The clinical relevance of LODDS for post-mastectomy radiotherapy in T1–2N1M0 breast cancer: a real-world propensity score-weighted analysis using the SEER database

Introduction

Background

Breast cancer is the most common malignant tumor in women worldwide and an important cause of cancer-related deaths in female patients (1). Postmastectomy radiotherapy (PMRT) is a core locoregional therapy for breast cancer (2). It can lower the risk of locoregional recurrence and improve clinical outcomes in high-risk patients following mastectomy (3). While PMRT is generally recommended for patients with ≥4 positive lymph nodes or T3–4 disease, its role in T1–2N1M0 breast cancer with 1–3 positive nodes remains uncertain (4,5). Despite being staged as node-positive, this patient cohort is clinically heterogeneous, with relatively early primary tumors and limited nodal burden (6,7). Therefore, it is vital to identify which patients in this group benefit from PMRT, to support individualized decisions about radiotherapy.

Rationale and knowledge gap

Nodal burden is an important factor in estimating recurrence risk and guiding adjuvant treatment decisions (7,8). Traditionally, nodal status in breast cancer has been assessed primarily according to the absolute number of positive lymph nodes. However, this approach may not fully capture the heterogeneity in prognosis among patients within the same nodal category. Previous studies have shown that the number of negative lymph nodes also provides important prognostic information (9,10). These findings suggest that positive lymph node count alone may be insufficient for comprehensive assessment of nodal disease and prognostic stratification.

The log odds of positive lymph nodes (LODDS) is an emerging nodal burden metric (11). Unlike simple positive node count, LODDS can further stratify risk among patients in the same pN1 category (12). LODDS has been shown to have prognostic value in several cancers, including breast cancer (11,13,14). But most existing studies only focus on its prognostic performance. For T1–2N1M0 breast cancer after mastectomy, its application is still not well established.

Objective

We used data from the Surveillance, Epidemiology, and End Results (SEER) database for this study and enrolled patients with T1–2N1M0 breast cancer who had undergone mastectomy. We applied inverse probability of treatment weighting (IPTW) to balance baseline characteristics between the PMRT and non-PMRT groups. This study aims to provide individualized PMRT guidance based on LODDS stratification for T1–2N1M0 mastectomy patients, to help surgeons and radiation oncologists in routine clinical practice. We present this article in accordance with the STROBE reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0343/rc).

Methods

Patients

Data for this retrospective study were extracted from the SEER database. Patients diagnosed with breast cancer between 2010 and 2018 were identified. The inclusion criteria were as follows: (I) female patients; (II) pathologically confirmed primary breast cancer; (III) receipt of mastectomy; (IV) American Joint Committee on Cancer (AJCC) stage T1–2N1M0 disease, excluding pN1mi; (V) one to three positive regional lymph nodes; and (VI) available information on year of diagnosis, age at diagnosis, race, histological type, T stage, number of positive regional lymph nodes, number of examined regional lymph nodes, pathological grade, chemotherapy status, radiotherapy status, and survival outcomes. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Patients were excluded if they met any of the following criteria: (I) male breast cancer; (II) distant metastatic disease; (III) no surgery, breast-conserving surgery, or unknown surgical procedure; (IV) unclear radiotherapy information, non-postoperative radiotherapy, or receipt of radioactive implants or radioisotope therapy; (V) missing key clinicopathological variables; or (VI) missing survival time or survival time of 0 months. The patient selection process is shown in Figure 1.

Figure 1 Patient selection flowchart. PMRT, postmastectomy radiotherapy; SEER, Surveillance, Epidemiology, and End Results.

No a priori sample-size calculation was performed; all eligible records within the study period were included to maximize precision.

Variables and data sources

Demographic, tumor, treatment, and survival variables were obtained from SEER and were ascertained using the same registry definitions in both treatment groups. Patients recorded as receiving external-beam radiotherapy after mastectomy were assigned to the PMRT group, whereas those without postoperative radiotherapy were classified as the non-PMRT group. The number of negative lymph nodes was calculated as the number of examined regional lymph nodes minus the number of positive regional lymph nodes.

The SEER dataset records the total number of regional lymph nodes examined. However, it cannot reliably tell whether patients received sentinel lymph node biopsy (SLNB) alone or completion axillary lymph node dissection (ALND) for breast cancer. For this reason, we lacked clear information on the exact axillary surgery type.

LODDS was calculated as log [(number of positive lymph nodes + 0.5)/(number of negative lymph nodes + 0.5)]. For the primary effect-modification analysis, patients were divided according to the cohort tertiles into low (LODDS ≤−1.85), intermediate (−1.85< LODDS ≤−1.00), and high (LODDS >−1.00) groups. For subgroup analyses, age was categorized as <50 versus ≥50 years and the number of examined lymph nodes as <10 versus ≥10. Alternative LODDS quartiles and LODDS modeled continuously with restricted cubic splines (RCS) were examined in sensitivity analyses.

The primary outcome was breast cancer-specific survival (BCSS), defined as the interval from diagnosis to death attributed to breast cancer in SEER. The secondary outcome was overall survival (OS), defined as the interval from diagnosis to death from any cause. Patients without the relevant event were right-censored at their last recorded follow-up. Molecular subtype was classified from SEER estrogen receptor, progesterone receptor, and HER2 information as hormone receptor-positive/HER2-negative, HER2-positive, or triple-negative breast cancer (TNBC).

Bias, study size, and missing data

Because PMRT was not randomly assigned, confounding by indication was anticipated. Measured confounding was addressed with propensity-score weighting, weight trimming, covariate-balance assessment, subgroup interaction analyses, and sensitivity analyses using propensity-score matching (PSM) and competing-risk models. Explicit eligibility criteria and identical SEER variable definitions for both treatment groups were used to reduce selection and information bias.

Patients with missing data on eligibility criteria, PMRT administration, LODDS, molecular subtype, or survival endpoints were excluded, and the number of patients excluded at each screening stage is summarized in Figure 1. Unknown pathological grade was retained as a separate category, and the SEER chemotherapy category “No/Unknown” was analyzed as recorded. Loss to follow-up could not be distinguished from administrative censoring in the registry; observations were censored at the last recorded follow-up.

Statistical analysis

Baseline characteristics were compared between the PMRT and non-PMRT groups. Categorical variables were analyzed using the chi-square test or Fisher’s exact test, and continuous variables were compared using the Wilcoxon rank-sum test. Continuous variables were summarized as medians with interquartile ranges, and categorical variables as counts and percentages.

Kaplan-Meier methods were used to estimate BCSS and OS, and groups were compared with log-rank tests. Cox proportional hazards regression models were used to estimate the associations between PMRT and each survival outcome. Results are reported as hazard ratios with 95% confidence intervals (CIs). Multiplicative interaction terms were used to test whether the association between PMRT and survival differed across LODDS strata and other predefined subgroups.

IPTW was used to construct a weighted cohort and reduce measured confounding. Propensity scores were estimated using multivariable logistic regression, with receipt of PMRT as the dependent variable. Covariates included year of diagnosis, age at diagnosis, race, histological type, T stage, number of positive lymph nodes, pathological grade, molecular subtype, chemotherapy status, and LODDS group. Stabilized IPTW weights were calculated and trimmed at the 1st and 99th percentiles. Covariate balance before and after weighting was evaluated using standardized mean differences (SMDs), with an absolute SMD <0.1 indicating adequate balance.

Sensitivity analyses included LODDS quartile stratification, 1:1 PSM, RCS analysis, and Fine-Gray competing-risk models. For the PSM sensitivity analysis, propensity scores were estimated using the same covariates as in the IPTW model. Patients receiving PMRT were matched 1:1 with patients not receiving PMRT using nearest-neighbor matching without replacement, with exact matching on LODDS group and a caliper width of 0.2 standard deviations of the logit of the propensity score. After matching, 10,810 PMRT-exposed patients and 10,810 non-PMRT patients were retained. Post-matching covariate balance was assessed using SMDs, with an absolute SMD <0.1 indicating adequate balance. RCS analysis evaluated LODDS as a continuous variable. Fine-Gray models treated non-breast-cancer death as a competing event and breast-cancer-specific death as the event of interest. All statistical tests were two-sided, and P<0.05 was considered statistically significant. Analyses were conducted using R software.

Results

Baseline patient characteristics

A total of 28,864 patients with T1–2N1M0 breast cancer who underwent mastectomy were finally enrolled in this study. Among them, 12,481 patients received PMRT, and the remaining 16,383 did not. Baseline demographic and clinicopathological characteristics of the two groups are summarized in Table 1.

In the original cohort, notable differences were observed between the PMRT and non-PMRT groups. Patients in the PMRT group were younger, and presented with a higher proportion of T2 stage disease, 2 to 3 positive lymph nodes, chemotherapy administration and high LODDS levels.

After IPTW, baseline characteristics were well balanced between the two groups (Table 2).

Before weighting, several variables showed substantial imbalance, including chemotherapy status (SMD =0.598), age (SMD =0.421), number of positive lymph nodes (SMD =0.339), year of diagnosis (SMD =0.310) and LODDS stratification (SMD =0.249). All these variables achieved adequate balance after weighting, with absolute SMD values below 0.1.

Comparisons of survival outcomes between the PMRT group and the non-PMRT group

Kaplan-Meier survival curves for the original cohort revealed that patients receiving PMRT achieved significantly better BCSS and OS than those without PMRT (Figure 2A, all P<0.001).

Figure 2 Kaplan-Meier curves of BCSS and OS in patients with T1–2N1M0 breast cancer with or without PMRT before (A) and after (B) IPTW. BCSS, breast cancer-specific survival; IPTW, inverse probability of treatment weighting; OS, overall survival; PMRT, postmastectomy radiotherapy.

Unweighted Cox regression analyses further confirmed that PMRT was independently linked to reduced risks of breast cancer-specific death and all-cause death. The hazard ratio for BCSS was 0.803 (95% CI: 0.753–0.857, P<0.001), and the hazard ratio for OS was 0.645 (95% CI: 0.615–0.677, P<0.001, Table S1).

Consistent results were observed in the IPTW-weighted cohort. The weighted Kaplan-Meier curves still indicated superior BCSS and OS in the PMRT group (Figure 2B, all P<0.001). Findings from the IPTW-weighted Cox regression were in line with those of the original cohort (Table S1).

Collectively, PMRT was significantly associated with improved BCSS and OS among mastectomy patients with T1–2N1M0 breast cancer, both in the original and IPTW-weighted cohorts.

Survival benefits of PMRT across different LODDS strata

To explore whether the survival benefits of PMRT varied by LODDS level, we plotted weighted Kaplan-Meier survival curves for the PMRT and non-PMRT groups within each LODDS subgroup (Figure 3). In the low LODDS stratum, the BCSS curve was slightly higher in the PMRT group, yet the log‑rank difference did not reach statistical significance (P=0.09). By contrast, patients in the intermediate and high LODDS strata who received PMRT achieved significantly better BCSS (all P<0.001). Kaplan-Meier curves for OS revealed that PMRT was linked to improved OS across all LODDS subgroups, with the most pronounced separation observed in the high LODDS group (Figure 3).

Figure 3 Kaplan-Meier curves of BCSS (A) and OS (B) for patients with T1–2N1M0 breast cancer stratified by LODDS. Patients were divided into low (LODDS ≤−1.85, left), intermediate (−1.85< LODDS ≤−1.00, middle), and high (LODDS >−1.00, right) groups. BCSS, breast cancer-specific survival; LODDS, log odds of positive lymph nodes; OS, overall survival; PMRT, postmastectomy radiotherapy.

IPTW-weighted Cox regression analyses confirmed a significant interaction between PMRT and LODDS for BCSS (P for interaction =0.048). In the low LODDS group, PMRT was associated with a marginally lower risk of breast cancer-specific death, but the difference was not statistically significant (hazard ratio =0.895, 95% CI: 0.779–1.028, P=0.12). Consistent with the curve findings, PMRT conferred significant improvements in BCSS among patients in the intermediate and high LODDS subgroups (Table 3).

Five-year absolute survival benefits of PMRT across different LODDS strata

To further quantify the absolute survival benefits of PMRT across different LODDS subgroups, we compared the 5-year BCSS and OS rates between the PMRT and non-PMRT groups (Table 4). The magnitude of the 5-year absolute survival difference associated with PMRT increased progressively with rising LODDS levels. In the low LODDS stratum, the 5-year BCSS rates were 92.8% in the non-PMRT group and 93.4% in the PMRT group, yielding an absolute difference of only 0.5%. This benefit was notably smaller than that observed in the intermediate (2.8%) and high LODDS strata (4.1%).

Role of PMRT in subgroup analyses

To further examine whether the association between PMRT and survival outcomes was consistent across different clinical subgroups, IPTW-weighted subgroup analyses were performed. The subgroup variables included age, T stage, molecular subtype, number of positive lymph nodes, number of examined lymph nodes, chemotherapy status, and year of diagnosis. The results for PMRT versus non-PMRT within each subgroup are presented in Figure 4 and Table S2.

Figure 4 IPTW-weighted subgroup analyses of the association between PMRT and survival outcomes. (A) BCSS; (B) OS. BCSS, breast cancer‑specific survival; CI, confidence interval; HR+/HER2−, hormone receptor‑positive/human epidermal growth factor receptor 2‑negative; HR, hazard ratio; IPTW, inverse probability of treatment weighting; OS, overall survival; PMRT, postmastectomy radiotherapy; TNBC, triple‑negative breast cancer.

Overall, the association between PMRT and improved BCSS and OS was consistent in direction across most subgroups. For BCSS, PMRT was associated with significantly better outcomes in patients aged ≥50 years (hazard ratio =0.733, 95% CI: 0.672–0.799, P<0.001), but not in those aged <50 years (hazard ratio =0.902, 95% CI: 0.794–1.026, P =0.11). The interaction between age and PMRT was statistically significant (P for interaction =0.008). In molecular subtype subgroups, PMRT was linked to significantly better BCSS in hormone receptor-positive (HR⁺)/HER2⁻ and TNBC patients (all P<0.001), while the association did not reach statistical significance in HER2-positive patients (hazard ratio =0.841, 95% CI: 0.703–1.006, P=0.058). When stratified by the number of positive lymph nodes, PMRT was associated with significantly improved BCSS in patients with 1, 2, or 3 positive nodes. Similarly, PMRT was significantly associated with better BCSS in both patients with <10 examined nodes and those with ≥10 examined nodes. The results of the OS subgroup analyses were generally consistent with those of BCSS, showing that PMRT was associated with significantly improved OS across all predefined subgroups.

Sensitivity analyses

To verify the robustness of the heterogeneous survival benefits of PMRT across LODDS strata, we performed a series of sensitivity analyses using different LODDS categorization schemes. Overall, the sensitivity results were largely consistent with the primary analysis: PMRT-related survival benefits were primarily observed in patients with higher LODDS levels, while those in the lowest LODDS stratum showed limited benefit.

First, RCS analyses revealed that LODDS was associated with a progressively increasing risk of breast cancer-specific death and all-cause death. Both risks rose gradually with increasing LODDS, with a more pronounced upward trend observed as LODDS approached or exceeded −1.00 (Figure S1).

Next, we re-stratified patients using LODDS quartiles instead of tertiles (Table S3). For BCSS, PMRT did not confer a significant benefit in the lowest quartile (Q1) (hazard ratio =1.006, 95% CI: 0.865–1.170, P=0.94). In contrast, PMRT was significantly associated with improved BCSS in Q2, Q3, and Q4, with hazard ratios of 0.729 (95% CI: 0.629–0.845), 0.691 (95% CI: 0.602–0.794), and 0.722 (95% CI: 0.632–0.824), respectively (all P<0.001). The interaction test for BCSS was statistically significant (P for interaction =0.001). OS results mirrored those of BCSS.

Third, we replaced IPTW with 1:1 propensity score matching (PSM) (Table S4). In the PSM cohort, PMRT was not significantly associated with improved BCSS (hazard ratio =0.954, 95% CI: 0.822–1.108, P=0.54) or OS (hazard ratio =0.967, 95% CI: 0.862–1.084, P=0.56) in the low LODDS stratum. Conversely, PMRT remained significantly associated with better BCSS and OS in both intermediate and high LODDS groups.

Finally, Fine-Gray competing risk models were used to treat non-breast cancer death as a competing event (Table S5). The association between PMRT and breast cancer-specific mortality remained consistent with the primary analysis. In the low LODDS stratum, PMRT did not significantly reduce the risk of breast cancer-specific death [subdistribution hazard ratio (sHR) =0.906, 95% CI: 0.789–1.040, P=0.16]. In contrast, PMRT was associated with significantly lower breast cancer-specific mortality in the intermediate and high LODDS groups.

Discussion

Key findings

This SEER-based cohort study found that PMRT was associated with improved BCSS and OS among patients with T1–2N1M0 breast cancer treated with mastectomy. The main finding was that this association varied by LODDS. In the low LODDS group, PMRT was not significantly associated with improved BCSS and the reported absolute BCSS difference was small. In the intermediate and high LODDS groups, PMRT was consistently associated with improved survival. Sensitivity analyses using LODDS quartiles, PSM, RCS, and Fine-Gray models were directionally consistent.

Strengths and limitations

This study has several strengths. First, it focused on patients with T1–2N1M0 disease and one to three positive lymph nodes, a clinically important population in which PMRT decision-making remains controversial. Second, IPTW was used to reduce baseline differences between the PMRT and non-PMRT groups, and multiple sensitivity analyses were performed to test the robustness of the findings. Third, we reported both relative effects and 5-year absolute survival differences, improving the clinical interpretability of the results.

Several limitations should be noted. First, this was a retrospective registry-based study, and residual confounding cannot be fully excluded despite the use of IPTW and PSM. Second, the SEER database lacks information on margin status, lymphovascular invasion, extranodal extension, and locoregional recurrence. Although the number of examined regional lymph nodes was available, the exact type and extent of axillary surgery, including SLNB alone versus completion ALND, could not be reliably determined. Third, details of radiotherapy, including target volume, dose, fractionation, and regional nodal irradiation, were unavailable. Therefore, we could not determine whether patients with an undissected axilla received specific axillary or regional nodal irradiation. In addition, information on endocrine therapy, anti-HER2 therapy, and specific chemotherapy regimens was not captured. Finally, the LODDS cutoffs were derived from the present cohort and require validation in independent datasets and prospective studies.

Comparison with similar research

These results align with evidence that PMRT can improve outcomes in node-positive breast cancer, while also supporting the guideline emphasis on individualized treatment for patients with limited nodal disease (2,7,15). The findings suggest that pN1 disease contains clinically relevant heterogeneity. Most previous studies have assessed nodal burden using positive lymph node count or lymph node ratio (16). The present study extends this work by applying LODDS to stratify PMRT benefit in patients with T1–2N1M0 disease. Notably, this study does not claim LODDS to be a better prognostic model. Rather, it highlights its potential to distinguish subgroups with differential survival benefits from PMRT.

Axillary management has progressively shifted toward de-escalation for patients with limited sentinel lymph node involvement (17). In the mastectomy subgroup of the SINODAR-ONE randomized trial, patients with T1–2 breast cancer and one or two macrometastatic sentinel lymph nodes who underwent SLNB alone had OS and recurrence-free survival outcomes that were non-inferior to those of patients who underwent completion ALND (18). Future studies will require datasets with detailed operative records and radiotherapy target volume data. These data will help further clarify the role of LODDS.

Patients with nodal micrometastases represent a distinct, lower-burden group compared with patients with macrometastatic pN1 disease included in the present study (19,20). In a retrospective analysis of a multi-institutional prospective database, axillary recurrence was rare among patients with T1–2N1mi breast cancer treated with mastectomy and SLNB alone, with no significant differences in disease-free survival or OS compared with those undergoing ALND (21). Furthermore, among patients treated with SLNB alone, PMRT was not associated with significant differences in survival outcomes (22,23). These findings suggest that further axillary treatment and PMRT may potentially be omitted in appropriately selected patients with pN1mi disease. Importantly, patients with pN1mi disease were excluded from our cohort; therefore, our findings apply specifically to patients with macrometastatic pN1 disease and should not be extrapolated to patients with nodal micrometastases.

Explanations of findings

The OS benefit associated with PMRT is consistent with its role in reducing locoregional recurrence and improving outcomes in selected node-positive patients. However, an overall benefit does not mean that all patients with T1–2N1M0 disease benefit equally.

Low LODDS generally reflects fewer positive nodes and more negative nodes, suggesting lower regional tumor burden and more adequate nodal evaluation. These patients may already have a relatively low risk of breast cancer-specific mortality. In contrast, intermediate and high LODDS may indicate higher residual regional risk, which could explain the greater BCSS and OS benefit observed with PMRT in these groups.

The RCS analysis showed increasing mortality risk with higher LODDS, supporting its value as a continuous nodal burden marker. Fine-Gray competing-risk models also showed that PMRT was associated with reduced breast cancer-specific mortality in the intermediate and high LODDS groups, but not in the low LODDS group. These findings suggest that the main results were not driven by competing non-breast cancer deaths.

Implications and actions needed

These findings suggest that PMRT decision-making in T1–2N1M0 breast cancer should not rely solely on N-stage. LODDS is a simple metric derived from routine pathological data and may help identify patients who are more likely to benefit from PMRT.

For patients with intermediate or high LODDS, the consistent association between PMRT and improved BCSS and OS supports stronger consideration of PMRT. Patients with low LODDS may be suitable candidates for future radiotherapy de-escalation studies.

Future studies should validate this LODDS-based stratification strategy in independent cohorts and prospective trials. More detailed data on radiotherapy fields, systemic therapy, molecular subtype, and locoregional recurrence are also needed to develop more individualized PMRT decision-making models.

Conclusions

In this SEER-based IPTW cohort study, PMRT was associated with improved BCSS and OS in patients with T1–2N1M0 breast cancer after mastectomy, particularly among those with intermediate or high LODDS. Patients with low LODDS showed limited BCSS benefit. LODDS may help refine PMRT decision-making beyond N stage alone in clinical practice.

Acknowledgments

The authors thank the National Cancer Institute and the SEER Program for providing access to the database used in this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0343/rc

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0343/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0343/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Jimenez RB, Abdou Y, Anderson P, et al. Postmastectomy Radiation Therapy: An ASTRO/ASCO/SSO Clinical Practice Guideline. Pract Radiat Oncol 2025;15:549-71. [Crossref] [PubMed]
3. Overgaard M, Nielsen HM, Tramm T, et al. Postmastectomy radiotherapy in high-risk breast cancer patients given adjuvant systemic therapy. A 30-year long-term report from the Danish breast cancer cooperative group DBCG 82bc trial. Radiother Oncol 2022;170:4-13.
4. Verma R, Chandarana M, Barrett J, et al. Post-mastectomy radiotherapy for women with early breast cancer and one to three positive lymph nodes. Cochrane Database Syst Rev 2023;6:CD014463. [Crossref] [PubMed]
5. Kunkler IH, Russell NS, Anderson N, et al. Ten-Year Survival after Postmastectomy Chest-Wall Irradiation in Breast Cancer. N Engl J Med 2025;393:1771-83. [Crossref] [PubMed]
6. Sun GY, Wen G, Zhang YJ, et al. Risk factors to identify the indication for regional nodal irradiation in T1-2N1M0 breast cancer: A joint analysis of 4,243 real-world cases from two institutions. Front Oncol 2022;12:955381. [Crossref] [PubMed]
7. Sittenfeld SMC, Zabor EC, Hamilton SN, et al. A multi-institutional prediction model to estimate the risk of recurrence and mortality after mastectomy for T1-2N1 breast cancer. Cancer 2022;128:3057-66. [Crossref] [PubMed]
8. Jagsi R, Barlow WE, Woodward WA, et al. Radiotherapy Use and Incidence of Locoregional Recurrence in Patients With Favorable-Risk, Node-Positive Breast Cancer Enrolled in the SWOG S1007 Trial. JAMA Oncol 2023;9:1083-9. [Crossref] [PubMed]
9. Bahardoust M, Yarahmadi D, Rad FN, et al. Association of a high versus low number of negative lymph nodes removed with survival and recurrence-free survival after lymph node dissection in breast cancer: a systematic review and meta-analysis of observational studies. Clin Exp Med 2025;26:88. [Crossref] [PubMed]
10. Bahardoust M, Ranjbarpazuki A, Rad FN, et al. The effect of the number of negative lymph nodes removed during surgery on the survival and recurrence rates of patients with breast cancer after surgery. BMC Surg 2026;26:139. [Crossref] [PubMed]
11. Meng X, Hao F, Wang N, et al. Log odds of positive lymph nodes (LODDS)-based novel nomogram for survival estimation in patients with invasive micropapillary carcinoma of the breast. BMC Med Res Methodol 2024;24:90. [Crossref] [PubMed]
12. Huang X, Xu X, Xu A, et al. Exploring the most appropriate lymph node staging system for node-positive breast cancer patients and constructing corresponding survival nomograms. J Cancer Res Clin Oncol 2023;149:14721-30. [Crossref] [PubMed]
13. Li Y, Wu G, Liu J, et al. Log odds of positive lymph nodes as a novel prognostic predictor for gastric cancer: a systematic review and meta-analysis. BMC Cancer 2023;23:523. [Crossref] [PubMed]
14. Wang Z, Li F, Zhu M, et al. Prognostic prediction and comparison of three staging programs for patients with advanced (T2-T4) esophageal squamous carcinoma after radical resection. Front Oncol 2024;14:1376527. [Crossref] [PubMed]
15. Jang BS, Shin KH. Postmastectomy Radiation Therapy in Patients With Minimally Involved Lymph Nodes: A Review of the Current Data and Future Directions. J Breast Cancer 2022;25:1-12. [Crossref] [PubMed]
16. Zheng C, Huang J, Zhao S, et al. Log odds of positive lymph nodes as a prognostic marker in resectable triple negative breast cancer. Sci Rep 2025;15:30348. [Crossref] [PubMed]
17. de Boniface J, Filtenborg Tvedskov T, Rydén L, et al. Omitting Axillary Dissection in Breast Cancer with Sentinel-Node Metastases. N Engl J Med 2024;390:1163-75. [Crossref] [PubMed]
18. Tinterri C, Canavese G, Gatzemeier W, et al. Sentinel lymph node biopsy versus axillary lymph node dissection in breast cancer patients undergoing mastectomy with one to two metastatic sentinel lymph nodes: sub-analysis of the SINODAR-ONE multicentre randomized clinical trial and reopening of enrolment. Br J Surg 2023;110:1143-52. [Crossref] [PubMed]
19. Bilani N, Patel R, Crowley F, et al. Outcomes in Premenopausal Patients with HR+/HER2- Breast Cancer and Lymph Node Micrometastasis Based on the 21-Gene Recurrence Score. Oncologist 2023;28:1049-54. [Crossref] [PubMed]
20. Balla A, Weaver DL. Pathologic Evaluation of Lymph Nodes in Breast Cancer: Contemporary Approaches and Clinical Implications. Surg Pathol Clin 2022;15:15-27. [Crossref] [PubMed]
21. Lim SZ, Kusumawidjaja G, Mohd Ishak HM, et al. Outcomes of Stage I and II Breast Cancer with Nodal Micrometastases Treated with Mastectomy without Axillary Therapy. Breast Cancer Res Treat 2021;189:837-43. [Crossref] [PubMed]
22. Zheng J, Huang B, Chen Y, et al. Effect of post-mastectomy radiation therapy on survival in breast cancer with lymph nodes micrometastases: a meta-analysis and systematic review. Front Oncol 2025;15:1489390. [Crossref] [PubMed]
23. Luo H, Yang OO, He JL, et al. Impact of Post-Mastectomy Radiation Therapy for Sentinel Lymph Node Micrometastases in Early-Stage Breast Cancer Patients. Med Sci Monit 2022;28:e933275. [Crossref] [PubMed]