Time benefit in the assessment of recurrences following fractionated radiotherapy in an experimental tumour system using positron‐emission tomography with18F‐fluorodeoxyglucose

PURPOSE: To determine the sensitivity and specificity of 18F-fluorodeoxyglucose-positron-emission tomography (FDG-PET) in the diagnosis of R1H tumours after fractionated radiotherapy, and the dependency of sensitivity and specificity on time after therapy. In addition, the time benefit of FDG-PET concerning early recognition of recurrences after fractionated radiotherapy was assessed. MATERIAL AND METHODS: Subcutaneously growing rat rhabdomyosarcoma R1H tumours were irradiated by applying total doses of 80 or 85 Gy after reaching a start volume of 0.8 cm3. Twenty animals were treated. Tumour volume was determined twice a week. FDG-PET was performed weekly before, during and for 6 months after therapy using a conventional full-ring whole-body PET scanner. In total, 600 PET results were evaluated qualitatively using a six-scale score. PET results and actual tumour volumes were compared. The sensitivity and specificity of tumour detection by PET was calculated for different times after the onset of therapy. The optimal score for tumour detection and the influence of time after therapy on the quality of PET (time benefit) was evaluated using receiver-operating characteristics. RESULTS: After irradiation, 8/20 tumours (40%) were locally controlled, while 12/20 recurred. In this tumour model, evidence of relapse is assured when a volume of 0.1 cm3 is reached. Sensitivity of tumour diagnosis by PET increases with time, i.e. with the volume of recurrent tumours after the onset of therapy, mounting to > 0.95 after 100 days. Specificities of 0.95-1.0 were determined after therapy, showing no increase with time. Tumour diagnosis by PET is highly accurate when performed 80 days after the start of treatment. On average, tumours were recognized by PET on 31, 62, 74 and 81 days (median) before approaching volumes of 0.2, 0.5, 0.8 or 1.0 cm3, respectively. CONCLUSION: An experimental system was implemented that allows reproducible detection of recurrent R1H tumours after radiotherapy using FDG-PET. The usefulness of PET as a diagnostic test for R1H tumours is very good and a reliable resolution for PET is demonstrated for volumes < 1 cm3. The results indicate that FDG-PET enables early recognition of recurrences after fractionated radiotherapy.