Accuracy of gleason grading by practicing pathologists and the impact of education on improving agreement

MRI-based prostate cancer classification using 3D efficient capsule network

BACKGROUND

Prostate cancer (PCa) is the most common cancer in men and the second leading cause of male cancer-related death. Gleason score (GS) is the primary driver of PCa risk-stratification and medical decision-making, but can only be assessed at present via biopsy under anesthesia. Magnetic resonance imaging (MRI) is a promising non-invasive method to further characterize PCa, providing additional anatomical and functional information. Meanwhile, the diagnostic power of MRI is limited by qualitative or, at best, semi-quantitative interpretation criteria, leading to inter-reader variability.

PURPOSES

Computer-aided diagnosis employing quantitative MRI analysis has yielded promising results in non-invasive prediction of GS. However, convolutional neural networks (CNNs) do not implicitly impose a frame of reference to the objects. Thus, CNNs do not encode the positional information properly, limiting method robustness against simple image variations such as flipping, scaling, or rotation. Capsule network (CapsNet) has been proposed to address this limitation and achieves promising results in this domain. In this study, we develop a 3D Efficient CapsNet to stratify GS-derived PCa risk using T2-weighted (T2W) MRI images.

METHODS

In our method, we used 3D CNN modules to extract spatial features and primary capsule layers to encode vector features. We then propose to integrate fully-connected capsule layers (FC Caps) to create a deeper hierarchy for PCa grading prediction. FC Caps comprises a secondary capsule layer which routes active primary capsules and a final capsule layer which outputs PCa risk. To account for data imbalance, we propose a novel dynamic weighted margin loss. We evaluate our method on a public PCa T2W MRI dataset from the Cancer Imaging Archive containing data from 976 patients.

RESULTS

Two groups of experiments were performed: (1) we first identified high-risk disease by classifying low + medium risk versus high risk; (2) we then stratified disease in one-versus-one fashion: low versus high risk, medium versus high risk, and low versus medium risk. Five-fold cross validation was performed. Our model achieved an area under receiver operating characteristic curve (AUC) of 0.83 and 0.64 F1-score for low versus high grade, 0.79 AUC and 0.75 F1-score for low + medium versus high grade, 0.75 AUC and 0.69 F1-score for medium versus high grade and 0.59 AUC and 0.57 F1-score for low versus medium grade. Our method outperformed state-of-the-art radiomics-based classification and deep learning methods with the highest metrics for each experiment. Our divide-and-conquer strategy achieved weighted Cohen's Kappa score of 0.41, suggesting moderate agreement with ground truth PCa risks.

CONCLUSIONS

In this study, we proposed a novel 3D Efficient CapsNet for PCa risk stratification and demonstrated its feasibility. This developed tool provided a non-invasive approach to assess PCa risk from T2W MR images, which might have potential to personalize the treatment of PCa and reduce the number of unnecessary biopsies.

Multinational, Multicenter Evaluation of Prostate Cancer Tissue in Sub-Saharan Africa: Challenges and Opportunities

PURPOSE

Prostate cancer disproportionately affects men of African descent, yet their representation in tissue-based studies is limited. This multinational, multicenter pilot study aims to establish the groundwork for collaborative research on prostate cancer in sub-Saharan Africa.

METHODS

The Men of African Descent and Carcinoma of the Prostate network formed a pathologist working group representing eight institutions in five African countries. Formalin-fixed paraffin-embedded prostate tissue specimens were collected from Senegal, Nigeria, and Ghana. Histology slides were produced and digitally scanned. A central genitourinary pathologist (P.L.) and eight African general pathologists reviewed anonymized digital whole-slide images for International Society of Urological Pathology grade groups and other pathologic parameters. Discrepancies were re-evaluated, and consensus grading was assigned. A virtual training seminar on prostate cancer grading was followed by a second assessment on a subcohort of the same tissue set.

RESULTS

Of 134 tissue blocks, 133 had evaluable tissue; 13 lacked cancer evidence, and four were of insufficient quality. Post-training, interobserver agreement for grade groups improved to 56%, with a median Cohen's quadratic weighted kappa of 0.83 (mean, 0.74), compared with an initial 46% agreement and a quadratic weighted kappa of 0.77. Interobserver agreement between African pathologist groups was 40%, with a quadratic weighted kappa of 0.66 (95% CI, 0.51 to 0.76). African pathologists tended to overgrade (36%) more frequently than undergrade (18%) compared with the reference genitourinary pathologist. Interobserver variability tended to worsen with a decrease in tissue quality.

CONCLUSION

Tissue-based studies on prostate cancer in men of African descent are essential for a better understanding of this common disease. Standardized tissue handling protocols are crucial to ensure good tissue quality and data. The use of digital slide imaging can enhance collaboration among pathologists in multinational, multicenter studies.

Use of the ISUP e-learning module improves interrater reliability in prostate cancer grading

AIMS

Prostate cancer (PCa) grading is an important prognostic parameter, but is subject to considerable observer variation. Previous studies have shown that interobserver variability decreases after participants were trained using an e-learning module. However, since the publication of these studies, grading of PCa has been enhanced by adopting the International Society of Urological Pathology (ISUP) 2014 grading classification. This study investigates the effect of training on interobserver variability of PCa grading, using the ISUP Education web e-learning on Gleason grading.

METHODS

The ISUP Education Prostate Test B Module was distributed among Dutch pathologists. The module uses images graded by the ISUP consensus panel consisting of 24 expert uropathologists. Participants graded the same 10 images before and after e-learning. We included those who completed the tests before and after training. We evaluated variation in PCa grading in a fully crossed study design, using linearly weighted kappa values for each pathologist, comparing them to other pathologists and to the ISUP consensus panel. We analysed the improvement in median weighted kappas before and after training, using Wilcoxon's signed rank-test.

RESULTS

We included 42 pathologists. Inter-rater reliability between pathologists improved from 0.70 before training to 0.74 after training (p=0.01). When compared with the ISUP consensus panel, five pathologists improved significantly, whereas the kappa of one pathologist was significantly lower after training. All pathologists who improved significantly, graded with less than substantial agreement before training.

CONCLUSIONS

ISUP Prostate Test B e-learning reduces variability in PCa grading. E-learning is a cost-effective method for standardisation of pathology.

Web-grading-a tool to test personal grading of renal and prostate cancer

Only a few pathologists have the opportunity to verify their personal grading through objective assessment. This study introduces a web-based grading platform to facilitate and validate the grading of renal cell carcinoma and prostate cancer. Two representative images of two clinically annotated cohorts of 100 cases each of prostate and renal cell carcinoma were used. Each participant was asked to grade a tumor series utilizing a three tiered grading system. Finally, a Kaplan-Meier curve was drawn, and the log-rank test was used for statistical testing of the p-value. The grading of 22 participants (68%) achieved prognostic significance. Further analysis highlighted that only two pathologists were able to reliably separate low- and high-grade tumors from intermediate grades. The limitations of this study are the low number of participants in each of the cohorts and the potential selection bias of the tumor images. This web-based grading portal facilitates the assessment of the validity of grading by individual pathologists. The observation that most participants can only successfully identify high- or low-grade tumors but cannot discriminate between more subtle intermediate grades does indicate that there is a need for the development of more formal training programs for tumor grading.

Interlaboratory Gleason grading variation affects treatment: a Dutch historic cohort study in 30 509 patients with prostate cancer

AIM

Substantial variation in Gleason grading (GG) of prostate cancer (PCa) exists between Dutch pathology laboratories. This study investigates its impact on treatment strategies.

METHODS

Pathology reports of prostate needle biopsies and clinical data of patients with PCa diagnosed between 2017 and 2019 were retrieved from the Dutch nationwide network and registry of histopathology and cytopathology and The Netherlands Cancer Registry. We investigated the impact of grading variation on treatment strategy for patients whose grade was decisive in treatment choice. First, we evaluated the effect of grading practice (low, average or high grading) on active treatment (AT) versus active surveillance in patients with prostate-specific antigen (PSA) <10 ng/mL and cT1c/cT2a disease. Second, we assessed the association of grading practice with performance of pelvic lymph node dissection (PLND) in patients with PSA 10-20 ng/mL or cT2b disease. We used multivariable logistic regression to analyse the relation between laboratories' grading practices and AT or PLND.

RESULTS

We included 30 509 patients. GG was decisive in treatment strategy for 11 925 patients (39%). AT was performed significantly less often in patients diagnosed by laboratories that graded lower than average (OR=0.77, 95% CI 0.68 to 0.88). Conversely, patients received AT significantly more often when diagnosed in high-grading laboratories versus average-grading laboratories (OR=1.21, 95% CI 1.03 to1.43). PLND was performed significantly less often in patients diagnosed by low-grading versus average-grading laboratories (OR=0.66, 95% CI 0.48 to 0.90).

CONCLUSION

Our study shows that the odds that a patient undergoes AT or PLND, depends on laboratories' grading practices in a substantial number of patients. This likely influences patient prognosis and outcome, necessitating standardisation of GG to prevent suboptimal patient outcome.

AI Model for Prostate Biopsies Predicts Cancer Survival

An artificial intelligence (AI) algorithm for prostate cancer detection and grading was developed for clinical diagnostics on biopsies. The study cohort included 4221 scanned slides from 872 biopsy sessions at the HUS Helsinki University Hospital during 2016–2017 and a subcohort of 126 patients treated by robot-assisted radical prostatectomy (RALP) during 2016–2019. In the validation cohort (n = 391), the model detected cancer with a sensitivity of 98% and specificity of 98% (weighted kappa 0.96 compared with the pathologist’s diagnosis). Algorithm-based detection of the grade area recapitulated the pathologist’s grade group. The area of AI-detected cancer was associated with extra-prostatic extension (G5 OR: 48.52; 95% CI 1.11–8.33), seminal vesicle invasion (cribriform G4 OR: 2.46; 95% CI 0.15–1.7; G5 OR: 5.58; 95% CI 0.45–3.42), and lymph node involvement (cribriform G4 OR: 2.66; 95% CI 0.2–1.8; G5 OR: 4.09; 95% CI 0.22–3). Algorithm-detected grade group 3–5 prostate cancer depicted increased risk for biochemical recurrence compared with grade groups 1–2 (HR: 5.91; 95% CI 1.96–17.83). This study showed that a deep learning model not only can find and grade prostate cancer on biopsies comparably with pathologists but also can predict adverse staging and probability for recurrence after surgical treatment.

Significant Inter- and Intralaboratory Variation in Gleason Grading of Prostate Cancer: A Nationwide Study of 35,258 Patients in The Netherlands

Simple Summary

Gleason grading of prostate cancer is essential for treatment strategies and patient prognosis. Previous studies showed grading variation between pathologists when grading prostate cancer. Our study analyzed the presence and extent of grading variation between and within pathology laboratories in The Netherlands. In our nationwide retrospective study, we analyzed prostate needle biopsy reports of 35,258 patients in The Netherlands graded by 40 pathology laboratories. We found a considerable variation between and within pathology laboratories, as over half of the laboratories graded significantly different from the national mean. This likely affects treatment strategy and prognosis assessment of prostate cancer patients.

Abstract

Purpose: Our aim was to analyze grading variation between pathology laboratories and between pathologists within individual laboratories using nationwide real-life data. Methods: We retrieved synoptic (n = 13,397) and narrative (n = 29,377) needle biopsy reports from the Dutch Pathology Registry and prostate-specific antigen values from The Netherlands Cancer Registration for prostate cancer patients diagnosed between January 2017 and December 2019. We determined laboratory-specific proportions per histologic grade and unadjusted odds ratios (ORs) for International Society of Urological Pathologists Grades 1 vs. 2–5 for 40 laboratories due to treatment implications for higher grades. Pathologist-specific proportions were determined for 21 laboratories that consented to this part of analysis. The synoptic reports of 21 laboratories were used for analysis of case-mix correction for PSA, age, year of diagnosis, number of biopsies and positive cores. Results: A total of 38,321 reports of 35,258 patients were included. Grade 1 ranged between 19.7% and 44.3% per laboratory (national mean = 34.1%). Out of 40 laboratories, 22 (55%) reported a significantly deviant OR, ranging from 0.48 (95% confidence interval (CI) 0.39–0.59) to 1.54 (CI 1.22–1.93). Case-mix correction was performed for 10,294 reports, altering the status of 3/21 (14%) laboratories, but increasing the observed variation (20.8% vs. 17.7%). Within 15/21 (71%) of laboratories, significant inter-pathologist variation existed. Conclusion: Substantial variation in prostate cancer grading was observed between and within Dutch pathology laboratories. Case-mix correction did not explain the variation. Better standardization of prostate cancer grading is warranted to optimize and harmonize treatment.

Automated Gleason grading of prostate cancer using transfer learning from general-purpose deep-learning networks

Two deep-learning algorithms designed to classify images according to the Gleason grading system that used transfer learning from two well-known general-purpose image classification networks (AlexNet and GoogleNet) were trained on Hematoxylin–Eosin histopathology stained microscopy images with prostate cancer. The dataset consisted of 439 images asymmetrically distributed in four Gleason grading groups. Mean and standard deviation accuracy for AlexNet derivate network was of 61.17±7 and for GoogleNet derivate network was of 60.9±7.4. The similar results obtained by the two networks with very different architecture, together with the normal distribution of classification error for both algorithms show that we have reached a maximum classification rate on this dataset. Taking into consideration all the constraints, we conclude that the resulted networks could assist pathologists in this field, providing first or second opinions on Gleason grading, thus presenting an objective opinion in a grading system which has showed in time a great deal of interobserver variability.

Radio-pathomic mapping model generated using annotations from five pathologists reliably distinguishes high-grade prostate cancer

Purpose: Our study predictively maps epithelium density in magnetic resonance imaging (MRI) space while varying the ground truth labels provided by five pathologists to quantify the downstream effects of interobserver variability. Approach: Clinical imaging and postsurgical tissue from 48 recruited prospective patients were used in our study. Tissue was sliced to match the MRI orientation and whole-mount slides were stained and digitized. Data from 28 patients ( n = 33 slides) were sent to five pathologists to be annotated. Slides from the remaining 20 patients ( n = 123 slides) were annotated by one of the five pathologists. Interpathologist variability was measured using Krippendorff's alpha. Pathologist-specific radiopathomic mapping models were trained using a partial least-squares regression using MRI values to predict epithelium density, a known marker for disease severity. An analysis of variance characterized intermodel means difference in epithelium density. A consensus model was created and evaluated using a receiver operator characteristic classifying high grade versus low grade and benign, and was statistically compared to apparent diffusion coefficient (ADC). Results: Interobserver variability ranged from low to acceptable agreement (0.31 to 0.69). There was a statistically significant difference in mean predicted epithelium density values ( p < 0.001 ) between the five models. The consensus model outperformed ADC (areas under the curve = 0.80 and 0.71, respectively, p < 0.05 ). Conclusion: We demonstrate that radiopathomic maps of epithelium density are sensitive to the pathologist annotating the dataset; however, it is unclear if these differences are clinically significant. The consensus model produced the best maps, matched the performance of the best individual model, and outperformed ADC.

Agreement of two pre-trained deep-learning neural networks built with transfer learning with six pathologists on 6000 patches of prostate cancer from Gleason2019 Challenge

INTRODUCTION

While the visual inspection of histopathology images by expert pathologists remains the golden standard method for grading of prostate cancer the quest for developing automated algorithms for the job is set and deep-learning techniques have emerged on top of other approaches.

METHODS

Two pre-trained deep-learning networks, obtained with transfer learning from two general purpose classification networks - AlexNet and GoogleNet, originally trained on a proprietary dataset of prostate cancer were used to classify 6000 cropped images from Gleason2019 Challenge.

RESULTS

The average agreement between the two networks and the six pathologists was found to be substantial for AlexNet and moderate for GoogleNet. When tested against the majority vote of the six pathologists the agreement was perfect and moderate for AlexNet, and GoogleNet, respectively. Despite our expectations, the average inter-pathologist agreement was moderate, while between the two networks it was substantial. Resulted accuracy for AlexNet and GoogleNet when tested against the majority vote as ground truth was of 85.51% and 74.75%, respectively. This result was higher than the score obtained on the dataset that they were trained on, showing their generalization capabilities.

CONCLUSIONS

Both the agreement and the accuracy indicate a better performance of AlexNet over GoogleNet, making it suitable for clinical deployment thus could potentially contribute to faster, more accurate and with higher reproducibility prostate cancer diagnosis.

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 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.

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.

Review by urological pathologists improves the accuracy of Gleason grading by general pathologists

Backgrounds

Urologists use biopsy Gleason scores for patient counseling, prognosis prediction, and decision making. The accuracy of Gleason grading is very important. However, the variability of Gleason grading between general pathologists cannot be overlooked. Here we evaluate the discrepancy in the Gleason grading between 2 urologic pathologists and general pathologists as well as improvement in the accuracy of Gleason grading by general pathologists as a result of review by urologic pathologists.

Methods

The subjects enrolled in the study were 755 patients who underwent prostate needle biopsy at affiliate hospitals of Nara Medical University over a period of 2 years. The biopsy samples were diagnosed by general pathologists. All biopsy samples were sent to Nara Medical University where they were diagnosed by 2 urologic pathologists. The results were then returned to the general pathologists. We compared the diagnostic accuracy of the general pathologists with that of the urologic pathologists for the parameters of no malignancy, atypical small acinar proliferation, high grade prostatic intraepithelial neoplasia and Gleason score (6, 3 + 4, 4 + 3 and 8–10). We then evaluated the concordance rate between the general and urologic pathologists for each of four consecutive 6-month periods.

Results

The overall concordance rate of urologic pathologists and general pathologists in the first, second, third and last 6-month periods was 71.8 % (140/198), 79.8 % (168/225), 89.7 % (166/185) and 89.9 % (133/148), respectively. The concordance rate of the Gleason score between urologic pathologists and general pathologists in the first, second, third and last 6-month periods was 47.5 %(38/80), 62.6 %(57/91),76.9 %(50/65) and 78.7 %(48/61), respectively, and the kappa value was 0.55, 0.68, 0.81 and 0.84, respectively. The concordance rate improved significantly over the course of each period (P = 0.04).

Conclusion

The concordance rate of the Gleason grading between the general pathologists and the urologic pathologists was 47.5 %. However, improvement of the concordance rate as a result of review by the urological pathologist could be seen.

Frequency and determinants of disagreement and error in gleason scores: a population-based study of prostate cancer

BACKGROUND

To examine factors that affect accuracy and reliability of prostate cancer grade we compared Gleason scores documented in pathology reports and those assigned by urologic pathologists in a population-based study.

METHODS

A stratified random sample of 318 prostate cancer cases was selected to ensure representation of whites and African-Americans and to include facilities of various types. The slides borrowed from reporting facilities were scanned and the resulting digital images were re-reviewed by two urologic pathologists. If the two urologic pathologists disagreed, a third urologic pathologist was asked to help arrive at a final "gold standard" result. The agreements between reviewers and between the pathology reports and the "gold standard" were examined by calculating kappa statistics. The determinants of discordance in Gleason scores were evaluated using multivariate models with results expressed as odds ratios (OR) and 95% confidence intervals (CI).

RESULTS

The kappa values (95% CI) reflecting agreement between the pathology reports and the "gold standard," were 0.61 (95% CI: 0.54, 0.68) for biopsies, and 0.37 (0.23, 0.51) for prostatectomies. Sixty three percent of discordant biopsies and 72% of discordant prostatectomies showed only minimal differences. Using freestanding laboratories as reference, the likelihood of discordance between pathology reports and expert-assigned biopsy Gleason scores was particularly elevated for small community hospitals (OR = 2.98; 95% CI: 1.73, 5.14).

CONCLUSIONS

The level of agreement between pathology reports and expert review depends on the type of diagnosing facility, but may also depend on the level of expertise and specialization of individual pathologists.

Reducing interrater variability and improving health care: a meta-analytical review

OBJECTIVES

In the scientific literature about reliability, the main approach to increasing reliability seems to involve increasing the number of observers and improving the instrument used. Other aspects for improving reliability - like the training of raters - seem to receive less notice. It is worth asking whether this technical approach could be complemented by training the user of the instrument. A systematic meta-analytical review of the research literature was performed to answer this question and examine the effectiveness of planned interventions for improving interrater reliability of health care professionals.

METHOD

The databases of PubMed (MEDLINE), Embase, Omega and PsycINFO were searched. The inclusion criteria were met by 57 studies. Details extracted from the studies included the study design, the number of observers and the number of observed cases, the intervention, the type of instrument (whether or not it was highly technical), and statistical information about the agreement before and after the intervention. Interventions were categorized into three groups: training of professionals, improving the diagnostic instrument and a combination of training and improving the instrument. A meta-analysis was performed by means of linear regression.

RESULTS

The interventions were arranged according to their effectiveness in improving the diagnostic instrument (mean change: β = 0.13), training combined with improving the instrument (mean change: β = 0.10) and training (mean change: β = 0.09).

CONCLUSION

On average, although all types of interventions are effective, improving the diagnostic instrument seems to be the most effective. Especially when highly technical instruments were concerned, improvement proved to be very effective (β = 0.52). Because instrumental variables constitute a major source of error, improving the instrument is an important approach. However, this review offers solid arguments that can complement the literature and practice, with a focus on training the user of the instrument.

Predictors for clinically relevant Gleason score upgrade in patients undergoing radical prostatectomy

OBJECTIVE

To evaluate clinical predictors for Gleason score upgrade (GSU) in radical prostatectomy (RP) specimen, especially in patients with 'very' low risk PCA (T1c and biopsy Gleason score ≤6 and PSA <10 ng/ml and ≤2 positive biopsy cores and PSA density <0.15).

PATIENTS AND METHODS

402 consecutive patients undergoing RP between 2004 and 2006, including a subgroup of 62 patients with 'very' low risk PCA, were examined. Patients were categorized for clinically relevant GSU (defined as upgrade into a higher PCA risk category). Parameters including number of biopsy cores obtained, positive biopsy cores, prostate weight, PSA, DRE and pathology department were evaluated for their role as predictors. Furthermore, GSU in RP specimen was analyzed for its impact on pT-stage.

RESULTS

Clinically relevant GSU occurred in 38.1% in the whole cohort and in 32.3% in the 'very' low risk PCA subgroup. Gleason score downgrade (GSD) occurred in 4.7%. Number of biopsy cores obtained and prostate weight were independent negative predictors of GSU in all 402 patients (P = 0.02 and P = 0.03, respectively). In the 'very' low risk group, only number of biopsy cores obtained revealed as an independent negative predictor of GSU (P = 0.02). PSA, DRE, number of positive cores or pathology department were not associated to GSU. In the 'very' low risk group, GSU was related with extracapsular tumor extension (P = 0.05).

CONCLUSIONS

Clinically relevant GSU in RP specimen is still a challenging problem. Increasing the number of biopsy cores lower this risk significantly. GSD is rare and thus of minor importance for treatment decisions.

Reliability and reproducibility of the edmondson grading of hepatocellular carcinoma using paired core biopsy and surgical resection specimens

CONTEXT

It has been claimed that the Edmondson and Steiner grading system (EGS) values should be obtained preoperatively to select patients with hepatocellular carcinoma for liver transplantation. However, EGS reliability in biopsy specimens has been questioned.

OBJECTIVE

To verify the reliability of the EGS using core biopsy specimens and its reproducibility among pathologists.

DESIGN

Paired biopsy and surgical specimens obtained from 40 patients (subset 1) were retrieved by means of computer-aided search of the pathology records and blindly and independently reviewed. The EGS interrater agreement was measured using κ statistics. After having held a consensus meeting, pathologists graded an additional 21 paired hepatocellular carcinoma specimens (subset 2).

RESULTS

Analyzing subset 1, pathologists gave significantly lower EGS grades to the biopsy specimens (P < .001), for which the observed agreement was 32.5% (κ  =  0.021), which increased to 82.5% (κ  =  0.186) if only 2 categories were considered (low grade, EGS I-II; high grade, EGS III-IV). The observed agreement in the case of the surgical specimens was 52.5% (κ  =  0.199), which increased to 62.5% (κ  =  0.275) when the low- and high-grade scores were merged. The observed agreement between the assessments of paired biopsy and surgical specimens was 50.0% for pathologist 1 (κ  =  0.057) and 35.0% for pathologist 2 (κ  =  0.078). Merging the EGS grades did not improve the strength of the agreement. Analyzing subset 2 (after the consensus meeting), the observed agreement between pathologists improved more on biopsies (76.2%, κ  =  0.614) than on surgical specimens (61.9%, κ  =  0.434).

CONCLUSIONS

The EGS is easily underestimated in core biopsy specimens, and interrater disagreement between pathologists can be significant unless consensus meetings are held.

Interobserver reproducibility of Gleason grading: evaluation using prostate cancer tissue microarrays

OBJECTIVES

Due to PSA screening and increased awareness, prostate cancer (PCa) is identified earlier resulting in smaller diagnostic samples on prostate needle biopsy. Because Gleason grading plays a critical role in treatment planning, we undertook a controlled study to evaluate interobserver variability among German pathologists to grade small PCas using a series of tissue microarray (TMA) images.

METHODS

We have previously demonstrated excellent agreement in Gleason grading using TMAs among expert genitourinary pathologists. In the current study, we identified 331 TMA images (95% PCa and 5% benign) to be evaluated by an expert PCa pathologist and subsequently by practicing pathologists throughout Germany. The images were presented using the Bacus Webslide Browser on a CD-ROM. Evaluations were kept anonymous and participant's scoring was compared to the expert's results.

RESULTS

A total of 29 German pathologists analysed an average of 278 images. Mean percentage of TMA images which had been assigned the same Gleason score (GS) as done by the expert was 45.7%. GSs differed by no more than one point (+/-1) in 83.5% of the TMA samples evaluated. The respondents were able to correctly assign a GS into clinically relevant categories (i.e. <7, 7, >7) in 68.3% of cases. A total of 75.9% respondents under-graded the TMA images. Gleason grading agreement with the expert reviewer correlated with the number of biopsies evaluated by the pathologist per week. Years of diagnostic experience, self-description as a urologic pathologist or affiliation with a university hospital did not correlate with the pathologist's performance.

CONCLUSION

The vast majority of participants under-graded the small tumors. Clinically relevant GS categories were correctly assigned in 68% of cases. This raises a potentially significant problem for pathologists, who have not had as much experience evaluating small PCas.

Significant discrepancies between diagnostic and pathologic Gleason sums in prostate cancer: the predictive role of age and prostate-specific antigen

OBJECTIVES

To assess the discrepancies between diagnostic and pathologic Gleason sums and the predictive role of age and prostate-specific antigen (PSA) level on Gleason sum discrepancies.

METHODS

A total of 2963 patients receiving radical prostatectomy at Duke University from 1988 to 2006 were divided into two groups according to year of diagnosis: 1988 to 1999 and 2000 to 2006. The Gleason sum discrepancies were evaluated in the above groups. The predictive roles of diagnostic age (less than 50, 50 to 60, 60.1 to 70, and greater than 70 years), PSA level (less than 10, 10 to 20, and greater than 20 ng/mL), race, body mass index, and prostate weight on the discrepancies were analyzed.

RESULTS

Overall, 55.8% of diagnostic Gleason sums differed from those on final surgical pathology (58.6% in the 1988 to 1999 and 49.3% in the 2000 to 2006 groups). Diagnostic Gleason sums were undergraded in 41.2% of cases and overgraded in 12.8% of cases. Men older than 60 years were more likely to have their diagnostic Gleason sums undergraded than men younger than 50 (odds ratio in age groups less than 50, 50 to 60, 60.1 to 70, and greater than 70 years: 1.00, 2.30, 4.03, and 3.96, respectively). Biopsy Gleason sums in men with a high PSA level were more likely to be undergraded compared with the PSA group less than 10 ng/mL (odds ratio in PSA groups less than 10, 10 to 20, and greater than 20 ng/mL: 1.00, 2.11, and 3.64, respectively).

CONCLUSIONS

Significant discrepancies between diagnostic and pathologic Gleason sums remain in recent years. The rate of diagnostic Gleason sum undergrading was 3.2-fold that of overgrading. Advanced age and high PSA level were predictive of diagnostic Gleason sum undergrading, and caution should be exercised when recommending active surveillance in older men.

Reproducibility and reliability of tumor grading in urological neoplasms

Histopathologic tumor grading reflects the degree of differentiation of a given tumor and for most urological tumors grading is an important factor in predicting their biological aggressiveness. Consequently, the clinical management of tumor patients is often strongly influenced by the tumor grade, provided by pathologists. This implicates that an ideal grading system should not only be of high prognostic relevance, but also of high reproducibility among different pathologists. To this end individual histological grading systems have been developed for different tumor entities and even for a given tumor type several grading systems have been proposed. All of these grading systems possess an inherent degree of subjectivity and consequently, both intra- and interobserver variability exist. In this review, grading systems for the most frequent urological tumors (i.e. prostate cancer, renal cell carcinoma, and urothelial tumors) are mentioned and data on the reproducibility and reliability of the most commonly used grading systems are summarized.

Predicting unilateral prostate cancer based on biopsy features: implications for focal ablative therapy--results from the SEARCH database

PURPOSE

For men with low risk prostate cancer it was recently proposed that ablative treatment to the affected side may decrease morbidity, while maintaining good oncological outcomes. However, few studies have assessed the correlation between biopsy parameters and pathological outcome (unilateral vs bilateral disease).

MATERIALS AND METHODS

Using the Shared Equal Access Regional Cancer Hospital Database of men treated with radical prostatectomy at multiple equal access medical centers we retrospectively examined the records of 261 men with clinical stage T1c or T2a prostate cancer, prostate specific antigen less than 10 ng/ml, Gleason sum 6 or less and only 1 or 2 ipsilateral positive cores on at least sextant biopsy. We compared clinical characteristics between men with pathologically unilateral disease or less (pT2b or less) and men with pathologically bilateral disease or extraprostatic extension (pT2c or greater). To determine the significant predictors of pT2c or greater disease we used a multivariate logistic regression model.

RESULTS

Of the cohort of 261 men with low risk prostate cancer only 93 (35.1%) had unilateral or no evidence of disease following examination of radical prostatectomy specimens. Men with pathologically unilateral or less disease did not differ from those with bilateral or more advanced disease by age, prostate specific antigen, clinical stage, body mass index or number of positive biopsy cores (1 vs 2). On multivariate analysis no clinical feature was significantly related to pathologically unilateral or less vs bilateral or greater disease.

CONCLUSIONS

The majority of men with low risk prostate cancer and 1 or 2 ipsilateral positive biopsy cores have pathologically bilateral disease. Therefore, strategies for unilateral treatment of prostate cancer are unlikely to be curative for these men.

A study of Gleason score interpretation in different groups of UK pathologists; techniques for improving reproducibility

AIMS

To test the effectiveness of a teaching resource (a decision tree with diagnostic criteria based on published literature) in improving the proficiency of Gleason grading of prostatic cancer by general pathologists.

METHODS

A decision tree with diagnostic criteria was developed by a panel of urological pathologists during a reproducibility study. Twenty-four general histopathologists tested this teaching resource. Twenty slides were selected to include a range of Gleason score groups 2-4, 5-6, 7 and 8-10. Interobserver agreement was studied before and after a presentation of the decision tree and criteria. The results were compared with those of the panel of urological pathologists.

RESULTS

Before the teaching session, 83% of readings agreed within +/- 1 of the panel's consensus scores. Interobserver agreement was low (kappa = 0.33) compared with that for the panel (kappa = 0.62). After the presentation, 90% of readings agreed within +/- 1 of the panel's consensus scores and interobserver agreement amongst the pathologists increased to kappa = 0.41. Most improvement in agreement was seen for the Gleason score group 5-6.

CONCLUSIONS

The lower level of agreement among general pathologists highlights the need to improve observer reproducibility. Improvement associated with a single training session is likely to be limited. Additional strategies include external quality assurance and second opinion within cancer networks.

Trends in distribution and prognostic significance of Gleason grades on radical retropubic prostatectomy specimens between 1989 and 2001

BACKGROUND

The objectives of the current study were to examine time trends in the prevalence of Gleason grades of prostate cancer on radical retropubic prostatectomy (RRP) specimens and to assess the resultant impact on prognosis.

METHODS

The authors examined the prevalence over time of each grade and Gleason score (GS) on RRP specimens from 8750 patients who were treated between 1989 and 2001. Biochemical recurrence-free survival (BRFS), which was estimated by using Kaplan-Meier methodology, was examined in subgroups of patients defined by tumor grade and era of surgery.

RESULTS

The prevalence of Grade 3 prostate cancers increased (86% vs. 49% for primary Gleason grade and 71% vs. 47% for secondary Gleason grade; 1999-2001 vs. 1989-1990, respectively), whereas the prevalence of Grade 2 tumors decreased (0.4% vs. 38% for primary Gleason grade and 1.3% vs. 28% for secondary Gleason grade, respectively) over the study period, leading to fewer GS 4 and 5 tumors and more GS 6 and 7 tumors. BRFS improved over time for patients who had GS 5 tumors (hazards ratio [HR], 0.92 per year; P = .003) and GS 6 tumors (HR, 0.93; P < .001) but remained unchanged for GS 7 tumors (HR 0.99; P = .462) and GS 8-10 tumors (HR 1.02; P = .360). Patients who were treated in the recent era (1997-2001) had greater differentiation of BRFS based on GS or Gleason grade compared with patients who were treated earlier (1989-1991).

CONCLUSIONS

The current results confirmed that there were changes in the prevalence of Gleason grades on RRP specimens between 1989 and 2001. A chronological change in pathologic grading classification is suggested by evolving prognostic implications, which must be accounted for when comparing outcomes from different eras.

Gleason Scoring Varies Among Pathologists and this Affects Clinical Risk in Patients with Prostate Cancer

AIMS

To investigate whether our practice of specialist review of all diagnostic biopsies was necessary to prevent misgrading of referred prostate cancer patients, and whether this misclassification, if any, would have resulted in misclassification of clinical risk grouping (Seattle Risk Grouping [SRG]) and subsequent treatment strategy and prognosis.

MATERIALS AND METHODS

Important prognostic indicators for prostate cancer include the presenting prostate-specific antigen (PSA), clinical stage and Gleason sum of the tumour. These three variables are incorporated into the SRG cohorts to establish treatment strategy. Patients with prostate cancer referred for brachytherapy had their prostate biopsies reviewed by a reference pathologist (PD) with a special interest in prostate cancer. We compared the agreement between the scoring of the referring pathologists with that of PD, and evaluated if any differences changed the SRG and therefore the clinical risk and treatment strategy for the patients.

RESULTS

In only 52% (43/83) of cases, was there total agreement between the two sets of pathologists. The inter-rater agreement was statistically 'fair' (unweighted kappa statistic 0.27). In 90% (36/40) of cases with disagreement, PD assigned higher Gleason sums. In 40% (16/40) of cases with disagreement, the change in Gleason sum altered the SRG; in one out of 16 cases, the SRG was downgraded from 'intermediate' to 'low' risk disease; in six out of 16 cases, it was upgraded from 'low' to 'intermediate' risk, and, in nine out of 16, from 'intermediate' to 'high' risk.

CONCLUSION

Our findings confirm previous reports of only limited correlation between pathologists in reporting Gleason sums. In this study, 19% (16/83) of cases had their grading changed to a level that altered clinical risk, almost always (94%; 15/16) to one that worsened prognosis. This would have significantly affected treatment strategy for these patients, and thus we recommend that all centres ensure accurate Gleason grading by the use of pathologists with special interests in prostate cancer.

A comparison of interobserver reproducibility of Gleason grading of prostatic carcinoma in Japan and the United States

CONTEXT

Gleason grading is now the sole prostatic carcinoma grading system recommended by the World Health Organization. It is imperative that there be good interobserver reproducibility within this system worldwide. To our knowledge, there are no studies, using the same specimens, that compare the interobserver reproducibility of Gleason grading in Japan and the United States.

OBJECTIVE

To compare the interobserver reproducibility of Gleason grading of prostatic carcinoma in Japan and the United States using, in Japan, images from the identical biopsy glass slides that were originally graded in the United States.

DESIGN

Microsopic images from 37 needle biopsies of prostatic carcinoma were placed on CD-ROM and distributed to 14 Japanese pathologists for grading. These 14 physicians included 8 general pathologists and 6 pathologists with a special interest in urologic pathology. The needle biopsies had been previously reviewed so that a consensus diagnosis could be formed by a panel of urologic pathologists in the United States and Canada. Interobserver agreement with the consensus diagnoses was calculated by determining the overall kappa coefficient for the Japanese pathologists and then compared to the interobserver agreement among American general pathologists who had previously graded identical needle biopsies from which the CD-ROM images had been taken.

RESULTS

The interobserver agreement with the consensus diagnoses for the 4 Gleason grading groups (Gleason grades 2-4, 5-6, 7, and 8-10) among the Japanese urologic pathologists in this series of cases was substantial (overall kappa = 0.68), and for the Japanese general pathologists, it was moderate (overall kappa = 0.49), similar to that reported in the earlier study of American general pathologists (overall kappa = 0.44). The major interobserver reproducibility problem for both Japanese and American general pathologists is undergrading. The major areas of undergrading are the underdiagnosis of Gleason scores 5-6 as Gleason scores 2-4, and the underdiagnosis of cribriform sheets and fragments of cribriform Gleason pattern 4 carcinoma as Gleason pattern 3.

CONCLUSIONS

The interobserver reproducibility of the Gleason grading for this collection of specimens was similar among Japanese and American general pathologists. The overall kappa values for these generalists of 0.44 and 0.49 are only in the moderate (0.41-0.60) range of interobserver agreement when compared to 0.68, substantial (0.61-0.80) agreement, for Japanese urologic pathologists. Educational efforts to improve Gleason grading have been shown to be effective and are clearly warranted.

Representación del “score” gleason en la biopsia prostática del “score” gleason definitivo de la prostatectomía radical

INTRODUCTION

The Gleason score (Gs) for prostatic cancer has a good prognosis correlation after radical prostatectomy, for this reason its correlation with the Gs in the biopsy can be useful.

PATIENTS AND METHODS

Two hundred fifteen patients with blind evaluation among three pathologists of their Gs in biopsy and in the corresponding radical prostatectomy specimen are presented.

RESULTS

The exact coincidence is present in 49.7% of cases, 38.6% of cases are under graded in the biopsy and 11.6% of them over graded in the biopsy. No cases of Gs 2 in the biopsy are found. Any case with Gs 3 and 4 in the biopsy are reproduced in the radical prostatectomy specimen. The exact coincidence for biopsy Gs 5, 6, 7, 8 and 9 are 25%, 45%, 72.7%, 36.6% and 60% respectively (kappa 0.32 +/- 0.047, p<0.0001 in Gs 5 to 8). The Gleason pattern 4 is the less diagnosed in prostate biopsies [40% of cases with this pattern in the excision specimen it is missing in the biopsy).

CONCLUSIONS

The Gs in the needle prostatic biopsy has a good correspondence with the Gs in the radical prostatectomy specimen. For an increase of the reproducibility it is recommendable avoid the diagnosis of Gs 2, 3 and 4 in biopsy and a scrupulous search for the patterns 4 and 5.

Risk-Factors of Recurrence after Retropubic and Perineal Radical Prostatectomy

The Authors review their own surgical series about retropubic and perineal radical prostatectomy performed in 1993–2002 ys, with particular attention for risk factors of biochemical recurrence of disease.