21 Mar Pathologists Often Disagree On Atypia or DCIS in Breast Biopsies
MedicalResearch.com Interview with:
Joann G. Elmore M.D., M.P.H.
Professor of Medicine,
Adjunct Professor of Epidemiology,
University of Washington School of Medicine
Harborview Medical Center
Seattle, WA 98104-2499
MedicalResearch.com: What is the background for this study? What are the main findings?
Dr. Elmore: Our team began studying diagnostic agreement among pathologists while interpreting breast biopsies in 2009. Early findings from the Breast Pathology Study (B-Path) were published in March 2015 in the Journal of the American Medical Association and indicated strong agreement among pathologists when diagnosing invasive breast cancer or benign breast tissue. Agreement, however, was much lower for ductal carcinoma in situ (DCIS) and atypia. Results from this study raised concerns that a high percentage of breast biopsies may be inaccurately diagnosed. These concerns were amplified in the media with statements like “as many as one-in-four biopsies are incorrectly diagnosed.” Statements like this inaccurately depicted the results of our study, which included a test set weighted heavily with DCIS and atypia cases. It is important to consider the percentage that each outcome category contributes to the overall number of biopsies in the U.S. population as we found that the agreement rate of pathologists varies drastically across these diagnostic categories.
In the new work published in Annals of Internal Medicine, we have analyzed the B-Path results to reflect variation among diagnoses of women using U.S. population-adjusted estimates,
In an effort to help physicians and patients better understand what the B-Path results mean for women, we have analyzed the B-Path results to reflect variation among diagnoses of women using U.S. population-adjusted estimates. When adjusted using population-based predictive value estimates, the B-Path results indicate that pathologists’ overall interpretations of breast biopsies would be confirmed by an expert panel 92 out of 100 biopsies, with more of the initial diagnoses over-interpreted rather than under-interpreted.
Of concern, our results noted that among 100 breast biopsies given an initial diagnosis of atypia, less than half of these cases would be given a diagnosis of atypia after review by a panel of three experienced breast pathologists. Over half of the biopsies would be downgraded from atypia to a diagnosis of benign without atypia after review.
MedicalResearch.com: What should clinicians and patients take away from your report?
Dr. Elmore: Overall, pathologists’ initial interpretations of breast biopsy are likely to be confirmed by an expert panel. However, this is markedly less true for initial diagnoses of atypia or DCIS. When confronted with a diagnosis of atypia or DCIS, the patient may want to consider obtaining a second opinion before pursuing treatment options.
MedicalResearch.com: What recommendations do you have for future research as a result of this study?
Dr. Elmore: Pathologists are asked to make a definitive diagnosis. Our research shows that certain diagnostic categories are associated with high levels of uncertainty. Research is needed to assess whether diagnostic uncertainty can be objectively measured and integrated into breast disease management.
MedicalResearch.com: Is there anything else you would like to add?
Dr. Elmore: We care about providing the highest quality medical care. Getting the diagnosis right is an important first step.
MedicalResearch.com: Thank you for your contribution to the MedicalResearch.com community.
[Epub ahead of print 22 March 2016] doi:10.7326/M15-0964
Note: Content is Not intended as medical advice. Please consult your health care provider regarding your specific medical condition and questions.
Joann G. Elmore M.D., M.P.H. (2016). Pathologists Often Disagree On Atypia or DCIS in Breast Biopsies MedicalResearch.com