Breast-Conserving Surgery in Multicentric Breast Cancer: Evolving Evidence and Patient Selection

1. Introduction

Multicentric breast cancer has for decades been managed as a near-automatic indication for mastectomy. This position was never based on local recurrence risk in isolation; it reflected three intersecting concerns that were each real limitations of practice at the time — an inability to reliably exclude occult disease outside the resected specimen using the imaging available, a correspondingly higher risk of leaving a positive margin at one or more foci, and an inability to remove multiple tumour sites from one breast without producing a cosmetically unacceptable result [1,2]. Each of these three limitations has been substantially, though not completely, addressed by developments of the past two decades: cross-sectional imaging with MRI, oncoplastic reconstructive technique, and biologically tailored systemic therapy. This review examines the evidence behind the historical position, the evidence that has begun to qualify it, and the areas where the two remain in genuine tension.- The current question is no longer whether multicentric disease can ever be treated with breast conservation, but rather which patients can safely undergo conservation without compromising oncological outcomes.

2. Methods

This narrative review was conducted in accordance with the SANRA (Scale for the Assessment of Narrative Review Articles) recommendations, a checklist-based framework for appraising and reporting non-systematic reviews across dimensions such as justification of the review's importance, statement of aims, description of the literature search, and referencing quality. PubMed/MEDLINE, Embase, and Google Scholar were searched for English-language publications from January 2000 to February 2026 using combinations of the terms "multicentric breast cancer," "multifocal breast cancer," "multiple ipsilateral breast cancer," "breast-conserving surgery," "oncoplastic surgery," "extreme oncoplasty," "breast MRI," "neoadjuvant systemic therapy," and "BRCA breast conservation." The final search was conducted on 28 February 2026. Priority was given to prospective trials, systematic reviews and meta-analyses, and current professional guideline documents; reference lists of retrieved articles were hand-searched for additional relevant studies. Because this is a narrative rather than a systematic review, study selection was based on relevance to the surgical management of multicentric/multifocal disease and on methodological quality as judged by the author, rather than on a predefined systematic search protocol or formal risk-of-bias assessment. Where evidence was sparse or conflicting — most notably regarding the necessity of routine preoperative MRI — this is stated explicitly rather than resolved by selective citation.

Definitions, Terminology, and the Problem of Classification

Throughout this narrative review conducted in accordance with SANRA recommendations [3], multiple ipsilateral breast cancer (MIBC) is used as the umbrella term for two or more tumour foci within the same breast, encompassing both multifocal (MF) and multicentric (MC) presentations; the two subtypes are distinguished below, but no consensus definition separates them precisely, and the labels should not be read as implying a specific anatomical or prognostic meaning. Multifocal (MF) and multicentric (MC) breast cancer are frequently discussed together as "multiple ipsilateral breast cancer" (MIBC), a term increasingly preferred in the trial literature because the distinction between the two entities is not as settled as textbook definitions imply. In most contemporary usage, multifocal disease refers to two or more tumour foci within the same breast quadrant, and multicentric disease to foci in different quadrants; some groups instead define the two by an inter-tumour distance threshold (commonly cited thresholds cluster around 4–5 cm, though this figure is not standardised and different studies have used values from 1 cm to 5 cm). Neither the quadrant-based nor the distance-based definition has achieved universal consensus, and both are acknowledged in the literature to be anatomically somewhat arbitrary, since breast tissue has no true internal quadrant boundaries and lymphatic drainage does not respect them. In other words, breast quadrants are a surgical and radiological convenience for describing tumour location, not a biological or anatomical unit — a distinction that matters clinically, since resection planning, margin assessment, and mastectomy-conversion decisions are often framed in quadrant-based language despite that lack of biological grounding. This is not a semantic issue only: it is the principal reason reported incidence of MF/MC disease spans such a wide range in the literature — from roughly 10% to over 60% of breast cancers [4] — depending on whether ascertainment is by clinical examination, mammography, MRI, or whole-organ pathological sectioning, and on whether in situ lesions are counted as qualifying foci. Unless stated otherwise, the studies discussed in this review required at least one biopsy-proven invasive focus for eligibility; where in situ-only lesions were counted toward MIBC status, this is noted explicitly, since doing so measurably inflates reported incidence and is not applied consistently across the literature.

Multifocal disease has traditionally been assumed to represent intraductal or intramammary spread from a single tumour, and multicentric disease to represent independent, biologically distinct primaries. Molecular data over the past decade complicate this binary. Targeted and whole-exome sequencing studies of paired foci from the same breast have found that a meaningful proportion of multifocal cases show strong clonal relatedness across foci, consistent with intramammary spread, while others show discordant driver mutations and copy-number profiles consistent with independent origin [5] — and this heterogeneity is not fully predicted by the anatomical (same-quadrant vs. different-quadrant) classification. A targeted-sequencing series of 21 patients with multiple ipsilateral lesions found discordant intrinsic subtype in 10% of cases and copy-number differences in a third, despite closely matched histology between foci [6] — indicating that biomarker and genomic heterogeneity between foci can occur even when the lesions look identical under the microscope. This growing biological literature is one reason quadrant-based nomenclature is increasingly regarded as a surgical planning convenience rather than a biological classification, and why some authors now argue for describing this population primarily as MIBC rather than forcing every case into an MF/MC dichotomy.

Whole-organ pathological sectioning studies, dating from the 1980s and 1990s, found occult multicentric foci in a large proportion of mastectomy specimens even when preoperative assessment had suggested unifocal disease [7,8] — one frequently cited whole-organ series found multicentric foci beyond the index quadrant in nearly 80% of breasts that harboured any additional foci at all [7]. The clinically important counterpoint, drawn from long-term breast-conservation follow-up series, is that the great majority of true local recurrences occur at or near the index tumour bed rather than in these unsampled additional foci, suggesting that pathological multicentricity [9] and clinically meaningful local recurrence risk are not the same thing. This distinction is central to understanding why breast conservation can be oncologically reasonable in selected multicentric disease despite the historical concern about undetected additional foci.

3. Result

The Historical Case for Mastectomy

Cohort studies auditing reasons for exclusion from breast conservation have consistently identified multifocal or multicentric disease as the single most common disqualifying factor, ahead of diffusely abnormal mammographic findings and unfavourable tumour-to-breast-volume ratio. Early pooled analyses supported treating multicentricity with caution: a 2014 systematic review and meta-analysis found higher local recurrence after breast-conserving surgery in multifocal/multicentric disease compared with unifocal cancer [10], though this signal was driven predominantly by cohorts treated before 2000, before the routine use of adjuvant systemic therapy, modern radiotherapy planning, and oncoplastic reconstruction were established practice. This body of evidence is the origin of language still found in current guidance — the American Society of Breast Surgeons' resource guide, for example, continues to list multicentric disease not amenable to oncoplastic surgery among the absolute contraindications to breast conservation [11] — even as the same documents increasingly qualify that position, as discussed later in this article.

A less commonly articulated but clinically important corollary of this history is that mastectomy is not itself risk-free. Beyond the immediate surgical burden, mastectomy carries its own consequences relative to breast conservation — loss of native breast sensation, the additional morbidity and potential complications of reconstruction, delays to adjuvant therapy where reconstruction is complex, and measurable psychological effects on body image and quality of life. The clinical argument here is not that mastectomy should be avoided in multicentric disease; it is that mastectomy should not be offered on the basis of an indication that reflects outdated assumptions about achievable margins, imaging accuracy, or cosmetic feasibility rather than a contemporary, individualised assessment. Where oncoplastic technique, current imaging, and modern systemic therapy have closed the gap that originally justified reflexive mastectomy, continued routine use of mastectomy in all such cases may represent persistence of historical practice patterns rather than evidence-based individualisation.

The Evidence Base for Breast Conservation in Multicentric Disease

No single study settles the question of oncologic safety, and the evidence is best read as several independent lines that converge on a broadly consistent, though not unanimous, conclusion.

Prospective trial evidence. ACOSOG Z11102 (Alliance) remains the only prospective trial to address this question directly. It enrolled 270 women aged 40 years and older with two to three biopsy-proven, cN0–N1 ipsilateral foci, each separated by at least 2 cm of normal tissue and confined to two quadrants, treated with lumpectomy to negative margins followed by whole-breast radiotherapy with a boost to every lumpectomy bed; neoadjuvant therapy was not permitted, and preoperative MRI, initially mandatory, was later made optional. Among 204 evaluable patients, the 5-year cumulative incidence of local recurrence was 3.1% (95% CI 1.3–6.4%), below the prespecified acceptability threshold of 8%. Adherence to endocrine therapy in hormone-receptor-positive patients was associated with lower local recurrence; age, number of preoperative biopsy sites, and HER2 status were not independent predictors [2]. As a single-arm design without a mastectomy comparator, the trial cannot quantify relative risk against mastectomy directly, and its statistical framing drew published commentary and a formal reply from the investigators [12].

Imaging-focused prospective data. The MIPA (Multicentre International Prospective Analysis) study, a large prospective cohort examining how screening and diagnostic breast MRI affect surgical treatment, provides complementary evidence [13] on how imaging strategy — rather than tumour biology — shapes the surgical pathway in this population; its findings are discussed together with other MRI trials later in this review, since they speak primarily to the imaging question rather than to oncologic safety of conservation itself.

Oncoplastic comparative data. A retrospective matched-cohort study compared 100 patients with multicentric or multifocal tumours treated with oncoplastic breast-conserving surgery against 100 matched patients treated with mastectomy. Overall and disease-free survival were similar between groups; local events were somewhat more frequent after oncoplastic surgery and regional events somewhat more frequent after mastectomy, but neither difference reached statistical significance, and distant recurrence was similar [14]. The authors characterised this as the best available evidence to date supporting oncoplastic breast conservation in this population, while explicitly calling for randomised confirmation. [15]

Retrospective single-centre cohorts. Two recently reported series add real-world texture using different selection strategies. A retrospective Argentinian cohort found breast conservation feasible and oncologically comparable in stage I–II multifocal/multicentric disease (n = 91, 7.7% of the overall cohort) versus unifocal disease (n = 1,097): 5-year locoregional recurrence was 3.3% versus 5.9%, with similar distant recurrence (3.3% vs. 4.8%). This cohort achieved these outcomes without routine preoperative MRI, and excluded patients who received neoadjuvant chemotherapy. A Turkish single-centre series identified 246 of 2,155 patients with multifocal/multicentric disease over a recent three-year period, of whom 68 (5.4%) were selected for breast conservation; 90% achieved negative margins at first surgery, two ultimately required completion mastectomy, complication rates were 19% (including seroma), and 90% of 46 surveyed patients rated their cosmetic outcome as excellent or good, with no local recurrences at a median follow-up of only 12 months. Unlike the Argentinian cohort, this series used routine preoperative MRI and structured multidisciplinary evaluation.

Systematic reviews. A 2022 systematic review and meta-analysis concluded that multicentric/multifocal breast cancer does not universally carry worse survival than unifocal disease [16], although some included analyses suggested a small increase in mortality odds that the authors attributed at least partly to confounding by stage and molecular subtype. Methodological heterogeneity across the underlying studies — inconsistent definitions of multicentricity, variable radiotherapy boost practice, and differing margin standards — was identified as a limiting factor across this literature, a caveat echoed in a separate 2025 narrative synthesis of more than fifty studies on this topic [17].

Comparative meta-analysis. The most direct synthesis to date is a 2026 systematic review and meta-analysis that pooled study-level risk ratios for BCS versus mastectomy specifically in MIBC, rather than comparing multifocal/multicentric against unifocal disease [15]. It included 17 comparative studies — 16 retrospective and one nominally randomised series that was analysed as non-randomised because of incomplete reporting of allocation and intention-to-treat methods — comprising 29,711 patients (9,587 BCS, 20,124 mastectomy) drawn mainly from European and North American cohorts published between 1999 and 2025, with a median follow-up of 62 months and a median Newcastle–Ottawa quality score of 8 out of 9. Pooled analysis found no significant difference between BCS and mastectomy in local recurrence (RR 1.06, 95% CI 0.74–1.51), locoregional recurrence (RR 1.02, 95% CI 0.68–1.52), distant recurrence (RR 0.78, 95% CI 0.54–1.12), or disease-free survival (RR 0.82, 95% CI 0.60–1.13). Overall survival significantly favoured BCS in the primary analysis (RR 0.60, 95% CI 0.43–0.82), but this advantage was not maintained in a sensitivity analysis restricted to upfront-surgery cohorts (RR 0.77, 95% CI 0.31–1.91), suggesting the survival signal is more plausibly explained by selection and treatment heterogeneity — particularly inconsistent inclusion of neoadjuvant systemic therapy across studies — than by a genuine oncological advantage of breast conservation. The authors themselves highlight important limitations: the evidence base remains almost entirely retrospective and heterogeneous in its definitions of multifocality and multicentricity, margin standards, and radiotherapy technique; the single largest contributing cohort supplied overall-survival data only and was excluded from the recurrence and disease-free-survival analyses — by our reading of the study-level numbers in the source publication, this one cohort accounted for roughly 90% of the patients contributing to the overall-survival analysis specifically [18], so the pooled overall-survival estimate is driven overwhelmingly by a single registry dataset rather than corroborated independently across the 17 included studies; and subgroup analysis by MF versus MC status, or by higher-risk biology (HER2-positive and triple-negative disease, tumours larger than 5 cm, or three or more foci), was not possible. These limitations temper, without negating, the reassuring headline findings, and a leave-one-out or influence analysis excluding this dominant cohort would help clarify how much of the overall-survival signal survives its removal. Readers should not interpret the overall-survival finding as evidence that BCS is oncologically superior to mastectomy in MIBC; the more defensible reading is that BCS is not inferior to mastectomy for oncological safety in appropriately selected patients, not that it is the preferable option.

A further source of heterogeneity between the neoadjuvant and upfront-surgery analyses deserves consideration: patients receiving neoadjuvant systemic therapy almost universally undergo preoperative breast MRI for staging, response assessment, and surgical planning, whereas MRI utilisation among patients treated with upfront surgery is considerably more variable. In the Alliance ACOSOG Z11102 trial, an exploratory subgroup analysis found a substantially lower 5-year local recurrence rate among patients who underwent preoperative MRI compared with those who did not (1.7% vs 22.6%) [2]. Although this analysis was neither randomised nor prespecified and should therefore be regarded as hypothesis-generating, it raises the possibility that differences in MRI utilisation — and consequently in patient selection for breast-conserving surgery — may have contributed to the heterogeneity observed between analyses that include neoadjuvant cohorts and those restricted to upfront surgery. This same imaging-driven staging heterogeneity plausibly bears on the overall-survival finding specifically, not only on local recurrence: because patients receiving neoadjuvant therapy are almost uniformly staged with MRI, the neoadjuvant-inclusive analyses may reflect a more consistently and accurately characterised multifocal/multicentric population than the upfront-surgery sensitivity analysis, where variable MRI use raises the possibility that some patients were imprecisely classified; were preoperative MRI applied uniformly across the upfront-surgery cohorts, the overall-survival advantage associated with BCS might plausibly emerge as a more consistent finding rather than the attenuated signal currently observed — consistent with separate evidence from the neoadjuvant setting, where breast conservation with radiotherapy has been associated with improved overall survival compared with mastectomy in a cohort staged with the more uniform imaging that neoadjuvant pathways entail [19]. Future comparative studies should consider preoperative imaging strategy, alongside systemic therapy, as a potential confounder when evaluating the oncological outcomes of BCS in multiple ipsilateral breast cancer. This aligns with current guideline recommendations that MRI is most valuable when multicentricity is already suspected and breast conservation is being actively considered, rather than used universally for all patients.

Read together, these independent sources — one prospective single-arm trial, one prospective imaging cohort, one matched retrospective comparison against mastectomy, two real-world single-centre series using different imaging strategies, pooled systematic-review data, and the largest comparative meta-analysis to date — converge on the conclusion that carefully selected patients can achieve local recurrence rates in the low single digits at 5 years and oncological outcomes broadly comparable to mastectomy, but none individually, nor their combination, replaces the randomised comparison that is still absent from this literature. It is also worth emphasising that virtually all of this evidence — including Z11102 and the pooled meta-analysis above — concerns disease with two or three ipsilateral foci; data supporting breast conservation for four or more separate foci are essentially absent, and this should be treated as a de facto boundary for patient selection outside highly individualised, oncoplastically favourable cases.

The apparent equivalence between BCS and mastectomy across this evidence base should not be interpreted as evidence that the two procedures are interchangeable for all patients. Patients selected for BCS are not simply healthier; they are selected on the basis of favourable surgical anatomy — breast size, tumour-to-breast-volume ratio, lesion distribution and inter-focal distance, lesion accessibility, MRI findings, surgeon expertise, and patient preference — variables that influence both whether BCS is performed and the oncological outcomes subsequently observed. Current evidence therefore reflects outcomes among patients already judged suitable for conservation, creating an unavoidable confounding by indication that no amount of retrospective adjustment can fully resolve; this is a structural limitation of the evidence base itself, not merely of any one study within it.

Table 1. Key characteristics of the principal studies underlying the evidence base.

Study/source	Design	Population/eligibility	Imaging strategy	Follow-up	Key finding
ACOSOG Z11102 (Alliance) [2]	Prospective, single-arm	2–3 biopsy-proven ipsilateral foci, ≤2 quadrants, no NST permitted	MRI initially mandatory, later optional	5 years (median)	5-yr local recurrence 3.1%
MIPA study [13]	Prospective cohort	Patients undergoing screening or diagnostic breast MRI	MRI (index test)	Not stated	MRI changes the surgical plan more often than mammography/ultrasound
De Lorenzi et al. matched-cohort [14]	Retrospective, matched	100 oncoplastic BCS vs. 100 mastectomy, MC/MF disease	Not specified	94 months (median)	OS/DFS similar between groups; local events non-significantly higher after BCS
Argentinian cohort	Retrospective	Stage I–II MF/MC (n = 91) vs. unifocal (n = 1,097)	No routine MRI	Not stated	5-yr locoregional recurrence 3.3% vs. 5.9%
Turkish cohort	Retrospective	246/2,155 MF/MC; 68 selected for BCS	Routine MRI	12 months (median)	No local recurrences; 90% negative margins at first surgery
Comparative meta-analysis (Althawadi et al.) [15]	Systematic review and meta-analysis, 17 studies	29,711 patients (9,587 BCS, 20,124 mastectomy)	Mixed across studies	62 months (median)	No difference in LR/LRR/DR/DFS; OS advantage lost in upfront-surgery sensitivity analysis

Table 1. Key characteristics of the principal studies underlying the evidence base.

Study/source	Design	Population/eligibility	Imaging strategy	Follow-up	Key finding
ACOSOG Z11102 (Alliance) [2]	Prospective, single-arm	2–3 biopsy-proven ipsilateral foci, ≤2 quadrants, no NST permitted	MRI initially mandatory, later optional	5 years (median)	5-yr local recurrence 3.1%
MIPA study [13]	Prospective cohort	Patients undergoing screening or diagnostic breast MRI	MRI (index test)	Not stated	MRI changes the surgical plan more often than mammography/ultrasound
De Lorenzi et al. matched-cohort [14]	Retrospective, matched	100 oncoplastic BCS vs. 100 mastectomy, MC/MF disease	Not specified	94 months (median)	OS/DFS similar between groups; local events non-significantly higher after BCS
Argentinian cohort	Retrospective	Stage I–II MF/MC (n = 91) vs. unifocal (n = 1,097)	No routine MRI	Not stated	5-yr locoregional recurrence 3.3% vs. 5.9%
Turkish cohort	Retrospective	246/2,155 MF/MC; 68 selected for BCS	Routine MRI	12 months (median)	No local recurrences; 90% negative margins at first surgery
Comparative meta-analysis (Althawadi et al.) [15]	Systematic review and meta-analysis, 17 studies	29,711 patients (9,587 BCS, 20,124 mastectomy)	Mixed across studies	62 months (median)	No difference in LR/LRR/DR/DFS; OS advantage lost in upfront-surgery sensitivity analysis

Preoperative Imaging: A More Selective Role for MRI Than Sometimes Assumed

Breast MRI has the highest sensitivity of any modality for detecting additional occult foci, and this is both its principal value and the source of its principal controversy in this context. Meta-analyses have shown that MRI identifies additional malignant lesions, and correspondingly changes the surgical plan, more often than mammography and ultrasound alone [20]; the MIPA cohort demonstrates this effect prospectively at scale [13]. However, a meaningful proportion of MRI-detected additional findings are false positive on subsequent biopsy or histology — in the COMICE trial, roughly a third of MRI-prompted mastectomies were later found to have no malignancy in the additional tissue removed. Randomised evidence is more circumspect still: The BREAST-MRI trial, which randomised 524 women eligible for breast conservation to preoperative MRI or standard imaging, found that MRI increased the initial mastectomy rate from 0.4% to 8.3%, without measurable improvement in 5-year local relapse-free survival, overall survival, or reoperation rates. [21].

The exploratory subgroup finding from Z11102 is often cited as evidence that MRI should be routine in this population: patients not imaged with MRI had a 5-year local recurrence of 22.6% versus 1.7% in those who were [2] — but this comparison was neither randomised nor prespecified, involved a small number of unimaged patients, and cannot exclude confounding by indication (patients selected for surgery without MRI may have differed in other clinically relevant ways). This finding is best regarded as hypothesis-generating rather than definitive proof that MRI is mandatory in this population, particularly given that adequately powered randomised trials of preoperative MRI have shown no corresponding survival benefit. Current guideline language, and the balance of evidence reviewed here, supports a more selective position than "MRI should be done" or "MRI should be routine for everyone": MRI is most useful, and most likely to change management appropriately, in patients where multicentricity is already suspected or confirmed and breast conservation is being actively considered — that is, precisely the population under discussion in this review — rather than as a reflexive addition to every preoperative workup. Critically, any additional lesion identified on MRI should be confirmed histologically before it is allowed to alter the surgical plan; imaging findings alone, in the absence of biopsy confirmation, should not convert a patient from a breast-conservation to a mastectomy candidate.

Margin Assessment in Multisite Resection

The margin standard for invasive breast cancer treated with whole-breast irradiation — "no ink on tumour" — was established by the 2014 SSO-ASTRO consensus guideline on the basis that wider clear margins do not further reduce local recurrence [22], and this standard applies without modification to each individual focus in a multisite resection. In practice, this means that achieving "no ink on tumour" at every excised focus, rather than at a single dominant lesion, is the operative surgical target in multicentric disease, and multifocality/multicentricity has itself been shown in retrospective series to be independently associated with higher rates of positive margins and higher conversion to mastectomy at first surgery [23] — a technical, rather than purely oncologic, argument for the additional value of oncoplastic planning and, in some cases, of intraoperative margin assessment. Where DCIS accompanies invasive disease at any focus, the corresponding SSO-ASTRO-ASCO consensus recommendation of a 2 mm margin for DCIS treated with whole-breast irradiation should be applied to that focus specifically, since the two margin standards are not interchangeable within the same specimen.

Oncoplastic Techniques: Making Multisite Conservation Feasible

The principal practical enabler of breast conservation in multicentric disease has been the maturation of oncoplastic technique. As defined by the ASBrS, oncoplastic breast surgery combines an oncologic partial mastectomy with immediate repair of the resulting defect using volume-displacement or volume-replacement reconstructive technique [25]. Volume-displacement approaches (local tissue rearrangement, therapeutic mammaplasty, Wise-pattern reduction with contralateral symmetrisation) and volume-replacement approaches (pedicled or perforator flaps, including intercostal artery perforator flaps) allow resection of two or more separate tumour beds — in some series removing 20–50% or more of breast volume in aggregate — while preserving contour, nipple-areolar position, and symmetry with the contralateral breast. "Extreme oncoplasty" series applied specifically to multifocal/multicentric and locally advanced disease, in patients with adequate breast volume, report favourable BREAST-Q outcomes alongside acceptable margin rates, extending eligibility to patients who would previously have been offered only mastectomy [24]. This is consistent with AGO guidance, which states that oncoplastic surgery is oncologically safe and carries complication rates comparable to standard breast-conserving surgery, and on this basis recommends that oncoplastic technique can replace mastectomy across several indications, including multicentric or multifocal tumours [26]. Anatomically demanding sites — particularly the upper inner and central/retroareolar quadrants, where local tissue mobility is limited — remain the most technically challenging scenarios for multisite conservation and are the subject of an active case-series and technique-development literature, and cases involving these sites are often best referred to specialist oncoplastic centres rather than managed as routine multisite resections, reflecting that oncoplastic feasibility, rather than oncologic principle alone, is often the practical rate-limiting factor in offering breast conservation for multicentric disease. As above, the matched-cohort data from De Lorenzi et al. provide the strongest currently available comparative evidence that this approach does not compromise survival relative to mastectomy [14], while the authors themselves note that a double radiotherapy boost after double lumpectomy cavities, and other technical questions specific to multisite oncoplasty, remain without prospective, randomised evidence. Delivering this approach at scale also has resource implications worth acknowledging for guideline implementation: multisite oncoplastic resection, double-boost radiotherapy planning, and multidisciplinary case discussion require specialist-centre capacity and training that is not uniformly available, which may itself be a rate-limiting factor independent of oncologic or technical feasibility.

Radiotherapy Planning Considerations

Z11102's protocol — whole-breast irradiation with a boost delivered to every lumpectomy bed — is the only prospectively tested radiotherapy approach in this population and is the de facto standard reflected in most subsequent series [2]. Several technical questions remain unresolved and are addressed inconsistently across the literature: whether two separate cavity boosts are needed when lumpectomy sites are close together versus widely separated within the breast, how cavities should be localised and delineated when more than one resection has occurred (clip placement at the time of surgery is widely used for this purpose but is not standardised across the multicentric-disease literature specifically), and whether cumulative boost dose to overlapping treatment volumes carries additional toxicity. In the absence of dedicated randomised data, radiotherapy planning for multisite resection is generally extrapolated from single-site boost practice, and this extrapolation — while clinically reasonable — is acknowledged in the oncoplastic and radiation oncology literature as a genuine, not merely theoretical, gap.

Neoadjuvant Systemic Therapy and Response-Guided Surgery

Neoadjuvant systemic therapy (NST) has reshaped surgical planning in breast cancer generally, and multicentric disease is no exception. For biologically aggressive subtypes — triple-negative and HER2-positive disease in particular — and for patients with an unfavourable tumour-to-breast-volume ratio or multifocal/multicentric disease, NST offers the opportunity to downstage individual foci and convert patients who would otherwise require mastectomy into breast-conservation candidates; professional guidance now explicitly encourages consideration of NST for this purpose [25]. Practical surgical planning around NST in this population depends on response-guided principles rather than a fixed protocol: imaging (mammography, ultrasound, and MRI) is repeated after treatment to reassess the extent and distribution of residual disease at each focus; radio-opaque or other localisation clips are placed in each individual tumour bed before treatment begins, since foci that respond well — including those achieving a complete radiological response — can otherwise become impossible to localise for targeted excision; and final resection is planned around the post-treatment extent of residual disease at each site rather than the pretreatment extent of disease, provided each original clip-marked site remains excisable to a negative margin.

Confirming that all disease has been removed depends on more than imaging and clip placement alone: intraoperative specimen radiography of each excised focus, correlation between the radiological and pathological extent of disease at each site, and explicit radiology-pathology discussion at multidisciplinary meeting are practical safeguards against under-excision in multisite resection, and take on particular importance after neoadjuvant therapy, when the post-treatment radiological appearance may understate residual viable tumour at an individual focus.

The evidence base specific to breast conservation after NST in multiple ipsilateral breast cancer is smaller than for upfront surgery — Z11102 itself excluded patients who received neoadjuvant therapy — though a large cohort from the GeparTrio, GeparQuattro, and GeparQuinto neoadjuvant chemotherapy trials (n = 6,134) has examined the impact of focality on surgery type and survival outcomes after NST [26], and more recent single-institution series addressing breast conservation specifically after NST in this population are only beginning to report outcomes. Guideline bodies remain correspondingly cautious: AGO guidance explicitly notes that multicentricity may still justify mastectomy even when a good response to neoadjuvant therapy has been achieved at each focus [27], reflecting persistent uncertainty about whether a favourable radiological or pathological response reliably predicts durable local control after conservation in this specific population. This caution is not unique to AGO — it reflects a broader evidence gap shared across guideline bodies, since long-term local-control data specific to breast conservation after a good neoadjuvant response in multicentric disease remain sparse regardless of which group is asked.

Hereditary Breast Cancer: A Distinct Consideration

Multicentricity does not exist in isolation from hereditary cancer risk, and the two considerations should not be conflated in surgical planning. Pathogenic variants in BRCA1, BRCA2, and PALB2 are associated with an increased likelihood of multicentric and bilateral disease, and germline testing is indicated in this population according to standard clinical and family-history criteria regardless of whether the presenting tumour is unifocal or multicentric. Among BRCA1/BRCA2 carriers who undergo breast-conserving therapy, cohort data show a higher rate of ipsilateral breast events compared with non-carriers — a landmark comparative series reported a 15-year cumulative ipsilateral risk of 23.5% after breast conservation versus 5.5% after mastectomy [28] — without a corresponding difference in overall survival, since most subsequent ipsilateral events in carriers represent new primary cancers rather than true recurrences of the treated tumour. Framed this way, the relevant clinical conversation for a BRCA carrier with multicentric disease is not whether breast conservation is oncologically unsafe — the survival data do not support that — but how much residual, largely new-primary risk the patient is willing to accept relative to the more radical risk reduction offered by mastectomy. This means the decision between breast conservation and mastectomy in a BRCA carrier with multicentric disease is properly framed as a risk-reduction and shared decision-making discussion — encompassing contralateral risk-reducing mastectomy, surveillance intensity, and the patient's own risk tolerance — rather than as a strictly oncologic determination based on the index tumour's focality alone.

The Current Guideline Landscape

No major guideline body has abandoned caution around multicentric disease, but the direction of travel across guidelines is consistent: from a categorical contraindication toward a conditional, multidisciplinary pathway.

Guideline body	Position on BCS in multicentric disease	Key caveat
American Society of Breast Surgeons (ASBrS)	Multicentric disease not amenable to oncoplastic surgery remains listed as an absolute contraindication, but oncoplastic technique and neoadjuvant therapy are explicitly recognised as expanding eligibility, and practice is expected to adapt to patient preference and evolving evidence [11].	Caveat applies only where oncoplastic surgery cannot achieve the goal — not a blanket rule against conservation.
NCCN	No disease-specific rule for multicentricity; general early breast cancer framework applies, with the surgical decision left to individualised, multidisciplinary judgement.	No disease-specific caveat — defers entirely to multidisciplinary judgement.
ESMO	Similarly does not issue a prescriptive rule; emphasises multidisciplinary team (MDT) selection within its general early breast cancer framework.	No disease-specific caveat — no prescriptive criteria issued for MIBC.
AGO (German Cancer Society, Breast Committee)	Permits breast-conserving therapy in multifocal/multicentric disease conditional on R0 resection at every focus, but explicitly notes multicentricity may still justify mastectomy even after a good neoadjuvant response [27,29].	Caution applies even after a good neoadjuvant response, reflecting limited long-term local-control data in that specific setting.

The practical convergence across these frameworks is that imaging adequacy, achievability of negative margins at every focus, and cosmetic feasibility function as the operative gatekeeping criteria, applied through multidisciplinary discussion rather than a fixed anatomical rule.

A further point deserves emphasis: guideline caution and clinical equivalence are not the same thing. Guideline bodies appropriately retain caution around multicentric disease because prospective comparative data against mastectomy remain absent, and this caution is a defensible response to genuine uncertainty. However, that caution should not be read as evidence that the historical, categorical contraindication and a contemporary, evidence-based selective approach carry equal clinical validity — the retrospective and pooled evidence reviewed here has moved materially beyond the assumptions that originally justified reflexive mastectomy, even though no guideline has yet been able to formalise that shift into a prescriptive rule. Guideline conservatism, in other words, reflects the absence of a particular kind of evidence (randomised comparison) rather than the presence of evidence against conservation.

Patient-Reported Outcomes and the Cosmetic Dimension

Cosmesis is intrinsic to the rationale for offering conservation at all in this population — if multisite excision routinely produced unacceptable deformity, the argument for avoiding mastectomy would be considerably weaker regardless of local recurrence data. The Z11102 cosmesis substudy, using a four-point cosmesis scale and the BREAST-Q, found patient- and surgeon-reported cosmetic outcomes after multisite lumpectomy and whole-breast radiation broadly comparable to single-site breast conservation [30], and extreme-oncoplasty series report similarly favourable BREAST-Q results [24]. However, formal comparative patient-reported outcome data specific to multisite resection remain limited relative to the volume of oncologic outcome data; most published series are single-centre, lack a contemporaneous mastectomy-with-reconstruction comparator, and report outcomes at relatively short follow-up, which limits how confidently patients can currently be counselled on relative cosmetic outcomes between the two approaches.

Synthesis: A Practical, Shared-Decision Framework for Patient Selection

Drawing the prospective, retrospective, and guideline evidence together, five conditions recur across the literature as practical prerequisites for offering breast conservation in multicentric disease:

• Imaging adequacy with histologic confirmation. Preoperative imaging — MRI included where multicentricity is suspected and conservation is being considered — has characterised the extent and distribution of disease, and any additional lesion that would change management has been confirmed histologically rather than acted upon by imaging appearance alone.
• Complete excision with acceptable cosmesis. All identified foci can be excised to a negative margin ("no ink on tumour" for invasive disease, 2 mm for accompanying DCIS) using standard or oncoplastic technique, while preserving an acceptable cosmetic result — in practice usually requiring access to oncoplastic expertise. Two practical gatekeepers deserve explicit attention: the tumour-to-breast-volume ratio, since even skilled oncoplastic technique cannot always remove multiple large-volume foci without compromising cosmesis; and anatomical site, since central/retroareolar and upper-inner-quadrant foci are technically more challenging and may warrant referral to a specialist oncoplastic centre.
• Adjuvant radiotherapy. Whole-breast radiotherapy with a boost to each lumpectomy bed is delivered, acknowledging that some technical aspects of multisite boost planning remain without dedicated evidence.
• Guideline-concordant systemic therapy. Endocrine, HER2-targeted, or chemotherapy is administered according to standard biomarker-driven indications, given the specific Z11102 signal linking endocrine therapy adherence to local recurrence risk [2].
• Shared decision-making, informed by hereditary risk where relevant. The patient has been counselled on the comparative evidence — including its genuine limitations — and, where a pathogenic variant is identified or strongly suspected, on how that risk should inform the choice between conservation and mastectomy independent of focality.

Table 2. Minimum requirements before offering breast-conserving surgery in multicentric disease.

Domain	Minimum requirement before offering BCS
Imaging	Disease extent defined by preoperative imaging; any suspicious MRI lesion confirmed histologically before it changes the surgical plan
Biology	Guideline-concordant systemic therapy plan established according to biomarker status (endocrine, HER2-targeted, or chemotherapy)
Surgery	All identified foci excisable to a negative margin (“no ink on tumour” for invasive disease; 2 mm for accompanying DCIS)
Reconstruction	Access to oncoplastic expertise for volume-displacement or volume-replacement technique where multisite resection is planned
Radiation	Whole-breast radiotherapy with a boost to each lumpectomy bed is feasible
Patient factors	Informed preference reached after multidisciplinary discussion, including hereditary risk where relevant

Table 2. Minimum requirements before offering breast-conserving surgery in multicentric disease.

Domain	Minimum requirement before offering BCS
Imaging	Disease extent defined by preoperative imaging; any suspicious MRI lesion confirmed histologically before it changes the surgical plan
Biology	Guideline-concordant systemic therapy plan established according to biomarker status (endocrine, HER2-targeted, or chemotherapy)
Surgery	All identified foci excisable to a negative margin (“no ink on tumour” for invasive disease; 2 mm for accompanying DCIS)
Reconstruction	Access to oncoplastic expertise for volume-displacement or volume-replacement technique where multisite resection is planned
Radiation	Whole-breast radiotherapy with a boost to each lumpectomy bed is feasible
Patient factors	Informed preference reached after multidisciplinary discussion, including hereditary risk where relevant

The value of this framework is best illustrated by contrast. Two patients with multicentric disease may appropriately receive different recommendations even though both nominally meet a MIBC definition. Consider a patient with a small breast, three foci in difficult anatomical locations, and an expected cosmetic result that oncoplastic technique cannot meaningfully improve — mastectomy, with or without reconstruction, is the more appropriate recommendation. Contrast this with a patient with a larger breast, two well-separated small tumours, straightforward oncoplastic feasibility, and a good response to systemic therapy — breast conservation is oncologically reasonable and is the recommendation this review supports. Both patients carry the same anatomical label; the appropriate surgery differs. This is precisely why anatomical labels alone — multifocal, multicentric, or the number of foci in isolation — are inadequate as a basis for the conservation-versus-mastectomy decision, and why the five conditions above, applied through multidisciplinary discussion, are the more clinically meaningful unit of assessment.

When these conditions are met, the aggregate evidence reviewed here suggests local recurrence rates in the low single digits at 5 years, broadly comparable to unifocal disease treated by breast conservation — a materially different clinical picture from a categorical "multicentricity equals mastectomy" position, while still requiring more deliberate, multidisciplinary evaluation than is typical for unifocal disease.

4. Gaps in the Evidence and Future Directions

Several questions remain genuinely unresolved. No randomised trial has directly compared breast conservation with mastectomy in multicentric disease, and equipoise and enrolment challenges make it plausible that none ever will, meaning the field will likely continue to depend on single-arm and retrospective comparative data for the foreseeable future. [15] Although prospective randomised clinical trials remain the ideal way to confirm this, this paradigm shift is evolving rapidly, and as clinical evidence continues to accumulate alongside advances in oncoplastic surgery, it is becoming increasingly difficult to justify — or even to conduct — a randomised controlled trial in this setting. The current body of evidence provides reassuring support for the oncological safety of breast-conserving surgery in appropriately selected patients with multicentric disease. The necessity of routine preoperative MRI remains unresolved for the reasons discussed earlier in this review, and this review has deliberately not resolved that tension, since the underlying trials do not resolve it either. Technical questions in oncoplastic and radiotherapy planning — optimal boost strategy after multisite resection, sequencing of contralateral symmetrisation, and management of anatomically difficult sites — remain addressed mainly by case series and single-centre reports rather than comparative studies. Evidence on breast conservation after NST specifically in multicentric disease is the least mature of the areas reviewed here, since the pivotal prospective trial excluded these patients entirely. Finally, growing evidence of genomic heterogeneity between foci — including cases where multifocal lesions are demonstrably clonally unrelated despite matched histology [5] — raises the possibility that future classification of this disease will shift from anatomical toward biological criteria, which could refine, in either direction, which patients are genuinely appropriate for breast conservation. Such classification may ultimately determine not only surgical eligibility but also whether all tumour foci require identical local and systemic treatment strategies.

Among these open questions, four are the most immediately actionable for future research: (1) oncological outcomes of breast conservation after neoadjuvant therapy specifically in multicentric disease; (2) optimal radiotherapy boost planning for multiple lumpectomy beds; (3) whether routine preoperative MRI offers a net benefit in this specific subgroup; and (4) the clinical implications of genomic heterogeneity between separate foci for treatment selection.

Limitations of This Review

This is a narrative rather than a systematic review, and it carries the limitations inherent to that methodology: study selection was based on relevance and quality as judged by the author rather than a predefined, reproducible search protocol, formal risk-of-bias assessment was not performed, and publication or citation bias cannot be excluded. The underlying evidence itself is heterogeneous, with inconsistent definitions of multifocal and multicentric disease, variable imaging strategies, and differing systemic therapy approaches across the studies discussed, which limits how precisely any single figure quoted here should be interpreted.

5. Conclusions

The categorical historical position that multicentric breast cancer mandates mastectomy is no longer supported by the totality of the evidence [3], but it has not been replaced by an equally categorical rule in the other direction, nor should it be. What has emerged instead is a conditional, multidisciplinary pathway: for appropriately selected patients with multicentric disease, breast-conserving surgery appears to provide oncological outcomes that are not inferior to those reported after mastectomy in appropriately selected patients, although direct randomised comparison remains unavailable, provided that complete excision with negative margins is achieved, adjuvant radiotherapy is administered, and an acceptable cosmetic outcome can be attained. This approach depends on adequate, histologically confirmed preoperative imaging, resectable disease amenable to oncoplastic techniques, access to appropriate adjuvant radiotherapy, guideline-concordant systemic therapy, and—where relevant—informed consideration of hereditary cancer risk. Under these conditions, local recurrence rates comparable to those reported in contemporary breast-conservation series of selected unifocal breast cancers can be achieved. The open questions identified here—particularly the necessity of routine MRI, radiotherapy planning for multisite resection, and outcomes after neoadjuvant therapy—mean that patient selection, informed consent, and multidisciplinary discussion remain as important to outcomes as surgical technique itself. The future of MIBC surgery is unlikely to be defined by a universal preference for conservation or mastectomy, but by accurate disease characterisation, multidisciplinary selection, and matching the extent of surgery to tumour biology rather than anatomical multiplicity alone.