Personalized survival prediction in young Asian American breast cancer

Introduction

Breast cancer is the most commonly diagnosed malignancy among women worldwide and remains a leading cause of cancer-related mortality (1). While the incidence of breast cancer increases with age, young women diagnosed with breast cancer often face unique clinical challenges compared to their older counterparts (2,3). Specifically, young breast cancer patients are more likely to present with aggressive tumor subtypes, including triple-negative breast cancer (TNBC) and human epidermal growth factor receptor 2 (HER2) positive disease, both of which are associated with poorer prognoses (4). Additionally, treatment decisions in young patients must account for factors such as fertility preservation, long-term endocrine therapy adherence, and the psychosocial impact of the disease.

Asian American women represent a rapidly growing United States (US) demographic with concerning breast cancer trends. Recent data show that young Asian American women’s breast cancer incidence has risen to match that of White women (5). Since 2000, breast cancer incidence in Asian American and Pacific Islander women under 50 has increased by 50%, with an annual growth rate exceeding 2% since 2012 (6). This trend likely stems from westernized lifestyle changes, delayed childbearing, dietary shifts, and increased screening (7). Asian American women’s higher breast density further complicates early detection while increasing cancer risk (8). Immigration status significantly impacts risk, with immigrant Asian women facing nearly double the risk compared to US-born counterparts due to lifestyle and environmental changes (9). Moreover, dietary patterns, such as high soy intake and lower consumption of red meat and refined carbohydrates, have been suggested as potential protective factors, though further research is needed to confirm these associations (10). Significant heterogeneity exists among Asian subgroups, with Korean, Chinese, Filipino, and South Asian American women showing higher incidence rates. Filipino and Pacific Islander women face 30% higher mortality rates than White women, highlighting the need for targeted interventions (11).

Despite these unique epidemiological characteristics, there remains a significant gap in prognostic models specifically tailored to young Asian American breast cancer patients. Traditional staging systems, such as the traditional tumor-node-metastasis (TNM) classification, provide general prognostic information but do not fully capture the heterogeneity of tumor biology and patient-specific factors in this population (12). Advances in breast cancer research have led to improved targeted therapies, novel imaging techniques, and enhanced prevention strategies, yet the representation of Asian American women in clinical research remains disproportionately low (13,14). Given the rising incidence and unique risk factors in this population, it is imperative to develop more precise prognostic tools to improve survival prediction and guide personalized treatment approaches.

To address this gap, we aimed to develop and validate a prognostic nomogram specifically designed for young Asian American breast cancer patients. By leveraging data from the Surveillance, Epidemiology, and End Results (SEER) database, we identified independent prognostic factors and constructed a predictive model that estimates overall survival (OS) at 1, 3, and 5 years. This study seeks to provide a clinically applicable tool to aid personalized risk stratification and optimize treatment planning for this unique and historically underserved patient population. We present this article in accordance with the TRIPOD reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0083/rc).

Methods

Patients

This retrospective cohort study utilized data from the SEER database, which covers approximately 35% of the US population (15). We analyzed data from 17 SEER registries for the period 2000–2021, accessed through the SEER*Stat Database. The initial study population comprised 109,317 Asian or Pacific Islander breast cancer cases diagnosed between 2000 and 2021. We excluded patients with multiple primary malignancies (n=27,886), those aged over 40 years or with unknown age (n=35,100), and male patients (n=365). Further exclusions were made for cases with unknown marital status (n=3,344), unknown household income (n=1), and unknown tumor grade (n=9,042). Cases identified solely through death certificates or autopsy (n=87), those with unknown breast cancer subtype (n=25,187), incomplete staging information (M1/MX, T0/TX, NX) (n=2,884), and cases without standard surgical treatment [breast-conserving surgery (BCS) or mastectomy] (n=1,737) were also excluded. Additionally, we removed cases with survival time less than 1 month or unknown survival status (n=512). After applying all inclusion and exclusion criteria, our final analytical cohort consisted of 3,172 patients (Figure S1 and Table 1).

This study used de-identified data from the SEER public database and was exempt from institutional review board approval. All data were accessed and analyzed in accordance with SEER data use agreements. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Data collection and follow-up

For the analysis, marital status was categorized into three groups: married, divorced/separated/widowed (DSW), and unmarried. The household income variable, representing county-level annual median income rather than individual data, was stratified into three levels: low (≤$49,999), medium ($50,000–69,999), and high (≥$70,000). Demographic variables included age at diagnosis (analyzed as a continuous variable), marital status, and income level (categorized based on the median household income of the patient’s county of residence). Tumor characteristics included T stage (T1, T2, T3, T4), N stage (N0, N1, N2, N3), and M stage (M0, M1) according to the American Joint Committee on Cancer (AJCC) 7^th edition staging system. Molecular subtypes were classified based on hormone receptor (HR) and HER2 status into four categories: HR⁺/HER2⁻, HR⁺/HER2⁺, HR⁻/HER2⁺, and HR⁻/HER2⁻ (TNBC). HR positivity was defined as estrogen receptor (ER) and/or progesterone receptor (PR) positivity. Treatment variables included surgical approach (BCS, mastectomy), radiotherapy (yes, no), and chemotherapy (yes, no).

Statistical analysis

The total study population (n=3,172) was randomly divided into a training cohort (n=2,224, 70%) and a validation cohort (n=948, 30%) using a computer-generated randomization sequence. Baseline characteristics between cohorts were compared using chi-square tests for categorical variables and t-tests for continuous variables, with Kaplan-Meier survival analysis and log-rank tests used to compare survival distributions. The primary outcome was OS, defined as the time from diagnosis to death from any cause or the last follow-up date (December 31, 2021). Patients alive at the last follow-up were censored. Univariate Cox proportional hazards regression analysis was performed to identify potential prognostic factors associated with OS. Variables with P<0.05 in univariate analysis were included in the multivariate Cox regression model to identify independent prognostic factors. The proportional hazards assumption was verified using Schoenfeld residuals. A prognostic nomogram was constructed based on independent factors identified in multivariate Cox regression analysis, with each variable assigned a score based on its regression coefficient to predict 1-, 3-, and 5-year OS probabilities. The nomogram’s predictive performance was evaluated using the concordance index (C-index) for discrimination, calibration plots to assess agreement between predicted and observed survival probabilities, time-dependent receiver operating characteristic (tROC) analysis for discriminative ability at different time points, and decision curve analysis (DCA) to evaluate clinical utility. To assess the potential for model overfitting, the events-per-variable (EPV) ratio was evaluated, with a widely accepted minimum safety threshold of 10. Internal validation was performed using bootstrap resampling with 1,000 replicates to assess potential overfitting and calculate an optimism-corrected C-index, while split-sample validation was conducted by applying the nomogram to the internal validation cohort. The nomogram’s performance was compared with the TNM staging system using C-index, tROC analysis, and DCA. All statistical analyses were performed using R software version 4.0.3 with the “rms”, “survival”, “timeROC”, and “rmda” packages, with two-sided P values <0.05 considered statistically significant.

Results

Patient characteristics

Among the initial cohort of 109,317 patients, the missing rates were as follows: molecular subtype 23.0%, tumor grade 8.3%, marital status 3.1%, tumor stage 2.6%, and age and county-level income both <0.1%. A total of 3,172 eligible young Asian American breast cancer patients were included in the study, randomly divided into training (n=2,224) and validation (n=948) cohorts. Baseline characteristics, including demographic factors (marital status, income) and clinical variables (tumor grade, tumor stage, molecular subtype, surgical approach, treatment modalities), showed no significant differences between cohorts (Table 1), minimizing selection bias.

Prognostic factors influencing OS in young Asian American breast cancer patients

Median OS was 66 months (IQR, 35–101 months), with 151 deaths recorded. The 1-, 3-, and 5- OS rates were 99.8%, 97.6%, and 94.5%, respectively. Kaplan-Meier survival analysis demonstrated comparable survival probabilities between cohorts, confirming model generalizability [hazard ratio =0.896; 95% confidence interval (CI): 0.662–1.213; P=0.48; Figure 1].

Figure 1 Kaplan-Meier survival curves comparing training and validation cohorts. Kaplan-Meier curves showing OS probability between training (n=2,224) and validation (n=948) cohorts. No significant difference was observed between the two groups. Numbers at risk were shown below the graph at 12-month intervals. CI, confidence interval; HR, hazard ratio; OS, overall survival.

Univariate and multivariate analyses of OS

In the univariate analysis, eight variables demonstrated a significant association with OS: tumor grade, tumor stage, T stage, N stage, molecular subtype, surgical approach, radiotherapy, and chemotherapy (Table 2). To prevent potential multicollinearity, the overall tumor stage was excluded from further analysis. Subsequently, the remaining seven significant factors, along with marital status, were incorporated into the multivariable Cox regression model. After adjusting for confounding effects, four variables—T stage, N stage, molecular subtype, and surgical approach—emerged as independent prognostic factors for OS.

Development of the integrated prognostic model

We developed a prognostic nomogram incorporating T stage, N stage, molecular subtype, and surgical approach to predict 1-, 3-, and 5-year OS for young Asian American breast cancer patients (Figure 2). The full specifications of the final multivariate Cox model, including the regression coefficients and standard errors, are detailed in Table S1. Additionally, the 1- to 5-year baseline survival estimates for both cohorts are provided in Table S2.

Figure 2 Nomogram development and performance evaluation in the training cohort. Nomogram for predicting 1-, 3-, and 5-year OS. Nomogram incorporating T stage, N stage, molecular subtype, and surgery type for predicting individual patient survival probability. Points are assigned for each variable, and total points correspond to predicted survival probabilities at different time points. BCS, breast-conserving surgery; HER2, human epidermal growth factor receptor 2; HR, hormone receptor; N, node; OS, overall survival; T, tumor.

Assessment of predictive performance of the prognostic model

The prognostic model displayed moderate discriminative performance, achieving a C-index of 0.759 (95% CI: 0.719–0.799) in the training group and 0.763 (95% CI: 0.698–0.827) in the validation group. Comparatively, the traditional staging system showed lower C-index values of 0.736 (95% CI: 0.678–0.793) and 0.726 (95% CI: 0.631–0.821) for the training and validation cohorts, respectively. Compared to traditional TNM staging, our nomogram demonstrated a modest yet clinically meaningful improvement in predictive accuracy. Furthermore, we rigorously assessed the potential for model overfitting. Given that a total of 151 outcome events (deaths) were observed and four independent variables were incorporated into the final multivariate Cox model, EPV ratio was 37.75. This value was well above the recommended threshold of 10, indicating a low risk of model overfitting. Internal validation via 1,000 bootstrap resamples showed minimal performance decay between the training and validation cohorts, with optimism-corrected C-indices of 0.753 in the training cohort and 0.756 in the validation cohort. Calibration plots for 1-, 3-, and 5-year survival showed excellent agreement between predicted and observed outcomes in the training group (Figure 3A). Calibration plots demonstrated strong agreement between predicted and observed survival probabilities at 1-, 3-, and 5-year time points, with curves closely following the ideal reference line in the validation group (Figure 3B). tROC analysis yielded area under the receiver operating characteristic curve (AUC) values of 0.842, 0.813, and 0.795 for 1-, 3-, and 5-year predictions, respectively, consistently outperforming TNM staging (0.798, 0.761, and 0.742) in the training group (Figure 3C). tROC analysis yielded AUC values of 0.831, 0.802, and 0.785 for 1-, 3-, and 5-year predictions, respectively, consistently exceeding TNM staging performance (0.784, 0.753, and 0.731) in the validation group (Figure 3D). DCA confirmed higher net benefits across various threshold probabilities compared to TNM staging and default strategies in the training group (Figure 3E). DCA confirmed higher net benefits compared to TNM staging and default strategies across threshold probabilities from 10% to 80% in the validation group (Figure 3F).

Figure 3 Assessment of predictive performance of the prognostic model. (A) Calibration plot for the training cohort. (B) Calibration plot for the validation cohort. (C) tROC curves for the training cohort. (D) tROC curves for the validation cohort. (E) DCA for the training cohort. (F) DCA for the validation cohort. AUC, area under the receiver operating characteristic curve; DCA, decision curve analysis; OS, overall survival; TNM, tumor-node-metastasis; tROC, time-dependent receiver operating characteristic.

Validation of the prognostic model in the split-sample cohort

The minimal degradation in performance metrics between training and validation cohorts confirms the nomogram’s internal reproducibility and clinical utility for personalized risk assessment in this specific patient population.

Discussion

Our study developed and validated a novel prognostic nomogram for young Asian American breast cancer patients that significantly outperformed the traditional TNM staging system. The nomogram incorporated four independent prognostic factors: T stage, N stage, molecular subtype, and surgical approach. This model demonstrated reasonable discrimination and calibration performance, with consistent superiority over TNM staging across multiple validation metrics.

The prognostic significance of tumor burden in our cohort aligns with established literature but provides specific risk quantification for young Asian American women. T stage showed a clear dose-response relationship with mortality, with hazard ratio increasing progressively from T2 (hazard ratio =1.778; 95% CI: 1.133–2.790) to T4 (hazard ratio =3.946; 95% CI: 1.786–8.715) (16,17). Similarly, lymph node involvement demonstrated a strong prognostic impact, with N2 disease carrying the highest risk (hazard ratio =4.968; 95% CI: 2.944–8.383), followed by N3 (hazard ratio =3.401; 95% CI: 1.723–6.712) and N1 (hazard ratio =1.878; 95% CI: 1.209–2.916) (18,19). These findings are consistent with previous studies in broader populations but highlight the pronounced impact of nodal burden in this specific demographic.

Tumor characteristics showed strong prognostic significance in our cohort. Histological analysis revealed 88.1% of young Asian American breast cancer patients had grades 2–3 tumors, consistent with previously reported rates (82.9–94%) in young women across broader populations (20-22), suggesting the biological aggressiveness of young-onset breast cancer transcends ethnic boundaries. Molecular subtype distribution showed 21.5% HER2-positive and 10.6% TNBC, mirroring previous studies of young breast cancer patients and indicating conserved biological drivers across ethnic groups (22-24). However, TNBC patients faced significantly worse outcomes (hazard ratio =2.640; 95% CI: 1.697–4.105) compared to other subtypes after multivariate adjustment, a concerning finding given the therapeutic challenges of this aggressive subtype.

Interestingly, our study revealed that age, while often considered a critical prognostic factor in breast cancer, did not emerge as an independent predictor of mortality in our multivariate analysis of young Asian American patients. Previous studies have consistently demonstrated that younger age at diagnosis is associated with worse outcomes in the general breast cancer population (25,26), with patients under 40 years exhibiting more aggressive disease features and poorer survival compared to older counterparts. However, our findings suggest that within the young age group (under 40 years), the specific age at diagnosis may not significantly modify risk once other clinicopathological factors are accounted for. This indicates that the prognostic impact of young age may operate primarily through its association with more aggressive tumor biology and advanced stage at presentation, rather than representing an independent risk factor itself within this demographic.

Several investigations have reported that young breast cancer patients with lower socioeconomic status (SES) generally experience worse outcomes, attributed to barriers in healthcare access, delayed diagnosis, and suboptimal treatment (27,28). In light of this context, regarding socioeconomic factors, a notable limitation is our reliance on county-level data as a proxy for individual SES. The absence of a significant survival association with county-level income does not inherently preclude a meaningful SES effect among young Asian American patients. Rather, the aggregate nature of SEER data fails to capture within-county socioeconomic heterogeneity, thereby potentially obscuring the true prognostic impact of individual-level SES. Future prospective studies incorporating granular, individual-level socioeconomic metrics are warranted to definitively elucidate this relationship. Similarly, marital status showed a significant association with survival in univariate analysis but lost statistical significance after multivariate adjustment, suggesting its effect may be mediated through other factors such as earlier detection, treatment compliance, or psychological well-being.

The association between surgical approach and survival outcomes warrants careful interpretation (29-32). Notably, our multivariable analysis indicated an apparent association between mastectomy and worse OS (hazard ratio =1.794; 95% CI: 1.147–2.808). However, it must be strongly emphasized that this finding reflects profound confounding by indication rather than a true causal relationship. In real-world clinical practice, patients who undergo mastectomy frequently present with a substantially higher baseline disease burden—such as larger tumor sizes, multifocal lesions, or specific contraindications to BCS. Consequently, the observed survival disparity is inherently driven by the more aggressive tumor biology necessitating the mastectomy, rather than the surgical modality itself.

Our parsimonious four-variable nomogram was developed through a systematic strategy: integrating significant variables (P<0.05) from univariate Cox regression into a multivariate analysis, followed by backward stepwise selection. Comprehensive model performance was evaluated in both the training and validation cohorts using tROC curves, calibration plots, and DCA. Furthermore, an adequate EPV ratio and 1,000 bootstrap resamplings were employed to prevent overfitting and verify internal stability. Consequently, our validated nomogram offers clinicians a practical tool for personalized risk assessment in young Asian American breast cancer patients, enabling informed prognostic discussions, tailored treatment decisions, and optimized follow-up strategies. Looking forward, future research should enhance this foundation by incorporating genomic signatures, immune parameters, and Asian-specific biomarkers, with prospective validation in contemporary cohorts receiving current standard treatments.

Despite its strengths, there are several limitations in our study. The registry-based retrospective nature of the SEER database analysis introduces potential selection bias and missing data issues. Information on specific systemic therapy regimens, treatment compliance, and recurrence patterns was not available, potentially limiting the comprehensiveness of our prognostic model. A potential limitation of our study is the exclusion of patients with missing data, most notably for molecular subtype (23.0%) and tumor grade (8.3%). We opted for complete-case analysis over multiple imputation because this missingness was considered not at random, which violates key imputation assumptions. Nevertheless, the remaining sample size (n=3,172) and a high EPV ratio (37.75) ensured robust statistical power and reliable model stability without the need for data imputation. Additionally, while focusing exclusively on Asian Americans addresses their historical underrepresentation, it precludes assessing the model’s performance in other ethnicities. Furthermore, stratifying our cohort of 3,172 patients (with only 151 total events) into multiple ethnic subgroups would severely fragment the data. This fragmentation would drastically compromise statistical power and lead to an inadequate EPV ratio, rendering any subgroup-specific analyses statistically unreliable and prone to overfitting. Finally, a notable limitation of this study is the lack of true external validation. Therefore, more studies utilizing external datasets are strictly required to verify the true generalizability and clinical utility of this prognostic model. In the future, evolving treatment paradigms, particularly in HER2-targeted therapy and immunotherapy, may influence the applicability of our findings to patients treated with the most current approaches.

Conclusions

Our study presents reasonable, internally validated prognostic nomogram specifically designed for young Asian American breast cancer patients. By integrating tumor characteristics, molecular subtypes, and treatment approaches, this tool moderately outperforms traditional TNM staging in predicting survival outcomes. The identified independent prognostic factors provide valuable insights into the unique risk profiles of this demographic group and highlight opportunities for tailored interventions. Implementation of this nomogram in clinical practice could facilitate more personalized treatment decisions and ultimately improve outcomes for young Asian American women with breast cancer. Although internally validated with favorable results, the true generalizability of this proposed nomogram must be confirmed through independent, external cohorts in future research.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0083/rc

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-1-0083/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-2026-1-0083/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. Wilkinson AN, Ellison LF, Billette JM, et al. Impact of Breast Cancer Screening on 10-Year Net Survival in Canadian Women Age 40-49 Years. J Clin Oncol 2023;41:4669-77. [Crossref] [PubMed]
2. Wang X, Xia C, Wang Y, et al. Landscape of young breast cancer under 35 years in China over the past decades: a multicentre retrospective cohort study (YBCC-Catts study). EClinicalMedicine 2023;64:102243. [Crossref] [PubMed]
3. Laas E, Dumas E, Hamy AS, et al. The influence of treatment intervals on prognosis in young breast cancer patients: Insights from the French National cohort. Eur J Surg Oncol 2025;51:109373. [Crossref] [PubMed]
4. Poorvu PD, Gelber SI, Rosenberg SM, et al. Prognostic Impact of the 21-Gene Recurrence Score Assay Among Young Women With Node-Negative and Node-Positive ER-Positive/HER2-Negative Breast Cancer. J Clin Oncol 2020;38:725-33. [Crossref] [PubMed]
5. Primm KM, Zhao H, Hernandez DC, et al. A Contemporary Analysis of Racial and Ethnic Disparities in Diagnosis of Early-Stage Breast Cancer and Stage-Specific Survival by Molecular Subtype. Cancer Epidemiol Biomarkers Prev 2022;31:1185-94. [Crossref] [PubMed]
6. Giaquinto AN, Sung H, Newman LA, et al. Breast cancer statistics 2024. CA Cancer J Clin 2024;74:477-95. [Crossref] [PubMed]
7. Kim J, Kim J, Seo KH, et al. Survival outcomes of young-age female patients with early breast cancer: an international multicenter cohort study. ESMO Open 2024;9:103732. [Crossref] [PubMed]
8. Houssami N, Hofvind S, Soerensen AL, et al. Interval breast cancer rates for digital breast tomosynthesis versus digital mammography population screening: An individual participant data meta-analysis. EClinicalMedicine 2021;34:100804. [Crossref] [PubMed]
9. Falcone M, Salhia B, Hughes Halbert C, et al. Impact of Structural Racism and Social Determinants of Health on Disparities in Breast Cancer Mortality. Cancer Res 2024;84:3924-35. [Crossref] [PubMed]
10. Wu AH, Yu MC, Tseng CC, et al. Dietary patterns and breast cancer risk in Asian American women. Am J Clin Nutr 2009;89:1145-54. [Crossref] [PubMed]
11. Wen KY, Fang CY, Ma GX. Breast cancer experience and survivorship among Asian Americans: a systematic review. J Cancer Surviv 2014;8:94-107. [Crossref] [PubMed]
12. de Jong VMT, Wang Y, Ter Hoeve ND, et al. Prognostic Value of Stromal Tumor-Infiltrating Lymphocytes in Young, Node-Negative, Triple-Negative Breast Cancer Patients Who Did Not Receive (neo)Adjuvant Systemic Therapy. J Clin Oncol 2022;40:2361-74. [Crossref] [PubMed]
13. Nguyen TT, Somkin CP, Ma Y, et al. Participation of Asian-American women in cancer treatment research: a pilot study. J Natl Cancer Inst Monogr 2005;102-5.
14. Ma GX, Seals B, Tan Y, et al. Increasing Asian American participation in clinical trials by addressing community concerns. Clin Trials 2014;11:328-35. [Crossref] [PubMed]
15. Che WQ, Li YJ, Tsang CK, et al. How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology. Mil Med Res 2023;10:50. [Crossref] [PubMed]
16. Sun R, Huang Y, Chen X, et al. A nomogram for predicting overall survival of breast cancer with regional lymph node metastasis in young women. Transl Cancer Res 2024;13:542-57. [Crossref] [PubMed]
17. Wang J, Luo T, Xiang ZZ, et al. Survival and Trends in Annualized Hazard Function by Age at Diagnosis Among Chinese Breast Cancer Patients Aged ≤40 Years: Case Analysis Study. JMIR Public Health Surveill 2023;9:e47110. [Crossref] [PubMed]
18. Taparra K, Dee EC, Dao D, et al. Disaggregation of Asian American and Pacific Islander Women With Stage 0-II Breast Cancer Unmasks Disparities in Survival and Surgery-to-Radiation Intervals: A National Cancer Database Analysis From 2004 to 2017. JCO Oncol Pract 2022;18:e1255-64. [Crossref] [PubMed]
19. Zhang D, Zheng Y, Wang T, et al. Lymph node ratio-based model for predicting survival and assessing the benefit of adjuvant chemotherapy in postoperative duodenal adenocarcinoma. Surgery 2025;178:108847. [Crossref] [PubMed]
20. Men DX, Li HZ, Dong J, et al. Correlation between ultrasonography and elastography parameters and molecular subtypes of breast cancer in young women. Ann Med 2025;57:2443041. [Crossref] [PubMed]
21. Redmond CE, Healy GM, Murphy CF, et al. The use of ultrasonography and digital mammography in women under 40 years with symptomatic breast cancer: a 7-year Irish experience. Ir J Med Sci 2017;186:63-7. [Crossref] [PubMed]
22. Bullier B, MacGrogan G, Bonnefoi H, et al. Imaging features of sporadic breast cancer in women under 40 years old: 97 cases. Eur Radiol 2013;23:3237-45. [Crossref] [PubMed]
23. Li Y, Tao X, Ye Y, et al. Prognostic nomograms for young breast cancer: A retrospective study based on the SEER and METABRIC databases. Cancer Innov 2024;3:e152. [Crossref] [PubMed]
24. Huang J, Lin Q, Cui C, et al. Correlation between imaging features and molecular subtypes of breast cancer in young women (≤30 years old). Jpn J Radiol 2020;38:1062-74. [Crossref] [PubMed]
25. Anders CK, Fan C, Parker JS, et al. Breast carcinomas arising at a young age: unique biology or a surrogate for aggressive intrinsic subtypes? J Clin Oncol 2011;29:e18-20. [Crossref] [PubMed]
26. Yang R, Wu Y, Qi Y, et al. A nomogram for predicting breast cancer specific survival in elderly patients with breast cancer: a SEER population-based analysis. BMC Geriatr 2023;23:594. [Crossref] [PubMed]
27. Mark C, Pujara V, Boyle MK, et al. Demographic and clinical trends of young breast cancer patients from the national cancer database: disproportionate effect on minority populations. Breast Cancer Res Treat 2025;210:521-8. [Crossref] [PubMed]
28. Myers SP, Zheng Y, Dibble K, et al. Financial Difficulty Over Time in Young Adults With Breast Cancer. JAMA Netw Open 2024;7:e2446091. [Crossref] [PubMed]
29. Chu J, Yang D, Wang L, et al. Nomograms predicting survival for all four subtypes of breast cancer: a SEER-based population study. Ann Transl Med 2020;8:544. [Crossref] [PubMed]
30. Ye JC, Yan W, Christos PJ, et al. Equivalent Survival With Mastectomy or Breast-conserving Surgery Plus Radiation in Young Women Aged < 40 Years With Early-Stage Breast Cancer: A National Registry-based Stage-by-Stage Comparison. Clin Breast Cancer 2015;15:390-7. [Crossref] [PubMed]
31. Lee CM, Zheng H, Tan VK, et al. Surgery for early breast cancer in the extremely elderly leads to improved outcomes - An Asian population study. Breast 2017;36:44-8. [Crossref] [PubMed]
32. Cortadellas T, Córdoba O, Gascón A, et al. Surgery improves survival in elderly with breast cancer. A study of 465 patients in a single institution. Eur J Surg Oncol 2015;41:635-40.