International prospective evaluation of scintimammography with technetium-99m sestamibi: interim results

The purpose of this study is to evaluate the validity and reliability of scintimammography with technetium-99m (99mTc)-labeled sestamibi as an adjunct radiodiagnostic agent in the diagnosis of breast cancer.

This is a multicenter prospective cohort clinical trial that was initiated in June 1999 and will terminate May 2001. All patients with a physical or mammographic finding visiting any of the participating clinics are enrolled in the study. Patients who are younger than 18 years, pregnant, not willing to sign the consent form, or who have undergone an invasive procedure on the breast 30 days or less before scintimammography are excluded. At the time of the first interim report, 633 patients had been recruited from six Canadian centers. To date complete data have been collected for 530 patients.

The mean (SD) age of the patients is 56 (12) years with a range between 24 and 85. There were 122 (23%) of the women in the sample who were postmenopausal and 36 (7%) who were perimenopausal. Breast density was graded as dense for 127 (24%), normal for 260 (50%), and fatty for 138 (26%). Of the 530 women in the study 156 (29%) had a palpable mass at physical examination. The mammographic results were classified as 76 (14%) breast imaging reporting and data system (BIRADS) 5, 60 (11%) BIRADS 4, 70 (13%) BIRADS 3 and the remaining BIRADS 2 or 1. The scintimammography results were positive for 118 (23%) of the patients and negative for 412 (78%). The histopathology showed malignant breast disease for 66 (12.5%) of the patients in the sample. Sensitivity and specificity of scintimammography for the detection of breast cancer was estimated 90.9% and 87.5%, respectively. A positive predictive value (PPV) of 50.8% with a negative predictive value of 98.5% and an accuracy of 87.5% were calculated. The PPV is interpreted as the posttest probability of disease given a positive test result. In this sample the pretest probability of disease would be best estimated by the prevalence of disease, which is equal to 12.5%. A positive scintimammography result would change the estimated probability to 51%, which is equivalent to a 400% change from the pretest value. Therefore a positive scintimammography result significantly increases our ability to predict the presence of malignant disease in this population.

The interim results of the present study suggest that scintimammography with 99mTc-sestamibi is accurate and potentially useful as an adjunct to mammography for the detection of breast cancer.