Efficacy of endocrine therapy alone versus endocrine therapy combined with radiotherapy after breast-conserving surgery in older patients with early-stage estrogen receptor-positive breast cancer: a meta-analysis

Abstract

Background

In the general adult female population, the benefit of adjuvant radiotherapy following breast-conserving surgery (BCS) is well established. Nevertheless, whether radiotherapy should be routinely added after BCS for older women with estrogen receptor-positive (ER+) breast cancer remains controversial. The present meta-analysis aimed to evaluate the efficacy and clinical value of endocrine therapy (ET) plus radiotherapy relative to ET alone in this population.

Methods

A systematic search was conducted across the Cochrane Library, PubMed, Embase, and Web of Science. Studies comparing ET plus radiotherapy versus ET alone in older (≥ 60 years) women with ER+ breast cancer after BCS were eligible. The primary conclusions were drawn exclusively on randomized controlled trials (RCTs), with findings from cohort studies serving as supporting evidence. Methodological quality was evaluated utilizing the National Institutes of Health Quality Assessment Tools. The meta-analysis was conducted in STATA (v15.0). Pooled hazard ratios (HRs) and 95% confidence intervals (CIs) were computed. Appropriate models were chosen for pooling results depending on between-study heterogeneity. Leave-one-out sensitivity analyses were performed to evaluate the robustness of findings. Outcomes comprised breast cancer-specific survival (BCSS), overall survival (OS), distant metastasis rate (DMR), local recurrence rate (LRR), and 5-year disease-free survival (DFS). Subgroup analyses were conducted by age, geographic region, ET regimen, radiotherapy regimen, and nodal status. For LRR, the absolute risk reduction (ARR) and number needed to treat (NNT) were calculated.

Results

Twenty studies (9 RCTs, 11 cohort studies) met the eligibility criteria. The primary analysis based on RCTs revealed that radiotherapy plus ET reduced the risk of local recurrence (HR = 0.33, 95% CI: 0.18–0.60); the corresponding ARR was 3.2%, and NNT was 31. No statistically significant differences were observed between the two groups in OS (HR = 1.00, 95% CI: 0.89–1.13), BCSS (HR = 0.96, 95% CI: 0.48–1.93), DMR (HR = 1.12, 95% CI: 0.67–1.87), or 5-year DFS (HR = 0.64, 95% CI: 0.35–1.16). The results from cohort studies were reported separately as supporting evidence.

Conclusion

For older women with early-stage ER+ breast cancer after BCS, adjuvant radiotherapy combined with ET reduces the risk of local recurrence, yet this combination demonstrates no statistically significant improvement in OS, BCSS, DMR, or 5-year DFS. The absolute benefit of radiotherapy is modest. Treatment decisions should involve individualized trade-offs incorporating the risk of recurrence, burden of comorbidities, life expectancy, and patient preferences.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-026-04417-0.

Background

Breast carcinoma ranks as the most prevalent malignancy in females, with incidence peaking between the ages of 45 and 55 [1]. In older adults, breast cancer exhibits clinical characteristics distinct from those in younger individuals. Management in this population requires a comprehensive evaluation considering physiological status, disease profile, presence of comorbidities, treatment tolerance, and estimated life expectancy. Breast tumors in older adults typically feature lower histological grade and a higher proportion of estrogen receptor-positive (ER+) and human epidermal growth factor receptor 2 (HER2)-negative cases [2]. Consequently, the sensitivity and potential benefit of endocrine therapy (ET) are comparatively more defined in this population.

Breast-conserving surgery (BCS), a widely adopted surgical approach, preserves breast morphology and alleviates the psychological impact related to mastectomy. BCS followed by adjuvant radiotherapy yields recurrence and mortality rates comparable to those of mastectomy [3]. Statistics indicate that over half of individuals with stage I-II disease are candidates for BCS plus radiotherapy [4]. This combined modality can reduce the 5-year recurrence rate of ipsilateral breast tumors to a range of 2% to 36% [5].

Treatments for older adults with breast cancer usually deviate from standard guidelines, and the likelihood of receiving post-BCS radiotherapy is lower [6]. The adverse effects of radiotherapy, such as cutaneous, cardiac, and hematological toxicities, are often less tolerable for older than for younger individuals. A careful appraisal of the benefits versus risks of radiotherapy for older individuals is therefore essential. Some randomized controlled trials (RCTs) have demonstrated the safety of omitting radiotherapy after BCS for specific individuals. However, these RCTs primarily enrolled individuals under 65 years, and the generalizability of their findings to the older population is constrained [7]. Debate persists regarding the necessity of post-BCS radiotherapy for older individuals. Research by Jennifer and Eleanor et al. suggests that adjuvant radiotherapy post-BCS reduces mortality in older adults, supporting its necessity [8]. Similarly, the PRIMR II trial finds that among older individuals with ER+ tumors after BCS, the local recurrence rate (LRR) is significantly higher in those receiving ET alone than in those receiving ET plus radiotherapy, whereas overall survival (OS) and distant metastasis did not differ significantly between groups [9].

Existing RCTs on this topic have typically used local recurrence as the primary endpoint and may have insufficient statistical power to assess survival outcomes. Meta-analysis, by pooling data across multiple studies, can enhance the statistical power for evaluating survival endpoints and assess the possible survival benefit of radiotherapy more reliably. This is also a key motivation for the present research.

Methods

This meta-analysis was implemented per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and registered a priori on the International Prospective Register of Systematic Reviews (CRD420251003559) [10], https://www.crd.york.ac.uk/PROSPERO/recorddashboard.

Data sources

A systematic search was conducted across the Cochrane Library, Embase, PubMed, and Web of Science up to November 14, 2025. The search strategy (Table S1) combined medical subject headings and free-text words associated with breast cancer, estrogen receptor-positive, breast-conserving surgery, endocrine therapy, and radiotherapy. No language restrictions were imposed. Table 1 (RCTs) and Table 2 (cohort studies) present the details.

Table 1.

Baseline characteristics of studies (RCTs)

Study	Country	Design	Stage	LN	ET (N)	ET Age	ET + RT (N)	ET + RT Age	Outcome
Kunkler 2015	UK	RCT	T1-T2	N0	668	70	658	69	LR, DM, LR
Hughes 2004	UK	RCT	T1N	N0	319	73.62	317	74.24	OS, DM
Winzer 2010	DE	RCT	T1	N0	80	66.5	94	65.3	DFS
Fastner 2019	AU	RCT	T1-T2	N0	430	67.5	439	63.2	OS, LR, DFS
PÖTTER 2006	USA	RCT	T1-T2	N0	417	61	414	61	OS, DM, LR, DFS
Hughes 2013	USA	RCT	T1	N0	319	70	317	70	OS, BCSS, DM, LR
Kunkler 2023	USA	RCT	T1-T2	N0	668	71.1	658	78	BCSS, DM, LR
Williams 2024	USA	RCT	T1-T2	N0-N1	294	65	291	66	OS, LR
Dalberg 1997	USA	RCT	T1-T2	N0	219	60.3	213	60.7	OS, BCSS, DM, LR

Table 2.

Baseline characteristics of studies (cohort studies)

Study	Country	Design	Stage	LN	ET (N)	ET Age	ET + RT (N)	ET + RT Age	Outcome
Goldberg 2019	USA	Cohort	T1-T2	N0	719	74	1964	70	LR
Nichol 2017	UK	Cohort	T1	N0	507	75.5	3294	74	OS, DFS
Zhi 2019	CN	Cohort	T1-T2	N0-N1	108	70	70	72	OS, DM, LR, DFS
Wei 2021	USA	Cohort	T1-T2	N0	131	72	230	72	OS
Ali 2018	USA	Cohort	T1-2	N0	1549	72.8	4139	72.05	OS
Khan 2013	USA	Cohort	T1	N0	122	73	102	71	OS, BCSS, DM, LR
Yang 2022	CN	Cohort	T1-2	N0	9725	75.6	15,591	74.7	OS, BCSS
Morris 2024	USA	Cohort	T1	N0	148	74.4	307	74.9	OS, DM, LR, DFS
Han 2023	CN	Cohort	T1-2	N0-N1	4837	76.6	12,310	74.7	OS, BCSS
Tringale 2021	USA	Cohort	T1	N0	233	75	398	69	OS, BCSS, LR, DM
Soyder 2013	Greece	Cohort	T1-T2	N0	21	74.5	19	74.5	LR

Abbreviations: LN Lymph node, ET Endocrine therapy, RT Radiotherapy

Country codes: CN (China), USA (United States), UK (United Kingdom), JP (Japan), DE (Germany), AU (Australia)

Study type: RCT (Randomized Controlled Trial)

Outcome measures: LR (Local Recurrence), DM (Distant Metastasis), OS (Overall Survival), DFS (Disease-Free Survival), BCSS (Breast Cancer-Specific Survival)

Tumor staging [41]: T1: Tumor size < 2 cm; T2: Tumor size 2–5 cm; N0: No axillary lymph node metastasis; N1: Metastasis in 1–3 movable ipsilateral axillary lymph nodes without conglomeration; M0: No distant metastasis

Literature screening

Wang and Huang independently screened the retrieved records using EndNote, resolving discrepancies through discussion. The inclusion criteria, formulated according to the PICOS framework, were: Participants (P): females aged ≥ 60 years with ER+ early-stage breast cancer treated with BCS; Intervention (I): postoperative ET combined with radiotherapy; Comparison (C): postoperative ET alone; Outcomes (O): primary outcomes encompassed breast cancer-specific survival (BCSS) and OS; secondary outcomes comprised recurrence of ipsilateral breast tumor, DMR, and 5-year disease-free survival (DFS); Study design (S): RCTs and cohort studies.

Exclusion criteria encompassed: (i) duplicate records or studies with overlapping data; (ii) research lacking outcome data; (iii) non-original research (e.g., letters, comments, reviews, guidelines); (iv) animal studies, conference abstracts, case reports, and single-arm studies.

Risk of bias assessment

Wang and Huang evaluated the risk of bias of eligible studies utilizing the National Institutes of Health (NIH) Quality Assessment Tools. Liu then aggregated the results. For RCTs, the NIH tool for intervention studies was adopted to evaluate aspects such as clarity of the research question, appropriateness of randomization and allocation concealment, comparability of baseline characteristics, intervention clarity, blinding, outcome definition and measurement, follow-up completeness, and statistical analysis. For cohort studies, the NIH tool for cohort studies was applied to evaluate study question and population definition, participation rate, objectivity and reliability of exposure and outcome measurement, follow-up duration, and identification and control of potential confounders. Based on overall scores, studies were categorized as good, fair, or poor quality.

Data extraction

Luo and Huang independently extracted data using a predefined form. Disagreements were settled through discussion with Wu. Collected information encompassed publication year, study design, first author, country or region, sample size, age, disease stage, surgical details, and intervention specifics. Primary outcomes were OS and BCSS. Secondary outcomes were recurrence of ipsilateral breast tumor, local recurrence, DMR, and 5-year DFS.

Data synthesis and statistical analyses

Hazard ratios (HRs) with 95% confidence intervals (CIs) were used as the effect measure. RCTs were defined a priori as the primary evidence base for causal inference, whereas cohort studies were analyzed separately and used only as supportive evidence. For studies not directly reporting HRs, survival data were derived from Kaplan-Meier (K-M) curves using Engauge Digitizer, and HRs with 95% CIs were estimated using the method by Tierney et al. [11]. Meta-analysis, generation of forest plots, and pooling of effect sizes were completed in Stata/MP (v15.1). Heterogeneity was evaluated via the Q-test and I² statistic. A fixed-effects model was adopted for low heterogeneity (I² ≤ 50% or P > 0.05), and a random-effects model for substantial heterogeneity (I² > 50% or P < 0.05). Pre-specified subgroup analyses by country/region, age, and study design were implemented to investigate the sources of heterogeneity. Leave-one-out sensitivity analyses were carried out to evaluate the robustness of the findings. Publication bias was evaluated using funnel plots and Egger’s test when more than 10 studies were available for a given outcome. A trim-and-fill method was applied to estimate the possible influence of missing studies if substantial publication bias was noted.

Results

Study selection

The initial search yielded 1,697 articles. After deduplication, 1,645 records were screened by title and abstract. A total of 103 articles remained for full-text review. Of these, 83 were subsequently excluded for not meeting the PICOS criteria (n = 54), inappropriate study design (n = 14), unavailable data (n = 12), and mean age outside the eligibility range (n = 3). The remaining 20 studies—9 RCTs and 11 cohort studies—were included in the final meta-analysis [9, 12–30]. Figure 1 illustrates the screening process.

Fig. 1.

PRISMA flowchart of literature screening

Study and patient characteristics

The 20 eligible studies (9 RCTs, 11 cohort studies) were published from 1997 to 2024. Participants were females aged ≥ 60 years with ER+, stage T1-T2, node-negative or node-positive (N0-N1) breast cancer treated with BCS. ET regimens comprised tamoxifen or aromatase inhibitors. Radiotherapy regimens comprised whole-breast irradiation or fractionated radiotherapy. None of the 20 studies reported the status of HER2. Tables 1 and 2 present the details.

Risk of bias assessment

The current meta-analysis encompassed 20 studies. Methodological quality was evaluated utilizing the NIH Assessment Tools for RCTs and cohort studies. Among the 9 RCTs, 8 were rated as good and 1 as fair. The fair rating was attributable to a lack of blinding of personnel, participants, and outcome assessors, insufficient reporting on the similarity of baseline characteristics, a dropout rate exceeding 15% at the study endpoint, and the absence of a pre‑specified subgroup analysis plan. Of the 11 cohort studies, 1 was rated as good, 8 as fair, and 2 as poor. The poor ratings stemmed from an incompletely defined study population, the absence of a power calculation, imprecise exposure measurement, and a lack of repeated exposure assessments.

Analysis of outcomes

OS

Six RCTs contributed data for the analysis of OS, yielding a pooled effect estimate of HR = 1.00 (95% CI: 0.89–1.13). This indicated that adjuvant radiotherapy plus ET did not significantly improve OS relative to ET alone in older women with early-stage ER+ breast cancer after BCS (Fig. 2). No notable between-study heterogeneity was detected (I² = 0.0%). Pre-specified subgroup analyses by age, geographic region, ET regimen, radiotherapy regimen, and nodal status (N0/N1) yielded consistent effect directions across all subgroups, with no statistically significant benefit of radiotherapy on OS observed, suggesting no differential effects across subgroups (Table S2). To evaluate the possible influence of reconstructing HRs from Kaplan–Meier curves, RCTs requiring such reconstruction were further excluded. After exclusion, the pooled HR for OS was 0.94 (95% CI: 0.71–1.26), remaining consistent with the primary analysis and failing to reach statistical significance.

Fig. 2.

Overall effect for OS based on RCTs

A separate analysis of 9 cohort studies yielded a pooled HR of 0.83 (95% CI: 0.61–1.14) for OS, again revealing no significant difference between radiotherapy plus ET and ET alone. Given the design of cohort studies, these results served only as supporting evidence (Fig. S1).

LRR

Seven RCTs were included in the analysis of LRR. The pooled HR was 0.33 (95% CI: 0.18–0.60), suggesting that radiotherapy plus ET reduced the risk of local recurrence relative to ET alone (Fig. 3). Considerable between-study heterogeneity was observed (I² = 73.5%). According to the pre-specified subgroup analyses stratified by age, geographic region, ET regimen, radiotherapy regimen, and nodal status (N0/N1), radiotherapy plus ET was associated with a lower risk of local recurrence in the 60–70 years group, the Americas region, the N0 subgroup, the tamoxifen subgroup, the aromatase inhibitor subgroup, and the whole-breast radiotherapy subgroup. In contrast, no statistically significant differences were found in the 70–80 years group, Europe, Australia, or the fractionated radiotherapy subgroup (Table S3). Because several subgroups comprised only 1 or 2 studies, these results should be cautiously interpreted. The ARR and NNT calculated from the RCT data were 3.2% and 31, respectively.

Fig. 3.

Overall effect for LRR based on RCTs

A separate analysis of 6 cohort studies produced a pooled HR of 0.75 (95% CI: 0.25–2.20) for LRR, which did not attain statistical significance (Figure S2). Given the potential for selection bias inherent in cohort studies, these results served only as supporting evidence and did not inform the primary inference.

BCSS

Three RCTs were available for the analysis of BCSS. The pooled HR was 0.96 (95% CI: 0.48–1.93), indicating no significant difference between radiotherapy plus ET and ET alone in terms of BCSS. Between-study heterogeneity was low to moderate (I² = 39.8%). Owing to the limited number of RCTs, no further subgroup analyses were conducted. The overall forest plot for BCSS based on RCTs is illustrated in Fig. 4.

Fig. 4.

Overall effect for BCSS based on RCTs

A separate analysis of 4 cohort studies yielded a pooled HR of 0.85 (95% CI: 0.38–1.90) for BCSS, likewise failing to reach statistical significance. These findings served only as supporting evidence in light of potential selection bias. The forest plot for BCSS from cohort studies is presented in Fig. S3.

DMR

Six RCTs provided data for the analysis of DMR. The pooled HR was 1.12 (95% CI: 0.67–1.87), suggesting no significant difference in the risk of distant metastasis between radiotherapy plus ET and ET alone (Fig. 5). Between-study heterogeneity ranged from moderate to high (I² = 64.3%). Pre-specified subgroup analyses by age, geographic region, ET regimen, and radiotherapy regimen revealed no statistically significant differences in any subgroup (Table S4). Because certain subgroups contained only 1 study, these results should be interpreted with caution.

Fig. 5.

Overall effect for DMR based on RCTs

A separate analysis of 4 cohort studies produced a pooled HR of 0.76 (95% CI: 0.44–1.29) for DMR, which similarly did not reach statistical significance (Fig. S4).

5-year DFS

Three RCTs were incorporated into the analysis of 5-year DFS. The pooled HR was 0.64 (95% CI: 0.35–1.16), suggesting no statistically significant difference between radiotherapy plus ET and ET alone in terms of 5-year DFS. Between-study heterogeneity was high (I² = 63.4%). Further subgroup analyses were not carried out due to the limited number of RCTs. The overall forest plot for 5-year DFS from RCTs is provided in Fig. 6.

Fig. 6.

Overall effect for 5-year DFS based on RCTs

A separate analysis of 3 cohort studies yielded a pooled HR of 1.46 (95% CI: 1.04–2.04) for 5-year DFS (Fig. S5). Given potential selection bias, these results served only as supporting evidence and did not contribute to the primary inference.

Leave-one-out sensitivity analyses were conducted for each outcome. Sequentially omitting any single study did not materially alter the direction of the pooled effect for any outcome. Detailed results are presented Figs. S6–S10.

Additionally, studies requiring data extraction from curves and estimation of HRs were excluded to assess the potential impact of reconstructing HRs from Kaplan–Meier curves. After exclusion, the pooled HR for OS was 0.94 (95% CI: 0.71–1.26), remaining statistically non-significant. The pooled HR for LRR was 0.24 (95% CI: 0.11–0.48), retaining statistical significance. The pooled HR for DMR was 1.12 (95% CI: 0.46–2.70), showing no statistically significant difference. For BCSS, only two RCTs remained after excluding relevant studies, and the pooled analysis was not performed. For 5-year DFS, none of the included RCTs required reconstruction of HRs from Kaplan–Meier curves; thus, this sensitivity analysis was not performed.

Owing to the limited number of included RCTs, formal publication bias testing was not undertaken for the primary outcomes.

Summary of results

In aggregate, the primary analysis based on RCTs demonstrated that radiotherapy plus ET reduced the risk of local recurrence, whereas no statistically significant effects emerged for OS, BCSS, DMR, or 5-year DFS. The results from cohort studies were reported separately as supporting evidence.

Discussion

Drawing on RCTs as the primary evidence source, the present meta-analysis evaluated the efficacy of radiotherapy combined with ET versus ET alone in older women (≥ 60 years) with early-stage ER+ breast cancer after BCS. The findings indicate that radiotherapy plus ET reduces the risk of local recurrence without producing significant improvements in OS, BCSS, DMR, or 5-year DFS.

These results align with the general conclusions of landmark RCTs, such as PRIME II [9] and CALGB 9343 [13], which indicate that in older women with ER+ breast cancer after BCS, the principal benefit of radiotherapy lies in reducing the risk of local recurrence, with no clear long-term survival advantage. The ASKAL RCT [31] and other studies focused on populations with low risk of baseline recurrence similarly suggest that adding radiotherapy does not confer a statistically significant benefit in OS or 5-year DFS in these patients [28]. A retrospective cohort study by Buszek et al. [32] likewise reports no statistically significant improvement in OS with ET alone or radiotherapy alone. From a clinical standpoint, the gain in local control conferred by radiotherapy plus ET is accompanied by a relatively limited absolute benefit (ARR = 3.2%, NNT = 31), indicating that approximately 31 patients would require treatment to avert 1 local recurrence. For older patients, this modest benefit must be weighed against the burden of comorbidities, life expectancy, treatment tolerance, quality-of-life considerations, and economic costs. Prior work also suggests that the risk of local recurrence after ET alone is not uniform across all patients and may vary with individual clinicopathological features [33]. Consequently, the decision to add radiotherapy should rest on individualized risk stratification rather than a uniform criterion.

Radiotherapy is an established adjuvant treatment after BCS. According to the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), adjuvant breast irradiation following BCS reduces local recurrence and may be associated with a reduction in long-term breast cancer mortality, and greater benefit emerges in patients with larger absolute reductions in local recurrence [3]. In the present study, the reduction in LRR observed with radiotherapy plus ET was broadly consistent with the direction of evidence from prior randomized and observational studies [34–36]. Given that older patients with ER+ breast cancer more commonly present with biologically less aggressive tumors and tend to have lower baseline risks of ipsilateral recurrence [37], the absolute clinical value of improved local control may vary across individuals. In this setting, local radiotherapy further reduces the risk of local recurrence by inducing DNA damage and suppressing the proliferation of residual tumor cells [38]. After the exclusion of studies requiring reconstruction of HRs from Kaplan–Meier curves, the direction of the LRR effect was not materially altered, and the result remained statistically significant, suggesting the robustness of this finding in sensitivity analysis. Nevertheless, the result for LRR should still be interpreted cautiously, as heterogeneity for this outcome remained substantial. In clinical practice, any potential local-control benefit of radiotherapy should be weighed against treatment burden, including toxicity, quality-of-life effects, inconvenience, and financial cost, particularly in older patients with substantial comorbidity or limited life expectancy.

Notably, most included studies employed conventional five-week whole-breast radiotherapy, whereas modern radiotherapy departments increasingly use hypofractionated regimens (e.g., three-week or one-week schedules) or accelerated partial breast irradiation. These modalities offer lower toxicity and better tolerability—features particularly suited to older patients. The present findings indicate that ET combined with radiotherapy improves the control of local recurrence. Future investigations could focus on the benefits of modern approaches of radiotherapy in this population.

DMRs exhibited no significant differences between groups. This may be related to the follow-up duration relative to life expectancy in older populations. Furthermore, DMRs are higher in younger patients [39] and strongly linked to molecular subtypes like triple-negative or HER2 + breast cancer, which more commonly metastasize to the brain or lung [40]. In the included studies, follow-up duration was five to ten years for most and three years for a minority. Although the meta-regression of follow-up duration for distant metastasis showed no statistical significance, insufficient follow-up in some studies may have led to the underestimation of distant metastases. This suggests a direction for future research using extended follow-up.

The clinical relevance of this meta-analysis rests on its focus on patients aged ≥ 60 years, a population often underrepresented in clinical trials. The findings suggest that in older women with early-stage ER+ breast cancer undergoing BCS, the primary advantage of radiotherapy consists in lowering the risk of local recurrence, with no clear long-term survival benefit. Treatment decisions should therefore place greater weight on individualized trade-offs. For patients with a low risk of local recurrence, substantial burden of comorbidities, limited life expectancy, or poor radiotherapy tolerance, omitting radiotherapy may help reduce treatment burden. For those with a higher risk of local recurrence, adding radiotherapy may enhance local control. This approach aligns with the principles of shared decision-making in current clinical practice, emphasizing individual risk profiles and patient preferences.

Several limitations merit consideration. First, the number of included RCTs was relatively modest (n = 9), particularly for outcomes such as BCSS and 5-year DFS. This limits statistical power and prevents further subgroup analyses for certain endpoints. Second, substantial to high heterogeneity was present for LRR, DMR, and 5-year DFS. The corresponding results should thus be interpreted with caution. Third, some HRs were reconstructed from Kaplan–Meier curves, which may introduce additional uncertainty. Although sensitivity analyses indicated that the direction of effects for the primary outcomes remained largely unchanged, potential bias stemming from this source cannot be entirely ruled out. Fourth, the majority of primary studies provided insufficient reporting of comorbidity burden, functional status, adverse events, quality of life, and adherence to ET, limiting a comprehensive assessment of the overall clinical value of radiotherapy. Furthermore, follow-up duration varied across studies, and long-term outcomes such as 10-year OS and late toxicities were inconsistently reported. This limits the evaluation of the long-term impact of radiotherapy. Finally, the sample size for patients aged over 80 years was clearly inadequate. The optimal treatment strategy for this subgroup awaits further investigation.

Conclusion

In women aged ≥ 60 years with early-stage ER+ breast cancer after BCS, adjuvant radiotherapy combined with ET reduces LRR but does not yield improvements in OS, BCSS, DMR, or 5-year DFS. Cohort studies provide supporting evidence, yet these should not be used to infer primary conclusions given the constraints of selection bias. The absolute benefit of radiotherapy is modest, and treatment decisions should be highly individualized to integrate the risk of recurrence, comorbidities, life expectancy, and treatment preferences. Well-designed RCTs with extended follow-up durations are necessary to corroborate these findings and to refine de-escalation strategies for this specific population.

Abbreviations

BCS

Breast-conserving surgery

BCSS

Breast cancer-specific survival

CIs

Confidence intervals

DFS

Disease-free survival

DMR

Distant metastasis rate

ER+

Estrogen receptor-positive

ET

Endocrine therapy

HER2

Human epidermal growth factor receptor 2

HRs

Hazard ratios

LRR

Local recurrence rate

NIH

National Institutes of Health

OS

Overall survival

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

RCTs

Randomized controlled trials

Authors’ contributions

Xiuling Wu: Writing- Original draft preparation, Conceptualization, Methodology, Software. Yifu Wang: Data curation, Writing- Original draft preparation. Xiaobin Huang: Visualization, Investigation. Jiqing Liu: Supervision. Bo Luo: Software, Validation. Shifen Huang: Writing- Reviewing and Editing, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that they did not receive any funding from any source.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.