1 x 0.8 x 0.8 cm tubulolobular carcinoma, detected by gamma imaging, negative mammo reading

Dedicated dual-head gamma imaging significantly outperformed mammography in detecting breast cancer among a sample of women with mammographically dense breasts, yielding higher specificity, sensitivity and positive predictive values than mammography, a study published in the November edition of Radiology found.

Mammography is the mainstay of breast cancer screening, producing well-established reductions in mortality for women. Studies have shown, however, that mammographic sensitivity falls drastically in women with extremely high mammographic breast density percentages (greater than 75 percent).

As a result, "[i]t is likely that a proportion of interval cancers are mammographically occult but present at the time of the last screening, and early detection of these cancers by other methods may have a survival benefit," according to Deborah J. Rhodes, MD, and colleagues of Mayo Clinic in Rochester, Minn. Although the use of ultrasound as an adjunct modality with mammography increased screening sensitivity from 50 percent to 77.5 percent in one study of women with mammographically dense breasts, the addition of breast ultrasound also drove up false-positive findings, with the result that the combined mammography-ultrasound screening regiment generated a positive predictive value (PPV) of just 10 percent.

"The purpose of this study was to prospectively and independently compare performance characteristics of dedicated dual-head gamma imaging and mammography in screening women with mammographically dense breasts," Rhodes and colleagues noted. Women with heterogeneously or extremely dense breasts (with mammographic breast densities of 51 percent or higher) aged 25 years or older were enrolled in the study. Only women with at least one risk factor for breast cancer were included, in order to increase the comparative values of cancer detection between mammography and gamma imaging.

Two-view or higher gamma imaging was performed using dual-head systems consisting of two opposing CZT-based detectors. Patients were administered 740 MBq of technetium Tc 99m sestamibi five minutes prior to imaging. Each breast was imaged in both craniocaudal and mediolateral oblique positions for 10 minutes per view. A total of 936 women were screened using dual-head gamma imaging and mammography with the exam performed no more than 21 days before dual-head imaging.

Eleven patients were diagnosed with cancer, of which one case was discovered with mammography and missed by gamma imaging. Seven cancers were discovered with gamma imaging but missed by mammography. Two cases of cancer were observed with both modalities and one case was found only by means of other testing.

Of the eight cancers discovered with gamma imaging, three were invasive lobular carcinoma, two ductal carcinoma in situ (DCIS), two invasive ductal carcinoma and one was tubulolobular carcinoma. The median tumor size was 11 mm, corresponding to the same median value of all tumors discovered in participants.

Dual-head gamma imaging yielded a specificity of 82 percent, compared with 27 percent for mammography. Together, the modalities had a specificity of 91 percent. Gamma imaging was also more sensitive to invasive cancer than mammography, detecting seven of seven invasive cancers as opposed to mammography's detection of two of seven invasive cancers.

Additionally, gamma imaging was a more accurate diagnostic tool, with a PPV of 12 percent, whereas mammography's measured 3 percent. Total recall rates were almost equal, at 8 percent for gamma and 9 percent for mammography. Gamma imaging also produced a slightly superior specificity rate of 93 percent, compared with 91 percent for mammography.

The authors argued that because mammography readers were not blinded to clinical history, whereas gamma readers were, "the specificity of gamma imaging would likely increase with annual incidence screening when results of prior studies would be available for review."

Addressing the dose issue

The authors cautioned that, despite gamma's encouraging findings, the "annual administration of 740 MBq (20 mCi) 99mTc-sestamibi would pose a substantially higher cumulative radiation risk than would mammography."

In light of these concerns, the authors are presently conducting a similar but substantially larger study at Mayo Clinic, screening 2,400 patients with gamma imaging and mammography at a dose five times lower than the present study. "Early results are very similar to this study," indicating that gamma imaging may remain powerful in detecting cancer even with vastly reduced radiation doses, co-author Michael K. O'Connor, PhD, professor of radiologic physics at Mayo Clinic told Health Imaging News.

"Although mammography added little increased diagnostic yield to gamma imaging in this study [3.2 per 1,000 compared to 10.7 per 1,000 for gamma imaging], we consider gamma imaging as an adjunct rather than as an alternative to screening mammography."

Despite this caution, the authors concluded that "[t]he results in our study and in other studies indicate that the sensitivity of gamma imaging is not reduced in women with dense breasts...Gamma imaging has the capability to depict small, node-negative breast cancers that are not detected with mammography."