Should You Trust Breast Biopsy Results?

Women who’ve undergone a breast biopsy are likely to agree: Few things are more anxiety-inducing than waiting to learn whether you have cancer. In fact, a 2009 study found that awaiting breast biopsy results affected stress-hormone levels just as much as being diagnosed with cancer.

A study in the March 2015 issue of the Journal of the American Medical Association calls the accuracy of biopsy findings into question, potentially giving women cause for concern. The investigation found that the pathologists in the study who assessed biopsy samples didn’t agree on the diagnoses of certain types of abnormal cells about 25 percent of the time. Pathologists are the doctors who examine tissue, cell and body-fluid specimens; interpret test results; and render a diagnosis.

“While pathologists excel at detecting invasive breast cancer, a misinterpretation can lead to patients receiving unnecessary treatment for a harmless growth or not receiving treatment for a malignancy,” says Edward Wallach, M.D., professor of gynecology and obstetrics at the Johns Hopkins University School of Medicine in Baltimore. “But before we start adding more concern for women undergoing a biopsy, it is important to note that the study had some limitations that are in direct contrast to the way pathologists examine biopsy tissue in real-world clinical practice.”

However, Wallach adds, “The study does identify certain weaknesses in the way specimens are interpreted. Doctors and patients should avail themselves of the opportunity to obtain a second opinion regarding the pathological interpretation to help ensure the accuracy of the diagnosis. The findings of the study also stress the need to develop optimal ways to assess biopsy samples.”

Delving into the research

For the study, investigators invited 115 randomly selected practicing pathologists throughout eight states to diagnose a series of breast biopsy samples. The samples were obtained using either incisional biopsy, in which a sample of breast tissue is surgically removed, or core needle biopsy, which involves using a hollow needle to remove small pieces, or cores, of breast tissue.

Each pathologist received a single glass slide to interpret. The researchers compared the diagnoses of these pathologists with those of a panel of three internationally recognized breast pathologists. The tissue specimens consisted of four types of samples:

■ Invasive breast cancer, which commonly starts in a lobule or a milk duct

Ductal carcinoma in situ (DCIS), an early, noninvasive breast cancer that originates in the milk ducts and increases the risk of invasive cancer

■ Atypical hyperplasia, also called atypia, a noncancerous (benign) condition in which more cells than normal are present in the breast-duct lining and which increases cancer risk

■ Benign cells, which carry no increased risk for cancer

The 115 invited pathologists agreed with the expert panel diagnoses in only 75 percent of cases, broken down as follows:

■ Invasive cancer. The participating pathologists agreed with the expert panel in 96 percent of cases.

■ DCIS. The panel and participating pathologists agreed on a diagnosis in only 84 percent of cases. The participating pathologists over-interpreted 3 percent of DCIS samples as invasive cancer.

■ Atypia. Both groups agreed only 48 percent of the time. Practitioners said atypia was present in 17 percent of samples when it wasn’t, and they missed 35 percent of cases. Atypia represents about 10 percent of breast biopsies done each year.

■ Benign. Both groups concurred that no cancer was present in 87 percent of samples. But practitioners diagnosed DCIS in 13 percent of benign samples, which can lead to unnecessary procedures.

“Diagnoses based on well-defined criteria such as that of invasive breast cancer are more likely to be agreed upon, so logically, this finding makes sense,” says Wallach. “Whereas opinions based on interpreting more subtle signs such as that of atypia will vary based on training and other factors. A pathologist must decide how far from ‘normal’ a cell looks, but there’s a wide range of what’s considered normal.”

What does it mean when samples are under- or overdiagnosed? The consequences can be quite serious: A woman may undergo unnecessary testing, surgery, radiation therapy or hormonal therapy—or she may not get the treatment she needs.

Behind the ambiguities

In the study, women who had dense breast tissue were more likely to have received misdiagnoses. Dense breasts have more glandular and fibrous connective tissue than fatty tissue. (They also make it harder for radiologists to detect cancer on a mammogram.) Having dense breasts has been linked to an increased risk of breast cancer.

Pathologists in the study who interpreted a limited number of cases each week or worked in small practices or nonacademic settings were responsible for more differences in diagnoses than those who had a larger caseload and worked in larger settings. Sixty-eight pathologists practiced in laboratory groups that had 10 or fewer pathologists, 47 practiced in groups of 10 or more, and 29 pathologists were affiliated with an academic medical center.

The number of missed or incorrect diagnoses in reality may be much lower. The researchers placed certain restrictions on the pathologists in the study design, and these restrictions possibly lowered the pathologists’ odds of making an accurate diagnosis.

“In the study, the pathologists weren’t allowed to get a second consultative opinion, a common practice when complex tissue samples are involved,” Wallach says. “They also had to base their diagnoses on a single slide. In the real world, a pathologist reviews multiple slides per case and can also request additional slides or tests before arriving at a final diagnosis.”

Even the three members of the expert panel originally disagreed among themselves about 25 percent of the time. Only after discussion among the three did they reach a consensus of 90 percent.

Finally, the study didn’t report the outcome of the patients whose biopsy samples were examined. So there’s no evidence to tell whether the expert panel’s classifications were actually more accurate than the participating pathologists’ assessments.

What do the findings mean?

Most of the approximately 1.6 million women in the United States who undergo breast biopsies each year have benign results, according to the study. The researchers suggest that there are more women overdiagnosed with DCIS or atypia than there are women whose invasive cancer is missed. The fact that the pathologists agreed with the diagnosis of invasive breast cancer in nearly all cases should be reassuring to women.

If you’re diagnosed with atypia or DCIS, you may want to ask your doctor about seeking a second opinion about the interpretation before you begin to discuss treatment options. But keep in mind that the first pathologist might have already sought other consultative expert opinions. Atypia is sometimes treated with a surgical biopsy to obtain and examine a larger sample of abnormal cells. Your doctor may also suggest chemoprevention (powerful drugs to help prevent cancer). You may want to discuss alternatives to surgery or drugs, such as close monitoring of your condition with regular mammograms and clinical breast exams. DCIS generally requires a lumpectomy or a mastectomy to remove the cancerous tissue, often followed by radiation therapy and/or chemoprevention with tamoxifen. Breast surveillance isn’t an option, but asking for a second opinion is. Because the cancer isn’t invasive, you don’t have to rush into a treatment decision. “It’s important to maintain an open line of communication with your doctor,” Wallach says. “If you feel a lump or see changes in your breasts, err on the side of caution and get it checked out. Conversely, if your doctor says you don’t need a biopsy but you still think something is wrong, don’t hesitate to see another doctor for a second opinion.”

Meet Our Writer

HealthAfter50 was published by the University of California, Berkeley, School of Public Health, providing up-to-date, evidence-based research and expert advice on the prevention, diagnosis, and treatment of a wide range of health conditions affecting adults in middle age and beyond. It was previously part of Remedy Health Media's network of digital and print publications, which also include HealthCentral; HIV/AIDS resources The Body and The Body Pro; the UC Berkeley Wellness Letter; and the Berkeley Wellness website. All content from HA50 merged into in 2018.