Explore the vast range of titles available.

The development of decision support systems for pathology and their deployment in clinical practice have been hindered by the need for large manually annotated datasets. To overcome this problem, we present a multiple instance learning-based deep learning system that uses only the reported diagnoses as labels for training, thereby avoiding expensive and time-consuming pixel-wise manual annotations. We evaluated this framework at scale on a dataset of 44,732 whole slide images from 15,187 patients without any form of data curation. Tests on prostate cancer, basal cell carcinoma and breast cancer metastases to axillary lymph nodes resulted in areas under the curve above 0.98 for all cancer types. Its clinical application would allow pathologists to exclude 65–75% of slides while retaining 100% sensitivity. Our results show that this system has the ability to train accurate classification models at unprecedented scale, laying the foundation for the deployment of computational decision support systems in clinical practice.

In this review, we detail the changes and the relevant features that are applied to neuroendocrine neoplasms (NENs) in the 2022 WHO Classification of Endocrine and Neuroendocrine Tumors. Using a question-and-answer approach, we discuss the consolidation of the nomenclature that distinguishes neuronal paragangliomas from epithelial neoplasms, which are divided into well-differentiated neuroendocrine tumors (NETs) and poorly differentiated neuroendocrine carcinomas (NECs). The criteria for these distinctions based on differentiation are outlined. NETs are generally (but not always) graded as G1, G2, and G3 based on proliferation, whereas NECs are by definition high grade; the importance of Ki67 as a tool for classification and grading is emphasized. The clinical relevance of proper classification is explained, and the importance of hormonal function is examined, including eutopic and ectopic hormone production. The tools available to pathologists for accurate classification include the conventional biomarkers of neuroendocrine lineage and differentiation, INSM1, synaptophysin, chromogranins, and somatostatin receptors (SSTRs), but also include transcription factors that can identify the site of origin of a metastatic lesion of unknown primary site, as well as hormones, enzymes, and keratins that play a role in functional and structural correlation. The recognition of highly proliferative, well-differentiated NETs has resulted in the need for biomarkers that can distinguish these G3 NETs from NECs, including stains to determine expression of SSTRs and those that can indicate the unique molecular pathogenetic alterations that underlie the distinction, for example, global loss of RB and aberrant p53 in pancreatic NECs compared with loss of ATRX, DAXX, and menin in pancreatic NETs. Other differential diagnoses are discussed with recommendations for biomarkers that can assist in correct classification, including the distinctions between epithelial and non-epithelial NENs that have allowed reclassification of epithelial NETs in the spine, in the duodenum, and in the middle ear; the first two may be composite tumors with neuronal and glial elements, and as this feature is integral to the duodenal lesion, it is now classified as composite gangliocytoma/neuroma and neuroendocrine tumor (CoGNET). The many other aspects of differential diagnosis are detailed with recommendations for biomarkers that can distinguish NENs from non-neuroendocrine lesions that can mimic their morphology. The concepts of mixed neuroendocrine and non-neuroendocrine (MiNEN) and amphicrine tumors are clarified with information about how to approach such lesions in routine practice. Theranostic biomarkers that assist patient management are reviewed. Given the significant proportion of NENs that are associated with germline mutations that predispose to this disease, we explain the role of the pathologist in identifying precursor lesions and applying molecular immunohistochemistry to guide genetic testing.

Context.--Although most pancreatic and bile duct neoplasms are solid, mucinous cystic neoplasms and intraductal neoplasms have been increasingly recognized even when clinically silent, thanks to the increased use of sensitive imaging techniques. Cystic and intraductal neoplasms of the pancreas are often resectable and curable and constitute about 5% of all pancreatic neoplasms. Owing to their preinvasive nature and different biology, recognition of these entities remains a major priority. Mucinous cystic neoplasms are histologically and clinically distinct from other cystic pancreatic neoplasms. Pancreatic intraductal neoplasms encompass 3 major entities: intraductal papillary mucinous neoplasm, intraductal oncocytic papillary neoplasm, and intraductal tubulopapillary neoplasm. Intraductal papillary neoplasms of bile ducts are also preinvasive mass-forming neoplasms with both similarities and differences with their pancreatic counterparts. All of these pancreatobiliary neoplasms have diverse and distinctive clinicopathologic, genetic, and prognostic variations. Objective.--To review the clinical, pathologic, and molecular features of mucinous cystic and intraductal neoplasms of the pancreatobiliary tract. Data Sources.--Literature review, diagnostic manuals, and guidelines. Conclusions.--This review will briefly describe wellknown clinical and pathologic features and will focus on selected recently described aspects of morphology, grading, classification, and genomic alterations of cystic and intraductal neoplasms of the pancreatobiliary tract. (Arch Pathol Lab Med. 2022;146:298-311; doi: 10.5858/arpa.2021-0399-RA)

The commonly mutated genes in pancreatic neuroendocrine tumors (PanNETs) are ATRX , DAXX , and MEN1 . We genotyped 64 PanNETs and found 58% carry ATRX , DAXX , and MEN1 mutations (A-D-M mutant PanNETs) and this correlates with a worse clinical outcome than tumors carrying the wild-type alleles of all three genes (A-D-M WT PanNETs). We performed RNA sequencing and DNA-methylation analysis to reveal two distinct subgroups with one consisting entirely of A-D-M mutant PanNETs. Two genes differentiating A-D-M mutant from A-D-M WT PanNETs were high ARX and low PDX1 gene expression with PDX1 promoter hyper-methylation in the A-D-M mutant PanNETs. Moreover, A-D-M mutant PanNETs had a gene expression signature related to that of alpha-cells (FDR q -value < 0.009) of pancreatic islets including increased expression of HNF1A and its transcriptional target genes. This gene expression profile suggests that A-D-M mutant PanNETs originate from or transdifferentiate into a distinct cell type similar to alpha cells. In pancreatic neuroendocrine tumors (PanNETs) ATRX, DAXX, and MEN1 are commonly mutated (A-D-M mutant PanNETs). Here, the authors find in a cohort of PanNETS 58% are A-D-M mutant PanNETs, with a worse clinical outcome and differences in gene expression and methylation compared to A-D-M wild type cases- these gene expression differences suggest that A-D-M mutant PanNETs potentially originate from a cell type similar to alpha cells.

The analysis of histopathology images with artificial intelligence aims to enable clinical decision support systems and precision medicine. The success of such applications depends on the ability to model the diverse patterns observed in pathology images. To this end, we present Virchow, the largest foundation model for computational pathology to date. In addition to the evaluation of biomarker prediction and cell identification, we demonstrate that a large foundation model enables pan-cancer detection, achieving 0.95 specimen-level area under the (receiver operating characteristic) curve across nine common and seven rare cancers. Furthermore, we show that with less training data, the pan-cancer detector built on Virchow can achieve similar performance to tissue-specific clinical-grade models in production and outperform them on some rare variants of cancer. Virchow’s performance gains highlight the value of a foundation model and open possibilities for many high-impact applications with limited amounts of labeled training data. Trained on 1.5 million whole-slide images from 100,000 patients, a pathology foundation model is shown to improve performance of specialized models in detection of rare cancers.

Although most pancreatic and bile duct neoplasms are solid, mucinous cystic neoplasms and intraductal neoplasms have been increasingly recognized even when clinically silent, thanks to the increased use of sensitive imaging techniques. Cystic and intraductal neoplasms of the pancreas are often resectable and curable and constitute about 5% of all pancreatic neoplasms. Owing to their preinvasive nature and different biology, recognition of these entities remains a major priority. Mucinous cystic neoplasms are histologically and clinically distinct from other cystic pancreatic neoplasms. Pancreatic intraductal neoplasms encompass 3 major entities: intraductal papillary mucinous neoplasm, intraductal oncocytic papillary neoplasm, and intraductal tubulopapillary neoplasm. Intraductal papillary neoplasms of bile ducts are also preinvasive mass-forming neoplasms with both similarities and differences with their pancreatic counterparts. All of these pancreatobiliary neoplasms have diverse and distinctive clinicopathologic, genetic, and prognostic variations. To review the clinical, pathologic, and molecular features of mucinous cystic and intraductal neoplasms of the pancreatobiliary tract. Literature review, diagnostic manuals, and guidelines. This review will briefly describe well-known clinical and pathologic features and will focus on selected recently described aspects of morphology, grading, classification, and genomic alterations of cystic and intraductal neoplasms of the pancreatobiliary tract.

The field of anatomic pathology has been evolving in the last few decades and the advancements have been largely fostered by innovative technology. Immunohistochemistry enabled a paradigm shift in discovery and diagnostic evaluation, followed by booming genomic advancements which allowed for submicroscopic pathologic characterization, and now the field of digital pathology coupled with machine learning and big data acquisition is paving the way to revolutionize the pathology medical domain. Whole slide imaging (WSI) is a disruptive technology where glass slides are digitized to produce on-screen whole slide images. Specifically, in the past decade, there have been significant advances in digital pathology systems that have allowed this technology to promote integration into clinical practice. Whole slide images (WSI), or digital slides, can be viewed and navigated comparable to glass slides on a microscope, as digital files. Whole slide imaging has increased in adoption among pathologists, pathology departments, and scientists for clinical, educational, and research initiatives. Integration of digital pathology systems requires a coordinated effort with numerous stakeholders, not only within the pathology department, but across the entire enterprise. Each pathology department has distinct needs, use cases and blueprints, however the framework components and variables for successful clinical integration can be generalized across any organization seeking to undergo a digital transformation at any scale. This article will review those components and considerations for integrating digital pathology systems into clinical practice.

Prostate cancer (PrCa) is the second most common cancer among men in the United States. The gold standard for detecting PrCa is the examination of prostate needle core biopsies. Diagnosis can be challenging, especially for small, well-differentiated cancers. Recently, machine learning algorithms have been developed for detecting PrCa in whole slide images (WSIs) with high test accuracy. However, the impact of these artificial intelligence systems on pathologic diagnosis is not known. To address this, we investigated how pathologists interact with Paige Prostate Alpha, a state-of-the-art PrCa detection system, in WSIs of prostate needle core biopsies stained with hematoxylin and eosin. Three AP-board certified pathologists assessed 304 anonymized prostate needle core biopsy WSIs in 8 hours. The pathologists classified each WSI as benign or cancerous. After ~4 weeks, pathologists were tasked with re-reviewing each WSI with the aid of Paige Prostate Alpha. For each WSI, Paige Prostate Alpha was used to perform cancer detection and, for WSIs where cancer was detected, the system marked the area where cancer was detected with the highest probability. The original diagnosis for each slide was rendered by genitourinary pathologists and incorporated any ancillary studies requested during the original diagnostic assessment. Against this ground truth, the pathologists and Paige Prostate Alpha were measured. Without Paige Prostate Alpha, pathologists had an average sensitivity of 74% and an average specificity of 97%. With Paige Prostate Alpha, the average sensitivity for pathologists significantly increased to 90% with no statistically significant change in specificity. With Paige Prostate Alpha, pathologists more often correctly classified smaller, lower grade tumors, and spent less time analyzing each WSI. Future studies will investigate if similar benefit is yielded when such a system is used to detect other forms of cancer in a setting that more closely emulates real practice.

Proteomic analyses indicate that STAT1 protein (signal transducer and activator of transcription 1 or transcription factor ISGF-3 components p91/p84) is upregulated in some colorectal cancers. This study examined 736 colorectal cancer patients for the expression of STAT1 protein in tissue specimens, including 614 early stage patients and 122 advanced stage patients. Tissue microarrays were constructed, and STAT1 expression was examined by immunohistochemistry and scored semi-quantitatively. Among all cases, 9% of cases displayed high levels of cytoplasmic expression of STAT1 and 15% of cases had positive nuclear expression. Based on statistical analyses of a cohort of 559 early stage patients with survival data and no neoadjuvant therapy, we found that high levels of cytoplasmic expression of STAT1 correlated with shorter survival time in early stage colorectal cancer, particularly of the microsatellite instability (MSI) subtype. Additional analysis of a 244-case cohort of colorectal cancers from the Cancer Genome Atlas found that STAT1 gene expression correlated positively with PD-L1 (CD274) and PD-1 (PDCD1) but had no correlation with KRAS or BRAF mutation status. STAT1 expression showed no clear correlation with any of the 4 clinical diagnostic markers of mismatch repair, MLH1, MSH2, MSH6, and PMS2, suggesting its potential as an independent outcome marker for MSI cancers. Our findings suggest that STAT1 may be used as a potential prognostic protein marker for stratifying the outcome risk of early stage MSI colorectal cancer.