Optimizing Extended Adjuvant Endocrine Therapy in Early HR+/HER2− Breast Cancer: The Emerging Role of Genomic Assays and ctDNA MRD

Abstract

Hormone receptor-positive (HR+) and HER2-negative (HER2−) breast cancer represents the most prevalent subtype of early-stage breast cancer. Adjuvant endocrine therapy (ET) substantially reduces recurrence and breast cancer mortality; however, late relapse remains a major challenge, with a considerable proportion of recurrences occurring beyond 5 years after diagnosis. Although extended ET can modestly reduce late recurrence, it is associated with cumulative toxicities and impaired quality of life, highlighting the urgent need for biomarkers to identify patients who truly benefit from treatment escalation or extension, while sparing low-risk individuals from overtreatment. Tissue-based multigene expression assays, including the Breast Cancer Index (BCI), EndoPredict, Prosigna, Oncotype DX, and MammaPrint, have improved risk stratification and informed treatment decisions, with BCI demonstrating the strongest evidence for predicting benefit from extended endocrine therapy. Among currently available assays, BCI has the strongest level of evidence supporting its use in guiding extended endocrine therapy decisions. In parallel, liquid biopsy approaches, particularly circulating tumor DNA (ctDNA)/minimal residual disease (MRD) detection, have emerged as promising tools for dynamic monitoring and early detection of molecular relapse. This review summarizes current evidence supporting biomarker-guided decision-making in early-stage HR+/HER2− breast cancer, focusing on three clinically relevant questions: who requires treatment escalation, who benefits from extended endocrine therapy, and who may safely de-escalate. We further discuss challenges and future directions toward integrated models combining genomic assays with longitudinal ctDNA monitoring to refine personalized adjuvant endocrine strategies. However, current evidence remains limited, and prospective validation is required.

Key Summary Points

Late recurrence remains a critical unmet need in early-stage HR+/HER2− breast cancer, with a substantial proportion of relapses occurring beyond 5 years despite standard endocrine therapy.

This study evaluates the complementary roles of tissue-based genomic assays and circulating tumor DNA (ctDNA) minimal residual disease in guiding individualized adjuvant endocrine therapy strategies.

Genomic assays enable risk stratification and selection for extended therapy, while ctDNA provides dynamic monitoring of relapse risk; their integration supports a precision framework for therapy escalation, extension, and de-escalation.

Introduction

Hormone receptor-positive (HR+) and human epidermal growth factor receptor 2-negative (HER2−) breast cancer accounts for the majority of breast cancer diagnoses worldwide and is characterized by a relatively favorable early prognosis but a persistent long-term recurrence risk [1]. HR+/HER2− breast cancer predominantly affects women, accounting for the majority of cases, whereas male breast cancer is rare, comprising less than 1% of all breast cancers. Differences in menopausal status further influence disease biology and treatment strategies [2, 3]. Adjuvant ET, including agents such as tamoxifen, aromatase inhibitors (AIs), and ovarian function suppression (OFS), is the cornerstone of treatment for early-stage HR+/HER2− breast cancer [4, 5]. Despite these established benefits, optimal personalization of adjuvant ET remains a critical clinical challenge. Recent advances in genomic profiling and liquid biopsy technologies are increasingly being incorporated into clinical practice to improve patient outcomes by refining recurrence risk stratification and identifying patients who may benefit from extended therapy or de-escalation [6–8]. Not all patients benefit equally from extended therapy beyond the initial 5-year period [9, 10]. Over half of the recurrences in HR+/HER2− breast cancer occur more than 5 years after initial diagnosis, suggesting that extending ET could reduce late recurrence risk, though the benefit is modest [11]. Current National Comprehensive Cancer Network (NCCN) guidelines recommend extended therapy for high-risk patients, yet this approach is associated with increased toxicity and adverse effects, emphasizing the importance of identifying those who are most likely to benefit [11]. Given the persistent challenge of recurrence risk and the associated toxicity of long-term therapy, biomarkers have an increasingly pivotal role in guiding treatment decisions [12]. Specifically, biomarkers are needed to address three clinically relevant questions: (1) which patients require treatment escalation in the adjuvant setting, (2) which patients benefit from extended endocrine therapy beyond 5 years, and (3) which patients may safely de-escalate or discontinue therapy to avoid unnecessary toxicity. Traditional clinicopathologic factors provide valuable prognostic information but are insufficient to precisely identify endocrine-sensitive tumors with high late recurrence risk. Tools such as the Breast Cancer Index (BCI), which includes the HOXB13/IL17BR ratio and Molecular Grade Index (MGI), have shown potential in predicting late recurrence risk and identifying patients who are most likely to benefit from extended therapy [13–15]. Studies consistently demonstrate that high BCI scores are associated with a greater benefit from extended ET, particularly in postmenopausal women [15, 16].

Advances in sequencing technologies have enabled the integration of genomic data into clinical practice, facilitating the identification of clinically relevant mutations and alterations in HR+/HER2− breast cancer [17]. This development has enhanced our understanding of the molecular diversity within HR+/HER2− breast cancer, offering opportunities for personalized treatment strategies based on individual tumor characteristics. In parallel, liquid biopsy techniques, which analyze ctDNA or other biomarkers from blood samples, are emerging as a valuable tool in monitoring treatment response and detecting early signs of recurrence. Liquid biopsies hold promise in further refining treatment decisions, offering a less invasive method for assessing tumor dynamics over time [18, 19].

This review will explore recent developments in adjuvant endocrine treatments, the role of biomarkers in clinical decision-making, and the future directions for managing HR+/HER2− breast cancer. It will also discuss the growing importance of liquid biopsy and its potential in improving treatment precision, as well as the need for further research to broaden our understanding of biomarker utility across diverse patient populations. We highlight the clinical utility and limitations of tissue-based genomic assays and ctDNA testing, propose an integrated framework for decision-making, and discuss future directions for combining static genomic risk assessment with longitudinal molecular surveillance. This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Clinical Background: Why Late Recurrence Matters in HR+/HER2− Disease

A defining feature of HR+/HER2− breast cancer is the prolonged risk of recurrence extending for decades after initial diagnosis. Multiple studies have demonstrated that a large proportion of recurrences occur after completion of standard 5-year endocrine therapy [20, 21]. This late recurrence phenomenon is clinically significant because it challenges conventional surveillance and treatment paradigms and underscores the importance of long-term preventive strategies.

Biologically, late recurrence is thought to be driven by tumor cell dormancy, where disseminated tumor cells survive in a quiescent state within distant organs, escaping immune surveillance and resisting cytotoxic therapy. Over time, microenvironmental changes and acquired resistance mechanisms may reactivate these dormant cells, leading to metastatic relapse. In addition, endocrine therapy resistance may emerge through alterations in estrogen receptor signaling pathways, epigenetic changes, and activation of alternative growth factor pathways [21, 22].

Clinically, extended endocrine therapy can reduce late recurrence risk; however, the magnitude of benefit varies considerably among individuals. While high-risk patients may experience clinically meaningful reductions in distant relapse, many low-risk patients derive minimal incremental benefit and instead accumulate toxicity [23]. Therefore, accurately distinguishing patients who remain at substantial late recurrence risk and are likely to benefit from therapy extension is central to modern HR+/HER2− breast cancer management.

Current Treatment Options for Adjuvant Endocrine Therapy in Early-Stage HR+/HER2− Breast Cancer

Adjuvant ET is the standard treatment for early-stage HR+/HER2− breast cancer. Ongoing clinical trials aim to refine the optimal agents, duration, and combinations to further improve patient outcomes. A summary of all relevant trials is provided in Table 1.

Table 1.

current treatment options for adjuvant endocrine therapy in early-stage HR+/HER2− breast cancer

Menopausal status	Trial	Treatment	Duration	Multigene assay or ctDNA	Main findings
Premenopause	ASTRRA	Tam vs. Tam + OFS	5 years	N/A	5 years DFS rate:87.5% vs. 91.1%, HR 0.69; 95% CI 0.48–0.97; p value = 0.033; 5 year OS rate: 99.4% vs. 97.8%, hazard ratio, 0.31; 95% CI 0.10–0.94; P = 0.029
	SOFT	Tam vs. Tam + OFS	5 years	BCI (prognostic)	BCI(H/I)-low exhibited a 12-year absolute benefit in BCFI of 11.6% from exemestane plus OFS [hazard ratio (HR), 0.48 (95% CI 0.33–0.71)] and an absolute benefit of 7.3% from tamoxifen plus OFS (HR, 0.69 [95% CI 0.48–0.97)] relative to tamoxifen alone
	Combined analysis of SOFT and TEXT	Tam vs. Tam + OFS vs. exemestane + OFS (SOFT)Tam + OFS vs. exemestane + OFS (TEXT)	5 years	N/A	5 years DFS 86.8% in exemestane + OFS vs. 82.8% in Tam + OFS, HR 0.77; 95%CI 0.67–1.05; p value = < 0.001
	MonarchE	Abemaciclib + any ET	2 years of abemaciclib/placebo + 5–10 years of any ET	N/A	IDFS 92.2% vs. 88.7%, HR 0.75. 95% CI 0.60–0.93; p value = 0.01
	aTToM	Tam	5 year vs. 10 years	N/A	Recurrence rate 32% vs. 28%, RR 0.85. 95% CI 0.76–0.95; p value = 0.003
	Trans-aTTom	Tam	5 year vs. 10 years	BCI (predictive)	BCI (H/I)-High derived a significant benefit from 10 versus 5 years of tamoxifen treatment [hazard ratio (HR): 0.35; 95% confidence interval (CI) 0.15–0.86; 10.2% absolute risk reduction based on recurrence-free interval (RFI), P = 0.027]
	ATLAS trial	AIs	5 years vs. 10 years	N/A	Continued tamoxifen therapy significantly reduced the risk of recurrence [relative risk (RR) 0.84; 95% CI 0.76–0.94; p = 0.002] and overall mortality (639 vs. 722 deaths; p = 0.01)
Postmenopause	IDEAL	AIs	7 years vs. 10 years	N/A	163 vs. 152 DFS events, HR 0.92; 95% CI 0.78–1.45
	ATAC trial	AIs vs. Tam	5 years	N/A	Compare to Tam DFS: HR = 0.91, 95% CI 0.83–0.99, p = 0.04,TTR: HR = 0.84, 95% CI 0.75–0.93, p = 0.001 and time to distant recurrence: HR = 0.87, 95% CI 0.77–0.99, p = 0.03
	AERAS trial	AIs	5 years vs. 10 years	N/A	DFS rate (extend Ais group:91%) compared to the discontinuation group (86%) HR = 0.61
	NSABP B42 trial	AIs	5 years vs. 10 years	N/A	Letrozole extension improved DFS (HR = 0.85, 95% CI 0.74–0.96, p = 0.01) and reduced the risk of distant recurrence (HR = 0.72, 95% CI 0.55–0.92, p = 0.01), but did not affect overall survival
	NSABP B42 trial translational cohort	AIs	5 years vs. 10 years	BCI (predictive)	10-year absolute benefit of 2.4% in RFI compared to those with low BCI scores
	NSABP B42 trial translational cohort	AIs	5 years vs. 10 years	Mammaprint (predictive)	For patients with MP-low risk tumors, extended letrozole therapy improve 10-year DRFI by 3.7%, with a hazard ratio (HR) of 0.43 (95% CI 0.25–0.74; P = 0.002)
	ABCSG6 and ABCSG8 trials	AIs	7 years vs. 10 years	Endopredict (prognostic)	EndoPredict had a significantly better clinical outcome in terms of late recurrence compared to high-risk patients (P = 0.002)
Premenopause + postmenopause	Cutts R. et al.	ET or chemo (adjuvant)	NA	ctDNA (prognostic)	The absence of ctDNA at a single post-surgical timepoint is associated with improved DRFS
	Marla et al.	ET or ET + chemo (adjuvant)	NA	ctDNA (prognostic)	in patients with high-risk HR+ BC in the late adjuvant setting, ctDNA was identified a median of 1 year before all cases of distant metastasis

BCI Breast Cancer Index, DFS disease-free survival, OFS ovarian function suppression

Adjuvant Endocrine Therapy for Pre- and Perimenopausal Patients with HR+/HER2− Breast Cancer

Tamoxifen Monotherapy

Tamoxifen is the cornerstone of adjuvant therapy for premenopausal women with HR+ breast cancer. It significantly reduces recurrence and mortality, with benefits extending up to 15 years. However, its efficacy is lower in ER– tumors [24, 25]. The ATLAS trial [26] examined the impact of extending tamoxifen therapy for an additional 5 years in women who had completed 5 years of adjuvant tamoxifen. Among the 6846 women with ER + breast cancer, continued tamoxifen therapy significantly reduced the risk of recurrence [relative risk (RR) 0.84; 95% CI 0.76–0.94; p = 0.002] and overall mortality (639 vs. 722 deaths; p = 0.01). However, this benefit was accompanied by a small but significant increase in the risk of endometrial cancer-related mortality (0.2%). The aTToM study corroborated these findings, demonstrating that extending tamoxifen reduced recurrence rates but also increased the incidence of endometrial cancer (rate ratio 2.20, p < 0.0001) [27].

OFS Combined with Endocrine Therapy

OFS, achieved through surgical, radiological, or pharmacologic means (e.g., LHRH agonists), enhances the efficacy of tamoxifen or AIs. Several pivotal trials have demonstrated that combining OFS with endocrine therapy improves disease-free survival (DFS) and reduces recurrence risk [11, 28]. The SOFT trial [29] (n = 3066) showed that tamoxifen with OFS significantly improved DFS (83.2% vs. 78.9%, HR 0.76, p = 0.009) and overall survival (OS) (93.3% vs. 91.5%, HR 0.67, p = 0.01) compared to tamoxifen alone, with the greatest benefit in high-risk subgroups, such as women under 35 or those with large or high-grade tumors. In a translational analysis of the SOFT trial, the BCI demonstrated its prognostic value in premenopausal women with HR+ breast cancer. Notably, patients with BCI(H/I)-low tumors experienced greater benefit from adjuvant endocrine therapy that included OFS, compared to those with BCI(H/I)-high tumors. This suggests that BCI(H/I)-low status may serve as a useful biomarker to identify premenopausal patients who are more likely to benefit from the addition of OFS to their endocrine therapy regimen [30]. Similarly, the TEXT-SOFT trial [31] (n = 4690) demonstrated that exemestane with OFS improved DFS by 4.6% and disease recurrence-free interval (DRFI) by 1.8%, with greater benefits observed in younger patients and those with larger, high-grade tumors. Additionally, the ASTRRA trial (n = 1293) found that tamoxifen plus OFS improved 5-year DFS rate (91.1% vs. 87.5%, hazard ratio, 0.69; 95% CI 0.48–0.97, P = 0.033) and 5 year OS rate (99.4% vs. 97.8%, hazard ratio, 0.31; 95% CI 0.10–0.94; P = 0.029) compared to tamoxifen alone, further supporting the inclusion of OFS in high-risk patients [32].

Emerging Therapies: CDK4/6 Inhibition in Combination with Endocrine Therapy

MonarchE Trial

The MonarchE trial [33] evaluated the addition of abemaciclib, a CDK4/6 inhibitor, to standard endocrine therapy in high-risk HR+/HER2− breast cancer. After a median follow-up of 42 months, the combination of abemaciclib and endocrine therapy improved invasive disease-free survival (IDFS) by 6.4% (85.8% vs. 79.4%, HR 0.664, p < 0.0001), though no difference in overall survival was observed. Despite the absence of an OS benefit, the combination approach demonstrated a promising effect in high-risk patients, particularly those with nodal involvement or other high-risk features. The NCCN guidelines recommend adding CDK4/6 inhibitors, like abemaciclib, to endocrine therapy for high-risk patients with nodal involvement for a duration of 2 years [34].

Postmenopausal

For postmenopausal women with early-stage HR+ breast cancer, adjuvant ET typically involves third-generation AIs, such as letrozole, anastrozole, or exemestane, or selective estrogen receptor modulators like tamoxifen. In cases where AIs are poorly tolerated or contraindicated, a 5-year course of tamoxifen may be an alternative [35–38].

ATAC Trial

The ATAC trial [39] (n = 6241) compared anastrozole and tamoxifen in postmenopausal women with HR+ breast cancer. After 10 years, anastrozole demonstrated significantly improved DFS (HR = 0.91, 95% CI 0.83–0.99, p = 0.04), time to recurrence (TTR) (HR = 0.84, 95% CI 0.75–0.93, p = 0.001), and time to distant recurrence (HR = 0.87, 95% CI 0.77–0.99, p = 0.03) compared to tamoxifen. There was no significant difference in OS (HR = 0.95, 95% CI 0.84–1.06, p = 0.4). Anastrozole also had a higher incidence of fractures during treatment (451 vs. 351; OR = 1.33, p < 0.0001), though fracture rates were similar post-treatment (110 vs. 112; OR = 0.98, p = 0.9). The incidence of other cancers was similar between groups.

In summary, based on these findings, the NCCN guidelines recommend combining OFS with endocrine therapy for at least 5 years in premenopausal women with high-risk HR+ breast cancer, including those with young age, high-grade tumors, or lymph node involvement. After 5 years of OFS, continuation of therapy with tamoxifen is preferred over AIs. AIs are recommended as the preferred adjuvant therapy for postmenopausal women with HR+ breast cancer, either as monotherapy or in combination with CDK4/6 inhibitors for high-risk patients [34].

Biomarker-Guided Decision-Making: Prognostic vs. Predictive Tools

Extended endocrine therapy has been investigated through multiple randomized trials. The ATLAS and aTTom trials demonstrated that extending tamoxifen to 10 years reduced recurrence and mortality compared with stopping at 5 years, albeit with increased risk of endometrial cancer and thromboembolic events [26]. Similarly, AI extension trials such as NSABP B-42 and AERAS showed improvements in DFS and reduced distant recurrence, but without consistent overall survival benefit [40–43]. However, this approach comes with an increased risk of adverse events, underscoring the importance of patient stratification. Therefore, the central challenge is not whether extended therapy works, but rather which patients should receive it. Identifying patients who benefit most from prolonged therapy can help minimize toxicity.

Over the past two decades, genomic assays have emerged as valuable tools for identifying patients at lower risk of recurrence who may not benefit from adjuvant chemotherapy and could be treated with ET alone. Genomic tests, such as Oncotype DX, MammaPrint, BCI, EndoPredict, and Recurindex, have been integrated into clinical practice to guide treatment decisions by assessing both the risk of relapse and the expected benefit from ET [44–50]. Biomarker-guided selection has emerged as a rational strategy to maximize benefit-to-toxicity ratios (Table 2).

Table 2.

Comparison of commonly used tissue-based genomic assays and ctDNA-based MRD approaches

Assay/biomarker	Core gene signature/marker	Sample type	Technology platform	Clinical role	Key clinical output	Evidence level	Predictive value for extended ET	Guideline adoption	Best-supported population	Major strengths and limitations
Breast Cancer Index (BCI)	HOXB13/IL17BR ratio (H/I) + Molecular Grade Index (MGI)	FFPE tumor tissue	RT-qPCR	Prognostic + predictive	Late distant recurrence risk; benefit from extended endocrine therapy	Strong evidence (prospective-retrospective from RCTs: aTTom, NSABP B-42; prospective registry)	Validated	ASCO/NCCN recognize for extended ET decision support	Primarily postmenopausal; some data in premenopausal (SOFT translational)	Best validated assay for extended ET benefit prediction; relatively high cost; limited availability in some regions
Oncotype DX (21-gene Recurrence Score)	16 cancer genes + 5 reference genes	FFPE tumor tissue	RT-qPCR	Prognostic (early) + limited predictive	10-year recurrence risk; chemotherapy benefit prediction	Very strong evidence (TAILORx, RxPONDER); limited data for late recurrence/extended ET	Limited	Widely adopted (NCCN/ASCO/ESMO)	Node-negative and selected node-positive HR+/HER2− patients	Gold standard for chemo decision; not designed for late recurrence/extended ET prediction
MammaPrint (70-gene signature)	70-gene expression signature	FFPE or fresh tissue	Microarray/NGS-based	Prognostic	Early recurrence risk; identifies ultralow-risk biology	Strong evidence (MINDACT)	None	Recognized by NCCN/ESMO for risk stratification	Broad early-stage HR+/HER2− population	Useful for identifying ultralow-risk group; predictive value for extended ET still debated
EndoPredict (EP/EPclin)	12 genes (8 cancer-related + 4 reference) + clinical factors (EPclin)	FFPE tumor tissue	RT-qPCR		Late recurrence prediction (especially 5–15 years)	Strong evidence (ABCSG6/8; multiple validation cohorts)		Used in Europe; referenced in guidelines as risk tool	Mainly postmenopausal HR+/HER2− patients treated with ET	Good for late risk stratification; less direct predictive evidence for extended ET benefit
Prosigna (PAM50 Risk of Recurrence)	50-gene intrinsic subtype classifier	FFPE tumor tissue	NanoString nCounter		10-year recurrence risk; intrinsic subtype classification	Strong evidence (TransATAC, ABCSG8)		Included in NCCN/ASCO guidance for prognosis	Postmenopausal HR+/HER2− early breast cancer	Adds biologic subtype info; not validated as predictive for extended ET benefit
RecurIndex	Multi-gene panel (China-developed, includes ER-related and proliferation genes)	FFPE tumor tissue	RT-qPCR		Recurrence risk prediction in Chinese populations	Moderate evidence (retrospective + prospective cohorts, limited RCT-level validation)		Regional use (China), not globally adopted	Chinese HR+/HER2− early breast cancer patients	Potential regional advantage; needs broader external validation and predictive trials
Circulating tumor DNA (ctDNA)	Patient-specific somatic mutations from primary tumor	Plasma (blood)	NGS ultra-deep sequencing	Prognostic (emerging predictive)	MRD detection; recurrence prediction with lead time; dynamic monitoring	Growing evidence (observational cohorts; emerging interventional trials)	Limited	Not guideline standard for routine use; recommended in trials	High-risk early-stage patients; especially stage II–III	Most sensitive MRD strategy; costly and complex; requires tumor tissue sequencing. Lack of standardization and potential confounding from clonal hematopoiesis

Postmenopausal: Comparing 5 to 10 Years of Aromatase Inhibitors

The AERAS trial [51] assessed the benefit of extending anastrozole therapy for an additional 5 years in postmenopausal women who had completed initial treatment. The extended anastrozole group showed a higher DFS rate (91%) compared to the discontinuation group (86%) with a significant hazard ratio (0.61). While extended treatment reduced local recurrences and second primary cancers, it did not affect overall DFS. Adverse events were more frequent in the continuation group but were generally manageable.

Similarly, the NSABP B42 trial [42, 52] randomized postmenopausal women who had received 5 years of adjuvant ET with an AI to either continue or discontinue letrozole for another 5 years. Letrozole extension improved DFS (HR = 0.85, 95% CI 0.74–0.96, P = 0.01) and reduced the risk of distant recurrence (HR = 0.72, 95% CI 0.55–0.92, P = 0.01), but did not affect overall survival. Importantly, letrozole therapy benefitted patients in both the “upfront” and “switch” groups, suggesting flexibility in treatment strategy.

Biomarker-Guided Trials in Postmenopausal Women

The translational cohort of the NSABP B42 trial provided key insights into the role of biomarkers in predicting the benefit of extended letrozole therapy. In this cohort, patients were tested with the BCI, which helps predict the likelihood of recurrence and the benefit of extended therapy. Patients with high BCI scores (indicating a higher risk of recurrence) experienced a significant benefit from extended therapy, with a 10-year absolute benefit of 2.4% in recurrence-free interval (RFI) compared to those with low BCI scores. For patients with low BCI scores, the benefit of extended letrozole was less pronounced, emphasizing the role of BCI in identifying patients who may not need extend treatment. The predictive value of BCI was notably stronger in the HER2− subgroup, where extended therapy led to a significant reduction in recurrence (HR = 0.29, 95% CI 0.12–0.69, P < 0.01) [15].

MammaPrint in the NSABP B42 Trial

MammaPrint (MP), another genomic test used in the NSABP B42 trial, provided further insights into the role of biomarkers in guiding treatment decisions. Tumors were classified as either MP high-risk (MP-HR) or MP low-risk (MP-LR). For patients with MP-LR tumors, extended letrozole therapy significantly improved the 10-year distant recurrence-free interval (DRFI) by 3.7%, with a hazard ratio (HR) of 0.43 (95% CI 0.25–0.74; P = 0.002). In contrast, patients with MP-HR tumors did not show significant improvement from extended therapy (HR = 0.65, 95% CI 0.34–1.24; P = 0.19), suggesting that MP can help identify patients who may benefit from extend ET. Additionally, within the MP-LR group, patients with ultralow-risk tumors (MP-UL) derived less benefit from extended therapy compared to those with low non-ultralow-risk tumors (MP-LNUL), further emphasizing the need to refine risk stratification to optimize therapy [53].

EndoPredict in Postmenopausal Women

The EndoPredict assay, used in the ABCSG6 and ABCSG8 trials, also provided valuable prognostic information. In postmenopausal patients who had completed 5 years of tamoxifen or tamoxifen followed by an AI, EndoPredict identified those at high risk for late recurrence, thus guiding decisions about extended therapy. Patients classified as low-risk by EndoPredict had a significantly better clinical outcome in terms of late recurrence compared to high-risk patients (P = 0.002) [54].

Dynamic Monitoring Beyond Static Genomic Risk

Distant recurrences in HR+ breast cancer typically occur more than 5 years after the initial diagnosis, often arising from residual disease that is undetectable with current imaging and clinical assessments [24]. While the efficacy of adjuvant chemotherapy and ET in preventing recurrence is well-established, the detection of MRD remains a challenge in the late adjuvant setting [7]. Currently, recurrence risk is assessed based on clinical features and molecular testing at the time of surgery, but this risk is not routinely re-evaluated during the adjuvant phase.

The detection of ctDNA as a marker of MRD has gained attention for its potential to predict relapse and guide treatment decisions. Several studies have shown that ctDNA is associated with high recurrence risk and can be detected even years after definitive treatment. In a previous study [55] that included 48 patients with early-stage breast cancer (median age 50.5 years), comprising 34 HR+/HER2−, 5 HER2+, and 9 TNBC cases, it was found that the detection of ctDNA after surgery was associated with worse disease recurrence-free survival (hazard ratio = 5.5, 95% CI 1.1–28.5, P = 0.04). In a cohort of 103 high-risk stage II–III HR+ breast cancer patients, all diagnosed over 5 years ago and without clinical recurrence at the time of follow-up, 10% tested positive for MRD at some point. Of those who were MRD-positive, 7.2% developed distant metastatic recurrence, with all MRD-positive patients showing signs of MRD before clinical relapse, providing a median lead time of 12.4 months [18]. Together, these results further underscore the potential of ctDNA to provide early signals of recurrence and assist in identifying high-risk patients who might benefit from more intensive follow-up or early intervention.

Additionally, the OlympiA study demonstrated the efficacy of olaparib in improving 3-year invasive disease-free survival (iDFS) in patients with germline BRCA-mutated (gBRCA), HER2−, early-stage breast cancer (85.9% vs. 77.1%). Olaparib also reduced the risk of invasive disease recurrence or death by 42% (HR = 0.58, P < 0.001), meeting the primary endpoint. These findings highlight the importance of testing for germline BRCA1/2 mutations using blood specimens to identify patients who could benefit from PARP inhibitors, offering a more personalized approach to therapy. NCCN guidelines also recommended HR+/HER2− tumors, if (1) ≥ 4 positive lymph nodes after adjuvant chemotherapy (category 2A), or (2) residual disease after preoperative therapy and a clinical stage, pathologic stage, ER status, and tumor grade score ≥ 3. Adjuvant olaparib can be used concurrently with endocrine therapy [34].

In summary, ctDNA represents a promising tool for real-time monitoring of minimal residual disease and recurrence risk in HR+/HER2− breast cancer. Its incorporation into clinical practice could refine adjuvant treatment strategies, reduce overtreatment, and improve outcomes by personalizing therapy based on individual molecular profiles. Despite its promising role in detecting minimal residual disease and predicting recurrence, several important limitations currently restrict the clinical implementation of ctDNA in early-stage breast cancer [56–58]. First, the economic burden associated with serial longitudinal monitoring remains a significant concern, particularly when compared with one-time tissue-based genomic assays. Second, the sensitivity of ctDNA detection is influenced by tumor biology and disease burden, and low-shedding tumors may result in false-negative findings, thereby limiting its reliability in certain patient populations. Third, false-positive results may arise from non-tumor-derived alterations, particularly clonal hematopoiesis of indeterminate potential, which can confound interpretation if not properly accounted for. In addition, there is currently no consensus regarding optimal assay platforms, detection thresholds, or sampling intervals, resulting in variability across studies and limiting clinical standardization. Importantly, existing expert guidelines do not recommend routine use of ctDNA for clinical decision-making in early-stage breast cancer outside of clinical trials. Taken together, these limitations highlight that ctDNA, while biologically informative, remains an evolving biomarker that requires further technical refinement and prospective validation before widespread clinical adoption.

Discussion

Adjuvant ET remains the cornerstone of treatment for hormone receptor-positive (HR+) and human epidermal growth factor receptor 2-negative (HER2−) breast cancer, significantly reducing recurrence and mortality. However, over 50% of recurrences occur after the first 5 years, emphasizing the need for better strategies to guide treatment decisions and extend disease-free survival [11]. Although standard treatment regimens typically last 5 years, extending therapy may benefit certain patients, while others might be overtreated. This highlights the importance of refining our ability to predict recurrence risk and personalize treatment approaches. Biomarkers, including genomic assays and ctDNA, offer great promise in enhancing our ability to predict recurrent risk of breast cancer [58].

Genomic assays such as the BCI, Oncotype DX, and MammaPrint have significantly advanced our ability to assess recurrence risk and guide adjuvant therapy in HR+/HER2− breast cancer. As summarized in Table 2, genomic assays are not interchangeable in clinical decision-making. While multiple assays provide robust prognostic information for recurrence risk, only BCI has demonstrated predictive value for benefit from extended endocrine therapy in prospective–retrospective analyses of randomized trials. Accurately identifying women at high risk for late distant recurrence is crucial for optimizing treatment strategies, as some patients may be spared unnecessary extended endocrine therapy, while others could benefit from continued treatment. The risk and timing of distant recurrence vary among breast cancer patients. Clinicopathological factors, such as nodal status and tumor size, are well-established predictors of late recurrence, particularly in postmenopausal women with HR+ breast cancer. However, the incorporation of molecular signatures has enhanced the ability to predict late relapse more accurately. These tools can identify women, even those with node-positive disease, who are at sufficiently low risk and may not require extended endocrine therapy, allowing for more personalized treatment decisions. For example, BCI has shown strong potential in predicting late recurrence, with high BCI scores identifying patients who may benefit from extending endocrine therapy beyond the standard 5 years. In contrast, low BCI scores indicate patients who may safely reduce or omit therapy, thus minimizing unnecessary treatment and side effects [15, 44, 45].

However, the broader application of these tools faces several challenges. BCI, while useful for predicting late recurrence, requires specialized testing that may not be available in all clinical settings. Additionally, there is no universally accepted threshold for determining when to extend therapy, leading to potential variability in clinical practice [59, 60]. Oncotype DX and MammaPrint, while widely used for early recurrence risk and chemotherapy decisions, do not offer sufficient predictive power for late recurrences, which is critical for guiding extended endocrine therapy. In the NSABP B42 study, despite differences in sample sizes, BCI favored extended endocrine therapy for patients with high-risk scores, while MammaPrint identified the ultra-low and low-risk groups as more suitable candidates for extended therapy [11]. However, the findings of mammaprint translational cohort contrast with current global guidelines, which recommend extending therapy primarily for high-risk patients. This discrepancy highlights the differing predictive capabilities of multigene assays in assessing both near-term and long-term recurrence risk [53].

The use of these genomic assays is further limited by their cost and accessibility, particularly in lower-resource settings. As multigene testing evolves, future advancements are expected to provide more precise recommendations for individual treatment plans, offering evidence-based guidance for selecting the most appropriate assay based on each patient’s specific risk profile.

In recent years, ctDNA has emerged as a promising biomarker for monitoring MRD, and detecting early signs of recurrence in HR+/HER2− breast cancer. ctDNA has the advantage of being a non-invasive marker, allowing for real-time monitoring of disease status. Studies have demonstrated that ctDNA can detect residual disease up to 2 years before clinical signs of recurrence appear, offering a window of opportunity for intervention [18, 61]. While ctDNA has demonstrated consistent and strong prognostic value in detecting molecular residual disease and predicting relapse, its role as a predictive biomarker to guide therapeutic intervention remains unproven in early-stage breast cancer. Importantly, no prospective randomized trial has demonstrated that ctDNA-guided intervention improves overall survival, highlighting a critical evidence gap.

Despite its promising potential, ctDNA testing has several limitations. First, most breast cancer patients will undergo adjuvant therapy regardless of MRD detection. Therefore, the presence of MRD after surgery is most informative when the patient is already receiving or has completed standard adjuvant treatment. Clinical trials are needed to determine whether modifying the adjuvant treatment strategy based on MRD status can improve patient outcomes. Second, technical challenges, such as errors and background noise, must be considered. To minimize these issues, assays may undergo refinements such as barcoding, polishing, and free radical elimination to enhance accuracy [62]. Third, much of the research on ctDNA in early breast cancer has focused on its prognostic value in high-risk patients during the neoadjuvant phase, while studies exploring its role in guiding adjuvant endocrine therapy decisions are limited. Most studies in the adjuvant setting have primarily assessed the prognostic value of ctDNA, rather than its utility in informing treatment decisions. While ctDNA is effective in identifying high-risk patients, its role in guiding therapy de-escalation for low-risk patients is still under investigation. Additionally, the rate of MRD positivity in early-stage HR+/HER2− breast cancer is approximately 10% postoperative [18]. This relatively low rate may require large sample sizes for study designs, which has led many breast cancer studies to focus on populations with poorer prognoses, such as TNBC or those at high risk of requiring neoadjuvant therapy.

Currently, both BCI and MammaPrint have shown promise in predicting early recurrence of HR+/HER2− breast cancer, with BCI demonstrating stronger predictive efficacy for guiding extended adjuvant endocrine therapy. Most studies on extended endocrine therapy using multigene testing tools rely on baseline surgical specimens from HR+ early-stage breast cancer patients who have not experienced recurrence after 5 years of TAM treatment [15, 46, 53]. These specimens, typically analyzed with qPCR or microarrays, are used to assess recurrence risk after 5 years of TAM therapy. However, post-TAM tissue samples are generally unavailable. To address this, future studies on extended endocrine therapy may benefit from incorporating MRD testing or conducting comparative studies to evaluate the predictive value of MRD alongside multigene assays.

The future of biomarker-guided endocrine therapy is likely to involve an integrated model combining static tissue-based genomic profiling with dynamic ctDNA monitoring. Genomic assays can provide baseline estimation of late recurrence risk and endocrine sensitivity, whereas ctDNA offers real-time information regarding residual disease burden and molecular relapse dynamics.

An integrated approach may refine decision-making in three critical directions. First, patients with high-risk genomic profiles and persistent MRD positivity may represent a subgroup requiring escalation strategies beyond standard endocrine therapy. Second, patients with high genomic risk but persistently MRD-negative status may benefit from extended ET without additional escalation. Third, patients with low genomic risk and sustained MRD negativity may safely discontinue therapy after 5 years, reducing toxicity and improving quality of life (Fig. 1). An integrated approach combining genomic risk assessment with longitudinal ctDNA monitoring represents a promising conceptual strategy to refine decision-making; however, this framework remains hypothesis-generating and has not yet been validated in prospective clinical trials.

Fig. 1.

Biomarker-guided framework for tailoring adjuvant endocrine therapy in early-stage HR+/HER2− breast cancer. BCI Breast Cancer Index, ctDNA circulating tumor DNA, MRD minimal residual disease

Importantly, the clinical utility of biomarker-guided therapeutic strategies remains to be established in prospective interventional trials. Several ongoing studies, including DARE, LEADER, and TRAK-ER, are currently evaluating whether ctDNA-informed treatment decisions can improve patient outcomes in early-stage HR+/HER2− breast cancer [63–65]. Until such data become available, the integration of genomic assays and ctDNA should be considered exploratory and primarily hypothesis-generating rather than a basis for routine clinical decision-making. Beyond analytical validity and clinical utility, cost and accessibility represent important considerations for the integration of biomarker-driven strategies into routine practice. Tissue-based genomic assays, such as multigene expression panels, are generally performed as a one-time test, whereas ctDNA-based approaches rely on serial monitoring, which may substantially increase cumulative costs.

Conclusion

While significant progress has been made in using biomarkers to guide adjuvant endocrine therapy in HR+/HER2− breast cancer, several challenges remain. Genomic assays such as BCI, Oncotype DX, and MammaPrint offer valuable insights into recurrence risk, but their utility in guiding extended adjuvant endocrine therapy remains limited. The introduction of ctDNA testing marks a significant advancement in monitoring residual disease and predicting recurrence, but this strategy should be considered investigational and requires prospective validation before clinical adoption. Moreover, its broader application is still constrained by factors such as cost, accessibility, and the lack of standardized thresholds or monitoring schedules. Moving toward a more personalized approach to therapy will require further research into the combined use of genomic assays and ctDNA. This approach has the potential to refine treatment strategies, reduce overtreatment, and ultimately improve outcomes for patients with HR+/HER2− breast cancer.

Author Contributions

All authors made substantial contributions to this work and approved the final manuscript. Yang Liu and Mengxiao Wang contributed to the conception and design, data acquisition, data extraction, quality assessment, and drafting of the manuscript. Dongsheng Chen and Yuan Tan contributed to the data acquisition. Xing Zhang and Xiaohong Wang contributed to the conception, critical revision of the manuscript for important intellectual content, and final approval.

Funding

The Rapid Service Fee and Open Access fee will be covered by Prof. Xiaohong Wang.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Declarations

Conflict of Interest

All authors (Yang Liu, Mengxiao Wang, Dongsheng Chen, Yuan Tan, Xing Zhang, Xiaohong Wang) declare that they have no competing interests.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.