Send to

Choose Destination
Can Urol Assoc J. 2015 Nov-Dec;9(11-12):E853-8. doi: 10.5489/cuaj.2895. Epub 2015 Dec 14.

Investigating the ability of multiparametric MRI to exclude significant prostate cancer prior to transperineal biopsy.

Author information

Department of Radiology, Champalimaud Foundation, Lisbon, Portugal, and Department of Biochemistry, University of Cambridge and Cancer Research, Cambridge, UK;
Department of Radiology, Addenbrooke's Hospital, Cambridge, UK;; Department of Radiology, Addenbrooke's Hospital, Cambridge, UK;
Department of Urology, Addenbrooke's Hospital, Cambridge, UK;
Cancer Research Unit, Division of Health Sciences, Warwick Medical School, University of Warwick, Coventry, UK;
Department of Radiology, Addenbrooke's Hospital, Cambridge, UK;
Department of Histopathology, Addenbrooke's Hospital.
Department of Radiology, Addenbrooke's Hospital, Cambridge, UK;; University of Cambridge, Cambridge, UK;



We characterized false negative prostate magnetic resonance imaging (MRI) reporting by using histology derived from MRI-transrectal ultrasound (TRUS)-guided transperineal (MTTP) fusion biopsies.


In total, 148 consecutive patients were retrospectively reviewed. Men underwent multiparametric MRI (mpMRI), reported by a consultant/attending radiologist in line with European Society of Urogenital Radiology (ESUR) standards. MTTP biopsy of the lesions was performed according to the Ginsburg recommendations. Cases with an MRI-histology mismatch were identified and underwent a second read by an experienced radiologist. A third review was performed with direct histology comparison to determine a true miss from an MRI-occult cancer. Statistical analysis was performed with McNemar's test.


False negative lesions were identified in 29 MRI examinations (19.6%), with a total of 46 lesions. Most false negative lesions (21/46) were located in the anterior sectors of the prostate. The second read led to a significant decrease of false-negative lesions with 7/29 further studies identified as positive on a patient-by-patient basis (24.1% of studies, p = 0.016) and 11/46 lesions (23.9%; p = 0.001). Of these, 30 lesions following the first read and 23 lesions after the second read were considered significant cancer according to the University College London criteria. However, on direct comparison with histology, most lesions were MRI occult.


We demonstrate that MRI can fail to detect clinically relevant lesions. Improved results were achieved with a second read but despite this, a number of lesions remain MRI-occult. Further advances in imaging are required to reduce false negative results.

Supplemental Content

Full text links

Icon for PubMed Central
Loading ...
Support Center