| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
From the Department of Surgery, University of Toronto, Head, Breast Site, Princess Margaret Hospital, Toronto, Ontario, Canada.
Breast-conserving surgery, or lumpectomy, and postoperative radiation comprise the standard option for the management of most localized, early breast cancers. The procedure provides organ preservation with its attendant cosmetic and sensory benefits and has withstood the oncologic test of time. The 20-year follow-up of Bernard Fisher’s1 randomized trial demonstrated equivalent survival rates for breast conservation compared with mastectomy. In this trial, eligible lumpectomy-treated patients had the primary tumor removed with sufficient normal tissue to ensure that the inked margins were free of tumor. Although it has been demonstrated that breast cancer cells can reside in areas away from the primary tumor and that lumpectomy may not remove every last deposit,2,3 this, and other studies including a meta analysis,4 have demonstrated equivalent survival between breast conservation and breast amputation. Modern series of lumpectomy, breast radiation, and appropriate systemic management have produced long-term local control rates of 90% to 95% or greater5 and "efforts to expand eligibility. . .and to reduce the associated morbidity are well under way."6
The goal of breast-conserving surgery is to remove all evidence of the index breast cancer in a cosmetically optimized fashion that, when combined with postoperative irradiation, a minimal rate of local recurrence is achieved. A number of tools can be used to help excise all known disease, including (1) meticulous preoperative imaging; (2) precise surgical technique with adequate margins of normal-appearing surrounding tissue; (3) use of bracketing wires for larger, nonpalpable lesions; (4) postoperative imaging to ensure that the index lesion has been totally removed; and (5) careful margin assessment of the excised specimen.
Margin width or the distance from breast excised cancer cells (invasive or in situ) to the cut edge of the surgical specimen is not an end in itself. Although important, it is but one factor in the imprecise methodology used to estimate the amount of residual disease and the propensity of any residua to form a local recurrence despite other adjuvant treatments.
In this month’s Annals of Surgical Oncology, Cellini et al.7 have attempted to shed light on a vexing surgical problem: Do all patients with unsatisfactory specimen margins after initial lumpectomy need re-excision? The goal of their study was to identify a subset of patients who could be spared the second surgery. They reviewed 276 patients with positive, close (study definition: <1 mm) or undetermined surgical margins after lumpectomy from a total cohort of 403 patients. They state that all initial excisions were intended as definitive procedures. The 276 patients then had either a mastectomy or re-excision. The presence or absence of residual cancer in the pathology specimen was the study’s endpoint. They found that higher grade tumors, larger tumors, and younger age increased the chance of finding residual disease. The authors did not define a subgroup that could avoid re-excision.
Some aspects of this paper merit further discussion. The study’s major endpoint, the presence of residual cancer, is a pathologic finding and infers a higher rate of local recurrence. Chart review of the existing pathology reports provided the study’s data, yet no consistent, systematic methodology is described for specimen handling or pathologic evaluation of either the initial lumpectomy or the subsequent re-excision or mastectomy. Also, no attempt was made to estimate the volume of residual disease in the second surgery or location of disease within a mastectomy specimen.
Furthermore, I am not convinced that all the surgeons approached their lumpectomies with the same definitive intent as the report suggests. If the study cohort of 276 was taken from the overall group of 403 (as the results section states), then the overall rate of having a positive margin following lumpectomy in this series was 42% (168/403). This would be excessive. Even if the overall study denominator was not 403, the chance of obtaining a positive margin was almost twice (168/87) as likely as getting a negative, but close margin. If the intent is to perform a lumpectomy as the one and only breast procedure with attention to intraoperative visual and mammographic assessments, then I would expect fewer positive margins.
At Princess Margaret Hospital, our current rate of finding a positive margin at final pathology in patients having definitive breast-conserving surgery for invasive and in situ cancers, including wire localizations, is <5%. In <15% of cases, the closest tumor (including duct carcinoma in situ [DCIS]) to negative margin width was <1 mm. Most (88%) of the patients who had breast conservation had only one procedure, 4% had a re-excision and 8% opted to have a mastectomy for unsuspected widespread DCIS or multifocal cancer. Similar rates of re-excision and mastectomy following definitive breast-conserving surgery have been published using a population-based cohort in a managed cancer care system.8 The unexpectedly high rate of positive margins found by Cellini et al.7 may be secondary to the proportion of diagnostically driven lumpectomies that were used in the days before standard core biopsy diagnosis.
Of their patients, 53% with "close" margins (<1 mm) were found to have residual carcinoma. Whereas similar to previous reports in the literature, some may use this finding to state that patients with margins <1 mm wide must have a re-excision. In addition to some of the preceding caveats, a few more potential reasons are seen why this should not be a reflex decision. It is possible that some patients in the close margin category may have been truly margin positive. Although margin assessment is a standard component in a complete breast pathology report, its determination is not as precise as might be thought. No data are provided with respect to the number of slides or margins analyzed at the initial lumpectomy.
Cancers are generally not spherical but have protruding spicules of disease and are often surrounded by DCIS. Pathologic assessment is but a sampling of the area. A single hematoxylin and eosin (H&E) slide of a 2- to 3-mm block cut from a lumpectomy specimen visualizes less than one thousandth of the surface area and may not capture the largest microscopic protrusion of cancer cells that are closest to the edge of the cut surface. Report of a close margin made on only one or two sections can be falsely negative. This might explain the Cellini et al.7 finding that specimens inked with multiple colors were associated with a lower chance of residual disease. The inking method may have been a surrogate for precise surgical technique, specimen handling, and more intensive pathologic assessment. For these multi-inked specimens, close margins were more likely truly negative.
Margin positivity or closeness is but one of a number of factors that needs to be evaluated. Sometimes, a close or negative margin may be found, even when specimen and postoperative imaging clearly show residual malignant calcifications. Inclusion of such patients would increase the chance of residual disease. Also, many patients had mastectomy and the residual disease could have been found in quadrants away from the primary tumor as has been demonstrated.2,3 No data regarding these possibilities have been provided.
Despite the difficulties inherent in assessing margin status and width, it is still the most important and quantitative variable in assessing the completeness of tumor excision. It appears clear that the presence of a positive margin with tumor at the cut surface is associated with residual disease and predisposes to local recurrence.5 Secondary surgery is usually required. It is much less clear whether increasing the radial margin between the last seen tumor cell and the cut surface produces improved local recurrence rates in a clinically meaningful way. Whereas some centers have arbitrarily defined negative margins for invasive disease only in the context of a specific tumor-to-cut-surface distance, little difference is seen in their reported outcomes compared with the "no tumor at cut edge" definition when appropriate adjuvant treatments are given.1,5
Negative margins are but one factor to indicate that all relevant evidence of the tumor has been excised and the patient is eligible for adjuvant breast radiation. Tumor size, excision volume, patient’s age, appropriate postoperative imaging, extent of surrounding in situ disease, and margin assessment, including width, orientation, and number, all play a role in determining suitability and success in breast-conserving surgery. Despite evidence that some patients will harbor subclinical disease somewhere within the breast, a lumpectomy with a negative margin, in the context of otherwise favorable variables, does not require more surgery to achieve excellent local control and survival equal to mastectomy. We should not be looking for excuses to perform more mastectomies or re-excisions. Our goal should be to ensure that all appropriate patients are given the option of breast conservation with minimal morbidity and optimal cosmesis.
FOOTNOTES
Received August 5, 2004; accepted August 23, 2004.
Address correspondence to: David R. McCready, MD, MSc, Department of Surgery, Princess Margaret Hospital, 610 University Ave., Suite 3-130, Toronto, Ontario, M5G 2M9 Canada; Fax: 416-946-6590; E-mail: david.mccready{at}uhn.on.ca
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
