British Study Confirms the Role of mpMRI Before Biopsy

Terrific news and a great way to start off 2017! On January 17, the prestigious medical journal The Lancet issued online publication of a British study, “Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study” (Ahmed et al., 2017[i]). This large trial validated the importance of multiparametric MRI (mpMRI) before biopsy.

The study, named PROMIS, involved a research team from 11 British medical centers. The team enrolled over 700 men with suspicious PSA who had never had a previous prostate biopsy and were now scheduled for one. The study ran from May 2012 to November 2015; of the original participants, 576 men completed all study requirements. PROMIS had a dual purpose:

  1. Evaluate the accuracy of mpMRI at identifying significant PCa
  2. Determine how many men could safely avoid a biopsy

The rationale for conducting the study lies in the fact that PCa is the only tumor cancer diagnosed by a random, blind biopsy. Men are sent for biopsy based on an abnormal PSA and/or DRE. The problem is that conventional transrectal ultrasound (TRUS) guidance cannot detect exact tumor location, so doctors take 12 cores from a systematic scheme of the gland by inserting a needle randomly into each schematic area in hopes of capturing physical evidence of the disease. Because of this, according to the authors, “…many men without cancer undergo unnecessary biopsies, clinically insignificant cancers are often detected and clinically significant cancers are sometimes missed. TRUS-biopsy also carries significant morbidity [side effects] and can cause life-threatening sepsis [infection with bowel bacteria].”

 Study design

Each participant first had a 1.5T mpMRI scan. Then they underwent a combination biopsy procedure that consisted of a TRUS biopsy followed by a transperineal prostate mapping (TPM) biopsy. The TPM is recognized as the most thorough way to sample the entire gland, short of lab analysis of a radical prostatectomy specimen, so it served as the diagnostic benchmark. The doctors who performed the biopsies did not know the patient’s mpMRI results. The performance of TRUS, TPM and mpMRI would all be statistically compared.

The TPM biopsies determined the following numbers:

  • 40% were found to have significant PCa, while 60% either had no PCa or no significant PCa
  • 10% of participants had Gleason ? 4+3 = 7
  • 38% had Gleason > 3+4 = 7
  • 174 patients could be classified as having significant PCa based on core length even if their Gleason score was 3+3 = 6 or 3+4 = 7.

Just comparing the two biopsy methods, TPM identified 119 significant cases (either Gleason score or core length) that were missed by TRUS, whereas only 13 significant cases were picked up on TRUS but not TPM. Obviously, the more thorough TPM outperformed TRUS in detecting significant PCa.

Performance of mpMRI vs. TRUS

Imaging was compared with TRUS biopsy using four statistical measures:

  • Sensitivity – correctly identifies those with true significant PCa
  • Specificity – correctly identifies those without true significant PCa
  • Positive predictive value – Probability that those with a biopsy positive for significant PCa are truly positive
  • Negative predictive value – Probability that those with a biopsy negative for significant PCa are truly negative

In comparing TRUS vs. mpMRI, the PROMIS study determined these results:

  TRUS mpMRI
Sensitivity 48% 93%
Specificity 96% 41%
Positive predictive value 90% 51%
Negative predictive value 74% 89%

While these numbers may seem confusing, the authors explain that mpMRI was more accurate than TRUS biopsy in terms of correctly identifying the 40% of patients with significant disease and in correctly predicting which patients probably did not.

Implications of the data

The PROMIS study suggests that if men with a suspicious PSA have mpMRI before biopsy, as many as 27% could avoid a biopsy. The diagnosis of insignificant PCa would be reduced by 5%. In addition, if a first biopsy does not detect significant disease, using mpMRI scans to guide future repeat biopsies would increase detection of significant disease by as much as 18% compared with the rate of TRUS misses.

This is good news. The ability to distinguish who really needs a biopsy and who doesn’t, and to increase the accuracy of prostate biopsy using real time mpMRI guidance, is a game changer. It places the detection/diagnosis standard of care in line with other tumor cancers (see it, target it, diagnose it) rather than use the gland as a human pincushion and hope to hit the cancer. Our Center has the added advantage of a more powerful magnet (3T rather than 1.5T) and leading experience in scanning and interpreting the images. We are justly proud that we have been in the vanguard of this positive change. We acknowledge the participants and authors of this study for showing how powerfully real it is.


[i] DOI: http://dx.doi.org/10.1016/S0140-6736(16)32401-1