Risk reduction mastectomy

The concept of NSM was first popularized in the 1960s and 1970s as subcutaneous mastectomy. That procedure for risk reduction quickly fell out of favor for a number of good reasons. Most significantly, there were no scientifically based selection criteria and therefore no possibility of ever demonstrating efficacy. Beyond that, the reconstructive techniques of the day were relatively crude, the results were inconsistent, the best outcomes were not impressive, the complication rate was high, and a significant amount of breast tissue was intentionally left behind.

All this changed with the seminal report by Hartmann et al, published in the New England Journal of Medicine on January 14, 1999. Over the preceding 20 years or longer, the Mayo Clinic in Rochester, Minnesota, had been a center of sorts for prophylactic mastectomy where the techniques were largely of the historic subcutaneous mastectomy type, but where the medical records were excellent and data could be extracted and reassessed using computer models and other more modern techniques. The data from that series of 639 women demonstrated that prophylactic mastectomy did indeed have a protective benefit, reducing the risk of breast cancer in both high-risk and moderate-risk groups by 81–94%. Ninety percent of the mastectomies in that series were nipple-sparing of the subcutaneous variety. Breast cancer developed in seven women after prophylactic mastectomy. Six cancers were confined to the chest wall at diagnosis and were specifically not in the area of the NAC. One patient in the high-risk group presented with bone marrow metastases from adenocarcinoma with no evidence of breast disease. There was no statistically significant difference in the cancer-preventing benefit whether the nipple was removed or retained.

In the context of risk reduction, other series report similar experience. McDonnell et al, also from the Mayo Clinic, reported in 2001 in the Journal of Clinical Oncology on a series of 745 women with a first breast cancer and a strong family history of breast cancer who underwent unilateral prophylactic mastectomy between 1960 and 1993, resulting in a projected 94–96% risk reduction. All told, there were eight breast cancers versus 156 predicted. Forty-one percent of the mastectomies spared the nipple, while 59% did not. The eight cancers that developed were evenly split, four and four, between the nipple-sparing and nipple removal groups. None of the eight cancers that did occur developed near the nipple.

In 2004, Crowe reported on 17 nipple-sparing prophylactic mastectomies performed from 2001 to 2003. They recommend a lateral incision for improved nipple–areola survivability. There were no occult cancers seen on frozen or permanent examination beneath the nipple and no cancers have developed since.

In 2006, Sacchini published in the Journal of the American College of Surgeons a larger multi-center experience including the Memorial Sloan-Kettering Cancer Center in New York and major cancer centers in Sao Paulo (Brazil), Milan, and Padua (Italy). Altogether there were 55 patients who underwent nipple-sparing prophylactic mastectomies. There were no recurrent or new cancers in the nipple with mean follow-up of 24 months. Two cancers did develop after prophylactic mastectomy: one in the axillary tail 24 months postoperatively and one in the upper outer quadrant at 62 months. The majority of their nipple-sparing procedures were performed via a periareolar incision, which included coring out the nipple. Twenty-two out of 192 nipples suffered some degree of necrosis, nine of which resulted in loss of more than one-third of the NAC.

TR Rebbeck et al published a combined multi-center experience from the Prose study group of 483 BRCA 1 and 2 positive women, 105 of whom underwent bilateral prophylactic mastectomy, including 29 nipple-sparing. Two of those 105 developed breast cancer, in contrast to 184 cancers among 378 patients in the control group, yielding a 90% or more risk reduction. The two cancers that developed included one in the axilla and one in the breast. There was no statistically significant difference in the occurrence of cancer between the nipple-sparing and non-nipple-sparing groups.

Nipple-sparing mastectomy in the treatment of breast cancer

If surgeons have been reluctant to embrace NSM in the context of risk reduction, then they have been even more reluctant to do so in the face of established breast cancer. Nevertheless, the gradual acceptance of skin-sparing mastectomy has encouraged some to look more critically at the concept of NSM as a treatment for breast cancer in order to better define inclusion criteria, exclusion criteria, preoperative strategies, operative strategies, surgical technique, and overall value.

There were many inconsistent reports from the 1970s and 1980s regarding the likelihood of nipple involvement with cancer in the face of known ipsilateral breast cancer. Reports of nipple involvement with cancer varied widely from 0 to 58%. Much of that data comes from reports in the 1970s and 1980s based upon examination of the mastectomy specimens from an era of later diagnosis and more advanced disease. Further, the methodology for tissue examination of those studies and the criteria for describing ‘involvement’ were not uniform.

With the increased interest in skin-sparing mastectomy and even NSM, more relevant studies have recently been done. Christine Laronga et al from the MD Anderson Cancer Center in Houston reported in 1999 on 326 patients whose breasts were examined after skin-sparing mastectomy. They described how earlier studies correlated a greater risk of occult NAC involvement with the primary tumor being close to the NAC, larger than 2 cm, poorly differentiated, and associated with positive axillary lymph nodes. They found 16 (5.6%) instances of occult tumor involvement. They believe four of these would have been identified on frozen section if it had been done. The only statistically significant predictors of nipple involvement were location (subareolar or multicentric) and axillary nodal status. The authors speculate that their finding of a relatively low incidence of occult NAC involvement might reflect their selective preoperative criteria for NSM in the first place. They conclude that it would be appropriate to offer NAC preservation in axillary node negative patients with tumors located on the periphery of the breast. They estimate that in that group the probability of missing occult tumor in the NAC would be less than 2%.

Bernd Gerber and colleagues from Rostock, Germany, reported in the Annals of Surgery in 2003 their experience with skin-sparing mastectomy, including conservation of the NAC. At 59 months, there were six (5.4%) recurrences in 112 nipple-sparing mastectomies: two on the chest wall, two in the upper breast, one in the fold, and one non-invasive in the nipple. There were 11 (8.2%) recurrences among 134 women who had the nipple removed. They found NAC tumor involvement depended predominantly on the distance of the tumor to the NAC. They believe that their data justify preservation of the NAC in patients with preoperatively evaluated tumors at least 2 cm from the nipple, no extensive intraductal component (extensive defined as greater than 25%), and intraoperatively demonstrated clear margins.

Crowe at the Cleveland Clinic attempted NSM in 54 women from September 2001 until June 2003. Frozen section findings of disease in the nipple resulted in aborting the procedure in six patients. Of the 48 women who underwent NSM, three had partial nipple loss. Candidates for NSM were excluded if the tumor was 3.5 cm in diameter or greater, if the tumor was central, if the patient had undergone neoadjuvant chemotherapy, and if there was inflammatory breast cancer or Paget’s disease. Overall, they recommended lateral incisions whenever possible. Of note even with their exclusion criteria, of 37 patients who underwent NSM for cancer, six (16%) had neoplastic involvement based upon frozen section.

In Sacchini’s report of a multi-center, multinational experience with NSM in 68 patients with breast cancer, they excluded women with very large breasts, very ptotic breasts, and tumors within 1 cm of the areola. Fourteen patients of the original 82 were excluded and not included among the remaining 68 because of tumor involvement found on permanent pathological evaluations. They did not employ frozen sections in their series, and thus finding occult nipple involvement with tumor was only possible on permanent pathologic review.

More recently, Laura Esserman and colleagues from San Francisco described their experience with total skin-sparing mastectomy (nipple-sparing mastectomy) in 64 breasts in 43 women. Twenty-nine breasts were prophylactic and 24 were for invasive cancer or DCIS. They recommended high-resolution magnetic resonance imaging (MRI) with fat suppression to exclude tumor within 2 cm of the nipple. Nevertheless, even with MRI screening two occult cancers were found on frozen section. Further, they do not recommend NSM in cases with large, centrally located tumors, skin involvement by tumor, or MRI evidence of tumor within 2 cm of the nipple.

Of similar interest is the recommendation by Govindarajulu and colleagues from Bristol, UK, to screen patients for possible NSM by performing preoperative ultrasound-guided mammotome assessment. Thirty-three women had 36 procedures, and seven of the 36 had a positive biopsy. The histopathology of the mastectomy specimen correlated 100% with the mammotome biopsy.