Favorable Outcome with Close Margins in Patients Undergoing Nipple/Skin Sparing Mastectomy with Immediate Breast Reconstruction: 5-year Follow-up

Background: Implant-based breast reconstruction after mastectomy has recently been reported to be the preferred type of surgery among breast-specific surgeons and plastic surgeons. Aims: To explore the significant clinicopathological factors associated with long-term outcome related to local recurrences of the nipple among patients who underwent immediate breast reconstruction with tissue expander or implant after mastectomy. Study Design: Retrospective cohort. Methods: From January 2007 to January 2013, 51 breast cancer patients who underwent immediate breast reconstruction with tissue expander or implant were retrospectively analysed. Patients’ demographic data, clinicopathological characteristics, and clinical outcome by disease-free survival and disease-specific survival analyses were determined. Results: The median follow-up was 64 (31-114) months. Of the 57 mastectomies, 41 were skin sparing mastectomy (72%) and 16 were nipple-areola sparing mastectomy (28%). Immediate breast reconstruction surgery included tissue expander (n=46, 81%) or implant (n=11, 19%) placement. The molecular subgroups of 47 invasive cancers were as follows: luminal A (n=23, 49%), luminal B (n=16, 34%), non-luminal HER2 (n=5, 10.6), triple negative breast cancer (n=3, 6.4%). The 5-years disease-specific survival, disease-free survival, and locoregional recurrence-free survival rates were 96.8%, 90%, and 97.6% respectively. Patients with luminal A cancer were found to have an improved 5-year disease-free survival time than other (luminal A; 100% vs. non-luminal A; 78%; p=0.028). Of the 14 nipple-areola sparing mastectomy, 13 had a close median tumour distance to nipple-areola complex (<20 mm) with a 5-year locoregional recurrence free survival of 100%. Conclusion: Immediate breast reconstruction with implant or tissue expander can be safely applied in patients undergoing skin sparing mastectomy or nipple-areola sparing mastectomy. Patients with luminal-A type show the most favourable outcome. During the 5-year follow-up period, patients even with close margins (<20 mm) to nipple-areola complex with nipple-areola sparing mastectomy have excellent locoregional and overall survival when treated by contemporary multidisciplinary oncological management.

Mastectomy is performed in approximately 40-50% of women with breast cancer (1). Implant-based breast reconstruction after mastectomy has recently been reported to be the preferred type of surgery among plastic surgeons comprising approximately 79% of all breast reconstruction procedures in the USA (2). Skin-sparing mastectomy has resulted in higher levels of patient satisfaction and quality of life (3). Nipple-areola-sparing mastectomy (NSM), in which the nipple-areola complex (NAC) is preserved, coupled with reconstruction, can result in remarkable aesthetic results, with the improved techniques related to breast implants, acellular tissue matrices, and fat grafting (4). The most important advantages of skin sparing mastectomy (SSM) or nipple-areola sparing mastectomy (NSM) are the preservation of the breast contour, preservation of the submammary fold, and avoidance of skin differences. Many studies comparing local and distant recurrence in patients undergoing SSM with those undergoing conventional mastectomy have reported no significant differences, thus supporting the oncological safety of this approach (5,6). Although there are some controversies about the risk of local relapse, various clinical studies have shown that the NSM also is a safe procedure for selected cases (7)(8)(9)(10). These results have further been maintained in long-term outcome studies, where NSM shows oncologic outcomes similar to SSM (10,11). The locoregional recurrence (LRR) rate after NSM ranges from 0 to 11.7% (8,(11)(12)(13)(14)(15). The recurrence in the NAC is a rare event with incidence of 0-5%. Besides, considering that a LRR is more likely to occur within the first two years after primary surgery, the length of follow-up is also important (16). Our aim in this study has been to determine the significant clinicopathological factors associated with long-term outcome among our institutional patient cohort who underwent immediate breast reconstruction (IBR) with tissue expander (TE) or implant after mastectomy treated with contemporary oncological management.

Surgical technique
Sentinel lymph node biopsy was performed either with blue dye (patent blue) and/or radiocolloid injection by using gamma probe at the discretion of the surgeon. After the induction of general anaesthesia, blue dye was applied to the deep subareolar tissue followed by breast massage for 5 minutes. Axillary lymph node dissection was performed in patients with a positive finding in intraoperative pathological assessment of the sentinel lymph nodes. Surgery was decided as SSM for patients whose NAC is identified as involved by tumour either with radiologic assessment or intraoperative pathological assessment. Otherwise NSM is selected. The mammary incision was carried out by elevating the skin flaps in the same planes. Subdermal tissue with its vascular plexus was separated from breast tissue. Depending on the patient, a 2-5 mm thickness of posterior areolar tissue was left, aiming to assure the viability of the nipple areola complex. At this time, another intraoperative pathological assessment was performed from the posterior part of the nipple for cases with NSM. If the result was positive, the nipple-areola complex was removed. After circumferential elevation of the skin flaps, shaving off the pectoral fascia was performed, and the specimen was delivered out of the skin pocket. Following SSM or NSM, the pectoralis major muscle was elevated starting from its lateral border and both sternal and inferior attachments below the inframammary fold were released with meticulous dissection that allowed the surgeon to stay strictly within the submuscular plane. In addition, the serratus fascia was also elevated laterally to accommodate the lateral border of the implant or TE. After copious irrigation of the implant pocket and insertion of two (submuscular or subdermal) drains, the implant or TE was placed submuscularly. We prefer to use TE for the first-stage reconstruction if the patient was considered to receive postoperative radiotherapy. Otherwise, we prefer to use directly implant. The lateral border of the pectoralis and serratus muscle was sutured to allow for the complete coverage of the device. If the TE was used, it was inflated perioperatively to approximately half to 2/3 of its final volume, depending on the perfusion of the mastectomy flaps. Antibiotic prophylaxis was started with 4 g ampicillin/sulbactam daily and was continued until the drains were removed if the drainage was less than 30 cc daily. TE inflation was started on the third postoperative week and an average of 3 sessions was performed to reach the desired TE size before the start of radiotherapy. Radiotherapy was planned with TE in the fully inflated state. Exchange to implant with additional fat grafting to the breast pocket was performed between 2 and 8 months following radiotherapy.

Statistical analysis
All statistical analyses were performed using SPSS 17.0 for Windows software (SPSS Inc., Chicago, IL, USA). The Kaplan-Meier test was used to calculate the overall survival (OS), Disease-specific survival (DSS), disease-free survival (DFS), and LRR free survival rates. All p values were twosided, and a p-value less than 0.05 considered as a statistically significant difference. The statistical and/or clinically significant variables were further evaluated by Cox regression analysis.

RESULTS
Patient demographics and treatment modalities are shown in Table 1. The median age was 42 (range, 20-74). Eighteen patients had a family history of breast cancer (35%). Among the 11 patients who were tested for a BRCA mutation due to a strong family history of breast cancer (blood relative with known BRCA mutation, blood relative with two or more primary breast cancer, two or more relatives with breast cancer on the same side of the family, blood relative with ovarian cancer, blood relative with male breast cancer, patients under 40-year old that demonstrate triple negative molecular subtype), only 1 patient was found to have a BRCA1 mutation. Thirty-nine patients were premenopausal (76.5%), 9 were postmenopausal (17.6%) and 3 were perimenopausal (5.9%). Of the 57 mastectomy procedures, 6 were bilateral. Indications for bilateral mastectomy were bilateral breast cancer (n=4), the presence of a BRCA1 mutation (n=1), and the presence of high risk lesions such as atypical ductal hyperplasia in the contralateral breast (n=1). Surgical type of mastectomy was SSM in 41 cases (72%) and NSM in 16 cases (28%), including 2 cases with video endoscopic assistance. Two of them were prophylactic NSM. A tissue expander was used in 46 (81%) mastectomies, whereas an implant was placed in 11 (19%) mastectomies. Tumour characteristics are summarised in Table 1 Table 2). There was no significant difference in DSS analysis; however, the results of the DFS subgroup analysis revealed that patients with a tumour diameter smaller than 20 mm (p=0.045), grade I-II (p=0.026), luminal A molecular subtype (p=0.028, Figure 2), and ESBC (p=0.040, Figure 3) were more likely to have a better outcome (

DISCUSSION
A tendency towards implant-based breast reconstruction continues to rise (17). In a study by Yi et al. (18), a total of 1810 breast cancer patients who underwent SSM (n=799, 44.1%) or conventional mastectomy (n=1011, 55.9%) were compared. The local, regional, systemic recurrence rates and DFS between two groups did not differ significantly, so they concluded that SSM is an acceptable option for patients who are candidates for IBR. Furthermore, in their meta-analysis, Lanitis (Table  2). In a large series of 539 patients (565 breast cancers), Carlson et al. (24) reported LR of 5.5% (n=31) after a median follow-up of 61.6 months. LR rates were 0.6%, 3.0%, 10.4%, 11.1%, and 0% in stage 0, 1, 2, 3, and 4, respectively. In concordance with our findings, high tumour grade was significantly associated with recurrence.
In a multivariate analysis of 501 patient blocks and at a median follow-up of 10 years, Liu et al. (25) showed that luminal A subtype was significantly associated with low ipsilateral breast recurrence. In another study with a median follow-up of 12 years, 2985 patients' tissue blocks were classified according to molecular subtype. Similarly, luminal A tumours were associated with a low risk of local or regional recurrence among the other subtypes in multivariate analysis (26). In a systematic review by receptor phenotype, 12.592 breast cancer patients who underwent either breast conserving surgery (n=7.174) or mastectomy (n=5.418) were identified from 15 studies by Lowery et al. (27). They assessed that luminal tumours exhibit the lowest rates of LRR. In concordance with these previous studies, we also found that Luminal-A type breast cancer undergoing SSM or NSM with IBR showed excellent survival outcome.
Essentially, some studies have considered NSM safe in patients with small, peripherally located tumours, without multicentricity, or for those who undergo risk reduction surgery (9). Although most of the clinical series emphasise that the principal selection criteria for NSM is the tumour to NAC distance, the ideal distance has not yet been clarified (28,29). Even though there is still no consensus about the selection criteria, most of the studies include tumour sizes up to 3 cm with a lack of clinical involvement of the NAC and a tumour to nipple distance greater than 2 cm (29). In a largest series of NSM that includes 981 mastectomies of 633 patients, with a median follow-up time of 29 (14-54) months, the authors highlighted that even in LABC, NSM can be performed (21). In their series, the overall 5-year cumulative incidences of LRR were 3% and distant recurrences were 4.2%. There was no LR in the NAC skin. They conclude that oncological outcomes remain similar to previously published SSM series. On the other hand, longer follow-up is sine qua non for more robust findings, as we know from mastectomy series that most of the recurrences occur within 5 years. In our series, none of the patients with NSM diagnosed with invasive cancer have shown nipple involvement in either intraoperative or final pathological assessment. Interestingly, the majority of patients had a tumour distance to NAC <20 mm. At a median follow-up of 48 months, there was no LR, and only one systemic recurrence was detected. Even though our number of cases is small, our follow-up period is 4 years for NSM group and as we know from mastectomy series that timeframe for the peek hazard of LRR is 2-to 3-years after surgery (30). Thus, we can conclude that 2 cm distance is not a precise criterion to determine the safety distance between tumour and NAC. Larger series and long-term follow-up however are needed to clarify this issue. Considering the survival analyses, long-term complications and implant or TE loss rates, IBR can be safely applied in patients undergoing SSM or NSM. Furthermore, our results demonstrate that breast cancer patients even with close margins (<20 mm) to NAC who underwent NSM with IBR show excellent disease-free and overall survival. Patients with luminal-A type breast cancer undergoing SSM or NSM with IBR also showed excellent survival outcomes when treated by contemporary multidisciplinary oncological management.