An international validation study of the prognostic role of interim FDG PET/CT in diffuse large B-cell lymphoma

Objectives The role of interim FDG PET/CT in guiding therapeutic strategies in diffuse large B-cell lymphoma (DLBCL) is debated, mainly because interpretation rules vary among centers. This study aimed to validate the prognostic value of interim FDG PET/CT using the Deauville 2009 consensus criteria.

Methods The international validation study enrolled 120 patients with newly-diagnosed DLBCL, treated with a rituximab-containing regimen. All patients underwent FDG PET/CT both at baseline and after 2 cycles, with no therapy change based on the latter. Scans were interpreted by 3 observers using the same software, and the prognostic value of interim FDG PET/CT was assessed by Kaplan-Meier estimates of event-free survival (EFS).

Results Median follow-up was 39.8 months. Using the Deauville 5-point scale, overall Kappa was better with the reference background set at the liver (score 4, κ=0.65) compared to the mediastinal blood pool (score 3, κ=0.49). Using a 66% SUVmax reduction cutoff led to even better interobserver agreement (κ=0.81). Moreover, the highest prognostic values of interim FDG PET/CT were obtained using either a Deauville score of 5 (2-year estimates of EFS: 78.0% in PET-negative vs. 35.1% in PET-positive patients, P<0.0001) or a 66% SUVmax reduction cutoff (2-year estimates of EFS: 80.4% in PET-negative vs. 40.7% in PET-positive patients, P<0.0001).

Conclusions Although the prognostic value of interim FDG PET/CT is validated using the Deauville scale in DLBCL, computation of SUVmax reduction leads to similar performance and better reproducibility