Evaluation of Artificial Intelligence-Based Gleason Grading Algorithms "in the Wild".

The biopsy Gleason score is an important prognostic marker for prostate cancer patients. It is, however, subject to substantial variability among pathologists. Artificial intelligence (AI)-based algorithms employing deep learning have shown their ability to match pathologists' performance in assigning Gleason scores, with the potential to enhance pathologists' grading accuracy. The performance of Gleason AI algorithms in research is mostly reported on common benchmark data sets or within public challenges. In contrast, many commercial algorithms are evaluated in clinical studies, for which data are not publicly released. As commercial AI vendors typically do not publish performance on public benchmarks, comparison between research and commercial AI is difficult. The aims of this study are to evaluate and compare the performance of top-ranked public and commercial algorithms using real-world data. We curated a diverse data set of whole-slide prostate biopsy images through crowdsourcing containing images with a range of Gleason scores and from diverse sources. Predictions were obtained from 5 top-ranked public algorithms from the Prostate cANcer graDe Assessment (PANDA) challenge and 2 commercial Gleason grading algorithms. Additionally, 10 pathologists (A.C., C.R., J.v.I., K.R.M.L., P.R., P.G.S., R.G., S.F.K.J., T.v.d.K., X.F.) evaluated the data set in a reader study. Overall, the pairwise quadratic weighted kappa among pathologists ranged from 0.777 to 0.916. Both public and commercial algorithms showed high agreement with pathologists, with quadratic kappa ranging from 0.617 to 0.900. Commercial algorithms performed on par or outperformed top public algorithms.

Image analysis for bright-field HER2 in situ hybridization: validation for clinical use.

The aim of the present study was to develop and validate a quantitative image analysis (IA) algorithm to aid pathologists in assessing bright-field HER2 in situ hybridization (ISH) tests in solid cancers. A cohort of 80 sequential cases (40 HER2-negative and 40 HER2-positive) were evaluated for HER2 gene amplification with bright-field ISH. We developed an IA algorithm using the ISH Module from HALO software to automatically quantify HER2 and CEP17 copy numbers per cell as well as the HER2/CEP17 ratio. We observed a high correlation of HER2/CEP17 ratio, an average of HER2 and CEP17 copy number per cell between visual and IA quantification (Pearson's correlation coefficient of 0.842, 0.916, and 0.765, respectively). IA was able to count from 124 cells to 47,044 cells (median of 5565 cells). The margin of error for the visual quantification of the HER2/CEP17 ratio and of the average of HER2 copy number per cell decreased from a median of 0.23 to 0.02 and from a median of 0.49 to 0.04, respectively, in IA. Curve estimation regression models showed that a minimum of 469 or 953 invasive cancer cells per case is needed to reach an average margin of error below 0.1 for the HER2/CEP17 ratio or for the average of HER2 copy number per cell, respectively. Lastly, on average, a case took 212.1 s to execute the IA, which means that it evaluates about 130 cells/s and requires 6.7 s/mm2. The concordance of the IA software with the visual scoring was 95%, with a sensitivity of 90% and a specificity of 100%. All four discordant cases were able to achieve concordant results after the region of interest adjustment. In conclusion, this validation study underscores the usefulness of IA in HER2 ISH testing, displaying excellent concordance with visual scoring and significantly reducing margins of error.

Optimizing the management of thyroid specimens to efficiently generate whole slide images for diagnosis.

Transition from optical to digital observation requires an additional procedure in the pathology laboratory, the scanning of glass slides, leading to increased time and digital archive consumption. Thyroid surgical samples often carry the need to collect several tissue fragments that generate many slides to be scanned. This study evaluated the impact of using different inking colours for the surgical margin, section thickness, and glass slide type, in the consumption of time and archive. The series comprehended 40 nodules from 30 patients, including 34 benign nodules in follicular nodular disease, 1 NIFTP, and 5 papillary carcinomas. In 12 nodules, the dominant pattern was microfollicular/solid and in 28 it was macrofollicular. Scanning times/mm2 were longer in red-inked fragments in comparison to green (p = 0.04) and black ones (p = 0.024), and in blue-inked in comparison to green ones (p = 0.043). File sizes/mm2 were larger in red-inked fragments in comparison to green (p = 0.008) and black ones (p = 0.002). The dominant pattern microfollicular/solid was associated with bigger file size/mm2 in comparison with the macrofollicular one (p < 0.001). All scanner outputs increase significantly with the thickness of the section. All scanning outputs increase with the usage of adhesive glass slides in comparison to non-adhesive ones. Small interventions in thyroid sample management that can help optimizing the digital workflow include to prefer black and green inking colours for the surgical margins and 2 µm section in non-adhesive glass slides for increased efficiency.

HER2 and PD-L1 Expression in Gastric and Gastroesophageal Junction Cancer: Insights for Combinatorial Targeting Approaches.

Gastric and gastroesophageal junction adenocarcinomas (GA/GEJA) are associated with a poor prognosis, primarily due to late disease diagnosis. Human Epidermal Growth Factor Receptor 2 (HER2) overexpression and programmed death-ligand 1 (PD-L1) expression are important biomarkers for treatment selection in locally advanced unresectable and metastatic GA/GEJA, and there is increasing interest in their role in earlier stages of disease. In this study, we aimed to evaluate HER2 and PD-L1 expression in a curative-intent GA/GEJA cohort to describe their expression patterns and analyze the association between HER2 expression and clinicopathological features. HER2 expression was evaluated in surgical and endoscopic submucosal dissection tumor samples, and PD-L1 was evaluated in HER2-positive cases. The clinical cohort included 107 patients, with 8.4% testing positive for HER2 (seven of whom also exhibited a PD-L1 combined positive score of ≥1. HER2 status was not significantly associated with survival outcomes. A pathologist-guided, region-specific analysis revealed that PD-L1 expression rarely overlaps with HER2-positive tumor areas. While the therapeutic implications of these observations remain unknown, these findings suggest that combination strategies targeting HER2 and PD-L1 might be directed toward distinct tumor subclones. The herein disclosed region-specific biomarker expression patterns may have important therapeutic and prognostic impacts, warranting further evaluation.

Somatic Mutations in KEAP1-NRF2 Complex in Breast Cancer.

Breast cancer remains the leading cause of cancer deaths for women. Long-term estrogen exposure is considered carcinogenic due to semiquinone production and to compromised detoxification. Metabolic regulator polymorphisms, such as KEAP1 (rs1048290) and NRF2 (rs35652124, rs6721961, rs6706649), can be valuable in understanding the individual cytoprotection profile. Thus, we aim to genotype these polymorphisms in blood, tumours and surrounding tissue, to identify somatic mutations and correlate it to prognoses. A total of 23 controls and 69 women with histological confirmed breast cancer were recruited, and DNA from blood/surrounding/tumour tissue was genotyped. Genotyping and clinicopathological data were correlated. We verified that rs35652124 presents different genotype distribution between the blood/surrounding tissue (p-value = 0.023) and tumour/surrounding tissues (p-value = 0.041). Apart from rs35652124 and considering the histological grade, the other four polymorphisms have different distributions among different tissues. There is a tendency towards the loss of heterozygosity in the surrounding tissue when compared to blood and tumour tissues, and higher genotype variability in histologic grade 2. These somatic mutations and different distribution patterns may indicate a heterogeneous and active microenvironment, influencing breast cancer outcome. Additionally, it would be pertinent to evaluate the predictive value of the histologic grade 2 considering somatic mutation profiles and distributions.

The 1000 Mitoses Project: A Consensus-Based International Collaborative Study on Mitotic Figures Classification.

Introduction. The identification of mitotic figures is essential for the diagnosis, grading, and classification of various different tumors. Despite its importance, there is a paucity of literature reporting the consistency in interpreting mitotic figures among pathologists. This study leverages publicly accessible datasets and social media to recruit an international group of pathologists to score an image database of more than 1000 mitotic figures collectively. Materials and Methods. Pathologists were instructed to randomly select a digital slide from The Cancer Genome Atlas (TCGA) datasets and annotate 10-20 mitotic figures within a 2 mm2 area. The first 1010 submitted mitotic figures were used to create an image dataset, with each figure transformed into an individual tile at 40x magnification. The dataset was redistributed to all pathologists to review and determine whether each tile constituted a mitotic figure. Results. Overall pathologists had a median agreement rate of 80.2% (range 42.0%-95.7%). Individual mitotic figure tiles had a median agreement rate of 87.1% and a fair inter-rater agreement across all tiles (kappa = 0.284). Mitotic figures in prometaphase had lower percentage agreement rates compared to other phases of mitosis. Conclusion. This dataset stands as the largest international consensus study for mitotic figures to date and can be utilized as a training set for future studies. The agreement range reflects a spectrum of criteria that pathologists use to decide what constitutes a mitotic figure, which may have potential implications in tumor diagnostics and clinical management.