Learning generalizable AI models for multi-center histopathology image classification.

Investigation of histopathology slides by pathologists is an indispensable component of the routine diagnosis of cancer. Artificial intelligence (AI) has the potential to enhance diagnostic accuracy, improve efficiency, and patient outcomes in clinical pathology. However, variations in tissue preparation, staining protocols, and histopathology slide digitization could result in over-fitting of deep learning models when trained on the data from only one center, thereby underscoring the necessity to generalize deep learning networks for multi-center use. Several techniques, including the use of grayscale images, color normalization techniques, and Adversarial Domain Adaptation (ADA) have been suggested to generalize deep learning algorithms, but there are limitations to their effectiveness and discriminability. Convolutional Neural Networks (CNNs) exhibit higher sensitivity to variations in the amplitude spectrum, whereas humans predominantly rely on phase-related components for object recognition. As such, we propose Adversarial fourIer-based Domain Adaptation (AIDA) which applies the advantages of a Fourier transform in adversarial domain adaptation. We conducted a comprehensive examination of subtype classification tasks in four cancers, incorporating cases from multiple medical centers. Specifically, the datasets included multi-center data for 1113 ovarian cancer cases, 247 pleural cancer cases, 422 bladder cancer cases, and 482 breast cancer cases. Our proposed approach significantly improved performance, achieving superior classification results in the target domain, surpassing the baseline, color augmentation and normalization techniques, and ADA. Furthermore, extensive pathologist reviews suggested that our proposed approach, AIDA, successfully identifies known histotype-specific features. This superior performance highlights AIDA's potential in addressing generalization challenges in deep learning models for multi-center histopathology datasets.

Update on Silicosis.

Although silicosis has been an established disease with a recognized cause for more than 100 years, many workers continue to be exposed to silica and new outbreaks of disease continue to occur. This article describes some of the well-established and new exposures, including denim sandblasting, artificial stone cutting, and some forms of "coal worker's pneumoconiosis." The authors review the imaging and pathology of acute silicosis (silicoproteinosis), simple silicosis, and progressive massive fibrosis and summarize known and putative associations of silica exposure, including tuberculosis, lung cancer, connective tissue disease (especially systemic sclerosis), and vasculitis.

Guidelines for Pathologic Diagnosis of Mesothelioma.

CONTEXT.­: Mesothelioma is an uncommon tumor that can be difficult to diagnose. OBJECTIVE.­: To provide updated, practical guidelines for the pathologic diagnosis of mesothelioma. DATA SOURCES.­: Pathologists involved in the International Mesothelioma Interest Group and others with expertise in mesothelioma contributed to this update. Reference material includes peer-reviewed publications and textbooks. CONCLUSIONS.­: There was consensus opinion regarding guidelines for (1) histomorphologic diagnosis of mesothelial tumors, including distinction of epithelioid, biphasic, and sarcomatoid mesothelioma; recognition of morphologic variants and patterns; and recognition of common morphologic pitfalls; (2) molecular pathogenesis of mesothelioma; (3) application of immunohistochemical markers to establish mesothelial lineage and distinguish mesothelioma from common morphologic differentials; (4) application of ancillary studies to distinguish benign from malignant mesothelial proliferations, including BAP1 and MTAP immunostains; novel immunomarkers such as Merlin and p53; fluorescence in situ hybridization (FISH) for homozygous deletion of CDKN2A; and novel molecular assays; (5) practical recommendations for routine reporting of mesothelioma, including grading epithelioid mesothelioma and other prognostic parameters; (6) diagnosis of mesothelioma in situ; (7) cytologic diagnosis of mesothelioma, including use of immunostains and molecular assays; and (8) features of nonmalignant peritoneal mesothelial lesions.

Mesothelioma: morphologic and immunohistochemical findings.

This paper reviews some basic and some new concepts in the diagnosis of mesothelioma. The term "malignant mesothelioma" is no longer recommended; rather, any tumor labeled "mesothelioma" is presumed to be malignant. Clinical and radiologic information is very useful in the diagnosis of mesothelioma; in particular, nodular pleural thickening on CT is usually a marker of malignancy. The literature on markers that separate mesotheliomas from metastatic carcinomas has become very complex and frequently misleading, with many recommended markers actually demonstrating poor specificity. However, newer data show that a combination of HEG1 (clone SKM9-2) and claudin­4 staining provides extremely high accuracy in separating epithelioid mesotheliomas from non-small-cell lung carcinomas with just two immunostains. This combination works at other sites as well, but caution should be used when high-grade serous carcinoma is in the differential, because all "mesothelioma" markers can also stain high-grade serous carcinomas. There are, unfortunately, no sensitive or specific markers for sarcomatoid mesotheliomas. A variety of immunohistochemical and fluorescence in situ hybridization (FISH) markers are useful in separating benign from malignant mesothelial proliferations; immunohistochemal staining for BAP1, MTAP (or CDKN2A FISH), and NF2/Merlin (or NF2 FISH) will enable the diagnosis of most mesotheliomas. Mesothelioma in situ is now recognized as either a single layer of bland cuboidal mesothelial cells that have lost BAP1, and sometimes MTAP, on immunohistochemical staining, or a process that is morphologically identical to a well-differentiated papillary mesothelial tumor that has lost BAP1/MTAP. Mesothelioma in situ probably always progresses to invasive mesothelioma, but this process is often quite slow.

Lung imaging patterns in connective tissue disease-associated interstitial lung disease impact prognosis and immunosuppression response.

OBJECTIVES: Interstitial lung disease (ILD) in connective tissue diseases (CTD) have highly variable morphology. We aimed to identify imaging features and their impact on ILD progression, mortality and immunosuppression response. METHODS: Patients with CTD-ILD had high-resolution chest computed tomography (HRCT) reviewed by expert radiologists blinded to clinical data for overall imaging pattern (usual interstitial pneumonia [UIP]; non-specific interstitial pneumonia [NSIP]; organizing pneumonia [OP]; fibrotic hypersensitivity pneumonitis [fHP]; and other). Transplant-free survival and change in percent-predicted forced vital capacity (FVC) were compared using Cox and linear mixed effects models adjusted for age, sex, smoking, and baseline FVC. FVC decline after immunosuppression was compared with pre-treatment. RESULTS: Of 645 CTD-ILD patients, the frequent CTDs were systemic sclerosis (n = 215), rheumatoid arthritis (n = 127), and inflammatory myopathies (n = 100). NSIP was the most common pattern (54%), followed by UIP (20%), fHP (9%), and OP (5%). Compared with UIP, FVC decline was slower for NSIP (1.1%/year, 95%CI 0.2, 1.9) and OP (3.5%/year, 95%CI 2.0, 4.9), and mortality was lower for NSIP (HR 0.65, 95%CI 0.45, 0.93) and OP (HR 0.18, 95%CI 0.05, 0.57), but higher in fHP (HR 1.58, 95%CI 1.01, 2.40). The extent of fibrosis also predicted FVC decline and mortality. After immunosuppression, FVC decline was slower compared with pre-treatment in NSIP (by 2.1%/year, 95%CI 1.4, 2.8), with no change for UIP or fHP. CONCLUSION: Multiple radiologic patterns are possible in CTD-ILD, including a fHP pattern. NSIP and OP were associated with better outcomes and response to immunosuppression, while fHP had worse survival compared with UIP.

New developments in mesothelial pathology.

This review article examines some new and some problem areas in mesothelial pathology, four of which are discussed, as follows. (1) The concept of mesothelioma in situ: this lesion is defined as a single layer of bland mesothelial cells without evidence of invasion, but that have lost BAP1 and/or MTAP by immunohistochemistry. Benign reactions can exactly mimic mesothelioma in situ, but a hint to the correct diagnosis is a story of recurrent pleural effusions/ascites of unknown aetiology without radiological or direct visual evidence of tumour. (2) The nature of well-differentiated papillary mesothelial tumour (WDPMT): WDPMT has a long history of arguments regarding its behaviour, and this uncertainty can now be seen to arise, in part, from the observation that some forms of mesothelioma in situ microscopically look exactly like WDPMT. Hence, it is recommended to always run at least a BAP1 stain on any lesion that looks like WDPMT. Both flat and WDPMT-like mesothelioma in situ are strongly associated with eventual development of invasive mesothelioma, but this process is relatively slow. (3) New immunostains for separating mesothelioma from other tumours: here, it is proposed that in most cases, and particularly when the differential is epithelioid mesothelioma versus non-small cell lung cancer, one can make this separation with extremely high sensitivity and specificity using just two stains: HEG1 and claudin-4. (4) Markers for separating benign from malignant mesothelial proliferations: this topic is briefly reviewed, with an indication of which markers are generally accepted and the best utilisation and possible limitations of each marker.