Prospective comparison of a 1.5T fast magnetic resonance imaging (MRI) protocol and the 3T multi-parametric MRI ESUR protocol as triage test for men with an elevated PSA

INTRODUCTION & OBJECTIVES: The PROMIS study has confirmed that multi-parametric MRI (mpMRI) is an efficient triage test to select amongst men with an elevated PSA those requiring TRUS guided prostate biopsy (HU Ahmed, JClinOncol 34, 2016 a5000). Standard mpMRI is performed using a 3.0T magnet and includes T2W, DWI, and DCE sequences. This requires IV infusion and magnet occupation time of ± 45 min. A widespread use of prostate MRI as an initial triage test will require major streamlining of the procedure. He we report the interim results of a prospective trial comparing the diagnostic performance of a Fast MRI (fMRI) comprising T2 and DWI sequence only, performed without endorectal coil, on a 1.5T magnet in <15 min to standard 3,0T mpMRI. MATERIAL & METHODS: The study includes patients referred for an elevated PSA with a risk of prostate cancer ≥ 10% based on SWOP prostate cancer risk calculator 4 of the Prostate Cancer Research Foundation (Rotterdam, NL). After providing inform consent, patients have undergone on the same day the reference mpMRI and the fMRI. PI-RADS v. 2.0 were calculated by 2 experienced senior radiologists. Patients with PI-RADS lesions ≥3/5 on the mpMRI have undergone a random ultrasoundguided 12-core biopsy and a MR/US fusion-guided prostate biopsy (Urostation, Koelis, France). Patients without significant lesions underwent urological follow up. RESULTS: 90 consecutive patients were enrolled between 31/12/2014 and 20/10/2016; 82 have successfully completed the trials tests resulting in readable fMRI and mpMRI. MpMRI showed 73 PiRAD ≥ 3 areas in 48 (58%) of the patients; these were subsequently biopsied. On a patient based analysis, only 2 discordant results were reported: 1 fMRI+/mpMRI- and 1 fMRI/mpMRI+. Using mpMRI as Gold Standard, sensitivity and specificity of fMRI is 0.97 (95%CI 0.87-0.99) and 0.97 (95%CI 0.82 and 0.99); the positive and negative predictive values are 0.97 (95%CI 0.87-0.99) and 0.97 (95%CI 0.82-0.99), respectively. On a lesion based analysis, only 2 discordant results were reported: 1 fMRI+/mpMRI- and 1 fMRI/mpMRI+. Using mpMRI as Gold Standard, sensitivity and specificity of fMRI is 0.98 (95%CI 0.92-0.99) and 0.97 (95%CI 0.92 and 0.99); the positive and negative predictive values are 0.98 (95%CI 0.87-0.99) and 0.97 (95%CI 0.82-0.99), respectively. Fourteen lesions PI-RADS 3 on f MRI were graded PI-RADS 4 on mpMRI because of DCE-MRI was positive. One fMRI lesion was misclassified PI-RADS 2 and scored 3 on mpMRI, interestingly corresponding to a Gleason 3+3 foci on the biopsy. On a lesion base analysis, there is an excellent correlation between PIRADS score on mpMRI and fMRI (p for CHI2: 0,001, Kendall's tau-b 0.792, p: 0.000) (Table 1). CONCLUSIONS: When triaging men with high risk of prostate cancer, a fast MRI protocol (1.5 T magnet, T2 + DWI, < 15 minutes) can safety replace the traditional ESUR 3T mpMRI protocol, saving injection and reducing magnet occupation time.