Perioperative Breast MRI Does Not Lower Recurrence Risk for Early-Stage Patients

Melissa L. Pilewskie, MD

Patients with ductal carcinoma in situ (DCIS) who received an MRI added to mammography before or immediately after receiving a lumpectomy did not experience an improvement in the rate of disease recurrence, according to the results of a large, retrospective study. The data were presented at the 2013 Breast Cancer Symposium.

Although no published clinical guidelines currently exist for the use of MRI in patients with newly diagnosed breast cancer, the screening is often ordered to determine whether a clinician missed any areas of cancer or to determine if there was a discrepancy between a mammogram and a physical exam.

“Theoretically, treating this additional disease found by MRI could result in lower rates of local recurrence or contralateral breast cancer down the road,” said Melissa L. Pilewskie, MD, a breast surgeon at Memorial Sloan-Kettering Cancer Center (MSKCC) in New York City, and first author of the study. “And theoretically, this effect could be most pronounced in women treated with excision alone, meaning just having a lumpectomy and no radiation.”

Since no formal guidelines exist, the use of MRI in newly diagnosed patients varies widely among doctors and hospitals, although a recent survey of US surgeons found that 37% of them routinely use MRI for patients with DCIS. Pilewskie and colleagues hypothesized that this test may be unnecessary since studies have not shown decreased rates of re-excision, with some studies even reporting unnecessary increases in mastectomy rates.

In this study, researchers looked at locoregional recurrence (LRR) rates in women with DCIS and compared the rates in patients who received a perioperative MRI with those who did not. A prospectively maintained database at MSKCC included data on all women who underwent breast-conserving surgery for DCIS between 1997 and 2010, and identified 2321 cases for inclusion in the study. Of those cases, 596 received MRI and 1725 did not. Patients who received MRI tended to be younger, premenopausal, have a family history of breast cancer, have a clinical presentation, receive radiation therapy and endocrine therapy, be treated in later years, and have fewer close or positive margins.

At a median follow-up of 57 months, there were 184 instances of ipsilateral breast tumor recurrence. The 5-year LRR rates were 8.5% in patients who received MRI and 7.2% in patients who did not, and this difference was not statistically significant (P = .52). At 8 years, LRR rates were 14.6% and 10.2%, respectively. When adjusting for age, menopausal status, family history, presentation, adjuvant therapy, margin status, number of excisions, and year of surgery, the researchers did not find that MRI was associated with lower LRR rates. This was also true for patients who received radiation therapy.

Additionally, the researchers looked at contralateral breast cancer event-free rates. At 5 years, both groups had the same event-free rates of 3.5%, and at 8 years, the rate in the group that received MRI was 3.5% compared with 5.1% in patients who did not receive MRI (P = .858).

In the entire cohort, patients who received an MRI did not experience a lower rate of events (hazard ratio = 1.18; 95% CI, 0.79-1.78), whereas patients who received endocrine therapy and had a negative margin status did see improvements in their LRRs, which Pilewskie said was to be expected.

“We did not find an association between perioperative breast MRI and decreased rates of either locoregional recurrence or contralateral breast cancer,” Pilewskie said. “In the absence of evidence that MRI is improving our surgical management or, as we show here, long-term outcomes, the routine use of this test for DCIS should be questioned.”