1 Department of Surgical Oncology, The First People’s Hospital of Linhai, 317000 Taizhou, Zhejiang, China
2 Department of Epidemiology, Harvard T.H. Chan School of Public Health, Harvard University, Boston, MA 02138, USA
3 Department of General Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, 310009 Hangzhou, Zhejiang, China
4 Department of Breast Surgery and Oncology, The Second Affiliated Hospital, Zhejiang University School of Medicine, 310009 Hangzhou, Zhejiang, China
5 Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), The Second Affiliated Hospital, Zhejiang University School of Medicine, 310009 Hangzhou, Zhejiang, China
Neoadjuvant therapy enables tumor downstaging to facilitate breast conserving
surgery (BCS), allows assessment of treatment sensitivity, provides prognostic
information, and serves as an experimental platform in breast cancer research
[1, 2]. Although chemotherapy remains the predominant neoadjuvant modality,
responses vary widely across breast cancer subtypes [1]. Notably, hormone
receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2–)
disease, the most common subtype (
Initially reserved for elderly or chemotherapy-ineligible patients, neoadjuvant endocrine therapy (NET) has proven to be a safe and effective strategy in HR+/HER2– breast cancer, achieving objective response rates (ORRs) and BCS rates comparable to NCT [5]. Nevertheless, its clinical application remains limited [6]. Key challenges remain regarding optimal patient selection, treatment regimens and duration, reliable response endpoints, and strategies for subsequent adjuvant therapy.
Current guideline recommendations support consideration of NET for HR+/HER2– patients in whom the indications for chemotherapy are uncertain [7], and for patients with strong HR expression who have comorbidities or low-risk luminal tumors as determined by clinicopathologic/genomic features [8]. Since adjuvant chemotherapy does not benefit all patients with HR+/HER2– breast cancer and can be omitted in low-risk cases [9], the routine use of NCT for chemosensitivity testing or downstaging may be questioned. Consequently, tumors deemed low risk by guideline-endorsed genomic assays (e.g., the 21-gene Oncotype DX or 70-gene MammaPrint tests) [7, 8] are therefore logical candidates for less toxic neoadjuvant approaches such as NET [10]. Indeed, direct evidence links the 21-gene Recurrence Score with clinical response to preoperative letrozole [11], and data from the DANCER trial indicate that MammaPrint low-risk patients treated with cyclin-dependent kinase 4/6 (CDK4/6) inhibitor plus NET had lower preoperative endocrine prognostic index (PEPI) and residual cancer burden (RCB) scores [12]. Similarly, the PLATO trial demonstrated that triaging patients by MammaPrint (high-risk to NCT, low-risk to NET) substantially increased BCS eligibility while sparing low-risk patients unnecessary chemotherapy toxicity [13].
Three pivotal trials have demonstrated that aromatase inhibitor (AI)-based NET yields superior clinical benefit compared with tamoxifen in postmenopausal women, establishing AIs as the preferred agents in this population [14, 15, 16]. For premenopausal patients, current guidelines recommend either an AI combined with ovarian function suppression (OFS) or tamoxifen with or without OFS [8]. Although maximal response to NET may take 6–12 months to evolve [17, 18], the commonly employed neoadjuvant interval in clinical practice is 4–6 months [8, 10, 19]. Treatment may be extended in patients who continue to derive clinical benefit [10].
A major recent focus has been the incorporation of CDK4/6 inhibitors into NET. Multiple neoadjuvant trials have evaluated CDK4/6 inhibitor plus endocrine therapy—either in comparison with chemotherapy (e.g., NeoPal [20, 21], CORALLEEN [22, 23], and CARABELA [24, 25] trial ) (Table 1, Ref. [20, 21, 22, 23, 24, 25]) or versus endocrine monotherapy (e.g., PALLET [26], FELINE [27], and neoMONARCH trial [28]). The addition of CDK4/6 inhibitors enhances tumor proliferation suppression, yet fails to improve pCR rates, ORRs, disease control rates, or PEPI scores compared with NCT or NET alone [20, 22, 25, 26, 28, 29]. Further work is needed to identify subgroups most likely to benefit. The DANCER trial introduced a two-step, biomarker-driven framework utilizing baseline parameters and dynamic circulating tumor DNA monitoring to better stratify patients likely to benefit from treatment [12]. Targeting the PI3K/AKT/mTOR (PAM) pathway, an axis implicated in proliferation and endocrine resistance, has also been explored in the neoadjuvant setting, but whether PAM inhibition adds additional benefit to NET remains uncertain [30, 31].
| Trial | Disign | Population | Intervention | ITT N | Primary endpoint | Main findings | Ref |
| NeoPAL | Randomized (1:1), open-label, phase II | Stage II–IIIA, ER+/HER2–, PAM50 luminal B/A with node+, not eligible for BCS, postmenopausal | Palbociclib + letrozole vs. FEC→T | 106 (53/53) | RCB 0–I rate | RCB 0–I: 4 (7.7%) vs. 8 (15.7%); pCR: 3.8% vs. 5.9%; PEPI-0: 17.6% vs. 8.0%; CRR/BCS similar; 40-month PFS: 86.7% vs. 89.9% | Efficacy [20], survival [21] |
| CORALLEEN | Randomized (1:1), open-label, phase II | Stage I–IIIA, HR+/HER2–, PAM50 luminal B, tumor |
Ribociclib + letrozole vs. AC→wP | 106 (52/54) | PAM50 low-ROR rate | Low-ROR: 46.9% vs. 46.1%; RCB 0–I: 6.1% vs. 11.8%; PEPI-0: 22.4% vs. 17.3%; significantly better HRQoL | Efficacy [22], HRQoL [23] |
| CARABELA | Randomized (1:1), open-label, phase II | Stage II–III, HR+/HER2–, Ki67 |
Abemaciclib + letrozole ( |
200 (100/100) | RCB 0–I rate | RCB 0–I: 13.0% vs. 18.0%; CRR: 78% vs. 71%; PEPI-0: 14% vs. 26% (p = 0.03); Ki67 rebound (2w→surgery): 17% vs. 2% (p |
[24, 25] |
Abbreviations: ITT, intention-to-treat; ER, estrogen receptor; BCS, breast-conserving surgery;
FEC
pCR to neoadjuvant therapy provides important prognostic information, although its association is weakest in the HR+ subtype [32, 33, 34]. RCB after NCT has established prognostic value across breast cancer subtypes, including HR+/HER2– tumors [35], though its significance following NET remains unclear. In fact, pCR rates after preoperative treatment are extremely low in the HR+/HER2– population, even reaching 0% in some reports, irrespective of NCT or NET [13, 36, 37]. The largest NET trial to date, ALTERNATE, reported a pCR rate of only 0.3% (3/933) after 6 months of NET [38].
Early changes in Ki67 levels following short-term (2–4 weeks) endocrine therapy
exhibit predictive value for long-term outcomes and can assist in guiding
treatment decisions [39, 40, 41, 42, 43, 44]. Consequently, Ki67 and derived indices such as
complete cell cycle arrest (CCCA; Ki67
There is no universally accepted strategy for adjuvant systemic
therapy following NET, and whether to add adjuvant chemotherapy remains a
clinically difficult decision. Generally, patients with clinical or genomic
high-risk features and those with unexpectedly extensive residual disease after
NET (e.g.,
YZ and PJ conceived and designed the study. XL contributed to the interpretation of the results. YZ critically revised the manuscript, supervised the research, and guided the discussion. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.
Not applicable.
Not applicable.
This work was funded by the Key Research and Development Program of Zhejiang Province (Grant No. 2024C03183), the National Natural Science Foundation of China (Grant No. 82373437), and Jingyi Research Fund Phase II of Beijing Vlove Charity Foundation (Grant No. JVII2025-0200304035).
The authors declare no conflict of interest. Yunxiang Zhou is serving as one of the Editorial Board members and one of the Guest Editors of this journal. We declare that Yunxiang Zhou had no involvement in the review of this article and has no access to information regarding its review. Full responsibility for the editorial process for this article was delegated to Michael H. Dahan.
During the preparation of this work, the authors used ChatGPT-5.1 in order to check spelling and grammar. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
References
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.