Brain glucose metabolism in diffuse large B-cell lymphoma patients as assessed with FDG-PET: impact on outcome and chemotherapy effects

Background: There is a lack of data on the effect of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) therapy on brain glucose metabolism of diffuse large B-cell lymphoma (DLBCL) patients, as measured by 18F-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET). Moreover, the prognostic value of brain glucose metabolism measurements is currently unknown. Purpose: To investigate the use of FDG-PET for measurement of brain glucose metabolism in R-CHOP-treated DLBCL patients, and to assess its prognostic value. Material and Methods: This retrospective study included DLBCL patients who underwent FDG-PET including the brain. FDG-PET metabolic volume products (MVPs) of the entire brain, cerebral cortex, basal ganglia, and cerebellum were measured, before and after R-CHOP therapy. Whole-body total lesion glycolysis (TLG) was also measured. Results: Thirty-eight patients were included, of whom 18 had an appropriate end-of-treatment FDG-PET scan. There were no significant differences (P > 0.199) between pre- and post-treatment brain glucose metabolism metrics. Low basal ganglia MVP was associated with a significantly worse progression-free survival (PFS) and overall survival (OS) (P = 0.020 and P = 0.032), and low cerebellar MVP was associated with a significantly worse OS (P = 0.034). There were nonsignificant very weak correlations between pretreatment brain glucose metabolism metrics and TLG. In the multivariate Cox regression, only the National Comprehensive Cancer Network International Prognostic Index (NCCN-IPI) remained an independent predictor of PFS (hazard ratio 3.787, P = 0.007) and OS (hazard ratio 2.903, P = 0.0345). Conclusion: Brain glucose metabolism was not affected by R-CHOP therapy. Low pretreatment brain glucose metabolism was associated with a worse outcome, but did not surpass the predictive value of the NCCN-IPI.