Multiple studies have shown that multiparametric prostate MRI is helpful in detecting, localizing and staging prostate cancer [1-3]. For diagnostic and therapeutic purposes it is important to detect extracapsular tumor extension and get more insights into the tumor metabolism. This could be improved by the higher signal-to-noise-ratio (SNR) at ultra-high field systems like 7 tesla, giving access to higher temporal and/or spatial resolution or improve contrast to noise [4]. This results not only in the opportunity to detect smaller tumors as a result of reduced partial volume effects, but also paves the way for new methods to gain insight into tumor metabolism and aggressiveness, e.g. MR spectroscopy [5, 6].In this study, we demonstrate that a dipole transceive antenna array with loop receive arrays at 7T substantially outperforms state of the art 3T MRI of the prostate. Apart from demonstrating the design of the new setup, we show intrinsic SNR comparisons to the 3T setup obtained from 6 healthy volunteers. Furthermore high-resolution T2weighted images were acquired in one patient with prostate cancer.6 healthy volunteers (age: 24-56, BMI: 21.6-31.8) and one patient (62 years, PSA: 9.7 ng/ml) with biopsy-proven prostate cancer (Gleason score: 3+4 in 9/10 biopsies) were included in this study. All 7 subjects were scanned on both 3T and 7T after informed consent was obtained. For this study a 8-element transceive array consisting of fractioned dipole antennas combined with a detunable 16-element receive only loop coil array was used at 7T (Achieva, Philips Healthcare, Best, NL) [7], Figure 1. At 3T (Achieva, Philips Healthcare, Best, NL) ( a 32 element torso/cardiac coil was used. The healthy volunteer MRI protocol included on both 3T and 7T the following exams: A survey for localization purposes, B1 shim series to optimize B1 levels in the prostate, AFI B1 map [8], T2weighted imaging for anatomical images and a proton density gradient echo acquisition combined with a noise scan for SNR determination. The proton density gradient echo sequence was obtained using TR/TE=100/5 ms, a flip angle of 1º, FOV =250x430x25mm3, voxel size= 2x2x2.5mm3 and reconstruction voxel size = 1x1x1mm3. The echo time (TE) was kept short to exclude T2 effects, while for T1 exclusion the repetition time (TR) was relatively long compared to the very low flip angle used. Moreover, the SNR was normalized to the actual flip angle applied. The patient protocol included a T2weighted acquisition using the following parameters: TR/TE= 5900/100 ms, TSE factor=29, SENSE factor RL=1.5, FOV =200x200x90mm3, voxel size= 0.78x0.78x3 mm3 and reconstruction voxel size = 0.5x0.5x3mm3 on both 3 and 7T. At 7T also high resolution images were acquired using voxel sizes of 0.5x0.5x2mm3 and 0.35x0.35x2mm3 to explore the possibilities that the extra SNR provides.In Figure 2 the SNR maps of all six volunteers are shown for both 3 and 7T. The overview images clearly show the difference in signal homogeneity throughout the pelvis between 3 and 7T. When comparing the zoomed-in maps a difference in SNR in the prostate can clearly be observed. To determine the SNR in the prostate a rectangular ROI was delineated midgland with dimensions as large as the prostate size would allow. At 3 tesla an SNR of 19.8±3.7 was obtained, at 7 tesla an SNR of 41.6±8.6 was measured in the same volunteers. This results in an average gain of intrinsic SNR of 2.1 fold in the prostate, Figure 3. In one prostate cancer patient (62 years, PSA: 9.7 ng/ml, Gleason score: 3+4) T2weighted images were acquired at 7T after the clinical 3T MR examination, both with external coils only. When comparing the clinical 3T examination to the 7T scan with the same resolution, Figure 4, the T2w contrast at 7T shows a clear difference between healthy and tumor areas, furthermore an excellent detail within the prostate gland can be obtained at 7 tesla. Anatomical details and contrast between healthy and tumor areas remain visible even at the resolution of 0.35x0.35x2mm3 at 7 tesla, Figure 5.Using the 24 channel body array at 7T, we demonstrated for the first time the intrinsic SNR benefits of using the higher field strength of 7 tesla for prostate MR imaging compared to state-of-the-art 3T MRI. On average an overall gain in SNR of 2.1 fold was observed between 7T and 3T. Furthermore high resolution imaging was obtained at 7 tesla in one prostate cancer patient, to show the potential imaging resolutions when stepping up to ultra-high field prostate imaging.