ORLANDODynamic magnetic resonance imaging (MRI) can identify
women who have unsuspected multicen-tric/multifocal breast cancer and
are therefore not candidates for breast conservation therapy, Michael
J. Kerin, MD, reported in a plenary presentation at the 52nd Annual
Cancer Symposium of the Society of Surgical Oncology (SSO).
Dr. Kerin demonstrated that images produced by dynamic MRI are better
than those from conventional mammography. Dynamic MRI does not
require compression of the breast, and the resulting images are
particularly sensitive for detecting multicentric or multifocal disease.
Dynamic MRI vs Triple Assessment
Dr. Kerin reported data comparing the sensitivity and specificity of
dynamic MRI with that of conventional triple assessment (clinical,
cytological, and mam-mographic examination) in 334 women with
symptomatic breast disease.
The work was done in collaboration with S. Chatpadhyay and other
researchers at the University of Hull Academic Surgical Unit and the
Centre for MR Investigations, Hull, United Kingdom. This prospective
study was inspired by the fact that mammography examinations miss
about one-third of cases of multifocal breast disease.
All patients underwent triple assessment and dynamic MRI with
gadolinium contrast, using a 1.5 T whole body imager with phased
array bilateral breast coils. Of the 334 patients, 156 had benign
disease. The 178 with primary malignancies included 50 (28%) with
multifocal or multicentric tumors on histologic examination (Table).
Standard mammography did not detect multifocal disease in any
patient and, in fact, only detected multicentric disease in 25% of
patients who had that sort of disease, Dr. Kerin said.
Dynamic MRI detected all instances of multicentric and multifocal
disease but also gave false-positive readings for multifocal disease
in 18 of 178 patients (10%). However, 15 of those 18 did have
unifocal invasive breast cancer detected on conventional triple assessment.
Triple assessment had a very poor sensitivity. By contrast,
dynamic MRI had a sensitivity of 100% with a reduced specificity
because of the false-positive detection of multifocal disease,
Dr. Kerin said.
Because dynamic MRI is expensive and because of the false-positive
problems, Dr. Kerin emphasized that a prospective, randomized,
controlled clinical trial of dynamic MRI vs triple assessment is
needed and that the overall effect on local control and on survival
remains to be determined.
For example, Dr. Kerin suggested that more accurate diagnosis of
multifocal/multicentric disease might reduce the number of patients
who receive breast-conserving therapy but then must undergo
additional surgery to remove residual disease. However, a member of
the audience challenged the clinical relevance of this point by
pointing out that the local recurrence rate in patients treated with
lumpectomy plus radiation therapy is only about 10%.