| Objectives The aim of the study was to assess the diagnostic performance of fluorine-18-fluorodeoxyglucose domestic dedicated positron emission mammography(18F-FDG PEM) and positron emission tomography(PET) in breast carcinoma patients and correlate PEM/PET SUVmax with histopathological,inmunohistochemical prognostic factors and serum tumor markers. To compare the performance characteristics of FDG PEM and PET with conventional imaging mammography and ultrasound.Methods We enrolled consecutive women with breast lesions who received 18F-FDG PEM/PET examination. All the patients successively underwent PEM and PET by random. All PEM/PET results correlated with follow-up surgical pathology results.The PEM/PET efficency of detecting cancer were calculated and compared with X ray and ultrasonography. The efficency of all imaging were compared between groups based on tumors size. The tumor size and SUVmax between malignant and benign lesions were compared. Two patient groups were classified according to tumor size:tumor size ≤1 cm and tumor size >1 cm, The SUVmax between two groups were compared. The expression of ER, PR, Ki-67 and p53 was evaluated using immunohistochemical analysis as part of the routine pathological assessment at our institution. The levels of serum CA15-3, CA125 were measured. The correlation between FDG PEM/PET SUVmax in breast tumours and the above factors were analysed.Results There were 253 patients accepted biopsy or surgery within 2 weeks after PEM/PET scan. The mean age was 50.2±9.2(range 27-77) years. A total of 283 breast lesions were found. The lesions of 9 patients were out of filed on PEM, and 244 patients were enrolled in the study. The five lsions confirmed by fine needle biopsy colud not be measured on histopathlogy, and finally 278 lesions were enrolled in the study. PET had a sensitivity of 95.5%, specificity of 58.1%, positive predictive value of 91.4%, negative predictive value of 73.5% and accuracy of 88.9% for all breast lesions. PEM had a sensitivity of 96.5%, specificity of 53.5%, positive predictive value of 90.7%, negative predictive value of 76.7% and accuracy of 88.9% for all breast lesions. Ultrasound had a sensitivity of 85.6%, specificity of 79.5%, positive predictive value of 95.2%, negative predictive value of 53.4% and accuracy of 84.5%for all breast lesions. Mammography had a sensitivity of 91.8%, specificity of 85.7%,positive predictive value of 96.8%, negative predictive value of 69.2% and accuracy of 90.8% for all breast lesions. The PEM sensitivity was slightly higher than other imaging, the difference was not statisticlly significant. The concordance rate of PEM and PET in the diagnosis of breast cancer was high, which was 96.7%. PET had a sensitivity of 91.8%, specificity of 53.3%, positive predictive value of 92.8%,negative predictive value of 50.0% and accuracy of 86.7% for tumors>1 cm. PEM had a sensitivity of 93.4%, specificity of 50%, positive predictive value of 93.4%,negative predictive value of 53.6% and accuracy of 87.6% for tumors>1 cm.Ultrasound had a sensitivity of 85.8%, specificity of 73.5%, positive predictive value of 94.3%, negative predictive value of 46.9% and accuracy of 83.9% for tumors >1cm. Mammography had a sensitivity of 89.5%, specificity of 81.8%, positive predictive value of 96.5%, negative predictive value of 55.6% and accuracy of 88.4%for tumors >1 cm. The PEM sensitivity was slightly higher than other imaging, the difference was not statisticlly significant. PET had a sensitivity of 60.0%, specificity of 72.2%, positive predictive value of 75%, negative predictive value of 56.5% and accuracy of 65.1% for tumors ≤1 cm. PEM had a sensitivity of 72.0%, specificity of66.7%, positive predictive value of 75.0%, negative predictive value of 65.3% and accuracy of 69.8% for tumors ≤1 cm. Ultrasound had a sensitivity of 65.2%,specificity of 78.6%, positive predictive value of 88.3%, negative predictive value of52.4% and accuracy of 68.4% for tumors ≤1 cm. Mammography had a sensitivity of70.8%, specificity of 88.9%, positive predictive value of 90%, negative predictive value of 69.6% and accuracy of 78.6% for tumors ≤1 cm. The PEM sensitivity was slightly higher than other imaging, the difference was not statisticlly significant. PET had a sensitivity of 33.3%, specificity of 85.7%, positive predictive value of 66.7%,negative predictive value of 60% and accuracy of 61.5% for tumors ≤5mm. PEM had a sensitivity of 50%, specificity of 71.4%, positive predictive value of 60%, negative predictive value of 62.5% and accuracy of 61.5% for tumors ≤5mm. Ultrasound had a sensitivity of 16.7%, specificity of 85.7%, positive predictive value of 50%, negative predictive value of 54.5% and accuracy of 53.8% for tumors ≤5mm. Mammography had a sensitivity of 33.3%, specificity of 71.4%, positive predictive value of 50%,negative predictive value of 5.6% and accuracy of 53.8% for tumors ≤5mm. The PEM sensitivity was slightly higher than other imaging, but the difference was not statisticlly significant. Among the patients mis-diagnosed by ultrasound, the accuracy of PET was 83.3%, and the accuracy of PEM was 80.6%. Among the patients mis-diagnosed by mammography, the accuracy of PET was 77.3%, and the accuracy of PEM was 72.7%. The detection rate of PET and PEM were compared in this study.Among the 20 patients with multifocal breast cancer, PET detected 3 in 20(15%),while PEM detected 7 in 20(35%). Although it was not statistically significant(p=0.256), but PEM was slightly higher than PET. The mean PEM SUVmax was higher than PET SUVmax. The mean PEM/PET SUVmax was higher in malignant tumors than benign. The mean PEM/PET SUVmax was higher in larger malignant tumors than in smaller one.The PEM/PET SUVmax in ER-negative tumours were higher than in ER-positive breast cancer. A negtive relationship was found between PEM/PET SUVmax and ER status. The PEM/PET SUVmax in PR-negative tumours were higher than in PR-positive breast cancer. A negtive relationship was found between PEM/PET SUVmax and PR status. The PEM/PET SUVmax in p53-negative tumours were lower than in p53-positive breast cancer. A positive relationship was found between PEM/PET SUVmax and p53 status. The PEM/PET SUVmax was significantly higher in carcinomas with high levels of Ki67 than in those with low levels. A positive relationship was found between PEM/PET SUVmax and Ki67. There was no significant difference between PEM/PET SUVmax in high level serum CEA breast cancer and low level CEA. There was no relationship between PEM/PET SUVmax and serum CEA level. There was no significant difference between high level serum CA15-3 and low level CA15-3. There was no relationship between PEM/PET SUVmax and serum CA15-3 level. Breast tumors with higher histologic grade had significantly higher PEM/PET FDG uptake values than did those with lower grade,respectively. There was no relationship between PEM/PET SUVmax and histological grade. There was no relationship between PEM/PET SUVmax and tumor size.PEM/PET SUV was significantly higher in infiltrating ductal carcinomas than in non-infiltrating ductal carcinomas. FDG PEM/PET had a higher sensitivity for infiltrating ductal carcinomas than non-infiltrating ductal carcinomas, the difference was not statisticlly significant.Conclusion FDG PEM/PET has high diagnostic accuracy for breast lesions. The concordance rate of PEM and PET in the diagnosis of breast cancer was high. The lesions out of the field of view on PEM resulted in false negative diagnosis, which was caused by the design of PEM and should be improved. PEM was slightly superior to PET in the small lesions and the multiple lesions, more cases are needed to confirm it. The PEM/PET SUVmax is a preoperative and non-invasivemetabolic factor that correlates to some prognostic factors in breast cancer. There were no significant difference between diagnostic efficency of PEM/PET and conventional imaging. PEM/PET has a high accuracy in patients mis-diagnosed by conventional imaging. The diagnosed accuracy could be improved by combing PEM/PET with conventional imaging. PEM/PET is a good alternative to conventional imaging as a presurgical breast imaging option. |
PDF Full Text Request