MRI appearance of prostate following transrectal HIFU ablation of localized cancer

To evaluate the accuracy of gadolinium-enhanced MRI in gauging the extent of the tissue damage induced by transrectal high intensity focused ultrasound (HIFU) therapy and to assess how well the results obtained with this modality correlate with histological findings (control biopsies).

Twenty-one patients with biopsy-proven prostate cancer (T1-T2-T3a, Nx, M0) who gave informed consent were included in the protocol. They underwent pre- and postoperative (2-5 days after HIFU treatment) MR examinations. Fifteen patients also underwent a follow-up MR examination 1-5 months postoperatively. MR findings were compared with the results of postoperative transrectal biopsy examinations.

The prostate volume increased after the HIFU session from 43.9+/-18.6 to 52.1+/-21 cm(3) by day 2 (p<0.001). On fat-saturated gadolinium-enhanced T1-weighted images, the treated area appeared as a hypointense zone surrounded by a peripheral rim of enhancement in all patients. A positive correlation (r = 0.75) was found between the volume of the hypointense zone measured at days 2-5 (30+/-11 cm(3), 67% of the posttreatment prostate volume) and the theoretical target volume (22+/-5 cm(3), 61% of the initial prostate volume). MRI showed that the anterior part of the base was not reached by the ultrasound beam. The mean volume of the untreated zone (prostate without any MRI modifications) was 8 cm(3) (range, 0.4-36). No correlation was found between the MRI appearance of the treated area and the intensity of the necrosis or the presence of foci of residual, viable cancer.

Gadolinium-enhanced MRI is an accurate way of determining the extent of tissue damage induced in HIFU ablative therapy, but cannot predict histological results.