|Comparison of 18FDG-PET with CT scans in the evaluation of patients with residual and recurrent Hodgkin's lymphoma|
Authors: H. Dittmann, M. Sokler, C. Kollmannsberger, B. M. Dohmen, C. Baumann, A. Kopp, R. Bares, C. D. Claussen, L. Kanz, C. Bokemeyer
Department of Nuclear Medicine, University of Tuebingen, Tuebingen, Germany
The reliable assessment of residual masses after treatment as well as of new lesions suspected for relapse remains a diagnostic problem in patients with Hodgkin's disease (HD). The current study compares the results obtained by CT scan to FDG-PET imaging in a blind analysis with respect to the viability of residual masses and in case of suspected relapse. Between 1/94 and 10/99, 47 comparisons of PET and corresponding CT scans - 26 comparisons in 24 patients with residual tumors and 21 comparisons in 20 patients with suspected relapse of HD - were evaluated by independent reviewers blinded to he results of each other. Patients with primary diagnosis had been treated within trials of the German HD Trial study group. Relapsed patients received intensified salvage chemotherapy regimens. PET was assessed visually and by quantifying glucose uptake (SUV). Changes in size of tumor lesions as well as contrast medium enhancement served as criteria for assessment by CT scans. Results were validated either by histologic examination of a resected mass or biopsy (n=17) or by a clinical follow-up over 6 months following treatment (n=30). In 26 cases with residual lesions FDG-PET showed an increased tracer uptake in 8, 7 of which were true positive (TP) and 1 false positive (FP). Eighteen cases were classified as being negative (no viable HD), 17 true negative (TN) and 1 FN. In the blinded reading of the corresponding CT scans, 10 cases with residual lesions were considered to contain vital lymphoma (2 TP, 8 FP). Sixteen CT scans were classified as negative (10 TP, 6 FN). The resulting sensitivity and specificity of PET were 87.5% and 94.4% in contrast to only 25% and 56% for CT scans. The positive and negative predictive values of PET and CT scans were 87.5% and 94.4% and 20% and 62.5%, respectively. In patients with suspected relapse, sensitivity and positive predictive value for the diagnosis of the relapse were 100% and 86%, respectively, yielding the same results for both methods. FDG-PET performed in HD patients with residual masses appears to offer important additional information regarding the presence of viable HD in these residual lesions. In patients with suspected relapse of HD, FDG-PET seems not to offer any information over CT scans. Using SUVs is not superior to visual assessment of PET alone.