18F-FET PET compared with 18F-FDG PET and CT in patients with head and neck cancer.

Recent studies suggest a somewhat selective uptake of O-(2-18F]fluoroethyl)-L-tyrosine (FET) in cerebral gliomas and in squamous cell carcinoma (SCC) and a good distinction between tumor and inflammation. The aim of this study was to investigate the diagnostic potential of 18F-FET PET in patients with SCC of the head and neck region by comparing that tracer with 18F-FDG PET and CT. METHODS: Twenty-one patients with suspected head and neck tumors underwent 18F-FET PET, 18F-FDG PET, and CT within 1 wk before operation. After coregistration, the images were evaluated by 3 independent observers and an ROC analysis was performed, with the histopathologic result used as a reference. Furthermore, the maximum standardized uptake values (SUVs) in the lesions were determined. RESULTS: In 18 of 21 patients, histologic examination revealed SCC, and in 2 of these patients, a second SCC tumor was found at a different anatomic site. In 3 of 21 patients, inflammatory tissue and no tumor were identified. Eighteen of 20 SCC tumors were positive for both 18F-FDG uptake and 18F-FET uptake, one 0.3-cm SCC tumor was detected neither with 18F-FDG PET nor with 18F-FET PET, and one 0.7-cm SCC tumor in a 4.3-cm ulcer was overestimated as a 4-cm tumor on 18F-FDG PET and missed on 18F-FET PET. Inflammatory tissue was positive for 18F-FDG uptake (SUV, 3.7-4.7) but negative for 18F-FET uptake (SUV, 1.3-1.6). The SUVs of 18F-FDG in SCC were significantly higher (13.0 +/- 9.3) than those of 18F-FET (4.4 +/- 2.2). The ROC analysis showed significantly superior detection of SCC with (18)F-FET PET or 18F-FDG PET than with CT. No significant difference (P = 0.71) was found between 18F-FDG PET and 18F-FET PET. The sensitivity of 18F-FDG PET was 93%, specificity was 79%, and accuracy was 83%. 18F-FET PET yielded a lower sensitivity of 75% but a substantially higher specificity of 95% (accuracy, 90%). CONCLUSION: 18F-FET may not replace 18F-FDG in the PET diagnostics of head and neck cancer but may be a helpful additional tool in selected patients, because 18F-FET PET might better differentiate tumor tissue from inflammatory tissue. The sensitivity of 18F-FET PET in SCC, however, was inferior to that of 18F-FDG PET because of lower SUVs.