Quantitative susceptibility mapping at 7 Tesla in COVID-19: mechanistic and outcome associations (preprint)

Post mortem studies have shown that patients dying from severe SARS-CoV-2 infection frequently have pathological changes in their central nervous system, particularly in the brainstem. Many of these changes are proposed to result from para-infectious and/or post-infection immune responses. Clinical symptoms such as fatigue, breathlessness, and chest pain are frequently reported in post-hospitalized COVID-19 patients. We propose that these symptoms are in part due to damage to key neuromodulatory brainstem nuclei. While brainstem involvement has been demonstrated in the acute phase of the illness, the evidence of long-term brainstem change on magnetic resonance imaging (MRI) is inconclusive. We therefore used ultra-high field (7T) quantitative susceptibility mapping (QSM) to test the hypothesis that brainstem abnormalities persist in post-COVID patients and that these are associated with persistence of key symptoms. We used 7T QSM data from 30 patients, scanned 93 - 548 days after hospital admission for COVID-19 and compared them to 51 age-matched controls without prior history of COVID-19 infection. We correlated the patients QSM signals with disease severity (duration of hospital admission and COVID-19 severity scale), inflammatory response during the acute illness (C-reactive protein, D-Dimer and platelet levels), functional recovery (modified Rankin scale; mRS), depression (PHQ-9) and anxiety (GAD-7). In COVID-19 survivors the MR susceptibility increased in the medulla, pons and midbrain regions of the brainstem. Specifically, there was increased susceptibility in the inferior medullary reticular formation and the raphe pallidus and obscurus. In these regions, patients with higher tissue susceptibility had worse acute disease severity, higher acute inflammatory markers, and significantly worse functional recovery. Using non-invasive ultra-high field 7T MRI, we show evidence of brainstem pathophysiological changes associated with inflammatory processes in post-hospitalized COVID-19 survivors. This study contributes to understanding the mechanisms of long-term effects of COVID-19 and recovery.

Para-infectious brain injury in COVID-19 persists at follow-up despite attenuated cytokine and autoantibody responses (preprint)

We measured brain injury markers, inflammatory mediators, and autoantibodies in 203 participants with COVID-19; 111 provided acute sera (1-11 days post admission) and 56 with COVID-19-associated neurological diagnoses provided subacute/convalescent sera (6-76 weeks post-admission). Compared to 60 controls, brain injury biomarkers (Tau, GFAP, NfL, UCH-L1) were increased in acute sera, significantly more so for NfL and UCH-L1, in patients with altered consciousness. Tau and NfL remained elevated in convalescent sera, particularly following cerebrovascular and neuroinflammatory disorders. Acutely, inflammatory mediators (including IL-6, IL-12p40, HGF, M-CSF, CCL2, and IL-1RA) were higher in participants with altered consciousness, and correlated with brain injury biomarker levels. Inflammatory mediators were lower than acute levels in convalescent sera, but levels of CCL2, CCL7, IL-1RA, IL-2R, M-CSF, SCF, IL-16 and IL-18 in individual participants correlated with Tau levels even at this late time point. When compared to acute COVID-19 patients with a normal GCS, network analysis showed significantly altered immune responses in patients with acute alteration of consciousness, and in convalescent patients who had suffered an acute neurological complication. The frequency and range of autoantibodies did not associate with neurological disorders. However, autoantibodies against specific antigens were more frequent in patients with altered consciousness in the acute phase (including MYL7, UCH-L1, GRIN3B, and DDR2), and in patients with neurological complications in the convalescent phase (including MYL7, GNRHR, and HLA antigens). In a novel low-inoculum mouse model of SARS-CoV-2, while viral replication was only consistently seen in mouse lungs, inflammatory responses were seen in both brain and lungs, with significant increases in CCL4, IFN{gamma}, IL-17A, and microglial reactivity in the brain. Neurological injury is common in the acute phase and persists late after COVID-19, and may be driven by a para-infectious process involving a dysregulated host response.

Hospitalisation for COVID-19 predicts long lasting cerebrovascular impairment: A prospective observational cohort study (preprint)

Human coronavirus disease 2019 (COVID-19) due to severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has multiple neurological consequences, but its long-term effect on brain health is still uncertain. The cerebrovascular consequences of COVID-19 may also affect brain health. Here we assess cerebrovascular health in 45 hospitalised patients using the resting state fluctuation amplitudes (RSFA) from functional magnetic resonance imaging, in relation to disease severity and in contrast with 42 controls. Widespread changes in frontoparietal RSFA were related to the severity of the acute COVID-19 episode, as indexed by COVID-19 WHO Progression Scale, inflammatory and coagulatory biomarkers. This relationship was not explained by chronic cardiorespiratory dysfunction, age, or sex. Exploratory analysis suggests that the level of cerebrovascular dysfunction is associated with cognitive, mental, and physical health at follow-up. The principal findings were consistent across univariate and multivariate approaches. The results indicate chronic cerebrovascular impairment following severe acute COVID-19, with the potential for long-term consequences on cognitive function and mental wellbeing.

Brain Injury in COVID-19 is Associated with Autoinflammation and Autoimmunity (preprint)

COVID-19 has been associated with many neurological complications including stroke, delirium and encephalitis. Furthermore, many individuals experience a protracted post-viral syndrome which is dominated by neuropsychiatric symptoms, and is seemingly unrelated to COVID-19 severity. The true frequency and underlying mechanisms of neurological injury are unknown, but exaggerated host inflammatory responses appear to be a key driver of severe COVID-19 more broadly. We sought to investigate the dynamics of, and relationship between, serum markers of brain injury (neurofilament light [NfL], Glial Fibrillary Acidic Protein [GFAP] and total Tau) and markers of dysregulated host response including measures of autoinflammation (proinflammatory cytokines) and autoimmunity. Brain injury biomarkers were measured using the Quanterix Simoa HDx platform, cytokine profiling by Luminex (R&D) and autoantibodies by a custom protein microarray. During hospitalisation, patients with COVID-19 demonstrated elevations of NfL and GFAP in a severity-dependant manner, and there was evidence of ongoing active brain injury at follow-up 4 months later. Raised NfL and GFAP were associated with both elevations of pro-inflammatory cytokines and the presence of autoantibodies; autoantibodies were commonly seen against lung surfactant proteins as well as brain proteins such as myelin associated glycoprotein, but reactivity was seen to a large number of different antigens. Furthermore, a distinct process characterised by elevation of serum total Tau was seen in patients at follow-up, which appeared to be independent of initial disease severity and was not associated with dysregulated immune responses in the same manner as NfL and GFAP.

Proteome-wide Mendelian randomization identifies causal links between blood proteins and severe COVID-19 (preprint)

The COVID-19 pandemic death toll now surpasses two million individuals and there is a need for early identification of individuals at increased risk of mortality. Host genetic variation partially drives the immune and biochemical responses to COVID-19 that lead to risk of mortality. We identify and prioritise blood proteins and biomarkers that may indicate increased risk for severe COVID-19, via a proteome Mendelian randomization approach by collecting genome-wide association study (GWAS) summary statistics for >4,000 blood proteins. After multiple testing correction, troponin I3, cardiac type (TNNI3) had the strongest effect (odds ratio (O.R.) of 6.86 per standard deviation increase in protein level), with proteinase 3 (PRTN3) (O.R.=2.48), major histocompatibility complex, class II, DQ alpha 2 (HLA-DQA2) (O.R.=2.29), the C4A-C4B heterodimer (O.R.=1.76) and low-density lipoprotein receptor-related protein associated protein 1 (LRPAP1) (O.R.=1.73) also being associated with higher odds of severe COVID-19. Conversely, major histocompatibility complex class I polypeptide-related sequence A (MHC1A) (O.R.=0.6) and natural cytotoxicity triggering receptor 3 (NCR3) (O.R.=0.46) were associated with lower odds. These proteins are involved in heart muscle contraction, natural killer and antigen presenting cells, and the major histocompatibility complex. Based on these findings, it may be possible to better predict which patients may develop severe COVID-19 and to design better treatments targeting the implicated mechanisms.

Virological and serological characterization of critically ill patient with COVID-19 in the UK: a special focus on variant detection (preprint)

Background. Treatment of COVID-19 patients with convalescent plasma containing neutralising antibody to SARS-CoV-2 is under investigation as a means of reducing viral loads, ameliorating disease outcomes, and reducing mortality. However, its efficacy might be reduced in those infected with the emerging B.1.1.7 SARS-CoV-2 variant. Here, we report the diverse virological characteristics of UK patients enrolled in the Immunoglobulin Domain of the REMAP-CAP randomised controlled trial. Methods. SARS-CoV-2 viral RNA was detected and quantified by real-time PCR in nasopharyngeal swabs obtained from study subjects within 48 hours of admission to intensive care unit. Antibody status was determined by spike-protein ELISA. B.1.1.7 strain was differentiated from other SARS-CoV-2 strains by two novel typing methods detecting the B.1.1.7-associated D1118H mutation with allele-specific probes and by restriction site polymorphism (SfcI). Findings. Of 1260 subjects, 90% were PCR-positive with viral loads in nasopharyngeal swabs ranging from 72 international units [IUs]/ml to 1.7x10^11 IU/ml. Median viral loads were 45-fold higher in those who were seronegative for IgG antibodies (n=314; 28%) compared to seropositives (n=804; 72%), reflecting in part the latter group's possible later disease stage on enrolment. Frequencies of B.1.1.7 infection increased from early November (<1%) to December 2020 (>60%). Anti-SARS-CoV-2 seronegative individuals infected with wild-type SARS-CoV-2 had significantly higher viral loads than seropositives (medians of 1.2x10^6 and 3.4 x10^4 IU/ml respectively; p=2x10^-9). However, viral load distributions were elevated in both seropositive and seronegative subjects infected with B.1.1.7 (13.4x10^6 and 7.6x10^6 IU/ml; p=0.18). Interpretation. High viral loads in seropositive B.1.1.7-infected subjects are consistent with increased replication capacity and/or less effective clearance by innate or adaptive immune response of B.1.1.7 strain than wild-type. As viral genotype was associated with diverse virological and immunological phenotypes, metrics of viral load, antibody status and infecting strain should be used to define subgroups for analysis of treatment efficacy.