Prostate Imaging Reporting and Data System Version 2.1: 2019 Update of Prostate Imaging Reporting and Data System Version 2

Machine learning applications in prostate cancer magnetic resonance imaging

With this review, we aimed to provide a synopsis of recently proposed applications of machine learning (ML) in radiology focusing on prostate magnetic resonance imaging (MRI). After defining the difference between ML and classical rule-based algorithms and the distinction among supervised, unsupervised and reinforcement learning, we explain the characteristic of deep learning (DL), a particular new type of ML, including its structure mimicking human neural networks and its 'black box' nature. Differences in the pipeline for applying ML and DL to prostate MRI are highlighted. The following potential clinical applications in different settings are outlined, many of them based only on MRI-unenhanced sequences: gland segmentation; assessment of lesion aggressiveness to distinguish between clinically significant and indolent cancers, allowing for active surveillance; cancer detection/diagnosis and localisation (transition versus peripheral zone, use of prostate imaging reporting and data system (PI-RADS) version 2), reading reproducibility, differentiation of cancers from prostatitis benign hyperplasia; local staging and pre-treatment assessment (detection of extraprostatic disease extension, planning of radiation therapy); and prediction of biochemical recurrence. Results are promising, but clinical applicability still requires more robust validation across scanner vendors, field strengths and institutions.

PI-RADS version 2: optimal time range for determining positivity of dynamic contrast-enhanced MRI in peripheral zone prostate cancer

AIM

To analyse the optimal time cut-off for determining positivity of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) in peripheral zone (PZ) prostate cancer (PCa).

MATERIALS AND METHODS

A consecutive series of 89 patients with PZ PCa who had undergone diffusion-weighted imaging (DWI) and subtraction DCE MRI were included. An experienced reader visually analysed the earliest time after contrast medium injection to visualise the best contrast between an index tumour and normal PZ on DCE MRI (i.e., best contrast time). The best contrast time cut-off for clinically significant cancer (csPCa) according to Epstein criteria or International Society of Urological Pathology (ISUP) grade ≥2 was analysed by an experienced reader, and applied to a less-experienced reader. For the index lesion of DWI category 3, the added value of DCE MRI (increased true positive and negative rates of PI-RADSv2 for csPCa) was evaluated using the cut-off time.

RESULTS

The best contrast time cut-off for csPCa was ≤72 seconds for Epstein criteria and ≤56 seconds for ISUP grade ≥2 by an experienced reader. The weighted kappa to determine positivity of DCE MRI was 0.622 for ≤72 seconds and 0.527 for ≤56 seconds between the two readers. The added value of DCE MRI was 55-75% by an experienced reader and 39.1-69.6% by a less-experienced reader.

CONCLUSION

For interpreting PI-RADSv2, imaging findings within 60-72 seconds following contrast media injection seem to reliably determine positivity of DCE MRI in PZ, and have added value for detecting csPCa.

Three-year experience of a dedicated prostate mpMRI pre-biopsy programme and effect on timed cancer diagnostic pathways

AIM

To evaluate the effect of pre-biopsy magnetic resonance imaging (MRI) on cancer diagnostic times, and to report MRI-directed pathology outcomes.

MATERIALS AND METHODS

In total, 1483 patients were referred with prostate cancer suspicion during a 30-month period. Upfront MRI was performed in 745 patients: 332 MRIs in the 15 months prior to dedicated scanning slots (group 1), and 413 in the 15 months post-introduction (group 2). A further 88 patients had initial MRI following clinical assessment. Biopsy via the transrectal (TR) or transperineal (TP) approach was performed, with MRI/ultrasound fusion for MRI targets. Clinically significant cancer (csPCa) was defined as Gleason ≥3+4. Negative MRIs were defined as Likert 1-2. Per-case clinical decisions were taken to biopsy or not.

RESULTS

44.4% of patients avoided biopsy. 484/833 (58.1%) MRIs were negative; 37.4% of these patients had biopsy with a negative predictive value (NPV) of 92.8% for Gleason ≥3+4 and 98.3% for ≥4+3. Overall prostate cancer prevalence was 34.3% (24.6% csPCa). In 323 MRI-positive cases, any cancer was present in 78.9% (csPCa 60.4%). Of the 1483 patients, 1232 (83.1%) completed all diagnostic tests within 28 days. Upfront MRI patients met this standard in 621/833 (74.5%), improving from 66.9% to 81.1% with reserved slots (group 2) with a reduced diagnostic time from median 25.5 to 20.9 days. Biopsy scheduling delayed the pathway in 69.7%, with MRI responsible in 22.3%, reducing to 10.3% in group 2. TP biopsies met the 28-day standard in significantly less cases (29.7%), compared to TR (67.4%, p<0.0001).

CONCLUSION

Reserved MRI slots reduces time-to-diagnosis, and upfront MRI safely avoids biopsy in a significant proportion of men, whilst maintaining expected csPCa detection rates.

PI-RADS version 2.1: one small step for prostate MRI

Multiparametric (mp) prostate magnetic resonance imaging (MRI) is playing an increasingly prominent role in the diagnostic work-up of patients with suspected prostate cancer. Performing mpMRI before biopsy offers several advantages including biopsy avoidance under certain clinical circumstances and targeting biopsy of suspicious lesions to enable the correct diagnosis. The success of the technique is heavily dependent on high-quality image acquisition, interpretation, and report communication, all areas addressed by previous versions of the Prostate Imaging-Reporting and Data System (PI-RADS) recommendations. Numerous studies have validated the approach, but the widespread adoption of PI-RADS version 2 has also highlighted inconsistencies and limitations, particularly relating to interobserver variability for evaluation of the transition zone. These limitations are addressed in the recently released version 2.1. In this article, we highlight the key changes proposed in PI-RADS v2.1 and explore the background reasoning and evidence for the recommendations.

PI-RADS Steering Committee: The PI-RADS Multiparametric MRI and MRI-directed Biopsy Pathway

High-quality evidence shows that MRI in biopsy-naive men can reduce the number of men who need prostate biopsy and can reduce the number of diagnoses of clinically insignificant cancers that are unlikely to cause harm. In men with prior negative biopsy results who remain under persistent suspicion, MRI improves the detection and localization of life-threatening prostate cancer with greater clinical utility than the current standard of care, systematic transrectal US-guided biopsy. Systematic analyses show that MRI-directed biopsy increases the effectiveness of the prostate cancer diagnosis pathway. The incorporation of MRI-directed pathways into clinical care guidelines in prostate cancer detection has begun. The widespread adoption of the Prostate Imaging Reporting and Data System (PI-RADS) for multiparametric MRI data acquisition, interpretation, and reporting has promoted these changes in practice. The PI-RADS MRI-directed biopsy pathway enables the delivery of key diagnostic benefits to men suspected of having cancer based on clinical suspicion. Herein, the PI-RADS Steering Committee discusses how the MRI pathway should be incorporated into routine clinical practice and the challenges in delivering the positive health impacts needed by men suspected of having clinically significant prostate cancer.

Validation of IMPROD biparametric MRI in men with clinically suspected prostate cancer: A prospective multi-institutional trial

BACKGROUND

Magnetic resonance imaging (MRI) combined with targeted biopsy (TB) is increasingly used in men with clinically suspected prostate cancer (PCa), but the long acquisition times, high costs, and inter-center/reader variability of routine multiparametric prostate MRI limit its wider adoption.

METHODS AND FINDINGS

The aim was to validate a previously developed unique MRI acquisition and reporting protocol, IMPROD biparametric MRI (bpMRI) (NCT01864135), in men with a clinical suspicion of PCa in a multi-institutional trial (NCT02241122). IMPROD bpMRI has average acquisition time of 15 minutes (no endorectal coil, no intravenous contrast use) and consists of T2-weighted imaging and 3 separate diffusion-weighed imaging acquisitions. Between February 1, 2015, and March 31, 2017, 364 men with a clinical suspicion of PCa were enrolled at 4 institutions in Finland. Men with an equivocal to high suspicion (IMPROD bpMRI Likert score 3-5) of PCa had 2 TBs of up to 2 lesions followed by a systematic biopsy (SB). Men with a low to very low suspicion (IMPROD bpMRI Likert score 1-2) had only SB. All data and protocols are freely available. The primary outcome of the trial was diagnostic accuracy-including overall accuracy, sensitivity, specificity, negative predictive value (NPV), and positive predictive value-of IMPROD bpMRI for clinically significant PCa (SPCa), which was defined as a Gleason score ≥ 3 + 4 (Gleason grade group 2 or higher). In total, 338 (338/364, 93%) prospectively enrolled men completed the trial. The accuracy and NPV of IMPROD bpMRI for SPCa were 70% (113/161) and 95% (71/75) (95% CI 87%-98%), respectively. Restricting the biopsy to men with equivocal to highly suspicious IMPROD bpMRI findings would have resulted in a 22% (75/338) reduction in the number of men undergoing biopsy while missing 4 (3%, 4/146) men with SPCa. The main limitation is uncertainty about the true PCa prevalence in the study cohort, since some of the men may have PCa despite having negative biopsy findings.

CONCLUSIONS

IMPROD bpMRI demonstrated a high NPV for SPCa in men with a clinical suspicion of PCa in this prospective multi-institutional clinical trial.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02241122.

Supervised risk predictor of central gland lesions in prostate cancer using 1 H MR spectroscopic imaging with gradient offset-independent adiabaticity pulses

BACKGROUND

Due to the histological heterogeneity of the central gland, accurate detection of central gland prostate cancer remains a challenge.

PURPOSE

To evaluate the efficacy of in vivo 3D ¹ H MR spectroscopic imaging (3D ¹ H MRSI) with a semi-localized adiabatic selective refocusing (sLASER) sequence and gradient-modulated offset-independent adiabatic (GOIA) pulses for detection of central gland prostate cancer. Additionally four risk models were developed to differentiate 1) normal vs. cancer, 2) low- vs. high-risk cancer, 3) low- vs. intermediate-risk cancer, and 4) intermediate- vs. high-risk cancer voxels.

STUDY TYPE

Prospective.

SUBJECTS

Thirty-six patients with biopsy-proven central gland prostate cancer.

FIELD STRENGTH/SEQUENCE

3T MRI / 3D ¹ H MRSI using GOIA-sLASER.

ASSESSMENT

Cancer and normal regions of interest (ROIs) were selected by an experienced radiologist and ¹ H MRSI voxels were placed within the ROIs to calculate seven metabolite signal ratios. Voxels were split into two subsets, 80% for model training and 20% for testing.

STATISTICAL TESTS

Four support vector machine (SVM) models were built using the training dataset. The accuracy, sensitivity, and specificity for each model were calculated for the testing dataset.

RESULTS

High-quality MR spectra were obtained for the whole central gland of the prostate. The normal vs. cancer diagnostic model achieved the highest predictive performance with an accuracy, sensitivity, and specificity of 96.2%, 95.8%, and 93.1%, respectively. The accuracy, sensitivity, and specificity of the low- vs. high-risk cancer and low- vs. intermediate-risk cancer models were 82.5%, 89.2%, 70.2%, and 73.0%, 84.7%, 60.8%, respectively. The intermediate- vs. high-risk cancer model yielded an accuracy, sensitivity, and specificity lower than 55%.

DATA CONCLUSION

The GOIA-sLASER sequence with an external phased-array coil allows for fast assessment of central gland prostate cancer. The classification offers a promising diagnostic tool for discriminating normal vs. cancer, low- vs. high-risk cancer, and low- vs. intermediate-risk cancer.

LEVEL OF EVIDENCE

2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;50:1926-1936.

[Predictive biomarkers in oncologic uropathology]

BACKGROUND

Tumors of the genitourinary system are common. In recent years, our understanding of their molecular background and therefore the number of potential predictive biomarkers has massively increased.

OBJECTIVES

The aim of the current work is to give an overview of recent (molecular) developments and predictive biomarkers in urologic oncology and to give a perspective of what might become relevant in the future of the field.

MATERIAL AND METHODS

We considered the recent literature and study data and combined it with our own expertise in tumors of the urinary system, kidneys, and prostate.

RESULTS AND CONCLUSIONS

The molecular subtypes of muscle-invasive urothelial bladder cancer (MIBC) hold a predictive and prognostic significance and correlate with clinicopathological features. Immune therapy with checkpoint inhibitors (CPI) has a major role in urothelial carcinoma (UC), but also in renal cell carcinoma and a subgroup of prostate cancers. The first-line use in UC is restricted to PD-L1-"positive" cases (≥IC2/3, CPS ≥ 10). Further predictive markers are currently under evaluation, while the predictive significance of tumor mutational burden (TMB) is under debate. In addition to a subgroup of renal cell carcinomas, a subgroup of prostate carcinomas with alterations in the DNA repair system might benefit from a customized therapy approach (PARP inhibitors, platin-containing chemotherapy). The multitude of potentially therapy-relevant molecular alterations and related predictive biomarkers calls for the implementation of sophisticated molecular analyses in daily routine. This will lead to an even more rapid dynamic in the field of genitourinary pathology.

Multiparametric MRI - local staging of prostate cancer and beyond

Background Accurate local staging is critical for treatment planning and prognosis in patients with prostate cancer (PCa). The primary aim is to differentiate between organ-confined and locally advanced disease with the latter carrying a worse clinical prognosis. Multiparametric MRI (mpMRI) is the imaging modality of choice for the local staging of PCa and has an incremental value in assessing pelvic nodal disease and bone involvement. It has shown superior performance compared to traditional staging based on clinical nomograms, and provides additional information on the site and extent of disease. MRI has a high specificity for diagnosing extracapsular extension (ECE), seminal vesicle invasion (SVI) and lymph node (LN) metastases, however, sensitivity remains poor. As a result, extended pelvic LN dissection remains the gold standard for assessing pelvic nodal involvement, and there has been recent progress in developing advanced imaging techniques for more distal staging. Conclusions T2W-weighted imaging is the cornerstone for local staging of PCa. Imaging at 3T and incorporating both diffusion weighted and dynamic contrast enhanced imaging can further increase accuracy. "Next generation" imaging including whole body MRI and PET-MRI imaging using prostate specific membrane antigen (68Ga-PSMA), has shown promising for assessment of LN and bone involvement as compared to the traditional work-up using bone scintigraphy and body CT.