Non-invasive assessment of inter-and intrapatient variability of integrin expression in metastasized prostate cancer by PET

Purpose: Due to the high expression of the integrin alpha v beta 3 not only on endothelial cells, but also on mature osteoclasts and prostate cancer cells, imaging of osseous metastases with alpha v beta 3-targeted tracers seems promising. However, little is known about the patterns of alpha v beta 3-expression in metastasized prostate cancer lesions in-vivo. Thus we evaluated the uptake of the alpha v beta 3-specific PET tracer [F-18]Galacto-RGD for assessment of bone metastases in prostate cancer patients. Results: [F-18]Galacto-RGD PET identified 58/74 bone-lesions (detection rate of 78.4%) and lymph node metastases in 2/5 patients. The SUVmean was 2.12+/-0.94 (range 0.70-4.38; tumor/blood 1.36+/-0.53; tumor/muscle 2.82+/-1.31) in bone-lesions and 2.21+/-1.18 (range 0.75-3.56) in lymph node metastases. Good visualization and detection of bone metastases was feasible due to a low background activity of the surrounding normal bone tissue. Methods: 12 patients with known metastasized prostate cancer according to conventional staging (including bone-scintigraphy and contrast-enhanced CT; median PSA 68.63 ng/ml, range 3.72-1935) were examined with PET after i.v.-injection of [F-18] Galacto-RGD. Two blinded nuclear-medicine physicians evaluated the PET-scans in consensus concerning lesion detectability. Volumes-of-interest were drawn in the PET-scans over all metastases defined by conventional staging (maximum of 11 lesions/patient), over the left ventricle, liver and muscle and standardized-uptake-values (SUVs) were calculated. Conclusions: Our data show generally elevated uptake of [F-18] Galacto-RGD in bone metastases from prostate cancer with a marked inter-and intrapatient variability. While [F-18] Galacto-RGD PET is inferior to bone scintigraphy for detection of osseous metastases, it might be valuable in patient screening and monitoring of alpha v beta 3-targeted therapies due to the high variability of alpha v beta 3-expression.