Is it time to consider biparametric MRI (bpMRI) in the diagnosis of significant prostate cancer ?
Han et al. report the results of a retrospective single center series of 123 patients with suspicion of prostate cancer, a prostate specific antigen (PSA) serum level of 4-10 ng/ml, who underwent mpMRI, subsequent cognitive fusion TRUS guided biopsies and post-biopsy follow-up of more than one year (1). The authors compared the diagnostic performances of mpMRI to that of bpMRI in the diagnosis of significant cancer (any Gleason grade 4 component on biopsy cores) and evaluated the added value of PSA density, in particular in PI-RADS 3 lesions.
The article wishes to demonstrate that no significant added value can be expected from a Gadolinium IV injection. The value under the curve (AUC) of bpMRI (0.884) is significantly better, although only slightly higher, than that of the mpMRI (0.867). The specificity of bpMRI is higher than that of mpMRI, as the use of gadolinium in 13 PI-RADS score 3 lesions injection upgrades to 3+1 (equivalent to 4 in the PI-RADS classification) 8 lesions (21%), corresponding to focal prostatitis in 7 of them (figure 1). Only one significant tumor was therefore missed (3%), which does not affect the sensitivity of bpMRI. The remaining five PI-RADS 3 lesions did not enhance and a cancer was detected in one case (20%). The authors found an added value of the PSAD in these patients, the most important benefit being observed with the bpMRI+PSAD model. However, no threshold is available in the article to adjust the biopsy decision making. It is likely that a larger number of patients would be needed to find a PSAD cut-off value to decide whether to do an immediate biopsy or follow-up.
It can be noted that the high b-value used by the authors in the DWI sequence is 1400s/mm². The authors do not mention the rate of agreement between the two readers to differentiate, using DW images of the bpMRI protocol, a score 3 from a score 4. With a high b-value of 1400s/mm², it is often difficult to visually differentiate a score 3 (mildly hyperintense) from a score 4 (markedly hyperintense), resulting in an established low inter-reader agreement (2, 3). Higher computed b-values show a progressive extinction of the signal intensity of the benign prostate (3), and differentiating a “mild” from a “marked” signal intensity on DW images increases in accuracy with a significant improvement of the inter-reader agreement (2). A b value of 2000 or 2500 s/mm² seems now more consensual (4) and higher b-values can be useful in some patients (3000, 4000 or even more) (2, 5). When available, a simultaneous evaluation of an array of very high computed b-values, rather than a single computed b-value, allows for a better evaluation of the background signal suppression and conspicuity of prostatic lesions, without requiring any acquisition time (3, 6).
The added value of contrast injection can thus only be evaluated once a more reproducible definition of the category 3 of the PI-RADS is obtained. Only a high quality of the diffusion sequence can achieve this goal and it is therefore desirable that radiologists follow the current recommendations of the PI-RADS committee (7) : good rectal preparation to avoid as much as possible the presence of rectal air and the use of recommended parameters for the acquisition of DW images (i.e., three b-values without exceeding b-1000 s/mm² for the acquired multi-b sequence) to benefit from the complementary diagnostic value of ADC metrics (2). Computed high b-values ≥ 2000s/mm² achieve an excellent signal-to-noise ratio (SNR) and increase the conspicuity of prostate tumors (6). Failure to follow these recommendations causes confusion and produces poor quality DWI sequences, which explains why one so often relies on DCE to compensate for a poor technique used for DWI.
The conclusion of the article by Han et al may thus be supported at the condition that radiologists pay attention to details to obtain an impeccable quality of the DWI sequence, and routinely use computed b-values ≥ 2000s/mm², for a better definition of PI-RADS 3 category.
- Han C, Liu S, Qin XB, Ma S, Zhu LN, Wang XY. MRI combined with PSA density in detecting clinically significant prostate cancer in patients with PSA serum levels of 4 approximately 10ng/mL: Biparametric versus multiparametric MRI. Diagn Interv Imaging. 2020.
- Pierre T, Cornud F, Colleter L et al. Diffusion-weighted imaging of the prostate: should we use quantitative metrics to better characterize focal lesions originating in the peripheral zone? Eur Radiol. 2017;28(5):2236-45.
- Rosenkrantz AB, Parikh N, Kierans AS, et al. Prostate Cancer Detection Using Computed Very High b-value Diffusion-weighted Imaging: How High Should We Go? Acad Radiol. 2016;23(6):704-11.
- Jendoubi S, Wagner M, Montagne S, et al. MRI for prostate cancer: can computed high b-value DWI replace native acquisitions? Eur Radiol. 2019;29(10):5197-204.
- Vural M, Ertas G, Onay A et al. Conspicuity of Peripheral Zone Prostate Cancer on Computed Diffusion-Weighted Imaging: Comparison of cDWI1500, cDWI2000, and cDWI3000. Biomed Res Int. 2014;2014:768291.
- Maas MC, Futterer JJ, Scheenen TW. Quantitative evaluation of computed high B value diffusion-weighted magnetic resonance imaging of the prostate. Invest Radiol. 2013;48(11):779-86.
- Weinreb JC, Barentsz JO, Choyke PL et al. PI-RADS Prostate Imaging – Reporting and Data System: 2015, Version 2. Eur Urol. 2016;69(1):16-40.
Legend figure 1
68 years old man. Rising PSA (8ng/ml). Hyposignal on the T2 image (a, arrow) with a mild restricted diffusion on the ADC map (b, arrow) and the computed DW image with a b-value of 1600s/mm² (c, arrow) . The PI-RADS score is 3. If ultrahigh computed b-values (b3000 and 6000 s/mm²) are used, the signal intensity of the lesion gets suppressed (d, e; arrow) as much as the rest of the prostate, confirming a low restricted diffusion and suggesting thus a benign lesion. The lesion enhances (f, arrow) after contrast injection, upgrading the PI-RADS 3 to 3+1 (equivalent to 4 in the PI-RADS classification). Targeted biopsies showed benign tissue, resulting in a false positive diagnosis of PI-RADS 4 lesion. The array of ultrahigh DW images is a highly effective tool to define PI-RADS 3 lesions eligible for follow-up.