External validation of an artificial intelligence model for Gleason grading of prostate cancer on prostatectomy specimens.

OBJECTIVES: To externally validate the performance of the DeepDx Prostate artificial intelligence (AI) algorithm (Deep Bio Inc., Seoul, South Korea) for Gleason grading on whole-mount prostate histopathology, considering potential variations observed when applying AI models trained on biopsy samples to radical prostatectomy (RP) specimens due to inherent differences in tissue representation and sample size. MATERIALS AND METHODS: The commercially available DeepDx Prostate AI algorithm is an automated Gleason grading system that was previously trained using 1133 prostate core biopsy images and validated on 700 biopsy images from two institutions. We assessed the AI algorithm's performance, which outputs Gleason patterns (3, 4, or 5), on 500 1-mm2 tiles created from 150 whole-mount RP specimens from a third institution. These patterns were then grouped into grade groups (GGs) for comparison with expert pathologist assessments. The reference standard was the International Society of Urological Pathology GG as established by two experienced uropathologists with a third expert to adjudicate discordant cases. We defined the main metric as the agreement with the reference standard, using Cohen's kappa. RESULTS: The agreement between the two experienced pathologists in determining GGs at the tile level had a quadratically weighted Cohen's kappa of 0.94. The agreement between the AI algorithm and the reference standard in differentiating cancerous vs non-cancerous tissue had an unweighted Cohen's kappa of 0.91. Additionally, the AI algorithm's agreement with the reference standard in classifying tiles into GGs had a quadratically weighted Cohen's kappa of 0.89. In distinguishing cancerous vs non-cancerous tissue, the AI algorithm achieved a sensitivity of 0.997 and specificity of 0.88; in classifying GG ≥2 vs GG 1 and non-cancerous tissue, it demonstrated a sensitivity of 0.98 and specificity of 0.85. CONCLUSION: The DeepDx Prostate AI algorithm had excellent agreement with expert uropathologists and performance in cancer identification and grading on RP specimens, despite being trained on biopsy specimens from an entirely different patient population.

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.

Should Grade Group 1 (GG1) be called cancer?

INTRODUCTION: ISUP Grade Group 1 prostate cancer is the lowest histologic grade of prostate cancer with a clinically indolent course. Removal of the term 'cancer' has been proposed and has historical precedent both in urothelial and thyroid carcinoma. METHODS: Evidence-based review identifying arguments for and against Grade Group 1 being referred to as cancer. RESULTS: Grade Group 1 has histologic evidence of tissue microinvasion and 0.3-3% rate of extraprostatic extension. Genomic evaluation suggests overlap of a minority of Grade Group 1 cancers with those of Grade Group 2. Conversely, Grade Group 1 tumors appear to have distinct genetic and genomic profiles from Grade Group 3 or higher tumors. Grade Group 1 has no documented ability for regional or distant metastasis and long-term follow up after treatment or active surveillance is safe with excellent oncologic outcomes. DISCUSSION: Grade Group 1 prostate cancer, while showing evidence of neoplasia on histology has a remarkably indolent natural history more akin to non-neoplastic precursor lesions. Consideration should be given to renaming Grade Group 1 prostate cancer, which has the potential to minimize overtreatment, treatment-related side effects, patient anxiety, and financial burden on the healthcare system.

Head to head: should the intraductal component of invasive prostate cancer be graded?

The reporting of intraductal carcinoma of the prostate (IDCP) is controversial, with conflicting recommendations having recently been published by the International Society of Urological Pathology (ISUP) and the Genitourinary Pathology Society (GUPS). Both recommend that isolated (pure) IDCP should not be graded. However, the ISUP recommends incorporating the IDCP component of invasive prostate cancer in the Gleason score, whereas the GUPS recommends reporting IDCP as a comment, independently of the Gleason score. The arguments for and against incorporating the IDCP component of invasive prostate cancer in the Gleason score are discussed in detail.

Old men with prostate cancer have higher risk of Gleason score upgrading and pathological upstaging after initial diagnosis: a systematic review and meta-analysis.

BACKGROUND: To evaluate the predictive performance of age for the risk of Gleason score change and pathologic upstaging. EVIDENCE ACQUISITION: Ovid MEDLINE, Ovid Embase, and the Cochrane Library were searched from inception until May 2020. Quality of included studies was appraised utilizing the Newcastle-Ottawa Quality Assessment Scale for case-control studies. The publication bias was evaluated by funnel plots and Egger's tests. EVIDENCE SYNTHESIS: Our search yielded 27 studies with moderate-to-high quality including 84296 patients with mean age of 62.1 years. From biopsy to prostatectomy, upgrading and upstaging occurred in 32.3% and 9.8% of patients, respectively. Upgrading from diagnostic biopsy to confirmatory biopsy was found in 16.8%. Older age was associated with a significant increased risk of upgrading (OR 1.04, 95% CI 1.03-1.05), and similar direction of effect was found in studies focused on upgrading from diagnostic biopsy to confirmatory biopsy (OR 1.06, 95% CI 1.04-1.08). For pathologic upstaging within older men compared with younger, the pooled odds was 1.03 (95% CI 1.01-1.04). CONCLUSION: Thorough consideration of age in the context of effect sizes for other factors not only prompts more accurate risk stratification but also helps providers to select optimal therapies for patients with prostate cancer.

Prostate cancer with comedonecrosis is frequently, but not exclusively, intraductal carcinoma: a need for reappraisal of grading criteria.

AIMS: Comedonecrosis in prostate cancer has always been Gleason pattern 5. However, we aimed to evaluate how intraductal carcinoma (not graded) with comedonecrosis should be considered. METHODS AND RESULTS: From 52 radical prostatectomy patients, 40 were informative and evaluated with immunohistochemistry for basal cells. Clinical outcome was assessed for biochemical recurrence, metastatic disease and the need for adjuvant therapy. Comedonecrosis was predominantly located in intraductal carcinoma (24, 60%). However, nine (23%) had comedonecrosis within invasive cancer and seven (18%) within both invasive and intraductal carcinoma. Extraprostatic extension rarely showed comedonecrosis (5, 13%), but rather perineural invasion within cribriform glands. Tumours were largely high-stage (15, 38% pT3a and 19, 48% pT3b), with 15 (37%) having positive lymph nodes and four distant metastases. Most cases (25, 63%) had other patterns of Gleason pattern 5 (single cells, solid), although 10 were reclassified as containing no invasive pattern 5. Of these, most were pT3 (eight of 10), but none had positive lymph nodes. Lymph node metastases were more common in patients with invasive cancer containing comedonecrosis (P = 0.02), and the need for androgen deprivation was near significance (P = 0.07), but biochemical recurrence was not significantly different (P = 0.58). CONCLUSIONS: Prostate cancer with comedonecrosis is often intraductal; however, these tumours are largely high-stage, showing a higher rate of positive lymph nodes with invasive comedonecrosis. Immunohistochemistry may be considered when comedonecrosis may significantly change the tumour grade. However, it is not clear at present that excluding intraductal carcinoma from the grade is superior to including it in grading when it is associated with high-grade invasive cancer.

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.

Utility of computed diffusion-weighted MRI for predicting aggressiveness of prostate cancer.

PURPOSE: To investigate the value of computed (c) diffusion-weighted imaging (DWI) in assessing prostate cancer aggressiveness. MATERIALS AND METHODS: Fifty-five patients with peripheral zone prostate cancer who underwent prebiopsy 1.5T magnetic resonance imaging (including native DWI at b-values of 0 and 1000 s/mm2 ) were included. cDWI signal intensities of peripheral zone prostate cancer and nonmalignant prostate tissue were measured. Association between changes in monoexponentially calculated cDWI signals according to different b-values and primary Gleason grades were assessed. RESULTS: The cDWI signal intensity of prostate cancer was lower at b = 0 s/mm2 and higher at b = 1000 s/mm2 compared to nonmalignant prostate tissue. The b-value at which the signal intensities of prostate cancer and nonmalignant prostate tissue were equal was defined as the "iso-b-value." On multivariate analysis, only the iso-b-value was a significant predictor of primary Gleason grade 4/5 cancer (P = 0.001). The area under the curve (AUC) of the iso-b-value for diagnosing primary Gleason grade 4/5 cancer was 0.94, and significantly higher than that of the tumor apparent diffusion coefficient (ADC) value with an AUC of 0.68 (P < 0.001). CONCLUSION: cDWI with iso-b-value-based semiquantitative analysis was found to be useful for predicting the aggressiveness of prostate cancer and may potentially outperform tumor ADC measurements in this setting. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:490-496.

Quantified analysis of histological components and architectural patterns of gleason grades in apparent diffusion coefficient restricted areas upon diffusion weighted MRI for peripheral or transition zone cancer locations.

PURPOSE: To quantify and compare the histological components and architectural patterns of Gleason grades in cancerous areas with restriction on apparent diffusion coefficient (ADC) maps. MATERIALS AND METHODS: Twelve consecutive cases with 14 separate ADC restriction areas, positive for cancer in the peripheral zone (PZ) and transition zone (TZ) were included. All had 3 Tesla MRI and radical prostatectomy. Ten regions of interest (ROIs) within and outside the 14 ADC restriction areas positive for cancer were selected. For each ROI, we performed quantitative analysis of (a) prostate benign and malignant histological component surface ratios, including stroma, glands, epithelium, lumen, cellular nuclei; (b) percent of Gleason grades and measures of ADC values. Means of histological components according to ADC restriction for cancerous area were compared with analyses of variance with repeated measures. RESULTS: Independent predictors of the probability of cancer were median epithelium/ROI ratio (P = 0.001) and nuclei/ROI ratio (P = 0.03). Independent predictors of the probability of ADC restriction were malignant glands/ROI and luminal space/ROI (P < 0.0001). Effect of malignant glands/ROI area was different according to the localization of the ROI (P = 0.03). We observed an overall difference between the means for all of the histological components for the comparison of true positive and false negative (P < 0.0001), except for the percent of Gleason grade 4 (P = 0.18). In TZ cancers, a predominant grade 3 pattern was associated with low ADC values. In PZ cancers, a predominant grade 4 pattern was associated with low ADC values. CONCLUSION: Determinants of low ADC were high ratio of malignant glands/ROI area which may be seen in Gleason grades 3 or 4 cancers. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1786-1796.

Metastatic potential to regional lymph nodes with Gleason score ≤7, including tertiary pattern 5, at radical prostatectomy.

OBJECTIVES: To determine the risk of pelvic lymph node (LN) metastases at radical prostatectomy (RP) for Gleason score (GS) ≤7: 3 + 3 = 6 (grade group [GG]1); 3 + 4 = 7 (GG2); 3 + 4 = 7 (GG2) with tertiary pattern 5 (T5); 4 + 3 = 7 (GG3); 4 + 3 = 7 (GG3) with T5, using the 2014 modified Gleason grading system and the novel GG system. MATERIALS AND METHODS: We searched our RP database to indentify cases of GS ≤7 prostate cancer with simultaneous pelvic LN dissection (PLND) in the period between 2005 and 2014. Since 2005, we have graded all glomeruloid and cribriform cancer as Gleason pattern 4 and have graded mucinous adenocarcinoma based on the underlying architectural pattern, consistent with the 2014 modified Gleason grading system. All RPs were embedded in entirety, including the PLND. A total of 7 442 cases were identified, of which 73 had at least one positive LN (+LN). RESULTS: The incidence rates for regional LN metastases at RP for 3 + 3 = 6 (GG1), 3 + 4 = 7 (GG2), 3 + 4 = 7 (GG2) with T5, 4 + 3 = 7 (GG3) and 4 + 3 = 7 (GG3) with T5 were 0, 0.6, 0.4, 4.3 and 6.3%, respectively. There was a statistically significant difference in risk of +LNs at RP between the grade groups, as defined by the novel GG system. There was no statistically significant difference in risk of +LNs at RP for men with 3 + 4 (GG2) vs 3 + 4 (GG2) with T5 and for men with 4 + 3 (GG3) vs 4 + 3 (GG3) with T5. Non-pelvic LN involvement was identified in 0.2% of all RP cases. Two patients with GS 3 + 4 = 7 with <5% pattern 4 experienced LN metastases. CONCLUSION: This study supports our previous finding that men with GS 6 (GG1) at RP have no risk of LN metastases. These findings also support the 2014 revisions to the Gleason grading system where 3 + 3 with T4 (2005 modified grading system) is now considered 3 + 4 (GG2), with a comment on percent pattern 4, because <5% pattern 4 increases the risk of LN metastases. It also supports keeping 3 + 4 (GG2) with T5 (<5% pattern 5) and 4 + 3 (GG3) with T5 with their respective grade groups, with a notation of T5 because the <5% higher grade component did not increase the risk of LN metastases within a given GG. Our findings highlight that 3 + 4 (GG2) and 4 + 3 (GG3), even with a 5% higher grade component, are distinct groups with respect to LN metastases and should not be combined under the umbrella designation of GS 7 as is often the case in the literature with the Gleason grading system.

T2-weighted MRI-derived textural features reflect prostate cancer aggressiveness: preliminary results.

PURPOSE: To evaluate the diagnostic relevance of T2-weighted (T2W) MRI-derived textural features relative to quantitative physiological parameters derived from diffusion-weighted (DW) and dynamic contrast-enhanced (DCE) MRI in Gleason score (GS) 3+4 and 4+3 prostate cancers. MATERIALS AND METHODS: 3T multiparametric-MRI was performed on 23 prostate cancer patients prior to prostatectomy. Textural features [angular second moment (ASM), contrast, correlation, entropy], apparent diffusion coefficient (ADC), and DCE pharmacokinetic parameters (Ktrans and Ve) were calculated from index tumours delineated on the T2W, DW, and DCE images, respectively. The association between the textural features and prostatectomy GS and the MRI-derived parameters, and the utility of the parameters in differentiating between GS 3+4 and 4+3 prostate cancers were assessed statistically. RESULTS: ASM and entropy correlated significantly (p < 0.05) with both GS and median ADC. Contrast correlated moderately with median ADC. The textural features correlated insignificantly with Ktrans and Ve. GS 4+3 cancers had significantly lower ASM and higher entropy than 3+4 cancers, but insignificant differences in median ADC, Ktrans, and Ve. The combined texture-MRI parameters yielded higher classification accuracy (91%) than the individual parameter sets. CONCLUSION: T2W MRI-derived textural features could serve as potential diagnostic markers, sensitive to the pathological differences in prostate cancers. KEY POINTS: • T2W MRI-derived textural features correlate significantly with Gleason score and ADC. • T2W MRI-derived textural features differentiate Gleason score 3+4 from 4+3 cancers. • T2W image textural features could augment tumour characterization.

Gleason grade 4 prostate adenocarcinoma patterns: an interobserver agreement study among genitourinary pathologists.

AIMS: To assess the interobserver reproducibility of individual Gleason grade 4 growth patterns. METHODS AND RESULTS: Twenty-three genitourinary pathologists participated in the evaluation of 60 selected high-magnification photographs. The selection included 10 cases of Gleason grade 3, 40 of Gleason grade 4 (10 per growth pattern), and 10 of Gleason grade 5. Participants were asked to select a single predominant Gleason grade per case (3, 4, or 5), and to indicate the predominant Gleason grade 4 growth pattern, if present. 'Consensus' was defined as at least 80% agreement, and 'favoured' as 60-80% agreement. Consensus on Gleason grading was reached in 47 of 60 (78%) cases, 35 of which were assigned to grade 4. In the 13 non-consensus cases, ill-formed (6/13, 46%) and fused (7/13, 54%) patterns were involved in the disagreement. Among the 20 cases where at least one pathologist assigned the ill-formed growth pattern, none (0%, 0/20) reached consensus. Consensus for fused, cribriform and glomeruloid glands was reached in 2%, 23% and 38% of cases, respectively. In nine of 35 (26%) consensus Gleason grade 4 cases, participants disagreed on the growth pattern. Six of these were characterized by large epithelial proliferations with delicate intervening fibrovascular cores, which were alternatively given the designation fused or cribriform growth pattern ('complex fused'). CONCLUSIONS: Consensus on Gleason grade 4 growth pattern was predominantly reached on cribriform and glomeruloid patterns, but rarely on ill-formed and fused glands. The complex fused glands seem to constitute a borderline pattern of unknown prognostic significance on which a consensus could not be reached.

Recent advances in prostate cancer pathology: Gleason grading and beyond.

Since its initial description in 1966 by Dr Donald Gleason, Gleason grading has become the cornerstone in the management of prostate cancer (PCa). With widespread use of Prostate Specific Antigen (PSA) screening and needle core biopsy, the diagnosis and management of PCa have dramatically evolved. In addition, better understanding of the morphological spectrum of prostate cancer and its clinical significance have prompted the refinement of the grading criteria and reporting guidelines commensurate to contemporary practice. The modification of the Gleason grading system implemented by the International Society of Urological Pathology in 2005 and subsequent revision in 2014 has profoundly impacted how PCa is graded and managed. This review aims to provide a concise update on the refinement of the histological criteria for various Gleason patterns and problem areas of Gleason grading, and provide recommendations on how to improve the grading reproducibility. The new proposal to group Gleason scores into clinically meaningful "grade groups" will also be discussed. Finally, we will discuss how magnetic resonance imaging (MRI)-targeted biopsy and emerging genetic markers may help improve the Gleason grading accuracy and risk stratification currently based on clinicopathological parameters.

Grading of Prostate Cancer: Past, Present, and Future.

The grading of prostate cancer has undergone significant changes since the adoption of the Gleason grading system in the 1970s. Gleason patterns 1 and 2 are no longer in use and the current Gleason score 6 of 10 is the lowest grade possible. Several specific morphologies that were historically considered Gleason grade 3 are currently assigned a Gleason pattern 4. Consequently, current Gleason score 6 cancers have a better prognosis than historic ones. There is now ample literature that supports that Gleason score 7 includes patients with very different prognosis; those with Gleason score 3 + 4 have a much better prognosis than patients with Gleason score 4 + 3. Within patients with high-grade cancer, it is now also clear that patients with Gleason score 8 have a significantly better prognosis than men with Gleason scores 9-10. Additionally, more recent studies have demonstrated that there is no significant difference in the prognosis of patients with Gleason score 9 or 10, making the distinction between the two pointless. A new contemporary grading system has been proposed that addresses these changes/problems and provides a simpler system with only five grades that reflect more accurately the prognosis of each group. We review the different changes applied to the Gleason scoring system since its conception as well as the studies leading to a new contemporary grading system.

Outcomes of Gleason score 3 + 4 = 7 prostate cancer with minimal amounts (<6%) vs ≥6% of Gleason pattern 4 tissue in needle biopsy specimens.

OBJECTIVE: The International Society of Urological Pathology Gleason grading system was modified in 2005. Since the modified system was introduced, many cancers that previously would have been categorized as Gleason score (GS) 6 are now categorized as GS 7 based on biopsy specimens that only contain minimal amounts (<6%) of Gleason pattern (GP) 4 tissue. However, the clinical significance of observing <6% of GP 4 tissue in biopsies of GS 7 prostate cancer has not been studied. MATERIAL AND METHODS: This study was based on needle biopsy specimens that were categorized as GS 6 or GS 7 and were obtained from patients who underwent radical prostatectomy (RP) with available follow-up data. We assessed the quantity of GP 4 tissue in biopsy specimens of GS 7 prostate cancer. Further, we evaluated the correlation between the quantity of GP 4 tissue and disease progression after RP. RESULTS: GP 4 comprising 26-49% of the specimen, GS 4+3 and percentage of total core tissue scored as positive were significant and independent predictors of prostate-specific antigen (PSA) failure after RP, as assessed using a multivariate Cox regression model that included the quantity of GP 4 in the prostate biopsy specimen, preoperative PSA, perineural invasion, clinical stage, number of positive cores, and percentage of core tissue scored as positive. Cases with GS 3+3 and cases in which the observed GP 4 area was <6% did not differ significantly in terms of biochemical PSA recurrence (BPR) status. In contrast, cases with 6-25% GP 4 tissue, 26-49% GP 4 tissue, and GS 4+3 showed more frequent BPR than cases with GS 3+3. CONCLUSIONS: Our data suggest that the quantity of GP 4 tissue in GS 7 cancer has clinical significance. However, there is a need for larger studies of the clinical significance of biopsy specimens that include <6% GP 4 tissue. We should reconsider whether the amount of GP 4 should be included in standart pathology reports.

The evolving Gleason grading system.

The Gleason grading system for prostate adenocarcinoma has evolved from its original scheme established in the 1960s-1970s, to a significantly modified system after two major consensus meetings conducted by the International Society of Urologic Pathology (ISUP) in 2005 and 2014, respectively. The Gleason grading system has been incorporated into the WHO classification of prostate cancer, the AJCC/UICC staging system, and the NCCN guidelines as one of the key factors in treatment decision. Both pathologists and clinicians need to fully understand the principles and practice of this grading system. We here briefly review the historical aspects of the original scheme and the recent developments of Gleason grading system, focusing on major changes over the years that resulted in the modern Gleason grading system, which has led to a new "Grade Group" system proposed by the 2014 ISUP consensus, and adopted by the 2016 WHO classification of tumours of the prostate.

Not all gleason pattern 4 prostate cancers are created equal: A study of latent prostatic carcinomas in a cystoprostatectomy and autopsy series.

BACKGROUND: The Gleason grading system represents the cornerstone of the management of prostate cancer. Gleason grade 4 (G4) is a heterogeneous set of architectural patterns, each of which may reflect a distinct prognostic value. METHODS: We determined the prevalence of the various G4 architectural patterns and intraductal carcinoma (IDC) in latent prostate cancer in contemporary Russian (n = 220) and Japanese (n = 100) autopsy prostates and in cystoprostatectomy (CP) specimens (n = 248) collected in Italy. We studied the association of each G4 pattern with extraprostatic extension (EPE) and tumor volume to gain insight into their natural history. Presence of IDC and nine architectural features of Gleason grade 4 and 5 cancer were recorded. RESULTS: The prevalence of Gleason score ≥ 7 PC was higher in the autopsy series (11%) compared to the CP series (6.5%, P = 0.04). The prevalence of IDC and carcinoma with a cribriform architecture was 2.2% and 3.4% in the autopsy series and 0.8% and 3.6% in the cystoprostatectomy series, respectively. In multivariable analysis, cribriform architecture was significantly associated with increased tumor volume (P < 0.001) and EPE (OR:11.48, 95%CI:2.30-57.16, P = 0.003). IDC was also significantly associated with EPE (OR:10.08, 95%CI:1.58-64.28, P = 0.014). Small fused glands had a strong negative association with EPE in the autopsy series (OR:0.06, 95%CI:0.01-0.58, P = 0.015). DISCUSSION: Our study revealed that in latent prostate cancer both cribriform architecture and IDC are uniquely associated with poor pathological outcome features. In contrast, Gleason score 7 (3 + 4) cancers with small-fused gland pattern might possibly include some prostate cancers with a more indolent biology.

A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer.

The Gleason score is an important predictor of prognosis in prostate cancer. However, its subjective nature can result in over- or under-grading. Our objective was to train an artificial intelligence (AI)-based algorithm to grade prostate cancer in specimens from patients who underwent radical prostatectomy (RP) and to assess the correlation of AI-estimated proportions of different Gleason patterns with biochemical recurrence-free survival (RFS), metastasis-free survival (MFS), and overall survival (OS). Training and validation of algorithms for cancer detection and grading were completed with three large datasets containing a total of 580 whole-mount prostate slides from 191 RP patients at two centers and 6218 annotated needle biopsy slides from the publicly available Prostate Cancer Grading Assessment dataset. A cancer detection model was trained using MobileNetV3 on 0.5 mm × 0.5 mm cancer areas (tiles) captured at 10× magnification. For cancer grading, a Gleason pattern detector was trained on tiles using a ResNet50 convolutional neural network and a selective CutMix training strategy involving a mixture of real and artificial examples. This strategy resulted in improved model generalizability in the test set compared with three different control experiments when evaluated on both needle biopsy slides and whole-mount prostate slides from different centers. In an additional test cohort of RP patients who were clinically followed over 30 years, quantitative Gleason pattern AI estimates achieved concordance indexes of 0.69, 0.72, and 0.64 for predicting RFS, MFS, and OS times, outperforming the control experiments and International Society of Urological Pathology system (ISUP) grading by pathologists. Finally, unsupervised clustering of test RP patient specimens into low-, medium-, and high-risk groups based on AI-estimated proportions of each Gleason pattern resulted in significantly improved RFS and MFS stratification compared with ISUP grading. In summary, deep learning-based quantitative Gleason scoring using a selective CutMix training strategy may improve prognostication after prostate cancer surgery.

Risk of Cancer-related Death for Men with Biopsy Grade Group 1 Prostate Cancer and High-risk Features: A European Multi-institutional Study.

International Society of Urological Pathology grade group 1 (GG 1) prostate cancer (PCa) is generally considered insignificant, with recent suggestions that it should even be considered as "noncancerous". We evaluated outcomes for patients with GG 1 PCa on biopsy (bGG 1) and high-risk features (prostate-specific antigen [PSA] >20 ng/ml and/or cT3-4 stage) to challenge the hypothesis that every case of bGG 1 PCa has a benign disease course. We used the multi-institutional EMPaCT database, which includes data for 9508 patients with high-risk PCa undergoing surgery. We included patients with bGG 1 PCa (n = 848) in our analysis and divided them into three groups according to PSA >20 ng/ml, cT3-4 stage, or both. The estimated 10-yr cancer-specific survival (CSS) rate was 96% in the overall population, 88% in the group with both PSA >20 ng/ml and cT3-4 stage, 97% in the group with PSA >20 ng/ml alone, and 98% in the group with cT3-4 stage alone. Similar CSS outcomes were found in subgroups with GG 1 PCa on pathology (n = 502) and with GG 1 on biopsy diagnosed after 2005 (n = 253). Study limitations include the lack of magnetic resonance imaging (MRI) staging and MRI-targeted biopsies. In conclusion, patients with GG 1 and either PSA >20 ng/ml or cT3-4 stage have a low risk of dying from their cancer after surgery. However, patients with GG 1 PCa and both PSA >20 ng/ml and cT3-4 stage are at higher risk of cancer-specific mortality and active treatment should be discussed for this subgroup. Patient summary: We assessed outcomes for patients diagnosed with low-grade prostate cancer on biopsy who also had one or two factors associated with high risk disease. Men with both of those risk factors had a higher risk of dying from their prostate cancer. Active treatment should be discussed for this subgroup of patients.

Predictors of Gleason Grading Group Upgrading in Low-Risk Prostate Cancer Patients From Transperineal Biopsy After Radical Prostatectomy.

RATIONALE AND OBJECTIVES: To investigate the predictors of Gleason Grading Group (GGG) upgrading in low-risk prostate cancer (Gleason score=3 + 3) from transperineal biopsy after radical prostatectomy (RP). MATERIALS AND METHODS: The clinical data of 160 patients who underwent transperineal biopsy and RP from January 2017 to December 2022 were retrospectively analyzed. First, univariate and multivariate logistic regression analysis were used to obtain independent predictors of postoperative GGG upgrading. Then receiver operating characteristic curve was used to evaluate the diagnostic efficacy of predictors. Finally, Linear-by-Linear Association test was used to analyze the risk trends of patients in different predictor groups in the postoperative GGG. RESULTS: In this study, there were 81 cases (50.6%) in the GGG concordance group and 79 cases (49.4%) in the GGG upgrading group. Univariate analysis showed age, free/total prostate-specific antigen (f/tPSA), proportion of positive biopsies, positive target of magnetic-resonance imaging (MRI) and positive target of contrast-enhanced ultrasound had significant effects on GGG upgrading (all P < .05). In multivariate logistic regression analysis, age (odds ratio [OR]=1.066, 95%CI=1.007-1.127, P = .027), f/tPSA (OR=0.001, 95%CI=0-0.146, P = .001) and positive target of MRI (OR=3.005, 95%CI=1.353-76.674, P = .007) were independent predictors. The prediction model (area under curve=0.751 P < .001) had higher predictive efficacy than all independent predictors. The proportion of patients in exposed group of different GGG increased with the level of GGG, but decreased in nonexposed group, and the linear trend was significantly different (all P < .001). CONCLUSION: Age, f/tPSA, and positive target of MRI were independent predictors of postoperative GGG upgrading. The predictive model constructed had the best diagnostic efficacy.