Aim: Of the present retrospective study was to validate the clinical value of FDG-PET for therapy control of malignant lymphoma.
Method: 72 patients (41 non-Hodgkin lymphomas, 29 Hodgkin's disease, 2 unclassified) received static FDG-PET scans of initially involved regions (n = 53) or of the entire neck and trunk (n = 19) after therapy. CT imaging (n = 70) and serum LDH measurement (n = 64) were also performed. Results were validated either by biopsy (n = 7) or by clinical follow-up (n = 65). The predictive value of PET was analysed in relation to different prognostic factors (stage, recurrence status, number of prior therapy regimen).
Results: PET obtained a sensitivity of 88%, a specificity of 83% and an overall accuracy of 85% for detection of residual disease. The values for CT were 84%, 31% resp. 54%, and for serum LDH 50%, 92% and 73%. The predictive value of PET was related to the prevalence of residual disease. PET predicted complete remission in more than 90% of patients with moderate risk (stage I-III, no relapse, no more than two different therapy regimens). In high risk patients, however, the negative predictive value of PET was 50-67%.
Conclusion: FDG-PET is more accurate than CT imaging and LDH measurement for therapy monitoring of malignant lymphoma. Therapy success can be reliably predicted in patients with moderate risk.