Background and purpose To evaluate the input of FDG-PET data in the implementation of the involved-node radiotherapy concept and dose painting. Materials and methods Patients with early-stage Hodgkin lymphoma treated with combined modality treatments. First, patients underwent a PET/CT before chemotherapy in the treatment position using a head and shoulder immobilization mask. Second, all patients had a CT simulation for treatment planning. The CT simulation was coregistered with the prechemotherapy CT and FDG-PET scan. All prechemotherapy volumes were superimposed on the CT simulation. The initially involved lymph node areas to be irradiated were delineated on the CT simulation scan. Chemotherapy-induced shrinkage rates of the tumor masses visible on CT scan and on FDG-PET were determined and compared. Results Before chemotherapy, FDG-PET-avid areas represented 25% of the total volume on CT. After chemotherapy, the influence of initial FDG-PET data on the delineation of involved-node radiotherapy fields was significant and was due to the fact that in 36% of the patients, FDG-PET helped pinpoint lymph nodes that were undetected on CT. After chemotherapy, the rates of tumor volume shrinkage on CT and FDG-PET were similar. This finding suggests similar chemosensitivity for FDG-PET-avid and non-avid areas. There was no correlation between initial FDG-PET-avid volumes and the clinical outcome. Conclusion Prechemotherapy FDG-PET data are essential for correctly implementing the involved-node radiotherapy concept but seem to be of minimal value for applying the concept of dose painting.