FDG PET for the routine follow-up in NHL: First prospective evaluation.
G. Jerusalem,R. Silvestre,Y. Beguin,R. Hustinx,M. Fassotte,G. Fillet
DOI: https://doi.org/10.1200/JCO.2006.24.18_SUPPL.7569
IF: 45.3
2006-06-20
Journal of Clinical Oncology
Abstract:7569 Background: We have shown that FDG PET is able to detect preclinical relapses in patients with Hodgkin's disease but the high rate of false-positivity (55%) was a problem (Jerusalem et al, Ann Oncol, 2003; 14 : 123-130). The aim of this study was to determine the accuracy of PET during follow-up after treatment in non-Hodgkin's lymphoma (NHL).
METHODS
Patients were eligible if they had a positive baseline FDG PET and a negative end of treatment work-up including a negative PET study. They underwent FDG PET every 4-6 months until 3 years after diagnosis. Forty-five aggressive NHL (diffuse large B-cell lymphoma : 34, other : 11) and eighteen low grade NHL (follicular lymphoma : 16, other : 2) were recruited prospectively between 5/94 and 12/01.
RESULTS
A total of 197 FDG PET studies were realized. Six patients relapsed (aggressive NHL : 4, low grade NHL : 2) and 3 patients died (aggressive NHL : 2, pancreatic cancer : 1) during the time interval of the study protocol. Four asymptomatic relapses (aggressive NHL : 2, low grade NHL : 2) and one symptomatic relapse (aggressive NHL) was detected by FDG PET. Disease recurrence became symptomatic or clinically detectable a week, 3 months (both aggressive NHL), 12 and 17 months (both low grade NHL) after the first positive PET study. The latter had 4 positive FDG PET studies before biopsy-proven relapse. The time interval between the only false-negative PET and clinical relapse was 5 months in a patient with aggressive NHL. We observed for the whole study population a sensitivity of 89% (8/9), a specificity of 93% (174/188), a positive predictive value of 36% (8/22), a negative predictive value (NPV) of 99% (174/175) and an accuracy of 92% (182/197). In low grade NHL PET had a sensitivity of 100% (5/5), a specificity of 93% (50/54), a PPV of 56% (5/9), a NPV of 100% (50/50) and an accuracy of 93% (55/59). In aggressive NHL, PET had a sensitivity of 75% (3/4), a specificity of 93% (124/134), a PPV 23% (3/13), a NPV of 99% (124/125) and an accuracy of 92% (127/138).
CONCLUSIONS
In aggressive NHL, routine follow-up by FDG PET is not very useful and we stopped it now in unselected aggressive NHL. In contrast, FDG PET can detect a relapse several months before the development of clinical symptoms in low grade NHL suggesting further evaluation of the role of PET in the routine follow-up of these patients. No significant financial relationships to disclose.