Vision Transformers for Breast Cancer Human Epidermal Growth Factor Receptor 2 Expression Staging without Immunohistochemical Staining.

Accurate staging of human epidermal growth factor receptor 2 (HER2) expression is vital for evaluating breast cancer treatment efficacy. However, it typically involves costly and complex immunohistochemical staining, along with hematoxylin and eosin staining. This work presents customized vision transformers for staging HER2 expression in breast cancer using only hematoxylin and eosin-stained images. The proposed algorithm comprised three modules: a localization module for weakly localizing critical image features using spatial transformers, an attention module for global learning via vision transformers, and a loss module to determine proximity to a HER2 expression level based on input images by calculating ordinal loss. Results, reported with 95% CIs, reveal the proposed approach's success in HER2 expression staging: area under the receiver operating characteristic curve, 0.9202 ± 0.01; precision, 0.922 ± 0.01; sensitivity, 0.876 ± 0.01; and specificity, 0.959 ± 0.02 over fivefold cross-validation. Comparatively, this approach significantly outperformed conventional vision transformer models and state-of-the-art convolutional neural network models (P < 0.001). Furthermore, it surpassed existing methods when evaluated on an independent test data set. This work holds great importance, aiding HER2 expression staging in breast cancer treatment while circumventing the costly and time-consuming immunohistochemical staining procedure, thereby addressing diagnostic disparities in low-resource settings and low-income countries.

Prediction of MET Overexpression in Lung Adenocarcinoma from Hematoxylin and Eosin Images.

Mesenchymal epithelial transition (MET) protein overexpression is a targetable event in non-small cell lung cancer and is the subject of active drug development. Challenges in identifying patients for these therapies include lack of access to validated testing, such as standardized immunohistochemistry assessment, and consumption of valuable tissue for a single gene/protein assay. Development of prescreening algorithms using routinely available digitized hematoxylin and eosin (H&E)-stained slides to predict MET overexpression could promote testing for those who will benefit most. Recent literature reports a positive correlation between MET protein overexpression and RNA expression. In this work, a large database of matched H&E slides and RNA expression data were leveraged to train a weakly supervised model to predict MET RNA overexpression directly from H&E images. This model was evaluated on an independent holdout test set of 300 overexpressed and 289 normal patients, demonstrating a receiver operating characteristic area under curve of 0.70 (95th percentile interval: 0.66 to 0.74) with stable performance characteristics across different patient clinical variables and robust to synthetic noise on the test set. These results suggest that H&E-based predictive models could be useful to prioritize patients for confirmatory testing of MET protein or MET gene expression status.

Comparing Accuracy of Helicobacter pylori Identification Using Traditional Hematoxylin and Eosin-Stained Glass Slides With Digital Whole Slide Imaging.

With advancements in the field of digital pathology, there has been a growing need to compare the diagnostic abilities of pathologists using digitized whole slide images against those when using traditional hematoxylin and eosin (H&E)-stained glass slides for primary diagnosis. One of the most common specimens received in pathology practices is an endoscopic gastric biopsy with a request to rule out Helicobacter pylori (H. pylori) infection. The current standard of care is the identification of the organisms on H&E-stained slides. Immunohistochemical or histochemical stains are used selectively. However, due to their small size (2-4 µm in length by 0.5-1 µm in width), visualization of the organisms can present a diagnostic challenge. The goal of the study was to compare the ability of pathologists to identify H. pylori on H&E slides using a digital platform against the gold standard of H&E glass slides using routine light microscopy. Diagnostic accuracy rates using glass slides vs digital slides were 81% vs 72% (P = .0142) based on H&E slides alone. When H. pylori immunohistochemical slides were provided, the diagnostic accuracy was significantly improved to comparable rates (96% glass vs 99% digital, P = 0.2199). Furthermore, differences in practice settings (academic/subspecialized vs community/general) and the duration of sign-out experience did not significantly impact the accuracy of detecting H. pylori on digital slides. We concluded that digital whole slide images, although amenable in different practice settings and teaching environments, does present some shortcomings in accuracy and precision, especially in certain circumstances and thus is not yet fully capable of completely replacing glass slide review for identification of H. pylori. We specifically recommend reviewing glass slides and/or performing ancillary stains, especially when there is a discrepancy between the degree of inflammation and the presence of microorganisms on digital images.

A Multicenter Study Validates the WHO 2022 Classification for Conjunctival Melanocytic Intraepithelial Lesions With Clinical and Prognostic Relevance.

Several nomenclature and grading systems have been proposed for conjunctival melanocytic intraepithelial lesions (C-MIL). The fourth "WHO Classification of Eye Tumors" (WHO-EYE04) proposed a C-MIL classification, capturing the progression of noninvasive neoplastic melanocytes from low- to high-grade lesions, onto melanoma in situ (MIS), and then to invasive melanoma. This proposal was revised to the WHO-EYE05 C-MIL system, which simplified the high-grade C-MIL, whereby MIS was subsumed into high-grade C-MIL. Our aim was to validate the WHO-EYE05 C-MIL system using digitized images of C-MIL, stained with hematoxylin and eosin and immunohistochemistry. However, C-MIL cases were retrieved from 3 supraregional ocular pathology centers. Adequate conjunctival biopsies were stained with hematoxylin and eosin, Melan-A, SOX10, and PReferentially expressed Antigen in Melanoma. Digitized slides were uploaded on the SmartZoom platform and independently scored by 4 ocular pathologists to obtain a consensus score, before circulating to 14 expert eye pathologists for independent scoring. In total, 105 cases from 97 patients were evaluated. The initial consensus diagnoses using the WHO-EYE04 C-MIL system were as follows: 28 benign conjunctival melanoses, 13 low-grade C-MIL, 37 high-grade C-MIL, and 27 conjunctival MIS. Using this system resulted in 93% of the pathologists showing only fair-to-moderate agreement (kappa statistic) with the consensus score. The WHO-EYE05 C-MIL system (with high-grade C-MIL and MIS combined) improved consistency between pathologists, with the greatest level of agreement being seen with benign melanosis (74.5%) and high-grade C-MIL (85.4%). Lowest agreements remained between pathologists for low-grade C-MIL (38.7%). Regarding WHO-EYE05 C-MIL scoring and clinical outcomes, local recurrences of noninvasive lesions developed in 8% and 34% of the low- and high-grade cases. Invasive melanoma only occurred in 47% of the cases that were assessed as high-grade C-MIL. This extensive international collaborative study is the first to undertake a comprehensive review of the WHO-EYE05 C-MIL scoring system, which showed good interobserver agreement and reproducibility.

Prediction of Epidermal Growth Factor Receptor Mutation Subtypes in Non-Small Cell Lung Cancer From Hematoxylin and Eosin-Stained Slides Using Deep Learning.

Accurate assessment of epidermal growth factor receptor (EGFR) mutation status and subtype is critical for the treatment of non-small cell lung cancer patients. Conventional molecular testing methods for detecting EGFR mutations have limitations. In this study, an artificial intelligence-powered deep learning framework was developed for the weakly supervised prediction of EGFR mutations in non-small cell lung cancer from hematoxylin and eosin-stained histopathology whole-slide images. The study cohort was partitioned into training and validation subsets. Foreground regions containing tumor tissue were extracted from whole-slide images. A convolutional neural network employing a contrastive learning paradigm was implemented to extract patch-level morphologic features. These features were aggregated using a vision transformer-based model to predict EGFR mutation status and classify patient cases. The established prediction model was validated on unseen data sets. In internal validation with a cohort from the University of Science and Technology of China (n = 172), the model achieved patient-level areas under the receiver-operating characteristic curve (AUCs) of 0.927 and 0.907, sensitivities of 81.6% and 83.3%, and specificities of 93.0% and 92.3%, for surgical resection and biopsy specimens, respectively, in EGFR mutation subtype prediction. External validation with cohorts from the Second Affiliated Hospital of Anhui Medical University and the First Affiliated Hospital of Wannan Medical College (n = 193) yielded patient-level AUCs of 0.849 and 0.867, sensitivities of 79.2% and 80.7%, and specificities of 91.7% and 90.7% for surgical and biopsy specimens, respectively. Further validation with The Cancer Genome Atlas data set (n = 81) showed an AUC of 0.861, a sensitivity of 84.6%, and a specificity of 90.5%. Deep learning solutions demonstrate potential advantages for automated, noninvasive, fast, cost-effective, and accurate inference of EGFR alterations from histomorphology. Integration of such artificial intelligence frameworks into routine digital pathology workflows could augment existing molecular testing pipelines.

Deep Learning-Based H-Score Quantification of Immunohistochemistry-Stained Images.

Immunohistochemistry (IHC) is a well-established and commonly used staining method for clinical diagnosis and biomedical research. In most IHC images, the target protein is conjugated with a specific antibody and stained using diaminobenzidine (DAB), resulting in a brown coloration, whereas hematoxylin serves as a blue counterstain for cell nuclei. The protein expression level is quantified through the H-score, calculated from DAB staining intensity within the target cell region. Traditionally, this process requires evaluation by 2 expert pathologists, which is both time consuming and subjective. To enhance the efficiency and accuracy of this process, we have developed an automatic algorithm for quantifying the H-score of IHC images. To characterize protein expression in specific cell regions, a deep learning model for region recognition was trained based on hematoxylin staining only, achieving pixel accuracy for each class ranging from 0.92 to 0.99. Within the desired area, the algorithm categorizes DAB intensity of each pixel as negative, weak, moderate, or strong staining and calculates the final H-score based on the percentage of each intensity category. Overall, this algorithm takes an IHC image as input and directly outputs the H-score within a few seconds, significantly enhancing the speed of IHC image analysis. This automated tool provides H-score quantification with precision and consistency comparable to experienced pathologists but at a significantly reduced cost during IHC diagnostic workups. It holds significant potential to advance biomedical research reliant on IHC staining for protein expression quantification.

Spatial transcriptomics prediction from histology jointly through Transformer and graph neural networks.

The rapid development of spatial transcriptomics allows the measurement of RNA abundance at a high spatial resolution, making it possible to simultaneously profile gene expression, spatial locations of cells or spots, and the corresponding hematoxylin and eosin-stained histology images. It turns promising to predict gene expression from histology images that are relatively easy and cheap to obtain. For this purpose, several methods are devised, but they have not fully captured the internal relations of the 2D vision features or spatial dependency between spots. Here, we developed Hist2ST, a deep learning-based model to predict RNA-seq expression from histology images. Around each sequenced spot, the corresponding histology image is cropped into an image patch and fed into a convolutional module to extract 2D vision features. Meanwhile, the spatial relations with the whole image and neighbored patches are captured through Transformer and graph neural network modules, respectively. These learned features are then used to predict the gene expression by following the zero-inflated negative binomial distribution. To alleviate the impact by the small spatial transcriptomics data, a self-distillation mechanism is employed for efficient learning of the model. By comprehensive tests on cancer and normal datasets, Hist2ST was shown to outperform existing methods in terms of both gene expression prediction and spatial region identification. Further pathway analyses indicated that our model could reserve biological information. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.

Deep Learning-Based Stratification of Gastric Cancer Patients from Hematoxylin and Eosin-Stained Whole Slide Images by Predicting Molecular Features for Immunotherapy Response.

Determining the molecular characteristics of cancer patients is crucial for optimal immunotherapy decisions. The aim of this study was to screen immunotherapy beneficiaries by predicting key molecular features from hematoxylin and eosin-stained images based on deep learning models. An independent data set from Asian gastric cancer patients was included for external validation. In addition, a segmentation model (Horizontal-Vertical Network) was used to quantify the cellular composition of tumor stroma. The model performance was evaluated by measuring the area under the curve (AUC). The tumor extraction model achieved an AUC of 0.9386 and 0.9062 in the internal and external test sets, respectively. The stratification model could predict the immunotherapy-sensitive subtypes (AUC range, 0.8685 to 0.9461), the genetic mutations (AUC range, 0.8283 to 0.9225), and the pathway activity (AUC range, 0.7568 to 0.8612) fairly accurately. In external validation, the prediction performance of Epstein-Barr virus and programmed cell death ligand 1 expression status achieved AUCs of 0.7906 and 0.6384, respectively. The segmentation model identified a relatively high proportion of inflammatory cells and connective cells in some immunotherapy-sensitive subtypes. The deep learning-based models potentially may serve as a valuable tool to screen for the beneficiaries of immunotherapy in gastric cancer patients.

A Deep Learning-Based System Trained for Gastrointestinal Stromal Tumor Screening Can Identify Multiple Types of Soft Tissue Tumors.

The accuracy and timeliness of the pathologic diagnosis of soft tissue tumors (STTs) critically affect treatment decision and patient prognosis. Thus, it is crucial to make a preliminary judgement on whether the tumor is benign or malignant with hematoxylin and eosin-stained images. A deep learning-based system, Soft Tissue Tumor Box (STT-BOX), is presented herein, with only hematoxylin and eosin images for malignant STT identification from benign STTs with histopathologic similarity. STT-BOX assumed gastrointestinal stromal tumor as a baseline for malignant STT evaluation, and distinguished gastrointestinal stromal tumor from leiomyoma and schwannoma with 100% area under the curve in patients from three hospitals, which achieved higher accuracy than the interpretation of experienced pathologists. Particularly, this system performed well on six common types of malignant STTs from The Cancer Genome Atlas data set, accurately highlighting the malignant mass lesion. STT-BOX was able to distinguish ovarian malignant sex-cord stromal tumors without any fine-tuning. This study included mesenchymal tumors that originated from the digestive system, bone and soft tissues, and reproductive system, where the high accuracy of migration verification may reveal the morphologic similarity of the nine types of malignant tumors. Further evaluation in a pan-STT setting would be potential and prospective, obviating the overuse of immunohistochemistry and molecular tests, and providing a practical basis for clinical treatment selection in a timely manner.

Polarization enhancement mechanism from tissue staining in multispectral Mueller matrix microscopy.

Mueller matrix microscopy can provide comprehensive polarization-related optical and structural information of biomedical samples label-freely. Thus, it is regarded as an emerging powerful tool for pathological diagnosis. However, the staining dyes have different optical properties and staining mechanisms, which can put influence on Mueller matrix microscopic measurement. In this Letter, we quantitatively analyze the polarization enhancement mechanism from hematoxylin and eosin (H&E) staining in multispectral Mueller matrix microscopy. We examine the influence of hematoxylin and eosin dyes on Mueller matrix-derived polarization characteristics of fibrous tissue structures. Combined with Monte Carlo simulations, we explain how the dyes enhance diattenuation and linear retardance as the illumination wavelength changed. In addition, it is demonstrated that by choosing an appropriate incident wavelength, more visual Mueller matrix polarimetric information can be observed of the H&E stained tissue sample. The findings can lay the foundation for the future Mueller matrix-assisted digital pathology.

Inflammation Is a Histological Characteristic of Skeletal Muscle in Chronic Limb Threatening Ischemia.

BACKGROUND: The loss of skeletal muscle is a prognostic factor in several diseases including in patients with chronic limb threatening ischemia (CLTI). Patients with CLTI also have a lower skeletal mass and area when compared to those with claudication. However, there are no currently available data regarding the histological characteristics of core muscles in patients with CLTI. This study aims to determine the differences in core skeletal muscles between patients with claudication and those with CLTI. The second aim is to evaluate the differences in myokines, which are molecules secreted by skeletal muscle, between patients with claudication and those with CLTI. METHODS: An observational, prospective study was conducted from January 2018 to July 2022 involving consecutive patients with peripheral arterial disease (PAD). The clinical characteristics were registered. In PAD patients with surgical indication for common femoral artery approach, samples of sartorius skeletal muscle (and not from the limb muscles directly involved in the ischemic process) were collected. The samples were submitted to histological characterization on hematoxylin-eosin and to immunohistochemical analysis to detect CD45+ leukocytes and CD163+ macrophages. The extent of the inflammatory cells (leukocytes and macrophages) was semiquantitatively assessed using a 0-to-4 grade scale as follows: absent (0†), mild (†), moderate (††), severe (†††), and very severe (††††). Serum levels of myokines: irisin, myostatin, IL-8, and lL-6 were determined with multiplex bead-based immunoassay. RESULTS: 119 patients (mean age: 67.58 ± 9.60 years old, 79.80% males) 64 with claudication and 54 with CLTI were enrolled in the study. No differences were registered between patients with claudication and those with CLTI on age, gender, cardiovascular risk factors, and medication, except on smoking habits. There was a significantly higher prevalence of smokers and a higher smoking load in the claudication group. Samples of sartorius skeletal muscle from 40 patients (14 with claudication and 26 with CLTI) were submitted to histological analysis. No differences were found in skeletal muscle fibers preservation, trauma, or hemorrhage (on hematoxylin-eosin staining). However, in the immunohistochemistry study, we found more inflammatory cells CD45+ leukocytes in patients with CLTI when compared to those with claudication [CD45+ ≥ moderate (††): claudication (n = 14): 4; 28.57%; CLTI (n = 25): 16; 64.00%; P = 0.034]. Patients with CLTI also had higher tissue levels of CD163+ macrophages, but this difference was not significant [CD163+ ≥ moderate (††): claudication (n = 13): 7; 53.85%; CLTI (n = 27): 21; 77.78%; P = 0.122]. The serum levels of the myokines, irisin, and myostatin were below the lower limit of detection, in the majority of patients, so no valid results were obtained. However, patients with CLTI had a higher serum level of Interleukin (IL)-6 and IL-8. CONCLUSIONS: CLTI patients exhibit increased quantities of leukocytes in their sartorius muscle, as well as elevated serum levels of myokines IL-8 and IL-6. Inflamed skeletal muscle can contribute to the loss of muscle mass and account for the lower density of skeletal muscle observed in CLTI. Additionally, inflamed skeletal muscle may contribute to the development of systemic inflammation through the secretion of pro-inflammatory cytokines into the systemic circulation. Halting the inflammatory process could eventually improve the prognosis of CLTI patients.

Intestinal metaplasia in follow-up endoscopies among Asian patients with short-segment Barrett's esophagus: Regression, sampling error, and associated factors.

BACKGROUND: The percentage of and factors associated with the regression of Barrett's esophagus (BE) or its characteristic intestinal metaplasia (IM) remain unclear, and conflicting results have been reported because of diverse regression and sampling error definitions. Thus, we investigated the rates of IM regression, sampling error, and associated factors. METHODS: Forty-two patients with proven short-segment BE with IM who underwent two follow-up endoscopies with biopsies of Barrett's mucosa were retrospectively analyzed. Additional Alcian blue and MUC2 staining were done on the biopsy specimens without IM in hematoxylin-eosin staining. Only patients with negative hematoxylin-eosin, Alcian blue, and MUC2 staining for IM in both follow-up endoscopies were considered to have true regression. When all three stains were negative for IM in the first, but positive in the second follow-up endoscopy, we considered IM persisting and declared sampling error. RESULTS: Among the 18 patients without IM at the first follow-up endoscopy, only five (11.9%) were judged to have true regression. Prolonged proton-pump inhibitor use was significantly associated with regression. Limited experience of the endoscopist, and insufficient biopsy number were significantly related to sampling error. Receiver operating characteristic (ROC) curve analysis showed the best cut-off value of the biopsy number/maximal-length (cm) ratio to predict sampling error was 2.25. CONCLUSION: In our patients with short-segment BE, 11.9% experienced regression of IM. Prolonged proton-pump inhibitors treatment was associated with regression. An insufficient biopsy number was related to a missed IM, which may be eliminated by maintaining biopsy number/maximal-length (cm) ratio ≥2.25.

Application of a Novel Miniaturized Histopathologic Microscope for Ex Vivo Identifying Cerebral Glioma Margins Rapidly During Surgery: A Parallel Control Study.

PURPOSE: The purpose of our study is to assess the clinical performance of the DiveScope, a novel handheld histopathologic microscope in rapidly differentiating glioma from normal brain tissue during neurosurgery. METHODS: Thirty-two ex vivo specimens from 18 patients were included in the present study. The excised suspicious tissue was sequentially stained with sodium fluorescein and methylene blue and scanned with DiveScope during surgery. The adjacent tissue was sent to the department of pathology for frozen section examination. They would eventually be sent to the pathology department later for hematoxylin and eosin staining for final confirmation. The positive likelihood ratio, negative likelihood ratio, sensitivity, specificity, and area under the curve of the device were calculated. In addition, the difference in time usage between DiveScope and frozen sections was compared for the initial judgment. RESULTS: The sensitivity and specificity of the DiveScope after analyzing hematoxylin and eosin -staining sections, were 88.29% and 100%, respectively. In contrast, the sensitivity and specificity of the frozen sections histopathology were 100% and 75%, respectively. The area under the curve of the DiveScope and the frozen sections histopathology was not significant ( P =0.578). Concerning time usage, DiveScope is significantly much faster than the frozen sections histopathology no matter the size of tissue. CONCLUSION: Compared with traditional pathological frozen sections, DiveScope was faster and displayed an equal accuracy for judging tumor margins intraoperatively.

Morphological Evaluation and Immunohistochemical Analysis of the Reparative Potential of the Buccal Fat Pad.

Background and Objectives: There are many surgical techniques for oroantral communication treatment, one of which is the buccal fat pad. Of particular interest is the high reparative potential of the buccal fat pad, which may be contributed to by the presence of mesenchymal stem cells. The purpose of this work is to evaluate the reparative potential of BFP cells using morphological and immunohistochemical examination. Materials and Methods: 30 BFP samples were provided by the Clinic of Maxillofacial and Plastic Surgery of the Russian University of Medicine (Moscow, Russia) from 28 patients. Morphological examination of 30 BFP samples was performed at the Institute of Clinical Morphology and Digital Pathology of Sechenov University. Hematoxylin-eosin, Masson trichrome staining and immunohistochemical examination were performed to detect MSCs using primary antibodies CD133, CD44 and CD10. Results: During staining with hematoxylin-eosin and Masson's trichrome, we detected adipocytes of white adipose tissue united into lobules separated by connective tissue layers, a large number of vessels of different calibers, as well as the general capsule of BFP. The thin connective tissue layers contained neurovascular bundles. Statistical processing of the results of the IHC examination of the samples using the Mann-Whitney criterion revealed that the total number of samples in which the expression of CD44, CD10 and CD133 antigens was confirmed was statistically significantly higher than the number of samples where the expression was not detected (p < 0.05). Conclusions: During the morphological study of the BFP samples, we revealed statistically significant signs of MSCs presence (p < 0.05), including in the brown fat tissue, which proves the high reparative potential of this type of tissue and can make the BFP a choice option among other autogenous donor materials when eliminating OAC and other surgical interventions in the maxillofacial region.

Unstained Tissue Imaging and Virtual Hematoxylin and Eosin Staining of Histologic Whole Slide Images.

Tissue structures, phenotypes, and pathology are routinely investigated based on histology. This includes chemically staining the transparent tissue sections to make them visible to the human eye. Although chemical staining is fast and routine, it permanently alters the tissue and often consumes hazardous reagents. On the other hand, on using adjacent tissue sections for combined measurements, the cell-wise resolution is lost owing to sections representing different parts of the tissue. Hence, techniques providing visual information of the basic tissue structure enabling additional measurements from the exact same tissue section are required. Here we tested unstained tissue imaging for the development of computational hematoxylin and eosin (HE) staining. We used unsupervised deep learning (CycleGAN) and whole slide images of prostate tissue sections to compare the performance of imaging tissue in paraffin, as deparaffinized in air, and as deparaffinized in mounting medium with section thicknesses varying between 3 and 20 µm. We showed that although thicker sections increase the information content of tissue structures in the images, thinner sections generally perform better in providing information that can be reproduced in virtual staining. According to our results, tissue imaged in paraffin and as deparaffinized provides a good overall representation of the tissue for virtually HE-stained images. Further, using a pix2pix model, we showed that the reproduction of overall tissue histology can be clearly improved with image-to-image translation using supervised learning and pixel-wise ground truth. We also showed that virtual HE staining can be used for various tissues and used with both 20× and 40× imaging magnifications. Although the performance and methods of virtual staining need further development, our study provides evidence of the feasibility of whole slide unstained microscopy as a fast, cheap, and feasible approach to producing virtual staining of tissue histology while sparing the exact same tissue section ready for subsequent utilization with follow-up methods at single-cell resolution.

Using Artificial Intelligence to Identify Tumor Microenvironment Heterogeneity in Non-Small Cell Lung Cancers.

Lung cancer heterogeneity is a major barrier to effective treatments and encompasses not only the malignant epithelial cell phenotypes and genetics but also the diverse tumor-associated cell types. Current techniques used to investigate the tumor microenvironment can be time-consuming, expensive, complicated to interpret, and often involves destruction of the sample. Here we use standard hematoxylin and eosin-stained tumor sections and the HALO AI nuclear phenotyping software to characterize 6 distinct cell types (epithelial, mesenchymal, macrophage, neutrophil, lymphocyte, and plasma cells) in both murine lung cancer models and human lung cancer samples. CD3 immunohistochemistry and lymph node sections were used to validate lymphocyte calls, while F4/80 immunohistochemistry was used for macrophage validation. Consistent with numerous prior studies, we demonstrated that macrophages predominate the adenocarcinomas, whereas neutrophils predominate the squamous cell carcinomas in murine samples. In human samples, we showed a strong negative correlation between neutrophils and lymphocytes as well as between mesenchymal cells and lymphocytes and that higher percentages of mesenchymal cells correlate with poor prognosis. Taken together, we demonstrate the utility of this AI software to identify, quantify, and compare distributions of cell types on standard hematoxylin and eosin-stained slides. Given the simplicity and cost-effectiveness of this technique, it may be widely beneficial for researchers designing new therapies and clinicians working to select favorable treatments for their patients.

Utility of Machine Learning to Detect Cytomegalovirus in Digital Hematoxylin and Eosin-Stained Slides.

Rapid and accurate cytomegalovirus (CMV) identification in immunosuppressed or immunocompromised patients presenting with diarrhea is essential for therapeutic management. Due to viral latency, however, the gold standard for CMV diagnosis remains to identify viral cytopathic inclusions on routine hematoxylin and eosin (H&E)-stained tissue sections. Therefore, biopsies may be taken and "rushed" for pathology evaluation. Here, we propose the use of artificial intelligence to detect CMV inclusions on routine H&E-stained whole-slide images to aid pathologists in evaluating these cases. Fifty-eight representative H&E slides from 30 cases with CMV inclusions were identified and scanned. The resulting whole-slide images were manually annotated for CMV inclusions and tiled into 300 × 300 pixel patches. Patches containing annotations were labeled "positive," and these tiles were oversampled with image augmentation to account for class imbalance. The remaining patches were labeled "negative." Data were then divided into training, validation, and holdout sets. Multiple deep learning models were provided with training data, and their performance was analyzed. All tested models showed excellent performance. The highest performance was seen using the EfficientNetV2BO model, which had a test (holdout) accuracy of 99.93%, precision of 100.0%, recall (sensitivity) of 99.85%, and area under the curve of 0.9998. Of 518,941 images in the holdout set, there were only 346 false negatives and 2 false positives. This shows proof of concept for the use of digital tools to assist pathologists in screening "rush" biopsies for CMV infection. Given the high precision, cases screened as "positive" can be quickly confirmed by a pathologist, reducing missed CMV inclusions and improving the confidence of preliminary results. Additionally, this may reduce the need for immunohistochemistry in limited tissue samples, reducing associated costs and turnaround time.

Targeted Quantitative Mass Spectrometry Analysis of Protein Biomarkers From Previously Stained Single Formalin-Fixed Paraffin-Embedded Tissue Sections.

Formalin-fixed, paraffin-embedded tissues represent a majority of all biopsy specimens commonly analyzed by histologic or immunohistochemical staining with adhesive coverslips attached. Mass spectrometry (MS) has recently been used to precisely quantify proteins in samples consisting of multiple unstained formalin-fixed, paraffin-embedded sections. Here, we report an MS method to analyze proteins from a single coverslipped 4-µm section previously stained with hematoxylin and eosin, Masson trichrome, or 3,3'-diaminobenzidine-based immunohistochemical staining. We analyzed serial unstained and stained sections from non-small cell lung cancer specimens for proteins of varying abundance (PD-L1, RB1, CD73, and HLA-DRA). Coverslips were removed by soaking in xylene, and after tryptic digestion, peptides were analyzed by targeted high-resolution liquid chromatography with tandem MS with stable isotope-labeled peptide standards. The low-abundance proteins RB1 and PD-L1 were quantified in 31 and 35 of 50 total sections analyzed, respectively, whereas higher abundance CD73 and HLA-DRA were quantified in 49 and 50 sections, respectively. The inclusion of targeted ß-actin measurement enabled normalization in samples where residual stain interfered with bulk protein quantitation by colorimetric assay. Measurement coefficient of variations for 5 replicate slides (hematoxylin and eosin stained vs unstained) from each block ranged from 3% to 18% for PD-L1, from 1% to 36% for RB1, 3% to 21% for CD73, and 4% to 29% for HLA-DRA. Collectively, these results demonstrate that targeted MS protein quantification can add a valuable data layer to clinical tissue specimens after assessment for standard pathology end points.

Quantitative Pathologic Analysis of Digitized Images of Colorectal Carcinoma Improves Prediction of Recurrence-Free Survival.

BACKGROUND & AIMS: To examine whether quantitative pathologic analysis of digitized hematoxylin and eosin slides of colorectal carcinoma (CRC) correlates with clinicopathologic features, molecular alterations, and prognosis. METHODS: A quantitative segmentation algorithm (QuantCRC) was applied to 6468 digitized hematoxylin and eosin slides of CRCs. Fifteen parameters were recorded from each image and tested for associations with clinicopathologic features and molecular alterations. A prognostic model was developed to predict recurrence-free survival using data from the internal cohort (n = 1928) and validated on an internal test (n = 483) and external cohort (n = 938). RESULTS: There were significant differences in QuantCRC according to stage, histologic subtype, grade, venous/lymphatic/perineural invasion, tumor budding, CD8 immunohistochemistry, mismatch repair status, KRAS mutation, BRAF mutation, and CpG methylation. A prognostic model incorporating stage, mismatch repair, and QuantCRC resulted in a Harrell's concordance (c)-index of 0.714 (95% confidence interval [CI], 0.702-0.724) in the internal test and 0.744 (95% CI, 0.741-0.754) in the external cohort. Removing QuantCRC from the model reduced the c-index to 0.679 (95% CI, 0.673-0.694) in the external cohort. Prognostic risk groups were identified, which provided a hazard ratio of 2.24 (95% CI, 1.33-3.87, P = .004) for low vs high-risk stage III CRCs and 2.36 (95% CI, 1.07-5.20, P = .03) for low vs high-risk stage II CRCs, in the external cohort after adjusting for established risk factors. The predicted median 36-month recurrence rate for high-risk stage III CRCs was 32.7% vs 13.4% for low-risk stage III and 15.8% for high-risk stage II vs 5.4% for low-risk stage II CRCs. CONCLUSIONS: QuantCRC provides a powerful adjunct to routine pathologic reporting of CRC. A prognostic model using QuantCRC improves prediction of recurrence-free survival.

Deep Learning for Predicting Effect of Neoadjuvant Therapies in Non-Small Cell Lung Carcinomas With Histologic Images.

Neoadjuvant therapies are used for locally advanced non-small cell lung carcinomas, whereby pathologists histologically evaluate the effect using resected specimens. Major pathological response (MPR) has recently been used for treatment evaluation and as an economical survival surrogate; however, interobserver variability and poor reproducibility are often noted. The aim of this study was to develop a deep learning (DL) model to predict MPR from hematoxylin and eosin-stained tissue images and to validate its utility for clinical use. We collected data on 125 primary non-small cell lung carcinoma cases that were resected after neoadjuvant therapy. The cases were randomly divided into 55 for training/validation and 70 for testing. A total of 261 hematoxylin and eosin-stained slides were obtained from the maximum tumor beds, and whole slide images were prepared. We used a multiscale patch model that can adaptively weight multiple convolutional neural networks trained with different field-of-view images. We performed 3-fold cross-validation to evaluate the model. During testing, we compared the percentages of viable tumor evaluated by annotator pathologists (reviewed data), those evaluated by nonannotator pathologists (primary data), and those predicted by the DL-based model using 2-class confusion matrices and receiver operating characteristic curves and performed a survival analysis between MPR-achieved and non-MPR cases. In cross-validation, accuracy and mean F1 score were 0.859 and 0.805, respectively. During testing, accuracy and mean F1 score with reviewed data and those with primary data were 0.986, 0.985, 0.943, and 0.943, respectively. The areas under the receiver operating characteristic curve with reviewed and primary data were 0.999 and 0.978, respectively. The disease-free survival of MPR-achieved cases with reviewed and primary data was significantly better than that of the non-MPR cases (P<.001 and P=.001), and that predicted by the DL-based model was almost identical (P=.005). The DL model may support pathologist evaluations and can offer accurate determinations of MPR in patients.