Post-COVID Trajectory of Pentraxin 3 Plasma Levels Over 6 Months and Their Association with the Risk of Developing Post-Acute Depression and Anxiety.

BACKGROUND AND OBJECTIVES: Clinical manifestations of coronavirus disease 2019 (COVID-19) often persist after acute disease resolution. Underlying molecular mechanisms are unclear. The objective of this original article was to longitudinally measure plasma levels of markers of the innate immune response to investigate whether they associate with and predict post-COVID symptomatology. METHODS: Adult patients with previous severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection during the first pandemic wave who underwent the 6-month multidisciplinary follow-up were included. Plasma levels of pentraxin 3 (PTX3), the complement components C3a and C5a, and chitinase-3 like-protein-1 (CHI3L1) were measured at hospital admission during acute disease (baseline) and at 1 and 6 months after hospital discharge. Associations with post-COVID-19 sequelae at 6 months were investigated using descriptive statistic and multiple regression models. RESULTS: Ninety-four COVID-19 patients were included. Baseline PTX3, C5a, C3a, and CHI3L1 did not predict post-COVID-19 sequelae. The extent of the reduction of PTX3 over time (delta PTX3) was associated with lower depressive and anxiety symptoms at 6 months (both p < 0.05). When entering sex, age, need for intensive care unit or non-invasive ventilation during hospital stay, psychiatric history, and baseline PTX3 as nuisance covariates into a generalized linear model (GLM), the difference between baseline and 6-month PTX3 levels (delta PTX3) significantly predicted depression (χ2 = 4.66, p = 0.031) and anxiety (χ2 = 4.68, p = 0.031) at 6 months. No differences in PTX3 levels or PTX3 delta were found in patients with or without persistent or new-onset other COVID-19 symptoms or signs at 6 months. Plasma levels of C3a, C5a, and CHI3L1 did not correlate with PTX3 levels at either time point and failed to associate with residual or de novo respiratory or systemic clinical manifestations of the disease at 6 months. CONCLUSIONS: A lower reduction of plasma PTX3 after acute COVID-19 associates with the presence of depression and anxiety, suggesting an involvement of inflammation in post-COVID-19 psychopathology and a potential role of PTX3 as a biomarker.

Pentraxin 3 in Myocarditis: Proof-of-Principle Assessment as a Diagnostic and Prognostic Biomarker.

Pentraxin 3 (PTX3) is an acute phase protein produced in various tissues in response to microbial and sterile stimuli, which regulates the inflammation outcomes. PTX3 has not been investigated in myocarditis. Our aim was to assess circulating and cardiac tissue expression of PTX3 in 55 patients with myocarditis proven by magnetic resonance and/or endomyocardial biopsy. A major proportion of patients with myocarditis displayed significantly increased plasma PTX3 levels as compared with controls (26/30 vs. 0/10), with higher diagnostic yield than conventional biomarkers in the study group. Cardiac tissue analysis revealed PTX3 expression in all patients (40/40), with viral myocarditis exhibiting higher signal intensity than autoimmune myocarditis, and with a predominant localization in cardiomyocytes. Abnormal plasma PTX3 was associated with systolic dysfunction and heart failure at presentation. Interestingly, patients who recovered by 12 months had higher baseline PTX3 levels. Our preliminary data support the potential use of PTX3 as a biomarker in myocarditis.

Histone-Specific CD4+ T Cell Plasticity in Active and Quiescent Systemic Lupus Erythematosus.

OBJECTIVE: The aim of this study was to assess whether circulating histone-specific T cells represent tools for precision medicine in systemic lupus erythematosus (SLE). METHODS: Seroprevalence of autoantibodies and HLA-DR beta (DRB) 1 profile were assessed among 185 patients with SLE and combined with bioinformatics and literature evidence to identify HLA-peptide autoepitope couples for ex vivo detection of antigen-specific T cells through flow cytometry. T cell differentiation and polarization was investigated in patients with SLE, patients with Takayasu arteritis, and healthy controls carrying HLA-DRB1*03:01 and/or HLA-DRB1*11:01. SLE Disease Activity Index 2000 and Lupus Low Disease Activity State were used to estimate disease activity and remission. RESULTS: Histone-specific CD4+ T cells were selectively detected in patients with SLE. Among patients with a history of anti-DNA antibodies, 77% had detectable histone-specific T cells, whereas 50% had lymphocytes releasing cytokines or upregulating activation markers after in vitro challenge with histone peptide antigens. Histone-specific regulatory and effector T helper (Th) 1-, Th2-, and atypical Th1/Th17 (Th1*)-polarized cells were significantly more abundant in patients with SLE with quiescent disease. In contrast, total Th1-, Th2-, and Th1*-polarized and regulatory T cells were similarly represented between patients and controls or patients with SLE with active versus quiescent disease. Histone-specific effector memory T cells accumulated in the blood of patients with quiescent SLE, whereas total effector memory T cell counts did not change. Immunosuppressants were associated with expanded CD4+ histone-specific naive T (TN) and terminally differentiated T cells. CONCLUSION: Histone-specific T cells are selectively detected in patients with SLE, and their concentration in the blood varies with disease activity, suggesting that they represent innovative tools for patient stratification and therapy.

Choroid plexus and perivascular space enlargement in neuropsychiatric systemic lupus erythematosus.

Choroid plexus (CP) enlargement is proposed as a marker of neuroinflammation in immune-mediated conditions. CP involvement has also been hypothesized in the immunopathology of systemic lupus erythematosus (SLE). We investigated whether CP enlargement occurs in SLE patients and its association with neuropsychiatric involvement. Additionally, we explored abnormalities along the glymphatic system in SLE patients through enlarged perivascular space (PVS) quantification. Clinical assessment and 3 Tesla brain dual-echo and T1-weighted MRI scans were obtained from 32 SLE patients and 32 sex and age-matched healthy controls (HC). CPs were manually segmented on 3D T1-weighted sequence and enlarged PVS (ePVS) were assessed through Potter's score. Compared to HC, SLE patients showed higher normalized CP volume (nCPV) (p = 0.023), with higher CP enlargement in neuropsychiatric SLE (NPSLE) (n = 12) vs. non-NPSLE (p = 0.027) patients. SLE patients with antiphospholipid antibodies (APA) positivity (n = 18) had higher nCPV compared to HC (p = 0.012), while APA negative ones did not. SLE patients also had higher Potter's score than HC (p < 0.001), with a tendency towards a higher number of basal ganglia ePVS in NPSLE vs. non-NPSLE patients. Using a random forest analysis, nCPV emerged as a significant predictor of NPSLE, together with T2-hyperintense white matter (WM) lesion volume (LV) and APA positivity (out-of-bag AUC 0.81). Our findings support the hypothesis of a role exerted by the CP in SLE physiopathology, especially in patients with neuropsychiatric involvement. The higher prevalence of ePVS in SLE patients, compared to HC, suggests the presence of glymphatic system impairment in this population.

Association between Perceived Health-Related Quality of Life and Depression with Frailty in the FRASNET Study.

Frailty is a major challenge facing the aging world. The phenotype of the frail subject is still far from being satisfactorily defined. We report data on mood, cognition, and quality of life (QoL) in relation to anamnestic factors, health, and socio-economic status in the FRASNET geriatric population (1204 subjects in stable health conditions), which is an observational cohort study that includes fairly balanced groups of Italian frail (421, 35%), pre-frail (449, 37.3%) and robust (334, 27.7%) subjects. A conditional inference tree analysis revealed a substantial influence of psychological variables on frailty. The physical indicator of QoL (Short Form Survey-36-Physical Component Summary, SF-36-PCS) was the predominant variable in the full model (threshold at 39.9, p < 0.001): higher frailty was found in subjects with a caregiver and lower SF-36-PCS. Frailty was also associated with the mental indicator of QoL (Short Form Survey-36-Mental Component Summary, SF-36-MCS), depression (Geriatric Depression Scale, GDS-15), leisure activities, and level of education. In support of the prominent role of inflammation in aging and mental illness, the SF-36-PCS score was correlated with the blood concentration of C-X-C motif chemokine ligand 10 (CXCL10) (r Pearson -0.355, p = 0.015), a critical signal in cell senescence and inflammaging, while the rs7567647 variant in FN1 gene encoding a glycoprotein in the extracellular matrix was significantly associated with frailty in a multivariable model (p = 0.0006). The perception of health-related QoL and subclinical depression contribute to frailty. Their assessment could improve the identification of older patients at increased risk of adverse outcomes.

Chitinase-3-like protein-1 at hospital admission predicts COVID-19 outcome: a prospective cohort study.

Infectious and inflammatory stimuli elicit the generation of chitinase-3-like protein-1 (CHI3L1), involved in tissue damage, repair and remodeling. We evaluated whether plasma CHI3L1 at disease onset predicts clinical outcome of patients with Coronavirus 2019 (COVID-19) disease. Blood from 191 prospectively followed COVID-19 patients were collected at hospital admission between March 18th and May 5th, 2020. Plasma from 80 survivors was collected one month post-discharge. Forty age- and sex-matched healthy volunteers served as controls. Primary outcome was transfer to intensive care unit (ICU) or death. CHI3L1 was higher in COVID-19 patients than controls (p < 0.0001). Patients with unfavorable outcome (41 patients admitted to ICU, 47 died) had significantly higher CHI3L1 levels than non-ICU survivors (p < 0.0001). CHI3L1 levels abated in survivors one month post-discharge, regardless of initial disease severity (p < 0.0001), although remaining higher than controls (p < 0.05). Cox regression analysis revealed that CHI3L1 levels predict primary outcome independently of age, sex, comorbidities, degree of respiratory insufficiency and systemic inflammation or time from symptom onset to sampling (p < 0.0001). Kaplan-Meier curve analysis confirmed that patients with CHI3L1 levels above the median (361 ng/mL) had a poorer prognosis (log rank test, p < 0.0001). Plasma CHI3L1 is increased in COVID-19 patients and predicts adverse outcome.

Chromogranin A plasma levels predict mortality in COVID-19.

BACKGROUND: Chromogranin A (CgA) and its fragment vasostatin I (VS-I) are secreted in the blood by endocrine/neuroendocrine cells and regulate stress responses. Their involvement in Coronavirus 2019 disease (COVID-19) has not been investigated. METHODS: CgA and VS-I plasma concentrations were measured at hospital admission from March to May 2020 in 190 patients. 40 age- and sex-matched healthy volunteers served as controls. CgA and VS-I levels relationship with demographics, comorbidities and disease severity was assessed through Mann Whitney U test or Spearman correlation test. Cox regression analysis and Kaplan Meier survival curves were performed to investigate the impact of the CgA and VS-I levels on in-hospital mortality. RESULTS: Median CgA and VS-I levels were higher in patients than in healthy controls (CgA: 0.558 nM [interquartile range, IQR 0.358-1.046] vs 0.368 nM [IQR 0.288-0.490] respectively, p = 0.0017; VS-I: 0.357 nM [IQR 0.196-0.465] vs 0.144 nM [0.144-0.156] respectively, p<0.0001). Concentration of CgA, but not of VS-I, significantly increased in patients who died (n = 47) than in survivors (n = 143) (median 0.948 nM [IQR 0.514-1.754] vs 0.507 nM [IQR 0.343-0.785], p = 0.00026). Levels of CgA were independent predictors of in-hospital mortality (hazard ratio 1.28 [95% confidence interval 1.077-1.522], p = 0.005) when adjusted for age, number of comorbidities, respiratory insufficiency degree, C-reactive protein levels and time from symptom onset to sampling. Kaplan Meier curves revealed a significantly increased mortality rate in patients with CgA levels above 0.558 nM (median value, log rank test, p = 0.001). CONCLUSION: Plasma CgA levels increase in COVID-19 patients and represent an early independent predictor of mortality.

Biobanking for COVID-19 research.

BACKGROUND: Biobanks are imperative infrastructures, particularly during outbreaks, when there is an obligation to acquire and share knowledge as quick as possible to allow for implementation of science-based preventive, diagnostic, prognostic, and therapeutic strategies. METHODS: We established a COVID-19 biobank with the aim of collecting high-quality and well-annotated human biospecimens, in the effort to understand the pathogenic mechanisms underlying COVID-19 and identify therapeutic targets (COVID-BioB, NCT04318366). Here we describe our experience and briefly review the characteristics of the biobanks for COVID-19 that have been so far established. RESULTS: A total of 46,677 samples have been collected from 913 participants (63.3% males, median [IQR] age 62.2 [51.2-74.0] years) since the beginning of the program. Most patients (66.9%) had been admitted to hospital for COVID-19, with a median length of stay of 15.0 (9.0-27.0) days. A minority of patients (13.3% of the total) had been admitted for other reasons and subsequently tested positive for SARS-CoV-2. The remainder were managed at home after being seen at the Emergency Department. CONCLUSIONS: Having a solid research infrastructure already in place, along with flexibility and adaptability to new requirements, allowed for the quick building of a COVID-19 biobank that will help expand and share the knowledge of SARS-CoV-2.

Unconventional CD147-dependent platelet activation elicited by SARS-CoV-2 in COVID-19.

BACKGROUND: Platelet activation and thrombotic events characterizes COVID-19. OBJECTIVES: To characterize platelet activation and determine if SARS-CoV-2 induces platelet activation. PATIENTS/METHODS: We investigated platelet activation in 119 COVID-19 patients at admission in a university hospital in Milan, Italy, between March 18 and May 5, 2020. Sixty-nine subjects (36 healthy donors, 26 patients with coronary artery disease, coronary artery disease, and seven patients with sepsis) served as controls. RESULTS: COVID-19 patients had activated platelets, as assessed by the expression and distribution of HMGB1 and von Willebrand factor, and by the accumulation of platelet-derived (plt) extracellular vesicles (EVs) and HMGB1+ plt-EVs in the plasma. P-selectin upregulation was not detectable on the platelet surface in a fraction of patients (55%) and the concentration of soluble P-selectin in the plasma was conversely increased. The plasma concentration of HMGB1+ plt-EVs of patients at hospital admission remained in a multivariate analysis an independent predictor of the clinical outcome, as assessed using a 6-point ordinal scale (from 1 = discharged to 6 = death). Platelets interacting in vitro with SARS-CoV-2 underwent activation, which was replicated using SARS-CoV-2 pseudo-viral particles and purified recombinant SARS-CoV-2 spike protein S1 subunits. Human platelets express CD147, a putative coreceptor for SARS-CoV-2, and Spike-dependent platelet activation, aggregation and granule release, release of soluble P-selectin and HMGB1+ plt-EVs abated in the presence of anti-CD147 antibodies. CONCLUSIONS: Hence, an early and intense platelet activation, which is reproduced by stimulating platelets in vitro with SARS-CoV-2, characterizes COVID-19 and could contribute to the inflammatory and hemostatic manifestations of the disease.

Platelet Phagocytosis via P-selectin Glycoprotein Ligand 1 and Accumulation of Microparticles in Systemic Sclerosis.

OBJECTIVE: It is unclear why activated platelets and platelet-derived microparticles (MPs) accumulate in the blood of patients with systemic sclerosis (SSc). This study was undertaken to investigate whether defective phagocytosis might contribute to MP accumulation in the blood of patients with SSc. METHODS: Blood samples were obtained from a total of 81 subjects, including 25 patients with SSc and 26 patients with stable coronary artery disease (CAD). Thirty sex- and age-matched healthy volunteers served as controls. Studies were also conducted in NSG mice, in which the tail vein of the mice was injected with MPs, and samples of the lung parenchyma were obtained for analysis of the pulmonary microvasculature. Tissue samples from human subjects and from mice were assessed by flow cytometry and immunochemical analyses for determination of platelet-neutrophil interactions, phagocytosis, levels and distribution of P-selectin, P-selectin glycoprotein ligand 1 (PSGL-1), and HMGB1 on platelets and MPs, and concentration of byproducts of neutrophil extracellular trap (NET) generation/catabolism. RESULTS: Activated P-selectin+ platelets and platelet-derived HMGB1+ MPs accumulated in the blood of SSc patients but not in the blood of healthy controls. Patients with CAD, a vasculopathy independent of systemic inflammation, had fewer P-selectin+ platelets and a negligible number of MPs. The expression of the receptor for P-selectin, PSGL-1, in neutrophils from SSc patients was significantly decreased, raising the possibility that phagocytes in SSc do not recognize activated platelets, leading to a failure of phagocytosis and continued neutrophil release of MPs. As evidence of this process, activated platelets were not detected in the neutrophils from SSc patients, whereas they were consistently present in the neutrophils from patients with CAD. HMGB1+ MPs elicited generation of NETs, which were only detected in the plasma of SSc patients. In mice, P-selectin-PSGL-1 interaction resulted in platelet phagocytosis in vitro and influenced the ability of MPs to elicit NETs, endothelial activation, and migration of leukocytes through the pulmonary microvasculature. CONCLUSION: The clearance of activated platelets via PSGL-1 limits the undesirable effects of MP-elicited neutrophil activation. This balance is disrupted in patients with SSc. Its reconstitution might curb vascular inflammation and prevent fibrosis.

CXCL10 levels at hospital admission predict COVID-19 outcome: hierarchical assessment of 53 putative inflammatory biomarkers in an observational study.

BACKGROUND: Host inflammation contributes to determine whether SARS-CoV-2 infection causes mild or life-threatening disease. Tools are needed for early risk assessment. METHODS: We studied in 111 COVID-19 patients prospectively followed at a single reference Hospital fifty-three potential biomarkers including alarmins, cytokines, adipocytokines and growth factors, humoral innate immune and neuroendocrine molecules and regulators of iron metabolism. Biomarkers at hospital admission together with age, degree of hypoxia, neutrophil to lymphocyte ratio (NLR), lactate dehydrogenase (LDH), C-reactive protein (CRP) and creatinine were analysed within a data-driven approach to classify patients with respect to survival and ICU outcomes. Classification and regression tree (CART) models were used to identify prognostic biomarkers. RESULTS: Among the fifty-three potential biomarkers, the classification tree analysis selected CXCL10 at hospital admission, in combination with NLR and time from onset, as the best predictor of ICU transfer (AUC [95% CI] = 0.8374 [0.6233-0.8435]), while it was selected alone to predict death (AUC [95% CI] = 0.7334 [0.7547-0.9201]). CXCL10 concentration abated in COVID-19 survivors after healing and discharge from the hospital. CONCLUSIONS: CXCL10 results from a data-driven analysis, that accounts for presence of confounding factors, as the most robust predictive biomarker of patient outcome in COVID-19.

Adiponectin to leptin ratio reflects inflammatory burden and survival in COVID-19.

AIM: Obesity is a risk factor for COVID-19, but the underlying mechanisms are unclear. We investigated the role of adiponectin (an anti-inflammatory adipokine), leptin (a pro-inflammatory adipokine) and their ratio (Adpn/Lep) in this context. DESIGN: Single-centre, prospective observational study. METHODS: Adiponectin and leptin were measured in 60 COVID-19 patients with mild (not hospitalised, n=11), moderate (hospitalised but not requiring intensive care, n=25) and severe (admission to the intensive care unit [ICU] or death, n=24) disease. RESULTS: Adiponectin and leptin levels were similar across severity groups, but patients with moderate severity had the highest Adpn/Lep ratio (1.2 [0.5; 2.0], 5.0 [1.6; 11.2], 2.1 [1.0; 3.6] in mild, moderate and severe disease; P = 0.019). Adpn/Lep, but not adiponectin or leptin alone, correlated with systemic inflammation (C reactive protein, CRP: Spearman's rho 0.293, P = 0.023). When dividing patients into Adpn/Lep tertiles, adiponectin was highest, whereas leptin was lowest in the third (highest) tertile. Patients in the highest Adpn/Lep tertile had numerically lower rates of obesity, diabetes and hypertension, and lower rates of death or admission to ICU versus other tertiles. At linear regression in the whole cohort, CRP significantly predicted Adpn/Lep (ß 0.291, P = 0.022), while female gender (ß -0.289, P = 0.016), diabetes (ß -0.257, P = 0.028), and hypertension (ß -239, P = 0.043) were negative predictors. CONCLUSIONS: We speculate that the rise in Adpn/Lep, due to increased adiponectin and reduced leptin, is a compensatory response to systemic inflammation. In patients with worse cardiometabolic health (e.g. diabetes, hypertension) this mechanism might be blunted, possibly contributing to higher mortality.

Blood neurofilament light chain and total tau levels at admission predict death in COVID-19 patients.

BACKGROUND AND AIMS: Patients infected with SARS-CoV-2 range from asymptomatic, to mild, moderate or severe disease evolution including fatal outcome. Thus, early predictors of clinical outcome are highly needed. We investigated markers of neural tissue damage as a possible early sign of multisystem involvement to assess their clinical prognostic value on survival or transfer to intensive care unit (ICU). METHODS: We collected blood from 104 patients infected with SARS-CoV-2 the day of admission to the emergency room and measured blood neurofilament light chair (NfL), glial fibrillary acidic protein (GFAP), ubiquitin carboxy-terminal hydrolase L1 (UCH-L1), and total tau protein levels. RESULTS: We found that NfL, GFAP, and tau were significantly increased in patients with fatal outcome, while NfL and UCH-L1 in those needing ICU transfer. ROC and Kaplan-Meier curves indicated that total tau levels at admission accurately predict mortality. CONCLUSIONS: Blood neural markers may provide additional prognostic value to conventional biomarkers used to predict COVID-19 outcome.

Patients with COVID-19: in the dark-NETs of neutrophils.

SARS-CoV-2 infection poses a major threat to the lungs and multiple other organs, occasionally causing death. Until effective vaccines are developed to curb the pandemic, it is paramount to define the mechanisms and develop protective therapies to prevent organ dysfunction in patients with COVID-19. Individuals that develop severe manifestations have signs of dysregulated innate and adaptive immune responses. Emerging evidence implicates neutrophils and the disbalance between neutrophil extracellular trap (NET) formation and degradation plays a central role in the pathophysiology of inflammation, coagulopathy, organ damage, and immunothrombosis that characterize severe cases of COVID-19. Here, we discuss the evidence supporting a role for NETs in COVID-19 manifestations and present putative mechanisms, by which NETs promote tissue injury and immunothrombosis. We present therapeutic strategies, which have been successful in the treatment of immunο-inflammatory disorders and which target dysregulated NET formation or degradation, as potential approaches that may benefit patients with severe COVID-19.

Mer tyrosine kinase as a possible link between resolution of inflammation and tissue fibrosis in IgG4-related disease.

OBJECTIVES: IgG4-related disease (IgG4-RD) is a systemic fibro-inflammatory disorder characterized by a dysregulated resolution of inflammation and wound healing response that might develop after an apoptotic insult induced by cytotoxic T lymphocytes (CTLs). Mer receptor tyrosine kinase (MerTK) and its ligand, protein S (ProS1), have a pivotal role in the resolution of inflammation, being implicated in the clearance of apoptotic cells, quenching of the immune response and development of tissue fibrosis. In the present work we aimed to investigate a possible involvement of the MerTK signalling pathway in the pathogenesis of IgG4-RD and development of tissue fibrosis. METHODS: MerTK and ProS1 expression patterns in IgG4-RD lesions were evaluated by immunohistochemistry and immunofluorescence studies. Circulating MerTK+ monocytes, soluble Mer and MerTK ligands were measured in the peripheral blood of IgG4-RD patients and healthy controls by flow cytometry and ELISA, respectively. RESULTS: MerTK was highly expressed by macrophages infiltrating IgG4-RD lesions. MerTK+ macrophages were more abundant in IgG4-RD than in Sjögren's syndrome and interacted with apoptotic cells and ProS1-expressing T and B lymphocytes. Moreover, they expressed the pro-fibrotic cytokine TGF-ß and their numbers declined following rituximab-induced disease remission. Circulating MerTK+ monocytes, soluble Mer and MerTK ligands were not increased in the peripheral blood of patients with IgG4-RD. CONCLUSIONS: The MerTK-ProS1 axis is activated in IgG4-RD lesions, possibly leading to persistent stimulation of processes involved in the resolution of inflammation and tissue fibrosis.

Resting state network functional connectivity abnormalities in systemic lupus erythematosus: correlations with neuropsychiatric impairment.

Neuropsychiatric manifestations are highly prevalent in systemic lupus erythematosus (SLE)-patients. We aimed to unravel the substrates of these manifestations by investigating abnormalities of resting state (RS) functional connectivity (FC) and their correlations with neuropsychiatric variables in SLE-patients. Thirty-two SLE-patients and 32 age- and sex-matched healthy controls (HC) underwent brain 3T RS fMRI. Neuropsychological assessment was performed for all SLE-patients. The main large-scale cognitive and psychiatric functional networks were derived and between-group comparisons and correlations with neuropsychological measures were performed. Compared to HC, SLE-patients exhibited increased RS FC in the right middle cingulate cortex and decreased RS FC in the left precuneus within default-mode network (DMN). They also showed increased RS FC in the left cerebellar crus I and left posterior cingulate cortex, and decreased RS FC in the left angular gyrus within working-memory networks (WMN). Compared to HC, SLE-patients exhibited increased RS FC in the left insular cortex and decreased RS FC in the right anterior cingulate cortex within salience network (SN), as well as decreased RS FC in the right middle frontal gyrus within executive-control network (ECN). Correlation analysis indicated a maladaptive role for left angular gyrus and cerebellar RS FC abnormalities in WMN, affecting memory and executive functions; and for precuneus and insular abnormalities in DMN and SN for psychiatric symptoms. Cingulate cortex modifications within DMN and SN correlated with better memory and global cognitive performance. Significant RS FC alterations in relevant cognitive and psychiatric networks occur in SLE-patients and participate in the pathophysiology of neuropsychiatric symptoms.

Deep Learning on Conventional Magnetic Resonance Imaging Improves the Diagnosis of Multiple Sclerosis Mimics.

OBJECTIVES: The aims of this study were to present a deep learning approach for the automated classification of multiple sclerosis and its mimics and compare model performance with that of 2 expert neuroradiologists. MATERIALS AND METHODS: A total of 268 T2-weighted and T1-weighted brain magnetic resonance imagin scans were retrospectively collected from patients with migraine (n = 56), multiple sclerosis (n = 70), neuromyelitis optica spectrum disorders (n = 91), and central nervous system vasculitis (n = 51). The neural network architecture, trained on 178 scans, was based on a cascade of 4 three-dimensional convolutional layers, followed by a fully dense layer after feature extraction. The ability of the final algorithm to correctly classify the diseases in an independent test set of 90 scans was compared with that of the neuroradiologists. RESULTS: The interrater agreement was 84.9% (Cohen κ = 0.78, P < 0.001). In the test set, deep learning and expert raters reached the highest diagnostic accuracy in multiple sclerosis (98.8% vs 72.8%, P < 0.001, for rater 1; and 81.8%, P < 0.001, for rater 2) and the lowest in neuromyelitis optica spectrum disorders (88.6% vs 4.4%, P < 0.001, for both raters), whereas they achieved intermediate values for migraine (92.2% vs 53%, P = 0.03, for rater 1; and 64.8%, P = 0.01, for rater 2) and vasculitis (92.1% vs 54.6%, P = 0.3, for rater 1; and 45.5%, P = 0.2, for rater 2). The overall performance of the automated method exceeded that of expert raters, with the worst misdiagnosis when discriminating between neuromyelitis optica spectrum disorders and vasculitis or migraine. CONCLUSIONS: A neural network performed better than expert raters in terms of accuracy in classifying white matter disorders from magnetic resonance imaging and may help in their diagnostic work-up.

Quantitative MRI adds to neuropsychiatric lupus diagnostics.

OBJECTIVE: Attributing neuropsychiatric manifestations to SLE is often challenging. Brain white matter lesions are frequent in SLE at MRI, but their diagnostic role is unclear. Here, we assessed whether white matter lesions count, volume and distribution measurement can help in the diagnosis of neuropsychiatric systemic lupus erythematosus (NPSLE). METHODS: Brain dual-echo and 3D T1-weighted sequences were acquired from 32 patients with SLE and 32 healthy controls with a 3 T-scanner and employed to derive T2-hyperintense lesion volume (T2LV), number (T2LN) and probability maps (LPM) using a semi-automatic local thresholding segmentation technique. NPSLE was classified as per the ACR nomenclature, the Italian Society for Rheumatology algorithm and by clinical impression. Clinical descriptors including the SLE International Collaborating Clinics/ACR damage index (SDI) were also recorded. RESULTS: Higher T2LV were observed in SLE vs healthy controls (P < 0.001) and in NPSLE vs other SLE (P =0.006). Patients with NPSLE also had higher T2LN (P =0.003) compared with other SLE. In SLE, T2LPM revealed a high prevalence of lesions in the splenium of the corpus callosum, right superior longitudinal fasciculus and right corona radiata. T2LV and T2LN correlated with SLE duration (rho = 0.606; P <0.001 and rho = 0.483; P =0.005, respectively) and age (rho = 0.478; P =0.006 and rho = 0.362; P = 0.042, respectively). T2LV also correlated with SDI (rho = 0.352; P =0.048). SLE patients with fatigue had lower T2LN (P =0.038) compared with patients without fatigue. Thresholds of T2LV ≥ 0.423 cm3 or of T2LN ≥ 12 were associated with definite NPSLE and improved the classification of patients with possible NPSLE per clinical impression. CONCLUSION: Brain white matter lesions (WML) quantitation adds to NPSLE diagnostics.