Abstract
This study evaluates the predictive value of post-therapy 18-fluoro-deoxyglucose positron emission tomography (FDG-PET), including indeterminate studies, following curative-intent therapy in diffuse large B-cell lymphoma (DLBCL). Consecutive patients from September 2002 to December 2005 were prospectively offered enrollment in an observational registry. Available FDG-PET reports after primary therapy were interpreted by hematologist-oncologists as positive, negative, or indeterminate. One hundred twenty-five patients with DLBCL had a median follow-up of 35.2 months. Ninety-three percent were treated with R-CHOP-like therapy. Twenty percent of PET reports were judged indeterminate. Event-free survival (EFS) at 3 years for the negative and indeterminate groups was 85% and 71%, respectively (p = 0.28 by log-rank). Overall survival (OS) at 3 years for negative, indeterminate, and positive groups was 89%, 88%, and 48%. Combining the pre-therapy International Prognostic Index (IPI) with the post-therapy FDG-PET result added to the predictive value of the study for patients. Three-year EFS for patients with low or low-intermediate IPI risk and an indeterminate FDG-PET report was 93%, while for those with high or high-intermediate pre-therapy IPI the 3-year EFS was 45% (p < 0.02). Interpreting FDG-PET reports following curative-intent chemotherapy in patients is informative but imprecise, and incorporation of pre-therapy prognosis can improve predictive utility.