A UK-based investigation of inter- and intra-observer reproducibility of Gleason grading of prostatic biopsies

Applications of artificial intelligence in prostate cancer histopathology

The diagnosis of prostate cancer (PCa) depends on the evaluation of core needle biopsies by trained pathologists. Artificial intelligence (AI) derived models have been created to address the challenges posed by pathologists' increasing workload, workforce shortages, and variability in histopathology assessment. These models with histopathological parameters integrated into sophisticated neural networks demonstrate remarkable ability to identify, grade, and predict outcomes for PCa. Though the fully autonomous diagnosis of PCa remains elusive, recently published data suggests that AI has begun to serve as an initial screening tool, an assistant in the form of a real-time interactive interface during histological analysis, and as a second read system to detect false negative diagnoses. Our article aims to describe recent advances and future opportunities for AI in PCa histopathology.

Visual Assessment of 2-Dimensional Levels Within 3-Dimensional Pathology Data Sets of Prostate Needle Biopsies Reveals Substantial Spatial Heterogeneity

Prostate cancer prognostication largely relies on visual assessment of a few thinly sectioned biopsy specimens under a microscope to assign a Gleason grade group (GG). Unfortunately, the assigned GG is not always associated with a patient's outcome in part because of the limited sampling of spatially heterogeneous tumors achieved by 2-dimensional histopathology. In this study, open-top light-sheet microscopy was used to obtain 3-dimensional pathology data sets that were assessed by 4 human readers. Intrabiopsy variability was assessed by asking readers to perform Gleason grading of 5 different levels per biopsy for a total of 20 core needle biopsies (ie, 100 total images). Intrabiopsy variability (Cohen κ) was calculated as the worst pairwise agreement in GG between individual levels within each biopsy and found to be 0.34, 0.34, 0.38, and 0.43 for the 4 pathologists. These preliminary results reveal that even within a 1-mm-diameter needle core, GG based on 2-dimensional images can vary dramatically depending on the location within a biopsy being analyzed. We believe that morphologic assessment of whole biopsies in 3 dimension has the potential to enable more reliable and consistent tumor grading.

A pathway activity-based proteomic classifier stratifies prostate tumors into two subtypes

Prostate cancer (PCa) is the second most common cancer in males worldwide. The risk stratification of PCa is mainly based on morphological examination. Here we analyzed the proteome of 667 tumor samples from 487 Chinese PCa patients and characterized 9576 protein groups by PulseDIA mass spectrometry. Then we developed a pathway activity-based classifier concerning 13 proteins from seven pathways, and dichotomized the PCa patients into two subtypes, namely PPS1 and PPS2. PPS1 is featured with enhanced innate immunity, while PPS2 with suppressed innate immunity. This classifier exhibited a correlation with PCa progression in our cohort and was further validated by two published transcriptome datasets. Notably, PPS2 was significantly correlated with poor biochemical recurrence (BCR)/metastasis-free survival (log-rank P-value < 0.05). The PPS2 was also featured with cell proliferation activation. Together, our study presents a novel pathway activity-based stratification scheme for PCa.

Cancer detection and complications of transperineal prostate biopsy with antibiotics when indicated

OBJECTIVES

To describe the prostate cancer (PCa) detection rate, including clinically significant prostate cancer (csPCa), in a large cohort of patients who underwent transperineal ultrasonography-guided systematic prostate biopsy (TPB-US) using a probe-mounted transperineal access system, with magnetic resonance imaging (MRI) cognitive fusion in case of a Prostate Imaging-Reporting and Data System grade 3-5 lesion, under local anaesthesia in an outpatient setting. Additionally, to compare the incidence of procedure-related complications with a cohort of patients undergoing transrectal ultrasonography-guided (TRB-US) and transrectal MRI-guided biopsies (TRB-MRI).

PATIENTS AND METHODS

This was an observational cohort study in men who underwent TPB-US prostate biopsy in a large teaching hospital. For each participant, prostate-specific antigen level, clinical tumour stage, prostate volume, MRI parameters, number of (targeted) prostate biopsies, biopsy International Society of Uropathology (ISUP) grade and procedure-related complications were assessed. csPCa was defined as ISUP grade ≥2. Antibiotic prophylaxis was only given in those with an increased risk of urinary tract infection.

RESULTS

A total of 1288 TPB-US procedures were evaluated. The overall detection rate for PCa in biopsy-naive patients was 73%, and for csPCa it was 63%. The incidence of hospitalization was 1% in TPB-US (13/1288), compared to 4% in TRB-US (8/214) and 3% in TRB-MRI (7/219; P = 0.002).

CONCLUSIONS

Contemporary combined systematic and target TPB-US with MRI cognitive fusion is easy to perform in an outpatient setting, with a high detection rate of csPCa and a low incidence of procedure-related complications.

Reliability of whole mount radical prostatectomy histopathology as the ground truth for artificial intelligence assisted prostate imaging

The development of artificial intelligence-based imaging techniques for prostate cancer (PCa) detection and diagnosis requires a reliable ground truth, which is generally based on histopathology from radical prostatectomy specimens. This study proposes a comprehensive protocol for the annotation of prostatectomy pathology slides. To evaluate the reliability of the protocol, interobserver variability was assessed between five pathologists, who annotated ten radical prostatectomy specimens consisting of 74 whole mount pathology slides. Interobserver variability was assessed for both the localization and grading of PCa. The results indicate excellent overall agreement on the localization of PCa (Gleason pattern ≥ 3) and clinically significant PCa (Gleason pattern ≥ 4), with Dice similarity coefficients (DSC) of 0.91 and 0.88, respectively. On a per-slide level, agreement for primary and secondary Gleason pattern was almost perfect and substantial, with Fleiss Kappa of .819 (95% CI .659-.980) and .726 (95% CI .573-.878), respectively. Agreement on International Society of Urological Pathology Grade Group was evaluated for the index lesions and showed agreement in 70% of cases, with a mean DSC of 0.92 for all index lesions. These findings show that a standardized protocol for prostatectomy pathology annotation provides reliable data on PCa localization and grading, with relatively high levels of interobserver agreement. More complicated tissue characterization, such as the presence of cribriform growth and intraductal carcinoma, remains a source of interobserver variability and should be treated with care when used in ground truth datasets.

Impact of Magnetic Resonance Imaging Targeting on Pathologic Upgrading and Downgrading at Prostatectomy: A Systematic Review and Meta-analysis

CONTEXT

The evidence supporting multiparametric magnetic resonance imaging (MRI) targeting for biopsy is nearly exclusively based on biopsy pathologic outcomes. This is problematic, as targeting likely allows preferential identification of small high-grade areas of questionable oncologic significance, raising the likelihood of overdiagnosis and overtreatment.

OBJECTIVE

To estimate the impact of MRI-targeted, systematic, and combined biopsies on radical prostatectomy (RP) grade group concordance.

EVIDENCE ACQUISITION

PubMed MEDLINE and Cochrane Library were searched from July 2018 to January 2022. Studies that conducted systematic and MRI-targeted prostate biopsies and compared biopsy results with pathology after RP were included. We performed a meta-analysis to assess whether pathologic upgrading and downgrading were influenced by biopsy type and a net-benefit analysis using pooled risk difference estimates.

EVIDENCE SYNTHESIS

Both targeted only and combined biopsies were less likely to result in upgrading (odds ratio [OR] vs systematic of 0.70, 95% confidence interval [CI] 0.63-0.77, p < 0.001, and 0.50, 95% CI 0.45-0.55, p < 0.001), respectively). Targeted only and combined biopsies increased the odds of downgrading (1.24 (95% CI 1.05-1.46), p = 0.012, and 1.96 (95% CI 1.68-2.27, p < 0.001) compared with systematic biopsies, respectively. The net benefit of targeted and combined biopsies is 8 and 7 per 100 if harms of up- and downgrading are considered equal, but 7 and -1 per 100 if the harm of downgrading is considered twice that of upgrading.

CONCLUSIONS

The addition of MRI-targeting results in lower rates of upgrading as compared to systematic biopsy at RP (27% vs 42%). However, combined MRI-targeted and systematic biopsies are associated with more downgrading at RP (19% v 11% for combined vs systematic). Strong heterogeneity suggests further research into factors that influence the rates of up- and downgrading and that distinguishes clinically relevant from irrelevant grade changes is needed. Until then, the benefits and harms of combined MRI-targeted and systematic biopsies cannot be fully assessed.

PATIENT SUMMARY

We reviewed the ability of magnetic resonance imaging (MRI)-targeted biopsies to predict cancer grade at prostatectomy. We found that combined MRI-targeted and systematic biopsies result in more cancers being downgraded than systematic biopsies.

Clinical Trial Protocol: Developing an Image Classification Algorithm for Prostate Cancer Diagnosis on Three-dimensional Multiparametric Transrectal Ultrasound

Introduction and hypothesis

The tendency toward population-based screening programs for prostate cancer (PCa) is expected to increase demand for prebiopsy imaging. This study hypothesizes that a machine learning image classification algorithm for three-dimensional multiparametric transrectal prostate ultrasound (3D mpUS) can detect PCa accurately.

Design

This is a phase 2 prospective multicenter diagnostic accuracy study. A total of 715 patients will be included in a period of approximately 2 yr. Patients are eligible in case of suspected PCa for which prostate biopsy is indicated or in case of biopsy-proven PCa for which radical prostatectomy (RP) will be performed. Exclusion criteria are prior treatment for PCa or contraindications for ultrasound contrast agents (UCAs).

Protocol overview

Study participants will undergo 3D mpUS, consisting of 3D grayscale, 4D contrast-enhanced ultrasound, and 3D shear wave elastography (SWE). Whole-mount RP histopathology will provide the ground truth to train the image classification algorithm. Patients included prior to prostate biopsy will be used for subsequent preliminary validation. There is a small, anticipated risk for participants associated with the administration of a UCA. Informed consent has to be given prior to study participation, and (serious) adverse events will be reported.

Statistical analysis

The primary outcome will be the diagnostic performance of the algorithm for detecting clinically significant PCa (csPCa) on a per-voxel and a per-microregion level. Diagnostic performance will be reported as the area under the receiver operating characteristic curve. Clinically significant PCa is defined as the International Society of Urological grade group ≥2. Full-mount RP histopathology will be used as the reference standard. Secondary outcomes will be sensitivity, specificity, negative predictive value, and positive predictive value for csPCa on a per-patient level, evaluated in patients included prior to prostate biopsy, using biopsy results as the reference standard. A further analysis will be performed on the ability of the algorithm to differentiate between low-, intermediate-, and high-risk tumors.

Discussion and summary

This study aims to develop an ultrasound-based imaging modality for PCa detection. Subsequent head-to-head validation trials with magnetic resonance imaging have to be performed in order to determine its role in clinical practice for risk stratification in patients suspected for PCa.

Multi-Stage Classification-Based Deep Learning for Gleason System Grading Using Histopathological Images

In this work, we introduced an automated diagnostic system for Gleason system grading and grade groups (GG) classification using whole slide images (WSIs) of digitized prostate biopsy specimens (PBSs). Our system first classifies the Gleason pattern (GP) from PBSs and then identifies the Gleason score (GS) and GG. We developed a comprehensive DL-based approach to develop a grading pipeline system for the digitized PBSs and consider GP as a classification problem (not segmentation) compared to current research studies (deals with as a segmentation problem). A multilevel binary classification was implemented to enhance the segmentation accuracy for GP. Also, we created three levels of analysis (pyramidal levels) to extract different types of features. Each level has four shallow binary CNN to classify five GP labels. A majority fusion is applied for each pixel that has a total of 39 labeled images to create the final output for GP. The proposed framework is trained, validated, and tested on 3080 WSIs of PBS. The overall diagnostic accuracy for each CNN is evaluated using several metrics: precision (PR), recall (RE), and accuracy, which are documented by the confusion matrices.The results proved our system's potential for classifying all five GP and, thus, GG. The overall accuracy for the GG is evaluated using two metrics, PR and RE. The grade GG results are between 50% to 92% for RE and 50% to 92% for PR. Also, a comparison between our CNN architecture and the standard CNN (ResNet50) highlights our system's advantage. Finally, our deep-learning system achieved an agreement with the consensus grade groups.

Bilateral Seminal Vesicle Invasion Is Not Associated with Worse Outcomes in Locally Advanced Prostate Carcinoma

Background and Objectives: Patients with seminal vesicle invasion (SVI) are a highly heterogeneous group. Prognosis can be affected by many clinical and pathological characteristics. Our aim was to study whether bilateral SVI (bi-SVI) is associated with worse oncological outcomes. Materials and Methods: This is an observational retrospective study that included 146 pT3b patients treated with radical prostatectomy (RP). We compared the results between unilateral SVI (uni-SVI) and bi-SVI. The log-rank test and Kaplan–Meier curves were used to compare biochemical recurrence-free survival (BCR), metastasis-free survival (MFS), and additional treatment-free survival. Cox proportional hazard models were used to identify predictors of BCR-free survival, MFS, and additional treatment-free survival. Results: 34.93% of patients had bi-SVI. The median follow-up was 46.84 months. No significant differences were seen between the uni-SVI and bi-SVI groups. BCR-free survival at 5 years was 33.31% and 25.65% (p = 0.44) for uni-SVI and bi-SVI. MFS at 5 years was 86.03% vs. 75.63% (p = 0.1), and additional treatment-free survival was 36.85% vs. 21.93% (p = 0.09), respectively. In the multivariate analysis, PSA was related to the development of BCR [HR 1.34 (95%CI: 1.01–1.77); p = 0.03] and metastasis [HR 1.83 (95%CI: 1.13–2.98); p = 0.02]. BCR was also influenced by lymph node infiltration [HR 2.74 (95%CI: 1.41–5.32); p = 0.003]. Additional treatment was performed more frequently in patients with positive margins [HR: 3.50 (95%CI: 1.65–7.44); p = 0.001]. Conclusions: SVI invasion is an adverse pathology feature, with a widely variable prognosis. In our study, bilateral seminal vesicle invasion did not predict worse outcomes in pT3b patients despite being associated with more undifferentiated tumors.

Discriminatory Gleason grade group signatures of prostate cancer: An application of machine learning methods

One of the most precise methods to detect prostate cancer is by evaluation of a stained biopsy by a pathologist under a microscope. Regions of the tissue are assessed and graded according to the observed histological pattern. However, this is not only laborious, but also relies on the experience of the pathologist and tends to suffer from the lack of reproducibility of biopsy outcomes across pathologists. As a result, computational approaches are being sought and machine learning has been gaining momentum in the prediction of the Gleason grade group. To date, machine learning literature has addressed this problem by using features from magnetic resonance imaging images, whole slide images, tissue microarrays, gene expression data, and clinical features. However, there is a gap with regards to predicting the Gleason grade group using DNA sequences as the only input source to the machine learning models. In this work, using whole genome sequence data from South African prostate cancer patients, an application of machine learning and biological experiments were combined to understand the challenges that are associated with the prediction of the Gleason grade group. A series of machine learning binary classifiers (XGBoost, LSTM, GRU, LR, RF) were created only relying on DNA sequences input features. All the models were not able to adequately discriminate between the DNA sequences of the studied Gleason grade groups (Gleason grade group 1 and 5). However, the models were further evaluated in the prediction of tumor DNA sequences from matched-normal DNA sequences, given DNA sequences as the only input source. In this new problem, the models performed acceptably better than before with the XGBoost model achieving the highest accuracy of 74 ± 01, F1 score of 79 ± 01, recall of 99 ± 0.0, and precision of 66 ± 0.1.

Infiltration pattern predicts metastasis and progression better than the T-stage and grade in pancreatic neuroendocrine tumors: a proposal for a novel infiltration-based morphologic grading

The advancing edge profile is a powerful determinant of tumor behavior in many organs. In this study, a grading system assessing the tumor-host interface was developed and tested in 181 pancreatic neuroendocrine tumors (PanNETs), 63 of which were <=2 cm. Three tumor slides representative of the spectrum (least, medium, and most) of invasiveness at the advancing edge of the tumor were selected, and then each slide was scored as follows. Well-demarcated/encapsulated, 1 point; Mildly irregular borders and/or minimal infiltration into adjacent tissue, 2 points; Infiltrative edges with several clusters beyond the main tumor but still relatively close, and/or satellite demarcated nodules, 3 points; No demarcation, several cellular clusters away from the tumor, 4 points; Exuberantly infiltrative pattern, scirrhous growth, dissecting the normal parenchymal elements, 5 points. The sum of the rankings on the three slides was obtained. Cases with scores of 3-6 were defined as "non/minimally infiltrative" (NI; n = 77), 7-9 as "moderately infiltrative" (MI; n = 68), and 10-15 as "highly infiltrative" (HI; n = 36). In addition to showing a statistically significant correlation with all the established signs of aggressiveness (grade, size, T-stage), this grading system was found to be the most significant predictor of adverse outcomes (metastasis, progression, and death) on multivariate analysis, more strongly than T-stage, while Ki-67 index did not stand the multivariate test. As importantly, cases <=2 cm were also stratified by this grading system rendering it applicable also to this group that is currently placed in "watchful waiting" protocols. In conclusion, the proposed grading system has a strong, independent prognostic value and therefore should be considered for integration into routine pathology practice after being evaluated in validation studies with larger series.

Comparative Pathobiology of Canine and Human Prostate Cancer: State of the Art and Future Directions

First described in 1817, prostate cancer is considered a complex neoplastic entity, and one of the main causes of death in men in the western world. In dogs, prostatic carcinoma (PC) exhibits undifferentiated morphology with different phenotypes, is hormonally independent of aggressive character, and has high rates of metastasis to different organs. Although in humans, the risk factors for tumor development are known, in dogs, this scenario is still unclear, especially regarding castration. Therefore, with the advent of molecular biology, studies were and are carried out with the aim of identifying the main molecular mechanisms and signaling pathways involved in the carcinogenesis and progression of canine PC, aiming to identify potential biomarkers for diagnosis, prognosis, and targeted treatment. However, there are extensive gaps to be filled, especially when considering the dog as experimental model for the study of this neoplasm in humans. Thus, due to the complexity of the subject, the objective of this review is to present the main pathobiological aspects of canine PC from a comparative point of view to the same neoplasm in the human species, addressing the historical context and current understanding in the scientific field.

Code-Free Development and Deployment of Deep Segmentation Models for Digital Pathology

Application of deep learning on histopathological whole slide images (WSIs) holds promise of improving diagnostic efficiency and reproducibility but is largely dependent on the ability to write computer code or purchase commercial solutions. We present a code-free pipeline utilizing free-to-use, open-source software (QuPath, DeepMIB, and FastPathology) for creating and deploying deep learning-based segmentation models for computational pathology. We demonstrate the pipeline on a use case of separating epithelium from stroma in colonic mucosa. A dataset of 251 annotated WSIs, comprising 140 hematoxylin-eosin (HE)-stained and 111 CD3 immunostained colon biopsy WSIs, were developed through active learning using the pipeline. On a hold-out test set of 36 HE and 21 CD3-stained WSIs a mean intersection over union score of 95.5 and 95.3% was achieved on epithelium segmentation. We demonstrate pathologist-level segmentation accuracy and clinical acceptable runtime performance and show that pathologists without programming experience can create near state-of-the-art segmentation solutions for histopathological WSIs using only free-to-use software. The study further demonstrates the strength of open-source solutions in its ability to create generalizable, open pipelines, of which trained models and predictions can seamlessly be exported in open formats and thereby used in external solutions. All scripts, trained models, a video tutorial, and the full dataset of 251 WSIs with ~31 k epithelium annotations are made openly available at https://github.com/andreped/NoCodeSeg to accelerate research in the field.

Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge

Artificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent validation. With this in mind, we organized the PANDA challenge-the largest histopathology competition to date, joined by 1,290 developers-to catalyze development of reproducible AI algorithms for Gleason grading using 10,616 digitized prostate biopsies. We validated that a diverse set of submitted algorithms reached pathologist-level performance on independent cross-continental cohorts, fully blinded to the algorithm developers. On United States and European external validation sets, the algorithms achieved agreements of 0.862 (quadratically weighted κ, 95% confidence interval (CI), 0.840-0.884) and 0.868 (95% CI, 0.835-0.900) with expert uropathologists. Successful generalization across different patient populations, laboratories and reference standards, achieved by a variety of algorithmic approaches, warrants evaluating AI-based Gleason grading in prospective clinical trials.

Prediction of Clinically Significant Cancer Using Radiomics Features of Pre-Biopsy of Multiparametric MRI in Men Suspected of Prostate Cancer

Background: Texture features based on the spatial relationship of pixels, known as the gray-level co-occurrence matrix (GLCM), may play an important role in providing the accurate classification of suspected prostate cancer. The purpose of this study was to use quantitative imaging parameters of pre-biopsy multiparametric magnetic resonance imaging (mpMRI) for the prediction of clinically significant prostate cancer. Methods: This was a prospective study, recruiting 200 men suspected of having prostate cancer. Participants were imaged using a protocol-based 3T MRI in the pre-biopsy setting. Radiomics parameters were extracted from the T2WI and ADC texture features of the gray-level co-occurrence matrix were delineated from the region of interest. Radical prostatectomy histopathology was used as a reference standard. A Kruskal–Wallis test was applied first to identify the significant radiomic features between the three groups of Gleason scores (i.e., G1, G2 and G3). Subsequently, the Holm–Bonferroni method was applied to correct and control the probability of false rejections. We compared the probability of correctly predicting significant prostate cancer between the explanatory GLCM radiomic features, PIRADS and PSAD, using the area under the receiver operation characteristic curves. Results: We identified the significant difference in radiomic features between the three groups of Gleason scores. In total, 12 features out of 22 radiomics features correlated with the Gleason groups. Our model demonstrated excellent discriminative ability (C-statistic = 0.901, 95%CI 0.859–0.943). When comparing the probability of correctly predicting significant prostate cancer between explanatory GLCM radiomic features (Sum Variance T2WI, Sum Entropy T2WI, Difference Variance T2WI, Entropy ADC and Difference Variance ADC), PSAD and PIRADS via area under the ROC curve, radiomic features were 35.0% and 34.4% more successful than PIRADS and PSAD, respectively, in correctly predicting significant prostate cancer in our patients (p < 0.001). The Sum Entropy T2WI score had the greatest impact followed by the Sum Variance T2WI. Conclusion: Quantitative GLCM texture analyses of pre-biopsy MRI has the potential to be used as a non-invasive imaging technique to predict clinically significant cancer in men suspected of having prostate cancer.

Learning a Triplet Embedding Distance to Represent Gleason Patterns

Gleason grade stratification is the main histological standard to determine the severity and progression of prostate cancer. Nonetheless, there is a high variability on disease diagnosis among expert pathologists (kappa lower than 0.44). End-to-end deep representations have recently deal with the automatic classification of Gleason grades, where each grade is limited to namely code high-visual-variability sharing patterns among classes. Such limitation on models may be attributed to the relatively few labels to train the representation, as well as, to the natural imbalanced sets, available in clinical scenarios. To overcome such limitation, this work introduces a new embedding representation that learns intra and inter-Gleason relationships from more challenging class samples (grades tree and fourth). The proposed strategy implements a triplet loss scheme building a hidden embedding space that correctly differentiates close Gleason levels. The proposed approach shows promising results achieving an average accuracy of 74% to differentiate between degrees three and four. For classification of all degrees, the proposed approach achieves an average accuracy of 62%.

A Deep Learning Pipeline for Grade Groups Classification Using Digitized Prostate Biopsy Specimens

Prostate cancer is a significant cause of morbidity and mortality in the USA. In this paper, we develop a computer-aided diagnostic (CAD) system for automated grade groups (GG) classification using digitized prostate biopsy specimens (PBSs). Our CAD system aims to firstly classify the Gleason pattern (GP), and then identifies the Gleason score (GS) and GG. The GP classification pipeline is based on a pyramidal deep learning system that utilizes three convolution neural networks (CNN) to produce both patch- and pixel-wise classifications. The analysis starts with sequential preprocessing steps that include a histogram equalization step to adjust intensity values, followed by a PBSs' edge enhancement. The digitized PBSs are then divided into overlapping patches with the three sizes: 100 × 100 (CNNS), 150 × 150 (CNNM), and 200 × 200 (CNNL), pixels, and 75% overlap. Those three sizes of patches represent the three pyramidal levels. This pyramidal technique allows us to extract rich information, such as that the larger patches give more global information, while the small patches provide local details. After that, the patch-wise technique assigns each overlapped patch a label as GP categories (1 to 5). Then, the majority voting is the core approach for getting the pixel-wise classification that is used to get a single label for each overlapped pixel. The results after applying those techniques are three images of the same size as the original, and each pixel has a single label. We utilized the majority voting technique again on those three images to obtain only one. The proposed framework is trained, validated, and tested on 608 whole slide images (WSIs) of the digitized PBSs. The overall diagnostic accuracy is evaluated using several metrics: precision, recall, F1-score, accuracy, macro-averaged, and weighted-averaged. The (CNNL) has the best accuracy results for patch classification among the three CNNs, and its classification accuracy is 0.76. The macro-averaged and weighted-average metrics are found to be around 0.70-0.77. For GG, our CAD results are about 80% for precision, and between 60% to 80% for recall and F1-score, respectively. Also, it is around 94% for accuracy and NPV. To highlight our CAD systems' results, we used the standard ResNet50 and VGG-16 to compare our CNN's patch-wise classification results. As well, we compared the GG's results with that of the previous work.

Predicting prostate cancer specific-mortality with artificial intelligence-based Gleason grading

Background

Gleason grading of prostate cancer is an important prognostic factor, but suffers from poor reproducibility, particularly among non-subspecialist pathologists. Although artificial intelligence (A.I.) tools have demonstrated Gleason grading on-par with expert pathologists, it remains an open question whether and to what extent A.I. grading translates to better prognostication.

Methods

In this study, we developed a system to predict prostate cancer-specific mortality via A.I.-based Gleason grading and subsequently evaluated its ability to risk-stratify patients on an independent retrospective cohort of 2807 prostatectomy cases from a single European center with 5–25 years of follow-up (median: 13, interquartile range 9–17).

Results

Here, we show that the A.I.’s risk scores produced a C-index of 0.84 (95% CI 0.80–0.87) for prostate cancer-specific mortality. Upon discretizing these risk scores into risk groups analogous to pathologist Grade Groups (GG), the A.I. has a C-index of 0.82 (95% CI 0.78–0.85). On the subset of cases with a GG provided in the original pathology report (n = 1517), the A.I.’s C-indices are 0.87 and 0.85 for continuous and discrete grading, respectively, compared to 0.79 (95% CI 0.71–0.86) for GG obtained from the reports. These represent improvements of 0.08 (95% CI 0.01–0.15) and 0.07 (95% CI 0.00–0.14), respectively.

Conclusions

Our results suggest that A.I.-based Gleason grading can lead to effective risk stratification, and warrants further evaluation for improving disease management.

Gleason grading is the process by which pathologists assess the morphology of prostate tumors. The assigned Grade Group tells us about the likely clinical course of people with prostate cancer and helps doctors to make decisions on treatment. The process is complex and subjective, with frequent disagreement amongst pathologists. In this study, we develop and evaluate an approach to Gleason grading based on artificial intelligence, rather than pathologists’ assessment, to predict risk of dying of prostate cancer. Looking back at tumors and data from 2,807 people diagnosed with prostate cancer, we find that our approach is better at predicting outcomes compared to grading by pathologists alone. These findings suggest that artificial intelligence might help doctors to accurately determine the probable clinical course of people with prostate cancer, which, in turn, will guide treatment.

Wulczyn et al. utilise a deep learning-based Gleason grading model to predict prostate cancer-specific mortality in a retrospective cohort of radical prostatectomy patients. Their model enables improved risk stratification compared to pathologists’ grading and demonstrates the potential for computational pathology in the management of prostate cancer.

Rapid Validation of Whole-Slide Imaging for Primary Histopathology Diagnosis

OBJECTIVES

The ongoing global severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic necessitates adaptations in the practice of surgical pathology at scale. Primary diagnosis by whole-slide imaging (WSI) is a key component that would aid departments in providing uninterrupted histopathology diagnosis and maintaining revenue streams from disruption. We sought to perform rapid validation of the use of WSI in primary diagnosis meeting recommendations of the College of American Pathologists guidelines.

METHODS

Glass slides from clinically reported cases from 5 participating pathologists with a preset washout period were digitally scanned and reviewed in settings identical to typical reporting. Cases were classified as concordant or with minor or major disagreement with the original diagnosis. Randomized subsampling was performed, and mean concordance rates were calculated.

RESULTS

In total, 171 cases were included and distributed equally among participants. For the group as a whole, the mean concordance rate in sampled cases (n = 90) was 83.6% counting all discrepancies and 94.6% counting only major disagreements. The mean pathologist concordance rate in sampled cases (n = 18) ranged from 90.49% to 97%.

CONCLUSIONS

We describe a novel double-blinded method for rapid validation of WSI for primary diagnosis. Our findings highlight the occurrence of a range of diagnostic reproducibility when deploying digital methods.

A second opinion pathology review improves the diagnostic concordance between prostate cancer biopsy and radical prostatectomy specimens

Objectives:

The Gleason scoring system is an essential tool for determining the treatment strategy in prostate cancer (PCa). However, the Gleason grade group (GGG) often differs between needle-core biopsy (NCB) and radical prostatectomy (RP) specimens. We investigated the diagnostic value of a second opinion pathology review using NCB specimens in PCa.

Materials and Methods:

We retrospectively evaluated 882 patients who underwent robot-assisted RP from January 2012 to September 2019. Of these, patients whose original biopsy specimens were obtained from another hospital and reviewed by the urological pathology expert at our institution were included in the study. Patients who received neoadjuvant hormonal therapy were excluded from the study. Weighted kappa (k) coefficients were used to evaluate the diagnostic accuracy of each review.

Results:

A total of 497 patients were included in this study. Substantial agreement (weighted k = 0.783) in the GGG between initial- and second-opinion diagnoses based on NCB specimens was observed in 310 cases (62.4%). Although diagnoses based on a single opinion showed moderate agreement with the GGG of RP specimens (initial: 35.2%, weighted k = 0.522; second opinion; 38.8%, weighted k = 0.560), matching initial and second opinion diagnoses improved the concordance (42.9%, 133/310 cases) to substantial agreement (weighted k = 0.626).

Conclusions:

A second opinion of PCa pathology helps to improve the diagnostic accuracy of NCB specimens. However, over half of diagnoses that matched between the initial and second opinions differed from the diagnosis of RP specimens.

Development and Validation of a Deep Learning Algorithm for Gleason Grading of Prostate Cancer From Biopsy Specimens

Question

How does a deep learning system for assessing prostate biopsy specimens compare with interpretations determined by specialists in urologic pathology and by general pathologists?

Findings

In a validation data set of 752 biopsy specimens obtained from 2 independent medical laboratories and a tertiary teaching hospital, this study found that rate of agreement with subspecialists was significantly higher for the deep learning system than it was for a cohort of general pathologists.

Meaning

The deep learning system warrants evaluation as an assistive tool for improving prostate cancer diagnosis and treatment decisions, especially where subspecialist expertise is unavailable.

Importance

For prostate cancer, Gleason grading of the biopsy specimen plays a pivotal role in determining case management. However, Gleason grading is associated with substantial interobserver variability, resulting in a need for decision support tools to improve the reproducibility of Gleason grading in routine clinical practice.

Objective

To evaluate the ability of a deep learning system (DLS) to grade diagnostic prostate biopsy specimens.

Design, Setting, and Participants

The DLS was evaluated using 752 deidentified digitized images of formalin-fixed paraffin-embedded prostate needle core biopsy specimens obtained from 3 institutions in the United States, including 1 institution not used for DLS development. To obtain the Gleason grade group (GG), each specimen was first reviewed by 2 expert urologic subspecialists from a multi-institutional panel of 6 individuals (years of experience: mean, 25 years; range, 18-34 years). A third subspecialist reviewed discordant cases to arrive at a majority opinion. To reduce diagnostic uncertainty, all subspecialists had access to an immunohistochemical-stained section and 3 histologic sections for every biopsied specimen. Their review was conducted from December 2018 to June 2019.

Main Outcomes and Measures

The frequency of the exact agreement of the DLS with the majority opinion of the subspecialists in categorizing each tumor-containing specimen as 1 of 5 categories: nontumor, GG1, GG2, GG3, or GG4-5. For comparison, the rate of agreement of 19 general pathologists’ opinions with the subspecialists’ majority opinions was also evaluated.

Results

For grading tumor-containing biopsy specimens in the validation set (n = 498), the rate of agreement with subspecialists was significantly higher for the DLS (71.7%; 95% CI, 67.9%-75.3%) than for general pathologists (58.0%; 95% CI, 54.5%-61.4%) (P < .001). In subanalyses of biopsy specimens from an external validation set (n = 322), the Gleason grading performance of the DLS remained similar. For distinguishing nontumor from tumor-containing biopsy specimens (n = 752), the rate of agreement with subspecialists was 94.3% (95% CI, 92.4%-95.9%) for the DLS and similar at 94.7% (95% CI, 92.8%-96.3%) for general pathologists (P = .58).

Conclusions and Relevance

In this study, the DLS showed higher proficiency than general pathologists at Gleason grading prostate needle core biopsy specimens and generalized to an independent institution. Future research is necessary to evaluate the potential utility of using the DLS as a decision support tool in clinical workflows and to improve the quality of prostate cancer grading for therapy decisions.

Interchangeability of light and virtual microscopy for histopathological evaluation of prostate cancer

Virtual microscopy (VM) holds promise to reduce subjectivity as well as intra- and inter-observer variability for the histopathological evaluation of prostate cancer. We evaluated (i) the repeatability (intra-observer agreement) and reproducibility (inter-observer agreement) of the 2014 Gleason grading system and other selected features using standard light microscopy (LM) and an internally developed VM system, and (ii) the interchangeability of LM and VM. Two uro-pathologists reviewed 413 cores from 60 Swedish men diagnosed with non-metastatic prostate cancer 1998–2014. Reviewer 1 performed two reviews using both LM and VM. Reviewer 2 performed one review using both methods. The intra- and inter-observer agreement within and between LM and VM were assessed using Cohen’s kappa and Bland and Altman’s limits of agreement. We found good repeatability and reproducibility for both LM and VM, as well as interchangeability between LM and VM, for primary and secondary Gleason pattern, Gleason Grade Groups, poorly formed glands, cribriform pattern and comedonecrosis but not for the percentage of Gleason pattern 4. Our findings confirm the non-inferiority of VM compared to LM. The repeatability and reproducibility of percentage of Gleason pattern 4 was poor regardless of method used warranting further investigation and improvement before it is used in clinical practice.

Interobserver reproducibility of perineural invasion of prostatic adenocarcinoma in needle biopsies

Numerous studies have shown a correlation between perineural invasion (PNI) in prostate biopsies and outcome. The reporting of PNI varies widely in the literature. While the interobserver variability of prostate cancer grading has been studied extensively, less is known regarding the reproducibility of PNI. A total of 212 biopsy cores from a population-based screening trial were included in this study (106 with and 106 without PNI according to the original pathology reports). The glass slides were scanned and circulated among four pathologists with a special interest in urological pathology for assessment of PNI. Discordant cases were stained by immunohistochemistry for S-100 protein. PNI was diagnosed by all four observers in 34.0% of cases, while 41.5% were considered to be negative for PNI. In 24.5% of cases, there was a disagreement between the observers. The kappa for interobserver variability was 0.67–0.75 (mean 0.73). The observations from one participant were compared with data from the original reports, and a kappa for intraobserver variability of 0.87 was achieved. Based on immunohistochemical findings among discordant cases, 88.6% had PNI while 11.4% did not. The most common diagnostic pitfall was the presence of bundles of stroma or smooth muscle. It was noted in a few cases that collagenous micronodules could be mistaken for a nerve. The distance between cancer and nerve was another cause of disagreement. Although the results suggest that the reproducibility of PNI may be greater than that of prostate cancer grading, there is still a need for improvement and standardization.

A grading dilemma; Gleason scoring system: Are we sufficiently compatible? A multi center study

Objective

Gleason scoring is the grading system which strongly predicts the prognosis of prostate cancer. However, even being one of the most commonly used systems, the presence of different interobserver agreement rates push the uropathologists update the definitons of the Gleason patterns. In this study, we aimed to determine the interobserver agreement variability among 7 general pathologists, and one expert uropathologist from 6 different centers.

Methods

A set of 50 Hematoxylin & Eosin stained slides from 41 patients diagnosed as prostate cancer were revised by 8 different pathologists. The pathologists were also grouped according to having their residency at the same institute or working at the same center. All pathologists' and the subgroups' Gleason scores were then compared for interobserver variability by Fleiss' and Cohen's kappa tests using R v3.2.4.

Results

There were about 8 pathologists from 6 different centers revised all the slides. One of them was an expert uropathologist with experience of 18 years. Among 7 general pathologists 4 had surgical pathology experience for over 5 years whilst 3 had under 5 years. The Fleiss' kappa was found as 0.54 for primary Gleason pattern, and 0.44 for total Gleason score (moderate agreement). The Fleiss' kappa was 0.45 for grade grouping system.

Conclusion

Assigning a Gleason score for a patient can be problematic because of different interobserver agreement rates among pathologists even though the patterns were accepted as well-defined.

Evaluation of the Use of Combined Artificial Intelligence and Pathologist Assessment to Review and Grade Prostate Biopsies

IMPORTANCE

Expert-level artificial intelligence (AI) algorithms for prostate biopsy grading have recently been developed. However, the potential impact of integrating such algorithms into pathologist workflows remains largely unexplored.

OBJECTIVE

To evaluate an expert-level AI-based assistive tool when used by pathologists for the grading of prostate biopsies.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study used a fully crossed multiple-reader, multiple-case design to evaluate an AI-based assistive tool for prostate biopsy grading. Retrospective grading of prostate core needle biopsies from 2 independent medical laboratories in the US was performed between October 2019 and January 2020. A total of 20 general pathologists reviewed 240 prostate core needle biopsies from 240 patients. Each pathologist was randomized to 1 of 2 study cohorts. The 2 cohorts reviewed every case in the opposite modality (with AI assistance vs without AI assistance) to each other, with the modality switching after every 10 cases. After a minimum 4-week washout period for each batch, the pathologists reviewed the cases for a second time using the opposite modality. The pathologist-provided grade group for each biopsy was compared with the majority opinion of urologic pathology subspecialists.

EXPOSURE

An AI-based assistive tool for Gleason grading of prostate biopsies.

MAIN OUTCOMES AND MEASURES

Agreement between pathologists and subspecialists with and without the use of an AI-based assistive tool for the grading of all prostate biopsies and Gleason grade group 1 biopsies.

RESULTS

Biopsies from 240 patients (median age, 67 years; range, 39-91 years) with a median prostate-specific antigen level of 6.5 ng/mL (range, 0.6-97.0 ng/mL) were included in the analyses. Artificial intelligence-assisted review by pathologists was associated with a 5.6% increase (95% CI, 3.2%-7.9%; P < .001) in agreement with subspecialists (from 69.7% for unassisted reviews to 75.3% for assisted reviews) across all biopsies and a 6.2% increase (95% CI, 2.7%-9.8%; P = .001) in agreement with subspecialists (from 72.3% for unassisted reviews to 78.5% for assisted reviews) for grade group 1 biopsies. A secondary analysis indicated that AI assistance was also associated with improvements in tumor detection, mean review time, mean self-reported confidence, and interpathologist agreement.

CONCLUSIONS AND RELEVANCE

In this study, the use of an AI-based assistive tool for the review of prostate biopsies was associated with improvements in the quality, efficiency, and consistency of cancer detection and grading.

Histopathological re-evaluations of biopsies in prostate cancer: a nationwide observational study

BACKGROUND

Grading prostate biopsies has an important role in determining treatment strategy. Histopathological evaluations suffer from interobserver variability and therefore biopsies may be re-evaluated.

OBJECTIVE

To provide insight into the extent of, characteristics associated with and clinical implications of prostate biopsy re-evaluations in daily clinical practice.

METHODS

Patients diagnosed with prostate cancer (PCa) by biopsy between October 2015 and April 2016 identified through the Netherlands Cancer Registry were included. The proportion of re-evaluations was assessed and characteristics were compared between patients with and without biopsy re-evaluation. Interobserver concordance of ISUP grade and EAU prognostic risk classification was determined by calculating Cohen's kappa.

RESULTS

Biopsy re-evaluation was performed in 172 (3.3%) of 5214 patients. Primary reason for re-evaluation in patients treated with curative intent was referral to another hospital. Most referred patients treated with curative intent (n = 1856) had no re-evaluation (93.0%, n = 1727). Patients with biopsy re-evaluation were younger and underwent more often prostatectomy compared to patients without re-evaluation. The disagreement rate for ISUP grade was 26.1% and interobserver concordance was substantial (κ-weighted = 0.74). Re-evaluation resulted in 21.1% (n = 14) of patients with localised PCa in a different prognostic risk group. More tumours were downgraded (57.1%) than upgraded (42.9%). Interobserver concordance was very good (κ _weighted = 0.85).

CONCLUSION

Pathology review of prostate biopsies is infrequently requested by clinicians in the Netherlands but in a non-negligible minority of patients with localised PCa the pathology review led to a change in prognostic risk group which might impact their treatment.

Inter-observer variability of cribriform architecture and percent Gleason pattern 4 in prostate cancer: relation to clinical outcome

The Grade group is an important parameter for clinical decision-making in prostate cancer. Recently, percent Gleason pattern 4 and presence of invasive cribriform and/or intraductal carcinoma (CR/IDC) have been recognized for their independent predictive value for prostate cancer outcome. There is sparse data on the inter-observer agreement for these pathologic features in practice. Our objectives were to investigate inter-observer variability of percent Gleason pattern and CR/IDC and to relate individual tumour scores to clinical outcome. Our cohort included 80 consecutive radical prostatectomies with a median follow-up 87.1 months (interquartile range 43.3-119.2), of which the slide with largest tumour volume was scored by six pathologists for Grade group (four tiers: 1, 2, 3 and 4/5), percent Gleason pattern 4 (four tiers: 0-25%, 26-50%, 51-75% and 76-100%) and presence of CR/IDC (two tiers: absent, present). The individual assignments were related to post-operative biochemical recurrence (20/80). Inter-observer agreement was substantial (Krippendorff's α 0.626) for assessment of Grade group and moderate for CR/IDC (α 0.507) and percent Gleason pattern 4 (α 0.551). For each individual pathologist, biochemical recurrence rates incremented by Grade group and presence of CR/IDC, although such relation was less clear for percent Gleason pattern 4. In conclusion, inter-observer agreement for CR/IDC and percent Gleason pattern 4 is lower than for Grade groups, indicating awareness of these features needs further improvement. Grade group and CR/IDC, but not percent Gleason pattern 4 was related to biochemical recurrence for each pathologist, indicating overall validity of individual grade assignments despite inter-observer variability.

Identification of areas of grading difficulties in prostate cancer and comparison with artificial intelligence assisted grading

The International Society of Urological Pathology (ISUP) hosts a reference image database supervised by experts with the purpose of establishing an international standard in prostate cancer grading. Here, we aimed to identify areas of grading difficulties and compare the results with those obtained from an artificial intelligence system trained in grading. In a series of 87 needle biopsies of cancers selected to include problematic cases, experts failed to reach a 2/3 consensus in 41.4% (36/87). Among consensus and non-consensus cases, the weighted kappa was 0.77 (range 0.68-0.84) and 0.50 (range 0.40-0.57), respectively. Among the non-consensus cases, four main causes of disagreement were identified: the distinction between Gleason score 3 + 3 with tangential cutting artifacts vs. Gleason score 3 + 4 with poorly formed or fused glands (13 cases), Gleason score 3 + 4 vs. 4 + 3 (7 cases), Gleason score 4 + 3 vs. 4 + 4 (8 cases) and the identification of a small component of Gleason pattern 5 (6 cases). The AI system obtained a weighted kappa value of 0.53 among the non-consensus cases, placing it as the observer with the sixth best reproducibility out of a total of 24. AI may serve as a decision support and decrease inter-observer variability by its ability to make consistent decisions. The grading of these cancer patterns that best predicts outcome and guides treatment warrants further clinical and genetic studies. Results of such investigations should be used to improve calibration of AI systems.

Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study

BACKGROUND

An increasing volume of prostate biopsies and a worldwide shortage of urological pathologists puts a strain on pathology departments. Additionally, the high intra-observer and inter-observer variability in grading can result in overtreatment and undertreatment of prostate cancer. To alleviate these problems, we aimed to develop an artificial intelligence (AI) system with clinically acceptable accuracy for prostate cancer detection, localisation, and Gleason grading.

METHODS

We digitised 6682 slides from needle core biopsies from 976 randomly selected participants aged 50-69 in the Swedish prospective and population-based STHLM3 diagnostic study done between May 28, 2012, and Dec 30, 2014 (ISRCTN84445406), and another 271 from 93 men from outside the study. The resulting images were used to train deep neural networks for assessment of prostate biopsies. The networks were evaluated by predicting the presence, extent, and Gleason grade of malignant tissue for an independent test dataset comprising 1631 biopsies from 246 men from STHLM3 and an external validation dataset of 330 biopsies from 73 men. We also evaluated grading performance on 87 biopsies individually graded by 23 experienced urological pathologists from the International Society of Urological Pathology. We assessed discriminatory performance by receiver operating characteristics and tumour extent predictions by correlating predicted cancer length against measurements by the reporting pathologist. We quantified the concordance between grades assigned by the AI system and the expert urological pathologists using Cohen's kappa.

FINDINGS

The AI achieved an area under the receiver operating characteristics curve of 0·997 (95% CI 0·994-0·999) for distinguishing between benign (n=910) and malignant (n=721) biopsy cores on the independent test dataset and 0·986 (0·972-0·996) on the external validation dataset (benign n=108, malignant n=222). The correlation between cancer length predicted by the AI and assigned by the reporting pathologist was 0·96 (95% CI 0·95-0·97) for the independent test dataset and 0·87 (0·84-0·90) for the external validation dataset. For assigning Gleason grades, the AI achieved a mean pairwise kappa of 0·62, which was within the range of the corresponding values for the expert pathologists (0·60-0·73).

INTERPRETATION

An AI system can be trained to detect and grade cancer in prostate needle biopsy samples at a ranking comparable to that of international experts in prostate pathology. Clinical application could reduce pathology workload by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores. An AI system with expert-level grading performance might contribute a second opinion, aid in standardising grading, and provide pathology expertise in parts of the world where it does not exist.

FUNDING

Swedish Research Council, Swedish Cancer Society, Swedish eScience Research Center, EIT Health.

Prognostic value of DNA ploidy and automated assessment of stroma fraction in prostate cancer

The combination of DNA ploidy and automatically estimated stroma fraction has been shown to correlate with recurrence and cancer death in colorectal cancer. We aimed to extend this observation and evaluate the prognostic importance of this combined marker in prostate cancer. DNA ploidy status was determined by image cytometry and the stroma fraction was estimated automatically on hematoxylin and eosin stained sections in three tumor samples from each patient to account for tumor heterogeneity. The optimal threshold for low (≤56%) and high (>56%) stroma fraction was identified in a discovery cohort (n = 253). The combined marker was validated in an independent patient cohort (n = 259) with biochemical recurrence as endpoint. The combined marker predicted biochemical recurrence independently in the validation cohort. Multivariable analysis showed that the highest risk of recurrence was observed for patients with samples that had both non-diploid ploidy status and a high stroma fraction (hazard ratio: 2.51, 95% confidence interval: 1.18-5.34). In conclusion, we suggest the combination of DNA ploidy and automatically estimated stroma fraction as a prognostic marker for the risk stratification of prostate cancer patients. It may also be a potential generic marker as concurrent results have been described in colorectal cancer.

Prostate health index density predicts aggressive pathological outcomes after radical prostatectomy in Taiwanese patients

BACKGROUND

There are models to predict pathological outcomes based on established clinical and prostate-specific antigen (PSA)-derived parameters; however, they are not satisfactory. p2PSA and its derived biomarkers have shown promise for the diagnosis and prognosis of prostate cancer (PCa). The aim of this study was to investigate whether p2PSA-derived biomarkers can assist in the prediction of aggressive pathological outcomes after radical prostatectomy (RP).

METHODS

We prospectively enrolled patients who were diagnosed with PCa and treated with RP between February 2017 and December 2018. Preoperative blood samples were analyzed for tPSA, free PSA (fPSA), percentage of fPSA (%fPSA), [-2]proPSA (p2PSA), and percentage of p2PSA (%p2PSA). Prostate health index (PHI) was calculated as (p2PSA/fPSA) × √tPSA. Prostate volume was determined by transrectal ultrasound using the ellipsoid formula, and PHI density was calculated as PHI/prostate volume. The areas under the receiver operating characteristic curve were estimated for various PSA/p2PSA derivatives. Aggressive pathological outcomes measured after RP were defined as pathological T3 or a Gleason score (GS) >6 as determined in RP specimens.

RESULTS

One hundred and forty-four patients were included for analysis. Postoperative GS was >6 in 86.1% of the patients, and pT stage was T3a or more in 54.2%. Among all PSA- and p2PSA-derived biomarkers, PHI density was the best biomarker to predict aggressive pathological outcomes after RP. The odds ratio of having an aggressive pathological outcome of RP was 8.796 (p = 0.001). In multivariate analysis, adding %fPSA to base model did not improve the accuracy (area under curve), but adding PHI and PHI density to base model improved the accuracy by 2% and 16%, respectively, in predicting pT3 stage or GS ≥ 7. The risk of pT3 stage or GS ≥ 7 was 20.8% for PHI density <1.125, and 64.6% for PHI density >1.125 (sensitivity: 74.6% and specificity: 88.9%).

CONCLUSION

PHI density may further aid in predicting aggressive pathological outcomes after RP. This biomarker may be useful in preoperative counseling and may have potential in decision making when choosing between definitive treatment and active surveillance of newly diagnosed PCa.

Pathological upgrading in prostate cancer treated with surgery in the United Kingdom: trends and risk factors from the British Association of Urological Surgeons Radical Prostatectomy Registry

Background

Accurate grading at the time of diagnosis if fundamental to risk stratification and treatment decision making in patients with prostate cancer. Whilst previous studies have demonstrated significant pathological upgrading and downgrading following radical prostatectomy (RP), these were based on historical cohorts and do not reflect contemporary patient selection and management practices. The aim of this national, multicentre observational study was to characterise contemporary rates and risk factors for pathological upgrading after RP in the United Kingdom (UK).

Methods

All RP entries on the British Association of Urological Surgeons (BAUS) Radical Prostatectomy Registry database of prospectively entered cases undertaken between January 2011 and December 2016 were extracted. Those patients with full preoperative PSA, clinical stage, needle biopsy and subsequent RP pathological grade information were included. Upgrade was defined as any increase in Gleason grade from initial needle biopsy to pathological assessment of the entire surgical specimen. Statistical analysis and multivariate logistic regression were undertaken using R version 3.5 (R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 17,598 patients met full inclusion criteria. Absolute concordance between initial biopsy and pathological grade was 58.9% (n = 10,364), whilst upgrade and downgrade rates were 25.5% (n = 4489) and 15.6% (n = 2745) respectively. Upgrade rate was highest in those with D’Amico low risk compared with intermediate and high-risk disease (55.7% versus 19.1 and 24.3% respectively, P < 0.001). Although rates varied between year of surgery and geographical regions, these differences were not significant after adjusting for other preoperative diagnostic variables using multivariate logistic regression.

Conclusions

Pathological upgrading after RP in the UK is lower than expected when compared with other large contemporary series, despite operating on a generally higher risk patient cohort. As new diagnostic techniques that may reduce rates of pathological upgrading become more widely utilised, this study provides an important benchmark against which to measure future performance.

Prognostic value and reproducibility of different microscopic characteristics in the WHO grading systems for pTa and pT1 urinary bladder urothelial carcinomas

Background

European treatment guidelines for pTa and pT1 urinary bladder urothelial carcinoma depend highly on stage and WHO-grade. Both the WHO73 and the WHO04 grading systems show some intra- and interobserver variability. The current pilot study investigates which histopathological features are especially sensitive for this undesired lack of reproducibility and the influence on prognostic value.

Methods

Thirty-eight cases of primary non-muscle invasive urothelial carcinomas, including thirteen cases with stage progression, were reviewed by three pathologists. Thirteen microscopic features were extracted from pathology textbooks and evaluated separately. Reproducibility was measured using Gwet’s agreement coefficients. Prognostic ability regarding progression was estimated by the area under curve (AUC) of the receiver operating characteristics (ROC) function.

Results

The best reproducible features (Gwet’s agreement coefficient above 0.60) were papillary architecture, nuclear polarity, cellular maturation, nuclear enlargement and giant nuclei. Nucleoli was the strongest prognostic feature, and the only feature with an AUC above 0.70 for both grading systems, but reproducibility was not among the strongest. Nuclear polarity also had prognostic value with an AUC of 0.70 and 0.67 for the WHO73 and WHO04, respectively. The other features did not have significant prognostic value.

Conclusions

The reproducibility of the histopathological features of the different WHO grading systems varied considerably. Of all the features evaluated, only nuclear polarity was both prognostic and significantly reproducible. Further validation studies are needed on these features to improve grading of urothelial carcinomas.

Precision and accuracy of magnetic resonance imaging for lobar classification of benign prostatic hyperplasia

PURPOSE

To validate the application of a magnetic resonance imaging (MRI)-based lobar classification of benign prostatic hyperplasia (BPH) for use in research and clinical management.

METHODS

Two radiologists with 5 and 11 years post-fellowship experience were trained in the lobar classification of BPH using an internally developed atlas of prostate anatomy with example MRI images edited by a third senior radiologist designated as the "administrator" of the study. A study population of 140 patients referred to a tertiary academic medical center with known or suspected prostate cancer was selected by the administrator to test the interrater reliability (IRR; precision) of the classification as well as accuracy of the two readers compared to the administrator as the "gold" standard. The intrarater reliability of repeat readings of the administrator was also examined. Percentage of agreement, proportion of agreement, and Cohen's κ were applied. This was a retrospective IRB-approved study.

RESULTS

IRR (precision) between the two interpreting radiologists was 64% agreement, corresponding to unweighted κ of 0.52. Composite proportion of agreement across all BPH types (categories) for interpreting radiologists was 0.67. Observer accuracy was 62% agreement, unweighted κ 0.49, for observer 1 and 67%, unweighted κ 0.58, for observer 2. Intrarater reliability for the administrator was 87% agreement, unweighted κ 0.81 with composite proportion of agreement across all categories of 0.87.

CONCLUSIONS

MRI lobar classification of BPH is a reproducible and reliable tool for research and clinical applications.

Prostatic Adenocarcinoma: A Grading from Gleason to the New Grade-Group System: A Historical and Critical Review

The introduction of the Gleason grading system revolutionised prognostic parameters and determination of patient treatment regiments for prostatic adenocarcinomas, and has become synonymous with prostate cancer, almost universally applied in clinical settings to predict radical prostatectomy specimen findings, potential biochemical failure, local recurrences, lymph nodes or distant metastases in patients not receiving any treatment as well as those receiving treatment including radiation therapy, surgical treatment such as radical prostatectomy and other therapies etc,. However, characterisation and classification of prostate cancer is very different compared to 40-50 years ago when Gleason scores were first introduced. Despite this radical shift in classification, the Gleason system has remained one of the most important prognostic factors in prostate cancer, only possible as a result of timely and appropriate modifications to this characterisation system made in 2005 and 2014. However, even after these modifications, certain limitations of the Gleason system remain, due to which a new prostate cancer prognostic grade group system was introduced in 2014, which was widely accepted in the 2014 ISUP consensus conference, and incorporated into the WHO classification of thetumor of the Urinary System and Male Genital Tract in 2016. Herein, this article will discuss how this new prognostic grade group system, which is regarded as simpler and more accurate than the Gleason system risk stratification groups, will be used in conjunction with the Gleason system to improve patient prognosis and treatment.

Circulating miRNAs in localized/locally advanced prostate cancer patients after radical prostatectomy and radiotherapy

BACKGROUND

Overtreatment is a well-known clinical challenge in local prostate cancer (PCa). Although risk assessment models have contributed to a better stratification of patients with local PCa, a tailored management is still in its infancy. Over the last few decades, microRNAs (miRNAs) have shown promising results as biomarkers in PCa. The aim of this study was to investigate circulating miRNAs after management of local PCa.

METHODS

The relative expression of four miRNAs (miRNA-21, -93, -125b, and miRNA-221) was assessed in plasma from 149 newly diagnosed patients with local or locally advanced PCa. Real-time polymerase chain reaction was used for analysis. A baseline sample at time of diagnosis and a follow-up sample after 6 months were assessed. The patients were grouped in an interventional cohort (radical prostatectomy, curative intent radiotherapy, or androgen-deprivation therapy alone) and an observational cohort (watchful waiting or active surveillance).

RESULTS

In the interventional cohort, levels of both miRNA-93 and miRNA-221 were significantly lower in the follow-up samples compared to baseline z = -2.738, P = 0.006, and z = -4.498, P < 0.001, respectively. The same observation was recorded for miRNA-125b in the observational cohort (z = -2.656, P = 0.008). Both miRNA-125b and miRNA-221 were correlated with risk assessment r = 0.23, P = 0.015, and r = 0.203, P = 0.016 respectively, while miRNA-93 showed tendency to significant correlation with the prostatectomy Gleason score (r = 0.276, P = 0.0576).

CONCLUSIONS

The current results indicate a possible role of miRNA-93 and miRNA-221 in disease monitoring in localized and locally advanced PCa. Larger studies are warranted to assess the clinical impact of these biomarkers.

PROSTATEx Challenges for computerized classification of prostate lesions from multiparametric magnetic resonance images

Grand challenges stimulate advances within the medical imaging research community; within a competitive yet friendly environment, they allow for a direct comparison of algorithms through a well-defined, centralized infrastructure. The tasks of the two-part PROSTATEx Challenges (the PROSTATEx Challenge and the PROSTATEx-2 Challenge) are (1) the computerized classification of clinically significant prostate lesions and (2) the computerized determination of Gleason Grade Group in prostate cancer, both based on multiparametric magnetic resonance images. The challenges incorporate well-vetted cases for training and testing, a centralized performance assessment process to evaluate results, and an established infrastructure for case dissemination, communication, and result submission. In the PROSTATEx Challenge, 32 groups apply their computerized methods (71 methods total) to 208 prostate lesions in the test set. The area under the receiver operating characteristic curve for these methods in the task of differentiating between lesions that are and are not clinically significant ranged from 0.45 to 0.87; statistically significant differences in performance among the top-performing methods, however, are not observed. In the PROSTATEx-2 Challenge, 21 groups apply their computerized methods (43 methods total) to 70 prostate lesions in the test set. When compared with the reference standard, the quadratic-weighted kappa values for these methods in the task of assigning a five-point Gleason Grade Group to each lesion range from - 0.24 to 0.27; superiority to random guessing can be established for only two methods. When approached with a sense of commitment and scientific rigor, challenges foster interest in the designated task and encourage innovation in the field.

Histopathology: ditch the slides, because digital and 3D are on show

Due to the growing field of digital pathology, more and more digital histology slides are becoming available. This improves the accessibility, allows teleconsultations from specialized pathologists, improves education, and might give urologist the possibility to review the slides in patient management systems. Moreover, by stacking multiple two-dimensional (2D) digital slides, three-dimensional volumes can be created, allowing improved insight in the growth pattern of a tumor. With the addition of computer-aided diagnosis systems, pathologist can be guided to regions of interest, potentially reducing the workload and interobserver variation. Digital (3D) pathology has the potential to improve dialog between the pathologist and urologist, and, therefore, results in a better treatment selection for urologic patients.

Accuracy of standardized 12-core template biopsies versus non-standardized biopsies for detection of Epstein Grade 5 prostate cancer regarding the histology of the prostatectomy specimen

OBJECTIVE

To evaluate the effectiveness of EAU Guideline compliant transrectal ultrasound-guided 12-core prostate biopsies for detection of highly aggressive Epstein Grade 5 (Gleason Score 9-10) prostate cancer.

METHODS

Two hundred ninety-nine patients, treated by radical prostatectomy for prostate cancer, have been prospectively recorded in a database and were evaluated for this study. Pre-operatively, all patients received transrectal ultrasound-guided biopsies according to inhomogeneous templates chosen by the referring urologist. We evaluated the outcomes according to a stratified group-analysis: Group 1 received less than 12 biopsies, Group 2 received more than 12 biopsies, and Group 3 received exactly 12 biopsies, according to the EAU Guidelines template. After surgical removal of the prostate, 12 EAU Guideline-templated biopsies were performed in all prostatectomy specimens, directly after the surgery. Pre-operative and post-operative Epstein Grade 5 biopsy detection rates were thereafter correlated with these prostatectomy specimens.

RESULTS

In prostatectomy specimens, the histology of 12 patients (4.0%) were Epstein Grade 1, 31 patients (10.5%) were Epstein Grade 2, 190 patients (63.5%) were Epstein Grade 3, 27 patients (9%) were Epstein Grade 4, and 39 patients (13%) were Epstein Grade 5. The detection rate of Epstein Grade 5 compared to the radical prostatectomy specimen was: Group 1: 23.0% pre-operatively and 61.5% post-operatively, Group 2: 33.3% pre-operatively and 58.3% post-operatively; and Group 3: 57.1% pre-operatively and 64.2% post-operatively.

CONCLUSION

Detection rates of highly aggressive Epstein Grade 5 prostate cancer vary considerably according to the biopsy technique. EAU Guideline compliant 12-core template biopsies increase the detection rates of Epstein Grade 5 prostate cancer.

Standardized Interpretation of Chest Radiographs in Cases of Pediatric Pneumonia From the PERCH Study

Background.

Chest radiographs (CXRs) are a valuable diagnostic tool in epidemiologic studies of pneumonia. The World Health Organization (WHO) methodology for the interpretation of pediatric CXRs has not been evaluated beyond its intended application as an endpoint measure for bacterial vaccine trials.

Methods.

The Pneumonia Etiology Research for Child Health (PERCH) study enrolled children aged 1–59 months hospitalized with WHO-defined severe and very severe pneumonia from 7 low- and middle-income countries. An interpretation process categorized each CXR into 1 of 5 conclusions: consolidation, other infiltrate, both consolidation and other infiltrate, normal, or uninterpretable. Two members of a 14-person reading panel, who had undertaken training and standardization in CXR interpretation, interpreted each CXR. Two members of an arbitration panel provided additional independent reviews of CXRs with discordant interpretations at the primary reading, blinded to previous reports. Further discordance was resolved with consensus discussion.

Results.

A total of 4172 CXRs were obtained from 4232 cases. Observed agreement for detecting consolidation (with or without other infiltrate) between primary readers was 78% (κ = 0.50) and between arbitrators was 84% (κ = 0.61); agreement for primary readers and arbitrators across 5 conclusion categories was 43.5% (κ = 0.25) and 48.5% (κ = 0.32), respectively. Disagreement was most frequent between conclusions of other infiltrate and normal for both the reading panel and the arbitration panel (32% and 30% of discordant CXRs, respectively).

Conclusions.

Agreement was similar to that of previous evaluations using the WHO methodology for detecting consolidation, but poor for other infiltrates despite attempts at a rigorous standardization process.

The updated grading system of prostate carcinoma: an inter-observer agreement study among general pathologists in an academic practice

In 2016, the grading criteria for Gleason scoring (GS) have been updated in the WHO classification of tumors of the prostate, and a new set of grade groups (GG) was introduced. As the inter-observer discordance is a well-known concern in Gleason grading before the update and no reproducibility study testing the grade groups exists, we planned to evaluate the inter-observer agreement of the most updated grading system. Four pathologists assessed 126 cores of prostatic carcinoma, and Kappa (k) test was calculated. The agreements for both GS and GG were substantial (k = 0.753 and 0.752; respectively). Discerning GG 2 from 3 also attained reasonable outcome (k = 0.675). Based on our results, the updated grading system seems to be reproducible, with satisfactory inter-observer concordance rate.

Heterogeneity of miRNA expression in localized prostate cancer with clinicopathological correlations

INTRODUCTION

In the last decade microRNAs (miRNAs) have been widely investigated in prostate cancer (PCa) and have shown to be promising biomarkers in diagnostic, prognostic and predictive settings. However, tumor heterogeneity may influence miRNA expression. The aims of this study were to assess the impact of tumor heterogeneity, as demonstrated by a panel of selected miRNAs in PCa, and to correlate miRNA expression with risk profile and patient outcome.

MATERIAL AND METHODS

Prostatectomy specimens and matched, preoperative needle biopsies from a retrospective cohort of 49 patients, who underwent curatively intended surgery for localized PCa, were investigated with a panel of 6 miRNAs (miRNA-21, miRNA-34a, miRNA-125b, miRNA-126, miRNA-143, and miRNA-145) using tissue micro-array (TMA) and in situ hybridization (ISH). Inter- and intra-patient variation was assessed using intra-class correlation (ICC).

RESULTS

Four miRNAs (miRNA-21, miRNA-34a, miRNA-125, and miRNA-126) were significantly upregulated in PCa compared to benign prostatic hyperplasia (BPH), and except for miRNA-21 these miRNAs documented a positive correlation between the expression level in PCa cores and their matched BPH cores, (r > 0.72). The ICC varied from 0.451 to 0.764, with miRNA-34a showing an intra-tumoral heterogeneity accounting for less than 50% of the total variation. Regarding clinicopathological outcomes, only miRNA-143 showed potential as a prognostic marker with a higher expression correlating with longer relapse-free survival (p = 0.016).

CONCLUSION

The present study documents significant upregulation of the expression of miRNA-21, miRNA-34a, miRNA-125, and miRNA-126 in PCa compared to BPH and suggests a possible prognostic value associated with the expression of miRNA-143. The results, however, document intra-tumoral heterogeneity in the expression of various miRNAs calling for caution when using these tumor tissue biomarkers in prognostic and predictive settings.

Nuclear morphometry in histological specimens of canine prostate cancer: Correlation with histological subtypes, Gleason score, methods of collection and survival time

Ten normal prostates, 22 benign prostatic hyperplasia (BPH) and 29 prostate cancer (PC) were morphometrically analyzed with regard to mean nuclear area (MNA), mean nuclear perimeter (MNP), mean nuclear diameter (MND), coefficient of variation of the nuclear area (NACV), mean nuclear diameter maximum (MDx), mean nuclear diameter minimum (MDm), mean nuclear form ellipse (MNFe) and form factor (FF). The relationship between nuclear morphometric parameters and histological type, Gleason score, methods of sample collection, presence of metastases and survival time of canine PC were also investigated. Overall, nuclei from neoplastic cells were larger, with greater variation in nuclear size and shape compared to normal and hyperplastic cells. Significant differences were found between more (small acinar/ductal) and less (cribriform, solid) differentiated PCs with regard to FF (p<0.05). MNA, MNP, MND, MDx, and MDm were significantly correlated with the Gleason score of PC (p<0.05). MNA, MNP, MDx and MNFe may also have important prognostic implications in canine prostatic cancer since negatively correlated with the survival time. Biopsy specimens contained nuclei that were smaller and more irregular in comparison to those in prostatectomy and necropsy specimens and therefore factors associated with tissue sampling and processing may influence the overall morphometric evaluation. The results indicate that nuclear morphometric analysis in combination with Gleason score can help in canine prostate cancer grading, thus contributing to the establishment of a more precise prognosis and patient's management.

Accuracy of Grading Gleason Score 7 Prostatic Adenocarcinoma on Needle Biopsy: Influence of Percent Pattern 4 and Other Histological Factors

BACKGROUND

Recognition of Gleason pattern 4 in prostatic needle biopsies is crucial for both prognosis and therapy. Recently, it has been recommended to record percent pattern 4 when Gleason score 7 cancer is the highest grade in a case.

METHODS

Four hundred and five prostate needle core biopsies received for a second opinion at our institution from February-June, 2015 were prospectively diagnosed with prostatic adenocarcinoma Gleason score 7 as the highest score on review by a consultant urological pathologist. Percentage of core involvement by cancer, percentage of Gleason pattern 4 per core, distribution of Gleason pattern 4 (clustered, scattered), morphology of pattern 4 (cribriform, non-cribriform), and whether the cancer was continuous or discontinuous were recorded.

RESULTS

Better agreement was noted between the consultant and referring pathologists when pattern 4 was clustered as opposed to dispersed in biopsies (P = 0.009). The percentage of core involvement by cancer, morphology of pattern 4, and continuity of cancer did not affect the agreement between the consultant and referring pathologists. There was a trend (P = 0.06) for better agreement based on the percent of pattern 4.

CONCLUSIONS

When pattern 4 is scattered amongst pattern 3 as opposed to being discrete foci, there is less interobserver reproducibility in grading Gleason score 7 cancer, and in this setting pathologists should consider obtaining second opinions either internally within their group or externally. Prostate 77: 681-685, 2017. © 2017 Wiley Periodicals, Inc.

Grade Group Underestimation in Prostate Biopsy: Predictive Factors and Outcomes in Candidates for Active Surveillance

OBJECTIVE

We intended to analyze the outcomes and predictive factors for underestimating the prostate cancer (PCa) grade group (GG) from prostate biopsies in a large monocentric cohort of patients treated by minimally invasive radical prostatectomy (RP).

MATERIALS AND METHODS

Using a monocentric prospectively maintained database, we included 3062 patients who underwent minimally invasive RP between 2006 and 2013. We explored clinicopathologic features and outcomes associated with a GG upgrade from biopsy to RP. Multivariate logistic regression was used to develop and validate a nomogram to predict upgrading for GG1.

RESULTS

Biopsy GG was upgraded after RP in 51.5% of cases. Patients upgraded from GG1 to GG2 or GG3 after RP had a longer time to biochemical recurrence than those with GG2 or GG3 respectively, on both biopsy and RP, but a shorter time to biochemical recurrence than those who remained GG1 after RP (P < .0001). In multivariate analyses, variables predicting upgrading for GG1 PCa were age (P = .0014), abnormal digital rectal examination (P < .0001), prostate-specific antigen density (P < .0001), percentage of positive cores (P < .0001), and body mass index (P = .037). A nomogram was generated and validated internally.

CONCLUSIONS

Biopsy grading system is misleading in approximately 50% of cases. Upgrading GG from biopsy to RP may have consequences on clinical outcomes. A nomogram using clinicopathologic features could aid the probability of needing to upgrade GG1 patients at their initial evaluation.

Interobserver variability in Gleason histological grading of prostate cancer

OBJECTIVE

The aims of this study were to evaluate the reproducibility of the Gleason grading system and to compare its interobserver variability with the novel Gleason grade grouping proposal using a large sample volume.

MATERIALS AND METHODS

In total, 407 pathology slides of prostate needle biopsies from 34 consecutive patients with prostate cancer were re-evaluated. The International Society of Urological Pathology 2005 modified Gleason grading system with Epstein's modification was used. Two pathologists, blind to each other and to the initial pathology report, performed the pathological evaluation. To determine interobserver concordance, the kappa (κ) coefficient test was used.

RESULTS

Pathologist 1 and pathologist 2 detected a tumor in 202 and 231 cores, respectively (p < 0.001). The two pathologists disagreed on the presence of a tumor in 31 cores. Of these 31 cores, 74% (n = 23/31) were Gleason pattern 3. The mean length of the cancer foci in these 31 disputed cores was 1.54 ± 0.8 mm. Concordance rates between the two observers for primary and secondary Gleason patterns were 63.96% (κ = 0.34) and 63.45% (κ = 0.37), respectively. Concordance with respect to the Gleason sum was 57.9% (κ = 0.43). When the Gleason scores were classified into the novel Gleason grade grouping, concordance was found to be 51.7% (κ = 0.39).

CONCLUSIONS

The agreement between observers on the Gleason sum was moderate. The novel Gleason grade grouping did not improve interobserver agreement. Further studies are needed to confirm these results on interobserver variability.