Comparison of RNA In Situ Hybridization and Immunohistochemistry Techniques for the Detection and Localization of SARS-CoV-2 in Human Tissues

Coronavirus disease-19 (COVID-19) is caused by a newly discovered coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Although SARS-CoV-2 is visualized on electron microscopy, there is an increasing demand for widely applicable techniques to visualize viral components within tissue specimens. Viral protein and RNA can be detected on formalin-fixed paraffin-embedded (FFPE) tissue using immunohistochemistry (IHC) and in situ hybridization (ISH), respectively. Herein, we evaluate the staining performance of ISH for SARS-CoV-2 and an IHC directed at the SARS-CoV nucleocapsid protein and compare these results to a gold standard, tissue quantitative real-time polymerase chain reaction (qRT-PCR). We evaluated FFPE sections from 8 COVID-19 autopsies, including 19 pulmonary and 39 extrapulmonary samples including the heart, liver, kidney, small intestine, skin, adipose tissue, and bone marrow. We performed RNA-ISH for SARS-CoV-2 on all cases with IHC for SARS-CoV and SARS-CoV-2 qRT-PCR performed on selected cases. Lungs from 37 autopsies performed before the COVID-19 pandemic served as negative controls. The ISH and IHC slides were reviewed by 4 observers to record a consensus opinion. Selected ISH and IHC slides were also reviewed by 4 independent observers. Evidence of SARS-CoV-2 was identified on both the IHC and ISH platforms. Within the postmortem lung, detected viral protein and RNA were often extracellular, predominantly within hyaline membranes in patients with diffuse alveolar damage. Among individual cases, there was regional variation in the amount of detectable virus in lung samples. Intracellular viral RNA and protein was localized to pneumocytes and immune cells. Viral RNA was detected on RNA-ISH in 13 of 19 (68%) pulmonary FFPE blocks from patients with COVID-19. Viral protein was detected on IHC in 8 of 9 (88%) pulmonary FFPE blocks from patients with COVID-19, although in 5 cases the stain was interpreted as equivocal. From the control cohort, FFPE blocks from all 37 patients were negative for SARS-CoV-2 RNA-ISH, whereas 5 of 13 cases were positive on IHC. Collectively, when compared with qRT-PCR on individual tissue blocks, the sensitivity and specificity for ISH was 86.7% and 100%, respectively, while those for IHC were 85.7% and 53.3%, respectively. The interobserver variability for ISH ranged from moderate to almost perfect, whereas that for IHC ranged from slight to moderate. All extrapulmonary samples from COVID-19-positive cases were negative for SARS-CoV-2 by ISH, IHC, and qRT-PCR. SARS-CoV-2 is detectable on both RNA-ISH and nucleocapsid IHC. In the lung, viral RNA and nucleocapsid protein is predominantly extracellular and within hyaline membranes in some cases, while intracellular locations are more prominent in others. The intracellular virus is detected within pneumocytes, bronchial epithelial cells, and possibly immune cells. The ISH platform is more specific, easier to analyze and the interpretation is associated with the improved interobserver agreement. ISH, IHC, and qRT-PCR failed to detect the virus in the heart, liver, and kidney.