Histological Images Research Articles

Generating and Analyzing High-Parameter Histology Images with Histoflow Cytometry.

The usage of histology to investigate immune cell diversity in tissue sections such as those derived from the central nervous system (CNS) is critically limited by the number of fluorescent parameters that can be imaged at a single time. Most immune cell subsets have been defined using flow cytometry by using complex combinations of protein markers, often requiring four or more parameters to conclusively identify, which is beyond the capabilities of most conventional microscopes. As flow cytometry dissociates tissues and loses spatial information, there is a need for techniques that can retain spatial information while interrogating the roles of complex cell types. These issues are addressed here by creating a method for expanding the number of fluorescent parameters that can be imaged by collecting the signals of spectrally overlapping fluorophores and using spectral unmixing to separate the signals of each individual fluorophore. These images are then processed using an analysis pipeline to take high-parameter histology images and extract single cells from these images so that the unique fluorescent properties of each cell can be analyzed at a single-cell level. Using flow cytometry-like gating strategies, cells can then be profiled into subsets and mapped back onto the histology sections to not only quantify their abundance, but also establish how they interact with the tissue environment. Overall, the simplicity and potential of using histoflow cytometry to study complex immune populations in histology sections is demonstrated.

Overview of the Multispecies Ovary Tissue Histology Electronic Repository†.

The Multispecies Ovary Tissue Histology Electronic Repository (MOTHER) is a publicly accessible repository of ovary histology images. MOTHER includes hundreds of images from nonhuman primates, as well as ovary histology images from an expanding range of other species. Along with an image, MOTHER provides metadata about the image, and for selected species, follicle identification annotations. Ongoing work includes assisting scientists with contributing their histology images, creation of manual and automated (via machine learning) processing pipelines to identify and count ovarian follicles in different stages of development, and the incorporation of that data into the MOTHER database (MOTHER-DB). MOTHER will be a critical data repository storing and disseminating high-value histology images that are essential for research into ovarian function, fertility, and intra-species variability.

Effects of Sapindus mukorossi Seed Oil on Bone Healing Efficiency: An Animal Study.

Natural products have attracted great interest in the development of tissue engineering. Recent studies have demonstrated that unsaturated fatty acids found in natural plant seed oil may exhibit positive osteogenic effects; however, few in vivo studies have focused on the use of plant seed oil for bone regeneration. The aim of this study is to investigate the effects of seed oil found in Sapindus mukorossi (S. mukorossi) on the osteogenic differentiation of mesenchymal stem cells and bone growth in artificial bone defects in vivo. In this study, Wharton-jelly-derived mesenchymal stem cells (WJMSCs) were co-cultured with S. mukorossi seed oil. Cellular osteogenic capacity was assessed using Alizarin Red S staining. Real-time PCR was carried out to evaluate ALP and OCN gene expression. The potential of S. mukorossi seed oil to enhance bone growth was assessed using an animal model. Four 6 mm circular defects were prepared at the parietal bone of New Zealand white rabbits. The defects were filled with hydrogel and hydrogel-S. mukorossi seed oil, respectively. Quantitative analysis of micro-computed tomography (Micro-CT) and histological images was conducted to compare differences in osteogenesis between oil-treated and untreated samples. Although our results showed no significant differences in viability between WJMSCs treated with and without S. mukorossi seed oil, under osteogenic conditions, S. mukorossi seed oil facilitated an increase in mineralized nodule secretion and upregulated the expression of ALP and OCN genes in the cells (p < 0.05). In the animal study, both micro-CT and histological evaluations revealed that new bone formation in artificial bone defects treated with S. mukorossi seed oil were nearly doubled compared to control defects (p < 0.05) after 4 weeks of healing. Based on these findings, it is reasonable to suggest that S. mukorossi seed oil holds promise as a potential candidate for enhancing bone healing efficiency in bone tissue engineering.

PATH-13. LEARNED RESIZING WITH EFFICIENT TRAINING (LRET) FACILITATES IMPROVED PERFORMANCE OF LARGE-SCALE BRAIN TUMOR HISTOLOGY IMAGE CLASSIFICATION MODELS

Abstract BACKGROUND Histologic examination is vital in oncology research and diagnostics. The adoption of digital scanning of whole slide images (WSI) has created an opportunity to leverage deep learning-based image classification methods to enhance diagnosis and risk stratification. However, technical limitations prevent training and deployment of accurate comprehensive multiclass deep convolutional neural networks (DCNN) models for histopathology image classification. The input dimensions of DCNN architectures are small compared to the typical pathologist field of view, degrading performance by excluding important architectural features. Furthermore, data requirements for comprehensive models are sufficiently large to overwhelm the system memory during training. METHODS A method termed Learned Resizing with Efficient Training (LRET) was developed to address the main limitations of traditional histopathology classification model training. The LRET method couples efficient training techniques with image resizing to facilitate seamless integration of larger histology image patches into state-of-the-art classification models while preserving important structural information. The LRET method was coupled with two distinct resizing techniques to train three diverse histology image datasets using five different DCNN architectures. Performance metrics were compared on cross validation and hold out test sets. RESULTS LRET-trained models were flexible to multiple input patch dimensions and DCNN models. We demonstrated performance improvement across all datasets while significantly reducing the training time and resources over traditional methods. Using a large-scale, multiclass brain tumor classification dataset consisting of 74 distinct histopathologic classes, LRET-trained models outperformed existing methods by 15-28% in accuracy, yielding 94% accuracy for the best model. CONCLUSION The LRET method for DCNN training significantly enhances the performance of large-scale multiclass histopathology image classification. The implications of this work extend to broader applications within medical imaging and beyond, where efficient integration of high-resolution images into deep learning pipelines is paramount for driving advancements research and clinical practice.

Explainable Histopathology Image Classification with Self-organizing Maps: A Granular Computing Perspective

The automatic analysis of histology images is an open research field where machine learning techniques and neural networks, especially deep architectures, are considered successful tools due to their abilities in image classification. This paper proposes a granular computing methodology for histopathological image classification. It is based on embedding tiles of histopathology images using deep metric learning, where a self-organizing map is adopted to generate the granular structure in this learned embedding space. The SOM enables the implementation of an explainable mechanism by visualizing a knowledge space that the experts can use to analyze and classify the new images. Additionally, it provides confidence in the classification results while highlighting each important image fragment, with the benefit of reducing the number of false negatives. An exemplary case is when an image detail is indicated, with small confidence, as malignant in an image globally classified as benign. Another implemented feature is the proposal of additional labelled image tiles sharing the same characteristics to specify the context of the output decision. The proposed system was tested using three histopathology image datasets, obtaining the accuracy of the state-of-the-art black-box methods based on deep learning neural networks. Differently from the methodologies proposed so far for the same purpose, this paper introduces a novel explainable method for medical image analysis where the advantages of the deep learning neural networks used to build the embedding space for the image tiles are combined with the intrinsic explainability of the granular process obtained using the clustering property of a self-organizing map.

Inferring single-cell spatial gene expression with tissue morphology via explainable deep learning.

The spatial arrangement of cells is vital in developmental processes and organogenesis in multicellular life forms. Deep learning models trained with spatial omics data uncover complex patterns and relationships among cells, genes, and proteins in a high-dimensional space, providing new insights into biological processes and diseases. State-of-the-art in silico spatial multi-cell gene expression methods using histological images of tissue stained with hematoxylin and eosin (H&E) to characterize cellular heterogeneity. These computational techniques offer the advantage of analyzing vast amounts of spatial data in a scalable and automated manner, thereby accelerating scientific discovery and enabling more precise medical diagnostics and treatments. In this work, we developed a vision transformer (ViT) framework to map histological signatures to spatial single-cell transcriptomic signatures, named SPiRiT ( S patial Omics P rediction and R eproducibility integrated T ransformer). Our framework was enhanced by integrating cross validation with model interpretation during hyper-parameter tuning. SPiRiT predicts single-cell spatial gene expression using the matched histopathological image tiles of human breast cancer and whole mouse pup, evaluated by Xenium (10x Genomics) datasets. Furthermore, ViT model interpretation reveals the high-resolution, high attention area (HAR) that the ViT model uses to predict the gene expression, including marker genes for invasive cancer cells ( FASN ), stromal cells ( POSTN ), and lymphocytes ( IL7R ). In an apple-to-apple comparison with the ST-Net Convolutional Neural Network algorithm, SPiRiT improved predictive accuracy by 40% using human breast cancer Visium (10x Genomics) dataset. Cancer biomarker gene prediction and expression level are highly consistent with the tumor region annotation. In summary, our work highlights the feasibility to infer spatial single-cell gene expression using tissue morphology in multiple-species, i.e., human and mouse, and multi-organs, i.e., mouse whole body morphology. Importantly, incorporating model interpretation and vision transformer is expected to serve as a general-purpose framework for spatial transcriptomics.

Development of dissolving microneedles loaded with fucoidan for enhanced anti-aging activity: An in vivo study in mice animal model

Skin aging occurs naturally as essential skin components gradually decline, leading to issues such as fine lines, wrinkles, and pigmentation. Fucoidan, a natural bioactive compound, holds potential for addressing these age-related concerns. However, its hydrophilic nature and substantial molecular weight hinder its absorption into the skin. In this study, we utilized polyvinyl pyrrolidone K30 (PVP) and polyvinyl alcohol (PVA) as polymers to fabricate dissolving microneedles loaded with fucoidan (DMN-F). The DMN-F formulations were examined for physical characteristics, stability, permeability, toxicity, and efficacy in animal models. These formulations exhibited consistent polymer blends with a conical structure and uniform cone-shaped design. Microneedle structure and penetration capability gradually decreased with increasing fucoidan concentration, with storage recommended at approximately 33 % relative humidity (RH). Ex vivo studies showed that DMN-F efficiently delivered up to 95.03 ± 2.36 % of the total fucoidan concentration into the skin. In vivo investigations revealed that DMN-F effectively reduced wrinkles, improved skin elasticity, maintained moisture levels, and increased epidermal thickness. Histological images provided additional evidence of DMN-F’s positive effects on these aging parameters. The results confirm that the DMN-F formulation effectively delivers fucoidan into the skin, allowing it to treat and mitigate signs of aging.

A deep learning-based compression and classification technique for whole slide histopathology images

This paper presents an autoencoder-based neural network architecture to compress histopathological images while retaining the denser and more meaningful representation of the original images. Current research into improving compression algorithms is focused on methods allowing lower compression rates for Regions of Interest (ROI-based approaches). Neural networks are great at extracting meaningful semantic representations from images and, therefore can select the regions to be considered of interest for the compression process. In this work, we focus on the compression of whole slide histopathology images. The objective is to build an ensemble of neural networks that enables a compressive autoencoder in a supervised fashion to retain a denser and more meaningful representation of the input histology images. Our proposed system is a simple and novel method to supervise compressive neural networks. We test the compressed images using transfer learning-based classifiers and show that they provide promising accuracy and classification performance.

IIMPACT: integrating image and molecular profiles for spatial transcriptomics analysis

Current clustering analysis of spatial transcriptomics data primarily relies on molecular information and fails to fully exploit the morphological features present in histology images, leading to compromised accuracy and interpretability. To overcome these limitations, we have developed a multi-stage statistical method called iIMPACT. It identifies and defines histology-based spatial domains based on AI-reconstructed histology images and spatial context of gene expression measurements, and detects domain-specific differentially expressed genes. Through multiple case studies, we demonstrate iIMPACT outperforms existing methods in accuracy and interpretability and provides insights into the cellular spatial organization and landscape of functional genes within spatial transcriptomics data.

Reproductive biology and nuptial coloration of Trachinus draco (Trachinidae) from Atlantic Spanish waters (SW Europe)

Studies focusing on reproductive biology play a key role in deciphering fish population dynamics. We studied the main features of the reproductive biology of the greater weever Trachinus draco based on gonad histology and digital image analysis techniques and the main hypothesis of the study is that males of T. draco exhibit a nuptial colouration. The biological samples of 311 individuals along the Atlantic coast of Spain, covering the twelve months of the year (from 2020 to 2022), in combination with the gonadosomatic index trend revealed a spawning season from May to August, consistent with similar studies in different geographical areas. Length-weight relationship presented differences by sex (females reach higher weight at the same length). The maturity ogive for females indicated that 50 % were mature at a length of 182 mm. Histological analysis confirmed asynchronous development of secondary growth oocytes, pointing to a batch spawning strategy. Digital image analysis of the nuptial coloration showed that the presence of a black spot above the pectoral fin in males was significantly correlated with the spawning capable phase, while blueish coloration was influenced by sex, size and maturity phase, with larger spawning males having the largest areas of blueish coloration. The results of this study represent a significant improvement of our understanding of the reproductive biology of T. draco.

The effect of aerobic-resistance training on the population of cardiac multipotent cells, thickness, and diameter of the ventricular wall through the c-kit pathway in male rats during the growth stages

Background. The purpose of the present study was to investigate the effect of aerobic-resistance training on the population of cardiac multipotent cells, thickness, and diameter of the ventricular through the C-Kit pathway of male rats during the growth stages. Methods. Overall, 30 male Wistar rats in three age groups of 2 (n = 10), 8 (n = 10), and 96 (n = 10) weeks were accidentally divided into exercise and control groups. Resistance training programs (resistance ladder, 3 days a week) and aerobics (treadmill running, 3 days a week) were performed for 6 weeks. Hematoxylin-Eosin staining was done the heart tissue and histological images were prepared. Afterward, C-Kit gene expression was examined by the real-time polymerase chain reaction method. The statistical method of one-way analysis of variance (P ≥ 0.05) and Tukey’s test were used at a significant level (P ≥ 0.01). Results. The trained neonatal group had higher values in heart weight to body weight index, and the trained neonatal and trained young adult groups had higher values in left ventricular thickness (P ≥ 0.01). The decrease in left ventricular internal diameter in the trained young adult group was significant compared to the control group (P ≥ 0.01). There was a significant increase in C-Kit gene expression in trained young adult and trained old groups (P ≥ 0.01). Conclusion. Aerobic-resistance training can be an effective stimulus for increasing the number of cardiac stem cells and creating structural adaptations of the heart, including increasing thickness and ventricular diameter in neonatal and young adult groups. Practical Implications. Participation in aerobic-resistance training programs leads to the improvement of the cardiac structure and an increase in cardiac stem cells.

Histopathology-based breast cancer prediction using deep learning methods for healthcare applications.

Breast cancer (BC) is the leading cause of female cancer mortality and is a type of cancer that is a major threat to women's health. Deep learning methods have been used extensively in many medical domains recently, especially in detection and classification applications. Studying histological images for the automatic diagnosis of BC is important for patients and their prognosis. Owing to the complication and variety of histology images, manual examination can be difficult and susceptible to errors and thus needs the services of experienced pathologists. Therefore, publicly accessible datasets called BreakHis and invasive ductal carcinoma (IDC) are used in this study to analyze histopathological images of BC. Next, using super-resolution generative adversarial networks (SRGANs), which create high-resolution images from low-quality images, the gathered images from BreakHis and IDC are pre-processed to provide useful results in the prediction stage. The components of conventional generative adversarial network (GAN) loss functions and effective sub-pixel nets were combined to create the concept of SRGAN. Next, the high-quality images are sent to the data augmentation stage, where new data points are created by making small adjustments to the dataset using rotation, random cropping, mirroring, and color-shifting. Next, patch-based feature extraction using Inception V3 and Resnet-50 (PFE-INC-RES) is employed to extract the features from the augmentation. After the features have been extracted, the next step involves processing them and applying transductive long short-term memory (TLSTM) to improve classification accuracy by decreasing the number of false positives. The results of suggested PFE-INC-RES is evaluated using existing methods on the BreakHis dataset, with respect to accuracy (99.84%), specificity (99.71%), sensitivity (99.78%), and F1-score (99.80%), while the suggested PFE-INC-RES performed better in the IDC dataset based on F1-score (99.08%), accuracy (99.79%), specificity (98.97%), and sensitivity (99.17%).

An unsupervised H&amp;E-based machine-learning approach for precise prediction of tumor microenvironment subtypes.

e13593 Background: Precision oncology increasingly relies on accurate classification of the tumor microenvironment (TME), which influences therapeutic responses and prognostication in cancer. While histological observation via H&amp;E staining remains the gold standard due to the detailed spatial context and morphological information provided, its lack of reproducibility limits its use in clinics. Next-generation sequencing (NGS) reliably provides valuable insights into the molecular landscape of tumors. Here we debut an unsupervised machine-learning approach that leverages both the rich detail in H&amp;E slides and the reproducibility of NGS to enhance TME classification. Methods: An ML-based image analysis algorithm was developed using a convolutional neural network (CNN) deep learning model. Using over 600 images from the TCGA COAD and READ datasets and from 139 colorectal cancer (CRC) samples in an internal cohort, we trained the CNN model to curate features from small patches of whole slide images (WSIs). Then, clusters were created from these H&amp;E WSIs without prior labeling. The clusters were validated by experienced pathologists to confirm cluster homogeneity and filter artifacts. We trained a classifier to categorize molecular features from the WSIs into TME subtypes. Performance was measured using the area under the receiver operating characteristic curve (AUC-ROC), based on the model's ability to correctly classify the Fibrotic (F) and Immune Desert (D) TME subtypes (1), as well as MSI status identified by whole exome sequencing. The F subtype corresponds with high stromal content, low immune infiltration, and the worst prognosis. The D subtype corresponds with a cold immune TME and an intermediate prognosis. Progression-free survival (PFS) data from TCGA were used to examine prognostic significance. Results: We obtained 0.86, 0.79, 0.72, and 0.71 AUC scores for predicting the ID and F TMEs, MSI status, and KRAS mutation status, respectively. PFS data analysis showed our H&amp;E-based approach to yield similar or better prognostic significance compared to NGS-based TME classification (ID vs others, p=0.05 (H&amp;E) and 0.2 (NGS); F vs others, p=0.07 (H&amp;E) and 0.09 (NGS)). Our findings showed a notable correlation with RNA- and DNA-based predictions of ID and F TMEs and MSI, with enhanced precision and more robust subtype identification while achieving high embedding consistency. Conclusions: By harnessing the inherently rich detail in histological images, our H&amp;E-based approach complements NGS-based TME analysis. Our findings advocate for the integration of advanced image analysis into the standard TME characterization workflow to refine prognostic and therapeutic strategies for CRC. We will expand this algorithm for use with other cancer types to realize its potential as a promising adjunct to NGS for precise TME classification. 1. Bagaev et al., 2021.

A multi-modal artificial intelligence model trained on H&amp;E tissue images and clinical patient data for prediction of BRAF mutations in colorectal cancer.

e15161 Background: Analyzing genetic mutations is essential for personalized therapy decisions in oncology and beyond. In colorectal cancer (CRC), mutations in BRAF ( BRAFmut) render patients eligible for targeted treatment with BRAF inhibitors and immunotherapy. BRAFmut is routinely analyzed via genomic sequencing, which, however, is costly and does not provide diagnostic results immediately. Reliable detection of BRAFmut status from histological images using artificial intelligence (AI) could provide a more accessible and faster diagnostic alternative. Previous approaches to image-based BRAFmut analysis, however, failed to reach a critical level of accuracy on independent test cohorts, with reported area under the receiver operating characteristic curve (AUROC) values typically below 0.8. Methods: We hypothesized that combining tissue image analysis with clinical patient data could improve diagnostic accuracy for BRAFmut. Following our hypothesis, we developed a multi-modal AI model on n = 426 CRC cases from the TCGA database using BRAFmut data, H&amp;E-stained tissue images, and information on patient parameters that are readily available in clinical practice and that are known to be associated with BRAFmut frequency. The model was based on a vision transformer architecture incorporating pathology-specific information using a pretrained embedding model for encoding tissue image features. Model performance was validated on an independent hold-out TCGA cohort of n = 107 samples and an additional external cohort of n = 104 CRC samples from the CPTAC database. Results: Our multi-modal AI model reached an AUROC of 0.85 ± 0.02 on the TCGA hold-out test set. Notably, on the second, external CPTAC dataset the predictive accuracy level was similar (AUROC of 0.83 ± 0.02) indicating the robustness of our approach to generalize to different datasets. Additional analyses showed that the significant performance boost over previous approaches was gained by the multi-modal analysis setup, as predicting BRAFmut based on either tissue, or clinical data alone led to lower accuracy values. Explainability analysis of AI results confirmed that relevant biological features on the histological slides, mainly tumor areas, were considered by the model for decision making. Conclusions: The combination of H&amp;E images with clinically readily available patient parameters enabled us to develop a model predicting BRAFmut in CRC that surpasses previous approaches in diagnostic accuracy on independent evaluation cohorts. This indicates that combining histological tissue information with clinical parameters can increase the ability of AI models to predict genetic variants from tissue and more generally highlights the potential of multi-modal deep learning approaches to enhance predictive AI models in digital pathology.

Neural Network Method for Detecting Blur in Histological Images

Neural Network Method for Detecting Blur in Histological Images

Prediction of melanoma metastasis using dermatoscopy deep features: An international multicenter cohort study.

9565 Background: Machine learning algorithms, and especially convolutional neural networks, demonstrated modest accuracy on the prediction of melanoma metastasis, based on histological images and clinicopathological information. Whether dermatoscopy deep features could serve as biomarker for the prediction of melanoma metastasis, remains an underexplored area in medical research. Methods: An international, multicenter, cohort study of cutaneous melanoma patients from 3 different continents was conducted. Patients with cutaneous melanoma, who had available clinical and dermatoscopic images and an adequate follow-up time for the development of metastasis (both locoregional and distant) were included. We utilized a support vector machine (SVM) classifier, to distinguish between melanomas that metastasized and those that did not. We used a pre-trained ResNet 50 network, we separated dataset into training set and testing set, stratified by TNM-stage, and to ensure robustness and guard against biased data selection, the stratified split into training and testing sets was repeated five times, resulting in five different training-test sets. The primary outcome was the comparison of the prognostic performance of deep dermatoscopy features based on SVM (model 1) to the performance of established prognostic factors of melanoma, such as Breslow thickness and ulceration (model 2) and to a combined model using deep features and histopathologic factors (model 3). A secondary aim was to examine the diagnostic performance of model 1 in stage IIB and IIC patients at diagnosis. The prognostic performance was assessed using the Area Under the Curve (AUC) and the True Positive Rate (TPR) at a True Negative Rate at 70%. Results: 712 patients were included, 465 (65.3%) non-metastatic and 247 (34.7%) metastatic, within a median follow-up of 60 months. The SVM model demonstrated mean AUC 0.84 (95% CI 0.80 – 0.87) and TPR 0.81 (95%CI 0.73 – 0.90). Similar results were shown for model 2 and model 3, and no statistically significant differences among models were detected in terms of AUC (De Long’s test, p&gt;0.05 and ANOVA Kruskal-Wallis p&gt;0.05). Regarding IIB/IIC patients and combining data from five test sets, SVM correctly classified as metastatic 21 out of 23 (91.3%) stage IIB and 21 out of 23 (91.3%) stage IIC patients, who eventually developed metastasis during follow-up. Conclusions: Our findings suggest that dermatoscopy deep features could offer an immediate, in vivo prediction of melanoma metastasis prior to excision. This advancement holds significant clinical importance, prioritizing high-risk patients for neoadjuvant treatment or guiding selection of patients who might benefit from adjuvant therapy.

Joint Super-resolution and Tissue Patch Classification for Whole Slide Histological Images

Joint Super-resolution and Tissue Patch Classification for Whole Slide Histological Images

Differences Between Medication-Related Osteonecrosis of the Jaw Caused by Bisphosphonates and Denosumab: Histological, Molecular Biological, and Clinical Studies.

Purpose Medication-related osteonecrosis of the jaw (MRONJ) is a serious side effect of antiresorptive agents such as bisphosphonates (BPs) and denosumab (DMB). We investigated whether a difference exists between BP- and DMB-related osteonecrosis of the jaw (ONJ). Patients and methods Histological images of 30 patients with BP-related ONJ and 13 patients with DMB-related ONJ were observed using hematoxylin-eosin and cathepsin K staining. Moreover, bone metabolism markers in the blood and bone mineral density were measured in 18 patients with BP-related ONJ and five patients with DMB-related ONJ. Furthermore, we conducted a quantitative analysis of local bone metabolism-related genes using surgical specimens through real-time reverse transcription polymerase chain reaction. Additionally, a retrospective study of 298 patients with MRONJ examined the differences in the characteristics of BP- and DMB-related ONJ and the factors associated with treatment outcomes. Results Histological examination revealed that patients treated with DMB had more severe osteoclast suppression than those treated with BP. No significant difference was observed in blood-bone metabolism markers between the two drugs; however, the suppression of local bone metabolism-related genes was stronger in patients treated with DMB. Clinical studies indicate that DMB-related ONJ is more frequently observed without osteolysis. Conclusion BP-associated ONJ and DMB-associated ONJ were shown to differ slightly. Clinical studies indicate that osteolysis is often unclear in DMB-related ONJ, and methods of bone resection during surgery need to be established.

Therapeutic response and outcomes with less common breast cancer subtypes in the I-SPY trial 2011-2022.

582 Background: Uncommon histologies are over-represented among high-risk breast cancer (BC) and denote an area with limited trial data and of significant unmet medical need. To better understand trial outcomes in this group and identify signals of tumor responsiveness, we report pathologic complete response (pCR) and early event-free survival (EFS) by disease subtype in the I-SPY2 trial. Additionally, the I-SPY2 trial currently utilizes a combination of tumor molecular signature and receptor status to determine response predictive subtype (RPS) first developed from 987 I-SPY2 patients (1). We report disease response rates for those who received what is now known to be optimal RPS guided therapy. Methods: The I-SPY2 platform trial tests novel agents given neo-adjuvantly with a chemotherapy backbone in high-risk BC (HER2 positive, triple negative and high molecular risk estrogen receptor positive BC). Histologic images of research biopsies, local biopsy and surgical pathology reports were reviewed centrally by I-SPY pathologists. Receptor subtype distribution and pCR rates were summarized. EFS was estimated using the Kaplan Meier method. Association between pCR and EFS was evaluated using the Cox proportional hazard model with significance assessed by the log rank test. Results: 144/2118 (7%) of I-SPY2 participants were identified with metaplastic (60), lobular (55), mucinous (9), micropapillary (8), neuroendocrine (4) and other (8) BCs. Tumor receptor status, pCR rate and EFS by tumor type are shown in the Table. For the full cohort, pCR was associated with better EFS (hazard ratio (95% CI): 0.12 (0.02 - 0.88), p = 0.01). Within metaplastic (metapBC) 11/32 (34%) and 5/28 (18%) patients had a pCR with or without checkpoint blockade, respectively. By RPS group, 9 of 18 (50%) (13 metapBC, 3 other, 2 lobular) in the HER2-Immune+ group who received RPS optimized therapy had a pCR. In this group 6/13 (46%) with metapBC had a pCR. In the HER2+ driven RPS groups, pCR rate in the HER+HER2orBasal was 88% (7/8) and 0% (0/3) in the HER2+Luminal. Conclusions: High pCR rates observed in metapBC and other among subsets of often difficult to treat BC subtypes support novel approaches and provide a roadmap for future study of uncommon BCs. Outcomes were improved when therapy matched RPS vulnerabilities. Participants presenting with uncommon BC subtypes will be prospectively identified in the I-SPY2.2 trial to further develop effective approaches for this group. 1. Wolf et al, Cancer Cell 2022. Clinical trial information: NCT01042379 . [Table: see text]

Masked pre-training of transformers for histology image analysis

In digital pathology, whole-slide images (WSIs) are widely used for applications such as cancer diagnosis and prognosis prediction. Vision transformer (ViT) models have recently emerged as a promising method for encoding large regions of WSIs while preserving spatial relationships among patches. However, due to the large number of model parameters and limited labeled data, applying transformer models to WSIs remains challenging. In this study, we propose a pretext task to train the transformer model in a self-supervised manner. Our model, MaskHIT, uses the transformer output to reconstruct masked patches, measured by contrastive loss. We pre-trained MaskHIT model using over 7000 WSIs from TCGA and extensively evaluated its performance in multiple experiments, covering survival prediction, cancer subtype classification, and grade prediction tasks. Our experiments demonstrate that the pre-training procedure enables context-aware understanding of WSIs, facilitates the learning of representative histological features based on patch positions and visual patterns, and is essential for the ViT model to achieve optimal results on WSI-level tasks. The pre-trained MaskHIT surpasses various multiple instance learning approaches by 3% and 2% on survival prediction and cancer subtype classification tasks, and also outperforms recent state-of-the-art transformer-based methods. Finally, a comparison between the attention maps generated by the MaskHIT model with pathologist's annotations indicates that the model can accurately identify clinically relevant histological structures on the whole slide for each task.