[P3-1] The role of magnetic resonance imaging in the breast cancer patient.

Morrow M Memorial Sloan-Kettering Cancer Center, New York, NY

Breast conserving therapy (BCT) is well established. Using standard guidelines, most patients are eligible, and those who are not can be identified with a high degree of accuracy with clinical evaluation and mammography. Local recurrence rates have decreased since the initial trials of BCT. Magnetic resonance imaging (MRI) identifies cancer not found by clinical exam or mammography. Improved definition of the extent of carcinoma in the breast has the potential to improve patient outcomes in the short term by decreasing the number of patients requiring conversion from BCT to mastectomy, or decreasing the number of excisions needed to obtain negative margins. Elimination of patients with extensive disease would be expected to decrease the rate of local recurrence of BCT in the long term.
We examined the effect of MRI on short term outcomes in 577 patients seen between 2004-2006. MRI did not decrease positive margins at initial lumpectomy (21.6% MRI vs 14% no MRI) or conversion from BCT to mastectomy (9.8%MRI vs 5.9% no MRI). MRI was associated with a 19-day delay in surgery, and after controlling for stage, the odds ratio of mastectomy was 1.8 in the MRI group (p= .024). Solin et al (JCO, 2008) examined the effect of MRI on local recurrence in 756 patients. After adjustment for patient and tumor features, 8-year local recurrence rates were 3% and 4% in the MRI and no-MRI groups, and contralateral cancer rates did not differ. A meta-analysis of 2763 breast cancers found that MRI detected additional disease in 16% of cases, a rate significantly higher than the 5-8% 10-year rates of local recurrence observed in most modern studies of BCT. These observations raise important questions regarding unnecessarily radical surgery performed for MRI findings which may not result in improved outcomes.
MRI is of benefit in selected, relatively uncommon circumstances including:
a) Axillary adenopathy with occult primary tumor
b) Paget's disease without evidence of tumor
c) Assessing response to neoadjuvant therapy in patients desiring BCT
d) Major discrepancy among clinical and conventional imaging findings
Conclusion: Breast cancer is well documented by pathologic studies to be a multifocal /centric disease. This occult disease has been successfully treated with RT. MRI is a technology capable of finding some of this disease, but how this benefits patients is unclear. In the absence of this information, MRI should not be a part of routine pre-surgical evaluation.

Saturday, December 13, 2008 9:00 AM

PLENARY III (9:00 AM-9:30 AM)