Combination of baseline metabolic tumour volume and early response on PET/CT improves progression-free survival prediction in DLBCL

Abstract

Background: The study objectives were to assess the prognostic value of quantitative PET and to test whether combining baseline metabolic tumour burden with early PET response could improve predictive power in DLBCL.

Methods: A total of 147 patients with DLBCL underwent FDG-PET/CT scans before and after two cycles of RCHOP. Quantitative parameters including metabolic tumour volume (MTV) and total lesion glycolysis (TLG) were measured, as well as the percentage change in these parameters. Cox regression analysis was used to test the relationship between progression-free survival (PFS) and the study variables. Receiver operator characteristics (ROC) analysis determined the optimal cut-off for quantitative variables, and Kaplan-Meier survival analysis was performed.

Results: The median follow-up was 3.8 years. As MTV and TLG measures correlated strongly, only MTV measures were used for multivariate analysis (MVA). Baseline MTV (MTV-0) was the only statistically significant predictor of PFS on MVA. The optimal cut-off for MTV-0 was 396 cm(3). A model combing MTV-0 and Deauville score (DS) separated the population into three distinct prognostic groups: good (MTV-0 < 400; 5-year PFS > 90 %), intermediate (MTV-0 ≥ 400+ DS1-3; 5-year PFS 58.5 %) and poor (MTV-0 ≥ 400+ DS4-5; 5-year PFS 29.7 %)

Conclusions: MTV-0 is an important prognostic factor in DLBCL. Combining MTV-0 and early PET/CT response improves the predictive power of interim PET and defines a poor-prognosis group in whom most of the events occur.

Keywords: Deauville score; Diffuse large B-cell lymphoma; Metabolic tumour volume; Positron emission tomography; Response assessment.

Fig. 1

Treatment protocol. CMR = complete metabolic response (defined as Deauville score 1–3). Partial response was defined as Deauville score 4–5 with improvement compared to baseline. No response/progression was defined as Deauville score 4–5 with no improvement or progression compared to baseline

Fig. 2

Receiver operator characteristics analysis

Fig. 3

Relationship between initial bulk and MTV-0

Fig. 4

Kaplan–Meier progression-free survival curves. a: IPI groups b: DS c: MTV-0 d: TLG-0 e: SUVmax-2 f: ΔSUVmax

Fig. 5

Prognostic model a: Model combining MTV-0 + DS (PFS). b: Model combining MTV-0 + ΔSUVmax (PFS). c: Model combining MTV-0 + DS (OS). d: Model combining MTV-0 + ΔSUVmax (OS)