Mitosis Count Research Articles

PRAME and Historical Immunohistochemical Antibodies Ki-67, P16, and HMB-45 in Ambiguous Melanocytic Tumors.

Ambiguous melanocytic lesions/tumors (AMLs) can be simply described as melanocytic neoplasms that cannot be differentiated as either a melanoma or a nevus. Preferentially expressed antigen in melanoma (PRAME) is a novel antibody that can help differentiate between nevi and melanomas. However, its usefulness remains controversial in AMLs. The aim of this study was to demonstrate the importance of PRAME and diagnostic auxiliary antibodies (Ki-67, p16, HMB-45) in the diagnosis of melanocytic lesions, especially in AMLs. This study included 52 ambiguous melanocytic lesions, 40 nevi, and 40 melanomas. All immunohistochemical studies were performed automatically using the Universal Alkaline Phosphatase Red Detection Kit. Different analytic approaches were used for each antibody based on the literature. Statistically, the multinomial forward stepwise elimination logistic regression analysis was used to create a statistical model to predict the diagnosis of melanocytic lesions based on clinical, morphological, and immunohistochemical data. PRAME positivity was very strong and diffuse in the melanoma group and statistically significantly higher than that of the AML and nevus groups. There was no statistically significant difference between the nevus and AML groups. The Ki-67 proliferation index and HMB-45 staining pattern provided valuable indications for distinguishing between these 3 groups. The P16 antibody was limited in supporting the differential diagnosis. Our statistical model showed that a high mitosis count, central pagetoid spread, and PRAME positivity increased the probability of melanoma against an AML diagnosis. This study showed the advantages of evaluating the PRAME antibody together with morphological features and other immunohistochemical markers (Ki-67 and HMB-45) in the differential diagnosis of melanocytic lesions.

Mitosis detection, fast and slow: Robust and efficient detection of mitotic figures

Counting of mitotic figures is a fundamental step in grading and prognostication of several cancers. However, manual mitosis counting is tedious and time-consuming. In addition, variation in the appearance of mitotic figures causes a high degree of discordance among pathologists. With advances in deep learning models, several automatic mitosis detection algorithms have been proposed but they are sensitive to domain shift often seen in histology images. We propose a robust and efficient two-stage mitosis detection framework, which comprises mitosis candidate segmentation (Detecting Fast) and candidate refinement (Detecting Slow) stages. The proposed candidate segmentation model, termed EUNet, is fast and accurate due to its architectural design. EUNet can precisely segment candidates at a lower resolution to considerably speed up candidate detection. Candidates are then refined using a deeper classifier network, EfficientNet-B7, in the second stage. We make sure both stages are robust against domain shift by incorporating domain generalization methods. We demonstrate state-of-the-art performance and generalizability of the proposed model on the three largest publicly available mitosis datasets, winning the two mitosis domain generalization challenge contests (MIDOG21 and MIDOG22). Finally, we showcase the utility of the proposed algorithm by processing the TCGA breast cancer cohort (1,124 whole-slide images) to generate and release a repository of more than 620K potential mitotic figures (not exhaustively validated).

Efficient mitosis detection: leveraging pre-trained faster R-CNN and cell-level classification

The assessment of mitotic activity is an integral part of the comprehensive evaluation of breast cancer pathology. Understanding the level of tumor dissemination is essential for assessing the severity of the malignancy and guiding appropriate treatment strategies. A pathologist must manually perform the intricate and time-consuming task of counting mitoses by examining biopsy slices stained with Hematoxylin and Eosin (H&E) under a microscope. Mitotic cells can be challenging to distinguish in H&E-stained sections due to limited available datasets and similarities among mitotic and non-mitotic cells. Computer-assisted mitosis detection approaches have simplified the whole procedure by selecting, detecting, and labeling mitotic cells. Traditional detection strategies rely on image processing techniques that apply custom criteria to distinguish between different aspects of an image. Additionally, the automatic feature extraction from histopathology images that exhibit mitosis using neural networks.Additionally, the possibility of automatically extracting features from histopathological images using deep neural networks was investigated. This study examines mitosis detection as an object detection problem using multiple neural networks. From a medical standpoint, mitosis at the tissue level was also investigated utilising pre-trained Faster R-CNN and raw image data. Experiments were done on the MITOS-ATYPIA- 14 dataset and TUPAC16 dataset, and the results were compared to those of other methods described in the literature.

Keeping Pathologists in the Loop and an Adaptive F1-Score Threshold Method for Mitosis Detection in Canine Perivascular Wall Tumours

Performing a mitosis count (MC) is the diagnostic task of histologically grading canine Soft Tissue Sarcoma (cSTS). However, mitosis count is subject to inter- and intra-observer variability. Deep learning models can offer a standardisation in the process of MC used to histologically grade canine Soft Tissue Sarcomas. Subsequently, the focus of this study was mitosis detection in canine Perivascular Wall Tumours (cPWTs). Generating mitosis annotations is a long and arduous process open to inter-observer variability. Therefore, by keeping pathologists in the loop, a two-step annotation process was performed where a pre-trained Faster R-CNN model was trained on initial annotations provided by veterinary pathologists. The pathologists reviewed the output false positive mitosis candidates and determined whether these were overlooked candidates, thus updating the dataset. Faster R-CNN was then trained on this updated dataset. An optimal decision threshold was applied to maximise the F1-score predetermined using the validation set and produced our best F1-score of 0.75, which is competitive with the state of the art in the canine mitosis domain.

Enhancing mitosis quantification and detection in meningiomas with computational digital pathology

Mitosis is a critical criterion for meningioma grading. However, pathologists’ assessment of mitoses is subject to significant inter-observer variation due to challenges in locating mitosis hotspots and accurately detecting mitotic figures. To address this issue, we leverage digital pathology and propose a computational strategy to enhance pathologists’ mitosis assessment. The strategy has two components: (1) A depth-first search algorithm that quantifies the mathematically maximum mitotic count in 10 consecutive high-power fields, which can enhance the preciseness, especially in cases with borderline mitotic count. (2) Implementing a collaborative sphere to group a set of pathologists to detect mitoses under each high-power field, which can mitigate subjective random errors in mitosis detection originating from individual detection errors. By depth-first search algorithm (1) , we analyzed 19 meningioma slides and discovered that the proposed algorithm upgraded two borderline cases verified at consensus conferences. This improvement is attributed to the algorithm’s ability to quantify the mitotic count more comprehensively compared to other conventional methods of counting mitoses. In implementing a collaborative sphere (2) , we evaluated the correctness of mitosis detection from grouped pathologists and/or pathology residents, where each member of the group annotated a set of 48 high-power field images for mitotic figures independently. We report that groups with sizes of three can achieve an average precision of 0.897 and sensitivity of 0.699 in mitosis detection, which is higher than an average pathologist in this study (precision: 0.750, sensitivity: 0.667). The proposed computational strategy can be integrated with artificial intelligence workflow, which envisions the future of achieving a rapid and robust mitosis assessment by interactive assisting algorithms that can ultimately benefit patient management.

Uninformed Teacher-Student for hard-samples distillation in weakly supervised mitosis localization

Background and Objective:Mitotic activity is a crucial biomarker for diagnosing and predicting outcomes for different types of cancers, particularly breast cancer. However, manual mitosis counting is challenging and time-consuming for pathologists, with moderate reproducibility due to biopsy slide size, low mitotic cell density, and pattern heterogeneity. In recent years, deep learning methods based on convolutional neural networks (CNNs) have been proposed to address these limitations. Nonetheless, these methods have been hampered by the available data labels, which usually consist only of the centroids of mitosis, and by the incoming noise from annotated hard negatives. As a result, complex algorithms with multiple stages are often required to refine the labels at the pixel level and reduce the number of false positives. Methods:This article presents a novel weakly supervised approach for mitosis detection that utilizes only image-level labels on histological hematoxylin and eosin (H&E) images, avoiding the need for complex labeling scenarios. Also, an Uninformed Teacher-Student (UTS) pipeline is introduced to detect and distill hard samples by comparing weakly supervised localizations and the annotated centroids, using strong augmentations to enhance uncertainty. Additionally, an automatic proliferation score is proposed that mimicks the pathologist-annotated mitotic activity index (MAI). The proposed approach is evaluated on three publicly available datasets for mitosis detection on breast histology samples, and two datasets for mitotic activity counting in whole-slide images. Results:The proposed framework achieves competitive performance with relevant prior literature in all the datasets used for evaluation without explicitly using the mitosis location information during training. This approach challenges previous methods that rely on strong mitosis location information and multiple stages to refine false positives. Furthermore, the proposed pipeline for hard-sample distillation demonstrates promising dataset-specific improvements. Concretely, when the annotation has not been thoroughly refined by multiple pathologists, the UTS model offers improvements of up to ∼4% in mitosis localization, thanks to the detection and distillation of uncertain cases. Concerning the mitosis counting task, the proposed automatic proliferation score shows a moderate positive correlation with the MAI annotated by pathologists at the biopsy level on two external datasets. Conclusions:The proposed Uninformed Teacher-Student pipeline leverages strong augmentations to distill uncertain samples and measure dissimilarities between predicted and annotated mitosis. Results demonstrate the feasibility of the weakly supervised approach and highlight its potential as an objective evaluation tool for tumor proliferation.

Enhancing cancer detection and prevention mechanisms using advanced machine learning approaches

Cancer is a major global health challenge, emphasizing the critical need for early detection to enhance patient outcomes. This study thoroughly investigates the applications of advanced machine learning methods for cancer detection and prevention, aiming to develop robust algorithms that can accurately identify cancerous cells and assess cancer severity based on key parameters. The authors synthesize insights from previous cancer detection and prevention research through an in-depth literature review. The study sets specific objectives, including creating and evaluating innovative algorithms for classifying cancer cells.The authors employed machine learning techniques to analyze parameters, such as cell size, shape, nucleus characteristics, and additional factors like cell texture, mitosis count, tumour progression, metastasis, gene expression patterns, and biological markers. This methodology is distinguished by its effective use of diverse data types and automated feature extraction to improve cancer detection and prediction accuracy. Advanced machine learning methods enhance the precision and reliability of current cancer cell classification algorithms. The research underscores the importance of timely and accurate cancer detection, which enables early intervention and significantly improves patient survival rates. The results and discussion section meticulously analyzes the findings, demonstrating the approach's effectiveness in accurately identifying cancerous cells and assessing cancer severity. This study is a valuable resource for medical professionals, supporting early-stage cancer detection and classification of cancer cells at different stages. The proposed algorithms show promising results in improving the accuracy and efficiency of cancer detection systems, paving the way for future advancements in cancer research and offering essential insights for healthcare practitioners and researchers.

Prognostic importance of mitosis quantification and PHH3 expression in oral epithelial dysplasia.

Oral epithelial dysplasia (OED) is diagnosed and graded using a range of histological features, making grading subjective and challenging. Mitotic counting and phosphohistone-H3 (PHH3) staining have been used for the prognostication of various malignancies; however, their importance in OED remains unexplored. This study conducts a quantitative analysis of mitotic activity in OED using both haematoxylin and eosin (H&E)-stained slides and immunohistochemical (IHC) staining for PHH3. Specifically, the diagnostic and prognostic importance of mitotic number, mitotic type and intra-epithelial location is evaluated. Whole slide images (WSI) of OED (n = 60) and non-dysplastic tissue (n = 8) were prepared for analysis. Five-year follow-up data was collected. The total number of mitosis (TNOM), mitosis type and intra-epithelial location was manually evaluated on H&E images and a digital mitotic count performed on PHH3-stained WSI. Statistical associations between these features and OED grade, malignant transformation and OED recurrence were determined. Mitosis count increased with grade severity (H&E: p < 0.005; IHC: p < 0.05), and grade-based differences were seen for mitosis type and location (p < 0.05). The ratio of normal-to-abnormal mitoses was higher in OED (1.61) than control (1.25) and reduced with grade severity. TNOM, type and location were better predictors when combined with histological grading, with the most prognostic models demonstrating an AUROC of 0.81 for transformation and 0.78 for recurrence, exceeding conventional grading. Mitosis quantification and PHH3 staining can be an adjunct to conventional H&E assessment and grading for the prediction of OED prognosis. Validation on larger multicentre cohorts is needed to establish these findings.

Clinicopathological Features of Patients with Malignant Melanoma Diagnosis and Prognostic and Predictive Importance of Neuthrophil-Lymphocyte Ratio

In this study, the effect of demographic, laboratory and clinicopathological parameters along with neutrophil-lymphocyte ratio (NLR) on prognosis and survival and correlation with other parameters was researched in patients with malignant melanoma (MM) diagnosis. In our study, 107 patients monitored for MM diagnosis in Eskişehir Osmangazi University Medical Oncology clinic from 2010-2017 were retrospectively assessed. Age, gender, LDH level, pathological parameters, BRAF mutation status, neutrophil-lymphocyte ratio (NLR) and the effects of these parameters on overall survival (OS) and disease-free survival (DFS) and correlations with each other were researched. At time of diagnosis, 86% of patients were in the early stage. The dominant types identified were cutaneous MM and nodular MM. Median NLR cut-off value was identified as 1.97. All non-cutaneous MM cases were BRAF negative (p&amp;lt;0.0001). High NLR was associated with advanced stage (p=0.001), advanced age (p=0.008), ulceration presence (p=0.011), and high mitosis count (p=0.05). High NLR (p&amp;lt;0.0001), high LDH level (p=0.04), increased Breslow thickness (p=0.01), increased Clark level (p=0.01), high mitosis count (p=0.02), and lymph node (LN) involvement (p=0.04) were correlated with significantly shorter OS durations. Cox multivariate regression analysis identified the most effective independent parameters on OS were LN involvement (HR: 3.4, p=0.01) and high NLR (HR: 4.6, p=0.04). Nodal involvement was also identified as the most predictive independent parameter for recurrence (HR: 3.2, p=0.03). In addition to classic parameters, NLR appears to be a biomarker which can predict prognosis. Patients with nodal involvement and high NLR values should be monitored more closely in clinics. Data require support with broad-scale studies.

A retrospective study on expression and clinical significance of PHH3, Ki67 and P53 in bladder exophytic papillary urothelial neoplasms.

Exophytic papillary urothelial neoplasms (EPUN) are difficult to diagnose pathologically and are well-known for their heterogeneous prognoses. Thus, searching for an objective and accurate diagnostic marker is of great clinical value in improving the outcomes of EPUN patients. PHH3 was reported to be expressed explicitly in the mitotic phase of the cell cycle, and recent studies have shown that PHH3 expression was associated with the differential diagnosis and prognosis of many tumors. However, its significance in EPUN remains unclear. This study aimed to determine the expression of PHH3 in different EPUN, compare its expression with cell-cycle related proteins Ki67 and P53, and analyze its significance in the differential diagnosis and prognostic value for high-grade papillary urothelial carcinoma (HGPUC), low-grade papillary urothelial carcinoma (LGPUC), papillary urothelial neoplasm of low malignant potential (PUNLMP) and urothelial papilloma (UP). We retrospectively analyzed the pathological diagnosis and clinical features of 26 HGPUC cases, 43 LGPUC cases, 21 PUNLMP cases and 11 UP cases. PHH3, Ki67 and P53 were detected by immunohistochemistry in 101 EPUN cases samples. The cut-off values of PHH3 mitosis count (PHMC), HE mitosis count (HEMC), Ki67 and P53 in the different EPUN were determined using the ROC curve. The distribution of counts in each group and its relationship with clinical parameters and prognosis of EPUN patients were also analyzed. The determination coefficient (R2=0.9980) of PHMC were more potent than those of HEMC (R2=0.9734) in the EPUN mitotic counts microscopically by both pathologists. Of the 101 EPUN cases investigated, significant positive linear correlations were found between PHMC and HEMC, PHMC and Ki67, and HEMC and Ki67 (P<0.0001). In HGPUC, LGPUC, PUNLMP and UP, a decreasing trend was observed in the median and range of PHMC/10HPFs, HEMC/10HPFs, Ki67 (%) and P53 (%). PHMC, HEMC, Ki67 and P53 were associated with different clinical parameters of EPUN. PHMC, HEMC, Ki67 and P53 were found to exhibit substantial diagnostic values among different EPUN and tumor recurrence. Based on the ROC curve, when PHMC was >48.5/10HPFs, a diagnosis of HGPUC was more likely, and when PHMC was >13.5/10HPFs, LGPUC was more likely. In addition, when PHMC was >5.5/10HPFs, the possibility of non-infiltrating LGPUC was greater. Kaplan-Meier survival curve analysis showed that the median recurrence-free survival (RFS) for cases with PHMC > 13.5/10HPFs and HEMC > 14.5/10HPFs were 52.5 and 48 months, respectively, and their respective hazard ratio was significantly higher (Log-rank P<0.05). PHH3 exhibited high specificity and sensitivity in diagnosing EPUN. Combined with HEMC, Ki67 and P53, it can assist in the differential diagnosis of EPUN and estimate its clinical progression with high predictive value to a certain extent.

Abstract 5423: AI powered quantification of mitotic rate in H&amp;E stained tissue detects significant differences between treatment groups of preclinical pancreas cancer xenografts

Abstract Background: Mitotic rate is a readout routinely used for characterization of tumor samples. Standard methods to quantify cell division include manual pathologist counts based on hematoxylin and eosin (H&amp;E) staining and either manual or automated assessment of immunohistochemistry (IHC) staining using phospho-histone H3 (pHH3). These suffer the drawbacks of high inter-pathologist variability in the case of H&amp;E assessment and time inefficiency and false positive calls in case of pHH3 staining. Aiming to overcome these issues, we developed a mitosis detection model based on H&amp;E-stained tissue sections alone. Methods: To develop and evaluate the model, we used 1032 H&amp;E-stained tissue sections (156 used for model training) of pre-clinical pancreas cancer xenografts originating from mice that have undergone a series of experiments conducted to examine the pharmacodynamic effect of several anti-cancer protocols. We trained (i) a tissue segmentation model (segmenting tissue regions into ‘carcinoma’, ‘stroma’, ‘necrosis’, and ‘other’) and (ii) a segmentation model for pixel-level mitosis detection (segmenting ‘mitosis’ vs. ‘non-mitosis’). Regions predicted as mitosis were post-processed to represent individual dividing cells. The tissue segmentation model served as a filter to predict mitotic rate for carcinoma tissue areas only, which has not been accounted for in previous AI-based methods for mitotic rate prediction on H&amp;E tissue. To evaluate the model on detecting mitotic events, the model was compared against a 5-pathologist consensus of mitotic count. The mitotic rates predicted by the model were used to infer differences between treatment groups (various treatments and dosages vs. control). Results: The mitosis detection model for quantifying rates of cell division in carcinoma regions of H&amp;E-stained tissue sections showed a notable agreement with the 5-pathologist mitotic count consensus (Pearson correlation 0.92). Furthermore, the model correlated well with mitosis counts based on pHH3 IHC staining (Pearson correlation 0.78), which were available for 63 cases. Finally, when used to characterize the entire CDX cohort, the case level mitotic rate predicted by the model showed significant differences between treatment groups in line with or better than using a pHH3 IHC stain. Conclusion: This study demonstrates the efficacy and scalability of AI-based models for the quantification of mitotic rate based on H&amp;E-stained tissue alone, presenting a time- and cost-efficient approach that also mitigates inter-annotator variability. Citation Format: Sharon Ruane, Lukas Ruff, Brian Reichholf, Christina Aigner, Emil Barbuta, Stephan Tietz, Olivér Atanaszov, Rosemarie Krupar, Simon Schallenberg, Maximilian Alber, Francesca Trapani, Frederick Klauschen. AI powered quantification of mitotic rate in H&amp;E stained tissue detects significant differences between treatment groups of preclinical pancreas cancer xenografts. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5423.

Multi CNN based automatic detection of mitotic nuclei in breast histopathological images

In breast cancer diagnosis, the number of mitotic cells in a specific area is an important measure. It indicates how far the tumour has spread, which has consequences for forecasting the aggressiveness of cancer. Mitosis counting is a time-consuming and challenging technique that a pathologist does manually by examining Hematoxylin and Eosin (H&E) stained biopsy slices under a microscope. Due to limited datasets and the resemblance between mitotic and non-mitotic cells, detecting mitosis in H&E stained slices is difficult. By assisting in the screening, identifying, and labelling of mitotic cells, computer-aided mitosis detection technologies make the entire procedure much easier. For computer-aided detection approaches of smaller datasets, pre-trained convolutional neural networks are extensively employed. The usefulness of a multi CNN framework with three pre-trained CNNs is investigated in this research for mitosis detection. Features were collected from histopathology data and identified using VGG16, ResNet50, and DenseNet201 pre-trained networks. The proposed framework utilises all training folders of the MITOS dataset provided for the MITOS-ATYPIA contest 2014 and all the 73 folders of the TUPAC16 dataset. Each pre-trained Convolutional Neural Network model, such as VGG16, ResNet50 and DenseNet201, provides an accuracy of 83.22%, 73.67%, and 81.75%, respectively. Different combinations of these pre-trained CNNs constitute a multi CNN framework. Performance measures of multi CNN consisting of 3 pre-trained CNNs with Linear SVM give 93.81% precision and 92.41% F1-score compared to multi CNN combinations with other classifiers such as Adaboost and Random Forest.

Abstract P2-11-16: Computerized Measurements of Nuclear Morphology Features, Mitosis Rate, and Tubule Formation from H&amp;E Images Predicts Disease-Free Survival in Patients with HR+ &amp; LN+ Invasive Breast Cancer from SWOG S8814

Abstract Background: Lymph node (LN) involvement is a strong indicator of poor prognosis for breast cancer (BC), with adjuvant chemotherapy remaining fundamental to management of these patients. SWOG S8814 was a Phase III randomized trial of postmenopausal patients with pathologic LN-positive BC who were hormone receptor positive (HR+). The objectives of the clinical trial were to compare disease free survival (DFS) and overall survival (OS) of 1) these postoperative patients treated with a combination of cyclophosphamide, doxorubicin, fluorouracil (CAF) plus tamoxifen versus tamoxifen alone; and 2) patients treated with CAF followed by tamoxifen versus CAF plus concurrent tamoxifen. In this study we sought to evaluate the potential of applying computational image analysis on whole slide images (WSI) for predicting DFS and OS in SWOG S8814. Methods: A cohort of 135 patients (N=53 DFS event) diagnosed with HR+ &amp; LN+ BC from clinical trial ECOG 2197 was utilized as training set D1. Validation set D2 comprised 630 patients (N=260 DFS event, N=195 death) with HR+&amp; LN+ BC from SWOG S8814. Three deep learning models were employed to respectively detect nuclei, mitosis, and tubules in WSIs. Subsequently, a total of 1,810 features relating to nuclear morphology (e.g., spatial distribution, shape, texture, orientation), mitotic activity (e.g., mitosis hotspot, mitotic rates) and tubule formation (e.g., tubular nuclei distribution, ratio of tubule to non-tubule area) were extracted from each WSI. A lasso regularized Cox regression model (IbRiS) was trained on D1 to respectively identify four features from each of the feature categories (nuclei morphology, mitotic activity, and tubule formation) most strongly associated with DFS, a continuous risk score based on the selected features was then constructed. An optimal risk threshold was identified on D1 to dichotomize the risk scores into high vs. low risk of recurrence categories. Blinded validation of the machine learning model on SWOG S8814 using Cox regression was performed by SWOG to evaluate its performance in terms of DFS and OS. Results: In D2, patients identified as high risk of recurrence by IbRiS had a significantly worse prognosis in terms of DFS with hazard ratio=1.30 (p=0.039, 95% CI=1.01-1.66). IbRiS was also found to be significantly prognostic of OS with hazard ratio=1.38 (p=0.026, 95% CI=1.04-1.83). IbRiS was however, neither prognostic of DFS (HR = 1.20; 95% CI 0.93-1.54) nor OS (HR = 1.28; 95% CI 0.96-1.71) in multivariable analysis adjusting for treatment, tumor size, and number of positive nodes. IbRiS was also not a significant predictor of chemotherapy benefit (DFS p=0.45; OS p=0.25). Conclusion: We developed a prognostic model (IbRiS) based on the combined features of nuclear morphology, mitosis count, and tubule formation that can help further risk stratify HR+ &amp; LN+ BC patients by only using H&amp;E slides. Citation Format: Yuli Chen, William E. Barlow, Haojia Li, Cheng Lu, Andrew Janowczyk, German Corredor, Shridar Ganesan, Michael Feldman, Pingfu Fu, Hannah Gilmore, Kathy S. Albain, Lajos Pusztai, James Rae, Daniel Hayes, Andrew K. Godwin, Alastair M. Thompson, Anant Madabhushi. Computerized Measurements of Nuclear Morphology Features, Mitosis Rate, and Tubule Formation from H&amp;E Images Predicts Disease-Free Survival in Patients with HR+ &amp; LN+ Invasive Breast Cancer from SWOG S8814 [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P2-11-16.

Abstract P2-11-34: Mitotic spindle hotspot counting using deep learning networks is highly associated with clinical outcomes in patients with early-stage triple-negative breast cancer who did not receive systemic therapy

Abstract Background: Triple-negative breast cancers (TNBC) exhibit high rates of recurrence and mortality. However, recent studies suggest that a subset of patients (pts) with early-stage TNBC enriched in tumor-infiltrating lymphocytes (TILs) have excellent clinical outcomes even in the absence of systemic therapy. Additional histological biomarkers that could identify pts for future systemic therapy escalation/de-escalation strategies are of great interest. TNBC are frequently highly proliferative with abundant mitoses. However, classic markers of proliferation (manual mitosis counting and Ki-67) appear to offer no prognostic value. Here, we evaluated the prognostic effects of automated mitotic spindle hotspot (AMSH) counting on RFS in independent cohorts of systemically untreated early-stage TNBC. Methods: AMSH counting was conducted with a state-of-the-art deep learning algorithm trained on the detection of mitoses within 2 mm2 areas with the highest mitotic density (i.e. hotspots) in digital H&amp;E images. Details of the development, training and validation of the algorithm were published previously [1] in a cohort of unselected TNBC. We obtained AMSH counts in a centrally confirmed TNBC cohort from Mayo Clinic [2] and focused our analysis on pts who received locoregional therapy but no systemic therapy. Using a fractional polynomial analysis with a multivariable proportional hazards regression model, we confirmed the assumption of linearity in the log hazard for the continuous variable AMSH and evaluated whether AMSH counts were prognostic of RFS. We corroborated our findings in an independent cohort of systemically untreated TNBC pts from the Radboud University Medical Center in the Netherlands (Radboud Cohort). Results are reported at a median follow-up of 8.1 and 6.7 years for the Mayo and Netherlands cohorts, respectively. Results: Among 182 pts with who did not receive systemic therapy in the Mayo Cohort, 140 (77%) with available AMSH counts were included. The mean age was 61 (range: 31-94), 71% were postmenopausal, 67% had tumors ≤ 2cm, and 83% were node-negative. As expected, most tumors were Nottingham grade 3 (84%) and had a high Ki-67 proliferation index (54% with Ki-67 &amp;gt;30%). Most tumors (73%) had stromal TILs ≤ 30%. The median AMSH count was 18 (IQR: 8, 42). AMSH counts were linearly associated with grade and tumor size, with the proportion of pts with grade 3 tumors and size &amp;gt; 2 cm increasing as the AMSH counts increased (p=0.007 and p=0.059, respectively). In a multivariate model controlling for nodal status, tumor size, and stromal TILs, AMSH counts were independently associated with RFS (p&amp;lt; 0.0001). For every 10-point increase in the AMSH count, we observed a 17% increase in the risk of experiencing an RFS event (HR 1.17, 95% CI 1.08-1.26). We corroborated our findings in the Radboud Cohort (n=126). The mean age was 68 (range: 40-96), and 81% were node-negative. While the median AMSH count was 36 (IQR: 16-63), higher than in the Mayo Cohort (p=0.004), the prognostic impact was similar, with a significant association between AMSH count and RFS (p=0.028) in a multivariate model corrected for nodal status, tumor size, and stromal TILs. For every 10-point increase in the AMSH count in the Netherlands cohort, we observed a 9% increase in the risk of experiencing an RFS event (HR 1.09, 95% CI 1.01-1.17). RFS rates according to AMSH counts for both cohorts are shown in the Table. Conclusions: AMSH counting is a new proliferation biomarker that provides prognostic value independent of nodal status, tumor size, and stromal TILs in systemically untreated early-stage TNBC. Plans are underway to evaluate AMSH counts in additional cohorts of systemically untreated TNBC, and in other disease settings such as prior to neoadjuvant systemic therapy. If validated, this biomarker should be prospectively evaluated as a potential selection biomarker in clinical trials of systemic therapy de-escalation.

Counting mitoses in gastrointestinal stromal tumours (GISTs): variable practices in the real-world setting and their clinical implications

Mitotic count (MC) is an important prognostic indicator in gastrointestinal stromal tumours (GISTs). Though MC evaluation was initially proposed in 50 HPFs, recent international guidelines recommend that MC be performed on 5 mm2 because HPFs may have different areas depending on the ocular field number (FN) of the utilized light microscope. Performing MC on different areas leads to a non-standardized evaluation and erroneous risk stratification. The aim of the study was to audit real-life MC practices with special emphasis on possible risk stratification errors. A survey was administered to Italian pathologists to evaluate the following: method used for MC (5 mm2 versus 50 HPF); FN of the light microscope; prognostic scheme for risk stratification. Based on the results of the survey, 100 GISTs (25/risk class using Miettinen prognostic scheme) were retrieved and MC performed using 5 mm2 versus the corresponding mm2 area sizes of 50 HPFs with variable FNs (18, 20, 22). The survey demonstrated that the majority of pathologists (64.5%) use 50 HPFs with various FNs leading to excessive area size. The most frequently used prognostic scheme is that by Miettinen. Using this prognostic scheme and counting mitoses in 5 mm2 versus 50 HPFs with FNs 18, 20 and 22, a change in risk class was identified ranging from 10 to 41%, depending on FN. In conclusion, this study demonstrates that MC is still frequently performed on 50 HPF, with area sizes exceeding the specified 5 mm2 by far.

PSlugS158 Immunohistochemistry is Useful in Grading Meningioma.

Snail family transcriptional repressor 2 (SNAI2, Slug) is a transcription factor that belong to the Slug/Snail superfamily. Site specific phosphorylation of slug (pSlugS158) is detected during the M phase, and thus, this phosphorylated protein is considered a novel marker for detecting mitotic figures. Herein, we investigated whether the detection of mitosis using pSlugS158 expression can be used in the histological grading of meningioma. We performed immunohistochemistry for pSlugS158 and PHH3 in tissue samples of 61 patients with meningioma and examined the association between mitotic counts using pSlugS158 and recurrence-free survival (RFS). The nuclear expression of pSlugS158 was observed in the cell with mitotic figures. Tumor grading based on pSlugS158 was significantly associated with the RFS (p < .001). It can be concluded that pSlugS158 is a useful and practical marker to detect mitosis and seems to be reliable for the counting of mitoses in histological grading of meningioma.

A Novel Deep Learning-Based Mitosis Recognition Approach and Dataset for Uterine Leiomyosarcoma Histopathology.

Simple SummaryIn this paper, we present an artificial Intelligence (AI) based automatic detection of mitoses in Uterine Leiomyosarcoma. Mitotic count is one of the important biomarkers in the field of histopathology. A dataset is also made available to research community which consists of images having moitotically active region. These regions are labeled by a trained AI expert in coordination with a senior histopathologist. Preliminary results show AI as promising solution for detection of mitotically active regions mitotic region in Uterine leiomyosarcoma cases and can be used as a second opinion system.Uterine leiomyosarcoma (ULMS) is the most common sarcoma of the uterus, It is aggressive and has poor prognosis. Its diagnosis is sometimes challenging owing to its resemblance by benign smooth muscle neoplasms of the uterus. Pathologists diagnose and grade leiomyosarcoma based on three standard criteria (i.e., mitosis count, necrosis, and nuclear atypia). Among these, mitosis count is the most important and challenging biomarker. In general, pathologists use the traditional manual counting method for the detection and counting of mitosis. This procedure is very time-consuming, tedious, and subjective. To overcome these challenges, artificial intelligence (AI) based methods have been developed that automatically detect mitosis. In this paper, we propose a new ULMS dataset and an AI-based approach for mitosis detection. We collected our dataset from a local medical facility in collaboration with highly trained pathologists. Preprocessing and annotations are performed using standard procedures, and a deep learning-based method is applied to provide baseline accuracies. The experimental results showed 0.7462 precision, 0.8981 recall, and 0.8151 F1-score. For research and development, the code and dataset have been made publicly available.

The associations between serum high-density lipoprotein cholesterol levels and malignant behavior in pancreatic neuroendocrine neoplasms

BackgroundThe role of serum high-density lipoprotein cholesterol (HDL-c) in tumorigenesis are observed in several endocrine-related cancers. However, its role in pancreatic neuroendocrine neoplasms (PNENs) has not been understood. In the current study, the relationship between HDL-c levels and malignant behavior in PNENs was explored.MethodsOne hundred ninety-seven patients with histopathology confirmed PNENs were included. PNENs were divided into three grades (G1, G2 and G3) as 2017 WHO classification based on ki67 index and mitosis count. The demographic data, clinical information, tumor morphological and pathological features (organs invasion, lymph node metastasis, vascular invasion and perineural invasion), and serum tumor biomarkers were collected. The relationships between HDL-c levels and malignant behaviors in PNENs were analyzed using logistic regression analysis. Models were also developed for the identification of high grade PNENs.ResultsThe levels of serum HDL-c in G2/G3 tumor were significantly lower than that in G1 tumor (P = 0.031). However, no such difference was found between G3 and G1/G2. The proportions of low HDL-c (≤ 0.9 mmol/L) were higher in high-grade PNENs (G2/G3 or G3) than those in low-grade (G1 or G1/G2) (29.0 vs 15.2%, P = 0.032; 37.0 vs 20.5%, P = 0.023). The risk of G2/G3 tumors in patients with high serum HDL-c levels was decreased (odds ratio (OR) = 0.35, 95% confidence interval (CI): 0.12–0.99). Similarly, the risk of G3 PNENs increased in patients with low HDL-c levels (OR = 2.51, 95%CI:1.12–5.60). HDL-c level was also associated with a high ki67 index (> 55%) (OR = 0.10, 95%CI: 0.02–0.51) and neuroendocrine carcinoma G3 (OR = 0.21, 95%CI: 0.06–0.80). The area under the curve (AUC) of HDL-c + tumor size + age was 0.85 (95% CI: 0.79–0.91) in identifying G2/G3 PNENs, and HDL-c (> 0.9 mmol/L) + tumor size + age had an AUC of 0.77 (95% CI: 0.70–0.84) in identifying G3 PNENs. HDL-c level was associated with lymph node metastasis (OR = 0.24, 95%CI:0.08–0.99).ConclusionSerum HDL-c levels were significantly associated with malignant behaviors in PNENs, in particular to tumor grade and lymph node metastasis.

Pathologic findings of acantholytic squamous cell carcinoma coexisting with cutaneous cryptococcosis in a Houbara bustard (Chlamydotis undulata)

A 2-year-old Asian Houbara bustard was presented with a solitary well-defined, firm cutaneous mass on the hock region. Grossly, the mass protruded from the surface was located on the hairless and unpigmented areas of the right hock joint with ulceration and dried hemorrhagic foci. On microscopic examination, ulceration, hemorrhage, as well as hyperkeratosis were observed. Large round, oval to polygonal neoplastic cells extended into the dermis were arranged to form cords, trabeculae, islands or glandular-like structures without keratin pearls. These pseudoglandular structures were composed of pseudolumina containing acantholytic and detached tumor cells. Necrosis of the neoplastic cells was accompanied by infiltration of inflammatory cells particularly heterophils. Unlike pleomorphic tumor cells, mitosis count was almost frequent. No evidence of other abnormalities and tumor metastasis was found. These gross and microscopic features appeared to be suggestive of a rare histologic variant of squamous cell carcinoma (SCC), acantholytic SCC.

Defining the area of mitoses counting in invasive breast cancer using whole slide image

AbstractAlthough counting mitoses is part of breast cancer grading, concordance studies showed low agreement. Refining the criteria for mitotic counting can improve concordance, particularly when using whole slide images (WSIs). This study aims to refine the methodology for optimal mitoses counting on WSI. Digital images of 595 hematoxylin and eosin stained sections were evaluated. Several morphological criteria were investigated and applied to define mitotic hotspots. Reproducibility, representativeness, time, and association with outcome were the criteria used to evaluate the best area size for mitoses counting. Three approaches for scoring mitoses on WSIs (single and multiple annotated rectangles and multiple digital high-power (×40) screen fields (HPSFs)) were evaluated. The relative increase in tumor cell density was the most significant and easiest parameter for identifying hotspots. Counting mitoses in 3 mm2 area was the most representative regarding saturation and concordance levels. Counting in area <2 mm2 resulted in a significant reduction in mitotic count (P = 0.02), whereas counting in area ≥4 mm2 was time-consuming and did not add a significant rise in overall mitotic count (P = 0.08). Using multiple HPSF, following calibration, provided the most reliable, timesaving, and practical method for mitoses counting on WSI. This study provides evidence-based methodology for defining the area and methodology of visual mitoses counting using WSI. Visual mitoses scoring on WSI can be performed reliably by adjusting the number of monitor screens.