Dose to the axillary-lateral thoracic vessel junction predicts breast cancer-related lymphedema after postmastectomy radiotherapy: development and temporal validation of NTCP and Nomogram models

Dose to the axillary-lateral thoracic vessel junction predicts breast cancer-related lymphedema after postmastectomy radiotherapy: development and temporal validation of NTCP and Nomogram models | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Dose to the axillary-lateral thoracic vessel junction predicts breast cancer-related lymphedema after postmastectomy radiotherapy: development and temporal validation of NTCP and Nomogram models Nan Xiang, Fang Wu, Chi Zhang, Hongyi Gu, Zhenjun Jin, Chong Yu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8886686/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 27 Apr, 2026 Read the published version in World Journal of Surgical Oncology → Version 1 posted 9 You are reading this latest preprint version Abstract Background Breast cancer-related lymphedema (BCRL) is a disabling late complication after postmastectomy radiotherapy (PMRT). This study evaluated the axillary-lateral thoracic vessel junction (ALTJ) as a functional organ-at-risk (OAR), established its dose-response relationship with BCRL, and developed validated predictive models to guide individualized risk mitigation. Methods 271 patients treated with PMRT from 2019 to 2022 constituted the development cohort, and 45 independent patients treated in 2023 formed the temporal validation cohort. All patients underwent modified radical mastectomy. The ALTJ was contoured on planning CT according to Gross et al. Candidate clinical factors and ALTJ dose–volume histogram (DVH) parameters were analyzed. A normal tissue complication probability (NTCP) model was developed using LASSO-based screening followed by multivariable logistic regression, and a Cox regression–based nomogram was built using multi-method consensus feature selection. Both models were evaluated and validated without refitting in the temporal cohort. Results The 2-year cumulative BCRL incidence was 25.1% in the development and 22.2% in the validation cohort. Multivariable analysis identified the number of dissected lymph nodes (LNDno) and ALTJ V30 as the strongest predictors. The final NTCP model achieved an AUC of 0.816 in the development cohort and 0.860 in the validation cohort, with Brier scores of 0.135 and 0.111, respectively. A clinically actionable risk stratification system was derived using thresholds of LNDno > 13 and ALTJ V30 > 51.75%, identifying high-, moderate-, and low-risk groups with 2-year BCRL rates of 58.8%/54.5%, 26.4%/18.2%, and 5.3%/0% in the development and validation cohorts, respectively. A nomogram, integrating LNDno with ALTJ V25, V30, V35, and Dmean, achieved C-indices of 0.948 and 0.894 in the two cohorts, respectively. Conclusions This study identifies ALTJ V30 and surgical extent as important predictors of BCRL in postmastectomy patients receiving radiotherapy. The findings support the consideration of ALTJ as a quantifiable OAR and provide an evidence-based dose–volume constraint (V30 < 51.75%). The validated NTCP model and nomogram offer practical tools for individualized risk estimation and may inform targeted surveillance and preventive strategies. Breast cancer Lymphedema OAR ALTJ Predictive model Dose–volume histogram Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1 Background Long-term survival rate among patients with breast cancer has improved substantially in recent decades [ 1 ] . As survivorship has increased, greater emphasis has shifted toward preventing treatment-related chronic morbidity. Breast cancer-related lymphedema (BCRL) remains a major, often irreversible complication, causing chronic swelling, pain, functional limitation, and psychological distress [ 2 ] . Reported incidence ranges from 10% to 30% in general survivor populations and may exceed 50% in high-risk subgroups [ 3 , 4 ] . Axillary lymph node dissection (ALND) is a well-established primary risk factor, directly damaging the axillary lymphatic network and associates with a BCRL incidence as high as 28.5% [ 5 ] . Postoperative radiotherapy can further amplify risk through microvascular endothelial injury and progressive fibrosis, thereby compromising residual lymphatic channels and soft-tissue compliance [ 6 ] . Historically, radiotherapy planning and research have focused on dosage of broad anatomical regions, lacking precision in delineating and sparing specific functional substructures within the complex axillary anatomy [ 7 ] . Furthermore, existing predictive models for BCRL rely predominantly on clinical parameters such as age, body mass index (BMI), and nodal status [ 8 , 9 ] , largely omitting critical dosimetric information. Even radiotherapy-specific nomograms rarely incorporate detailed dose-volume data from key axillary substructures [ 10 , 11 ] , limiting their utility for personalized, dosimetry-driven risk assessment. Recent anatomical investigation has highlighted the axillary-lateral thoracic vessel junction (ALTJ) as a critical lymphatic confluence, serving as the principal hub for drainage from axillary Level I to central lymphatics [ 12 ] . This functional role supports the ALTJ as a novel, potential organ-at-risk (OAR) for BCRL. Preliminary studies have suggested a correlation between radiation dose to this area and lymphedema risk [ 13 , 14 ] ; however, these studies are limited by heterogeneous patient cohorts that include both breast-conserving surgery and mastectomy patients, obscuring the risk profile for the most vulnerable population. Accordingly, this study aimed to: 1) definitively establish the dose-response relationship between ALTJ dose-volume histogram (DVH) parameters and BCRL risk in a homogeneous cohort of post-mastectomy patients; 2) develop and independently validate integrative normal tissue complication probability (NTCP) and nomogram models that synergize ALTJ dosimetry with established clinical features; and 3) propose an optimized, evidence-based dose constraint for the ALTJ as a functional OAR to guide precision radiotherapy planning. 2 Materials and Methods 2.1 Study design and population This retrospective, single-center cohort study was approved by the Ethics Committee of Changshu No.1 People’s Hospital, and the requirement for informed consent was waived due to the retrospective design. Consecutive patients treated with postoperative radiotherapy after modified radical mastectomy between January 2019 and December 2022 were included as the development cohort (n = 271) for model building. An independent temporal validation cohort consisted of consecutive patients treated between January and December 2023 (n = 45) were included to assess model generalizability without refitting. Eligibility criteria were: (1) female, age 80; (3) status post modified radical mastectomy; (4) pathologically confirmed invasive breast carcinoma; (5) indication for postoperative radiotherapy defined as tumor size > 5 cm or ≥ 4 positive axillary lymph nodes, or T1–2 disease with 1–3 positive axillary lymph nodes plus high-risk features (fewer than 10 axillary lymph nodes retrieved, lymphovascular tumor emboli, triple-negative breast cancer, or age < 35 years); (6) no contraindications to radiotherapy; and (7) no evidence of upper limb lymphedema prior to radiotherapy. 2.2 Radiotherapy technique and ALTJ delineation All patients underwent CT simulation in the supine position using a dedicated breast board with the affected arm abducted and externally rotated. Radiotherapy plans were generated using either three-dimensional conformal radiotherapy (3D-CRT) or intensity-modulated radiotherapy (IMRT) techniques. The clinical target volumes (CTVs) were delineated according to European Society for Radiotherapy and Oncology (ESTRO) recommendations [ 15 ] and Radiation Therapy Oncology Group (RTOG) consensus guidelines [ 16 ] . Target volumes included the ipsilateral chest wall and supra-/infraclavicular regions; inclusion of the internal mammary drainage region was performed when clinically indicated. A conventional fractionation schedule of 50 Gy delivered in 25 fractions was prescribed. The ALTJ was retrospectively delineated on the original planning CT images by two radiation oncologists, each with more than 10 years of subspecialty experience, in accordance with the definition and contouring guidance proposed by Gross et al. [ 12 ] . The ALTJ was defined as the anatomical confluence where the lateral thoracic vein and subscapular vein drain into the axillary vein, located superior to axillary Level I and inferior to the humeral head. All contours underwent blinded review by a senior radiation oncologist with over 15 years of experience, and discrepancies were resolved by consensus. 2.3 Lymphedema assessment and follow-up The primary endpoint was the development of clinically significant upper limb BCRL. Diagnosis was based on serial circumferential measurements, defined as a persistent difference of ≥ 2 cm between the affected and contralateral limbs at any of the standard measurement points: 5 cm and 15 cm proximal to the olecranon process, or 10 cm distal to it [ 17 ] . Assessments were performed at the completion of radiotherapy and subsequently at each 3-month follow-up visit. The maximum follow-up was 28 months after radiotherapy. Patients without lymphedema were censored at the date of last follow-up. 2.4 Clinical and dosimetric variables Based on prior literature [ 8 – 11 ] , four clinical covariates were prespecified as candidate predictors: body mass index (BMI), number of dissected lymph nodes (LNDno), number of pathologically positive lymph nodes, and receipt of chemotherapy. ALTJ dosimetric parameters were extracted from the treatment planning system and included maximum dose (Dmax), minimum dose (Dmin), mean dose (Dmean), and Vx parameters (V5, V10, V15, V20, V25, V30, V35, V40, V45, and V50), where Vx denotes the percentage volume of ALTJ receiving ≥ x Gy. 2.5 Statistical analysis Patient, tumor and treatment characteristics were summarized using descriptive statistics. Continuous variables were reported as median with interquartile range (IQR) or mean ± standard deviation (SD) as appropriate, and categorical variables were reported as frequencies and percentages. Differences in baseline characteristics between the development and validation cohorts were compared using the Pearson’s χ² test or Fisher’s exact test for categorical variables, and the Mann-Whitney U test for continuous variables. Predictive performance for binary outcomes (lymphedema vs. no lymphedema) was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC). For time-to-event outcomes, Harrell’s C-index was used. Calibration was assessed with the Hosmer-Lemeshow goodness-of-fit test (P > 0.05 indicating good fit). Kaplan–Meier methods were used to estimate cumulative BCRL incidence across risk strata, with comparisons by log-rank test. Statistical analyses were performed using SPSS (version 27.0.1.0) and R (version 4.5.1), with a two-sided P < 0.05 considered statistically significant. 2.6 Model development and validation In accordance with TRIPOD principles [ 18 ] , two distinct prediction models were developed using the development cohort and subsequently validated in the independent temporal validation cohort. 2.6.1 NTCP model Least absolute shrinkage and selection operator (LASSO) regression was used for initial variable screening among candidate clinical and dosimetric predictors. Variables meeting prespecified significance criteria were entered into a multivariable logistic regression framework. Forward stepwise selection guided by the Akaike information criterion (AIC) was then performed: the subset with the lowest AIC was selected as the final NTCP model estimating probability of BCRL. 2.6.2 Nomogram Model For time-to-event prediction, candidate predictors were screened using a consensus strategy comprising three approaches to ensure robustness: (1) variables with P < 0.05 in univariable Cox regression; (2) significant factors identified by the LASSO regression; and (3) features identified as important by the Boruta algorithm, a wrapper method built around a random forest classifier. Variables confirmed by all three approaches were included in the final multivariable Cox model. A nomogram was constructed based on this final model to predict the 2-year cumulative lymphedema incidence rate. 3 Results 2.7 Patient characteristics A total of 316 patients were included (development cohort, n = 271; independent validation cohort, n = 45). Baseline characteristics were generally balanced between cohorts (Table 1 ). In the development and validation cohorts, mean age was 56.63 ± 9.64 years and 56.58 ± 10.31 years, respectively. Mean BMI was 23.97 ± 3.27 kg/m² and 24.53 ± 3.12 kg/m², respectively. Mean LNDno was 13.50 ± 4.68 and 13.78 ± 2.63, and mean number of positive lymph nodes was 4.22 ± 5.04 and 4.84 ± 4.94, respectively. No significant differences were observed in tumor laterality, T/N stage, hormone receptor status, or other key variables. Table 1 Baseline patient, tumor, and treatment characteristics of the patient cohorts Development Cohort N = 271 1 Validation Cohort N = 45 1 P-value 2 Age 56.63(9.64) 56.58 (10.31) 0.967 Location 0.138 Left 145 (53.5%) 30 (66.7%) Right 126 (46.5%) 15 (33.3%) T Stage 0.154 1 104 (38.4%) 11 (24.4%) 2 141 (52.0%) 27 (60.0%) 3 24 (8.9%) 6 (13.3%) 4 2 (0.7%) 1 (2.2%) N Stage 0.679 0 21 (7.7%) 2 (4.4%) 1 153 (56.5%) 23 (51.1%) 2 65 (24.0%) 13 (28.9%) 3 32 (11.8%) 7 (15.6%) Stage 0.337 1 3 (1.1%) 0 (0%) 2 160 (59.0%) 22 (48.9%) 3 108 (39.9%) 23 (51.1%) Number of Dissected Lymph Nodes 13.50 (4.68) 13.78 (2.63) 0.565 Number of Pathologically Positive Lymph Nodes 4.22 (5.04) 4.84 (4.94) 0.438 Chemotherapy 262(96.7%) 43(95.6%) 0.824 ER 0.636 Negative 95 (35.1%) 18 (40.0%) Positive 176 (64.9%) 27 (60.0%) PR > 0.9 Negative 133 (49.1%) 22 (48.9%) Positive 138 (50.9%) 23 (51.1%) Her2 0.743 Negative 173 (63.8%) 27 (60.0%) Positive 98 (36.2%) 18 (40.0%) Ki-67 > 0.9 ≤ 5 20 (7.4%) 3 (6.7%) 5 ~ 30 111 (41.0%) 18 (40.0%) >30 140 (51.6%) 24 (53.3%) BMI(kg/m 2 ) 23.97 (3.27) 24.53 (3.12) 0.271 BMI Group 0.232 ˃ 23.9 133 (49.1%) 27 (60.0%) ≤ 23.9 138 (50.9%) 18 (40.0%) Lymphedema Event 68 (25.1%) 10 (22.2%) 0.821 1 Mean (SD); n (%). 2 Wilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test Abbreviations: BMI=body mass index; SD= standard deviation. 2.8 Lymphedema incidence In the development cohort, 68 patients (25.1%) developed BCRL during follow-up, with a median time to onset of 13.5 months. The cumulative incidence rates at 12 and 24 months were 11.4% and 25.1%, respectively. In the validation cohort, 10 patients (22.2%) developed BCRL (median time: 10.5 months), with 12- and 24-month cumulative incidences of 13.3% and 22.2%. The difference in 24-month incidence between the two cohorts was not statistically significant (P = 0.821). 2.9 ALTJ dosimetric parameters Dosimetric parameters for ALTJ are detailed in Table 2 . In the development cohort, the median ALTJ volume was 14.9 cm³ (IQR: 13.3–17.7). Median Dmax, Dmean, and Dmin were 49.0 Gy (46.2–52.1), 26.8 Gy (19.6–33.6), and 2.6 Gy (1.7–8.6), respectively. Median ALTJ V30 was 41.4% (26.2–60.0). In the validation cohort, the corresponding values were 12.2 cm³ (9.8–15.1), 49.6 Gy (48.2–50.6), 30.1 Gy (26.7–33.1), 10.2 Gy (6.0–17.9), and 49.2% (41.5–54.6), respectively. Overall distributions of ALTJ DVH metrics were broadly similar between cohorts, although selected metrics (including ALTJ volume and Dmin) differed. Table 2 ALTJ dosimetric parameters in the development and validation cohorts Development Cohort N = 271 1 Validation Cohort N = 45 1 P-value 2 ALTJ volume(cm 3 ) 14.9 (13.3, 17.7) 12.2 (9.8, 15.1) < 0.001 ALTJ Dmax(Gy) 49.0 (46.2, 52.1) 49.6 (48.2, 50.6) 0.595 ALTJ Dmean(Gy) 26.8 (19.6, 33.6) 30.1 (26.7, 33.1) 0.052 ALTJ Dmin(Gy) 2.6 (1.7, 8.6) 10.2 (6.0, 17.9) < 0.001 ALTJ V5(%) 79.9 (69.3, 95.9) 84.5 (81.9, 90.0) 0.051 ALTJ V10(%) 73.6 (57.4, 91.5) 78.4 (74.7, 79.8) 0.076 ALTJ V15(%) 68.8 (52.4, 86.8) 71.9 (69.9, 75.3) 0.124 ALTJ V20(%) 62.7 (46.9, 81.3) 67.2 (64.1, 70.0) 0.105 ALTJ V25(%) 52.1 (35.6, 68.5) 56.2 (51.4, 61.2) 0.126 ALTJ V30(%) 41.4 (26.2, 60.0) 49.2 (41.5, 54.6) 0.111 ALTJ V35(%) 43.4 (26.6, 57.3) 44.4 (40.5, 50.3) 0.613 ALTJ V40(%) 34.2 (17.3, 45.4) 36.2 (28.7, 45.7) 0.294 ALTJ V45(%) 21.2 (4.0, 29.4) 21.5 (13.8, 30.4) 0.231 ALTJ V50(%) 10.6 (1.5, 15.4) 5.5 (2.4, 11.2) 0.031 1 Median (IQR); 2 Wilcoxon rank sum test Abbreviations: ALTJ= axillary-lateral thoracic vessel junction; Dmax= maximum dose; Dmin=minimum dose; Dmean=mean dose; Vx denotes the percentage volume of ALTJ receiving ≥ x Gy; IQR= interquartile range. 2.10 Model performance 2.10.1 NTCP model Univariable analysis followed by LASSO regression identified LNDno, ALTJ V30, ALTJ Dmean, and ALTJ V25 as significant predictors. In multivariable logistic regression, ALTJ V30 (OR = 3.47, 95% CI: 2.00–6.00, P < 0.001), LNDno (OR = 1.55, 95% CI: 1.13–2.21, P < 0.001) and ALTJ V25 (OR = 0.322, 95% CI: 0.194–0.532, P < 0.001) were significantly associated with BCRL and remained independent predictors ( Fig. 1 ). LNDno (OR = 1.30, 95% CI: 1.19–1.43, P < 0.001) and ALTJ V30 (OR = 1.03, 95% CI: 1.02–1.05, P < 0.001) were used as 2 highly predictive variable to develop the NTCP model. The final NTCP model was: $$\:{NTCP}_{Logit}=\frac{1}{1+{e}^{-Logit\left(P\right)}}$$ Logit(P) = 1.03×ALTJ V30 + 1.30×LNDno − 6.53 Figure 1. LASSO regression multi-factor importance analysis The model demonstrated robust performance with an AUC of 0.816 and a Brier score of 0.135 in the development cohort. In the independent validation cohort, the model maintained high predictive accuracy with an AUC of 0.860 and a Brier score of 0.111, supporting its external validity (Fig. 2 ). For clinical implementation, the optimal cut-off was determined by the Youden index. Patients were stratified into three risk groups based on LNDno and ALTJ V30: (1) High risk: LNDno > 13 and ALTJ V30 > 51.75%; (2) Moderate risk: one of the two factors present; (3) Low risk: neither factor present. In the development cohort, the 2-year BCRL incidence was 58.8% in the high-risk group, 26.4% in the moderate-risk group, and 5.3% in the low-risk group. The corresponding rates in the validation cohort were 54.5%, 18.2%, and 0%, respectively. Kaplan–Meier curves confirmed a significantly different cumulative incidence of lymphedema among these groups. (log-rank P < 0.001, Fig. 3 ). 2.10.2 Nomogram model After consensus feature selection (including LASSO and Boruta), LNDno and multiple ALTJ DVH parameters (including Dmean, V25, V30, V35, and V50) were evaluated in multivariable Cox regression (Fig. 4 ). The final predictors used to construct the nomogram for estimating 2-year BCRL risk were LNDno (HR = 0.99, 95%CI: 0.95–1.04), ALTJ Dmean (HR = 1.13, 95%CI: 1.03–1.23), ALTJ V25 (HR = 0.75, 95%CI: 0.69–0.81), ALTJ V30 (HR = 1.54, 95%CI: 1.41–1.68), and ALTJ V35 (HR = 0.85, 95%CI: 0.81–0.89) (Fig. 5 ). The nomogram demonstrated excellent discrimination, with a C-index of 0.948 in the development cohort and 0.894 in the validation cohort. Moreover, decision curve analysis indicated a favorable net benefit, and the calibration plots showed good agreement between predicted and observed risk (Fig. 6 ). 4 Discussion In this homogeneous cohort of patients treated with modified radical mastectomy and followed by postoperative radiotherapy, we identified a synergistic relationship between surgical extent and functionally targeted axillary dosimetry in predicting BCRL. Specifically, the number of dissected lymph nodes (LNDno) and ALTJ V30 emerged as independent, dominant predictors and formed the basis of a parsimonious, clinically interpretable NTCP model that retained strong performance in an independent temporal validation cohort. We further developed a clinical–dosimetric nomogram integrating LNDno with multiple ALTJ intermediate-dose metrics (Dmean, V25, V30, V35), achieving high discrimination and demonstrable clinical utility. Importantly, risk stratification based on two readily accessible thresholds (ALTJ V30 > 51.75% and LNDno > 13) delineated a marked gradient in 2-year BCRL risk. This approach identifies a patient subgroup for whom targeted preventive strategies and ALTJ-focused plan optimization may be particularly impactful. A key strength of the present study is its exclusive focus on patients treated with modified radical mastectomy. Prior ALTJ-related studies often included heterogeneous cohorts combining breast-conserving surgery and mastectomy [ 12 – 14 ] . Because breast-conserving surgery typically preserves more axillary lymphatic architecture, baseline lymphatic reserve—and therefore BCRL susceptibility—differs substantially from that of patients undergoing more extensive axillary surgery [ 19 – 21 ] . By restricting the cohort to modified radical mastectomy, our analysis targets a clinically relevant setting of substantial baseline lymphatic disruption. Within this context, the ALTJ—conceptualized as a critical lymphatic confluence for axillary Level I drainage [ 12 ] —may represent a disproportionately important residual pathway. Radiation-related endothelial injury and fibrosis at this junction, even at the intermediate-dose range, could plausibly impair lymphatic decompensation. Our findings support the "dual-hit" pathophysiological model, where the initial surgical insult is critically exacerbated by subsequent radiation injury to this pivotal structure [ 22 , 23 ] . The stability and interpretability of the combined LNDno–ALTJ V30 signal in our models may reflect this biologically and anatomically constrained vulnerability. Our findings also refine that how radiotherapy dose should be conceptualized in relation to BCRL. Historically, many studies have focused on dose to broad axillary anatomical regions (Levels I–III) [ 24 ] . This approach, by averaging dose over large volumes, may mask clinically significant heterogeneity within functionally critical substructures. By delineating the ALTJ as a distinct OAR and evaluating its DVH identified, we identified ALTJ V30 as a robust, clinically relevant predictor in the NTCP model. While prior studies by Park et al. [ 14 ] and Kim et al. [ 13 ] identified correlations with parameters like V35 or Dmax, we identify V30 as the dominant factor, particularly in a mastectomy-only cohort, thereby refining the understanding of the dose-risk relationship. The clinical impact of this work is twofold. First, the two-parameter stratification (LNDno and ALTJ V30) provides an immediately implementable approach to identify patients at high risk (2-year risk > 54%), in whom early physiotherapy referral, structured surveillance, and preventive counseling could be prioritized. Second, the proposed dose constraint—maintaining ALTJ V30 below 51.75%—provides a clear, quantitative objective for the radiotherapy planner. Using techniques such as IMRT or proton therapy to achieve this constraint while preserving target coverage provides a tangible strategy for reducing the risk of BCRL. Furthermore, the nomogram provides additional nuance by incorporating multiple intermediate-dose metrics (V25, V30, V35) and Dmean, which collectively suggest a continuous dose–response relationship across approximately 25–35 Gy rather than an exclusively threshold-driven phenomenon. This continuous-risk profile likely explains why the multivariable nomogram provides better discrimination than the simpler, clinical-only tools reported in prior studies [ 25 ] . Methodologically, we strengthened predictor identification by combining conventional regression screening with LASSO and Boruta selection. This multi-method consensus strategy was designed to improve feature stability in the presence of collinearity among DVH parameters and to limit overfitting when evaluating high-dimensional dosimetric feature sets [ 18 , 26 ] . The maintenance of performance in an independent temporally cohort, without refitting, supports the transportability of these models within comparable PMRT practice settings. Several limitations should be acknowledged. Firstly, the retrospective, single-center design limits causal inference and introduces potential selection bias. Although the use of a temporally independent validation cohort is valuable, it cannot fully address the fundamental constraints of the retrospective approach. Secondly, the sample size—particularly in the validation cohort—was modest; therefore, multicenter validation across institutions and involving patient populations with diverse racial backgrounds will be essential to establish broader generalizability. Thirdly, BCRL was defined using circumferential measurements, a clinically standard method [ 17 ] , which may be less sensitive than bioimpedance spectroscopy (BIS) for detecting subclinical lymphedema [ 27 ] . Future studies incorporating BIS-based endpoints, together with emerging strategies such as radiomics [ 28 ] and circulating biomarkers [ 29 ] , may enable earlier risk identification and facilitate further refinement of ALTJ constraints and risk-adapted preventive interventions. 5 Conclusions In conclusion, this study establishes the ALTJ as a pivotal functional OAR, whose radiation dose, specifically the V30, interacts synergistically with surgical extent to determine BCRL risk following mastectomy. We developed and validated an NTCP model and a Cox-based nomogram in an independent temporal cohort, providing clinically implementable tools that translating ALTJ dosimetry into individualized risk estimates. This evidence translates into a specific planning constraint: limiting ALTJ V30 to < 51.75%. Collectively, these findings represent a pivotal advancement towards function-preserving breast cancer radiotherapy. Prospective trials are warranted to confirm that adherence to these dosimetric principles yield measurable improvements in patient-reported lymphedema outcomes without compromising oncologic efficacy. Abbreviations BCRL breast cancer-related lymphedema PMRT postmastectomy radiotherapy ALND axillary lymph node dissection BMI body mass index ALTJ axillary-lateral thoracic vessel junction OAR organ at risk DVH dose-volume histogram NTCP normal tissue complication probability KPS Karnofsky performance status 3D-CRT three-dimensional conformal radiotherapy IMRT intensity-modulated radiotherapy CTV clinical target volume ESTRO European Society for Radiotherapy and Oncology RTOG Radiation Therapy Oncology Group LNDno number of dissected lymph nodes Dmax maximum dose Dmin minimum dose Dmean mean dose IQR interquartile range SD standard deviation ROC receiver operating characteristic AUC area under the ROC curve LASSO least absolute shrinkage and selection operator AIC Akaike information criterion BIS bioimpedance spectroscopy Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of Changshu No.1 People’s Hospital, and the requirement for informed consent was waived due to the retrospective design. All procedures performed in this study were in accordance with the ethical standards of the institutional research committee and the principles of the Declaration of Helsinki. Consent for publication Not applicable. Availability of data and materials The data are available from the corresponding author on reasonable request. Competing interests The authors declare no competing interests. Funding This study was approved by Changshu Health Commission Science and Technology Plan Sponsored Projects and Changshu Science and Technology Development Plan (Medical and Health) Guidance Projects (CSWS202304) Authors' contributions N.X and C.Y. performed data statistics and drafted the main manuscript. F.W., C.Z. and H.G. were involved in data collection and data statistics. Z.J. reviewed the target regions, and participated in the revision and review of the manuscript. Acknowledgements Not applicable. References Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49. Miller KD, Nogueira L, Devasia T, et al. Cancer treatment and survivorship statistics, 2022. CA Cancer J Clin. 2022;72(5):409–36. Shen A, Lu Q, Fu X, et al. Risk factors of unilateral breast cancer-related lymphedema: an updated systematic review and meta-analysis of 84 cohort studies. Support Care Cancer. 2022;31(1):18. Published 2022 Dec 14. Naoum GE, Roberts S, Brunelle CL, et al. Quantifying the Impact of Axillary Surgery and Nodal Irradiation on Breast Cancer-Related Lymphedema and Local Tumor Control: Long-Term Results From a Prospective Screening Trial. J Clin Oncol. 2020;38(29):3430–8. Ahn HR, Jeong HE, Jeong C, et al. Incidence and risk factors of breast cancer-related lymphedema in Korea: a nationwide retrospective cohort study. Int J Surg. 2024;110(6):3518–26. Kwon S, Janssen CF, Velasquez FC, et al. Radiation Dose-Dependent Changes in Lymphatic Remodeling. Int J Radiat Oncol Biol Phys. 2019;105(4):852–60. Lin Y, Xu Y, Wang C, et al. Loco-regional therapy and the risk of breast cancer-related lymphedema: a systematic review and meta-analysis. Breast Cancer. 2021;28(6):1261–72. Gross JP, Whelan TJ, Parulekar WR, et al. Development and Validation of a Nomogram to Predict Lymphedema After Axillary Surgery and Radiation Therapy in Women With Breast Cancer From the NCIC CTG MA.20 Randomized Trial. Int J Radiat Oncol Biol Phys. 2019;105(1):165–73. Li MM, Wu PP, Qiang WM, et al. Development and validation of a risk prediction model for breast cancer-related lymphedema in postoperative patients with breast cancer. Eur J Oncol Nurs. 2023;63:102258. Luo X, Ye J, Xiao T, et al. Development and validation of a predictive nomogram for postoperative upper limb lymphedema in breast cancer patients: a retrospective cohort study. Sci Rep. 2025;15(1):24609. Byun HK, Chang JS, Im SH, et al. Risk of Lymphedema Following Contemporary Treatment for Breast Cancer: An Analysis of 7617 Consecutive Patients From a Multidisciplinary Perspective. Ann Surg. 2021;274(1):170–8. Gross JP, Lynch CM, Flores AM, et al. Determining the Organ at Risk for Lymphedema After Regional Nodal Irradiation in Breast Cancer. Int J Radiat Oncol Biol Phys. 2019;105(3):649–58. Suk Chang J, Ko H, Hee Im S, et al. Incorporating axillary-lateral thoracic vessel juncture dosimetric variables improves model for predicting lymphedema in patients with breast cancer: A validation analysis. Clin Transl Radiat Oncol. 2023;41:100629. Park YI, Chang JS, Ko H, et al. Development and Validation of a Normal Tissue Complication Probability Model for Lymphedema After Radiation Therapy in Breast Cancer. Int J Radiat Oncol Biol Phys. 2023;116(5):1218–25. Offersen BV, Boersma LJ, Kirkove C, et al. ESTRO consensus guideline on target volume delineation for elective radiation therapy of early stage breast cancer. Radiother Oncol. 2015;114(1):3–10. Breast cancer atlas for radiation therapy planning. Consensus definitions. Available at: https://www.rtog.org/corelab/contouringatlases/breastcanceratlas.aspx . Accessed January 31, 2026. Position Statement of the National Lymphedema Network. Topic:Screening and Measurement for Early Detection of Breast Cancer Related Lymphedema. Available at: https://lymphnet.org/positionpapers . Accessed January 31, 2026. Collins GS, Moons KGM, Dhiman P, et al. TRIPOD + AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. Published 2024 Apr 16. Giuliano AE, Ballman KV, McCall L, et al. Effect of Axillary Dissection vs No Axillary Dissection on 10-Year Overall Survival Among Women With Invasive Breast Cancer and Sentinel Node Metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial. JAMA. 2017;318(10):918–26. Galimberti V, Cole BF, Viale G, et al. Axillary dissection versus no axillary dissection in patients with breast cancer and sentinel-node micrometastases (IBCSG 23 – 01): 10-year follow-up of a randomised, controlled phase 3 trial. Lancet Oncol. 2018;19(10):1385–93. Whelan TJ, Olivotto IA, Parulekar WR, et al. Regional Nodal Irradiation in Early-Stage Breast Cancer. N Engl J Med. 2015;373(4):307–16. 10.1056/NEJMoa1415340 . Hong SW, Kim EK, Shin HC, et al. Relationship of Immediate Breast Reconstruction and the Development of Lymphedema in Breast Cancer Patients With Radiation Therapy. Int J Radiat Oncol Biol Phys. 2025;123(3):753–64. Rockson SG. Lymphedema after Breast Cancer Treatment. N Engl J Med. 2018;379(20):1937–44. Almahariq MF, Maywood MJ, Levitin RB, et al. Mapping of Metastatic Level I Axillary Lymph Nodes in Patients with Newly Diagnosed Breast Cancer. Int J Radiat Oncol Biol Phys. 2020;106(4):811–20. Jiang Q, Hu H, Liao J, et al. Development and validation of a nomogram for breast cancer-related lymphedema. Sci Rep. 2024;14(1):15602. Kursa MB, Rudnicki WR. Feature Selection with the Boruta Package. J Stat Softw. 2010;36(11):1–13. Shah C, Boyages J, Koelmeyer L, et al. Timing of Breast Cancer Related Lymphedema Development Over 3 Years: Observations from a Large, Prospective Randomized Screening Trial Comparing Bioimpedance Spectroscopy (BIS) Versus Tape Measure. Ann Surg Oncol. 2024;31(11):7487–95. Ar GK. J, R H, Breast Cancer-Related Lymphedema: Magnetic Resonance Imaging Evidence of Sparing Centered Along the Cephalic Vein. J Reconstr Microsurg. 2021;37(6). Visser J, van Geel M, Cornelissen AJM, et al. Breast Cancer-Related Lymphedema and Genetic Predisposition: A Systematic Review of the Literature. Lymphat Res Biol. 2019;17(3):288–93. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 27 Apr, 2026 Read the published version in World Journal of Surgical Oncology → Version 1 posted Editorial decision: Revision requested 21 Mar, 2026 Reviews received at journal 15 Mar, 2026 Reviews received at journal 12 Mar, 2026 Reviewers agreed at journal 12 Mar, 2026 Reviewers agreed at journal 12 Mar, 2026 Reviewers invited by journal 12 Mar, 2026 Editor assigned by journal 18 Feb, 2026 Submission checks completed at journal 16 Feb, 2026 First submitted to journal 15 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8886686","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":606476500,"identity":"f0a1eb34-f660-4ee6-aac9-d39a46af2c05","order_by":0,"name":"Nan Xiang","email":"","orcid":"","institution":"Changshu Hospital Affiliated to Soochow University, Changshu No.1 People’s Hospital","correspondingAuthor":false,"prefix":"","firstName":"Nan","middleName":"","lastName":"Xiang","suffix":""},{"id":606476501,"identity":"53ccf187-9e09-4c2a-90da-7afe44824fd1","order_by":1,"name":"Fang Wu","email":"","orcid":"","institution":"Changshu Hospital Affiliated to Soochow University, Changshu No.1 People’s Hospital","correspondingAuthor":false,"prefix":"","firstName":"Fang","middleName":"","lastName":"Wu","suffix":""},{"id":606476503,"identity":"11b4c0a6-fdc7-49cd-8e5b-07ccc83615e4","order_by":2,"name":"Chi Zhang","email":"","orcid":"","institution":"Changshu Hospital Affiliated to Soochow University, Changshu No.1 People’s Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chi","middleName":"","lastName":"Zhang","suffix":""},{"id":606476505,"identity":"b1c393b6-8f79-4f79-ba68-817361bf79d8","order_by":3,"name":"Hongyi Gu","email":"","orcid":"","institution":"Suzhou Health Vocational and Technical College","correspondingAuthor":false,"prefix":"","firstName":"Hongyi","middleName":"","lastName":"Gu","suffix":""},{"id":606476507,"identity":"d46bacc0-4e8e-4fef-952c-acdab353db58","order_by":4,"name":"Zhenjun Jin","email":"","orcid":"","institution":"Changshu Hospital Affiliated to Soochow University, Changshu No.1 People’s Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhenjun","middleName":"","lastName":"Jin","suffix":""},{"id":606476509,"identity":"ed0aea71-581d-46b1-a103-f60b2bca9b43","order_by":5,"name":"Chong Yu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYBACCQbmBiB1AIiZDxz48IMoLYwwLWyJB2f2kKaFx/gwBxsRWiRnJDY+5vlzR86cf82Hwww8DPL8Ygfwa5GWSGw25m17Zmw54+2GwwUWDIYzZyfg1yInkdgmzdtwOHHDjbMbDs/gYUgwuE2MFp4/h+s33Djz4DAPGxFapMFa2A4nGJzvYSBOi2TPw2bDuW2HDTfcYDMABrIEYb9IHE8++ODNn8PyBucPP/7w4YeNPL80AS1ImsEqJYhVDgL8B0hRPQpGwSgYBSMJAADiLEviPvyEGQAAAABJRU5ErkJggg==","orcid":"","institution":"Changshu Hospital Affiliated to Soochow University, Changshu No.1 People’s Hospital","correspondingAuthor":true,"prefix":"","firstName":"Chong","middleName":"","lastName":"Yu","suffix":""}],"badges":[],"createdAt":"2026-02-15 14:38:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8886686/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8886686/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12957-026-04372-w","type":"published","date":"2026-04-27T15:56:50+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":104872687,"identity":"49abb3e2-b284-4717-9606-47f315b40c94","added_by":"auto","created_at":"2026-03-18 08:22:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44613,"visible":true,"origin":"","legend":"\u003cp\u003eLASSO regression multi-factor importance analysis\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/9baeb7eeaa95c333ebb86b3a.png"},{"id":104872683,"identity":"58b03b42-fcde-4b72-b194-06ed4670824a","added_by":"auto","created_at":"2026-03-18 08:22:40","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":65358,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves of NTCP model in development and validation cohorts\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/ba7060606089367322a3994b.png"},{"id":104872686,"identity":"90bd7986-2a9e-4c98-ab9a-3aa3b335a34e","added_by":"auto","created_at":"2026-03-18 08:22:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":93361,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative incidence of lymphedema stratified by clinical and dosimetric risk factors. High risk: LNDno \u0026gt; 13 and ALTJ V30 \u0026gt; 51.75%, Moderate risk: LNDno \u0026gt; 13, ALTJ V30 ≤ 51.75% or LNDno ≤ 13, ALTJ V30 \u0026gt; 51.75%, Low risk: LNDno ≤ 13, ALTJ V30 ≤ 51.75%. ALTJ= axillary-lateral thoracic vessel junction, ALTJ V30= percentage volume of ALTJ receiving ≥ 30 Gy.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/c95add8a3e3b30d78436d72e.png"},{"id":104872685,"identity":"6bb67711-f7fa-4a6c-88dd-b0174ee94392","added_by":"auto","created_at":"2026-03-18 08:22:40","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":175795,"visible":true,"origin":"","legend":"\u003cp\u003eA: Factors selected by LASSO regression. B: Importance distribution of factors identified by the Boruta algorithm. C: Results of the multivariate Cox regression model.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/8440bfcc9c0079e08dc8139a.png"},{"id":104872682,"identity":"6f5c9e36-014a-445f-93c6-d9513f7475ab","added_by":"auto","created_at":"2026-03-18 08:22:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":48201,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram model\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/d4b1fde1e3121e3bffe0cc24.png"},{"id":104872684,"identity":"e4177b5b-6bdf-4e5d-ab8e-7d9a94823703","added_by":"auto","created_at":"2026-03-18 08:22:40","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":95558,"visible":true,"origin":"","legend":"\u003cp\u003eA: ROC curve of the nomogram model. B: Decision curve analysis (DCA) curve. C: Calibration curve in the development cohort. D: Calibration curve in the validation cohort.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/ecd73de8d0b0317a49127c7e.png"},{"id":108438453,"identity":"2703f0de-4998-4a59-b34c-422584f264f3","added_by":"auto","created_at":"2026-05-04 16:09:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":870228,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8886686/v1/3a6e9fcc-186a-4b37-bce0-f717eb1d2132.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dose to the axillary-lateral thoracic vessel junction predicts breast cancer-related lymphedema after postmastectomy radiotherapy: development and temporal validation of NTCP and Nomogram models","fulltext":[{"header":"1 Background","content":"\u003cp\u003eLong-term survival rate among patients with breast cancer has improved substantially in recent decades\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. As survivorship has increased, greater emphasis has shifted toward preventing treatment-related chronic morbidity. Breast cancer-related lymphedema (BCRL) remains a major, often irreversible complication, causing chronic swelling, pain, functional limitation, and psychological distress\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Reported incidence ranges from 10% to 30% in general survivor populations and may exceed 50% in high-risk subgroups\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAxillary lymph node dissection (ALND) is a well-established primary risk factor, directly damaging the axillary lymphatic network and associates with a BCRL incidence as high as 28.5%\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. Postoperative radiotherapy can further amplify risk through microvascular endothelial injury and progressive fibrosis, thereby compromising residual lymphatic channels and soft-tissue compliance\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e. Historically, radiotherapy planning and research have focused on dosage of broad anatomical regions, lacking precision in delineating and sparing specific functional substructures within the complex axillary anatomy\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Furthermore, existing predictive models for BCRL rely predominantly on clinical parameters such as age, body mass index (BMI), and nodal status\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e, largely omitting critical dosimetric information. Even radiotherapy-specific nomograms rarely incorporate detailed dose-volume data from key axillary substructures\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e, limiting their utility for personalized, dosimetry-driven risk assessment.\u003c/p\u003e \u003cp\u003eRecent anatomical investigation has highlighted the axillary-lateral thoracic vessel junction (ALTJ) as a critical lymphatic confluence, serving as the principal hub for drainage from axillary Level I to central lymphatics\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. This functional role supports the ALTJ as a novel, potential organ-at-risk (OAR) for BCRL. Preliminary studies have suggested a correlation between radiation dose to this area and lymphedema risk\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e; however, these studies are limited by heterogeneous patient cohorts that include both breast-conserving surgery and mastectomy patients, obscuring the risk profile for the most vulnerable population.\u003c/p\u003e \u003cp\u003eAccordingly, this study aimed to: 1) definitively establish the dose-response relationship between ALTJ dose-volume histogram (DVH) parameters and BCRL risk in a homogeneous cohort of post-mastectomy patients; 2) develop and independently validate integrative normal tissue complication probability (NTCP) and nomogram models that synergize ALTJ dosimetry with established clinical features; and 3) propose an optimized, evidence-based dose constraint for the ALTJ as a functional OAR to guide precision radiotherapy planning.\u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study design and population\u003c/h2\u003e \u003cp\u003e This retrospective, single-center cohort study was approved by the Ethics Committee of Changshu No.1 People\u0026rsquo;s Hospital, and the requirement for informed consent was waived due to the retrospective design. Consecutive patients treated with postoperative radiotherapy after modified radical mastectomy between January 2019 and December 2022 were included as the development cohort (n\u0026thinsp;=\u0026thinsp;271) for model building. An independent temporal validation cohort consisted of consecutive patients treated between January and December 2023 (n\u0026thinsp;=\u0026thinsp;45) were included to assess model generalizability without refitting.\u003c/p\u003e \u003cp\u003eEligibility criteria were: (1) female, age\u0026thinsp;\u0026lt;\u0026thinsp;80 years; (2) Karnofsky performance status (KPS)\u0026thinsp;\u0026gt;\u0026thinsp;80; (3) status post modified radical mastectomy; (4) pathologically confirmed invasive breast carcinoma; (5) indication for postoperative radiotherapy defined as tumor size\u0026thinsp;\u0026gt;\u0026thinsp;5 cm or \u0026ge;\u0026thinsp;4 positive axillary lymph nodes, or T1\u0026ndash;2 disease with 1\u0026ndash;3 positive axillary lymph nodes plus high-risk features (fewer than 10 axillary lymph nodes retrieved, lymphovascular tumor emboli, triple-negative breast cancer, or age\u0026thinsp;\u0026lt;\u0026thinsp;35 years); (6) no contraindications to radiotherapy; and (7) no evidence of upper limb lymphedema prior to radiotherapy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Radiotherapy technique and ALTJ delineation\u003c/h2\u003e \u003cp\u003eAll patients underwent CT simulation in the supine position using a dedicated breast board with the affected arm abducted and externally rotated. Radiotherapy plans were generated using either three-dimensional conformal radiotherapy (3D-CRT) or intensity-modulated radiotherapy (IMRT) techniques. The clinical target volumes (CTVs) were delineated according to European Society for Radiotherapy and Oncology (ESTRO) recommendations\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e and Radiation Therapy Oncology Group (RTOG) consensus guidelines\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. Target volumes included the ipsilateral chest wall and supra-/infraclavicular regions; inclusion of the internal mammary drainage region was performed when clinically indicated. A conventional fractionation schedule of 50 Gy delivered in 25 fractions was prescribed.\u003c/p\u003e \u003cp\u003eThe ALTJ was retrospectively delineated on the original planning CT images by two radiation oncologists, each with more than 10 years of subspecialty experience, in accordance with the definition and contouring guidance proposed by Gross et al.\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. The ALTJ was defined as the anatomical confluence where the lateral thoracic vein and subscapular vein drain into the axillary vein, located superior to axillary Level I and inferior to the humeral head. All contours underwent blinded review by a senior radiation oncologist with over 15 years of experience, and discrepancies were resolved by consensus.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Lymphedema assessment and follow-up\u003c/h2\u003e \u003cp\u003eThe primary endpoint was the development of clinically significant upper limb BCRL. Diagnosis was based on serial circumferential measurements, defined as a persistent difference of \u0026ge;\u0026thinsp;2 cm between the affected and contralateral limbs at any of the standard measurement points: 5 cm and 15 cm proximal to the olecranon process, or 10 cm distal to it\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e. Assessments were performed at the completion of radiotherapy and subsequently at each 3-month follow-up visit. The maximum follow-up was 28 months after radiotherapy. Patients without lymphedema were censored at the date of last follow-up.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Clinical and dosimetric variables\u003c/h2\u003e \u003cp\u003eBased on prior literature\u003csup\u003e[\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e, four clinical covariates were prespecified as candidate predictors: body mass index (BMI), number of dissected lymph nodes (LNDno), number of pathologically positive lymph nodes, and receipt of chemotherapy.\u003c/p\u003e \u003cp\u003eALTJ dosimetric parameters were extracted from the treatment planning system and included maximum dose (Dmax), minimum dose (Dmin), mean dose (Dmean), and Vx parameters (V5, V10, V15, V20, V25, V30, V35, V40, V45, and V50), where Vx denotes the percentage volume of ALTJ receiving\u0026thinsp;\u0026ge;\u0026thinsp;x Gy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Statistical analysis\u003c/h2\u003e \u003cp\u003ePatient, tumor and treatment characteristics were summarized using descriptive statistics. Continuous variables were reported as median with interquartile range (IQR) or mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD) as appropriate, and categorical variables were reported as frequencies and percentages. Differences in baseline characteristics between the development and validation cohorts were compared using the Pearson\u0026rsquo;s χ\u0026sup2; test or Fisher\u0026rsquo;s exact test for categorical variables, and the Mann-Whitney U test for continuous variables.\u003c/p\u003e \u003cp\u003ePredictive performance for binary outcomes (lymphedema vs. no lymphedema) was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC). For time-to-event outcomes, Harrell\u0026rsquo;s C-index was used. Calibration was assessed with the Hosmer-Lemeshow goodness-of-fit test (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05 indicating good fit). Kaplan\u0026ndash;Meier methods were used to estimate cumulative BCRL incidence across risk strata, with comparisons by log-rank test. Statistical analyses were performed using SPSS (version 27.0.1.0) and R (version 4.5.1), with a two-sided P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 considered statistically significant.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Model development and validation\u003c/h2\u003e \u003cp\u003eIn accordance with TRIPOD principles\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e, two distinct prediction models were developed using the development cohort and subsequently validated in the independent temporal validation cohort.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.6.1 NTCP model\u003c/h2\u003e \u003cp\u003eLeast absolute shrinkage and selection operator (LASSO) regression was used for initial variable screening among candidate clinical and dosimetric predictors. Variables meeting prespecified significance criteria were entered into a multivariable logistic regression framework. Forward stepwise selection guided by the Akaike information criterion (AIC) was then performed: the subset with the lowest AIC was selected as the final NTCP model estimating probability of BCRL.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.6.2 Nomogram Model\u003c/h2\u003e \u003cp\u003eFor time-to-event prediction, candidate predictors were screened using a consensus strategy comprising three approaches to ensure robustness: (1) variables with P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 in univariable Cox regression; (2) significant factors identified by the LASSO regression; and (3) features identified as important by the Boruta algorithm, a wrapper method built around a random forest classifier. Variables confirmed by all three approaches were included in the final multivariable Cox model. A nomogram was constructed based on this final model to predict the 2-year cumulative lymphedema incidence rate.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Patient characteristics\u003c/h2\u003e \u003cp\u003eA total of 316 patients were included (development cohort, n\u0026thinsp;=\u0026thinsp;271; independent validation cohort, n\u0026thinsp;=\u0026thinsp;45). Baseline characteristics were generally balanced between cohorts (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In the development and validation cohorts, mean age was 56.63\u0026thinsp;\u0026plusmn;\u0026thinsp;9.64 years and 56.58\u0026thinsp;\u0026plusmn;\u0026thinsp;10.31 years, respectively. Mean BMI was 23.97\u0026thinsp;\u0026plusmn;\u0026thinsp;3.27 kg/m\u0026sup2; and 24.53\u0026thinsp;\u0026plusmn;\u0026thinsp;3.12 kg/m\u0026sup2;, respectively. Mean LNDno was 13.50\u0026thinsp;\u0026plusmn;\u0026thinsp;4.68 and 13.78\u0026thinsp;\u0026plusmn;\u0026thinsp;2.63, and mean number of positive lymph nodes was 4.22\u0026thinsp;\u0026plusmn;\u0026thinsp;5.04 and 4.84\u0026thinsp;\u0026plusmn;\u0026thinsp;4.94, respectively. No significant differences were observed in tumor laterality, T/N stage, hormone receptor status, or other key variables.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline patient, tumor, and treatment characteristics of the patient cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDevelopment Cohort \u003c/p\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;271\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValidation Cohort \u003c/p\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;45\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP-value\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e56.63(9.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e56.58 (10.31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.967\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.138\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e145 (53.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (66.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e126 (46.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT Stage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.154\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e104 (38.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (24.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e141 (52.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (60.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (8.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (13.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (0.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (2.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN Stage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.679\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21 (7.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (4.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e153 (56.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23 (51.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e65 (24.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (28.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32 (11.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (15.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.337\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3 (1.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e160 (59.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (48.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e108 (39.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23 (51.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of Dissected Lymph Nodes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.50 (4.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13.78 (2.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.565\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of Pathologically Positive Lymph Nodes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.22 (5.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.84 (4.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.438\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChemotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e262(96.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43(95.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.824\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eER\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.636\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e95 (35.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (40.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e176 (64.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (60.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;0.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e133 (49.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (48.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138 (50.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23 (51.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHer2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.743\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e173 (63.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (60.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e98 (36.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (40.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKi-67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;0.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (7.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (6.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u0026thinsp;~\u0026thinsp;30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e111 (41.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (40.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e140 (51.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24 (53.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI(kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23.97 (3.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24.53 (3.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.271\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI Group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.232\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e˃ 23.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e133 (49.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (60.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;23.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138 (50.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (40.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphedema Event\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e68 (25.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.821\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003e1\u003c/sup\u003eMean (SD); n (%).\u003csup\u003e2\u003c/sup\u003eWilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eAbbreviations: BMI=body mass index; SD= standard deviation.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Lymphedema incidence\u003c/h2\u003e \u003cp\u003eIn the development cohort, 68 patients (25.1%) developed BCRL during follow-up, with a median time to onset of 13.5 months. The cumulative incidence rates at 12 and 24 months were 11.4% and 25.1%, respectively. In the validation cohort, 10 patients (22.2%) developed BCRL (median time: 10.5 months), with 12- and 24-month cumulative incidences of 13.3% and 22.2%. The difference in 24-month incidence between the two cohorts was not statistically significant (P\u0026thinsp;=\u0026thinsp;0.821).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e2.9 ALTJ dosimetric parameters\u003c/h2\u003e \u003cp\u003eDosimetric parameters for ALTJ are detailed in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. In the development cohort, the median ALTJ volume was 14.9 cm\u0026sup3; (IQR: 13.3\u0026ndash;17.7). Median Dmax, Dmean, and Dmin were 49.0 Gy (46.2\u0026ndash;52.1), 26.8 Gy (19.6\u0026ndash;33.6), and 2.6 Gy (1.7\u0026ndash;8.6), respectively. Median ALTJ V30 was 41.4% (26.2\u0026ndash;60.0). In the validation cohort, the corresponding values were 12.2 cm\u0026sup3; (9.8\u0026ndash;15.1), 49.6 Gy (48.2\u0026ndash;50.6), 30.1 Gy (26.7\u0026ndash;33.1), 10.2 Gy (6.0\u0026ndash;17.9), and 49.2% (41.5\u0026ndash;54.6), respectively. Overall distributions of ALTJ DVH metrics were broadly similar between cohorts, although selected metrics (including ALTJ volume and Dmin) differed.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eALTJ dosimetric parameters in the development and validation cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDevelopment Cohort N\u0026thinsp;=\u0026thinsp;271\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValidation Cohort N\u0026thinsp;=\u0026thinsp;45\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP-value\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ volume(cm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14.9 (13.3, 17.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.2 (9.8, 15.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ Dmax(Gy)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e49.0 (46.2, 52.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e49.6 (48.2, 50.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.595\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ Dmean(Gy)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26.8 (19.6, 33.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.1 (26.7, 33.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ Dmin(Gy)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.6 (1.7, 8.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.2 (6.0, 17.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V5(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e79.9 (69.3, 95.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.5 (81.9, 90.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V10(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e73.6 (57.4, 91.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.4 (74.7, 79.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.076\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V15(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e68.8 (52.4, 86.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71.9 (69.9, 75.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.124\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V20(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62.7 (46.9, 81.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.2 (64.1, 70.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.105\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V25(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52.1 (35.6, 68.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e56.2 (51.4, 61.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.126\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V30(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41.4 (26.2, 60.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e49.2 (41.5, 54.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.111\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V35(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43.4 (26.6, 57.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44.4 (40.5, 50.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.613\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V40(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34.2 (17.3, 45.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36.2 (28.7, 45.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.294\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V45(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21.2 (4.0, 29.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.5 (13.8, 30.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.231\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALTJ V50(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10.6 (1.5, 15.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.5 (2.4, 11.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003e1\u003c/sup\u003eMedian (IQR); \u003csup\u003e2\u003c/sup\u003eWilcoxon rank sum test\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eAbbreviations: ALTJ= axillary-lateral thoracic vessel junction; Dmax= maximum dose; Dmin=minimum dose; Dmean=mean dose; Vx denotes the percentage volume of ALTJ receiving\u0026thinsp;\u0026ge;\u0026thinsp;x Gy; IQR= interquartile range.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Model performance\u003c/h2\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003e2.10.1 NTCP model\u003c/h2\u003e \u003cp\u003eUnivariable analysis followed by LASSO regression identified LNDno, ALTJ V30, ALTJ Dmean, and ALTJ V25 as significant predictors. In multivariable logistic regression, ALTJ V30 (OR\u0026thinsp;=\u0026thinsp;3.47, 95% CI: 2.00\u0026ndash;6.00, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), LNDno (OR\u0026thinsp;=\u0026thinsp;1.55, 95% CI: 1.13\u0026ndash;2.21, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and ALTJ V25 (OR\u0026thinsp;=\u0026thinsp;0.322, 95% CI: 0.194\u0026ndash;0.532, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were significantly associated with BCRL and remained independent predictors (\u003cb\u003eFig.\u0026nbsp;1\u003c/b\u003e). LNDno (OR\u0026thinsp;=\u0026thinsp;1.30, 95% CI: 1.19\u0026ndash;1.43, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and ALTJ V30 (OR\u0026thinsp;=\u0026thinsp;1.03, 95% CI: 1.02\u0026ndash;1.05, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were used as 2 highly predictive variable to develop the NTCP model. The final NTCP model was:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{NTCP}_{Logit}=\\frac{1}{1+{e}^{-Logit\\left(P\\right)}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eLogit(P)\u0026thinsp;=\u0026thinsp;1.03\u0026times;ALTJ V30\u0026thinsp;+\u0026thinsp;1.30\u0026times;LNDno\u0026thinsp;\u0026minus;\u0026thinsp;6.53\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure\u0026nbsp;1.\u003c/b\u003e LASSO regression multi-factor importance analysis\u003c/p\u003e \u003cp\u003eThe model demonstrated robust performance with an AUC of 0.816 and a Brier score of 0.135 in the development cohort. In the independent validation cohort, the model maintained high predictive accuracy with an AUC of 0.860 and a Brier score of 0.111, supporting its external validity (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor clinical implementation, the optimal cut-off was determined by the Youden index. Patients were stratified into three risk groups based on LNDno and ALTJ V30: (1) High risk: LNDno\u0026thinsp;\u0026gt;\u0026thinsp;13 and ALTJ V30\u0026thinsp;\u0026gt;\u0026thinsp;51.75%; (2) Moderate risk: one of the two factors present; (3) Low risk: neither factor present.\u003c/p\u003e \u003cp\u003eIn the development cohort, the 2-year BCRL incidence was 58.8% in the high-risk group, 26.4% in the moderate-risk group, and 5.3% in the low-risk group. The corresponding rates in the validation cohort were 54.5%, 18.2%, and 0%, respectively. Kaplan\u0026ndash;Meier curves confirmed a significantly different cumulative incidence of lymphedema among these groups. (log-rank P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e2.10.2 Nomogram model\u003c/h2\u003e \u003cp\u003eAfter consensus feature selection (including LASSO and Boruta), LNDno and multiple ALTJ DVH parameters (including Dmean, V25, V30, V35, and V50) were evaluated in multivariable Cox regression (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The final predictors used to construct the nomogram for estimating 2-year BCRL risk were LNDno (HR\u0026thinsp;=\u0026thinsp;0.99, 95%CI: 0.95\u0026ndash;1.04), ALTJ Dmean (HR\u0026thinsp;=\u0026thinsp;1.13, 95%CI: 1.03\u0026ndash;1.23), ALTJ V25 (HR\u0026thinsp;=\u0026thinsp;0.75, 95%CI: 0.69\u0026ndash;0.81), ALTJ V30 (HR\u0026thinsp;=\u0026thinsp;1.54, 95%CI: 1.41\u0026ndash;1.68), and ALTJ V35 (HR\u0026thinsp;=\u0026thinsp;0.85, 95%CI: 0.81\u0026ndash;0.89) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe nomogram demonstrated excellent discrimination, with a C-index of 0.948 in the development cohort and 0.894 in the validation cohort. Moreover, decision curve analysis indicated a favorable net benefit, and the calibration plots showed good agreement between predicted and observed risk (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eIn this homogeneous cohort of patients treated with modified radical mastectomy and followed by postoperative radiotherapy, we identified a synergistic relationship between surgical extent and functionally targeted axillary dosimetry in predicting BCRL. Specifically, the number of dissected lymph nodes (LNDno) and ALTJ V30 emerged as independent, dominant predictors and formed the basis of a parsimonious, clinically interpretable NTCP model that retained strong performance in an independent temporal validation cohort. We further developed a clinical\u0026ndash;dosimetric nomogram integrating LNDno with multiple ALTJ intermediate-dose metrics (Dmean, V25, V30, V35), achieving high discrimination and demonstrable clinical utility. Importantly, risk stratification based on two readily accessible thresholds (ALTJ V30\u0026thinsp;\u0026gt;\u0026thinsp;51.75% and LNDno\u0026thinsp;\u0026gt;\u0026thinsp;13) delineated a marked gradient in 2-year BCRL risk. This approach identifies a patient subgroup for whom targeted preventive strategies and ALTJ-focused plan optimization may be particularly impactful.\u003c/p\u003e \u003cp\u003eA key strength of the present study is its exclusive focus on patients treated with modified radical mastectomy. Prior ALTJ-related studies often included heterogeneous cohorts combining breast-conserving surgery and mastectomy\u003csup\u003e[\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Because breast-conserving surgery typically preserves more axillary lymphatic architecture, baseline lymphatic reserve\u0026mdash;and therefore BCRL susceptibility\u0026mdash;differs substantially from that of patients undergoing more extensive axillary surgery\u003csup\u003e[\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e. By restricting the cohort to modified radical mastectomy, our analysis targets a clinically relevant setting of substantial baseline lymphatic disruption. Within this context, the ALTJ\u0026mdash;conceptualized as a critical lymphatic confluence for axillary Level I drainage\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e\u0026mdash;may represent a disproportionately important residual pathway. Radiation-related endothelial injury and fibrosis at this junction, even at the intermediate-dose range, could plausibly impair lymphatic decompensation. Our findings support the \"dual-hit\" pathophysiological model, where the initial surgical insult is critically exacerbated by subsequent radiation injury to this pivotal structure\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. The stability and interpretability of the combined LNDno\u0026ndash;ALTJ V30 signal in our models may reflect this biologically and anatomically constrained vulnerability.\u003c/p\u003e \u003cp\u003eOur findings also refine that how radiotherapy dose should be conceptualized in relation to BCRL. Historically, many studies have focused on dose to broad axillary anatomical regions (Levels I\u0026ndash;III)\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. This approach, by averaging dose over large volumes, may mask clinically significant heterogeneity within functionally critical substructures. By delineating the ALTJ as a distinct OAR and evaluating its DVH identified, we identified ALTJ V30 as a robust, clinically relevant predictor in the NTCP model. While prior studies by Park et al.\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e and Kim et al.\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e identified correlations with parameters like V35 or Dmax, we identify V30 as the dominant factor, particularly in a mastectomy-only cohort, thereby refining the understanding of the dose-risk relationship.\u003c/p\u003e \u003cp\u003eThe clinical impact of this work is twofold. First, the two-parameter stratification (LNDno and ALTJ V30) provides an immediately implementable approach to identify patients at high risk (2-year risk\u0026thinsp;\u0026gt;\u0026thinsp;54%), in whom early physiotherapy referral, structured surveillance, and preventive counseling could be prioritized. Second, the proposed dose constraint\u0026mdash;maintaining ALTJ V30 below 51.75%\u0026mdash;provides a clear, quantitative objective for the radiotherapy planner. Using techniques such as IMRT or proton therapy to achieve this constraint while preserving target coverage provides a tangible strategy for reducing the risk of BCRL.\u003c/p\u003e \u003cp\u003eFurthermore, the nomogram provides additional nuance by incorporating multiple intermediate-dose metrics (V25, V30, V35) and Dmean, which collectively suggest a continuous dose\u0026ndash;response relationship across approximately 25\u0026ndash;35 Gy rather than an exclusively threshold-driven phenomenon. This continuous-risk profile likely explains why the multivariable nomogram provides better discrimination than the simpler, clinical-only tools reported in prior studies\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. Methodologically, we strengthened predictor identification by combining conventional regression screening with LASSO and Boruta selection. This multi-method consensus strategy was designed to improve feature stability in the presence of collinearity among DVH parameters and to limit overfitting when evaluating high-dimensional dosimetric feature sets\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e. The maintenance of performance in an independent temporally cohort, without refitting, supports the transportability of these models within comparable PMRT practice settings.\u003c/p\u003e \u003cp\u003eSeveral limitations should be acknowledged. Firstly, the retrospective, single-center design limits causal inference and introduces potential selection bias. Although the use of a temporally independent validation cohort is valuable, it cannot fully address the fundamental constraints of the retrospective approach. Secondly, the sample size\u0026mdash;particularly in the validation cohort\u0026mdash;was modest; therefore, multicenter validation across institutions and involving patient populations with diverse racial backgrounds will be essential to establish broader generalizability. Thirdly, BCRL was defined using circumferential measurements, a clinically standard method\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e, which may be less sensitive than bioimpedance spectroscopy (BIS) for detecting subclinical lymphedema\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e. Future studies incorporating BIS-based endpoints, together with emerging strategies such as radiomics\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e and circulating biomarkers\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e, may enable earlier risk identification and facilitate further refinement of ALTJ constraints and risk-adapted preventive interventions.\u003c/p\u003e"},{"header":"5 Conclusions","content":"\u003cp\u003eIn conclusion, this study establishes the ALTJ as a pivotal functional OAR, whose radiation dose, specifically the V30, interacts synergistically with surgical extent to determine BCRL risk following mastectomy. We developed and validated an NTCP model and a Cox-based nomogram in an independent temporal cohort, providing clinically implementable tools that translating ALTJ dosimetry into individualized risk estimates. This evidence translates into a specific planning constraint: limiting ALTJ V30 to \u0026lt;\u0026thinsp;51.75%. Collectively, these findings represent a pivotal advancement towards function-preserving breast cancer radiotherapy. Prospective trials are warranted to confirm that adherence to these dosimetric principles yield measurable improvements in patient-reported lymphedema outcomes without compromising oncologic efficacy.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBCRL\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ebreast cancer-related lymphedema\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePMRT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003epostmastectomy radiotherapy\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eALND\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eaxillary lymph node dissection\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBMI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ebody mass index\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eALTJ\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eaxillary-lateral thoracic vessel junction\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOAR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eorgan at risk\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDVH\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003edose-volume histogram\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNTCP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003enormal tissue complication probability\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eKPS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eKarnofsky performance status\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003e3D-CRT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ethree-dimensional conformal radiotherapy\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIMRT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eintensity-modulated radiotherapy\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCTV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eclinical target volume\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eESTRO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eEuropean Society for Radiotherapy and Oncology\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRTOG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRadiation Therapy Oncology Group\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLNDno\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003enumber of dissected lymph nodes\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDmax\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003emaximum dose\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDmin\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eminimum dose\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDmean\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003emean dose\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eIQR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003einterquartile range\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003estandard deviation\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ereceiver operating characteristic\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003earea under the ROC curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLASSO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eleast absolute shrinkage and selection operator\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAIC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAkaike information criterion\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBIS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ebioimpedance spectroscopy\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Ethics Committee of Changshu No.1 People\u0026rsquo;s Hospital, and the requirement for informed consent was waived due to the retrospective design. All procedures performed in this study were in accordance with the ethical standards of the institutional research committee and the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by Changshu Health Commission Science and Technology Plan Sponsored Projects and Changshu Science and Technology Development Plan (Medical and Health) Guidance Projects (CSWS202304)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eN.X and C.Y. performed data statistics and drafted the main manuscript. F.W., C.Z. and H.G. were involved in data collection and data statistics. Z.J. reviewed the target regions, and participated in the revision and review of the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller KD, Nogueira L, Devasia T, et al. Cancer treatment and survivorship statistics, 2022. CA Cancer J Clin. 2022;72(5):409\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen A, Lu Q, Fu X, et al. Risk factors of unilateral breast cancer-related lymphedema: an updated systematic review and meta-analysis of 84 cohort studies. Support Care Cancer. 2022;31(1):18. Published 2022 Dec 14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaoum GE, Roberts S, Brunelle CL, et al. Quantifying the Impact of Axillary Surgery and Nodal Irradiation on Breast Cancer-Related Lymphedema and Local Tumor Control: Long-Term Results From a Prospective Screening Trial. J Clin Oncol. 2020;38(29):3430\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhn HR, Jeong HE, Jeong C, et al. Incidence and risk factors of breast cancer-related lymphedema in Korea: a nationwide retrospective cohort study. Int J Surg. 2024;110(6):3518\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKwon S, Janssen CF, Velasquez FC, et al. Radiation Dose-Dependent Changes in Lymphatic Remodeling. Int J Radiat Oncol Biol Phys. 2019;105(4):852\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin Y, Xu Y, Wang C, et al. Loco-regional therapy and the risk of breast cancer-related lymphedema: a systematic review and meta-analysis. Breast Cancer. 2021;28(6):1261\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGross JP, Whelan TJ, Parulekar WR, et al. Development and Validation of a Nomogram to Predict Lymphedema After Axillary Surgery and Radiation Therapy in Women With Breast Cancer From the NCIC CTG MA.20 Randomized Trial. Int J Radiat Oncol Biol Phys. 2019;105(1):165\u0026ndash;73.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi MM, Wu PP, Qiang WM, et al. Development and validation of a risk prediction model for breast cancer-related lymphedema in postoperative patients with breast cancer. Eur J Oncol Nurs. 2023;63:102258.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo X, Ye J, Xiao T, et al. Development and validation of a predictive nomogram for postoperative upper limb lymphedema in breast cancer patients: a retrospective cohort study. Sci Rep. 2025;15(1):24609.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eByun HK, Chang JS, Im SH, et al. Risk of Lymphedema Following Contemporary Treatment for Breast Cancer: An Analysis of 7617 Consecutive Patients From a Multidisciplinary Perspective. Ann Surg. 2021;274(1):170\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGross JP, Lynch CM, Flores AM, et al. Determining the Organ at Risk for Lymphedema After Regional Nodal Irradiation in Breast Cancer. Int J Radiat Oncol Biol Phys. 2019;105(3):649\u0026ndash;58.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuk Chang J, Ko H, Hee Im S, et al. Incorporating axillary-lateral thoracic vessel juncture dosimetric variables improves model for predicting lymphedema in patients with breast cancer: A validation analysis. Clin Transl Radiat Oncol. 2023;41:100629.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark YI, Chang JS, Ko H, et al. Development and Validation of a Normal Tissue Complication Probability Model for Lymphedema After Radiation Therapy in Breast Cancer. Int J Radiat Oncol Biol Phys. 2023;116(5):1218\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOffersen BV, Boersma LJ, Kirkove C, et al. ESTRO consensus guideline on target volume delineation for elective radiation therapy of early stage breast cancer. Radiother Oncol. 2015;114(1):3\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBreast cancer atlas for radiation therapy planning. Consensus definitions. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.rtog.org/corelab/contouringatlases/breastcanceratlas.aspx\u003c/span\u003e\u003cspan address=\"https://www.rtog.org/corelab/contouringatlases/breastcanceratlas.aspx\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed January 31, 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePosition Statement of the National Lymphedema Network. Topic:Screening and Measurement for Early Detection of Breast Cancer Related Lymphedema. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://lymphnet.org/positionpapers\u003c/span\u003e\u003cspan address=\"https://lymphnet.org/positionpapers\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed January 31, 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollins GS, Moons KGM, Dhiman P, et al. TRIPOD\u0026thinsp;+\u0026thinsp;AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. Published 2024 Apr 16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiuliano AE, Ballman KV, McCall L, et al. Effect of Axillary Dissection vs No Axillary Dissection on 10-Year Overall Survival Among Women With Invasive Breast Cancer and Sentinel Node Metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial. JAMA. 2017;318(10):918\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalimberti V, Cole BF, Viale G, et al. Axillary dissection versus no axillary dissection in patients with breast cancer and sentinel-node micrometastases (IBCSG 23\u0026thinsp;\u0026ndash;\u0026thinsp;01): 10-year follow-up of a randomised, controlled phase 3 trial. Lancet Oncol. 2018;19(10):1385\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhelan TJ, Olivotto IA, Parulekar WR, et al. Regional Nodal Irradiation in Early-Stage Breast Cancer. N Engl J Med. 2015;373(4):307\u0026ndash;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMoa1415340\u003c/span\u003e\u003cspan address=\"10.1056/NEJMoa1415340\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong SW, Kim EK, Shin HC, et al. Relationship of Immediate Breast Reconstruction and the Development of Lymphedema in Breast Cancer Patients With Radiation Therapy. Int J Radiat Oncol Biol Phys. 2025;123(3):753\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRockson SG. Lymphedema after Breast Cancer Treatment. N Engl J Med. 2018;379(20):1937\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlmahariq MF, Maywood MJ, Levitin RB, et al. Mapping of Metastatic Level I Axillary Lymph Nodes in Patients with Newly Diagnosed Breast Cancer. Int J Radiat Oncol Biol Phys. 2020;106(4):811\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang Q, Hu H, Liao J, et al. Development and validation of a nomogram for breast cancer-related lymphedema. Sci Rep. 2024;14(1):15602.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKursa MB, Rudnicki WR. Feature Selection with the Boruta Package. J Stat Softw. 2010;36(11):1\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShah C, Boyages J, Koelmeyer L, et al. Timing of Breast Cancer Related Lymphedema Development Over 3 Years: Observations from a Large, Prospective Randomized Screening Trial Comparing Bioimpedance Spectroscopy (BIS) Versus Tape Measure. Ann Surg Oncol. 2024;31(11):7487\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAr GK. J, R H, Breast Cancer-Related Lymphedema: Magnetic Resonance Imaging Evidence of Sparing Centered Along the Cephalic Vein. J Reconstr Microsurg. 2021;37(6).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVisser J, van Geel M, Cornelissen AJM, et al. Breast Cancer-Related Lymphedema and Genetic Predisposition: A Systematic Review of the Literature. Lymphat Res Biol. 2019;17(3):288\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"world-journal-of-surgical-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"wjso","sideBox":"Learn more about [World Journal of Surgical Oncology](http://wjso.biomedcentral.com)","snPcode":"12957","submissionUrl":"https://submission.nature.com/new-submission/12957/3","title":"World Journal of Surgical Oncology","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Breast cancer, Lymphedema, OAR, ALTJ, Predictive model, Dose–volume histogram","lastPublishedDoi":"10.21203/rs.3.rs-8886686/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8886686/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eBreast cancer-related lymphedema (BCRL) is a disabling late complication after postmastectomy radiotherapy (PMRT). This study evaluated the axillary-lateral thoracic vessel junction (ALTJ) as a functional organ-at-risk (OAR), established its dose-response relationship with BCRL, and developed validated predictive models to guide individualized risk mitigation.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003e271 patients treated with PMRT from 2019 to 2022 constituted the development cohort, and 45 independent patients treated in 2023 formed the temporal validation cohort. All patients underwent modified radical mastectomy. The ALTJ was contoured on planning CT according to Gross et al. Candidate clinical factors and ALTJ dose\u0026ndash;volume histogram (DVH) parameters were analyzed. A normal tissue complication probability (NTCP) model was developed using LASSO-based screening followed by multivariable logistic regression, and a Cox regression\u0026ndash;based nomogram was built using multi-method consensus feature selection. Both models were evaluated and validated without refitting in the temporal cohort.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe 2-year cumulative BCRL incidence was 25.1% in the development and 22.2% in the validation cohort. Multivariable analysis identified the number of dissected lymph nodes (LNDno) and ALTJ V30 as the strongest predictors. The final NTCP model achieved an AUC of 0.816 in the development cohort and 0.860 in the validation cohort, with Brier scores of 0.135 and 0.111, respectively. A clinically actionable risk stratification system was derived using thresholds of LNDno\u0026thinsp;\u0026gt;\u0026thinsp;13 and ALTJ V30\u0026thinsp;\u0026gt;\u0026thinsp;51.75%, identifying high-, moderate-, and low-risk groups with 2-year BCRL rates of 58.8%/54.5%, 26.4%/18.2%, and 5.3%/0% in the development and validation cohorts, respectively. A nomogram, integrating LNDno with ALTJ V25, V30, V35, and Dmean, achieved C-indices of 0.948 and 0.894 in the two cohorts, respectively.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThis study identifies ALTJ V30 and surgical extent as important predictors of BCRL in postmastectomy patients receiving radiotherapy. The findings support the consideration of ALTJ as a quantifiable OAR and provide an evidence-based dose\u0026ndash;volume constraint (V30\u0026thinsp;\u0026lt;\u0026thinsp;51.75%). The validated NTCP model and nomogram offer practical tools for individualized risk estimation and may inform targeted surveillance and preventive strategies.\u003c/p\u003e","manuscriptTitle":"Dose to the axillary-lateral thoracic vessel junction predicts breast cancer-related lymphedema after postmastectomy radiotherapy: development and temporal validation of NTCP and Nomogram models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-18 08:22:35","doi":"10.21203/rs.3.rs-8886686/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-21T15:57:34+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-15T22:51:33+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-12T15:04:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"197264772073015403506770688939274155039","date":"2026-03-12T12:49:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"97819548161816886608578211043579094979","date":"2026-03-12T10:27:03+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-12T10:19:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-19T04:30:52+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-16T08:46:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"World Journal of Surgical Oncology","date":"2026-02-15T14:30:13+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"world-journal-of-surgical-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"wjso","sideBox":"Learn more about [World Journal of Surgical Oncology](http://wjso.biomedcentral.com)","snPcode":"12957","submissionUrl":"https://submission.nature.com/new-submission/12957/3","title":"World Journal of Surgical Oncology","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"27581a41-48f7-4d00-b866-ee609c230285","owner":[],"postedDate":"March 18th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T16:09:09+00:00","versionOfRecord":{"articleIdentity":"rs-8886686","link":"https://doi.org/10.1186/s12957-026-04372-w","journal":{"identity":"world-journal-of-surgical-oncology","isVorOnly":false,"title":"World Journal of Surgical Oncology"},"publishedOn":"2026-04-27 15:56:50","publishedOnDateReadable":"April 27th, 2026"},"versionCreatedAt":"2026-03-18 08:22:35","video":"","vorDoi":"10.1186/s12957-026-04372-w","vorDoiUrl":"https://doi.org/10.1186/s12957-026-04372-w","workflowStages":[]},"version":"v1","identity":"rs-8886686","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8886686","identity":"rs-8886686","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}