key: cord-0938626-1qhen1bj authors: Young, Barnaby E; Ong, Sean W X; Ng, Lisa F P; Anderson, Danielle E; Chia, Wan Ni; Chia, Po Ying; Ang, Li Wei; Mak, Tze-Minn; Kalimuddin, Shirin; Chai, Louis Yi Ann; Pada, Surinder; Tan, Seow Yen; Sun, Louisa; Parthasarathy, Purnima; Fong, Siew-Wai; Chan, Yi-Hao; Tan, Chee Wah; Lee, Bernett; Rötzschke, Olaf; Ding, Ying; Tambyah, Paul; Low, Jenny G H; Cui, Lin; Barkham, Timothy; Lin, Raymond Tzer Pin; Leo, Yee-Sin; Renia, Laurent; Wang, Lin-Fa; Lye, David Chien title: Viral dynamics and immune correlates of COVID-19 disease severity date: 2020-08-28 journal: Clin Infect Dis DOI: 10.1093/cid/ciaa1280 sha: 5d6438e7f84eeb49e71d469f1e1bc72c16748897 doc_id: 938626 cord_uid: 1qhen1bj BACKGROUND: Key knowledge gaps remain in the understanding of viral dynamics and immune response of SARS-CoV-2 infection. METHODS: We evaluated these characteristics and established their association with clinical severity in a prospective observational cohort study of 100 patients with PCR-confirmed SARS-CoV-2 infection (mean age 46 years, 56% male, 38% with comorbidities). Respiratory samples (n=74) were collected for viral culture, serum samples for measurement of IgM/IgG levels (n=30), and plasma samples for levels of inflammatory cytokines and chemokines (n=81). Disease severity was correlated with results from viral culture, serologic testing, and immune markers. RESULTS: 57 (57%) patients developed viral pneumonia, of whom 20 (20%) required supplemental oxygen including 12 (12%) invasive mechanical ventilation. Viral culture from respiratory samples was positive for 19 of 74 patients (26%). No virus was isolated when the PCR cycle threshold (Ct) value was >30 or >14 days after symptom onset. Seroconversion occurred at a median of 12.5 days (IQR 9-18) for IgM and 15.0 days (IQR 12-20) for IgG; 54/62 patients (87.1%) sampled at day 14 or later seroconverted. Severe infections were associated with earlier seroconversion and higher peak IgM and IgG levels. Levels of IP-10, HGF, IL-6, MCP-1, MIP-1α, IL-12p70, IL-18, VEGF-A, PDGF-BB and IL-1RA significantly correlated with disease severity. CONCLUSION: We found virus viability was associated with lower PCR Ct value in early illness. A stronger antibody response was associated with disease severity. The overactive proinflammatory immune signatures offers targets for host-directed immunotherapy which should be evaluated in randomised controlled trials. A c c e p t e d M a n u s c r i p t The coronavirus disease 2019 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused great harm to health and the global economy [1, 2] . Understanding of SARS-CoV-2 pathogeneisis has advanced at an unprecedented speed, but key gaps remain and preliminary findings require validation. Studies of COVID-19 patients have described the inflammatory milieu in severe infections, with raised neutrophils, suppressed lymphocytes and elevated inflammatory mediators [3, 4] . However, most studies are limited to comparing severe against non-severe infections and lack serial data [5, 6] . A clearer definition of disease pathogenesis will support the development of risk stratification tools and therapeutics targeting critical pathways in the inflammatory cascade. Early seroconversion has been reported with IgG to the receptor binding domain (RBD) detected at day seven [7] [8] [9] [10] . However, evidence of correlation between antibody titres and disease severity is conflicting [8, [10] [11] [12] . There remains a need for more detailed assessment of antibody kinetics to help determine the impact of antibody-dependent enhancement in COVID-19 pathogenesis, as well as guide convalescent plasma harvesting and use of serological assays for diagnosis [13] . SARS-CoV-2 can be detected from the nasopharynx for a median of 2-3 weeks following onset of symptoms [14] . Several studies have reported that in immunocompetent individuals, virus is typically only cultured from respiratory samples during the first week of illness when viral loads are highest [8, 15, 16] . This suggests transmission risk declines in the second week. This finding requires confirmation in larger cohorts as it has important implications for infection control and isolation protocols [17] . A c c e p t e d M a n u s c r i p t In this multi-pronged study, we describe the serologic evolution, inflammatory response and pattern of viral shedding and viability in patients with virologically confirmed COVID-19 in Singapore, and analyse the contributions these make to severe infections. All individuals confirmed to have COVID-19 by SARS-CoV-2 real-time reverse transcriptasepolymerase chain reaction (RT-PCR) and admitted to any of seven public hospitals in Singapore were eligible for inclusion in this study. RT-PCR was performed on respiratory samples as previously described [18] (details in Supplementary Appendix). Clinical information was extracted from the medical record using a standardized data collection form adapted from the International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC) case record form [19] . Serial blood and respiratory samples were collected during hospitalisation and follow-up post-discharge (days 1, 3, 7, 14, 21 and 28 after enrolment). A stool and urine sample was also collected for SARS-CoV-2 PCR and culture on Day 1. At the time of the study, all patients with COVID-19 in Singapore were admitted to airborne infection isolation rooms regardless of disease severity. Supportive therapy including supplemental oxygen and symptomatic treatment were administered as required. Patients with moderate to severe hypoxia were transferred to ICU for further management and invasive mechanical ventilation as required. A c c e p t e d M a n u s c r i p t Patients were discharged from hospital only after resolution of symptoms and when two consecutive nasopharyngeal swabs >24 hours apart were negative for SARS-CoV-2 by RT-PCR. Follow-up visits were arranged on day 28 of enrolment. Material from nasopharyngeal swabs was collected in universal transport media and used to inoculate Vero-E6 cells (ATCC®CRL-1586TM) for virus isolation in an Animal Biological Safety Level 3 laboratory. Urine and stool samples were collected and transported fresh for virus culture. Stools were filtered before inoculation. Cells were cultured at 37 o C for seven days or less if cytopathic effect (CPE) was observed, and three blind passages were performed. CPE consisted of rounded cells and extensive cell death, usually by day four post-inoculation. Positive isolation was confirmed by the observation of CPE and virus-specific PCR. Total RNA was extracted from all samples using E.Z.N.A. Total RNA Kit I (Omega Bio-Tek Inc.) according to the manufacturer's instructions and samples were analyzed by RT-PCR for the detection of SARS-CoV-2 as previously described [20] . Serum collected during the acute and convalescent phases of infection were tested for SARS-CoV-2 receptor binding domain specific IgM and IgG using capture ELISA (details in Supplementary Appendix). Plasma samples were collected during acute and convalescent phases and treated with a solvent/detergent based on Triton TM X-100 (1%) for virus inactivation [21] . Immune mediator levels were measured using Cytokine/Chemokine/Growth Factor 45-Plex Human ProcartaPlex TM Panel 1 (ThermoFisher Scientific) (details in Supplementary Appendix). Cytokine levels were also measured in 23 healthy donor plasma as baseline controls. A c c e p t e d M a n u s c r i p t Patients were categorised into three groups for comparison: no pneumonia on chest radiographs (CXR) throughout admission; pneumonia on CXR without hypoxia; pneumonia with hypoxia (desaturation to ≤94%) needing supplemental oxygen. Day one was defined as the first day of symptom onset. Fold change (FC) of antibody titres compared with negative controls were calculated. Kruskal-Wallis test followed by Dunn's multiple comparison test was used to determine significance of antibody levels. Unpaired t-test was applied to ascertain significant difference in the immune mediator levels between the COVID-19 patients and healthy controls at different time points post illness onset. Oneway ANOVA analysis with post-hoc t-test corrected using the method of Bonferroni was used to discern the differences in immune mediator levels between the various disease severity groups. One-way ANOVA results were corrected for multiple testing using the method of Benjamini and Hochberg. A c c e p t e d M a n u s c r i p t Written informed consent was obtained from all participants. The study protocol at all sites was approved by the National Healthcare Group Domain Specific Review Board, Study Reference 2012/00917; additional study protocol at Singapore General Hospital approved by the SingHealth Centralised Institutional Review Board, Study Reference 2018/3045. 130 COVID-19 patients were diagnosed in Singapore from the first known case on 22 January 2020 to 6 March 2020 ( Figure 1 ). Among the 100 (77%) enrolled in this study, viral pneumonia was diagnosed by CXR in 57, of whom 20 required supplemental oxygen for hypoxia, and 12 required invasive mechanical ventilation (Table 1) . Following resolution of infection and viral shedding, 97 patients have been discharged and after 90 days of follow up, no re-infection or recrudescence has been detected. Three patients have died. Deaths occurred 27 days or more after hospital admission, and while all were assessed by the managing clinican as related to COVID-19 they followed viral clearance and prolonged invasive mechanical ventilation. SARS-CoV-2 was detectable from nasopharyngeal swabs by PCR up to 48 days after symptom onset. Further network analyses showed association between the aforementioned immune mediators and clinical parameters (pneumonia, hypoxia and ICU admission). The intertwined relationship of the cytokines is shown in Figure 5A . In addition, Ingenuity Pathway Analysis (IPA) revealed canonical pathways associated with these immune mediators and severity, with the top ten canonical pathways involved in inflammatory diseases and cell signalling ( Figure 5B ). The top canonical pathway highlights the common immune mediators between influenza and SARS-CoV-2 infection, including MCP-1, IL-1RA, IP-10 and IL-6 ( Figure 5C ). In addition to IPA, STRING prediction of proteinprotein interactions identified IL-6 as a direct interacting partner with other severity-associated immune mediators ( Figure 5D ). Longitudinal comparison of these immune mediators associated with severity in 12 ICU patients was performed to explore their role as prognostic markers for severe COVID-19 (Supplementary Figure 6 ). Four of these twelve patients (CT009, CT032, CT037 and CT057) recovered and were discharged at the time of study. Interestingly, HGF and VEGF-A were distinctly separated into two levels, with the four discharged patients having lower HGF and VEGF-A levels, and HGF approaching healthy baseline levels during convalescence, which further indicates that high levels of these cytokines were This study of 100 patients in the first few months of the COVID-19 pandemic in Singapore provides a detailed overview of clinical presentation, progress and outcomes. Case detection and contact tracing in Singapore is rigorous with the main gap possibly in missing asymptomatic cases [22] . From the cohort, 43% of patients never developed pneumonia, 37% developed pneumonia without hypoxia and 20% pneumonia with hypoxia. We found no relationship between illness severity and duration of viral shedding or PCR Ct values. The central role of the immune response to SARS-CoV-2 in COVID-19 was evident from the strong correlation between disease severity and levels of IgG/IgM and inflammatory immune mediators in our cohort. Viral shedding of SARS-CoV-2 from the respiratory tract has been observed to persist for several weeks and potentially up to months [18, [23] [24] [25] . Whether the virus remains infectious throughout this prolonged viral shedding is an important question to resolve. Smaller studies have found that viable virus was readily isolated from immunocompetent individuals during the first week of illness, but were unable to successfully isolate viruses in culture from day 8 onwards despite detectable viral load by PCR [8, 25, 26] . However, a case report from Taiwan showed it was possible to culture the virus up till day 18 [27] , while positive cultures up to Day 20 has been reported from a study of patients with severe infection [15] . We found that successful virus culture was associated with PCR Ct value ≤30. Virus was isolated up to Day 14 post-symptom onset, though the majority were cultured at day 10 or earlier. Using PCR Ct value to guide decision making on de-isolation may be an A c c e p t e d M a n u s c r i p t alternative to using day of illness and provide an additional level of reassurance. However, this requires further validation. We observed seroconversion of IgM at a median of 12.5 days and IgG at 15.0 days. Similar to a Hong Kong study of 16 patients, there were no significant differences of IgM and IgG titres by age or comorbidities [10] . We found IgM and IgG titres correlated with disease severity, similarly found in two Chinese studies of 173 [7] and 285 patients respectively [28] . Different sample sizes and antibody assays may account for differing seroconversion rates, which merits further investigation in a large cohort over a longer period. Serological testing is vital for determining the prevalence of infection in sero-surveys and as part of epidemiological investigations to understand transmission of clusters [29] . The timing of seroconversion may guide the timing of plasmapheresis for convalescent plasma [30, 31] . The role of antibodies in long-term immunity after infection needs further investigation. cytokine storm in a subset of patients with markedly increased levels of pro-inflammatory cytokines, chemokines and growth factors [32, 33] . Notably, higher levels of IL-6, MCP-1, IP-10, IL-18, IL-1RA, PDGF-BB, HGF, VEGF-A, IL-12p70 and MIP-1 were associated with severe disease in our study. This robust induction of pro-inflammatory mediators indicates that innate immune cell responses and anti-viral T-cell responses are responsible for SARS-CoV-2 pathogenesis in COVID-19 patients [34] . In addition, elevation of growth factors, including HGF [35] , PLGF-1 [36] and LIF [37] , illuminates the A c c e p t e d M a n u s c r i p t repair mechanisms following acute lung injury during SARS-CoV-2 infection. Previous studies showed that ICU patients had more significant cytokine activation compared with non-ICU patients [1] . Our longitudinal cytokine profiling in ICU patients further evaluated prognostic values of specific cytokines. We observed better prognosis in critically ill patients with lower levels of acute lung injury associated growth factors, HGF and VEGF-A at the time of admission to ICU. Our data suggest that the differences in degree of lung injury could reflect recovery rate of patients in ICU. HGF and VEGF-A may serve as early indicators of poor prognosis and may provide guidance to make pre-emptive clinical decisions in critically ill patients. A central role for IL-6 in lung injury has been postulated with higher levels associated with mortality in two separate studies and severe immune-mediated injury in lung tissues of patients who died from COVID-19 [38, 39] . Similarly, the IL-1 pathway has been highlighted to contribute towards SARS-CoV-2 pathogenesis [40] . Importantly, our study highlighted the upregulation of IL-1 pathway mediators IL-18, MCP-1, and VEGF-A in critically-ill patients, providing further evidence that dysregulation in the IL-1 pathway could contribute to the hyperinflammatory state, especially in fatal cases. STRING analysis revealed potential protein-to-protein interactions in severe COVID-19 infection, in which IL-6 is the direct interacting partner of other cytokines associated with disease severity. Thus, several approved IL-6 receptor antagonists could be repurposed to treat severe SARS-CoV-2 infection. Given the elevated level of VEGF-A, JAK inhibitors should also be investigated as a potential therapeutic option [41, 42] . However, the serial data presented here indicates that the therapeutic window for intervention is narrow. Immune modulators will need to be active before the inflammatory cascade causing acute lung injury develops. Our study has several limitations. Active case finding through contact tracing is likely to have identified the majority of symptomatic infections in Singapore over this time period. However, A c c e p t e d M a n u s c r i p t atypical or subclinical infections may have been missed. No infections were detected in children (<18 years) or residents of long term care facilities. Individuals were enrolled as soon as possible following admission, but while biological samples were collected serially, they were not all acquired at the same timepoint after symptom onset. Additionally, some laboratory data were incomplete, and clinical data such as date of symptom onset is subject to recall bias. We also did not investigate the effect of different SARS-CoV-2 lineages or mutations on study outcomes [43] . Finally, samples were processed consistently but numerous factors determine the success of viral cultures, and correlation between culturability and infectiousness is unclear. In conclusion, we found virus viability was associated with lower PCR Ct value in early illness. This Dr Young had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. BY reports personal fees from Roche and Sanofi, outside the submitted work. All other no interest declared. M a n u s c r i p t M a n u s c r i p t 130 patients diagnosed in Singapore as of 6 March 2020. *1 excluded as diagnosis was retrospective via serology rather than PCR. The 100 were enrolled from the following hospitals in Singapore: National Centre for Infectious Diseases (77); National University Hospital (8); Singapore General Hospital (8); Ng Teng Fong General Hospital (3); Changi General Hospital (2); Alexandra Hospital (1); Khoo Teck Puat Hospital (1). Fold of change (Fc) is calculated by dividing optical density (OD) reading of a test sample by the average OD reading of negative controls. Samples with Fc > 3 are considered positive. As sampling time and numbers were not uniform for all patients, we plotted the highest IgM/IgG from a single serum sample for each patient when multiple samples were available. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Interventions to mitigate early spread of SARS-CoV-2 in Singapore: a modelling study Clinical and immunological features of severe and moderate coronavirus disease 2019 Dysregulation of Immune Response in Patients With Coronavirus Longitudinal analyses reveal immunological misfiring in severe COVID-19 Longitudinal COVID-19 profiling associates IL-1RA and IL-10 with disease severity and RANTES with mild disease Antibody responses to SARS-CoV-2 in patients of novel coronavirus disease Virological assessment of hospitalized patients with COVID-2019 Severe Acute Respiratory Syndrome Coronavirus 2-Specific Antibody Responses in Coronavirus Disease Patients Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study Kinetics of SARS-CoV-2 specific IgM and IgG responses in COVID-19 patients Neutralizing Antibody Responses to Severe Acute Respiratory Syndrome Coronavirus 2 in Coronavirus Disease 2019 Inpatients and Convalescent Patients Impact of immune enhancement on Covid-19 polyclonal hyperimmune globulin therapy and vaccine development Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China Shedding of infectious virus in hospitalized patients with coronavirus disease-2019 (COVID-19): duration and key determinants SARS-CoV-2 Virus Culture and Subgenomic RNA for Respiratory Specimens from Patients with Mild Coronavirus Disease Temporal dynamics in viral shedding and transmissibility of COVID-19 Epidemiologic Features and Clinical Course of Patients Infected With SARS-CoV-2 in Singapore COVID-19 CRF -ISARIC. Available at Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR Evaluation of inactivation methods for severe acute respiratory syndrome coronavirus in noncellular blood products Using observational data to quantify bias of traveller-derived COVID-19 prevalence estimates in Wuhan, China SARS-CoV-2 Viral Load in Upper Respiratory Specimens of Infected Patients Detection of SARS-CoV-2 in Different Types of Clinical Specimens Profile of RT-PCR for SARS-CoV-2: a preliminary study from 56 COVID-19 patients Predicting infectious SARS-CoV-2 from diagnostic samples Prolonged virus shedding even after seroconversion in a patient with COVID-19 Antibody responses to SARS-CoV-2 in patients with COVID-19 Investigation of three clusters of COVID-19 in Singapore: implications for surveillance and response measures Treatment With Convalescent Plasma for Critically Ill Patients With Severe Acute Respiratory Syndrome Coronavirus 2 Infection Treatment of 5 Critically Ill Patients With COVID-19 With Convalescent Plasma Plasma inflammatory cytokines and chemokines in severe acute respiratory syndrome Comparative and kinetic analysis of viral shedding and immunological responses in MERS patients representing a broad spectrum of disease severity Distinct immune response in two MERS-CoV-infected patients: can we go from bench to bedside? The cellular kinetics of lung alveolar epithelial cells and its relationship with lung tissue repair after acute lung injury The role of placenta growth factor in the hyperoxia-induced acute lung injury in an animal model Leukemia inhibitory factor signaling is required for lung protection during pneumonia Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China Pathological findings of COVID-19 associated with acute respiratory distress syndrome Interleukin-1 blockade with high-dose anakinra in patients with COVID-19, acute respiratory distress syndrome, and hyperinflammation: a retrospective cohort study The trinity of COVID-19: immunity, inflammation and intervention COVID-19: combining antiviral and anti-inflammatory treatments Discovery and Genomic Characterization of a 382-Nucleotide Deletion in ORF7b and ORF8 during the Early Evolution of SARS-CoV-2 A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t Duration of symptoms before admission 6 (2 to 9) 5 (2 to 9) 6 (4 to 9) 0. A c c e p t e d M a n u s c r i p t