Initial evaluation of the applicability of PET response criteria in solid tumors (PERCIST 1.0)

Objectives To compare the stability of normal background FDG uptake in the liver in the “test-retest” setting and to determine the applicability of the proposed minimum threshold for metabolically measurable tumor uptake according to PERCIST (Wahl et al. J Nucl Med 2009).

Methods 33 patients (pts) had FDG PET/CT scans pre- and post radioimmunotherapy (RIT) for lymphoma. SUVlean (SUL) mean and standard deviation (SD) were determined in a ~3 cm spherical volume of interest in the right hepatic lobe pre- and post-RIT. Percent and absolute differences in liver SUL were calculated and correlated with differences in uptake time and lean body mass (LBM). SULpeak (maximal 1.2 cm diameter volume ROI) of the most metabolically active tumor was determined and compared to the minimum threshold for measurable tumor metabolic uptake (1.5*liver SUL + 2SD). Statistical analyses were performed with t-tests and Pearson correlation coefficients.

Results The average absolute difference in liver SULmean between scans was 0.19±0.18 and the average percent difference was 14±15%. In 4/33 pts, absolute difference in liver SUL was >0.3 SUL units and percent difference was >20%. In 3 pts, percent differences in liver SUL were >20%, but absolute differences were 0.3 SUL or less. Percent and absolute differences in liver SUL were not correlated with absolute differences in uptake time or LBM between scans. The mean minimum threshold for measurable tumor metabolic activity at baseline was 2.60 ± 0.43 (n=33). Tumor SULpeak was less than the minimal threshold in 5 pts (average tumor SULpeak - 2.21 ± 0.38).

Conclusions Liver SUL was stable in 79% of patients undergoing RIT using both the absolute and percent difference thresholds according to PERCIST and increased to 88% using the absolute difference threshold (<0.3 SUL units) alone. Most patients had “measurable” tumor metabolic activity according to PERCIST, but 15% with low level tumor uptake did not. Further study is needed to optimize response assessments for these patients and to determine if the PERCIST minimum tumor uptake metric can be decreased