Abstract
478
Introduction: Early detection of metastatic neuroendocrine disease with 68Ga DOTATATE PET / CT requires high spatial resolution imaging of small lesions, high tumor uptake, low background activity and relatively low background noise. With novel radiotracers, the normal biodistribution and tumor radiotracer uptake may differ greatly compared to standard imaging with 18F FDG PET. We previously characterized signal to noise ratio (S/N) relationships for both 18F and 68Ga using the ACR phantom with standard concentrations for the hot rods and background. We observed hot rod detectability when S/N was greater than 5, in agreement with previous reports. The purpose of this study is to determine if clinically detected lesions in 68Ga DOTATATE PET / CT similarly demonstrate S/N greater than 5, as predicted by phantom studies.
Methods: Clinical patient 68Ga DOTATATE PET / CT studies (n=50) were retrospectively reviewed by an expert nuclear medicine physician who identified tumor lesions to be quantified. These were verified as true positive lesions compared to the clinical reports, and analyzed using a semi-automated software package (MIM software). The normal background region and tumor lesion metrics included tumor uptake SUVmax and SUVmean (mean+1std), size, location, and background noise. The S/N was calculated as the difference in SUVmean between tumor and local background divided by the standard deviation of the background for lesions in hepatic (high background) and extrahepatic sites (low background; no splenic lesions).
Results: Of the 101 lesions, the S/N of the 43 hepatic lesions was very high (16.8+13.2; median 12.0; range 5.4-71.6). Of the 58 extrahepatic lesions, the S/N was also very high (80.5 +70.2; median 61.3; range=10.2-393.7). The S/N of the hepatic metastases was significantly lower compared to the extrahepatic metastases (p<0.0001); however, all lesions showed S/N of greater than 5.
Conclusions: In agreement with phantom studies, the S/N of detectable lesions in clinical 68Ga DOTATATE PET / CT studies was greater than 5 for all lesions. As expected, higher S/N was seen in extrahepatic compared to hepatic lesions. This data supports the use of S/N as a metric for lesion detectability applied to different radiotracers where normal biodistribution and tumor uptake may significantly differ compared to common standards developed for 18F FDG PET.