How Accurate Is the Pathologic Assessment of the Partial Mastectomy Specimen?

Get Permission

Stuart J. Schnitt, MD

Surgical Margins in Ductal Carcinoma in Situ

Margin assessment of lumpectomy specimens is very problematic. There are technical and methodologic issues, issues with definition and interpretation, and issues regarding the distribution of tumor in the breast.

—Stuart J. Schnitt, MD

The pathologic evaluation of lumpectomy margins is “fraught with problems and pitfalls,” said Stuart J. Schnitt, MD, Director of Anatomic Pathology at Beth Israel Deaconess Medical Center and Professor of Pathology at Harvard Medical School, Boston, who was part of a multidisciplinary discussion of margin assessment during the 2014 Breast Cancer Symposium.

“Quite frankly,” he said, “given the numerous limitations of margin assessment after lumpectomy, it’s amazing that we do as well as we do!”

Pathologic Evaluation of Margins

Dr. Schnitt cited a few basic assumptions in the assessment of lumpectomy specimens:

  • Margin evaluation is an exercise in probabilities, not absolutes.
  • Patients with positive margins are more likely to have residual disease at or near the primary site than those with negative margins.
  • A positive margin does not guarantee the presence of residual disease, nor does a negative margin preclude the presence of even extensive residual disease.

“The goal of margin evaluation is not to ensure that there is no residual tumor in the breast, but to identify patients more likely to have a large residual tumor burden who require further surgery, and, conversely, to identify those unlikely to have a large residual tumor burden and who therefore are suitable candidates for breast-conserving surgery without further excision,” he said.

Limitations of Margin Assessment

“Margin assessment of lumpectomy specimens is very problematic,” Dr. Schnitt said to the oncologists at the symposium. There are technical and methodologic issues, issues with definition and interpretation, and issues regarding the distribution of tumor in the breast.

One of the technical issues is the “pancake phenomenon,” a term that describes how gravity affects the specimen as it sits on the examination table. The typical specimen compresses, or flattens, and this can impact margin assessment. The relationship of margins to each other, as it was in vivo, can change once it arrives at the pathology lab, Dr. Schnitt said.

Specimen orientation, which is normally accomplished via the placement of short and long sutures by the surgeon, can be another problematic area for the pathologist reading the specimen. So can the application of ink to the specimen and the identification of tissue-marking dye colors on the microscopic slides used to assess margins present challenges. 

In one study in which eight pathologists examined specimens for color and color distinctness, certain colors—black, green, red, and blue—were accurately identified by most participants, but identifying violet and distinguishing between orange and yellow dyes proved difficult.1

“It was basically the flip of a coin whether the pathologist recognized the margin as orange or yellow.  Things like this have serious implications for calling a particular margin positive or negative,” he said.

Furthermore, ink can “stray” from its intended mark and render the specimen surface hard to read. And there is no uniform sampling method for lumpectomy specimens, therefore, this process is subject to sampling error, he added.

“Sampling ranges from very limited sectioning to total sequential embedding,” Dr. Schnitt said. “And even with total, sequential embedding, only a small proportion of the specimen is examined microscopically.”

For example, Dr. Schnitt explained, suppose a 4.2-cm lumpectomy specimen is cut at 3-mm intervals, resulting in 14 slices. Each slice is embedded in paraffin and cut at 5 microns, resulting in 14 sections, 5 microns each. This amounts to 70 microns of tissue examined from a 4.2-cm specimen, which totals only 0.2% of the specimen.

“Even when we entirely embed the specimen, we are examining less than 1% of it. Complete histologic examination of this 4.2-cm lumpectomy specimen would require 8,400 slides,” he said.

Issues With Definitions and Measurements

There is also no general agreement among surgeons or radiation oncologists as to what constitutes an adequate negative margin. In fact, there is no margin width about which more than 50% of these professionals agree is “adequate” or “negative,” Dr. Schnitt said.

All the available data are from retrospective studies; the issue has never been addressed in randomized trials and never will be, he added.

The lack of a standard definition of what constitutes an adequate or negative margin has resulted in large variability in reexcision rates following breast-conserving surgery.  A 2012 study of 54 surgeons found their reexcision rates for patients with negative margins varied from 0% to 70%.2 Almost half of the reexcisions, in fact, were performed on patients with negative margins (ie, no tumor at the inked tissue edge).

Another critical issue, he said, is “How well does any given margin measurement reflect reality?” Dr. Schnitt offered a case in point from his own practice.  The specimen showed the distance from tumor to ink to be 2 mm, however, a suspicious area provoked Dr. Schnitt to look further at additional sections obtained from the same tissue block. On these subsequent levels, he found the tumor was actually < 1 mm from the margin, which changed the treatment recommendation.

“If I had not pursued this, I would have said this was a 2-mm margin, which most surgeons and radiation oncologists would agree is fine, considering the patient would be irradiated. In reality, this was a margin of less than 1 mm.”

Microscopic foci of residual disease in the breast are frequent, even with clearly negative margins and even with T1 tumors. In one classic study, 42% of patients with tumors that were < 2 cm and excised with a 2-cm margin were found to have residual disease, with invasive cancer and ductal carcinoma in situ occurring equally.3 Residual disease was also found in 17% of patients with 3-cm margins and in 10% with 4-cm margins— “and these results are for margin widths that are all unacceptably large,” Dr. Schnitt said.

Does Margin Width Really Matter?

“So, given the foregoing information, do millimeters really matter in margin assessment?” Dr. Schnitt asked.  “In fact, in current clinical practice, margin width must be viewed in the context of other factors that influence the risk of local recurrence, such as breast cancer subtype and the impact of systemic therapy.”

Fortunately, multidisciplinary conferences help guide patient care in complex cases. A new understanding of risks imposed by subtype and the efficacy of modern systemic therapy also help to ameliorate pitfalls in the pathologic assessment, though systemic therapy, by itself, does not compensate for inadequate resection, Dr. Schnitt cautioned. ■

Disclosure: Dr. Schnitt reported no potential conflicts of interest.


1. Williams AS, Hache KD: Recognition and discrimination of tissue-marking dye color by surgical pathologists: Recommendations to avoid errors in margin assessment. Am J Clin Pathol 142:355-361, 2014.

2. McCahill LE, Single RM, Aiello Bowles EJ, et al: Variability in reexcision following breast conservation surgery. JAMA 307:467-475, 2012.

3. Holland R, Veling SH, Mravunac M, et al: Histologic multifocality of Tis, T1-2 breast carcinomas. Implications for clinical trials of breast-conserving surgery. Cancer 56:979-990, 1985.

Related Articles

Expert Point of View: Harold Burstein, MD

Harold Burstein, MD, Associate Professor of Medicine at Dana-Farber Cancer Institute, Boston, spoke to The ASCO Post about the multidisciplinary discussion of margin assessment during the 2014 Breast Cancer Symposium.

Should Oncologists Be Concerned?

“There is a real art to being a good...