Abstract
BACKGROUND: Tissues with low magnetic resonance (MR) signals, such as bones and lungs differ considerably in their attenuation properties, requiring special considerations for attenuation correction. We evaluated the impact of using the five-compartment segmentation model, which incorporates bones, in (68)Ga-PSMA-11 PET/MR studies in patients undergoing evaluation for prostate cancer. METHODS: Prostate cancer patients underwent dedicated prostate (68)Ga-PSMA-11 PET/MR followed by whole-body (68)Ga-PSMA-11 PET/CT. Coronal µmap images of the pelvis derived from four- and five-compartment segmentation models of magnetic resonance attenuation correction (MRAC) were produced. Standardized uptake values (SUV) calculated by the four and five-compartment MRAC models and by computed tomography attenuation correction (CTAC) were compared and correlated in normal prostate tissue, gluteus muscle, sacrum, intra-prostatic lesions and metastases (i.e., bone lesions and involved lymph nodes), and prostatic lesions to gluteus (L/G) ratio. RESULTS: Twenty-six patients (mean age 69.4±9.3 years) were included in the study. Twenty-five patients presented for prostate cancer staging and one patient was evaluated for recurrent disease. There was a statistically significant difference between SUVs of the gluteus, sacrum, prostatic lesions and normal prostate tissue measured by the four-compartment vs. the five-compartment MRAC models, with a medium effect size. Very good to good correlation between SUV measured using the four-compartment MRAC model and SUV measured using the five-compartment model were noted in all lesional and non-lesional areas. Very good to good correlation was noted between four-compartment MRAC and CTAC SUVs of prostatic lesions and L/G ratio and between five-compartment MRAC and CTAC SUVs of prostatic lesions, L/G ratio and metastatic lesions. CONCLUSIONS: (68)Ga-PSMA-11 PET/MR using the five-compartment segmentation model affects SUV measurements in prostate lesions and in the normal prostate and therefore patient follow-up studies must be conducted using the same segmentation model.