Reviewed by: Stephen Z. Sack, MD PhD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: September 25, 2005
Authors: Downes KJ, Glatt BS, Kanchwala SK, Mick R, Fraker DL, Fox KR, Solin LJ, Bucky LP, Czerniecki BJ.
Source: Cancer. 2005 Mar 1;103(5):906-13.
Skin-sparing mastectomy (SSM) has been a commonly performed procedure for greater than 10 years due to its aesthetic and psychological advantages. The aesthetic advantage is partially due to retention of the natural inframammary crease, allowing for a more anatomically accurate reproduction of the patient's preoperative anatomy. Furthermore, the immediate reconstruction is thought to alleviate some of the psychological stress associated with the time period following mastectomy and delayed reconstruction, so-called breast "mourning." In a 1999 paper, Kroll et al. presented 6-year follow-up data on early-stage breast cancer patients (stage T1 or T2) who underwent SSM, demonstrating no significant difference in local control or overall survival when comparing SSM to conventional mastectomy. These data were followed by a prospective study in 2002 by Foster et al. that looked at 25 patients who underwent SSM for more advanced disease. Thus, there are very limited data on the acceptability of SSM in the more advanced-stage breast cancer patients who commonly undergo intensive postoperative treatment with dose-dense chemotherapy and radiation therapy. Although SSM is an acceptable treatment in early-stage breast cancer patients, the questions addressed by this paper are whether SSM in more advanced breast cancer patients results in adequate local control and acceptable cosmesis, and whether patients undergoing SSM can be safely irradiated?
A retrospective review was performed of 38 patients who underwent SSM by one of two surgeons at the Hospital of the University of Pennsylvania from July 1996 through January 2002. These patients were selected from a pool of 119 women who underwent SSM during this time period due to the classification of their disease as "high risk", defined by: 1) greater than or equal to stage IIB disease, 2) primary tumor greater than 4 cm, or 3) greater than 3 positive lymph nodes. A majority of patients (>60%) were stage IIB. Almost 95% of these patients had positive lymph nodes, while greater than 10% had close margins (defined as less than 1 mm in this paper). The average age was 45.9 years with a range of 27 to 61 years.
Of patients undergoing SSM, more than 80% underwent TRAM flap reconstruction. The minimum follow up was 2 years, with a median follow up of 52.9 months. The median time from surgery to adjuvant therapy was 38 days (range 25 through 238 days).
Just over 65% of patients underwent postoperative radiation therapy. These patients were treated using 3- or 4-field techniques to deliver 50 Gy in 200 cGy fractions. A 1-cm bolus was used on an every- other-day schedule. The supraclavicular nodes were dosed to 46 Gy to a depth of 3 cm. The internal mammary nodes were not specifically targeted in these patients.
Chemotherapy before June 1998 consisted of doxorubicin and cyclophosphamide for 4 cycles. After June 1998, patients underwent an additional 4 rounds of chemotherapy using paclitaxel or docetaxel. If the primary disease was estrogen receptor (ER) positive, patients were also treated with tamoxifen for 5 years.
The authors conclude that the 7.9% locoregional recurrence rate seen in this study compares favorably with prior published data. Furthermore, no significant complications with irradiating SSM were reported (which is a dramatic improvement over prior published literature ).
Due to the paucity of published literature on this topic, this retrospective review provides valuable information into cosmetic and oncologic outcome after SSM. These data seem to refute the commonly held concern that irradiating reconstructed tissue is associated with significant toxicity. There are, however, some points which require critical review.
Although no comparisons are made to conventional mastectomy with respect to timing of adjuvant therapy, this timing issue may be of some importance to a minority of patients in this study. Many randomized clinical trials require adjuvant therapy to begin within 60 (and sometimes up to 84) days postoperatively. There are no large randomized trials suggesting a detriment to delaying treatment, however a recent British retrospective study  looking at 7800 patients questions the wisdom of delaying radiation by more than 63 days, demonstrating a statistically significantly increased relative risk of death when radiation therapy was delayed for more than 140 days. These data compliment a prior Canadian literature review which raised similar concerns  and found that only ~5% of patients were delayed more than 200 days before receiving adjuvant therapy (most likely chemotherapy followed by radiation), although 28% started adjuvant therapy greater than 60 days postoperatively. Would those patients who were somewhat delayed have been better served by undergoing delayed reconstruction? With the addition of paclitaxel/docetaxel chemotherapy, adjuvant postchemotherapy radiation is delayed by more than 5 months (or even more due to the common practice of waiting ~1 month postchemotherapy to begin radiation treatment). The median time of 38 days from mastectomy to commencing adjuvant therapy is impressive, and the authors suggest modern surgical techniques help decrease morbidity and hasten initiation of adjuvant therapy.
Chemotherapy was not delivered in a "dose-dense" schedule in these patients, and thus the possibility exists that this more aggressive chemotherapeutic approach may produce more postoperative complications especially when followed by radiation therapy. The combined adjuvant therapy effects on long-term cosmesis of reconstructed tissue are an issue that has not been fully resolved. The ~2% disease-free survival advantage of "dose-dense" chemotherapy has spurred its widespread use. A further advantage may also be the shorter interval to the initiation of postoperative radiation therapy.
The authors do not provide definitive criteria as to which patients were offered postoperative radiation therapy. The commonly accepted indications include T3 (>5cm) or T4 primary tumors, greater than 3 positive axillary nodes, incomplete axillary sampling, and close or positive margins. The authors also point out that patients with 1 to 3 positive lymph nodes, extracapsular extension, or younger age, may benefit from radiation therapy. An actual breakdown of irradiated patients is not provided for review.
As with all retrospective reviews, the issue of selection bias may be somewhat important. Furthermore, the reasoning behind classifying patients with > 4-cm primary tumors as high risk (whereas the staging system uses 5 cm as the cutoff for T3 disease) is not explained. No data regarding radiation effects on complications or cosmesis are mentioned, and this is one argument used against performing SSM on patients likely to require postmastectomy radiation treatments. Another unanswered question is the long-term efficacy and/or quality of the radiated reconstructed tissue. These short term data support minimal acute morbidity; however, no cosmesis data is presented. This argument is one reason patients at some institutions typically undergo an initial mastectomy, followed by radiation to the chest wall and then delayed breast reconstruction.
Despite these shortcomings, these data provide a valuable look at outcomes at the Hospital of the University of Pennsylvania should a patient elect to undergo SSM. These data are particularly significant for those patients who are discovered to require postoperative radiation only after surgical intervention (once the pathology has been reviewed). The authors point out that they use sentinel lymph node biopsy and MRI imaging to enhance their ability to predict the need for postoperative radiation therapy. However, it should be noted that 30% of these high-risk patients who underwent SSM were node positive (which is in itself an argument for postoperative radiation therapy , except probably in the one T1aN1M0 patient in this group). In this study however, over 20% of patients received postoperative radiation therapy, indicating the currently held belief that SSM will likely provide superior cosmetic outcome without affecting oncologic control. In fact, the aesthetic improvement is stated as being significant without significant toxicity. Multiple improvements in surgical technique and radiation therapy are attributed to the success seen in this study. However, the role of SSM in patients known to require postoperative radiation therapy still awaits a more definitive data set before guidelines can be established.