Dynamic risk assessment based on positron emission tomography scanning in diffuse large B-cell lymphoma: Post-hoc analysis from the PETAL trial

Christine Schmitz, Andreas Hüttmann, Stefan P. Müller, Maher Hanoun, Ronald Boellaard, Marcus Brinkmann, Karl-Heinz Jöckel, Ulrich Dührsen, Jan Rekowski

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Background: Valid prognostic tools are needed to guide risk-adjusted treatment approaches in patients with diffuse large B-cell lymphoma (DLBCL). Methods: We assessed total metabolic tumor volume (TMTV) and standardized uptake value (SUV)-based interim positron emission tomography (iPET) response in 510 patients with DLBCL participating in the positron emission tomography-guided therapy of aggressive non-Hodgkin lymphomas (PETAL) trial (NCT00554164). TMTV was analyzed with a relative (SUV41max) and a fixed thresholding method (SUV4), and iPET was evaluated using the ΔSUVmax procedure. We determined associations between TMTV and international prognostic index (IPI) factors using Welch's t-test, investigated effects of TMTV, iPET response, and the IPI factors on time to progression (TTP), progression-free survival (PFS), and overall survival (OS) by Cox regression, and estimated the outcome using Kaplan–Meier curves. Findings: TMTV was associated with all IPI factors except age. Irrespective of the thresholding method used, TMTV and iPET response were correlated with TTP, PFS, and OS, and remained the only independent outcome predictors in Cox regression analysis. By dichotomizing TMTV (cut-off: 328 cm³ by SUV41max) and iPET response (cut-off: 66% SUVmax reduction), we defined three groups at different risk of treatment failure (low [57.1% of patients]: low TMTV/good iPET response; intermediate [37.8%]: high TMTV/good iPET response or low TMTV/poor iPET response; and high [5.1%]: high TMTV/poor iPET response), with corresponding 2-year probabilities of 93.8% vs. 67.3% vs. 38.5% for TTP, 90.9% vs. 62.5% vs. 29.9% for PFS, and 95.5% vs. 77.4% vs. 37.1% for OS. Interpretation: The PET-based risk model proposed may help identify patients who may benefit from treatment modifications or novel approaches.
Original languageEnglish
Pages (from-to)25-36
Number of pages12
JournalEuropean Journal of Cancer
Volume124
DOIs
Publication statusPublished - 1 Jan 2020

Cite this