The new technique, called molecular breast imaging, uses a novel dual-head gamma-camera system and is sensitive enough to detect tumours with a diameter of less than 10 mm in 88% of cases. The investigators claim that the technology should provide a "valuable complementary imaging technique" to mammography.
Mammography uses low-dose X-rays to create images of breast tissue anatomy. But for very dense breasts it can only provide accurate tumour diagnosis in 30-50% of cases. Yet women with dense breasts are four to six times as likely to develop breast cancer. Early findings from an ongoing comparison with mammography showed that molecular breast imaging can find small cancers that are not seen on mammograms.
"Our ultimate goal is to detect small cancers that may be inconspicuous or invisible on a mammogram for high-risk women with dense breasts," said Mayo Clinic physicist Michael O'Connor, who presented the results of this study at the 2006 meeting of the San Antonio Breast Cancer Symposium last week.
Stephen Phillips, Mayo radiologist and coauthor, added: "We hope that our studies will eventually show our device to be almost as sensitive as MRI, which is probably the best diagnostic test available to date but is not widely used because of its expense."
An MRI scan costs up to 10 times as much as a traditional mammogram. The investigators say that their device is likely to be only slightly more expensive to use than mammography and should be much more comfortable for women because less pressure is needed to image a breast.
Tailored design
Conventional gamma cameras cannot be easily adapted for breast imaging. Instead, the investigators used cadmium zinc telluride gamma cameras incorporated into a new dual-head breast-imaging system. These semiconductor-based detectors are compact devices that offer superior energy resolution and spatial resolution. Images obtained with these gamma cameras are not affected by the presence of dense or fatty tissue.
A dual-head gamma-camera system offers some important benefits for breast imaging. The fall-off of spatial resolution with distance means that low-contrast lesions located opposite a single detector could be lost in the noise, whereas a second opposing detector should detect this lost lesion. In addition, recording two facing views provides a means to localize a lesion's depth and estimate its size.
The research team used the dual-head gamma-camera system to scan 100 patients who had suspicious breast lesions with a diameter of 20 mm or less. The women were injected with a small amount of a radioactive drug, sestamibi, which preferentially travels to tumours, which absorb the substance.
During the scan, each breast was lightly compressed between the gamma cameras, with just enough pressure to keep it stationary for 5–10 min while several pictures were taken. "It is much more comfortable for women because a force of only 15 lb is used, compared with the 45 lb force compression needed to take a mammogram," said O'Connor.
The images showed that a small amount of the sestamibi was usually absorbed throughout the breast. But in cancerous areas, the amount of drug absorption was significantly increased. Although some benign conditions, such as fibroadenomas, occasionally absorb the drug and create a false-positive result, the researchers believe that the error rate is less than the 10% found with traditional mammography.
Surgery later identified 82 cancers in 54 patients. Of these, the gamma camera detected 76 cancers, representing a 93% success rate in these cases. O'Connor noted that some lesions were missed because the breast was not properly positioned in the device or because the lesions were too small (2–4 mm) to detect with this technology. The gamma camera was, however, 88% accurate in picking up cancers less than 10 mm in diameter.
The researchers also presented preliminary findings from an ongoing, blinded clinical trial comparing the gamma camera with mammography in 2000 women who came to the Mayo Clinic for routine screening. In the first 250 patients the gamma camera detected four cancers. Three of these were not visible on a mammogram.
"Although these initial results are very exciting, we clearly need to image more patients to confirm the promise of the device," said coauthor Deborah Rhodes of the Mayo Clinic, adding that the team hopes to complete the ongoing trial in the next two years.