Central prostate pathology review: should it be mandatory?

International Society of Urological Pathology (ISUP) Gleason Grade Groups stratify outcomes in the CHHiP Phase 3 prostate radiotherapy trial

OBJECTIVES

To compare the results of Gleason Grade Group (GGG) classification following central pathology review with previous local pathology assessment, and to examine the difference between using overall and worst GGG in a large patient cohort treated with radiotherapy and short-course hormone therapy.

PATIENTS AND METHODS

Patients with low- to high-risk localized prostate cancer were randomized into the multicentre CHHiP fractionation trial between 2002 and 2011. Patients received short-course hormone therapy (≤6 month) and radical intensity-modulated radiotherapy (IMRT). Of 2749 consented patients, 1875 had adequate diagnostic biopsy tissue for blinded central pathology review. The median follow-up was 9.3 years. Agreement between local pathology and central pathology-derived GGG and between central pathology-derived overall and worst GGG was assessed using kappa (κ) statistics. Multivariate Cox regression and Kaplan-Meier methods were used to compare the biochemical/clinical failure (BCF) and distant metastases (DM) outcomes of patients with GGG 1-5.

RESULTS

There was poor agreement between local pathology- and central pathology-derived GGG (κ = 0.19) but good agreement between overall and worst GGG on central pathology review (κ = 0.89). Central pathology-derived GGG stratified BCF and DM outcomes better than local pathology, while overall and worst GGG on central pathology review performed similarly. GGG 3 segregated with GGG 4 for BCF, with BCF-free rates of 90%, 82%, 74%, 71% and 58% for GGGs 1-5, respectively, at 8 years when assessed using overall GGG. There was a progressive decrease in DM-free rates from 98%, 96%, 92%, 88% and 83% for GGGs 1-5, respectively, at 8 years with overall GGG. Patients (n = 57) who were upgraded from GGG 2-3 using worst GS had BCF-free and DM-free rates of 74% and 92% at 8 years. CHHiP eligibility criteria limit the interpretation of these results.

CONCLUSION

Contemporary review of International Society of Urological Pathology GGG successfully stratified patients treated with short-course hormone therapy and IMRT with regard to both BCF-free and DM-free outcomes. Patients upgraded from GGG 2 to GGG 3 using worst biopsy GS segregate with GGG 3 on long-term follow-up. We recommend that both overall and worst GS be used to derive GGG.

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.

Narrative Review of the Post-Operative Management of Prostate Cancer Patients: Is It Really the End of Adjuvant Radiotherapy?

Simple Summary

Among patients with prostate cancer who have been operated on, a subset harboring high-risk features will present with a biochemical recurrence (BCR). Adjuvant radiotherapy (aRT) was proven to significantly reduce the risk of BCR when compared to salvage radiotherapy (SRT) but suffered from several limitations: a lack of patient selection criteria, a higher treatment-related morbidity and an uncertain benefit for long-term clinical endpoints. In the same clinical setting, early SRT (eSRT) appears as non-inferior to aRT with a lower morbidity, replacing aRT as the preferred option. In this review, we insist on the need for multidisciplinary discussions to fully comprehend the individual characteristics of each patient and propose the best treatment strategy for every patient.

Abstract

Despite three randomized trials indicating a significant reduction in biochemical recurrence (BCR) in high-risk patients, adjuvant radiotherapy (aRT) was rarely performed, even in patients harboring high-risk features. aRT is associated with a higher risk of urinary incontinence and is often criticized for the lack of patient selection criteria. With a BCR rate reaching 30–70% in high-risk patients, a consensus between urologists and radiation oncologists was needed, leading to three different randomized trials challenging aRT with early salvage radiotherapy (eSRT). In these three different randomized trials with event-free survival as the primary outcome and a planned meta-analysis, eSRT appeared as non-inferior to aRT, answering, for some, this never-ending question. For many, however, the debate persists; these results raised several questions among urologists and radiation oncologists. BCR is thought to be a surrogate for clinically meaningful endpoints such as overall survival and cancer-specific survival but may be poorly efficient in comparison with metastasis-free survival. Imaging of rising prostate-specific antigen (PSA), post-operative persistent PSA and BCR was revolutionized by the broader use of MRI and nuclear imaging such as PET-PSMA; these imaging modalities were not analyzed in the previous randomized trials. A sub-group of very high-risk patients could possibly benefit from an adjuvant radiotherapy; but their usual risk factors such as high Gleason score or invaded surgical margins mean they are unable to be selected. More precise biomarkers of early BCR or even metastatic-relapse were developed in this setting and could be useful for the patients’ stratification. In this review, we insist on the need for multidisciplinary discussions to fully comprehend the individual characteristics of each patient and propose the best treatment strategy for every patient.

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.

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.

Factors predicting pathological upgrading after prostatectomy in patients with Gleason grade group 1 prostate cancer based on opinion-matched biopsy specimens

The present study investigated the concordance between Gleason scores assigned to prostate biopsy specimens by outside pathologists and a urological pathology expert, and determined the risk of upgrading between opinion-matched Gleason grade group (GGG) 1 biopsy specimens and radical prostatectomy specimens. Between January 2012 and May 2018, 733 patients underwent robot-assisted radical prostatectomy. Patients whose original biopsy specimens from outside hospitals were reviewed by a urological pathology expert Okayama University Hospital were included. Patients who had received neoadjuvant hormonal therapy were excluded. Logistic regression analysis was used to identify predictors of upgrading among GGG 1 diagnoses. A total of 403 patients were included in the present study. Agreement in GGG between initial and second-opinion diagnoses was present in 256 cases (63.5%). Although opinion-matched cases improved concordance between biopsy and prostatectomy specimen GGG compared with single-opinion cases (initial, 35.2%; second-opinion, 36.5%; matched, 41.4%), 71% (56/79) of cases classified as GGG 1 were upgraded after prostatectomy. Multivariate analysis revealed that prostate-specific antigen density and Prostate Imaging Reporting and Data System version 2 score were significant predictors of upgrading (odds ratio, 1.10; P=0.01; and odds ratio, 1.88; P=0.03, respectively). In conclusion, the GGG concordance rate between needle-core biopsy and radical prostatectomy specimens was higher in opinion-matched cases; however, 71% of opinion-matched GGG1 cases were upgraded after robot-assisted radical prostatectomy. Urologists should propose treatment strategies or further biopsy rather than active surveillance for patients with GGG1 and a high PSAD and/or PI-RADS score.

Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer

For prostate cancer patients, the Gleason score is one of the most important prognostic factors, potentially determining treatment independent of the stage. However, Gleason scoring is based on subjective microscopic examination of tumor morphology and suffers from poor reproducibility. Here we present a deep learning system (DLS) for Gleason scoring whole-slide images of prostatectomies. Our system was developed using 112 million pathologist-annotated image patches from 1226 slides, and evaluated on an independent validation dataset of 331 slides. Compared to a reference standard provided by genitourinary pathology experts, the mean accuracy among 29 general pathologists was 0.61 on the validation set. The DLS achieved a significantly higher diagnostic accuracy of 0.70 (p = 0.002) and trended towards better patient risk stratification in correlations to clinical follow-up data. Our approach could improve the accuracy of Gleason scoring and subsequent therapy decisions, particularly where specialist expertise is unavailable. The DLS also goes beyond the current Gleason system to more finely characterize and quantitate tumor morphology, providing opportunities for refinement of the Gleason system itself.

The Quantitative Criteria Based on the Fractal Dimensions, Entropy, and Lacunarity for the Spatial Distribution of Cancer Cell Nuclei Enable Identification of Low or High Aggressive Prostate Carcinomas

Background: Tumor grading, PSA concentration, and stage determine a risk of prostate cancer patients with accuracy of about 70%. An approach based on the fractal geometrical model was proposed to eliminate subjectivity from the evaluation of tumor aggressiveness and to improve the prediction. This study was undertaken to validate classes of equivalence for the spatial distribution of cancer cell nuclei in a larger, independent set of prostate carcinomas.

Methods: The global fractal capacity D₀, information D₁ and correlation D₂ dimension, the local fractal dimension (LFD) and the local connected fractal dimension (LCFD), Shannon entropy H and lacunarity λ were measured using computer algorithms in digitalized images of both the reference set (n = 60) and the test set (n = 208) of prostate carcinomas.

Results: Prostate carcinomas were re-stratified into seven classes of equivalence. The cut-off D₀-values 1.5450, 1.5820, 1.6270, 1.6490, 1.6980, 1.7640 defined the classes from C1 to C7, respectively. The other measures but the D₁ failed to define the same classes of equivalence. The pairs (D₀, LFD), (D₀, H), (D₀, λ), (D₁, LFD), (D₁, H), (D₁, λ) characterized the spatial distribution of cancer cell nuclei in each class. The co-application of those measures enabled the subordination of prostate carcinomas to one out of three clusters associated with different tumor aggressiveness. For D₀ < 1.5820, LFD < 1.3, LCFD > 1.5, H < 0.7, and λ > 0.8, the class C1 or C2 contains low complexity low aggressive carcinomas exclusively. For D₀ > 1.6980, LFD > 1.7644, LCFD > 1.7051, H > 0.9, and λ < 0.7, the class C6 or C7 contains high complexity high aggressive carcinomas.

Conclusions: The cut-off D₀-values defining the classes of equivalence were validated in this study. The cluster analysis suggested that the number of the subjective Gleason grades and the number of the objective classes of equivalence could be decreased from seven to three without a loss of clinically relevant information. Two novel quantitative criteria based on the complexity and the diversity measures enabled the identification of low or high aggressive prostate carcinomas and should be verified in the future multicenter, randomized studies.

Prevalence and management of prostate cancer among East Asian men: Current trends and future perspectives

OBJECTIVES

Previously East Asian men had been considered less likely to develop or die of prostate cancer. Emerging research and the onset of prostate-specific antigen screening in East Asian countries suggests that this may not be the case. We sought to analyze epidemiology and molecular genetic data and recent trends in the management of prostate cancer among East Asian men.

METHODS AND MATERIALS

We performed literature searches using PubMed, Embase, and Google Scholar to examine current literature on prostate cancer in East Asian men. Additionally, articles were searched for further references related to the topic.

RESULTS

Recent studies have reported increasing incidence of prostate cancer identified in East Asian men. Prostate cancer mortality has increased and is currently the fourth leading cause of death among men in Shanghai, China. Although prostate cancer was considered less aggressive among East Asian men, studies suggest that it is similarly aggressive to prostate cancer in Western populations. Molecular markers such as the TEMPRESS:ERG fusion gene and PTEN loss may provide novel methods of screening East Asian men for prostate cancer. National-level guidelines for prostate cancer screening and management are only available in Japan.

CONCLUSIONS

The prevalence of prostate cancer in East Asian men is likely similar to that in Western male populations. East Asian men present at higher stages of prostate cancer, likely because of a lack of standardized screening protocols. Urologists in Western countries should screen East Asian men for prostate cancer using the same standards as used for Western men.

On the relationship between tumor structure and complexity of the spatial distribution of cancer cell nuclei: a fractal geometrical model of prostate carcinoma

BACKGROUND

A risk of the prostate cancer patient is defined by both the objective and subjective criteria, that is, PSA concentration, Gleason score, and pTNM-stage. The subjectivity of tumor grading influences the risk assessment owing to a large inter- and intra-observer variability. Pathologists propose a central prostate pathology review as a remedy for this problem; yet, the review cannot eliminate the subjectivity from the diagnostic algorithm. The spatial distribution of cancer cell nuclei changes during tumor progression. It implies changes in complexity measured by the capacity dimension D0, the information dimension D1, and the correlation dimension D2.

METHODS

The cornerstone of the approach is a model of prostate carcinomas composed of the circular fractals CF(4), CF(6 + 0), and CF(6 + 1). This model is both geometrical and analytical, that is, its structure is well-defined, the capacity fractal dimension D0 can be calculated for the infinite circular fractals, and the dimensions D0, D1, D2 can be computed for their finite counterparts representing distribution of cell nuclei. The model enabled both the calibration of the software and the validation of the measurements in 124 prostate carcinomas. The ROC analysis defined the cut-off D0 values for seven classes of complexity.

RESULTS

The Gleason classification matched in part with the classification based on the D0 values. The mean ROC sensitivity was 81.3% and the mean ROC specificity 75.2%. Prostate carcinomas were re-stratified into seven classes of complexity according to their D0 values. This increased both the mean ROC sensitivity and the mean ROC specificity to 100%. All homogeneous Gleason patterns were subordinated to the class C1, C4, or C7. D0 = 1.5820 was the cut-off D0 value between the complexity class C2 and C3 representing low-risk cancers and intermediate-risk cancers, respectively.

CONCLUSIONS

The global fractal dimensions eliminate the subjectivity in the diagnostic algorithm of prostate cancer. Those complexity measures enable the objective subordination of carcinomas to the well-defined complexity classes, and define subgroups of carcinomas with very low malignant potential (complexity class C1) or at a large risk of progression (complexity ass C7).