Abstract
1123
Objectives: 68Ga-DOTATATE PET/CT: Assessment for the optimum internal reference for improving standardized uptake value (SUV) utilization
Introduction: 68Ga-DOTATATE PET/CT widely replaced 111In-Octreoscan in the assessment of neuroendocrine tumor due to higher sensitivity, detectability, spatial resolution and lower patient radiation. Standardized uptake value (SUV) is a semi quantitative measurement used in PET to assess radiopharmaceutical concentration in tumors. Accurate SUV measurement is critical for tumor characterization and treatment response assessment, yet SUVs are variable between different patients and among different studies of the same patient. Many factors related to patients and/or technique affect SUV measurements. Identifying an internal reference, that is the least variable but change under the influence of the same confounding factors would improve SUV utilization and reliability. The aim of this study is to identify the optimum internal reference among organs commonly shows physiologic uptake in 68Ga-DOTATATE PET/CT.
Methods: This IRB approved and HIPAA compliant study included 315 PET/CT studies in 180 consecutive patients presented to our PET/CT center between September-2018 and May-2019. Demographic data along with patient weight and height were recorded. Volume of Interests (VOI) were placed on liver and spleen to get mean SUV normalized to body weight (SUVbw) and to lean body mass (SUL). SUVs of liver were excluded in patients with extensive liver involvement and spleen SUVs were non-obtainable in patients who underwent splenectomy. Paired Grambsch test was used to assess for equality of variance between variables. Spearman Rank correlation test was used to assess for correlation between the different SUVs and the patient and technique related factors. SUV reproducibility was tested in a subset of patients with multiple TPs PET/CT studies; Bland-Altman plots were constructed for SUV measurements in 2 different time-points for the same patient.
Results: Among the 180 patients included in our study, 57.8% were females and 42.2% males and mean age was 60 + 12.9. Mean SUL variance was significantly lower than mean SUVbw variance in both liver and spleen (p-value < 0.05). At the same time, variance of both mean SUVbw and SUL of liver were significantly lower than spleen (p-value < 0.05). Uptake time was the only variable significantly correlated with all measured SUVs (p-value < 0.005), yet correlation was negative. On the other hand, Patient weight significantly correlated with mean SUVbw for both liver and spleen (p-value < 0.001) while was not significantly correlated with mean SUL.
Conclusions: Lean body mass normalized SUV (SUL) is a more reproducible, less variable and therefore more reliable quantitative measure than body weight normalized SUV (SUVbw). The mean liver SUL is the optimum internal reference to be used in 68Ga-DOTATATTE PET//CT studies giving the liver larger size, modest, less variable and more reproducible measurement.