Surgical margin status is an important predictor of local recurrence and affects survival of early-stage breast cancer, but no consensus exists on what constitutes an optimal margin. Its definition ranges widely, from "no tumor on ink" to "at least 1.0 cm clearance." It is intuitive that a wider margin will contribute to better outcome, but even "wide negative margins" cannot guarantee complete tumor removal due to architectural complexity of the mammary ductal system, multifocality of some tumors, and limitations of margin assessment methods.
In a study of 60 mastectomy specimens, Faverly and associates explored the 3-dimensional structure of the mammary ducts and the patterns of distribution of ductal carcinoma in situ (DCIS). Each mastectomy specimen was sectioned at 5-mm intervals and x-rayed, and the slices containing radiographically or grossly suspicious areas were carefully mapped and examined by conventional microscopy and stereomicroscopy to determine the extent of intraductal carcinoma and connections between different foci. The study demonstrated that 70% of low-grade lesions were multifocal, whereas 90% of high-grade DCIS showed continuous duct involvement.20 These findings suggest that negative margins may be easier to achieve and more reliable in cases of high-grade DCIS as compared to low-grade carcinoma. In the same study, the gaps between the foci of carcinoma ranged from 0.1 to 4 cm, but most foci (82%) were less than 0.5 cm apart; hence, a 0.5-cm margin would have been adequate for 82% of patients.20 Studies aimed to improve the conservative management of breast carcinoma have consistently demonstrated the benefit of radiation therapy in achieving local control, and a margin smaller than 5 mm is considered adequate in this setting.21 A recent study evaluated over 1000 patients with DCIS treated with lumpectomy and radiation and showed that the recurrence rate at 10 years was 8% in patients with a 2-mm margin.22
Several methods of margin evaluation have been proposed. They are outlined next, with discussion of advantages and disadvantages of each method.
Radial (Perpendicular) Margin Evaluation
The most common method of margin assessment is the perpendicular (or radial) margin technique, which allows for precise measurement of the distance separating the tumor from the closest inked margin (Fig. 15-1). With this method, the specimen is received oriented with side specification (right vs left breast). Usually, the surgeon designates the superior margin with a short suture (or a single metal clip), and the lateral margin with a long suture (or 2 metal clips). The 6 margins are differentially inked by the prosector. Ink should be applied lightly, with gentle dabbing, so not to artificially disrupt the tissue. Excess ink should be gently removed by pressing a gauze sponge on the specimen, to reduce the problem of running ink. Brief application of acetone on the inked surface seems to enhance staining and reduce the problem of running ink.
Perpendicular (radial) margin evaluation method. The specimen is received oriented (A) and the 6 margins are differentially inked (B). The tissue is then sequentially cut into 2- to 3-mm slices perpendicular to the long axis of the specimen, so that the perimeter of each slice contains 4 margins identified by different color inks (C). The margin sections are taken perpendicular to the inked surface (C). The extent of individual margin sampling is dictated by its proximity to the lesion. The tissue is embedded in such a manner that the distance from the lesion to the inked margin can be measured (D).
It is best to follow an established coloring scheme for margin designation, but color-coding should be always specified in the gross description for reference. Any color scheme is acceptable, as long as it specifically identifies each of the 6 margins. Medial and lateral margins can be inked with the same color and submitted in separate tissue blocks with margin designation. The inked specimen is then sequentially cut into 2- to 3-mm slices perpendicular to its long axis, so that the perimeter of each tissue slice contains 4 margins identified by ink of different colors. The extent of margin sampling is determined by its proximity to the index lesion. Margins within 0.5 cm of the index lesion are best evaluated in their entirety, and more distant margins can be representatively sampled. Sections should be taken preferably from fibrous, not fatty areas. With this method, the pathologist can report the exact microscopic distance from the tumor to each individual margin and distinguish between a truly positive margin ("tumor on ink"), a close margin (1-2 mm), and a margin 2 mm or greater, allowing the surgeon a greater level of judgment in determining the need of reexcision.23 The disadvantages of this method include imprecise margin orientation and seepage of ink. Breast tissue is soft and may be artifactually disrupted when the specimen is compressed to obtain a specimen radiograph, or in the course of gross examination. Seepage of ink inside the specimen and different color inks running into each other occur frequently, leading to possible overinterpretation and false-positive margins.
Shaved (En Face) Margins Obtained from the Specimen
With en face margins obtained from the specimen (Fig. 15-2), the oriented specimen can be inked entirely in 1 color, as long as the prosector is able to maintain specimen orientation, but use of different color inks for different margins is more reliable. The margins are shaved off parallel to the outer surface of the specimen (in a manner similar to peeling an orange) at a depth of 2 to 3 mm. Shaving of the margins is greatly facilitated by a short course (1-2 hours) of formalin fixation, as the outer tissue acquires firmer consistency. The shaved margins closest to the index lesion are submitted entirely, and the rest is submitted representatively. In some cases, the entire shaved surface of the specimen can be submitted for evaluation. The sections are embedded en face with the inked surface facing down, so that the histologic examination starts from the inner aspect of the shaved tissue. With this method, a margin is reported as positive when tumor is present in a section designated as shaved margin. This means that malignant cells are present within a 2- to 3-mm radius of the surgical margin, but the exact distance of the tumor to the margin cannot be assessed. If no tumor is identified, the margin is reported as negative. Microscopic examination for this method is straightforward and ink-related problems do not occur. The major disadvantage of this technique is that the pathologist cannot provide detailed information on the margin clearance, which limits the surgeon's ability to discriminate among patients with truly positive ("tumor at ink") or close margins, leading to a higher rate of reexcision.23
Shaved (en face) margins obtained from the specimen. The specimen is received oriented and may be inked entirely in one color as long as orientation is maintained (A). Then, the 6 margins are completely shaved off the specimen at a depth of 2 to 3 mm, parallel to the surface (B). Each margin is embedded en face with the inked surface facing down, so that the section for histologic examination is cut from the inner aspect of the shave (C and D).
How does a shaved margin compare with a conventional radial margin? To answer this question, Guidi and associates evaluated 22 surgical specimens using both methods. The specimens were inked, and the margins were first shaved and examined microscopically. Then, the tissue was extracted from the block, cut perpendicular to the inked surface and reembedded to assess radial inked margins. The study demonstrated that a negative shaved margin was highly predictive of a negative inked margin (98% concordance), but the positive predictive value was much lower because only 61% of the positive shaved margins were called positive (tumor on ink) by the radial margin method.24
Evaluation of separate cavity shaves obtained by the surgeon likely provides the best solution to margin assessment, as it combines the advantages of the 2 methods described above and it is ideally suited for treatment of malignant tumors, when the goal is to excise the lesion with a rim of benign tissue. With this method (Fig. 15-3), the surgeon resects the lesion and then takes separate shave margins from the surgical cavity. A separate anterior margin may or may not be submitted. The specimen containing the main tumor (central core) is received unoriented and does not necessarily need to be inked. The tumor in the central core is entirely sampled, including any prior biopsy site, if present; one or two representative sections of any grossly uninvolved breast tissue present are also submitted. Each shave specimen represents a margin, and is received oriented with a suture (or metal clip) designating the final margin surface; occasionally, 1 specimen can consist of 2 adjacent margins. Each margin is inked on the side designated with a suture, perpendicularly sectioned, and submitted either entirely or representatively in 10 blocks. This technique allows precise margin designation and accurate measurement of the margin width. These specimens are easily handled by the prosector and limited manipulation of the tissue and use of the same color ink contribute to reduce some of the problematic artifacts described for the other methods. A significant increase in the number of blocks and slides is the main disadvantage of this technique. A group of investigators recently reported that this method significantly reduces the rate of reexcision for close margin.25
Cavity shave method. The surgeon first removes the lesion and then takes separate margins from the remaining cavity. Each margin should be appropriately labeled and oriented with a suture indicating the actual margin surface (A and B). The surface with a suture is inked, and the margins are serially sectioned and entirely submitted (C). Several sections can fit in one block. The tissue is embedded in such a manner that the margin width can be reported (D).
The cavity shave method is usually adopted by the surgeon only in patients with documented diagnosis of malignancy (by core biopsy or fine-needle aspiration). Either the perpendicular margin or the shaved margin technique can be used in other cases.
Sampling of Excision Specimens
Small specimens (up to 3-4 cm) are usually submitted entirely, while larger breast excisions are selectively sampled. The extent of sampling is dictated by clinical, radiologic, and gross findings.
If the surgeon suspects malignancy, he or she will orient the specimen to allow assessment of margin status. The lesion corresponding to a palpable abnormality should be carefully examined and, in many cases, will need to be entirely submitted. Several sections (2 or 3) representative of unremarkable breast parenchyma away from the lesions also need to be submitted.
When excision is performed for a nonpalpable abnormality, the lesion is radiographically located with the aid of one or more needle wires. The specimen is intraoperatively x-rayed to document retrieval of the index lesion and guide the surgeon as to the need for additional tissue resection. The specimen radiograph is usually sent to pathology along with the specimen to facilitate histologic sampling. If an alphanumeric specboard was used to hold the specimen during x-ray, the tissue should be kept on it during transport to maintain specimen orientation, facilitate correlation with the accompanying specimen radiograph, and guide sampling.
The excision specimen is usually received oriented for margin assessment. The prosector measures and inks the specimen (see methods above), notes the presence and number of wires, and sequentially sections the specimen (usually perpendicular to the specimen main axis) to visualize the index lesion and/or core biopsy site. The area of interest is usually adjacent to the thick portion of the needle wire. If the lesion was previously sampled by core biopsy, a metal clip may be present, and its identification (or lack thereof) needs to be recorded. The suspicious area is widely blocked out and entirely submitted. If any gross abnormality is identified, the gross description should include its size and proximity to margin (specimen surface). Two to 3 representative sections from the grossly uninvolved parenchyma should be also submitted.
If the breast excision specimen contains calcifications, they need to be mapped and entirely submitted, noting the designation of the blocks. The remaining fibrous tissue can be representatively sampled in 2 or 3 blocks.
If no abnormality is found grossly, the entire tissue or 10 sections of its fibrous parenchyma are submitted. Schnitt and Wang studied optimal sampling of grossly benign breast lesions, and found that initial evaluation of 10 blocks of fibrous parenchyma was a cost-effective method, suitable to detect 97% of cases of microscopic atypia or carcinoma in situ.26 The remaining tissue can be examined histologically only if carcinoma or atypical hyperplasia are found among the initial 10 blocks.
The pathologist should consider submitting all tissue if the excision is performed for a core biopsy diagnosis of atypical ductal hyperplasia (ADH) or DCIS to evaluate the morphologic spectrum of the ductal proliferation, rule out stromal invasion, and assess surgical margins status. In cases of DCIS, its extent can be calculated if the specimen is serially sectioned at 2- to 3-mm intervals and entirely submitted in sequential order.
Submission of all tissue is routinely done for breast excisions performed for
- MRI-detected lesions. Selective sampling is not possible for these specimens, as no lesion is appreciated grossly.
- Calcifications not amenable to core biopsy. The suspicious calcifications may be very focal and could be missed on limited sampling. These specimens are received with an accompanying specimen x-ray, and need to be sampled accordingly. If feasible, these specimens should be submitted in their entirety.
- Any sizable, clinically or radiologically detected intraductal papilloma. These lesions should be examined in their entirety to exclude focal atypia or carcinoma. A papilloma usually arises in a large duct of the breast, often in a central/subareolar location. A breast excision performed for a papillary lesion is usually slightly cone-shaped and is received oriented with a stitch (metal clip) designating the apex of the specimen, which is usually the area closest to the nipple, and one designating the superior or lateral aspect of the specimen, as indicated by the surgeon. After the specimen is inked to reflect its orientation, it is best to section it from the nipple toward the deep aspect, rather than from lateral to medial. Transverse sectioning allows serial evaluation of the central ducts and maximizes the chances of identifying the papillary lesion, and of confirming its excision. In cross-section, a papilloma (or any other large papillary tumor involving a central duct) will appear as a collection of delicate fronds protruding into a cystic lumen, which consists of a dilated duct.
Discordant core biopsy and imaging findings. Atypia or malignant cells identified on fine-needle aspiration (FNA) without obvious mass on gross examination.