PT - JOURNAL ARTICLE AU - Sylvain Beaulieu AU - Paul Kinahan AU - Jeffrey Tseng AU - Lisa K. Dunnwald AU - Erin K. Schubert AU - Pam Pham AU - Barbara Lewellen AU - David A. Mankoff TI - SUV Varies with Time After Injection in <sup>18</sup>F-FDG PET of Breast Cancer: Characterization and Method to Adjust for Time Differences DP - 2003 Jul 01 TA - Journal of Nuclear Medicine PG - 1044--1050 VI - 44 IP - 7 4099 - http://jnm.snmjournals.org/content/44/7/1044.short 4100 - http://jnm.snmjournals.org/content/44/7/1044.full SO - J Nucl Med2003 Jul 01; 44 AB - The purpose of this study was to measure how 18F-FDG PET standardized uptake values (SUVs) change over time in breast cancer and to examine the feasibility of a method to adjust for modest variations in the time of uptake measurement experienced in clinical practice. Methods:18F-FDG PET was performed as 60-min dynamic imaging with an additional image acquired at ∼75 min after injection. For 20 newly diagnosed, untreated, locally advanced breast cancer patients, both the maximum SUV and the average SUV within the lesion were calculated with and without correction for blood glucose concentration. A linear regression analysis of the portion of the time–activity curves starting at 27 min after injection was used to estimate the rate of SUV change per minute during the interval from 27 to 75 min. The rate of SUV change with time was compared with the instantaneous SUV obtained at different times from 27 to 75 min. Results: In untreated breast cancer, 18F-FDG SUV values changed approximately linearly after 27 min at a rate ranging from −0.02 to 0.15 per minute. In addition, the rate of SUV change was linearly correlated with the instantaneous SUV measured at different times after injection (r2 ranged from 0.82 to 0.94; P &lt; 0.001). Using this information, an empirical linear model of SUV variation with time from injection to uptake measurement was formulated. The comparison method was then applied prospectively to a second set of 20 locally advanced breast cancer lesions not included in the initial analysis. The average percent error using the method to adjust for time differences was 8% and 5% for maximum SUVs and average SUVs ranging from 2 to 12. Conclusion: In untreated breast cancer, the SUV at any time point approximately predicts the rate of change of SUV over time. A comparison method based on this finding appears feasible and may improve the usefulness of the SUV by providing a means of comparing SUV acquired at different times after injection.