–––1.50)	<0.001, 0.52
QLQ-BR23**, mean (SD)	Body image, n = 62	92.35 (18.22)	93.16 (17.88)	0.637, 0.06
	Arm symptom, n = 58	4.84 (8.31)	7.73 (12.13)	0.109, 0.21
	Breast symptom, n = 58	8.10 (8.97)	22.85 (18.69)	<0.001, 0.79
QLQ-C30**, mean (SD)	Fatigue, n = 62	16.75 (19.69)	19.47 (20.38)	0.286, 0.14
	Pain, n = 61	7.95 (10.26)	13.61 (16.53)	0.012, 0.33
	Global health status, n = 57	78.32 (20.08)	78.00 (19.17)	0.889, 0.02

For symptom scales (e.g., pain, fatigue, breast/arm symptoms), higher scores indicate worse symptoms; for global health status and body image, higher scores indicate better functioning.

Paired analyses were performed using available complete pairs; the effective paired sample size may vary across domains due to missing data.

BCTOS, Breast Cancer Treatment Outcome Scale; QLQ-C30, EORTC Quality of Life Questionnaire Core 30; QLQ-BR23, EORTC Quality of Life Questionnaire Breast Cancer–Specific Module; IQR, interquartile range; SD, standard deviation.

^*BCTOS domains are presented as median (interquartile range Q3-Q1) and were compared using the Wilcoxon signed-rank test; effect size is reported as r = |Z|/√N, 0.1/0.3/0.5 indicate small/moderate/large effects.

^**EORTC QLQ-C30 and QLQ-BR23 scores are presented as mean (SD) and were compared using paired t tests; effect size is reported as Cohen’s d (absolute value shown), 0.2/0.5/0.8 indicate small/moderate/large effects.

Predictors of acute toxicity and lymphocyte kinetics

In exploratory analyses (Fig. 2), traditional anatomical factors such as body mass index and breast volume showed no significant association with acute radiation dermatitis. Also, no statistically significant association was observed between tumor bed boost and grade ≥2 dermatitis. Prior chemotherapy increased the risk of grade ≥2 dermatitis (OR 5.52, 95% CI 1.49–24.84, P = 0.010), while baseline systemic immune status was also a significant predictor. Higher baseline counts of total T cells and CD4⁺ T cells were associated with a reduced risk of grade ≥2 dermatitis (total T cells: OR, 0.74; 95% CI, 0.57–0.91, P = 0.003; CD4⁺ T cells: OR, 0.62; 95% CI, 0.39–0.89; P = 0.007). Lymphocyte kinetics are shown in Supplementary Fig. 1. Patients who developed grade ≥2 dermatitis exhibited lower absolute CD4⁺ T cell counts at both baseline and the end of radiotherapy compared with those in grade 0–1 toxicity.

Fig. 2.

Uni-variable Firth logistic regression of clinical/treatment factors and baseline lymphocyte subsets for prediction of grade ≥ 2 acute radiation dermatitis. OR = odds ratio; CI = confidence interval; ALC = absolute lymphocyte count; NK = natural killer; CD = cluster of differentiation.

Following radiotherapy, the absolute lymphocyte count (ALC) decreased from baseline with a median relative reduction of 32.0% (IQR: 24.5%–41.0%). Subgroup analysis revealed differential declines among lymphocyte subsets. B cells showed the greatest decline (median 43.5%, IQR 23.0–55.0%), followed by NK cells (median 40.0%, IQR 29.0–55.5%). T-cell subsets were relatively preserved, with CD8⁺ T cells decreasing by a median of 32.5% (IQR 21.0–43.5%) and CD4⁺ T cells showing the smallest decline (median 22.0%, IQR 15.0–33.5%).

Oncologic outcome

With a median follow-up of 18.2 months (IQR, 15.3–25.0 months), no local recurrences were observed. No distant metastases or deaths occurred during follow-up.

Discussion

This prospective phase II trial investigated a 10-fraction whole-breast radiotherapy regimen following breast-conserving surgery. Grade ≥2 radiation dermatitis occurred in 17.2% of patients, with no grade ≥3 events or treatment interruptions, and all cases resolved to grade 0–1 within 3 months post-radiotherapy. Cosmetic and patient-reported outcomes demonstrated transient increases in breast symptoms and pain scores at 2 weeks post-treatment, without significant deterioration in functional status or global quality of life.

While moderate hypofractionation is established as the standard of care for whole-breast radiotherapy in early-stage breast cancer, ultra-hypofractionated further reduce treatment time [7], [9]. However, the practical advantage of ultra-hypofractionated schedule is often offset in patients requiring a sequential tumor-bed boost delivered with conventional fractionation, thereby extending the overall treatment duration to 2–2.5 weeks in relatively higher-risk patients. This issue is particularly relevant in China, where breast cancer is diagnosed at a younger age [10] and a higher proportion of patients are candidates for boost radiotherapy. Moreover, the clinical applicability of ultra-hypofractionated schedules remains limited by their sensitivity to dose heterogeneity and potential impact on normal tissue tolerance, particularly in regimens exceeding 5 Gy per fraction [6], [7], [9], [15]. Under this consideration, a 10-fraction whole-breast radiotherapy represents a pragmatic intermediate approach, balancing treatment efficiency with a conservative fraction size and an integrated strategy for boost delivery.

The 10-fraction regimen in the study showed tolerated, with grade 2 acute dermatitis observed in 17.2% of patients and no grade ≥3 events, consistent with the acceptable acute toxicity reported for accelerated hypofractionated schedules [12], [13], [16]. A notable finding was the temporal pattern of skin reactions with peak dermatitis severity most frequently recorded at the 2-week post-treatment assessment rather than at radiotherapy completion. This delayed peak is biologically consistent with the evolution of radiation-induced epidermal injury and highlights the importance of post-treatment evaluation [16]. Interestingly, most acute skin reactions occurred in the nipple–areola complex, a finding seldom reported in prior trials. This region is particularly prone to higher surface dose and mechanical stress, which may explain its greater susceptibility. This finding underscores the importance of hotspot control, improved dose homogeneity, and targeted preventive care in short-course regimens. In addition, explicit dose constraints for the nipple–areola region may be warranted in future hypofractionated breast radiotherapy protocol. These physician-assessed findings were corroborated by patient-reported outcomes indicating only transient worsening of breast symptoms without meaningful deterioration in functional domains or global health status. Collectively, these results suggest that acute effects with this regimen are self-limited and manageable in routine practice.

In exploratory analyses, prior neo-/adjuvant chemotherapy and lower baseline lymphocyte counts—particularly reduced CD4⁺ T cells—were associated with an increased risk of grade ≥2 acute skin toxicity. The association between prior chemotherapy and acute skin toxicity may reflect residual effects on epidermal stem-cell recovery and immune modulation, both of which could exacerbate radiation-induced inflammatory responses. Regarding immune profile, whereas radiation-induced apoptosis of CD8⁺ T lymphocytes has been implicated in late normal-tissue effects such as fibrosis [17], [18], our findings suggest that baseline CD4⁺ T-cell status may be associated with acute radiation dermatitis. Acute radiation dermatitis is primarily driven by epithelial injury and subsequent inflammatory amplification. CD4⁺ T cells constitute a heterogeneous immune compartment that includes regulatory T cells (Tregs), which are involved in controlling cutaneous inflammation and facilitating tissue repair [19], [20]. Although CD4⁺ subsets were not specifically characterized in this study, a lower baseline CD4⁺ T-cell count may reflect reduced immunoregulatory reserve; this interpretation remains hypothesis-generating. In contrast, classical patient-related factors reported in prior studies, such as BMI and breast size [21], were not significant in our cohort. This likely reflects population and treatment-delivery differences, including generally lower BMI and smaller breast volumes in Chinese patients, and the use of hypofractionation with strict dosimetric constraints and hotspot control. We observed a median 32% reduction in total lymphocyte count after radiotherapy, slightly lower than the 40–60% decline commonly reported with conventional fractionation, suggesting that hypofractionated regimens may induce less systemic immune suppression [22]. Prior randomized studies demonstrated that radiation-induced lymphopenia predicts poorer prognosis in breast cancer [22], [23], underscoring the clinical relevance of immune preservation. In addition, heterogeneous changes among lymphocyte subsets were observed, highlighting the complexity of lymphocyte radiosensitivity [24], [25].

This study has several limitations. First, it was a single-center, single-arm analysis with a relatively small sample size, which may limit the generalizability of the findings. Second, the median follow-up time remains relatively short, and longer-term follow-up is required to confirm durable local control and late toxicity outcomes. Third, given the exploratory nature and limited sample size, the statistical power for subgroup analyses was limited, and potential confounding between prior chemotherapy and lymphocyte alterations could not be fully excluded. Fourth, the study population consisted exclusively of Chinese patients, which may limit the generalizability of the findings to other ethnic groups. These limitations warrant confirmation in future multicenter, controlled studies with larger cohorts and longer follow-up periods. Accordingly, the present 10-fraction regimen may represent a pragmatic shortened approach evaluated in a phase II setting, and its role should be further defined in larger prospective studies.

Conclusion

In this single-center phase II study, whole-breast irradiation with 37 Gy in 10 fractions was associated with acceptable acute toxicity. These findings support the feasibility of this shortened regimen after breast-conserving surgery. Continued follow-up and larger prospective studies are needed to further define its long-term oncologic efficacy and late toxicity profile.

Author contributions

YX, JT, SZ, CS and CG contributed equally to this work and share first authorship. YX, JT, SZ, CS and CG contributed to study conceptualization. JT, SZ, CS, CG, YP and YL were responsible for data collection and curation. YX performed data analysis and drafted the manuscript. WW was responsible for funding acquisition. JT and WW contributed to manuscript review and editing. All authors had access to and verify the underlying study data. The final version of the manuscript was read and approved by all authors.

Ethical approval

This study was approved by the Institutional Review Board (IRB) of Peking University Cancer Hospital (2023YJZ30).

Data sharing statement

The datasets used during the current study are available from the corresponding author upon reasonable request.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this manuscript, the authors used ChatGPT to assist with language editing and text refinement. The authors reviewed and edited the content carefully and take full responsibility for the accuracy and integrity of all aspects of the manuscript.

CRediT authorship contribution statement

Yirong Xiang: Conceptualization, Formal analysis, Methodology, Writing – original draft. Jian Tie: Conceptualization, Data curation, Methodology, Writing – review & editing. Siyuan Zhang: Conceptualization, Methodology. Chen Shi: Conceptualization, Data curation, Methodology. Changkuo Guo: Conceptualization, Data curation, Methodology. Yushuo Peng: Data curation. Yuhao Liu: Data curation. Weihu Wang: Funding acquisition, Writing – review & editing.

Funding

This work was supported by the National Key Research and Development Program of China (2025YFC3410204) and the Beijing Hospital Authority’s Ascent Plan (DFL20220902).

Declaration of competing interest

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.

Appendix A. Supplementary data

The following are the Supplementary data to this article: