Dedicated breast PET value to evaluate BI-RADS 4 breast lesions

Purpose: To evaluate the diagnostic value of dedicated breast PET (dbPET) parallel imaging in mammographically or sonographically detected BI-RADS 4 (Breast Imaging Reporting And Data Systems) lesions. Materials and methods: After institutional review board and patient approvals, 50 consecutive women with 60 BI-RADS 4 breast lesions were prospectively included in the study. All patients underwent Magnetic Resonance Imaging (MRI) and dbPET before biopsy and fusion of both MRI and dbPET images was performed to better locate corresponding lesions. Final findings were compared with histological results. Sensitivity and specificity for dbPET were determined along with their respective 95%-confidence intervals. Results: Histopathology examination revealed 18 malignant lesions (7 in situ and 11 invasive carcinomas) and 42 benign entities. The dedicated breast PET reported no evidence of malignancy in 41 patients, 9 of them with histological diagnosis of neoplasm. Besides, dbPET showed increased metabolically activity in 10 benign lesions and in 9 breast cancers. Two invasive carcinomas were located less than 1 cm from the pectoral muscle, which can explain that they were missed by dbPET because they were outside the field of view (FOV). There were other 6 false negative results, which corresponded to a 0.1 cm invasive lobular carcinoma and 5 in situ carcinomas. Sensitivity and specificity of dbPET were 50% and 76%, respectively. Conclusions: Our analysis does not allow the recommendation of dbPET for diagnosis of malignancy in BI-RADS 4 mammographic or US abnormalities, mainly due to its high false-negative rate for the detection of in situ carcinomas (85.7%). However, considering the lesions greater than 0.1 cm and included in the FOV, dbPET depicted all invasive carcinomas.