RESUMEN
Background: Better understanding of host inflammatory changes that precede development of severe COVID-19 could improve delivery of available antiviral and immunomodulatory therapies, and provide insights for the development of new therapies. Method(s): In plasma from individuals with COVID-19, sampled <=10 days from symptom onset from the All-Ireland Infectious Diseases Cohort study, we measured 61 biomarkers, including markers of innate immune and T cell activation, coagulation, tissue repair, lung injury, and immune regulation. We used principal component analysis (PCA) and k-means clustering to derive biomarker clusters, and univariate and multivariate ordinal logistic regression to explore association between cluster membership and maximal disease severity, adjusting for risk factors for severe COVID-19, including age, sex, ethnicity, BMI, hypertension and diabetes. Result(s): From March 2020-April 2021, we included 312 individuals, (median (IQR) age 62 (48-77) years, 7 (4-9) days from symptom onset, 54% male) in the analysis. PCA and clustering derived 4 clusters. Compared to cluster 1, clusters 2-4 were significantly older and of higher BMI but there were no significant differences in sex or ethnicity. Cluster 1 had low levels of inflammation, cluster 2 had higher levels of markers of tissue repair and endothelial activation (EGF, VEGF, PDGF, TGFalpha, serpin E1 and p-selectin). Cluster 3 and 4 were both characterised by higher overall inflammation, but compared to cluster 4, cluster 3 had downregulation of growth factors, markers of endothelial activation, and immune regulation (IL10, PDL1), but higher alveolar epithelial injury markers (RAGE, ST2). In univariate analysis, compared to cluster 1, cluster 3 had the highest odds of severe disease (OR (95% CI) 9.02 (4.62-18.31), followed by cluster 4: 5.59 (2.75-11.72) then cluster 2: 4.5 (2.38-8.81), all p < 0.05). Cluster 3 remained most strongly associated with severe disease in fully adjusted analyses;cluster 3: OR(95% CI) 5.99 (2.69-13.35), cluster 2: 3.14 (1.54-6.42), cluster 4: 3.13 (1.36-7.19), all p< 0.05). Conclusion(s): Distinct early inflammatory profiles predicted maximal disease severity independent of known risk factors for severe COVID-19. A cluster characterised by downregulation of growth factor and endothelial markers and early evidence of alveolar injury was associated with highest risk of developing severe COVID19. Whether this reflects a dysregulated inflammatory response that could improve targeted treatment requires further study. Heatmap of biomarker derived clusters and forest plot of association between clusters and disease severity. A: Heatmap demonstrating differences in biomarkers between clusters B: Forest plot demonstrating odds ratio of specific clusters for progressing to moderate or severe disease (reference Cluster 1), calculated using ordinal logistic regression. Odds ratio (95% CI) presented as unadjusted and fully adjusted (for age, sex, ethnicity, BMI, hypertension, diabetes, immunosuppression, smoking and baseline anticoagulant use). Maximal disease severity graded per the WHO severity scale.
RESUMEN
Background. Long COVID is a heterogenous condition. We previously demonstrated distinct phenotypes of long COVID, but the impact of later waves caused by SARS-CoV-2 variants on long COVID presentations has not been described. Methods. We selected individuals with ongoing symptoms > 4 weeks from PCR-confirmed COVID-19 in a multicentre, prospective cohort study. We used multiple correspondence analysis and hierarchical clustering on self-reported symptoms to identify symptom clusters, in individuals recruited during two periods;cohort 1 from March 2020 to April 2021, and cohort 2 from April 2021 to March 2022. We explored differences in symptoms by mapping acute infection to one of four COVID-19 waves in Ireland (table 1) as well as vaccination status, and used Chi2 test to compare symptoms frequencies. Results. Demographics are shown in Table 2. Cluster analysis of each cohort demonstrated 3 distinct clusters in both cohorts, which shared similar clinical characteristics;a musculoskeletal/pain symptom cluster, a cardiorespiratory cluster and a third less symptomatic cluster (Figure 1). While symptoms within clusters were similar across both periods, in the cardiorespiratory cluster, the frequency of palpitations decreased (56% vs 16%) and cough increased (14% vs 45%) between reporting periods (both P< 0.01). Furthermore, a greater proportion of palpitations were reported in those with COVID-19 from waves 1 and 2 (35% and 28%) compared to 3 and 4 (both 12%, P< 0.001), and a greater proportion of chest pain in waves 1, 2 and 4 compared to wave 3. There were no differences in other symptoms (Table 3). Additionally there were significantly less palpitations reported in those vaccinated at the time of review (17% vs 31% P=0.002), but not chest pain (30% vs 39% P=0.13). In multivariate analysis, infection in wave 3 and 4 but not vaccination status remained significantly associated with lower reported palpitations (OR (95% CI) 0.28 (0.13-0.97) and 0.5 (0.06-0.87) for waves 3 and 4, both P< 0.05), and wave 3 infection remained independently associated with lower reported chest pain (OR 0.3 (0.25-0.7)). Conclusion. Three symptom clusters define long COVID across the two cohorts, but characteristics of the cardiorespiratory phenotype have evolved over time with evolution of SARS-CoV-2 variants. (Table Presented).
RESUMEN
Background: Although reports suggest that most individuals with COVID-19 infection develop detectable antibodies post infection, the kinetics, durability, and relative differences between IgM and IgG responses remain poorly understood beyond the first few weeks after symptom onset. Methods: Within a large, well-phenotyped, diverse, prospective cohort of subjects with and without SARS-CoV-2 PCR-confirmed infection and historical controls derived from cohorts with high prevalence of viral coinfections and samples taken during prior flu seasons, we measured SARS-CoV-2 serological responses (both IgG and IgM) using three commercially available assays. We calculated sensitivity and specificity, relationship with disease severity and mapped the kinetics of antibody seropositivity and antibody levels over time using generalised additive models. Results: We analysed 1,001 samples (327 confirmed SARS-CoV-2, of whom 30% developed severe disease) from 752 subjects spanning a period of 90 days from symptom onset. Overall sensitivity was lower (44.1-47.1%) early (<10 days) after symptom onset but increased to >80% after 10 days. IgM positivity increased earlier than IgG-targeted assay but positivity peaked between day 32 and 38 post onset of symptoms and declined thereafter, a dynamic that was confirmed when antibody levels were analysed and was more rapid with IgM. Early (<10 days) IgM but not IgG levels were significantly higher in those who subsequently developed severe disease (signal / cut-off 4.20 (0.75-17.93) versus 1.07 (0.21-5.46), P=0.048). Conclusion: This study suggests that post-infectious antibody responses in those with confirmed COVID-19 infection begin to decline relatively early post infection and suggests a potential role for higher IgM levels early in infection predicting subsequent disease severity.