Long-term Oncologic Outcomes of Immediate Breast Reconstruction vs Conventional Mastectomy Alone for Breast Cancer in the Setting of Neoadjuvant Chemotherapy

Key Points

Question  How do the long-term oncologic outcomes of nipple- or skin-sparing mastectomy and immediate breast reconstruction (IBR) compare with those of conventional mastectomy (CM) alone for breast cancer after neoadjuvant chemotherapy?

Findings  In this case-control study, no significant difference was noted between the IBR and CM-alone groups in 5-year local recurrence–free, disease-free, and distant metastasis–free survival rates. In addition, there was no significant difference between the groups in 5-year overall survival rates.

Meaning  In this study, the long-term oncologic outcomes of nipple- or skin-sparing mastectomy and IBR for breast cancer appeared to be comparable to those of CM alone after neoadjuvant chemotherapy, suggesting the feasibility of nipple- or skin-sparing mastectomy and IBR in the current setting.

Importance  An increasing number of patients with breast cancer receiving neoadjuvant chemotherapy (NACT) undergo immediate breast reconstruction (IBR) with nipple-sparing mastectomy (NSM) or skin-sparing mastectomy (SSM) as surgical treatment. However, the oncologic efficacy and safety of this treatment sequencing strategy is unclear.

Objective  To compare the long-term oncologic outcomes of IBR with NSM/SSM and conventional mastectomy (CM) alone for breast cancer in the NACT setting.

Design, Setting, and Participants  A retrospective, propensity score–matched case-control study was conducted at Asan Medical Center, Seoul, Korea. A total of 1266 patients with breast cancer who underwent NACT followed by mastectomy with or without breast reconstruction between January 1, 2010, and November 30, 2016, were included. Data analysis was performed from July 1, 2019, to January 24, 2020. After propensity score matching, 323 patients who underwent IBR with NSM/SSM and 323 who underwent CM alone were selected for comparison of long-term oncologic outcomes.

Main Outcomes and Measures  The 5-year local recurrence–free survival, disease-free survival, distant metastasis–free survival, and overall survival rates were calculated using the Kaplan-Meier method and compared using log-rank tests. Hazard ratios (HRs) and 95% CIs were estimated using the Cox proportional hazards regression model.

Results  After matching, the median follow-up periods were 67 (range, 17-125) months for the IBR group and 68 (range, 17-126) months for the CM-alone group. Median age of the women in the IBR group was 42 (range, 23-72) years; median age of those in the CM-alone group was 46 (range, 30-75) years. No significant differences were observed between the IBR and CM-alone groups in local recurrence (3.7% vs 3.4%; P = .83), regional recurrence (7.1% vs 5.3%; P = .33), or distant metastasis (17.3% vs 18.6%; P = .68) rates. There was also no significant difference between the IBR and CM-alone groups in 5-year local recurrence–free survival (95.6% vs 96.7%; HR, 1.124; 95% CI, 0.495-2.549; P = .78), disease-free survival (76.5% vs 79.9%; HR, 1.089; 95% CI, 0.790-1.500; P = .60), distant metastasis–free survival (82.5% vs 82.5%; HR, 0.941; 95% CI, 0.654-1.355; P = .74), or overall survival (92.0% vs 89.3%; HR, 0.847; 95% CI, 0.530-1.353; P = .49) rates.

Conclusions and Relevance  The long-term oncologic outcomes of IBR with NSM/SSM for breast cancer in this study appeared to be comparable to those of CM alone after NACT, suggesting the feasibility of IBR with NSM/SSM in the NACT setting.

The surgical management of breast cancer has evolved substantially in recent decades. Breast-conserving surgery has replaced radical treatment approaches and is now an established standard of care. However, there has been a reversing trend in which the mastectomy rate is increasing owing to a growing preference for postmastectomy breast reconstruction and a paradigm shift in mastectomy patterns.^1,2 Nipple-sparing mastectomy (NSM) or skin-sparing mastectomy (SSM) with immediate breast reconstruction (IBR) has gained popularity in the treatment of selected patients with breast cancer owing to increasing recognition of the improved aesthetic results and quality of life without compromising oncologic safety.³

As another important component of modern breast cancer care, neoadjuvant chemotherapy (NACT) has been widely used for the treatment of locally advanced and early-stage breast cancer.⁴ A large proportion of patients receiving NACT undergo mastectomy as the surgical treatment either because breast-conserving surgery is not feasible or according to patient preference.⁵ Therefore, IBR with NSM/SSM techniques may be an important alternative surgical option after NACT. Nonetheless, the oncologic efficacy and safety of this treatment sequencing strategy is unclear.² NACT is associated with more frequent local recurrence (LR) than adjuvant chemotherapy after breast-conserving treatment.⁴ Furthermore, regarding the surgical techniques of NSM/SSM compared with those of conventional mastectomy (CM), there is the theoretical concern of less adequate resection of the breast tissue owing to limited incision, extended flaps, and potentially compromised exposure during the operation, which may result in increased risks of subsequent LR. To our knowledge, no randomized clinical trials have investigated the oncologic safety of IBR with NSM/SSM, and data on the oncologic outcomes of this treatment approach in the setting of NACT are insufficient. Previous studies have reported on the feasibility of IBR following NACT; however, most of these studies were small series with relatively short follow-up or lacked matched control groups.^6-14 This propensity score–matched case-control study compared the long-term oncologic outcomes of a large series of patients who underwent NSM/SSM and IBR with those of patients who underwent CM alone in the NACT setting.

A total of 1395 patients who underwent NACT followed by mastectomy with or without breast reconstruction between January 1, 2010, and November 30, 2016, at Asan Medical Center, Seoul, Korea, were identified from a prospectively maintained database. Of those patients, we excluded those with inflammatory breast cancer or pathologic T4 category disease (n = 90), synchronous distant metastasis (DM) (n = 37), or recurrent disease (n = 2). Accordingly, 1266 patients were included in the analysis. The study cohort comprised 526 patients who underwent IBR with NSM/SSM and 740 who underwent CM alone (Figure 1). The study was approved by the Asan Medical Center Institutional Review Board. Because of the retrospective design of the study, the requirement for informed consent was waived; data were deidentified. Data analysis was performed from July 1, 2019, to January 24, 2020. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for case-control studies.

All patients included in this study received NACT after the diagnosis of breast cancer. The NACT regimens were selected at the discretion of the treating oncologist. The indications for NSM or SSM were any category, any tumor size, and any tumor-to-nipple distance with indications for mastectomy. Patients with a clinically normal nipple and without skin involvement were counseled about NSM after NACT. In NSM cases, retroareolar frozen-section biopsy specimens were collected and examined intraoperatively. The nipple-areola complex was preserved if the shape, color, and palpated features of the nipple were normal and if the nipple margin was confirmed to be tumor-free on frozen-section biopsy. In cases in which the retroareolar tissue was positive for cancer in the frozen-section or permanent biopsy, the nipple with or without the areola was removed and the surgical procedure was converted to SSM. In our entire cohort, no treatment was converted to CM due to failure of the NSM/SSM procedure. The decision to undergo adjuvant radiotherapy was determined by the treating radiation oncologists after evaluation of pre- and post-NACT disease stages, tumor response to NACT, and other tumor biomarker status of the patients. Most patients who required postmastectomy radiotherapy after evaluation underwent simultaneous irradiation of the chest wall and supraclavicular region. Postoperatively, patients were regularly followed up every 3 to 6 months for the first 5 years and annually thereafter. In patients suspected of having an LR or regional recurrence (RR), punch needle or excisional biopsy was performed for pathologic confirmation. Patients who did not present for examination were contacted via telephone to confirm their survival status.

To reduce the effects of selection bias on the type of surgery and confounding factors on the comparisons of oncologic outcomes between the IBR and CM-alone groups, we used propensity score matching to create well-balanced groups. We included the following baseline covariates for matching: age at diagnosis, clinical T category, clinical N category, molecular subtype, and response to NACT. Propensity scores were calculated for each of the 1266 patients using a multivariate logistic regression model, and patients were matched 1:1 into IBR with the NSM/SSM and CM-alone groups using caliper restriction to the nearest neighbor without replacement. After propensity score matching, each group included 323 patients (Figure 1). Tumor stage and NACT response were determined according to the 8th edition of the American Joint Committee on Cancer AJCC Cancer Staging Manual.¹⁵

The end points of interest were local recurrence–free survival (LRFS), disease-free survival (DFS), distant metastasis–free survival (DMFS), and overall survival (OS). Local recurrence was defined as biopsy-proven cancer recurrence in the ipsilateral chest wall, breast skin, or nipple-areola complex. Regional recurrence was defined as a recurrence in the ipsilateral axillary, internal mammary, or supraclavicular lymph nodes. Any other site of recurrence was considered to be DM. Patients with initial RR or DM were excluded from the LR group. In cases of concurrent LR, RR, or DM, each recurrence was counted simultaneously as an event. A pathologic complete response was defined as no evidence of invasive cancer in the breast and axillary lymph node. Follow-up was calculated from the date of diagnosis. The 5-year LRFS, DFS, DMFS, and OS rates were calculated using the Kaplan-Meier method and compared using log-rank tests. The hazard ratios (HRs) and 95% CIs were estimated using the Cox proportional hazards regression model. The χ² or Fisher exact test was used for intergroup comparisons. All statistical analyses were performed using SPSS, version 24.0 (IBM Corp). Two-tailed P values <.05 were considered statistically significant.

A total of 1266 patients met the selection criteria for analysis. Baseline characteristics according to the surgical procedures for the entire cohort before matching are presented in Table 1. Among the unmatched groups, patients in the IBR group were more likely to be younger than 50 years, have cancer at earlier clinical and pathologic categories, and have hormone-positive disease than those in the CM-alone group. However, the response to NACT did not differ significantly between the groups.

After propensity score matching, 646 patients were 1:1 matched, comprising well-balanced IBR (n = 323) and CM-alone (n = 323) groups. Median age of the women in the IBR group was 42 (range, 23-72) years; median age of those in the CM-alone group was 46 (range, 30-75) years. Other patient, tumor, and treatment characteristics for the matched groups are reported in Table 2. There were no significant differences between the matched groups in the selected matching variables (ie, age, clinical T and N categories, molecular subtype, and response to NACT) and pathologic variables (ypT, ypN, histologic type, histologic grade, pathologic multifocality/multicentricity, or post-NACT Ki67 status). The pathologic complete response rates were 13.6% in the breast, 34.4% in the axilla, and 12.1% in both breast and axilla for the IBR group and 12.4% in the breast, 28.3% in the axilla, and 10.8% for both breast and axilla for the CM-alone group. Regarding treatment outcomes, the IBR group comprised 187 patients who underwent NSM and 136 individuals who underwent SSM. There were no significant differences between the matched groups in NACT regimens, adjuvant radiotherapy, adjuvant hormonal therapy, or adjuvant chemotherapy. However, patients in the IBR group were less likely to undergo axillary lymph node dissection than those in the CM-alone group (46.1% vs 57.6%; P = .004).

After matching, the median follow-up periods were 67 (range, 17-125) months for the IBR group and 68 (range, 17-126) months for CM-alone group. We observed no significant differences between the IBR and CM-alone groups in LR (3.7% vs 3.4%; P = .83), RR (7.1% vs 5.3%; P = .33), or DM (17.3% vs 18.6%; P = .68) rates. Of the 187 patients who underwent NSM in the IBR group, 2.1% of patients (n = 4) experienced recurrence at the nipple-areola complex, with rates of 4.8% (n = 9) for LR, 7.0% (n = 13) for RR, and 16% (n = 30) for DM. Of the 136 patients who underwent SSM in the IBR group, the rates of recurrence were 2.2% (n = 3) for LR, 7.4% (n = 10) for RR, and 19.1% (n = 26) for DM. Regarding survival outcomes, we observed no significant differences between the IBR vs CM-alone groups in 5-year LRFS (95.6% vs 96.7%; HR, 1.124; 95% CI, 0.495-2.549; P = .78), DFS (76.5% vs 79.9%; HR, 1.089; 95% CI, 0.790-1.500; P = .60), DMFS (82.5% vs 82.5%; HR, 0.941; 95% CI, 0.654-1.355; P = .74), or OS (92.0% vs 89.3%; HR, 0.847; 95% CI, 0.530-1.353; P = .49) rates (Figure 2).

We further compared the survival outcomes of the subset of patients without response to NACT in the matched cohort (n = 261 [IBR, 126; CM alone, 135]). There were no significant differences in terms of age at diagnosis, clinical and pathologic stages, and subtype between the IBR and CM-alone groups in this subset of the population. Moreover, we observed no significant differences between the groups in the 5-year LRFS (96.4% vs 95.1%; HR, 0.641; 95% CI, 0.187-2.197; P = .48), DFS (74.5% vs 74.8%; HR, 0.949; 95% CI, 0.595-1.514; P = .83), DMFS (79.5% vs 78.7%; HR, 0.924; 95% CI, 0.546-1.562; P = .77), or OS (91.4% vs 90.6%; HR, 1.032; 95% CI, 0.510-2.089; P = .93) rates (Figure 3).

In the present study, comparisons of the 5-year LRFS, DFS, DMFS, and OS rates between the propensity score–matched groups revealed comparable outcomes between IBR with NSM/SSM and CM alone in patients who received NACT. To our knowledge, this is the largest study comparing the oncologic outcomes of IBR and CM alone in the NACT setting.

Use of NSM or SSM combined with IBR is increasingly adopted as a therapeutic strategy for patients with breast cancer receiving NACT.^1,2 However, evidence for the oncologic safety of this treatment approach is insufficient.² Although randomized clinical trials comparing the safety of IBR and CM alone would yield the most powerful conclusions, such trials have not and are unlikely to be performed since both patient and surgeon preference may affect decisions regarding breast reconstruction, making blind randomization challenging.¹⁶ Previous observational studies have demonstrated comparable oncologic outcomes of IBR vs CM alone in the treatment of selected patients.^17-20 However, such results need to be weighed with care because there are generally substantial differences in terms of baseline characteristics and oncologic severity between patients who receive IBR and those who receive CM alone.^2,21 In the present study, before matching, patients in the IBR group tended to be younger, at earlier disease stages, and have favorable molecular subtypes (ie, ERBB2- [formerly HER2] or hormone receptor+/ERBB2+) than those in the CM-alone group. To reduce the selection bias on the type of surgical procedure and oncologic outcomes, we adjusted for baseline characteristics between the 2 groups by propensity score matching, resulting in 2 groups with similar matching and pathologic variables and adjuvant treatment characteristics.

Several studies have reported the feasibility of IBR with NSM/SSM in patients who received NACT.^6-14 However, most of these studies involved insufficient numbers of cases per center and were limited by relatively short follow-up durations. Ryu et al¹¹ compared 31 patients who underwent IBR following NSM/SSM with 85 patients who underwent CM alone after NACT in a matched case-control study. The investigators reported no significant differences in short-term OS, DFS, DMFS, and LRFS rates between the 2 groups. The median follow-up durations were 29.2 months for the IBR group and 38.8 months for the CM-alone group. Agresti et al¹² compared 61 patients who underwent IBR following NSM with 61 patients who underwent CM alone after NACT and reported no significant difference in the 4-year LRFS rate between the 2 matched groups (89% for IBR with NSM vs 93% for CM alone). Wengler et al¹³ retrospectively reviewed 280 patients who underwent IBR with NSM/SSM after receiving NACT and reported a locoregional recurrence rate of 3.2% and a DM rate of 13.2% in a median follow-up of 45 months. Their data support expanding the indications for IBR with NSM/SSM in the NACT setting with excellent local control; however, they did not perform a matched comparison with patients who underwent CM alone.

In the present study with a long-term follow-up period, we performed a propensity score–matched analysis of 646 patients treated with IBR with NSM/SSM or CM alone after NACT and found no significant differences in the recurrence and survival outcomes between the groups, even for the subset of patients without response to NACT, supporting the oncologic safety of IBR with NSM/SSM in the current setting. Our results suggest that the response to NACT should not be considered a contraindication to IBR and, even in selected patients without response to NACT, IBR can be a feasible option and should be discussed before surgical treatment. In the present study, patients in both groups had low rates of LR (3.7% for IBR and 3.4% for CM alone) but high rates of DM (17.3% for IBR and 18.6% for CM alone). This finding is consistent with those of previous studies on NACT, suggesting that patients in the current setting are more at risk of DM than LR.^4,10 These results also reflect that tumor biology and systemic treatment, rather than the surgical approach, may be the determinants of patient prognosis.

Although our data showed no statistically significant differences in clinical and pathologic N category between the matched groups, patients in the CM-alone group were more likely to undergo axillary lymph node dissection than those in the IBR group (57.6% vs 46.1%; P = .004). In our cohort, the pathologic complete response rate in the axilla was lower in the CM-alone group than that in the IBR group (28.3% vs 34.4%). This difference may reflect the overall poorer nodal response to NACT in the CM-alone group and, thus, was associated with the higher rate of axillary lymph node dissection in this group. Moreover, other factors may affect the choice of axillary lymph node dissection in the current setting, such as intraoperative assessment of the axillary lymph nodes, complexity of the therapeutic changes in nodal status, and discretion of the surgeons according to individual experience, which were not considered as matching covariates in this study. Depending on the presence or absence of pathologic complete response in the axilla, differences in DM or OS may occur. However, in this study, there was no significant difference in prognosis between the 2 groups during the follow-up period. Further studies should be conducted in more patients with axillary metastasis.

The major limitation of the present study was its retrospective design and the associated selection bias even after close matching of the key clinical factors. Furthermore, hidden confounders may not have been accounted for; thus, our results should be interpreted with caution. Randomized clinical trials on the oncologic efficacy of IBR with NSM/SSM are required but are difficult to perform owing to ethical concerns. Accordingly, in the absence of randomized clinical trials and limited long-term follow-up data, our propensity score–matched analysis of a robust series with durable follow-up provides the evidence of the oncologic safety of IBR with NSM/SSM after NACT.

The findings of this large-series, matched case-control study suggest comparable long-term oncologic outcomes between IBR with NSM/SSM and CM alone after NACT for breast cancer. The results suggest the feasibility of IBR with NSM/SSM in the NACT setting.

Accepted for Publication: June 10, 2020.

Corresponding Author: BeomSeok Ko, MD, PhD, Department of Surgery, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Republic of Korea (spdoctorko@gmail.com).

Published Online: October 14, 2020. doi:10.1001/jamasurg.2020.4132

Author Contributions: Drs Wu and Ko had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Wu, J.-S.-B. Kim, S.-B. Lee, Ahn, Ko.

Acquisition, analysis, or interpretation of data: Wu, H.-J. Kim, J.-W. Lee, Chung, J.-S. Kim, Son, Eom, S.-B. Kim, Jung, Gong, H.-H. Hee Kim, Ko.

Drafting of the manuscript: Wu, H.-H. Hee Kim.

Critical revision of the manuscript for important intellectual content: Wu, H.-J. Kim, J.-W. Lee, Chung, J.-S. Kim, S.-B. Lee, Son, Eom, S.-B. Kim, Jung, Gong, Ahn, Ko.

Statistical analysis: Wu.

Administrative, technical, or material support: Wu, H.-J. Kim, Son, Eom, Jung, H.-H. Kim, Ko.

Supervision: Wu, Chung, J.-S. Kim, S.-B. Lee, Gong, Ahn, Ko.

Conflict of Interest Disclosures: Dr Jung reported receiving personal fees from AstraZeneca Korea, Roche Korea, Celgene Korea, Novartis Korea, and Takeda Pharmaceuticals outside the submitted work. Dr Gong reported receiving personal fees from Roche outside the submitted work. No other disclosures were reported.