Molecular pathology selected abstracts

Create PDF

Editors: Donna E. Hansel, MD, PhD, chair of pathology, Oregon Health and Science University, Portland; Richard D. Press, MD, PhD, professor and director of molecular pathology, OHSU; James Solomon, MD, PhD, assistant professor, Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York; Sounak Gupta, MBBS, PhD, senior associate consultant, Mayo Clinic, Rochester, Minn.; Fei Yang, MD, assistant professor, Department of Pathology, OHSU; Andrés G. Madrigal, MD, PhD, molecular genetic pathology fellow, Department of Pathology, OHSU; and Erica Reinig, MD, assistant professor and medical director of molecular diagnostics, University of Wisconsin-Madison.

Use of DNA methylation profiling to diagnose malignant mesothelioma

July 2021—The accurate diagnosis of malignant pleural mesothelioma is important because of its association with asbestos inhalation and because it is an aggressive tumor with poor outcome despite multimodal treatment. Unfortunately, however, diagnosing malignant pleural mesothelioma is not straightforward. Initial diagnosis often occurs on small biopsies, and the disease has morphologic overlap not only with other neoplasms that affect the lungs and pleura, such as solitary fibrous tumor or synovial sarcoma, but also with reactive conditions, such as reactive mesothelial hyperplasia or sclerosing fibrous pleuritis. While immunohistochemical markers can provide ancillary information, many of them are nonspecific and can lead to diagnostic dilemmas. Sequencing of cancer genes can help in some situations but cannot provide a definitive diagnosis. DNA methylation profiling recently has demonstrated clinical utility in classifying and diagnosing brain tumors. By analyzing the methylation status of thousands of possible methylation sites across the genome, a DNA methylation profile can be determined. This profile can serve as a tumor-specific fingerprint for determining tumor type and site of origin. The authors conducted a study in which they evaluated methylation profiling for its ability to distinguish malignant pleural mesothelioma from other neoplastic or reactive mimics. In a discovery cohort, they compared the methylation profiles of 34 malignant mesothelioma specimens, including epithelioid, biphasic, and sarcomatoid subtypes, with those of 180 other entities that could present diagnostic dilemmas. The authors used two statistical methods for interpreting high-dimensional data to compare the methylation profiles: unsupervised hierarchical clustering and t-distributed stochastic neighbor embedding (t-SNE). Through this analysis, they demonstrated that mesotheliomas formed a well-defined cluster that was distinct from the other entities. This well-defined cluster was also seen in an independent validation series of 46 additional mesothelioma specimens. A few outlier cases did not cluster as expected. In some cases, low tumor content or DNA quality caused background noise in the analysis output, highlighting the importance of sample quality. However, the findings of one discrepant case were surprising. This involved a specimen thought to be reactive that was found to cluster within the mesothelioma group. The authors performed additional workups for this case, including molecular testing and retrospective morphologic review, both of which were highly suggestive but not necessarily diagnostic of malignant mesothelioma. The patient was diagnosed with malignant mesothelioma nine months later. Therefore, this case highlights how methylation profiling can provide additional information to aid in diagnosis. While additional steps must be taken before methylation profiling is used routinely in the clinical laboratory—for example,
developing a methylation classifier to report objective diagnostic scores for individual cases—this study demonstrates the potential and clinical utility of methylation classification for diagnosing malignant pleural meso­thelioma.

Bertero L, Righi L, Collemi G, et al. DNA methylation profiling discriminates between malignant pleural mesothelioma and neoplastic or reactive histological mimics [published online ahead of print April 19, 2021]. J Mol Diagn. 2021. doi:10.1016/j.jmoldx.2021.04.002

Correspondence: Luca Bertero at luca.bertero@unito.it

Spatial analysis of COVID-19 lung pathology

It is thought that the clinical severity of COVID-19 disease is related primarily to the patient’s immune response. SARS-CoV-2 infection can cause overactivation of the patient’s adaptive immune system, resulting in tissue and organ damage that leads to clinically severe acute respiratory distress syndrome (ARDS). However, a spatial analysis of how the infected cells and immune system interact has not been performed. The authors conducted a study in which they used imaging mass cytometry to further understand this pathologic process. Imaging mass cytometry allows for high-resolution imaging and quantitation of many different analytes across a tissue section. In the process, lanthanide metal-labeled antibodies are incubated with the tissue section and bind to their antigens. Then laser ablation allows for mass spectrometry to quantify each antibody in the minute area targeted by the laser. The quantitative data collected across a region of tissue can be converted into images to examine individual cells and tissue morphology in further downstream analyses. To investigate the cellular composition and spacial architecture of the lung, the authors designed a metal-labeled antibody panel for imaging mass cytometry comprising 36 biomarkers. The panel included phenotype markers of endothelial, epithelial, mesenchymal, and immune system cells, functional markers, and an antibody specific to the spike protein of SARS-CoV-2. This landmark study examined 23 autopsy patients, including 10 patients who died from COVID-19; nine control patients who died of ARDS resulting from influenza, bacterial infection, or acute bacterial pneumonia; and four control patients who did not have lung disease at the time of death. The authors analyzed 237 multiplexed images at 1-µm resolution, comprising 332 mm² of tissue and 664,006 single cells across all specimens. The spatially resolved data allowed for characterization of the lung pathology, and the authors made a number of notable observations. For example, alveolar epithelial cells showed the highest rate of SARS-CoV-2 positivity, suggesting that these are the predominant cells infected by the virus. Additionally, infected alveolar epithelial cells exhibited high levels of specific markers, including pSTAT3, KIT, IL-6, arginase-1, CASP3, and complement C5b-C9, that were not seen in uninfected cells. While some of these elevated markers were found in other lung infections, the CASP3 and complement markers were exclusive to SARS-CoV-2–infected alveolar epithelial cells. This suggests that the apoptotic and complement mediated host defense pathways with which these markers are associated are integral to COVID-19 pathology. This complement activation results in “off-target” tissue damage and continues to stimulate the cycle of inflammation. Further, spatial analysis demonstrated that activated macrophages interact with fibroblasts and mesenchymal cells in the alveolar walls, contributing to the fibrosis seen in late-stage COVID-19. The authors concluded that these findings not only have implications for future therapeutic intervention but also elucidate lung pathology more broadly.