Lung cancer: reproducibility of quantitative measurements for evaluating 2-[F-18]-fluoro-2-deoxy-D-glucose uptake at PET.

Abstract

To study the precision of repeated 2-[fluorine-18]-fluoro-2-deoxy-D-glucose (FDG) uptake measurements at positron emission tomography (PET) in patients with primary lung cancer. Ten patients with untreated lung cancer underwent two dynamic FDG PET examinations after a 4-hour fast within 1 week. Kinetic modeling of tumor FDG uptake was performed on the basis of a three-compartment model. The tumor concentration of F-18 (standardized uptake value calculated on the basis of predicted lean body mass [SUV-lean]) was also measured 50-60 minutes after injection of a tracer. Blood glucose, insulin, and free fatty acid levels were monitored. SUV-lean and the FDG influx constant Ki were measured with a mean +/- standard deviation difference of 10% +/- 7 and 10% +/- 8, respectively, over repeated PET scans. The mean difference was reduced to 6% +/- 6 and 6% +/- 5 by multiplying SUV-lean and Ki by plasma glucose concentration. SUV-lean and graphical Ki can be measured reproducibly, supporting their use in quantitative FDG PET algorithms.