Clinical, imaging, serological, and histopathological features of pulmonary post-acute sequelae after mild COVID-19 (PASC)

ABSTRACT Background A significant proportion of patients experience prolonged pulmonary, cardiocirculatory or neuropsychiatric symptoms after Coronavirus disease 2019 (COVID-19), termed post-acute sequelae of COVID (PASC). Lung manifestations of PASC include cough, dyspnea on exertion and persistent radiologic abnormalities and have been linked to viral persistence, ongoing inflammation and immune dysregulation. So far, there is limited data on lung histopathology and tissue-based immune cell subtyping in PASC. Methods 51 unvaccinated patients (median age, 40 years; 43% female) with a median of 17 weeks (range, 2-55 weeks) after mild SARS-CoV-2 infection (without hospitalization) underwent full clinical evaluation including high-resolution computed tomography (HR-CT) and transbronchial biopsy. We used RT-PCR/FISH and immunohistochemistry (nucleocapsid/spike/CD3/CD4/CD8) for residual SARS-CoV-2 detection and T lymphocyte subtyping, respectively. We assessed interstitial fibrosis and macrophage profiles by transmission electron microscopy (TEM) and immunofluorescence multiplex staining, while cytokine profiling in bronchoalveolar lavage (BAL) fluid was performed by legendplex immunoassay. Results Dyspnea on exertion was the leading symptom of pulmonary PASC in our cohort. In 16% and 42.9% of patients, FEV1 and MEF50 were ≤ 80% and 35.3% showed low attenuation volume (LAV) in >5% of lung area, in line with airflow obstruction. There was a significant correlation between oxygen pulse and time since COVID (p=0.009). Histopathologically, PASC manifested as organizing pneumonia (OP), fibrinous alveolitis and increased CD4+ T cell infiltrate predominantly around airways (bronchiolitis), while the residual virus components were detectable in only a single PASC patient (2%). T cell infiltrates around small airways were inversely correlated with time since COVID, however, this trend failed to reach statistical significance. We identified discrete interstitial fibrosis and a pro-fibrotic macrophage subtype (CD68/CD163/S100A9) as well as significantly elevated interleukin 1β in BAL fluid from PASC patients (p=0.01), but H-scores for fibrotic macrophage population did not correlate with severity of clinical symptoms or T cell infiltration. Interpretation We show decreased FEV1/MEF50 and increased LAV in line with obstructive lung disease due to CD4+ T cell-predominant bronchiolitis as well as evidence of pro-fibrotic signaling in a subset of unvaccinated PASC patients. Since our results point towards self-limiting inflammation of small airways without detectable viral reservoirs, it remains unclear whether pulmonary symptoms in PASC are SARS-CoV-2-specific or represent a general response to viral infection. Still, evidence of pro-fibrotic signaling should warrant clincal follow-up and further research into possible long-time fibrotic remodeling in PASC patients. Key points Dyspnea on exertion is the leading clinical manifestation of PASC in the lung a minority of pts have significantly impaired lung function (FVC/TLC≤80% or DLCO≤70%) in spiroergometry and/or radiologic abnormalities, oxygen pulse seems to normalize over time 16% and 42.9% of pts have FEV1 and MEF50≤80% and 35.3% have LAV>5% of lung area, in line with airflow obstruction due to bronchiolitis Residual virus was not detectable in the lung tissue of all but one PASC patient (2%) Histologically, PASC may manifest as T cell-mediated bronchiolitis, OP and fibrinous alveolitis There is evidence of fibrotic remodeling (ultrastructural interstitial fibrosis, pro-fibrotic macrophage subpopulation, pro-fibrotic cytokine IL-1β in BAL) but this did not correlate with the degree of T cell infiltrate/bronchiolitis