October 2012
Feature Story

Anne Paxton

In the pathology laboratory no less than in the law, a “bright line” rule is a highly desirable thing. But as breast pathologists, surgeons, and radiation oncologists can testify, it can be an elusive goal when it comes to surgical margins in breast cancer patients—the width of normal breast tissue that should be removed with the primary tumor during surgery. In this era of breast-conserving surgery, or lumpectomy, what is the appropriate surgical margin in patients undergoing surgery? Surprisingly, there’s little medical consensus on how to answer that question.

Pathologists assess the status of the surgical margin by determining the microscopic distance between tumor cells and the surface of the inked lumpectomy specimen. If there is no ink on any cancer cells, the margin is negative; if there is ink, the margin is positive. “Everybody has agreed for decades that if we see tumor cells, either DCIS [ductal carcinoma in situ] or invasive cancer, touching the inked tissue edge under the microscope, that’s a positive margin,” says Stuart J. Schnitt, MD, director of anatomic pathology at Beth Israel Deaconess Medical Center in Boston and a professor at Harvard Medical School. “But what constitutes a negative margin is still not well defined. It varies from surgeon to surgeon and radiation oncologist to radiation oncologist. That’s really where the controversy is.”

In a recent “Sounding Board” in the New England Journal of Medicine, “Surgical margins in lumpectomy for breast cancer—bigger is not better” (2012;367:79–82), Dr. Schnitt joined with co-authors Monica Morrow, MD, of Memorial Sloan-Kettering Cancer Center, and Jay R. Harris, MD, of Dana-Farber Cancer Institute and Brigham and Women’s Hospital, to map a road to consensus on surgical margins. They make the case that with the multi-modality breast cancer treatment that is common today, a clear margin wider than negative may not be as important to minimizing tumor recurrence in the breast (local recurrence) as it used to be. Whether there is local recurrence, they suggest, appears to be related less to the surgical margin width than to the underlying tumor biology and to the availability of effective adjuvant therapy, and there needs to be a “shift in thinking” to reflect that fact. “This shift could have a major impact on clinical practice,” Dr. Schnitt and his colleagues write.

Specifically, these physicians propose that the field adopt “no tumor at the inked tissue edge,” as used in the trials of the National Surgical Adjuvant Breast and Bowel Project (NSABP), as the standard definition of a negative margin. Subjective pathological descriptors such as “close,” the authors say, should be replaced with the measurement of the distance of the tumor cells from the inked specimen surface, without additional qualifications or judgment. Surgeons, radiation oncologists, and tumor boards should abandon routine requirements for margins wider than negative. And the use of reexcision rates as a quality measure should be eliminated. “We believe that the weight of evidence is sufficient to initiate these changes now,” the authors write.

Breast-conserving surgery or lumpectomy followed by radiation therapy is commonly used for the treatment of patients with invasive breast cancer and DCIS, Dr. Schnitt notes. But between 20 and 30 percent of women who undergo lumpectomy undergo additional breast surgery or reexcision, and the rate of reexcisions varies markedly from surgeon to surgeon. A 2011 survey (McCahill LE, et al. Variability in reexcision following breast conservation surgery. JAMA. 2012;307:467–475) found that the reexcision rate among surgeons in the U.S. ranged from zero percent to as high as 70 percent. This variation is a reflection of different practices across the country and what is accepted as a negative margin, says Jean Simpson, MD, professor of pathology at Vanderbilt University Medical Center and a member of the CAP Cancer Committee. She says the recently updated CAP breast cancer protocols follow the suggestion. The margin is reported as positive or uninvolved, with the distance to the closest margin listed (www.cap.org/cancerprotocols).

“What stimulated our writing the article in the NEJM,” Dr. Schnitt says, “was the sense that we and others have had that there are too many patients having reexcisions that probably don’t need them.” While the 2011 survey on reexcision rates didn’t discuss different cosmetic outcomes, “I would think that the surgeons who never do reexcisions probably have some pretty poor cosmetic results, because they do fairly big excisions up front.”

Despite the wide variation and the fact that there is no agreed upon margin width that requires reexcision, some hospitals use the reexcision rate as a quality measure for breast cancer care. “Since there is no agreement as to what constitutes a negative margin, there is no agreement as to what proportion of patients who undergo lumpectomy should be having reexcisions,” Dr. Schnitt says.

There was a time, in the early days of breast-conserving therapy, that pathologists did not routinely evaluate margins. “The surgeon just took out the lump and as long as the gross lump was removed, patients got radiation therapy and that was it. Nowadays, we have better imaging before surgery, and pathologists pay very careful attention to evaluating the margins. So the local recurrence rate is much lower,” he says.

But an even more important factor in the lower recurrence rate is that patients typically get systemic therapy, in addition to radiation therapy, for invasive breast cancer. “There seems to be a synergistic or additive effect between radiation and chemotherapy or hormonal therapy in reducing the risk of local recurrence,” Dr. Schnitt explains. “Patients treated with the combination of a lumpectomy, radiation therapy, and systemic therapy have lower local recurrence rates than patients who get a lumpectomy and radiation therapy without systemic therapy.”

This means, in effect, that minimizing the subclinical tumor burden is no longer as critical for reducing local recurrence. “It is likely that we can get away with smaller margins today than in the past. The problem is that a lot of people are still stuck in the old days when they thought you need a wider margin to reduce the local recurrence rate.” Often, surgeons and radiation oncologists decide upon margin distances with which they feel comfortable, “but those decisions are typically not evidence-based. Some take a ‘one size fits all’ attitude. They might say, ‘We need 2 mm or 3 mm margins around every cancer; otherwise the patient isn’t a good candidate for breast-conserving therapy.’ Or they may be under the misconception that there are definitive data indicating that 3 mm margins are better than 2 mm margins, and 5 mm margins are better than that.”

That there is no margin distance about which 50 percent or more of surgeons or radiation oncologists agree is significant, Dr. Schnitt adds. “When you really look at the data critically, there are no convincing data to indicate that increasing the margin width in 1 mm increments in the 1 to 5 mm range really relates to rate of local recurrence when you control for other factors.”

Not all patients require the same margin distance in order to get optimum local control, especially in this era of systemic therapy being added to radiation therapy, he says, pointing out that the likelihood of a heavy residual tumor burden in the breast varies according to the histologic type of the tumor and the extent of disease in proximity to the surgical margin. If a patient is not going to receive radiation therapy, “then you need a wider margin.” And there are other circumstances in which that’s true as well. “We know from clinical followup studies that some patients need a wider margin and those patients include those whose invasive cancers have a prominent component of DCIS, perhaps some patients with invasive lobular carcinoma, and some patients with pure DCIS, particularly those who are going to be treated with lumpectomy without radiation therapy.”

Much is unknown about what signs of cancer away from the index lesion might mean. With magnetic resonance imaging, for example, “we clearly identify lots of foci of tumor away from the index cancer, but whether or not those foci are clinically significant remains to be seen. Among patients with unicentric invasive breast cancers that are less than 5 cm in size and no clinical evidence that there’s cancer elsewhere in the breast, if you map the distribution of tumor in the breast away from the primary site, 43 percent of the patients have microscopic foci of invasive cancer and/or DCIS more than 2 cm away from the index lesion, and with MRIs we’re detecting many of these foci. In some patients they are probably clinically significant. But in many they’re probably not.”

As a starting point for a more evidence-based approach, Dr. Schnitt and colleagues propose that a standard definition of “negative margin” be adopted. “Everybody can agree that no tumor on ink is negative, although they may not agree on whether or not it’s an adequate negative margin. But right now pathologists and clinicians use terms like ‘close’ or ‘approximate’ which immediately suggest that the margin is bad—i.e. evidence of an inadequate resection. I can tell you from my own experience that there are lots of breast cancers that come very close to the margins, within less than 1 mm, and you do a reexcision and don’t find anything. Just because there is tumor close to the margin doesn’t necessarily mean there’s going to be a lot of residual tumor in the breast, or even any residual tumor in the breast.” Even mastectomy, he notes, which provides the largest margins that can be obtained, does not eliminate local recurrence.

Dr. Schnitt and colleagues advocate an end to the use of reexcision rates as a quality measure. “Nobody knows what the right rate is. If there is some guideline that says any surgeon who has a reexcision rate greater than 25 percent is not practicing properly, then the consequence of that could be that surgeon will do very wide excisions, and in turn that will result in an unnecessarily large amount of breast tissue being taken out at the initial surgical procedure and very possibly adverse cosmetic consequences for patients.”

The lack of standardization of specimen processing can be a major complicating factor in assessing surgical margins, Dr. Simpson says. “What happens is the specimen is removed and it is placed on a cutting board in the surgical pathology lab. It looks like a pancake stretched out, so it’s not really an accurate representation of what was in the patient. Or it may be removed in several pieces and the margins are difficult to determine. Ideally, when the specimen is removed it should be oriented by the surgeon as to which margin is which.”

The best practice, Dr. Simpson says, is for the surgeons to apply different inks to denote different margins—lateral, medial, anterior, and posterior. “More surgeons are doing this in the OR to more accurately reflect where the margin is as it relates to the specimen, so the pathologists know exactly what they are trying to analyze. Then, ideally, the specimen is carefully oriented, and the area is sequentially sectioned and mapped out to match up with the imaging studies.”

She believes that standardizing specimen processing will go a long way toward improving pathologic analysis of surgical margins. “The most important thing is to work with the surgeons to ensure that specimens are received intact and well oriented so the margins can be assessed.”

Dr. Schnitt and his co-authors would like to see an evidence-based consensus on surgical margins, including agreement on what constitutes an adequate margin and identification of the parameters for who needs reexcision, and in the wake of their New England Journal editorial, there is already a promising development in that direction. With a new agreement by the Susan G. Komen Foundation to provide funding, a consensus development process is being jointly put together by the American College of Surgeons’ oncology group and the American Society of Radiation Oncology, with representation from the CAP. The group is expected to begin meeting in spring 2013.

“For whatever reason, many clinicians and pathologists have blind faith in margins, and think that positive margins are always a bad thing associated with lots of residual disease, and negative margins are always a good thing and associated with no residual disease. Nothing could be further from the truth,” Dr. Schnitt emphasizes. “The idea that surgeons need to obtain the same arbitrary, pre-defined margin distance on every patient in order for that patient to be a suitable candidate for breast-conserving treatment needs to change. Just as systemic therapy is becoming more and more personalized, the way we look at margins needs to become more personalized and individualized as well.”