CHICAGO-Magnetic resonance imaging (MRI) was able to find tumors in two thirds of 47 patients with stage II or III occult primary breast cancer with a high degree of sensitivity, avoiding mastectomy in nearly half of the women, Elizabeth Morris, MD, reported at the 85th Annual Meeting of the Radiological Society of North America (RSNA).
CHICAGOMagnetic resonance imaging (MRI) was able to find tumors in two thirds of 47 patients with stage II or III occult primary breast cancer with a high degree of sensitivity, avoiding mastectomy in nearly half of the women, Elizabeth Morris, MD, reported at the 85th Annual Meeting of the Radiological Society of North America (RSNA).
Occult primary breast cancer is relatively rare, accounting for less than 1% of all breast cancers. It is defined as carcinoma that has metastasized to the axilla or other sites of the body in the absence of mammographic or physical findings of disease in the breast.
Patients with stage II or III occult disease (localized metastatic adenocarcinoma confined to the axilla) may be treated with ipsilateral radiotherapy. However, many institutions have chosen mastectomy as first-line therapy because of concern about high recurrence rates with radiotherapy, Dr. Morris said. Patients with stage IV occult disease (distal metastatic adenocarcinoma) typically are treated with chemotherapy.
In this series of patients who had been followed over a period of 4 years, breast MRI identified primary breast tumors in 30 of 47 women (64%) with stage II or III occult disease. Surgery or clinical follow-up confirmed the absence of breast cancer in 15 patients. There was one false-positive and one false-negative result.
Thus, MR breast imaging yielded a sensitivity of 97%, specificity of 73%, positive predictive value of 84%, and negative predictive value of 95% when used in women with stage II/III occult primary breast cancer, said Dr. Morris, associate attending radiologist, Memorial Sloan-Kettering Cancer Center.
Breast MRI altered the surgical management of 21 of these patients (45%), allowing 9 women (19%) to have lumpectomy rather than mastectomy and 12 (26%) to have no surgery at all.
Even more women with stage II/III occult primary breast cancer may be spared mastectomy on the basis of breast MRI, Dr. Morris said. She explained that the majority of the women who had mastectomy were early in our experience, and both the surgeons and the radiologists were not confident enough in our results to recommend a change from traditional therapy, which is mastectomy. Now, however, if a patient has a negative MRI scan, we treat with radiotherapy only.
The majority of women undergoing lumpectomy in this study had tumors that were undetectable by standard physical examination but could be palpated by the surgeon during resection. Once we were able to identify where the lesion was on MRI, the surgeons went back and examined the patient, and they were able to palpate something they felt comfortable about removing in the operating room, Dr. Morris said.
Five patients who underwent mastectomy had multicentric disease. One woman had a primary tumor with a spiculated mass and extensive ductal carcinoma in situ (DCIS) throughout the remainder of the breast. Another patient had metastatic involvement in the axillary lymph nodes and several ill-defined masses that represented multifocal invasive cancer as well as posterior enhancement that reflected DCIS.
Results in Stage IV Disease
Breast MRI was less successful in finding a primary tumor in women with stage IV occult breast cancer; it identified a malignancy in only 1 of 12 stage IV patients (8%). MRI was negative in 5 patients, confirmed on follow-up. There were six false-positive findings. Thus, MRI had a sensitivity of 100%, specificity of 45%, positive predictive value of 14%, and negative predictive value of 100% for detecting stage IV disease.
Because of the six false-positive findings, which necessitated percutaneous core or aspiration biopsy, Dr. Morris concluded that it is uncertain for stage IV patients whether MRI may be valuable in this setting. We need to evaluate this with future studies with larger numbers.
Breast MRI examinations in this study were done with a 1.5 T Signa GE magnet using a dedicated breast coil. Fat-suppressed T1-weighted images were acquired with a three-dimensional gradient-echo pulse sequence before and after gadolinium-DTPA was administered at a rate of 0.1 mmol/kg.