Margin status of lumpectomy specimens is related to frequency of local recurrence. Optimal surgical technique requires microscopic margins free of carcinoma by at least 2 mm. Recurrence following lumpectomy is associated with residual carcinoma secondary to inadequate resection.
To review our series of breast excisions to determine the frequency of residual carcinoma for positive, close, and negative margins.
We reviewed lumpectomies and excisional biopsies for invasive ductal carcinoma that had subsequent reexcisions. Margin status of specimens was recorded as positive, less than 1 mm, 1 to 2 mm, or greater than 2 mm.
A total of 123 lumpectomies and excisional biopsies of invasive ductal carcinoma with reexcision were reviewed. Residual invasive carcinoma was found in 44% (17), 25% (6), 28% (8), and 16% (5) of cases with positive, less than 1 mm, 1 to 2 mm, and greater than 2 mm margins, respectively. Residual invasive carcinomas were found in 57% (8), 100% (5), 67% (2), and 100% (2) of mastectomies with positive, less than 1 mm, 1 to 2 mm, and greater than 2 mm margins, respectively, in the initial lumpectomy or excisional biopsy.
Frequency of residual invasive carcinoma was related to margin status of the original lumpectomy/biopsy. Even when margins were positive, most reexcisions were free of carcinoma. Residual invasive carcinoma was found in greater than 25% of patients with margins less than 2 mm, supporting reexcision for patients with margins of less than 2 mm. Sixteen percent of cases with margins greater than 2 mm harbored residual invasive carcinoma. Evaluation of margin status was complicated by tissue distortion and fragmentation.
Breast-conserving surgery is now the standard of care for women with early stage carcinoma of the breast. This treatment modality has been shown to be equivalent to mastectomy1 and, when properly performed, optimizes cosmesis. The success of breast-conserving surgery depends on the attainment of adequately negative, but not excessively large, operative margin results.2 In general, surgeons performing breast-conserving surgery attempt to obtain a gross 1-cm margin around a nodule, with the expectation that the final microscopic margin will be clear by at least 2 mm.3 Standard surgical practice requires the attainment of clear operative margins, even if a second operative procedure is required.4 It has been assumed that reexcision of breast tissue around the lumpectomy site when the original surgical procedure had positive margins is as effective at reducing the risk of recurrence as the attainment of negative margins in the initial excision.4 Young patient age appears to be the variable most predictive of positive margins.4 Mirza et al5 found margin status, patient age less than 50 years, large tumor size, positive lymph nodes, absence of chemotherapy, and absence of hormone therapy all to be predictors of local recurrence. The most important of these variables may be final operative margin status.6,7 A study of 1430 patients undergoing lumpectomy showed that negative margins were achieved on initial resection in 1225 patients (86%).4 More recently, separate cavity margin sampling at the time of initial lumpectomy has been used to reduce the number of reexcisions for positive lumpectomy margins.8 There is a clear clinical relationship between rates of local recurrence and proximity of carcinoma to excision margin.9 Although this relationship between margin status and local recurrence is well documented, relatively few studies have correlated the initial margin status for lumpectomies and excisional biopsies with the frequency of residual invasive carcinoma in the reexcision specimens.10–13 Unfortunately, most studies addressing this issue did not stratify margin status by 1 mm increments, and many studies included examples of in situ carcinoma in their analyses. Traditionally, margin status of excisional biopsy and lumpectomy specimens has been evaluated by treating the pathology specimen as a hypothetical cube and marking each of the 6 margin faces with a unique color of ink.8 Orientation of the specimen is indicated by the surgeon using one or more suture marks. Unfortunately, specimens rarely have smooth, intact margins but are usually characterized by specimen fragmentation, irregular surfaces with fissures, and irregular protrusions of breast tissue, complicating assessment of the final inked margin.8 The frequency of residual carcinoma in the reexcision specimens has varied greatly. Kotwall et al2 found that close, but negative, margins were associated with a relatively low incidence of residual carcinoma in the reexcision specimen. When margins were negative, but free of carcinoma by less than 1 mm, microscopic residual carcinoma was found in 6% of cases. Similarly, when margins were greater than 2 mm in size, residual microscopic cancer was found in only 4% of patients. Surprisingly, lumpectomy margins free of carcinoma by 1 to 2 mm showed residual carcinoma in 15% of cases. These frequencies are considerably less than the 21% of patients with negative margins harboring residual carcinoma in resection specimens reported by Scopa et al.14 Cao et al,8 in a study of 126 patients with ductal carcinoma of the breast (103 invasive ductal carcinomas) undergoing lumpectomy with immediate excision of cavity margins, found residual carcinoma in 50% of the cavity margins when carcinoma was present within 2 mm of the original lumpectomy margin. In 2 of the 23 patients (9%) with initially negative lumpectomy margins (carcinoma was farther than 2 mm from the margin), foci of carcinoma were found in the additional cavity margins.5 Based on their findings, Kotwall et al2 stated that all clear margins were equivalent, irrespective of the distance in millimeters, and no further reexcision was necessary even for those tumors with margins less than 1 mm. However, Scopa et al14 stated that histologically negative, but close, margins in biopsy specimens do not guarantee complete excision of carcinoma. These authors believed that residual in situ carcinoma in the immediate vicinity of the previous biopsy site is difficult to completely excise and is most likely responsible for locally recurrent disease. The precise risk of residual carcinoma associated with close, but technically negative, margins is not clear. We studied the relationship between margin status in the initial resection specimen and the presence or absence of carcinoma in the reexcision specimen for a series of 123 cases of invasive ductal carcinomas. Herein, we report of the findings of that study.
MATERIALS AND METHODS
The surgical pathology files for the Department of Pathology at the University of Utah (Salt Lake City, Utah) were searched for all cases of invasive ductal carcinoma present in a lumpectomy or excisional biopsy specimen. This set of cases was then searched electronically for those cases that had additional margins taken at the time of lumpectomy (cavity margin) or had subsequent reexcision or mastectomy. For each of the cases identified in the electronic search, the surgical pathology reports and glass slides were obtained. The glass slides were rereviewed to confirm the diagnosis of invasive ductal carcinoma, both in the initial specimen and in subsequent reexcisions or mastectomies, and to measure the distance in millimeters between the inked operative margins and the closest focus of invasive ductal carcinoma. For study purposes, each specimen was considered to be a cube with 6 unique margin faces. Each face had been inked with a unique ink color. Orientation of the margins to superior, inferior, medial, lateral, superficial, and deep direction was based on suture orientation points supplied by the surgeon performing the lumpectomy or excisional biopsy. For analysis, the closest margins of the lumpectomies and excisional biopsies were stratified into the groups: positive, less than 1 mm, 1 to 2 mm, and greater than 2 mm. These margin distance categories were then correlated with the frequency of residual invasive ductal carcinoma in the reexcision and/or mastectomy specimens. The greater than 2 mm category was further subdivided into 2.1 to 4 mm, 4.1 to 9 mm, and greater than 9.0 mm.
Between January 2005 and February 2008, 173 lumpectomies and excisional biopsies were performed for invasive ductal carcinoma. Ninety-seven (56.1%) had positive margins or margins of less than 2 mm. One hundred twenty-three cases (71.1%) of invasive ductal carcinoma were identified in lumpectomy or excisional biopsy specimens that had inked operative margins and either intraoperative reexcisions (cavity margins) or delayed reexcision or mastectomy. Only these 123 cases were further studied. In 98 patients (79.7%), the reexcision specimens were collected during a single procedure (cavity margins), and in the remaining 25 cases (20.3%), delayed reexcision and/or mastectomy was performed (Table). Residual invasive ductal carcinoma was found in 44% (17), 25% (6), 28% (8), and 16% (5) of reexcision specimens with positive, less than 1 mm, 1 to 2 mm, and greater than 2 mm margins, respectively (Figure 1). Residual carcinoma was found in 35% of specimens performed during the same procedure (cavity margins) and in 43% of cases on review of permanent sections of cavity margins or when a delayed reexcision or mastectomy was performed (Figure 2). In 35 of 123 cases (28%), negative margins (>2 mm) were achieved at the time of lumpectomy/biopsy (Table). In the 24 cases (20%) where mastectomy was performed after a lumpectomy or excisional biopsy, residual invasive carcinoma was found in 57% (8 cases), 100% (5 cases), 67% (2 cases), and 100% (2 cases) of mastectomy specimens with positive, less than 1 mm, 1 to 2 mm, and greater than 2 mm margins, respectively in the initial lumpectomy or excisional biopsy (Figure 3). When the greater than 2 mm category was further subdivided, all residual carcinomas were found in the 2.1 to 4 mm group, even though it represented only 48% of the cases with margins greater than 2 mm.
Interestingly, residual carcinoma was found in only a minority of samples, including those with positive margins. Approximately 57% of reexcisioned specimens with positive margins showed no evidence of residual invasive carcinoma in the subsequent reexcision or mastectomy. In the present series, residual invasive carcinoma was found in a higher percentage of mastectomy specimens than in reexcisions. Hence, 57% of mastectomies in which the original excision had been positive for carcinoma showed residual invasive ductal carcinoma, whereas only 44% of all cases demonstrated residual carcinoma. Similarly, when margins were adequate (>2 mm), 100% of the mastectomy specimens demonstrated residual carcinoma in contrast to the 16% of residual carcinoma seen in all reexcision specimens.
Review of the initial lumpectomies and excisional biopsies showed most to have irregular, fragmented margins with deep fissures, tissue protrusions, or loose tissue fragments (Figure 4, A and B).
Breast-conserving therapy with adjuvant radiotherapy has been shown to be equivalent to mastectomy for 5- and 10-year survival rates.1 Although 5- and even 10-year survival rates are comparable to those of mastectomy, breast-conserving therapy is associated with more frequent local recurrence.15 A variety of factors have been found to be predictors of local recurrence following lumpectomy, including age less than 50 years, large tumor size, positive lymph nodes, absence of chemotherapy, absence of endocrine therapy, and positive margins in the original excision specimen.4,6,16 A positive or close operative margin appears to be a strong independent predictor for local recurrence. If final margins are negative, the 5-year risk of local recurrence is between 2% and 7%.7 However, if the margins are positive or close, the 5-year risk of local failure ranges from 0% to 22% and from 2% to 11%, respectively.7 Gage et al9 detected a close relationship between the positivity or closeness of an inked operative margin to carcinoma and frequency of local recurrence. When negative margins were present, the incidence of recurrence was approximately 2%. For close margins (≥1 mm), the frequency of local recurrence was also 2%. However, when margins were either focally or multifocally positive, the frequency of local recurrence was between 9% and 28%. Based on these findings, most surgeons performing breast-conserving therapy attempt to obtain a gross margin of 1 cm around a mass, with the expectation that the microscopic margin will be clear by at least 2 mm.3 Thus, margins free of carcinoma by 2 mm or more are considered clear, and lesser margins generally require reexcision.17 Margin status, as determined by histopathologic examination, acts as a surrogate marker for risk of residual disease. The percentage of patients with positive margins requiring local reexcision has ranged from 1% to 14.3%.4,5 Institutions practicing separate cavity margin sampling at the time of initial breast lumpectomy have a higher incidence of positive margins (82%) in the initial lumpectomy before cavity margin sampling is performed.8 Our rate of margin positivity for excisional biopsy and lumpectomy specimens was relatively high at 56.1%. Prior studies have reported rates of lumpectomy margin positivity of approximately 14%. Our high rate of margin positivity may reflect a mixture of specimens undergoing lumpectomy with cavity margin sampling, some of which had frozen-section evaluation and immediate reexcision, whereas others had cavity margins taken but were not examined intraoperatively. The former cases frequently had final intraprocedural margins free of invasive carcinoma by greater than 2 mm.
Positive resection margins are known to be associated with a higher local recurrence rate, and thus, immediate or delayed reexcisions are performed in most cases.6,16 The efficacy of reexcision for the prevention of local recurrence has not been well studied, but most authors have reported rates of local recurrence to be similar between initial complete excision and a negative margin following reexcision.18 Others have demonstrated a mildly elevated actuarial risk of local recurrence at 10 years for women who required reexcision of the lumpectomy cavity to achieve clear final resection margins.4,7 In these studies, the differences in recurrence rates between those with initially clear margins and those requiring reexcision were not statistically significant. Thus, breast-conserving therapy appears to be essentially equivalent for 5- and 10-year survival as well as for local recurrence rates whether reexcision is or is not required for establishment of clear margins.
Many authors believe that the risk of local recurrence following breast-conserving therapy is directly related to the presence or absence of invasive carcinoma in the residual breast tissue.2 Several studies have investigated the likelihood of carcinoma being present in reexcision specimens following positive and close margins.2,10–14 These studies were heterogeneous in the types of carcinoma studied as well as in the methods for quantitating margin status. Many studies included both in situ and invasive carcinomas, whereas others did not stratify margin distances by 1-mm increments. The presence or absence of invasive carcinoma in residual breast tissue following breast-conserving therapy is probably the most significant predictor for local recurrence.2 Thus, our study and that of Kotwall et al2 were restricted to the identification of invasive ductal carcinoma within the initial resection and the subsequent reexcision/mastectomies. The study by Scopa et al14 approximated our study and that of Kotwall et al because it reviewed 201 consecutive lumpectomy specimens of which 131 (65.2%) were various types of invasive ductal carcinoma. In the Kotwall et al review of 582 early stage invasive ductal carcinomas, margin status was assessed as positive, focally positive, margins less than 1 mm, margins between 1 and 2 mm, and margins greater than 2 mm.2 They found that when margins were multifocally positive, 30% of patients had residual carcinoma in their reexcisioned specimen.2 When margins were only focally positive, the frequency of residual carcinoma in excisioned specimens fell to 22%. When margins were clear by less than 1 mm, an 8% residual carcinoma rate was obtained, and rates of 15% and 4% were found for margin statuses of 1 to 2 mm and greater than 2 mm, respectively. From this, Kotwall et al2 concluded that once margins were definitively clear of carcinoma, the percentage of patients with residual carcinoma before reexcision was relatively similar, and reexcision was not necessary.2 We found a higher frequency of residual carcinoma in reexcisioned specimens when margins were positive (17 of 39; 44%). Of more significance was our finding that when margins were clear but close (<1 mm and 1 to 2 mm), a significant number of patients had residual carcinoma in reexcisioned specimens (25% [6 of 24] and 28% [8 of 28], respectively). This would support reexcision in patients where a margin was clear by 2 mm or less. This finding is in agreement with the report by Scopa et al,14 who found residual carcinoma in excisioned specimens of 21% of patients with negative margins. When the margins were described as close, fully 30% of these patients had residual carcinoma in the reexcision specimen.14 In this latter study,14 a significant percentage of the residual carcinomas were in situ. Nonetheless, approximately 1 in 4 patients with close, but negative, margins had residual carcinoma justifying reexcision.
A surprising finding in our study was that 16% (5 of 31) of patients with traditionally negative (>2 mm) operative margins had residual carcinoma in their reexcision or mastectomy specimens. This was substantially higher than the 4% reported by Kotwall et al.2 This would lend support to the suggestion by some European authors that a negative margin should be defined as greater than 5 mm.2 Unfortunately, we did not stratify our data for margins greater than 2 mm.
An interesting finding in our study and in that of Cao et al8 and Kotwall et al2 was the relatively low incidence of residual carcinoma when lumpectomy or excisional biopsy margins were positive. Residual carcinoma was found in no more than 55% of these cases and frequently less (22%–70%).2,10,13,19–27 The reasons for this relatively low rate of residual carcinoma in patients with positive margins are unknown but appear to be multifactorial.8 Potential reasons include destruction of residual carcinoma by the operative procedure and subsequent granulation tissue formation, incomplete sampling, and the potential for inaccurate interpretation of the original margins because of artifacts of inking, tissue fissures, or fragmentation and irregular operative margins. A number of our specimens did not resemble clean cubes but were composed of irregular, fissured, and fragmented specimens of breast tissue (Figure 4, A and B). Specimens having significant disruption, as pictured in Figure 4, A and B, comprised 8% (10 of 123) of our cases. This degree of disruption made accurate inking and assessment of the margins difficult. These residual carcinomas were located in the general area of the lumpectomy cavity. All were invasive carcinomas and of the same histologic appearance as the original tumor. Thus, in some of the specimens, margins appearing positive on histopathologic assessment might in actuality be negative biologically.
Another finding of interest in our series was that the positivity of reexcision was higher in patients undergoing mastectomy than in those with margin reexcision. These residual carcinomas were located in the general area of the lumpectomy cavity. All were invasive carcinomas and of the same histologic appearance as the original neoplasm. Thus, residual invasive carcinoma was found in 57% (8), 100% (5), 67% (2), and 100% (2) of mastectomy cases where the initial margins had been positive, less than 1 mm, 1–2 mm and greater than 2 mm respectively. This higher incidence of residual invasive carcinoma in mastectomy specimens than in more limited reexcisions appears to be due to foci of carcinoma farther removed from the lumpectomy cavity than is normally sampled by reexcision. Hence, a number of residual foci in mastectomies in our study were well beyond the zone characteristically removed by reexcision and also beyond the cut point for what would be considered a negative margin by either American or European standards. Given the success of breast-conservation therapy,1 it appears the presence of these residual foci do not negate the advantages of breast-conserving therapy in early stage breast cancer. However, this finding may imply a need for postlumpectomy radiotherapy in some patients.
Our study indicates that a significant number of patients with margins less than or equal to 2 mm have invasive carcinoma foci in their residual breast tissue. In our study, approximately a quarter of patients with negative margins up to 2 mm in size will have residual foci of invasive ductal carcinoma before reexcision. This supports the current approach of reexcising margins when clearance is less than 2 mm. Our study showed a 16% frequency of residual carcinoma, even when margins were greater than 2 mm. This raises the question as to whether a 2 mm margin is adequate. A number of European authors require a margin of 5 mm to consider a lumpectomy negative.2 Although we did not stratify margin positivity between 2 and 5 mm, our findings suggest that 5 mm margin may be of aid in reducing overall local recurrence.
Plan now to submit abstracts and case studies for the College of American Pathologists (CAP) 2010 meeting, which will be held September 26th through the 29th in Chicago, Illinois. Submissions for the CAP ‘10 Abstract Program will be accepted from:
Accepted submissions will appear in the September 2010 issue of the Archives of Pathology & Laboratory Medicine. Visit the CAP ‘10 Web site at www.cap.org/cap10 for additional abstract program information.
From the Department of Pathology (Dr Skripenova) and the Division of Anatomic Pathology, Department of Pathology, and the ARUP Laboratories (Dr Layfield), University of Utah School of Medicine, Salt Lake City.
The authors have no relevant financial interest in the products or companies described in this article.