The accuracy of conventional mammography dips dangerously when it comes to finding cancer in dense breasts. Sensitivities of 80% down to 50% and lower have been reported in various studies, depending on the cause of the density. Mammography's blind spot means that women with dense breast tissue are far more likely to have false-positive results from conventional screening mammography, leading to needless biopsies, extra costs, and unfounded worries.
Yet the denser breast parenchyma often found in younger women -- and in those who have undergone hormone replacement therapy or surgery -- can harbor high-grade cancers. This makes the development of adjunctive tools to screening mammography, such as MR mammography and scintimammography, all the more important.
A recent study in the European Journal of Radiology illustrated the point. "Dense parenchymal pattern is associated with high-grade cancers, and is both a risk factor and a reason for impaired screening sensitivity," the authors wrote (July 2001, Vol. 39:1, pp. 50-59).
Among the alternatives to conventional mammography, scintimammography has earned increasing interest in recent years, both from doctors hoping to find the lesions that mammography misses, and from radiopharmaceutical developers eager to "repurpose" existing products for use in the vast mammography market.
To date, most studies have centered on two technetium-based nuclear cardiology agents originally developed for use in the evaluation of ischemic heart disease: 99mTc-2-hexakis 2-methoxyiso-butyl-isonitrile (sestamibi, developed by DuPont as Cardiolite) and 99mTc-1, 2-bis[bis(2-ethoxyethyl)phosphino]ethane (tetrofosmin, developed by Nycomed Amersham as Myoview).
Research on both agents has shown promising, if imperfect, results that indicate a robust ability to detect cancer, especially in women with dense breasts.
Scintimammography vs. mammography
For example, a 1999 Turkish study compared mammography to tetrafosmin scintimammography for the detection of breast cancer and axillary node involvement in 128 women presenting with breast masses. Diagnosis was confirmed with either excision or fine-needle aspiration biopsy (European Journal of Surgery Dec. 1999, Vol. 165:12, pp. 1147-1153).
According to the results, the sensitivity, specificity, and positive and negative predictive value of scintimammography were 95%, 96%, 97%, and 92%, respectively, compared with 87%, 26%, 68% and 52%, respectively, for standard mammography.
More important, scintimammography changed the 34 false-positive mammograms into true negatives, and for detecting axillary lymph node metastases the technique had a sensitivity of 72% and specificity of 100%. The researchers concluded that scintimammography may help avoid unnecessary operations in patients with inconclusive mammography results.
A Taiwanese study last year found similar results in 60 young women with palpable breast masses detected by mammography or physical examination. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of tetrafosmin scintimammography were 93%, 93% 98%, 82%, and 93%, respectively. Mammography clocked in at 84%, 80%, 93%, 63%, and 83%, respectively. The authors found that scintimammography significantly improves the accuracy and differentiation of breast cancer in Taiwanese women (Anticancer Research, May-June 2000; Vol. 20:3B, pp. 2061-2064).
Another group concluded that "the detection of malignant breast tumors by....sestamibi scintimammography was independent of the density of the breast tissue" (Anticancer Research, September-October 2000, Vol. 20:5C, pp. 3755-3758).
Sestamibi vs. tetrofosmin
A few studies have put sestamibi and tetrofosmin head to head. In 1999, Austrian researchers concluded that the two agents had similar diagnostic value in a randomized prospective trial of 101 patients with 103 breast tumors. Both planar and SPECT images were acquired, and the results were compared to histology in all patients (European Journal of Nuclear Medicine, December 1999, Vol.26:12, pp. 1553-1559).
Both agents, 99m Tc-tetrofosmin and 99m Tc-sestamibi, were less sensitive in planar imaging (44% and 46%, respectively), than in SPECT (70% and 69%, respectively). Specificity in planar images was 83% and 87%, respectively, and fell to 70% and 78% respectively in SPECT. The positive predictive value, 76% and 81% respectively, was the same in both imaging methods, and the positive predictive value was higher in SPECT (65% and 67%), than in planar imaging (54% and 57%).
And last month an Israeli study found the two agents to be "accurate and equally efficient for the detection of breast malignancies" (Nuclear Medicine Communications, July 2001, Vol.22:7, pp. 807-811).
In 35 consecutive patients with abnormal mammographic or physical findings, the sensitivity, specificity, positive and negative predictive values, and total accuracy were 89.4%, 80%, 85%, 85.7%, and 85.7%, respectively, for sestamibi. They were 90%, 80%, 85.6%, 85.7%, and 85.7%, respectively, for tetrofosmin.
Multicenter trial
At the 2001 Society of Nuclear Medicine meeting in Toronto, Dr. P. Van Rijk from University Medical Center in Utrecht, the Netherlands, presented the results of the first large multicenter study of tetrofosmin scintimammography in the detection of malignant breast lesions.
In a study funded by Nycomed Amersham of Buckinghamshire, U.K., the group looked at 183 patients, mean age 56.3 years, in 19 centers throughout Europe.
"The primary objective was to assess the efficacy of tetrofosmin in breast patients," Van Rijk said. "Second and perhaps most important for Nycomed Amersham was the success or safety of tetrofosmin when used for breast imaging, because we're all aware it can be used safely in cardiac [patients]."
All of the patients included in the study had palpable masses or breast abnormalities detected in mammography. Beginning 10 minutes after administration of 500-750 Mbq of 99m Tc-tetrofosmin, planar anterior, left, and right lateral images were acquired over 10 minutes each. The location of uptake was recorded by an independent blind-read panel with varying experience levels, and arbitration was used to settle disagreements.
"We divided the patients into three groups: under 50, under 60, and over 60 because there are all of these questions about premenopausal and postmenopausal sensitivity," Van Rijk said. Additional subgroups were based on breast density and breast size.
In all, 180 patients were examined with excision biopsy, core biopsy, or fine-needle aspiration as reference standards, and the results compared with scintimammography findings.
Tetrofosmin's overall sensitivity was 69%, and was not affected by patient age or breast density. The variation in sensitivity among readers was small (kappa values=0.72-0.80). Overall specificity was 64%.
"For the last measure, specificity, you should be aware that there were only 34 patients with benign disease," which impacted the figure negatively, Van Rijk said. But the technique identified an additional 42% of malignant tumors which had been reported negative in mammography.
In all there were only three very minor reactions, including a rash at the injection site (foot) of one woman, according to Van Rijk, who concluded that tetrofosmin is safe for breast imaging.
"There is limited interreader variability, and it's not affected by age or breast density," Van Rijk said. "In summary, it's a useful diagnostic tool," especially in women with dense breasts for whom mammography is inconclusive.
In another presentation, a group from Pusan National University in Korea compared sestamibi scintimammography to ultrasound for the detection of primary breast cancer.
"Nowadays, ultrasound is widely used for the diagnosis of breast masses. However, its ability to find small lesions and microcalcifications is low," lead researcher Dr. S. Kim said, also noting the low specificity reported with ultrasound of the breast.
The study took 182 patients with mammographically suspicious masses and evaluated them with sestamibi scintimammography and ultrasound, using pathologic results or FNA biopsy as a reference standard.
According to the results, 110 of the masses were malignant and 48 were benign. The sensitivity, specificity, and positive and negative predictive values of ultrasound were 69%, 45%, 70%, and 45%, respectively. In comparison, the values for sestamibi scintimammography were 86%, 60%, 83%, and 66%, respectively.
Scintimammography was not only more sensitive and specific than ultrasound for the detection of cancer, it provided more useful information when ultrasound findings were indeterminate, Kim said.
By Eric BarnesAuntMinnie.com staff writer
August 30, 2001
Related Reading
MRI may be best diagnostic option for women at high risk of breast cancer, May 5, 2001
Software pairs MRI and mammography to detect subtle pathology, April 20, 2001
Gene profiling could refine breast imaging, March 22, 2001
Breast MRI has high negative predictive value in high-risk premenopausal women, March 7, 2001
Scintimammography may back up x-ray mammo better than US, February 26, 2001
.Copyright © 2001 AuntMinnie.com
