Innovative Standards in Surgery of the Breast after Neoadjuvant Systemic Therapy

Abstract

Background

The goal of neoadjuvant systemic therapy (NST) in breast cancer is to downstage tumors and downgrade treatment. Indications are constantly evolving. These changes raise practical questions for planning of surgery after NST.

Summary

In this review we discuss current evolving aspects of surgery of the breast after NST. Breast-conserving surgery (BCS) eligibility increases after NST − both neoadjuvant chemotherapy (NAC) and neoadjuvant endocrine therapy. Adequate margin width in NST and upfront surgery are similar − “no tumor on ink” for invasive cancer. Oncoplastic breast surgery after NST is feasible − both for BCS and mastectomy with reconstruction. There is increasing interest in the possibility of omitting surgery in patients with a complete response to NAC. Several trials are being conducted in aim of achieving acceptable prediction of pathological complete response, by combination of imaging and percutaneous biopsy of the tumor bed, as well as assessing the safety of such an approach.

Key Messages

Surgery of the breast after NST should be determined not only according to biologic and anatomic parameters at diagnosis, but is dynamic, and must be tailored according to the response to therapy. The omission of surgery in exceptional responders after NAC is being explored.

Introduction

Oncological safety of neoadjuvant systemic therapy (NST) for breast cancer in terms of locoregional recurrence and survival has been established, and it is being used increasingly [1, 2, 3]. The indications for NST have evolved over the last decades. Originally used to convert inoperable tumors to operable, and later to convert mastectomy to breast-conserving surgery (BCS) [3], it is now often used to decrease surgical morbidity by de-escalating axillary surgery, reducing rates of margin involvement and reducing the volume of tissue resected. NST also allows in vivo evaluation of treatment effect, and provides time for additional preoperative evaluation such as genetic testing [4]. The decision regarding NST must be tailored by a multidisciplinary team, as the different treatment modalities influence each other [5].

The goal of NST is to downstage the tumor in the breast and axilla. The ultimate downstaging is pathological complete response (pCR). Definitions of pCR range from no residual invasive to no residual invasive or in situ disease in the breast, with or without considering the axilla. pCR rates vary between tumor subtypes and NST regimens, and it is a positive prognostic marker for event-free survival and overall survival (OS) [3, 6, 7, 8]. In triple negative (TNBC) and HER2-enriched subtypes, additional adjuvant systemic therapy improves outcomes when pCR is not achieved [9, 10]. Many of these questions have been addressed by a panel of experts and presented as the “Lucerne Toolbox,” which covers initial workup, response evaluation, and surgical options [5].

The use of NST raises practical questions for surgery planning after treatment. In this review we discuss controversial and evolving aspects of surgery of the breast after NST. Axillary surgery after NST is discussed elsewhere.

BCS and Margins after NST

Increase in BCS Rates after NST

BCS, complete resection of the tumor with negative margins with adjuvant radiotherapy is equivalent to mastectomy after NST in terms of disease-free survival (DFS) and OS [7, 11, 12, 13].

NST has been shown to increase BCS rates [3]. In the NSABP B-18 trial, BCS rates increased from 60% to 68% [1, 14]. HER2-enriched and TNBC have even higher rates of conversion to BCS following neoadjuvant chemotherapy (NAC) [15, 16]. In a retrospective cohort, Fitzal et al. reported that patients who responded to NST and converted from mastectomy to BCS had a similar locoregional recurrence-free survival (LRFS), with a median follow up of 60 months [17].

Margins after NST

Inadequate margins are associated with higher local recurrence rates. Updated SSO-ASCO guidelines refer to “no ink on tumor” as adequate for upfront BCS in invasive breast cancer, but these guidelines do not refer to patients after NST [18]. Despite past studies that showed mixed results [3], recent studies did not demonstrate an advantage for margins wider than “no ink on tumor.” Choi et al. reported no association between margin width and LRFS, DFS, or OS. Therefore “no ink on tumor” can be applied after NST [11, 12]. In a recent study by Lin et al., 161 patients who underwent BCS after NAC showed no difference in LRFS between ≤1 mm and >1 mm margins [13].

Methods to reduce involved margins are similar to those used in upfront surgery. Intraoperative ultrasound guidance and wire localization have shown similar positive margin, re-excision, and local recurrence rates [19]. The MarginProbe device, which uses radiofrequency spectroscopy, has shown a trend towards reduction in re-excision rates for positive margins after NAC (31% vs. 6%) [20]. In a review from the American Society of Breast Surgeons registry, NST was the highest-ranked actionable factor associated with fewer reoperations for positive margins, a finding corroborated by several other studies [13, 21, 22].

Management of Calcifications

Currently, residual suspicious calcifications after NST need to be resected. Most calcifications associated with breast cancer persist on mammography after NST, and there is no accurate method of discriminating pCR from residual disease [23]. Feliciano et al. reported on 90 patients and found no correlation between changes in calcifications on mammogram and changes in enhancement on MRI. Loss of MRI enhancement strongly correlated with pCR (p < 0.0001), but no relationship was found between changes in calcifications and pCR rates [24].

Is It Necessary to Remove the Original Tumor Volume?

A survey of British breast cancer teams reported that 74% of centers practice response-adapted surgery, whereas 26% stated that they resect the original tumor footprint [25]. The 2017 St. Gallen consensus panel voted on de-escalation of surgery after NST: 82% of panelists voted that surgical resection should include only the post-NST extent of the tumor, and only 14% voted that the original (pretreatment) tumor bed needs to be excised [26].

Oncoplastic Breast Surgery following NST

Oncoplastic techniques apply principles of plastic surgery to facilitate breast conservation with adequate margins and improve cosmetic results. These techniques have been shown feasible also for locally advanced breast cancer after NST [27].

Breast-Conserving Surgery

Oncoplastic breast surgery (OPBS) allows resection of larger breast volumes while maintaining negative margins and acceptable cosmetic results, with similar re-excision, local recurrence, and distant relapse rates compared with non-oncoplastic procedures, and is increasingly used to facilitate breast conservation [5, 28].

In a study by Adamson et al., NAC was not associated with higher complication rates after OPBS [29]. Earlier studies reported higher rates of reoperation for postoperative complications after OPBS (2% vs. 1%, p = 0.007) [28].

Mastectomy with Reconstruction

Similar rates of locoregional recurrence, DFS, and distant metastasis-free survival were reported after NAC for simple mastectomy and mastectomy with immediate reconstruction [30].

Agresti et al. reported no significant difference in locoregional recurrence with a median follow-up of 46 months for NSM, NSM after NAC, or total mastectomy after NAC [31]. NAC was not associated with elevated rates of nipple loss or other perioperative complications after NSM compared with adjuvant chemotherapy [32, 33, 34, 35]. The Oncoplastic Breast Consortium (OPBC) is an international group of breast surgeons which aims to promote OPBS. In the OPBC consensus conference on NSM, the majority voted that oncologic safety does not depend on NST [36].

Autologous reconstruction with free flap microsurgical anastomosis following NAC is reported to have similar complication rates compared to upfront surgery [37, 38]. A 2019 study by Beugels et al. found no difference in the incidence of complications after mastectomy and deep inferior epigastric perforator reconstruction with or without NAC [39].

Omission of Surgery after NAC

As in other aspects of breast cancer treatment, de-escalation of surgery is attempted when oncologically safe. The option of omitting surgery after NAC for patients with complete clinical and radiological response was proposed in the past, but studies showed unacceptable false-negative rates (FNR) [40, 41]. However, some studies included only physical examination to assess response, did not discriminate between cancer subtypes, or used suboptimal imaging.

Advocates for omitting surgery claim there is no benefit for surgery when there is no residual disease after NST. In cases of pCR, surgery serves as a method of response assessment, and not a means of disease eradication. A retrospective study of patients receiving NAC compared surgery to nonsurgical treatment and showed no significant difference in 5-year OS when pCR or clinical complete response was achieved (92.5% vs. 96.8%, respectively, p = 0.15); however, this study was limited by a relatively short follow-up and small number of patients [42]. Advocates of omitting surgery also mention persistent pain and dissatisfaction with breast shape after surgery. Opponents mention the relatively small scope of surgery with BCS, the need for frequent follow-up exams, and patient preference and sense of security with surgery [43, 44].

Prediction of Response to NAC

In order to promote omitting surgery after NAC, it is crucial to identify patients who are potential exceptional responders, as well as those who have obtained a complete response. Biomarkers such as tumor subtype, Ki67, molecular subtype, and inflammation-based prognostic indexes may assist in response prediction and patient selection [45]. For example, patients with HER2-enriched tumors, TNBC, and N1 status (among other factors) are expected to have higher response rates [46, 47]. Even in ER-positive tumors, intrinsic subtyping can predict pCR, with luminal A tumors showing a significantly lower rate of pCR [48].

Methods that rely solely on imaging to assess response to NAC are inconsistent [49]. A review of 16 studies revealed methodologic and technical heterogeneity. Several studies relied on the appearance of the tumor pretreatment and others on changes in tumor appearance on mid-treatment imaging [50, 51]. A study of 102 patients with pre- and posttreatment MRI showed a sensitivity of 80.3%, specificity of 78.0%, and an FNR of 27.2% in predicting pCR [52]. Other imaging methods such as contrast-enhanced mammography or dedicated breast PET scans are also reported as possibly useful [53, 54].

Management of Residual Ductal Carcinoma in situ

Residual ductal carcinoma in situ (DCIS) might not affect long-term outcome [55] but has significant implications on the extent of surgery. Patients with DCIS on pretreatment core biopsy were less likely to achieve pCR in the breast (31% vs. 43%; p = 0.038), and commonly DCIS is not eradicated by chemotherapy [47, 56]. In the CALGB 90403 trial, DCIS volume decreased by 71% on MRI, and pCR was seen in 15% [57].

Surgery Omission Feasibility Studies

Since there is no single biomarker or imaging modality that accurately predicts pCR, studies of feasibility of surgery omission in patients with radiological complete response incorporate tumor bed targeted biopsies. Clinical trials have focused on methodology (assessment of tumor bed, needle thickness, number of cores, and biopsy method) and patient selection (tumor subtype, types of imaging abnormalities, patient preference) [58].

These trials aimed to determine whether pCR can be accurately predicted with a combination of posttreatment imaging and guided biopsies. All included a relatively small number of patients and were intended as pilot studies. Disappointingly, these studies mostly reported an unacceptable FNR of over 10% (Table 1) [59].

Table 1.

Selected publications of surgery omission after neoadjuvant chemotherapy

Completed studies
Author, study, year [reference]	Number and receptor status of patients (if mentioned)	Response inclusion criteria	Post-NAC assessment	Biopsy details	Results	Specific details
Heil, 2016 [61]	50	Clinical/radiological partial/complete response	Mammography, US, and MRI	US guided VAB, 9G, 6–12 biopsies	NPV − 76.7%, FNR − 25.9%; NPV − 94.4%, FNR − 4.8% when pathologically representative	Pathologically representative − residual tumor or tumor bed
Francis, NOSTRA PRELIM, 2017 [59]	20	No response criteria	Not specified	US guided, 2–6 cores	FNR–4/18	3 of 4 FN cases had only 3 cores
Kuerer, 2018 [60]	40, HER2 enriched or TNBC	Radiological partial/complete response	Mammography, US	Mammography or US-guided FNA and VAB, 9G, 4–14 cores (median 12)	NPV–95%, FNR–5%	pCR (ypT0) −47.5%, breast-axilla concordance − 97.5%
Narui, 2020 [64]	83	Radiological complete response	MRI	US guided, 14G, median 3 cores	NPV for ypT0 −66.1%, NPV for ypT0/is − 89.1 %, FNR for ypT0 − 50%, FNR for ypTO/is − 28%	Without clip localization
Lee, 2020 [63]	40	Radiological near complete response	MRI	US guided, CNB or VAB, 14G, at least 5 cores	NPV − 87.1%, FNR − 30.8%	NPV − 100% in ER positive, 14.8% had breast pCR and axillary residual disease
Sutton, 2021 [62]	20	Radiological complete response	MRI	MRI guided, 9G, 7–12 samples	NPV for ypT0 − 92.8%, NPV for ypT0/is − 85.8%, FNR for ypT0 − 20%, FNR for ypTO/is −11.1 %	85% HER2 enriched or TNBC

Ongoing studies
Author, study [reference]	Planned number and receptor status of patients (if mentioned)	Response inclusion criteria	Post-NAC assessment	Biopsy details	Interim results	Specific details
van der Noordaa, MICRA [65]	525, all hormonal subtypes	Radiological partial/complete response	MRI	US guided, 14G, 8 cores	167 patients included for analysis, FNR − 37% (of 78 patients with residual disease)
Heil, RESPONDER [66]	600	Radiological partial/complete response	Mammography or US	Mammography or US-guided VAB, 10G or less, at least 12 passes	FNR − 17.8% (398 patients), 32.4% of them DCIS, 54.1% of them ypT <5 mm	FNR − 0 with 7G biopsy
Basik, NRG study BR005 [67]	175	Clinical complete response, radiological complete/near complete response	Mammography, US, and MRI	Mammography-guided VAB	NPV–77.5% (105 patients)
Rea, NOSTRA [68]	150, HER2 enriched ER negative	Not stated	Not stated	US guided, up to 8	n/a	Outcomes: FNR, local and distant recurrence, OS
Kuerer [69]	50, HER2 enriched, TNBC	Exclusion: clinical progression of disease >20% in the breast or new nodal metastases	Not stated	Not stated	n/a	Surgery vs. WBRT + boost, primary outcomes: IBTR-free survival, OS

NAC, neoadjuvant chemotherapy; US, ultrasound; MRI, magnetic resonance imaging; VAB, vacuum-assisted biopsy; NPV, negative predictive value; FNR, false-negative rate; FN, false negative; TNBC, triple negative breast cancer; FNA, fine needle aspiration; pCR, pathological complete response; CNB, core needle biopsy; OS, overall survival; WBRT, whole breast radiation therapy; IBTR, ipsilateral breast tumor recurrence.

As expected, FNR and negative predictive value (NPV) were better in HER2-enriched and TNBC subtypes. Larger gauge needles and multiple biopsy samples also improved the FNR and NPV [60, 61, 62]. Some studies included near or partial response to NAC as well as complete response [63]. Outcomes also depended on the definition of pCR. Narui et al. reported an FNR of 50% for detection of ypT0, and 28% for ypT0+Tis [64]. The use of a combination of fine needle aspiration and vacuum-assisted biopsy of the tumor bed reduced the FNR to 5%.

Ongoing trials, like MICRA [65], RESPONDER [66], NRG-BR005 [67], and NOSTRA [68], are enrolling a larger number of patients in an attempt to identify protocols for post-NAC evaluation which will result in a lower FNR and improved NPV (Table 1).

Ongoing Clinical Studies of Surgery Omission after NAC

An MD Anderson study is currently enrolling 50 HER2-enriched and TNBC patients after NAC with image-guided biopsies without residual cancer. Patients will receive surgery followed by radiation versus standard breast radiotherapy alone. Primary endpoints are ipsilateral breast tumor recurrence and OS. Several secondary endpoints, including change in biomarkers such as circulating tumor cells and cDNA, as well as quality of life, will be assessed [69].

Despite the abovementioned trials, there is currently a clear recommendation for surgery after NAC in all cases except rare cases of patients at extremely high risk for surgery or those enrolled in a clinical trial.

Neoadjuvant Endocrine Therapy and Surgery of the Breast

Neoadjuvant endocrine therapy (NET) is a safe and more tolerable alternative compared with NAC in selected patients with ER-positive tumors. An increase in the use of NET has been reported lately [70], but it is still probably underutilized.

pCR after NET

The goals of NET are similar to those of NAC, although the presence of residual disease will usually not alter adjuvant treatment. Response to NET has less of a prognostic significance compared with NAC. Objective response rates for NET are high (over 70%), but rates of pCR are low and usually do not exceed 10% [71].

NET versus NAC for BCS

A common indication for NET is to convert mastectomy to BCS, although this is achieved at lower rates than NAC [25]. Semiglazov et al. reported a phase 2 trial of NET compared with NAC in 239 postmenopausal patients with ER-positive breast cancer, all ineligible for BCS at diagnosis. Rates of clinical complete response were 10% in both groups, pCR was 3% versus 6%, respectively (p > 0.05), and BCS was feasible in 33% versus 24% [72]. In ACOSOG Z1031 overall BCS rates were 68%, with conversion rates of 50.9% from “obligatory mastectomy” and 83.1% from “marginal for breast conservation” with no difference in BCS rates between luminal A and B tumors [73].

Montagna et al. conducted a study including patients borderline eligible or ineligible for BCS. Among non-BCS candidates after NET, 64% became eligible for BCS. Among borderline candidates, 88% became BCS eligible, which was successful in 95% of patients who elected BCS. Overall, 77% of the patients with a large tumor relative to breast size became eligible for BCS after NET, and 66% avoided a mastectomy [71].

A review of the National Cancer Data Base reported higher rates of BCS in NET compared with primary surgery for all T stages (46.4 vs. 43.9%; p = 0.02), persisting in multivariable analysis with an adjusted OR of 1.60 [74].

Conclusions

Multidisciplinary treatment planning is crucial for optimizing results in breast cancer surgery. Surgery of the breast after NST should be determined not only according to biologic and anatomic parameters at diagnosis, and response to therapy must be considered. The omission of surgery in exceptional responders after NAC is being explored, but not yet proven to be safe and should not be utilized in clinical practice outside of a clinical trial.

Conflict of Interest Statement

All authors have no conflicts of interest to declare.

Funding Sources

No funding was granted for this review.

Author Contributions

T.H. − literature review, designing, drafting and revising the paper. Approval of final version. M.K. − designing, drafting and revising the paper. Approval of final version. M.N. − designing, drafting and revising the paper. Approval of final version. T.M.A. − designing, drafting and revising the paper. Approval of final version.