Persistent tumor 18F-FDG uptake after a few cycles of polychemotherapy is predictive of treatment failure in non-Hodgkin's lymphoma.

BACKGROUND AND OBJECTIVE Early recognition of the ineffectiveness of chemotherapy could result in lower cumulative drug toxicity and tumor burden at the start of salvage therapy, which might improve clinical outcome. Therefore, we studied the value of (18)F-FDG PET for early evaluation of response in patients with non-Hodgkin's lymphoma (NHL). DESIGN AND METHODS We studied 28 patients by (18)F-FDG PET after a median of 3 cycles of polychemotherapy. The presence or absence of abnormal (18)F-FDG uptake was correlated to clinical outcome (median follow-up: 17.5 months, range 4-47 months). RESULTS Five of 28 patients still had increased (18)F-FDG uptake in one or more sites previously shown to be involved by lymphoma at baseline evaluation. Only one of these five patients entered complete remission (CR), whereas among the 23 patients with negative (18)F-FDG PET studies, two died of toxicity during chemotherapy and all the others entered clinical CR (p<0.00001). All five patients with and 7/21 patients without residual abnormal (18)F-FDG uptake relapsed or reprogressed (positive predictive value for relapse: 100%, negative predictive value: 67%). By Kaplan-Meier analysis, progression-free survival (PFS) at 1 and 2 years was respectively 20+/-18% and 0% for (18)F-FDG PET positive patients and 81+/-9% and 62+/-12% for (18)F-FDG PET negative patients (p=0.0001). Overall survival (OS) at 1 and 2 years was respectively 20+/-18% and 0% for (18)F-FDG PET positive and 87+/-7% and 68+/-11% for (18)F-FDG PET negative patients (p<0.0001). INTERPRETATION AND CONCLUSIONS Persistent tumoral (18)F-FDG uptake after a few cycles of polychemotherapy is predictive of CR, PFS and OS in NHL. Further studies are warranted to determine whether (18)F-FDG PET has a predictive value independent from conventional prognostic factors. However, the sensitivity of qualitative (18)F-FDG PET imaging in identifying patients with a poor outcome was insufficient. Earlier evaluation after only one cycle of chemotherapy and quantitative analysis might increase the sensitivity of 18F-FDG PET is predicting treatment failure.