A biopsy of a breast lump or a suspicious spot on an ultrasound has long been considered the way to get definitive answers in diagnosing breast cancer. This is still true for cases where the biopsy definitely shows cancerous cells, but biopsies appear to be a bit less reliable at identifying abnormalities that are more subtle, according to a new study.
The uncertain results tend to occur in biopsies that find cells that are abnormal, but are not cancerous. These would include cells that a pathologist considers to be atypia or ductal carcinoma in situ. Both conditions can progress to cancer, but are not cancerous yet.
The study asked 115 pathologists in eight states to examine a set of 60 biopsy slides collected from 240 women. The researchers then compared these diagnoses with the diagnosis given on each slide by a panel of three experts in breast pathology. They found that there was 96% agreement in cases of invasive cancer and 87% agreement for completely benign tissue, but 13% of benign tissue was diagnosed as abnormal.
In ductal carcinoma in situ, the agreement was 84%, with 13% of cases being missed. In atypia, the rate of agreement was only 48%, with 17% of the pathologists finding it where the panel of experts did not, and missing it in 35% of cases where the panel diagnosed it.
The study was published in JAMA: The Journal of the American Medical Association. It was accompanied by an editorial that called the findings disconcerting. The editorial recommended that women get a second opinion if biopsy results are ambiguous. In a second opinion, a different pathologist looks at the same slides that the first pathologist examined.
Although the findings of the study are of concern, it should be noted that the pathologists in the study were given only one slide to view for each case. In actual practice, pathologists usually ask for more slides to view and consult with colleagues.