Comparison of image-derived and arterial input functions for estimating the rate of glucose metabolism in therapy-monitoring 18F-FDG PET studies.

Abstract

The use of dynamic (18)F-FDG PET to determine changes in tumor metabolism requires tumor and plasma time-activity curves. Because arterial sampling is invasive and laborious, our aim was to validate noninvasive image-derived input functions (IDIFs). We obtained 136 dynamic (18)F-FDG PET scans of 76 oncologic patients. IDIFs were determined using volumes of interest over the left ventricle, ascending aorta, and abdominal aorta. The tumor metabolic rate of glucose (MRGlu) was determined with the Patlak analysis, using arterial plasma time-activity curves and IDIFs. MRGlu using all 3 IDIFs showed a high correlation with MRGlu based on arterial sampling. Comparability between the measures was also high, with the intraclass correlation coefficient being 0.98 (95% confidence interval, 0.97-0.99) for the ascending aorta IDIF, 0.94 (0.92-0.96) for the left ventricle IDIF, and 0.96 (0.93-0.98) for the abdominal aorta IDIF. The use of IDIFs is accurate and simple and represents a clinically viable alternative to arterial blood sampling.