PT - JOURNAL ARTICLE AU - Katherine Binzel AU - Jun Zhang AU - Nathan Hall AU - Michael Knopp TI - How does a technology upgrade affect PET SUV quantification? DP - 2011 May 01 TA - Journal of Nuclear Medicine PG - 2027--2027 VI - 52 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/52/supplement_1/2027.short 4100 - http://jnm.snmjournals.org/content/52/supplement_1/2027.full SO - J Nucl Med2011 May 01; 52 AB - 2027 Objectives As PET/CT systems become more critical components of clinical trials and biomarker-like readouts, we investigate the effect of a system upgrade on a 4 year old technology PET/CT system. We evaluate the consistency of standardized uptake value (SUV) measurements using both current and past default reconstruction protocols. Methods PET/CT data of 100 patients were collected on a Biograph 64 mCT (Siemens) that has been upgraded from a Biograph 16 scanner. PET raw data were reconstructed twice using both the current vendor recommended TrueX protocol (3 iterations, 24 subsets, 4 mm FWHM Gaussian filter) as well as the previous standard 3D OSEM protocol (2 iterations, 8 subsets, 7 mm FWHM Gaussian filter). Identical and matched regions of interest (ROIs) were placed on target lesions, liver, heart, and bladder of both reconstructed PET data sets. SUVmax was used as the primary measurement unit and compared for a total of 437 ROIs. Results Comparing SUVmax, TrueX gave an overall increase in SUVmax, on average 33%, over OSEM, while lesions alone led to an average increase of 48%, higher than bladder (23%), heart (34%), and liver (21%). Lesions smaller than 10 cm3 showed a greater increase, on average 57%, than larger lesions, on average 30%, for TrueX reconstruction. For normal tissues, no significant change in percent difference occurred as ROI volume increased. As SUVmax increased per ROI, only a small decrease in percent difference occurred. Conclusions Significant inconsistencies in SUV measurements were found between PET data reconstructed using previous standard protocol and newly introduced standard protocol implementations after a major system upgrade. The difference varied with tissue type and lesion size. While improved image quality is always desirable, advances in reconstructions can impact consistent measurement essential for accurate and reproducible therapy response assessment. Research Support OSUMC Imaging Signature Progra