Évaluation de la réponse thérapeutique des tumeurs solides par TEP-FDG : étude de la reproductibilité des critères EORTC et PERCIST sur une cohorte de patients atteints de cancer de vessie récidivant traités par temserolimus dans un essai de phase II (VESTOR)
Year: 2016
Type: dissertation
Abstract: Introduction: FDG-PET response assessment is a challenge due to published criteria heterogeneity. PERCIST and EORTC criteria were proposed to standardize metabolic response in solid tumors. Our aim was to evaluate criteria agreement in recurrent bladder cancer patients treated with temserolimus in a phase II trial (VESTOR). Material and Method: Among a prospective cohort of 51 patients, 12 from 2 centers underwent at least 2 PET: 1 baseline and 1 before the 3rd injection (31 PETs performed). 2 nuclear physicians determined metabolic response according to PERCIST and EORTC criteria and recorded several PET measures. Mean measure percentage difference and disagreement on response classification between two observers and criteria (with corresponding κ) were obtained. Results: Intra-observer and inter-observer (IRO) variability of PET indices were similar: very weak for SUVbw, SULpeak, and SUVbsa max (0,2-0,5 %), slighty higher for Liver SULmean and MTV segmentation threshold (3,5% to 5%) and clearly higher for FDG uptake extent and liver SULmean standard deviation (10-15%). IRO agreement was better when only the hottest lesion was assessed: 1 target EORTC SUVbsa max κ=0,83; 1 target PERCIST SULpeak κ= 0,67; 5 targets EORTC SUVbsa max κ= 0,62; 5 targets PERCIST SULpeak κ=0,56. The inter-criteria agreement was perfect (κ=1) with PERCIST and EORTC when evaluating only the hottest target, but lower with other ways of assessment: 5 targets EORTC/PERCIST κ=0,44; 5 targets PERCIST SULpeak/PERCIST TLG: κ=0,55; 5 targets EORTC SUVbsa max/EORTC SUVbsa mean κ= 0,77; 5 targets PERCIST SULpeak/1 target PERCIST SULpeak κ= 0,77; 5 targets EORTC SUVbsa max/1 target EORTC SUVbsa max κ= 0,31. Conclusion: IRO agreement was better with the simplest assessment method. Different ways of assessment proposed by PERCIST and the EORTC didn’t lead to strictly identical classifications; one single easy and widely accepted method would be useful for study comparisons and result reproducibility.
Source: HAL
Author: I. Galinier
Institution: Université de Bordeaux
Language: French
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2341278077