Total Skin Sparing (Nipple Sparing) Mastectomy: What is the Evidence?

The surgical treatment of breast cancer has evolved from radical mastectomy with routine removal of the nipple-areolar complex (NAC) to breast conservative therapy with preservation of the breast and NAC. Each step along this evolutionary process was met with criticism, skepticism, controversy, anger, emotion, and often bitter and impassioned debate. Today we find ourselves at yet another therapeutic decision point: the management of the skin of the nipple-areolar complex in mastectomy. Enhanced understanding of the pathogenesis of breast cancer coupled with rising interest in improved cosmesis has led to the investigation of the skin-sparing and nipple-sparing mastectomy as potential modifications to conventional mastectomy.

Physicians endeavoring to treat cancer have struggled to strike a balance between the efficacy of a particular treatment and the quality of life after that treatment. Nowhere has this struggle had a more varied and tortuous course, or been more controversial, than in the treatment of breast cancer.

Nothing has been as constant in the field of breast cancer treatment as change . The past two decades have witnessed dramatic conceptual changes in strategic and practical approaches to virtually all forms of breast cancer. These sea changes have been driven by improvements in technology, improvements in public awareness resulting in earlier stage at diagnosis, and improvements in our understanding of fundamental biologic mechanisms of the disease, most importantly, better understanding of disease subsets. The surgical treatment of breast cancer has evolved from radical mastectomy with routine removal of the nipple-areolar complex (NAC) to breast conservative therapy with preservation of the breast and NAC. Each step along this evolutionary process was met with criticism, skepticism, controversy, anger, emotion, and often bitter and impassioned debate. Today we find ourselves at yet another therapeutic decision point: the management of the skin of the nipple-areolar complex in mastectomy. Enhanced understanding of the pathogenesis of breast cancer coupled with rising interest in improved cosmesis has led to the investigation of the skin-sparing (SSM) and nipple-sparing mastectomy (NSM) as potential modifications to conventional mastectomy. There has been much debate regarding the oncologic safety of these procedures. Purists argue that leaving the skin of the NAC behind may increase the chance for local recurrence and is therefore contraindicated. A growing number of clinicians argue that the nipple is a rare site for end-organ carcinogenesis and think that leaving the skin of the nipple in-situ (after removing lactiferous ducts) adds little or no risk. What is the clinical evidence that suggests that a total skin-sparing approach is oncologically safe?

Literature review

There have been a growing number of series of total skin sparing mastectomy (TSSM) in recent years. Although all reported series lack the statistical power to reach definitive conclusions, taken in the aggregate they form the basis for continued clinical study and rational clinical practice. This information is important insofar as a randomized clinical trial will never be achievable. There is no evidence to date that suggests an increased oncologic risk associated with sparing the skin of the NAC.

In reviewing the efficacy of bilateral prophylactic mastectomy, Hartmann and colleagues conducted a retrospective study of 639 women at moderate-high risk for breast cancer undergoing prophylactic mastectomy. A total of 575 (90%) of these women underwent bilateral subcutaneous mastectomy, sparing residual breast tissue beneath an intact NAC, whereas the remaining 64 underwent bilateral total mastectomy. After 14 years of median follow-up, seven subjects (1.2%) versus no subjects (0%) developed breast cancer after prophylactic subcutaneous mastectomy and total mastectomy, respectively ( P = .32). Six tumors were found in the chest wall, whereas one subject was found to have bone metastases with no evidence of local disease. No subject developed breast cancer in the residual NAC. Although not their primary goal, this group’s data demonstrates the exceedingly low rate of breast cancer in the NAC, even in patients with a moderate-high risk.

Margulies and colleagues described their experience with 31 subjects undergoing 50 mastectomies, including four cases with centrally located tumors. Six attempted NSMs had to have the NAC excised because of tumor involvement on touch-preparation cytology (n = 4) or nipple necrosis (n = 2). Although this group reported a short follow-up time, no local or distant recurrences were observed in this series.

In 2006, Sacchini and colleagues reported a multi-institutional review of 123 subjects undergoing 192 TSSM, including 20 mastectomies for ductal carcinoma in situ (DCIS) and 44 for invasive cancer. All tumors were peripherally located and none showed preoperative evidence of disease less than 1cm from the areolar margins. After a median follow-up of 98.4 weeks, four subjects presented with a local recurrence, two after TSSM for invasive cancer, and two after prophylactic mastectomy. All recurrences were distant from the NAC, with three in the upper-outer quadrant and one in the axillary tail. A fifth subject succumbed to distant metastatic disease without evidence of loco-regional recurrence.

Crowe and colleagues selected 110 subjects to undergo NSM in 149 breasts as a procedure for treatment (73%) or prophylaxis (27%) of breast cancer. Of the 149 procedures, a total of 9 (6%) were eventually converted to total mastectomy secondary to tumor involvement of the NAC as demonstrated by intraoperative frozen section. Four subjects treated for invasive breast cancer developed recurrences, two local recurrences (LR) and two distant recurrences (DR), with no disease seen in the NAC after a median of 164 weeks of follow-up. Only three (2.7%) of the subjects with NSM experienced significant complications, including one subject who required subsequent removal of the NAC because of infection.

One of the earliest studies was conducted by the Karolinska group in Stockholm, Sweden in the late 1980s. Their prospective series included 216 subjects who underwent 184 NSM and 32 SSM with NAC. They enjoyed a robust follow-up of 676 weeks to evaluate recurrence rates. In general these were subjects with large tumors and extensive axillary metastases. The high LR and DR rates (24% and 20% respectively) seen in this study may stem from the fact that only 47 (22%) of enrolled subjects received radiation therapy (RT); in fact, the LR rate for these 47 subjects was only 8.5%. This group left a 2 cm plate of breast tissue of 5 mm thickness behind in an attempt to preserve NAC blood supply. Despite this technical point, no LRs were seen in the preserved NAC tissue.

Garwood and colleagues have prospectively reported two cohorts of 115 subjects undergoing 170 NSMs for prophylaxis and treatment of invasive breast cancer. This study used the results of early NSMs, grouped into cohort one, to expand subject selection and improve surgical technique. The second cohort of subjects included more subjects receiving adjuvant chemotherapy or radiotherapy, which was as expected considering 37 of 48 (54%) of the subjects had stage 2 or 3 disease. Despite this, necrotic complication rates dropped from 30% to 13% from cohort one to cohort two, respectively. Additionally, only 5% of cohort two subjects experienced nipple loss, which was significantly less than the 15% nipple loss seen in cohort one. The investigators contribute these differences to surgical-technique lessons learned from the first cohort of subjects. The second group had significantly fewer incisions crossing more than 30% of the NAC as these subjects seemed to have a lower rate of nipple necrosis. Also, this group identified reconstruction with immediate implants as a risk factor for skin flap necrosis and implant loss, presumably because of increased skin tension and ischemia. Significantly more subjects had tissue expander reconstructions in the second cohort when compared with cohort one, probably contributing to the decrease in complications. Fifty-two weeks of median follow-up revealed only one LR (0.6%) discovered in the axillary tail following NSM. Two subjects developed metastatic disease without evidence of local recurrence. No NAC recurrences were identified by this group.

Petit and colleagues have introduced a novel method of delivering intraoperative, single fraction electron boost radiotherapy, dubbed ELIOT. Patients are treated while under anesthesia at the time of TSSM. The surgeons reportedly preserve 5 to 10mm of breast tissue beneath the NAC to preserved vascular supply. A 2009 review of 1001 subjects has provided the largest single-institution experience of TSSM for invasive cancer (82%) and DCIS (18%). After a median follow-up of 80 weeks, only 14 (1.4%) local recurrences were documented, and 36 subjects developed distant metastasis, with 4 succumbing to their disease. No local recurrence was identified at the NAC. This finding is surprising considering final pathology showed cancer cells in 79 cases in which the NAC was ultimately preserved. Of the 1001 subjects, 800 underwent ELIOT and 201 received delayed radiotherapy, with no significant difference seen in the complication rates between these two groups. Total or partial NAC necrosis was seen in 30 (3%) and 55 (5.5%) of the subjects respectively, with 50 (5%) requiring subsequent NAC removal. Although the short follow-up time in this series precludes any definitive conclusions, the early results from this group are quite promising and extended follow-up will provide valuable insight into the oncologic safety of NSM with radiotherapy to the breast.

Gerber and colleagues reported a prospective series of 238 subjects with indications for mastectomy. A total of 112 NSM were attempted, however, 51 (45.5%) were converted to SSM because of NAC involvement by intraoperative frozen section (FS), resulting in 61 (54.5%) cases of NSM. The remaining 134 subjects underwent standard modified radical mastectomy (MRM). Despite no significant differences in the indications for surgery, overall disease stage, axillary node involvement, or adjuvant treatment in these three groups, LR rates were virtually indistinguishable at 11.7% (NSM), 10.4% (SSM), and 11.5% (MRM) at mean follow-up of 404 weeks. This series includes a rare subject with local recurrence in preserved NAC tissue that underwent nipple resection with areola preservation, and remained disease free at 364 weeks of follow-up. Aesthetic results as reported by subjects and surgeons were comparable among TSSM and SSM subjects.

Paepke and colleagues recently reported their prospective experience with 109 attempted NSM in 96 subjects, including 33 (30%) centrally located tumors and 36 of 78 invasive cancers of stage 2 or greater. All duct tissue was removed by nipple inversion and sharp dissection, with FS performed to evaluate for occult disease. Thirteen cases were found to have NAC involvement, and 12 of these subjects underwent SSM as the definitive procedure. One subject refused removal of the NAC and underwent NSM despite the positive FS; she remained disease free after 160 weeks of follow-up. Overall, two (2%) local recurrences were observed, neither within the NAC. Two (2%) other subjects developed distant recurrence. Only one (1%) subject had NAC necrosis severe enough to warrant surgical intervention. Based on the higher-risk subject profile presented in their series, the investigators offer clinical suspicion of skin or nipple involvement and inflammatory breast cancer as the remaining absolute contraindications for TSSM.

A thorough review of the literature provides many examples of high-volume surgeons performing NSM for prophylaxis and for the treatment of breast cancer in carefully selected patients ( Table 1 ). The oncologic and cosmetic outcomes of these series often recapitulate those seen for SSM and MRM while allowing for a more natural reconstruction. Although many of the aforementioned studies require further subject accrual and extended follow-up, these promising early results will no doubt increase the interest in this controversial arena.

Table 1

Summary of NSM review

Author	Year	NSM (n)	SSM (n)	Total (n)	LR (%)	Complications (%)
Paepke et al	2009	96	13	109	2.0	0.9
Gerber et al	2009	60	48	108	11.7 ^a, 10.4 ^b	NR
Petit et al	2009	1001	0	1001	1.4	13.3
Garwood et al	2009	170	0	170	0.6	5
Benediktsson & Perbeck	2008	184	32	216	24.1	NR
Crowe et al	2008	140	0	140	1.4	2.7
Sacchini et al	2006	192	0	192	0.0	8
Margulies et al	2005	44	6	50	7.9	NR