Breast MRI pays its way in preoperative planning

MRI has great value in guiding treatment of breast cancer patients and is well worth the extra expense when used appropriately. Accurate assessment of the extent of disease prior to breast surgery is essential if inappropriate procedures and repeat conservation surgeries are to be avoided, according to speakers at ECR 2007.

When it comes to the tricky subgroup of invasive lobular carcinoma, research has shown that MRI detects additional disease beyond that discovered by initial imaging assessments. This extra information can sometimes have strong implications for surgical planning. In light of such research, Bradford Royal Infirmary in the U.K. began in 2002 requiring breast MR, in addition to mammography and ultrasound, for all patients who had biopsy-proven cancer and who were set to undergo breast conservation surgery.

"With appropriate case selection, this is a good use of MR, which is a scarce and expensive resource," said Dr. Nisha Sharma, a consultant radiologist at Bradford.

A five-year retrospective review of cases shows the new protocol is paying off for patients. In 26 of 62 histologically confirmed cases of invasive lobular carcinoma, MR results changed treatment plans from wide local excision to mastectomy. Based on the MR study, another five patients underwent successful wide local excision for more extensive disease.

Sharma demonstrated cases in which cancer extent was drastically underestimated by mammography and more accurately characterized with MRI. She also noted that MRI is not perfect. The technique underestimated cancer in two patients. In another case, invasive lobular cancer was palpable but occult on MRI, as it was with other imaging modalities, but proven histologically after the patient elected to undergo mastectomy. One false positive on MR resulted in a mastectomy rather than breast conservation surgery.

"If you do find additional disease that influences patient management, it is important to perform image-guided biopsy for confirmation," Sharma said.

In another study presented during the same session, researchers at Seoul National University College of Medicine in Korea said the role of MRI for assessing ductal carcinoma in situ is controversial but the modality is potentially valuable.

The study, performed between October 2003 and September 2005, involved 72 women, the majority of whom (almost 80%) had dense breasts. Of these cases, 40 were imaged prior to breast conservation surgery and 32 before mastectomies. Researchers compared high-resolution 1.5T MR findings with full-field digital mammography and histological results. Assessments covered breast density, tumor grade, presence of comedo necrosis, microinvasion, and tumor size. MRI picked up 94% of 72 ductal carcinoma in situ lesions, whereas mammography found only 86%. Overall accuracy for MRI was 72% versus 43% for mammography. MRI underestimated size in 17% of cases and overestimated in 11%, whereas mammography underestimated size in 35% of cases and overestimated in 22%.

"MRI can provide more information for preoperative planning, particularly in patients with dense breasts," said Dr. Woo-Kyung Moon, an instructor of radiology at Seoul National University.

In a third study, Italian researchers found diffusion-weighted imaging is better than dynamic contrast-enhanced MRI for measuring breast cancer extension. The study, from the Institute for Cancer Research and Treatment in Torino, involved 110 patients enrolled between 2006 and 2007. Dynamic contrast-enhanced MRI resulted in more false positives and false negatives than diffusion-weighted MRI, according to Dr. Laura Martincich. Performance in characterizing lesions was much better for DWI than for dynamic contrast-enhanced MRI, with an accuracy of 94% versus 72%, sensitivity of 94% versus 76%, and specificity of 50% versus 25%.

Multicenter studies are needed to further evaluate DWI's utility in characterizing malignant lesions, she said.