Positron emission tomography in breast cancer: a clinicopathological correlation of results

We conducted a retrospective study to evaluate the sensitivity, specificity and accuracy of positron emission tomography (PET) scans in 109 patients w ith primary recurrent or metastatic breast cancer. All patients had a PET s can, X-ray or CT scan of the chest, an ultrasound or CT scan of the Liver a nd a bone scan. Mammography was available for 86 patients. Correlation betw een the PET scan result and histological findings were made. The sensitivit y, specificity and accuracy of the PET scan were calculated for both the pr imary tumour (T) and lymph nodes (N). In patients with metastasis (M) the a ccuracy of the PET scan was compared with other imaging modalities. Histolo gical results of the site in question were available in only 105 patients. Information for the primary tumour was available for 93 patients and for no des in 74. The PET scan was accurate in 89.2% for (T), with 3.2% false posi tive and 7.6% false negative. For (N) the PET scan was accurate in 90.5% wi th 9.5% false negative. In the 86 patients who underwent both mammography a nd PET scanning, the PET scan was more accurate in 89.5% versus 72% (p=0.00 03). In the 19 patients with metastasis, the PET scan was in agreement with other imaging modalities in 100% of cases. PET scanning is the only non-in vasive imaging procedure that will detect tumours in the breast, lymph node s, lung, liver, bone and bone marrow with high sensitivity, specificity and accuracy. It is a valuable tool in the management of patients in all stage s of breast cancer for diagnosis, staging and following treatment response.