For more than 100 years, axillary clearance was the standard in breast cancer care. Only in the last three decades has axillary management evolved from radical surgical clearance to minimally invasive sentinel staging. It is now advancing toward selective omission driven by biological risk stratification and therapeutic relevance. Few areas in surgical oncology illustrate such a sustained recalibration of necessity over such a short time.

The original rationale for axillary dissection was rooted in the Virchow-Halsted concept of centrifugal tumor spread. Breast cancer was viewed as a locoregional disease that disseminated stepwise through lymphatic channels. On this basis, Halsted introduced radical mastectomy with en bloc axillary lymph node removal in 1894, later refining the procedure in the early twentieth century [1, 2]. Axillary lymph node dissection (ALND) was therefore considered essential for cure and remained a cornerstone of treatment for decades.

Understanding of metastasis evolved in parallel with this surgical doctrine. As early as the eleventh century, Avicenna described the invasive and spreading nature of breast cancer [3]. In 1829, Récamier introduced the term “metastasis,” formally recognizing systemic dissemination [4]. Paget’s “seed and soil” hypothesis in 1889 proposed that metastatic spread depends on the interaction between tumor cells and a permissive microenvironment [5, 6]. Halsted interpreted this biology through a linear model of progression, assuming orderly spread from breast to lymph nodes and subsequently to distant organs [1]. However, alternative concepts emerged. Ewing proposed dissemination according to vascular and lymphatic flow patterns [7]. Later, experimental and clinical evidence demonstrated that only a very small fraction of circulating tumor cells formed metastases, a phenomenon termed metastatic inefficiency [5, 8]. Observations of tumor dormancy further challenged the linear progression; delayed metastases without local recurrence were described by Willis and Hadfield, suggesting a temporary mitotic arrest [5, 9]. In the 1960s, Fisher advanced the systemic disease hypothesis, proposing that breast cancer is often disseminated at diagnosis [10]. Hellman later refined this with the spectrum theory, describing metastatic behavior as a biological continuum [11].

Together, these insights progressively weakened the assumption that regional lymph nodes are obligatory gateways to systemic spread. Nodal metastases were increasingly interpreted as markers of tumor biology rather than the source of dissemination. Randomized evidence reinforced this shift. The National Surgical Adjuvant Breast and Bowel Project (NSABP) B-04 trial demonstrated no overall survival advantage for immediate ALND compared with delayed intervention [12, 13]. Some node-negative patients developed distant metastases, while some node-positive patients did not. These findings altered the perceived role of the axilla: its prognostic value remained, but its therapeutic indispensability was questioned.

The introduction of sentinel lymph node biopsy (SLNB) marked the next decisive transformation. Giuliano and colleagues demonstrated that targeted removal of the first draining lymph node could accurately stage the axilla [14]. Randomized trials, including NSABP B-32 and Axillary Lymphatic Mapping Against Nodal Axillary Clearance (ALMANAC), confirmed equivalent survival and regional control compared with ALND, with substantially lower morbidity [15–17]. SLNB replaced routine dissection in clinically node-negative disease.

The axilla thus transitioned from a therapeutic battlefield to a staging field. Further de-escalation followed in patients with limited nodal involvement. The International Breast Cancer Study Group 23-01 (IBCSG 23-01) trial demonstrated that completion of ALND could be safely omitted in patients with sentinel node micro-metastases [18, 19]. ACOSOG Z0011 extended this principle to selected patients with limited macro-metastatic disease undergoing breast-conserving surgery with whole-breast radiotherapy (WBRT) [20]. In parallel, the AMAROS trial showed that axillary radiotherapy provided regional control equivalent to ALND with lower morbidity [21].

These findings were reinforced by additional studies. The OTOASOR trial confirmed that axillary radiotherapy yields comparable long-term regional control to ALND in patients with limited sentinel node involvement [22]. The SENOMAC trial further broadened the evidence base, including patients with T1–T3 tumors and both breast-conserving surgery and mastectomy. Omission of completion ALND in patients with one or two positive sentinel nodes did not compromise recurrence-free survival [23].

Together, these studies showed that low-volume nodal disease can be safely managed without extensive axillary surgery when effective systemic therapy and radiotherapy are applied. As nodal involvement became increasingly a marker of tumor biology rather than a therapeutic target, the role of routine surgical staging itself came into question.

Initial support for omission of axillary surgery came from observational studies and the Society of Surgical Oncology Choosing Wisely recommendations in women over 70 years with low-risk disease [24–26]. This naturally led to the question of whether SLNB could also be omitted in other clinically node-negative patients with favorable tumor biology.

This question has now been addressed in three contemporary randomized trials: SOUND, INSEMA, and Dutch Breast Cancer Research Group (BOrstkanker Onderzoek Groep; BOOG) 2013-08 [27–30]. The SOUND trial demonstrated that omission of SLNB in patients with tumors ≤ 2 cm and a negative axillary ultrasound did not compromise distant disease-free survival (DFS) [27]. INSEMA expanded eligibility to tumors up to 5 cm and confirmed non-inferior invasive DFS, with only a small absolute increase in axillary recurrence [28]. The BOOG 2013-08 study provided additional supportive evidence in a representative breast-conserving population. Results have been presented at the San Antonio Breast Cancer Symposium 2026; however, full peer-reviewed publication is not yet available [29, 30]. These findings must be interpreted within the context of modern systemic therapy. In hormone receptor (HR)-positive, human epidermal growth factor receptor 2 (HER2)-negative disease, genomic assays frequently guide chemotherapy decisions independently of limited nodal involvement [31, 32]. Similarly, in HER2-positive and triple-negative disease, systemic therapy is primarily determined by tumor biology. Consequently, nodal status no longer changes adjuvant treatment in a substantial proportion of patients with early breast cancer. The available evidence provides the basis for omission of SLNB, but careful patient selection according to current guidelines remains essential for its safe implementation [33].

Axillary surgery has therefore evolved through distinct phases: radical clearance, sentinel staging, selective omission of completion dissection, and now consideration of complete omission in carefully defined patients. The SLNB procedure itself has also evolved. Indo-cyanine green (ICG) fluorescence provides high sentinel lymph node detection rates and represents an effective alternative to conventional mapping techniques. However, improvements in tracer technology do not change the principles of patient selection for omission of SLNB [34]. This progression reflects a broader oncologic principle: treatment intensity should correspond to therapeutic consequence.

With advances in precision oncology, the axilla is no longer an obligatory staging structure for every patient but a biological checkpoint whose assessment should be guided by its impact on treatment decisions. The challenge is to identify the patients in whom SLNB can be safely omitted. This position paper proposes a practical four-step framework to support that decision in daily clinical practice.

The historical evolution of axillary management leads to a straightforward clinical question: “In which patients can SLNB be safely omitted?” This position paper focuses on the patient population in whom omission is currently supported by randomized evidence. The proposed framework applies to patients with invasive breast cancer undergoing upfront breast-conserving surgery followed by planned WBRT. This treatment context is essential, as incidental irradiation of the lower axilla likely contributes to the low regional recurrence rates observed in the omission trials [20]. Patients should have a clinically node-negative axilla confirmed by negative preoperative axillary ultrasound; suspicious lymph nodes require image-guided biopsy before omission of SLNB is considered. The guiding principle is simple: SLNB should be omitted only when nodal information is unlikely to change systemic therapy or radiotherapy planning.

A literature search was performed using PubMed/MEDLINE to identify relevant publications up to May 2026. Current international guidelines, including those from ESMO, NCCN, ASCO, and EUSOMA, were also reviewed where relevant. Additional references were identified from the reference lists of relevant publications. Randomized controlled trials, prospective studies, meta-analyses, and international guidelines on axillary staging and omission of SLNB formed the basis of this narrative review. Historical landmark studies were included to provide the necessary background. Conference abstracts were generally excluded, with the exception of the BOOG 2013-08 trial, because no full peer-reviewed publication is currently available.

The randomized evidence is mature. The guidelines are converging. The remaining question is no longer whether omission of SLNB is oncologically safe in selected patients, but how to implement this safely and consistently in clinical practice [27–30].

Clinical implementation must be grounded in one guiding principle: therapeutic proportionality. Axillary surgery should be performed only when the information obtained is expected to alter management in a way that improves outcome. When nodal information does not change systemic therapy, does not meaningfully alter radiotherapy planning, and does not improve survival, surgical staging becomes difficult to justify as routine practice.

From tomorrow onward, axillary decision-making can follow a structured sequence (Figure 1).

First, confirm the biological phenotype. Is the patient postmenopausal? Is the tumor HR-positive/HER2-negative and of low to intermediate grade? Are clinicopathologic or genomic features such that systemic therapy is primarily biology-driven rather than node-driven? If the systemic treatment plan would not change with limited nodal involvement, the therapeutic dependence on SLNB is already weakened.

Second, confirm axillary selection. Is the axilla clinically node-negative and sonographically negative? Suspicious findings should be excluded by biopsy when appropriate. In an imaging-defined node-negative setting, the residual probability of clinically meaningful nodal burden is low, narrowing the incremental staging yield.

Third, define radiotherapy intent. Is WBRT planned? If so, incidental low-axillary coverage contributes to regional control. In contrast, omission cannot be generalized to mastectomy without radiotherapy.

Fourth, assess therapeutic consequence explicitly. Would nodal positivity trigger chemotherapy escalation? Alter regional nodal irradiation? Unlock CDK4/6 inhibitor eligibility? If the answer is “unlikely,” omission is proportionate.

This approach transforms SLNB from a reflex into a decision. It shifts the axilla from an obligatory staging field to a conditional decision node. Implementation does not require structural change, but explicit documentation. A concise multidisciplinary statement documenting clinical node-negative status, negative imaging, biology-driven systemic planning, radiotherapy intent, and informed patient consent is sufficient to operationalize omission safely. Patients should be engaged transparently. The discussion is not about reducing care, but about avoiding low-yield intervention. The trade-off should be communicated in absolute terms: a very small increase in isolated regional recurrence risk, no demonstrated survival disadvantage, and a measurable reduction in arm morbidity.

Clinical implementation is therefore neither abandonment nor enthusiasm-driven de-escalation. It is calibrated restraint. In biologically favorable, imaging-negative, postmenopausal patients undergoing breast-conserving surgery with planned WBRT, omission of SLNB is not experimental. It is evidence-aligned. Outside this phenotype, staging remains appropriate. The paradigm shift is subtle but profound: the axilla is no longer interrogated because it can be staged, but only when staging is expected to change management. That is therapeutic proportionality.

The routine use of SLNB in all clinically node-negative invasive breast cancers is no longer proportionate to its clinical yield.

Randomized evidence from SOUND, INSEMA, and BOOG 2013-08 demonstrates that omission of SLNB in carefully selected patients undergoing breast-conserving surgery with planned WBRT does not compromise distant disease control and results in only a minimal absolute increase in axillary recurrence, without survival disadvantage. When interpreted in absolute terms, the clinical benefit of SLNB is limited: many patients must undergo staging to prevent a single, typically salvageable regional event, while a smaller number experience avoidable morbidity. At the same time, nodal information alters adjuvant treatment in only a minority, largely depending on systemic therapy frameworks such as CDK4/6 inhibitor use. De-escalation therefore represents not omission of care, but refinement of intent. The axilla should no longer be staged by default, but only when nodal information is expected to change management.