Dataset Information

Computationally Derived Cribriform Area Index from Prostate Cancer Hematoxylin and Eosin Images Is Associated with Biochemical Recurrence Following Radical Prostatectomy and Is Most Prognostic in Gleason Grade Group 2.

ABSTRACT:

Background

The presence of invasive cribriform adenocarcinoma (ICC), an expanse of cells containing punched-out lumina uninterrupted by stroma, in radical prostatectomy (RP) specimens has been associated with biochemical recurrence (BCR). However, ICC identification has only moderate inter-reviewer agreement.

Objective

To investigate quantitative machine-based assessment of the extent and prognostic utility of ICC, especially within individual Gleason grade groups.

Design, setting, and participants

A machine learning approach was developed for ICC segmentation using 70 RP patients and validated in a cohort of 749 patients from four sites whose median year of surgery was 2007 and with median follow-up of 28 mo. ICC was segmented on one representative hematoxylin and eosin RP slide per patient and the fraction of tumor area composed of ICC, the cribriform area index (CAI), was measured.

Outcome measurements and statistical analysis

The association between CAI and BCR was measured in terms of the concordance index (c index) and hazard ratio (HR).

Results and limitations

CAI was correlated with BCR (c index 0.62) in the validation set of 411 patients with ICC morphology, especially those with Gleason grade group 2 cancer (n = 192; c index 0.66), and was less prognostic when patients without ICC were included (c index 0.54). A doubling of CAI in the group with ICC morphology was prognostic after controlling for Gleason grade, surgical margin positivity, preoperative prostate-specific antigen level, pathological T stage, and age (HR 1.19, 95% confidence interval 1.03-1.38; p = 0.018).

Conclusions

Automated image analysis and machine learning could provide an objective, quantitative, reproducible, and high-throughput method of quantifying ICC area. The performance of CAI for grade group 2 cancer suggests that for patients with little Gleason 4 pattern, the ICC fraction has a strong prognostic role.

Patient summary

Machine-based measurement of a specific cell pattern (cribriform; sieve-like, with lots of spaces) in images of prostate specimens could improve risk stratification for patients with prostate cancer. In the future, this could help in expanding the criteria for active surveillance.

SUBMITTER: Leo P

PROVIDER: S-EPMC8419103 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND:While the new Gleason grade grouping (GGG), which started in 2016, has been widely validated in prostate cancer, it does not incorporate the concept of tertiary Gleason pattern 5. Furthermore, no study has "quantified" the individual risk of each Gleason pattern, including tertiary Gleason pattern 5, after radical prostatectomy. METHODS:We reviewed 1022 men with adjuvant-treatment-naïve prostate cancer who underwent radical prostatectomy between 2005 and 2017. The primary endpoint was biochemical recurrence-free survival, defined as two consecutive prostate-specific antigen measurements ?0.2?ng/ml after surgery. The individual quantitative risk score (IQRS) of each amount (primary/secondary/tertiary) of each Gleason pattern (3/4/5) was calculated using the Cox regression model. On the basis of the IQRS, the modified Gleason grade grouping (mGGG) model was developed. As a robustness analysis of the mGGG model, salvage treatment-free survival was also assessed. RESULTS:During a median follow-up of 45?months, 229 of 1022 (22.4%) patients developed biochemical recurrence. The IQRS of each Gleason pattern was as follows: primary 5, 1.81 points (hazard ratio [HR] 6.13); secondary 5, 1.37 points (HR 3.92); tertiary 5, 0.87 points (HR 2.39); primary 4, 1.07 points (HR 2.91); secondary 4, 0.79 points (HR 2.21); and any Gleason pattern 3, 0 points (HR 1). Based on the IQRS, the mGGG model was developed, which classified patients into the following five groups: I (3?+?3 or less); II (3?+?4); III (4?+?3); IV (3?+?4?+?t5, 4?+?3?+?t5, 3?+?5, 5?+?3, and 4?+?4); V (4?+?4?+?t5, 4?+?5, 5?+?4, and 5?+?5). The c-index for biochemical recurrence-free survival was significantly improved from 0.655 of the original GGG model to 0.672 of the mGGG model (P?<?0.05). In the robustness analysis, the c-index for salvage treatment-free survival was also significantly improved from 0.619 of the original GGG model to 0.638 of the mGGG model (P?<?0.05). CONCLUSIONS:The quantitative risk of tertiary (<?5%) Gleason pattern 5 is slightly higher than that of secondary (5-50%) Gleason pattern 4. Our newly developed mGGG model more accurately predicts outcomes after radical prostatectomy than the original GGG model.

Project description:IntroductionPositive surgical margins (PSM) in cancer patients are commonly associated with worse prognosis and a higher risk of secondary treatment. However, the relevance of this parameter in prostate cancer patients undergoing radical prostatectomy (RP) remains controversial, given the inconsistencies in its ability to predict biochemical recurrence (BCR) and oncological outcomes. Hence, further assessment of the utility of surgical margins for prostate cancer prognosis is required to predict these outcomes more accurately. Over the last decade, studies have used the Gleason score (GS) of positive margins to predict outcomes. Herein, the authors aim to conduct a systematic review investigating the role of GS of PSM after radical prostatectomy in predicting BCR and oncological outcomes.Methods and analysisWe will perform a search using MEDLINE, EMBASE, SCOPUS and COCHRANE databases. The review will be reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We will screen titles and abstracts to select articles appropriate for full-text review. Studies discussing GS of PSM after RP will be included. Given the change in reporting of GS, only articles from 2005 to 2019 will be included. The quality of the studies chosen will be assessed using the Newcastle Ottawa tool for non-randomised and Cochrane risk of bias for randomised control studies. We will adopt the grading of recommendations, assessment, development and evaluation framework to comment on quality of cumulative evidence. The primary outcome measure will be time to BCR. Secondary outcome measures include secondary treatment, disease-specific survival, disease progression-free and overall mortality at follow-up period. We aim to perform a meta-analysis if the level of heterogeneity is acceptable (I2 <50%).Ethics and disseminationThe review does not require ethics approval as it is a review of published literature. The findings of the review will be submitted for peer-reviewed publications and presented at scientific meetings.Prospero registration numberCRD42019131800.

Project description:We assessed the utility of quantitative features of colon cancer nuclei, extracted from digitized hematoxylin and eosin-stained whole slide images (WSIs), to distinguish between stage II and stage IV colon cancers. Our discovery cohort comprised 100 stage II and stage IV colon cancer cases sourced from the University Hospitals Cleveland Medical Center (UHCMC). We performed initial (independent) model validation on 51 (143) stage II and 79 (54) stage IV colon cancer cases from UHCMC (The Cancer Genome Atlas's Colon Adenocarcinoma, TCGA-COAD, cohort). Our approach comprised the following steps: (1) a fully convolutional deep neural network with VGG-18 architecture was trained to locate cancer on WSIs; (2) another deep-learning model based on Mask-RCNN with Resnet-50 architecture was used to segment all nuclei from within the identified cancer region; (3) a total of 26 641 quantitative morphometric features pertaining to nuclear shape, size, and texture were extracted from within and outside tumor nuclei; (4) a random forest classifier was trained to distinguish between stage II and stage IV colon cancers using the five most discriminatory features selected by the Wilcoxon rank-sum test. Our trained classifier using these top five features yielded an AUC of 0.81 and 0.78, respectively, on the held-out cases in the UHCMC and TCGA validation sets. For 197 TCGA-COAD cases, the Cox proportional hazards model yielded a hazard ratio of 2.20 (95% CI 1.24-3.88) with a concordance index of 0.71, using only the top five features for risk stratification of overall survival. The Kaplan-Meier estimate also showed statistically significant separation between the low-risk and high-risk patients, with a log-rank P value of 0.0097. Finally, unsupervised clustering of the top five features revealed that stage IV colon cancers with peritoneal spread were morphologically more similar to stage II colon cancers with no long-term metastases than to stage IV colon cancers with hematogenous spread. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.