Intended for healthcare professionals

Research News

MRI scans could help one in four men with suspected prostate cancer avoid biopsy

BMJ 2017; 356 doi: (Published 20 January 2017) Cite this as: BMJ 2017;356:j329
  1. Susan Mayor
  1. London

Using multiparametric magnetic resonance imaging (MRI) in men with raised prostate specific antigen (PSA) levels could enable one in four to avoid having a biopsy and could reduce overdiagnosis of prostate cancer, a study validating this scanning technique against the most accurate reference test has shown.1

Men presenting with an elevated PSA level currently undergo transrectal ultrasound guided biopsy to check for prostate cancer. However, the PSA test is not always indicative of prostate cancer, so some of the 100 000 prostate biopsies in the UK each year are unnecessary. This biopsy technique often detects clinically insignificant cancers and can miss aggressive cancers.

Multiparametric MRI provides information on the size of any cancer detected in addition to its cellularity and vascularity, so it could help to differentiate between aggressive and harmless cancers that would not become clinically significant in a patient’s lifetime. The new study tested the diagnostic accuracy of this type of MRI scan and transrectal ultrasound guided biopsy against transperineal template prostate mapping biopsy, which takes multiple tissue samples across the entire prostate, as a reference standard.

Researchers investigated 576 men with elevated serum PSA levels, suspicious digital rectal examination, or family history (from a total of 740 men recruited at 11 NHS hospitals in the UK), using multiparametric MRI scans. The men then underwent transrectal ultrasound guided biopsy and prostate mapping biopsy.

The results, published in the Lancet, showed that prostate mapping biopsy detected cancer in 71% (408) of the 576 men investigated, and 40% (230) of those with cancer had clinically significant disease.

Multiparametric MRI was significantly more sensitive than transrectal ultrasound guided biopsy in correctly identifying clinically significant prostate cancer, having a sensitivity of 93% (95% confidence interval 88% to 96%), compared with 48% (42% to 55%) (P<0.001).

MRI was also more accurate in identifying men who had no cancer or a harmless cancer, having a specificity of 41% (36% to 46%), compared with 96% (94% to 98%) using ultrasound guided biopsy (P<0.001).

“Using multiparametric MRI to triage men might allow 27% of patients to avoid a primary biopsy and diagnosis of 5% fewer clinically insignificant cancers,” said the researchers, led by Hashim Ahmed, Medical Research Council clinician scientist and consultant urological surgeon at University College London.

If subsequent transrectal ultrasound directed biopsies were directed by multiparametric MRI findings, the research group said, an estimated 18% more cases of clinically significant prostate cancer might be detected than with the current standard pathway of starting with this type of biopsy in all patients.

“Multiparametric MRI should be used as a triage test before prostate biopsy in men who present with an elevated serum PSA,” the authors concluded, adding that a cost effectiveness analysis of their data using this approach was now under way.