Comparing oxygen-sensitive MRI (BOLD R2*) with oxygen electrode measurements: A pilot study in men with prostate cancer

Purpose: To explore the relationship between oxygen-sensitive Magnetic Resonance Imaging (MRI) and oxygen measurements in prostate cancer. Methods: Nine men underwent MRI examinations followed by needle oxygen measurements of tumor bearing region within prostate gland and five men further consented to biopsy. Median pO2 and hypoxic fraction5mm Hg (HP5) were derived. Biopsies were immunostained for Carbonic Anhydrase IX (CA IX), Hypoxia Inducible Factor-1 (HIF 1) and Glucose Transporter-1 (GLUT 1). Corresponding Regions-of-Interest (ROI) were delineated on T2-weighted (T2w) MRI by two observers. Median R2* was calculated for each ROI. Spearman correlation was calculated between R2* and HP5/pO2. Results: MRI quality evaluation resulted in exclusion of 4/18 ROI due to motion (n=2) and rectal air susceptibility artifact (n=2). Quality of remaining data was validated by concordance of R2* with T2w, indices and with secondary observer R2* (r=0.94, p=0.005). Correlation was observed between R2* and HP5 (r=0.76, p=0.02) and a trend was noted between R2* and pO2 (r=-0.66, p=0.07). GLUT 1 and HIF 1 were expressed in all patients, and CA IX was expressed in one patient with high HP5 (77%) and low pO2 (1.4mm Hg). Conclusions: MRI using R2* quantification is a promising tool for non-invasive imaging of prostate cancer hypoxia.