MR strengthens its role in breast cancer imaging

December 2, 2004

New data are refueling the drive for further utilization of MR imaging in the triage and characterization of breast cancer, according to studies presented Wednesday.

New data are refueling the drive for further utilization of MR imaging in the triage and characterization of breast cancer, according to studies presented Wednesday.

Dr. Constance D. Lehman and colleagues at the University of Washington, Seattle, reported results from the component of the International Breast MRI Consortium (IBMC) Trial that compared MR and screening mammography in women at genetically high risk for breast cancer. After screening 367 women, aged 25 and older, during a 30-month period, the researchers found that MR screening can detect breast cancers missed by clinical examination and x-ray mammography.

Following imaging recommendations, 27 biopsies were performed that revealed four cancers for an overall 1.1% cancer yield. MR imaging detected all the cancers, while mammography found only one. The biopsy recommendation rates and diagnostic yields for MR and mammography were 8.5% to 2.2%, and 1.1% to 0.3%, respectively. Positive predictive values for MR and mammography were 12.9% and 12.5%, respectively.

Investigators at the University of Pennsylvania, on the other hand, presented IBMC findings on breast MR's diagnostic performance. This part of the study included 821 women whose enrollment required biopsy referrals based on American College of Radiology guidelines or suspicious clinical or ultrasound findings.

Although MR's sensitivity proved inferior to that of previous reports, its specificity was relatively high according to Breast Imaging Reporting and Data System (BI-RADS) standards. In addition, the investigators found that MR imaging could predict diagnosis in patients with suspicious lesions identified before biopsy.

MR's sensitivity and specificity were 88.1% and 67.4%, respectively. Breast density, histology, and menopausal status did not significantly compromise MR's performance. MR and mammography's positive predictive values were 72.4% and 52.8%, respectively.

In another study, Dr. Sina Meisamy and colleagues at the University of Minnesota evaluated whether adding in vivo quantitative hydrogen-1 spectroscopy could improve a radiologist's accuracy in reading breast MR images.

The group evaluated 55 patients with pathologically confirmed lesions, using gadolinium-enhanced, fat-suppressed, T1-weighted sequences at 4T and H-1 MRS assessment of choline-containing compounds concentration. They found that adding quantitative H-1 MRS could bolster radiologists' ability to characterize tumors.

Thirty-five of the 55 lesions were invasive carcinomas and 20 were benign. H-1 MRS significantly improved the sensitivity, specificity, and accuracy of four independent readers.

San Francisco researchers assessed the value of measuring the contrast injection rate against the contrast washout of breast malignancies. Twenty-seven women with breast cancers underwent dynamic MR imaging with gadolinium at 1.5T.

The investigators found that faster contrast injection rates resulted in faster blood enhancement washout, confirming previous findings that showed a statistically significant increase in the enhancement rate of breast cancers in relation to the rate of contrast injection.

Although a faster blood washout does not directly translate to faster tumor washout, the delayed phase of tumor enhancement can vary in relation to the contrast injection rate. The injection rate assessment should therefore be included in the pharmacokinetic analysis of breast cancer, the investigators said.