Data from the SHAVE and SHAVE2 trials, in terms of long-term outcomes vis-à-vis local recurrence, may lend some insight into the impact of adjuvant therapy on reducing the ramifications of a positive margin.
Anees B. Chagpar, MD, MSc, MPH, MA, MBA
It is well-known that breast-conserving surgery yields the same survival as mastectomy for patients with breast cancer, and it has become the mainstay of surgical therapy for the majority of patients with early-stage disease. Critical to the success of breast-conserving surgery, however, is the achievement of negative margins; a meta-analysis has demonstrated that positive margins have been associated with a 2-fold increase in local recurrence rates.1
Hence, positive margins often result in reexcision. Neither surgeons nor patients want to return to the operating room for another procedure, yet positive margins are found in 20% to 40% of breast cancer cases,2,3 prompting reexcision— an event touted as the “other breast cancer epidemic.”4
Debate is ongoing about what constitutes a positive margin, but recent consensus statements have described a positive margin as tumor at ink for invasive carcinoma,5 or within 2 mm for ductal carcinoma in situ (DCIS).6 For DCIS that is concomitant with invasive disease, without an extensive intraductal component, margins are considered negative if DCIS is not at ink (even if within 2 mm).5 This revision in the definition of a positive margin has resulted in a decline of reexcision rates7; however, the long-term impact of this revision remains to be well elucidated.
Indeed, margin status does not (in absolute terms) predict residual disease.8 Roughly 50% of patients with positive margins will have no further disease and 12% to 14% of patients with negative margins will have cancer left in situ adjacent to the cavity resected. Given that margin evaluation is not an exact science, there remains debate, therefore, on whether all positive margins require reexcision, especially for focal disease,9 anterior margins,10 etc. Data are mixed on the oncologic safety of avoiding reexcision in such patients, so the maxim remains to recommend reexcision for all positive margins.A number of techniques have been evaluated for their value in reducing positive margin rates. To begin, some have felt that the use of preoperative MRI may better define extent of disease and may therefore reduce the positive margin rate by better guiding surgical resection. Sadly, the 2 randomized controlled trials that have evaluated this have not validated this hypothesis: The COMICE trial11 found no difference in positive margin rates between patients who had preoperative MRI and those who did not, and the MONET trial12 paradoxically demonstrated an increase in positive margins associated with the use of preoperative MRI. Some have argued that the value of preoperative MRI may be more evident in patients presenting with DCIS, which tends to be less cohesive and more likely to result in positive margins. However, a recent metaanalysis found that MRI did not significantly affect margin status nor reexcision rates in these patients.13
Others have evaluated various methods of localization of nonpalpable tumors. Although a Cochrane analysis found that no significant differences exist among wire localization, radio-occult lesion localization, and radioactive seed localization in terms of margin positivity and reexcision rates,14 use of intraoperative ultrasound seems to result in a lower positive margin rate. A meta-analysis of 7 prospective studies found that the odds of achievement of negative margins were significantly higher with the use of intraoperative ultrasound (risk ratio, 1.37; 95% CI, 1.18-1.59).15
Still others have argued that intraoperative evaluation of margins may be helpful in reducing positive margin rates. Some utilize intraoperative touch imprint cytology or frozen section, both of which require pathology expertise and may add to operative time. Many surgeons will resect margins where intraoperative imaging suggests a close margin. However, studies that have evaluated this in a robust fashion have not found this to reduce positive margin rates.16,17Resection of circumferential cavity shave margins, however, seems to consistently reduce positive margin and reexcision rates in both retrospective and prospective studies (Table). Three randomized controlled trials to date have all demonstrated that resection of cavity shave margins reduces positive margin and reexcision rates by at least 50%.17,18
The SHAVE trial was the first randomized controlled trial to evaluate this technique. Inclusion criteria included patients undergoing partial mastectomy for stage 0 to III breast cancer. For patients enrolled in the trial after informed consent, surgeons were advised to perform their best operation, including the resection of any margins deemed close on either intraoperative imaging or gross evaluation.
Once the surgeon was ready to close, patients were randomized intraoperatively to either “shave”—that is, have circumferential cavity shave margins resected—or “no shave” in which the surgeon was advised to close. The “shave” procedure added 10 minutes to operative time and did not result in an increase in postoperative seromas, hematomas, or infections. However, patients in the “shave” arm had a significantly lower rate of positive margins and need for reexcision than those in the “no shave” arm.
Although these findings were practice-changing for some, others criticized the study for a number of reasons. First, SHAVE was a singlecenter study, and some questioned whether the results would be externally generalizable. In addition, some argued that the technique of resecting cavity shave margins may be of limited utility for surgeons who already have a low positive margin rate (<25%), and/or who use oncoplastic surgery.
To address these concerns, SHAVE2 was conducted as a multicenter, randomized controlled trial that involved 9 diverse practices from across the United States. This trial demonstrated that the reductions in positive margin and reexcision rate associated with resection of cavity shave margins hold regardless of practice setting (eg, academic vs private practice), prior positive margin rate, or use of oncoplastic surgery.
Of note, the technique of resection of cavity shave margins does (by definition) yield a higher resection volume. However, the SHAVE trial found that this did not affect patient-reported cosmetic outcome postoperatively. Both the SHAVE and SHAVE2 trials will evaluate the impact of resection of cavity shave margins on cosmetic outcome over time, particularly after radiation therapy.
Further, both trials will evaluate local recurrence rate at 5 years. It is anticipated that the impact of cavity shave margins on local recurrence rate will be muted in both trials because all patients, regardless of randomization arm, were expected to be offered reexcision if a positive margin was found. Further, with the nearly ubiquitous rate of systemic therapy and radiation therapy, locoregional recurrence rates are expected to be low.
However, not all patients in the trial underwent adjuvant therapy, and in both the SHAVE and SHAVE2 trials, roughly 12% to 14% of patients in the “shave” arm were found to have residual disease in the cavity shave margins even when the initial margin was negative. Therefore, one would speculate that patients in the “no shave” arm may have a higher burden of residual disease; the impact of this on locoregional recurrence, however, remains to be seen.
How vigorously we pursue positive margins remains a question to be answered. The SHAVE trial found that of 31 margin-positive patients randomized to the “shave” arm in whom the positive margin was resected with a cavity shave margin, further disease was found in 16 (51.6%). The positive margin was rendered negative in 8 of those patients. In other words, a positive margin yielded no further disease in ~50% of cases, leading some to question the value of resection of these margins, and/or the ability to predict which cases would have further disease.
Data from the SHAVE and SHAVE2 trials, in terms of long-term outcomes vis-à-vis local recurrence, may lend some insight into the impact of adjuvant therapy on reducing the ramifications of a positive margin. For now, however, the achievement of negative margins remains a necessary part of breast-conserving surgery.