Early risk stratification for diffuse large B-cell lymphoma integrating interim Deauville score and International Prognostic Index

2019 
The aim of this study was to evaluate the prognostic relevance of early risk stratification in diffuse large B-cell lymphoma (DLBCL) using interim Deauville score on positron emission tomography-computed tomography (PET-CT) scan and baseline International Prognostic Index (IPI). This retrospective study included 220 patients (median age, 64 years; men, 60%) diagnosed with DLBCL between 2007 and 2016 at our institution, treated with rituximab-based chemotherapy. Interim PET-CT was performed after three cycles of immuno-chemotherapy. Interim Deauville score was assessed as 4 or 5 in 49 patients (22.3%), and 94 patients (42.7%) had high-intermediate or high-risk IPI scores. In multivariate analysis, interim Deauville score (1–3 and 4–5) and baseline IPI (low/low-intermediate and high-intermediate/high) were independently associated with progression-free survival (for Deauville score, hazard ratio [HR], 1.00 vs. 2.96 [95% confidence interval (CI), 1.83–4.78], P < 0.001; for IPI, HR, 1.00 vs. 4.84 [95% CI, 2.84–8.24], P < 0.001). We stratified patients into three groups: low-risk (interim Deauville scores 1–3 and low/low-intermediate IPI), intermediate-risk (Deauville scores 1–3 with high-intermediate/high IPI or Deauville scores 4–5 with low/low-intermediate IPI), and high-risk (Deauville scores 4–5 and high-intermediate/high IPI). This early risk stratification showed a strong association with progression-free survival (HR, 1.00 vs. 3.98 [95% CI 2.10–7.54] vs. 13.97 [95% CI 7.02–27.83], P < 0.001). Early risk stratification using interim Deauville score and baseline IPI predicts the risk of disease progression or death in patients with DLBCL. Our results provide guidance with interim PET-driven treatment intensification strategies.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    31
    References
    10
    Citations
    NaN
    KQI
    []