MRI Results After DCIS Biopsy Unreliable, Leads to Overtreatment

Submitted by admin5 on Thu, 05/04/2017 - 13:33

MRI results after percutaneous biopsy for ductal carcinoma in situ (DCIS) often overestimate the extent of breast cancer, resulting in unnecessary mastectomies, according to a presentation at the American Society of Breast Surgeons annual meeting (April 26-30, 2017; abstract 257324).


Related Content

Contralateral prophylactic mastectomy has higher complication rates

Breast-conserving therapy tops mastectomy for early breast cancer


Radiologists at the Mayo Clinic (Phoenix, Arizona) recently believed that there was an issue interpreting MRI findings after needle biopsy; they hypothesized that biopsy-related inflammation could make lesions appear larger on subsequent MRIs. However, the impact of percutaneous biopsy on MRI to accurately depict the degree of disease in patients with DCIS has not be evaluated.

Led by Barbara Pockaj, MD, senior investigator, surgical oncologist, researchers at the Mayo Clinic conducted a study to compare MRIs performed before versus after biopsy for accuracy in predicting degree of DCIS. Researchers reviewed 54 cases of women with mostly high- or intermediate-grade DCIS who underwent either pre-biopsy MRI (n = 16) or post-biopsy MRI (n = 38).

Of the 54 women, 7 (13%) had mastectomies as a result of post-biopsy MRI findings that were not indicated on final pathology.

Of the 38 patients who underwent post-biopsy MRI, 15 (39%) produced limited MRI evaluation for the reading radiologists as a result of the biopsy, which prohibited accurate measurements to be drawn.

Fourteen patients had their surgical approach changed because of their MRI results, 11 of whom (20% of the entire cohort) underwent mastectomy. According to Dr Pockaj, “Three really needed it, and 1 maybe half way,” but the remaining 7 demonstrated on pathology to have only needed lumpectomies.

Upon further investigation, researchers found that mean lesion size on preoperative MRI was 3.6 cm, whereas mean lesion size on pathologic specimen was 1.6 cm. Pre-biopsy MRI strongly correlated with surgical specimen (r = 0.561, P < .001) and post-biopsy MRI did not significantly correlate with the actual size of tumors on surgical specimen (r = 0.028, P = .921).

Researchers call for breast surgeons to be aware that in patients with post-biopsy MRI, tumor size may be different than anticipated, which could influence surgical decision-making. – Zachary Bessette