Purpose: To evaluate whether positron emission tomography (PET) with carbon-11-methionine (MET) can be used for detection of early response to external beam radiotherapy (RT) in untreated head and neck cancer using locoregional control and survival as study endpoints.
Materials: Fifteen patients with head and neck cancer underwent a MET PET study before RT and after a median dose of 24 Gy. Fractionation was standard (n = 6) or hyperfractionated (n = 9), and 13 out of 15 patients had planned surgery after RT. SUV was calculated for primary tumor (n = 13) or largest lymph node metastasis in two patients of whom one had his primary excised before study enrollment and one presented with unknown primary tumor syndrome.
Methods: Attenuation corrected PET scans acquired 20-40 min from tracer injection were used for evaluation of MET uptake in tumors. A quantitative MET uptake index was expressed as standardized uptake value (SUV) or SUV(lean) (corrected for lean body mass). The PET results were correlated with clinical follow-up data. The median follow-up time is currently 28 months (range 22-34).
Results: A total of 13 primary tumors and 12 metastatic lymph nodes were visually identified in MET PET. In the first PET study the median SUV in tumor was 8.6 (range, 5.5-14.0). In the second PET study performed during RT the median SUV decreased to 5.7 (range, 3.1-8.2, P = 0.001). Two out of 15 patients showed no radiation-induced decrease in SUV. The median tumor SUV ratio of patients remaining in local control (CR) after RT was 0.7 (range 0.6-0.8, n = 6), and that of relapsing patients similarly 0.7 (range 0.5-1.0, n = 9, NS). The SUV ratio was not associated with survival time. The MET uptake of submandibular salivary glands decreased in all patients during the first two or three weeks of RT (P = 0.03).
Conclusions: MET uptake in tumor shows a significant decrease during the first two to three weeks of RT of head and neck cancer. It appears that the rate of decrease in tracer uptake is comparable in relapsing patients and those who remain locally controlled and thus the use of MET PET for prediction of response to RT is limited.