Lawrentschuk and Fleshner [1] are to be congratulated on their editorial, which makes the argument for increased research into the role of MRI in the detection of prostate cancer [2] and its potential to guide treatment such as focal therapy [3]. There are several important issues to which we would like to draw attention. (i) Reports on MRI are limited due to the use of either inaccurate or inappropriate reference standards. Transrectal biopsies are inherently inaccurate; they miss cancer in 20–30% of men who have a negative biopsy and they under-sample most areas, but in particular the apex, transition zone and anterior horns of the peripheral zone. The use of radical prostatectomy whole-mount specimens as a reference standard introduces a selection or evaluation bias, as men with low-risk disease, or unfit for surgery, and those with no cancer are not evaluated. We propose that the reference standard to evaluate any imaging method in men with a clinical suspicion of prostate cancer is template transperineal prostate-mapping biopsies, sampling the gland every 5 mm. This has been shown to have a 95% accuracy in detecting clinically significant cancer [4,5].

(ii) The use of targeted biopsies in studies evaluating MRI introduces incorporation bias, by which the test under evaluation influences the reference standard. Designing a study using two biopsy operators might seem attractive, but has problems. If the MRI-directed biopsies are taken after the systematic biopsies there will be gland swelling and distortion, making it difficult to target the area that is suspicious on MRI. If the MRI-directed biopsies are taken first, then the needle tracts might influence where the second operator systematically samples the prostate. To overcome these biases, a randomized design is needed in which men are randomly allocated to one of three arms; (a) standard systematic biopsies; (b) systematic plus MRI-directed biopsies; or (c) MRI-directed biopsies alone (no biopsies if there is no lesion on MRI). The third arm in such a trial would only be ethically justified if the clinical validity of MRI for clinically important cancers was high (negative predictive value 90–95%) in a study that used an accurate and appropriate reference standard.

(iii) Studies using MRI to direct the biopsies in the gland have required a judgement as to which area to direct the needle. Image registration is necessary to allow MRI scans to be superimposed in real-time and dynamically, which takes account of prostate rotation, deformation and swelling. Research is ongoing in this area, but there have been some promising developments [6,7].

(iv)The recent multicentre trial assessing the accuracy of MR spectroscopy in 110 men, with eight independent expert readers, has shown that MR spectroscopy is no better than T2-weighted MRI (accuracy as measured by the area under the receiver operating curve of 58% and 60%, respectively) using whole-mount specimens as the reference standard [8]. It is clear that no single MRI variable is sufficient and the technique of multi-sequence, multi-parametric or multi-functional MRI is necessary. This combines T2-weighted scans with dynamic contrast enhancement, diffusion weighting and spectroscopy, allowing a non-invasive evaluation of anatomy, microvessel density, cellular density and cellular metabolic activity [9]. Such a scan might allow an accurate evaluation of disease volume, Gleason grading and staging.

Lawrentschuk and Fleshner are right in calling for research into this area to be given priority. With over 70 000 biopsies taken in the UK, and >1 million in the USA every year, the healthcare economic evaluation will be crucial. The balance between the initial costs of an MRI scan that can take 30–40 min, and requiring expert uroradiologists to report, needs to be weighed against the potential benefits; fewer men undergoing biopsy, fewer biopsies per man, fewer repeat biopsies, accurate determination of disease burden and staging, better treatment delivery and less harm as a result of treating clinically insignificant cancers. The latter advantage has been particularly emphasised by the recent conflicting evidence on the benefits of screening [10,11].

  • 1
    Lawrentschuk N, Fleshner N. The role of magnetic resonance imaging in targeting prostate cancer in patients with previous negative biopsies and elevated prostate-specific antigen levels. BJU Int 2009; 103: 7303
  • 2
    Ahmed HU, Kirkham A, Arya M, Illing RO, Allen C, Emberton M. Is it time to consider a role for MRI prior to prostate biopsy. Nature CP Onc 2009; 6 Doi 10.1038/nrclinonc.2009.18
  • 3
    Ahmed HU, Pendse D, Illing R, Allen C, Van Der Meulen JH, Emberton M. Will focal therapy become a standard of care for men with localized prostate cancer? Nat Clin Pract Oncol 2007; 4: 63242
  • 4
    Crawford ED, Wilson SS, Torkko KC et al. Clinical staging of prostate cancer: a computer-simulated study of transperineal prostate biopsy. BJU Int 2005; 96: 9991004
  • 5
    Ahmed HU, Barbouti O, Zacharakis E et al. Prostate cancer risk stratification and cancer mapping – template guided transperineal prostate mapping biopsies. AUA Annual Meeting 2008, Abstract 436
  • 6
    Shen F, Shinohara K, Kumar D et al. Three-dimensional sonography with needle tracking: role in diagnosis and treatment of prostate cancer. J Ultrasound Med 2008; 27: 895905
  • 7
    Hu Y, Morgan D, Ahmed HU et al. A statistical motion model based on biomechanical simulations for data fusion during image-guided prostate interventions. Med Image Comput Comput Assist Interv 2008; 11: 73744
  • 8
    Weinreb JC, Blume JD, Coakley FV et al. Prostate cancer: sextant localization at MR imaging and MR spectroscopic imaging before prostatectomy – results of ACRIN Prospective Multi-institutional Clinicopathologic Study. Radiology 2009; 251: 1
  • 9
    Kurhanewicz J, Vigneron D, Carroll P, Coakley F. Multiparametric magnetic resonance imaging in prostate cancer: present and future. Curr Opin Urol 2008; 18: 717
  • 10
    Schröder FH, Hugosson J, Roobol MJ et al. the ERSPC Investigators. Screening and prostate-cancer mortality in a randomized European Study. N Engl J Med 2009; Epub ahead of print PMID: 19297566
  • 11
    Andriole GL, Grubb RL 3rd, Buys SS et al. the PLCO Project Team. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009; Mar 18 Epub ahead of print PMID: 1929756