key: cord-1025875-ddk2xobb
authors: Tufa, Abdisa; Gebremariam, Tewodros Haile; Manyazewal, Tsegahun; Getinet, Tewodros; Webb, Dominic-Luc; Hellström, Per M.; Genet, Solomon
title: Cytokine and chemokine profile in patients hospitalized with COVID-19: A comparative study
date: 2022-03-18
journal: bioRxiv
DOI: 10.1101/2022.03.17.484837
sha: 82f0e09c27f65dbe6d3205d84c048cddd352b692
doc_id: 1025875
cord_uid: ddk2xobb

Abnormal cytokine and chemokine concentrations during SARS-CoV-2 infection may represent disease severity. We aimed to assess plasma cytokine and chemokine concentrations in patients with SARS-CoV-2 in Addis Ababa, Ethiopia. In this study, 260 adults: 126 hospitalized patients with confirmed COVID-19 sorted into severity groups: severe (n=68) and mild or moderate (n=58), and 134 healthy controls were enrolled. We quantified 39 plasma cytokines and chemokines using multiplex ELISA. Spearman rank correlation and Mann-Whitney U test were used to identify mechanistically coupled cytokines/chemokines and compare disease severity. Compared to healthy controls, patients with COVID-19 had significantly higher levels of interleukins 1α, 2, 6, 7, 8, 10 and 15, C-reactive protein (CRP), serum amyloid A (SAA), intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion protein 1 (VCAM-1), IFN-γ-inducible protein-10 (IP-10), macrophage inflammatory protein-1 alpha (MIP-1α), eotaxin-3, interferon-gamma (IFN-ϒ), tumor necrosis factor-α (TNF-α), basic fibroblast growth factor (bFGF), placental growth factor (PlGF), and fms-like tyrosine kinase 1 (Flt-1). Patients with severe COVID-19 had higher IL-10 and lower macrophage-derived chemokine (MDC) compared to the mild or moderate group (P<0.05). In the receiver operating characteristic curve, SAA, IL-6 and CRP showed strong sensitivity and specificity predicting the severity and prognosis of COVID-19. Greater age and higher CRP had a significant association with disease severity (P<0.05). Our findings reveal that CRP, SAA, VCAM-1, IP-10, MDC and IL-10 levels are promising biomarkers for COVID-19 disease severity, suggesting that plasma cytokines/chemokines could be used as warning indicators of COVID-19 severity, aid in COVID-19 prognosis and treatment. IMPORTANCE SARS-CoV-2 triggers inflammatory reaction resulting in respiratory discomfort and in critical case may result in death. Cytokines and chemokines are inflammatory biomarkers that regulate and determine the nature of immune responses. Measuring cytokine and chemokine levels is useful in stratification, management and treatment of COVID-19 patients as well as guide resource allocations and therapeutic options. Here, we examined ctytokine and chemokine profiles in COVID-19 patients. Understanding how distinct cytokines and chemokines change over time as COVID-19 disease progresses might aid clinicians in detecting severe illness earlier and thereby improve patient prognosis.

. The median age of mild or moderate (32 years of age, IQR 20-78) versus severe (60 87 years of age, IQR 22-86) groups was significant (Table 1 ). There were 64 COVID-19 patients 88 (50.8 %) with either single or multiple morbidities such as diabetes (16 %), hypertension (14 %), 89 cardiovascular disease (CVD) (17 %), cancer (17 %) and chronic lung disease (10 %). In the 90 severe group, the proportion of diabetes and chronic lung disease were considerably higher than 91 in the mild or moderate group (Table 2) . 92 Cytokines and chemokines plasma concentrations 93 Circulating cytokine and chemokine concentrations are summarized in table 3. Circulating 94 concentrations of 19 cytokines or chemokines (interleukins 1ɑ, 2, 6, 7, 8, 10 and 15, CRP, SAA, 95 ICAM-1, VCAM-1, IP-10, MIP-1ɑ, Eotaxin-3, IFN-Ƴ, TNF-ɑ, bFGF, PlGF and Flt-1) in 96 COVID-19 patients were increased compared to the healthy controls (P < 0.05, Table 3 ). Even 97 after age was adjusted ( comparing age < 40 years in both COVID-19 patients and healthy 98 control group), the majority of the median levels of cytokines and chemokines in the COVID-19 99 patients were greater than the healthy control group and statistically significant (P < 0.05, Table   100 5). 101 Severe COVID-19 patients had significantly higher levels of CRP, ICAM-1, SAA, VCAM-1, 102 IP-10, IL-10, IL-15, IL-16, IL-7, Flt-1 and VEGF-D than mild or moderate COVID-19 patients. 103 MDC, IL-12/IL-23p40, IL-17A and TNF-β levels, on the other hand, were greater in mild or 104 moderate COVID-19 patients than in severe COVID-19 patients (P < 0.05, Table 4 ). Between 105 COVID-19 patients and healthy controls, median chemokine concentrations (MCP-1 and MIP-1β) 106 were not statistically significant (P > 0.05, Table 3 ). 107 Based on the correlations of 30 significantly elevated cytokines and chemokines; CRP, SAA, 108 IL-6, IL-8, IP-10, TNF-ɑ, IL-10, VEGF-D, MDC and VCAM-1 were the ones that discriminated 109 between mild or moderate and severe cases ( Table 6 ). The diagnostic value of cytokines and 110 chemokines for illness severity was assessed using the receiver operating characteristic (ROC) 111 curve and the area under the ROC curve (AUC) ( Table 7 and Table 7) . TNF-ɑ and 117 VEGF-D have poor sensitivity and specificity when compared to CRP, SAA and IL-6 ( Figure 1 ). 118 The hierarchy among the 10 selected cytokines and chemokines in distinguishing between 119 healthy controls and COVID-19 patients, as well as between mild or moderate and severe 120 COVID-19 patients, was validated by ROC and AUC analyses (Table 7 & 8) . Figure 2   121 demonstrates the use of a ROC curve to distinguish between mild/moderate and severe 122 COVID-19 cases for severity prediction. 123 Table 9 shows a binary logistic regression analysis that links CRP and age to COVID-19 124 disease severity. These findings indicated that CRP and age were statistically significant (P < 125 0.05).

This study looks at 39 cytokines and chemokines in COVID-19 patients' inflammation and 128 disease severity. SARS-COV-2 infection can cause varied inflammatory response abnormalities. 129 The pathogenesis of many viral infections, including coronaviruses, is connected to aberrant 7 The average age of the severe group was much higher than mild/moderate group, in line with 138 earlier research (17). This showed that age could be a risk factor for COVID-19 severity and it 139 was assumed that ACE2 density was shown to be positively connected with age (18 ; 19). 140 Gender distribution in the group was not statistically significant (Table 1) . However, male 141 patients were more likely to develop severe illness in the severe group than in the other groups. 142 This could be due to gender linked immunological differences (20). 143 Pro-inflammatory cytokines and interferons (IFNs) were shown to be greater in COVID-19 144 patients (21). These cytokines aid in the elimination of infections as well as the maintenance of 145 cellular homeostasis. Dysregulated pro-inflammatory cytokine release, on the other hand, 146 contributes to cytokine storm, a potentially fatal condition caused by inflammatory cells' 147 excessive cytokine production (22). Inflammatory cytokine and chemokine levels were found to 148 be significantly higher in patients than in the healthy controls group. This study consistent with a 149 prior study comparing patients with severe COVID-19 to those with mild disease (23). In 150 comparison to mild/moderate cases of COVID-19, plasma from patients with severe cases 151 showed a greater tendency for levels of IL-6, IL-8 and TNF-ɑ indicating a pro-inflammatory 152 response. This research supports the findings of the prior research (24). 153 C-reactive protein (CRP) is increased in response to IL-6, IL-1β and TNF-ɑ activation during 154 infection and systemic inflammation (25). CRP levels were significantly greater in the severe 155 group compared to the other groups in this investigation, which is consistent with a prior study 156 that found CRP levels may represent disease severity in 27) . CRP is a CRP is linked to a poor . It can also be used as an early indicator of 161 infection and inflammation. According to the current study findings and previous researches, 162 increased levels of CRP have clinical diagnostic and prognostic value during infection. 164 Endothelial cells aid in the recruitment of leukocytes from the circulation to infection and 165 inflammatory sites (30). Inflammatory cytokines initiate different kinase cascades during 166 SARS-CoV-2 infection, leading to the activation of transcription factors like NF-β, which in turn 167 stimulate adhesion molecules like ICAM-1 and VCAM-1. ICAM-1 promotes leukocyte arrest 168 and firm adhesion, as well as monocyte and lymphocyte transmigration. In the current study, 169 COVID-19 patients had higher levels of ICAM-1 and VCAM-1 than healthy control and the 170 severe group had higher levels than the mild or moderate group, which was consistent with 171 earlier findings (31). This suggests that ICAM-1 and VCAM-1 could be used as a biomarker to 172 predict the severity of COVID-19 disease and may have a role in coagulation dysfunction. 173 In line with the previous study, SAA levels in the current study were considerably higher in the 174 COVID-19 patients than healthy controls. SAA is released by hepatocyte cells in response to 175 inflammatory cytokines such as IL-6, TNF-ɑ and IL-1β during an acute phase response. One of 176 the most common clinical symptoms of severe COVID-19 is ARDS (32). There were patients 177 with ARDS who had a considerably higher level of SAA. SAA could be used as a biomarker to 178 track the evolution of respiratory disorders like COVID-19, according to our findings. The 179 median levels of both CRP and SAA were greater in COVID-19 patients than healthy controls 180 after age adjustment (comparing age < 40 years in both COVID-19 patients and healthy controls). 181 CRP and SAA were both linked to disease severity; have a strong and positive linear correlation. 182 Since measurement of different cytokines/chemokines is time-consuming and costly, correlation analysis aids clinical prediction of one biomarker in terms of another. Thus, correlations were 184 performed in the current investigation, implying that plasma levels of CRP, SAA, IL-6, IP-10, 185 VCAM-1, and IL-10, among the linked, exhibit a significant and beneficial association with 186 COVID-19 patient progression. CRP, SAA, IL-6, IP-10, VCAM-1, TNF-ɑ and IL-10 all had a 187 negative relationship with plasma levels of In this study, in individuals with COVID-19, IP-10 levels were higher than in healthy controls 189 and the same is true after age adjustment. Furthermore, when the viral infection was linked with 190 a pulmonary pathology ( e.g., influenza and SARS-COV-2), IP-10 levels were higher than when 191 the infection was associated with a non-pulmonary pathology ( e.g., human rhinovirus) (13).

IP-10 levels were significantly greater in severe SARS-CoV-2-positive patients than in mild or 193 moderate patients, implying that IP-10 could be a useful biomarker for distinguishing between 194 mild/moderate and severe COVID-19 patients and predicting disease severity. 195 In SARS, Middle East respiratory syndrome-CoV and SARS-CoV-2 infected patients, excessive 196 cytokine production has been associated to pulmonary inflammation and acute lung damage (15, 197 33). IL-10 is an anti-inflammatory cytokine that was shown to be higher in with severe illness in this study.This was in line with the findings of a previous study (23). It is 199 important for immune system homeostasis. IL-10 is a cytokine with many functions that reduces initially attributed to a negative feedback mechanism which is suggestive of anti-inflammatory 204 response (36). Furthermore, as compared to healthy controls and mild/moderate COVID-19 205 patients, plasma IL-7 levels were considerably higher in severe COVID-19 patients in this study. 206 This research supports the findings of the prior research (37). T-cell growth and functions are 207 dependent on IL-7. Increased levels of IL-7 in the blood have been linked to a reduction in the 208 number of T cells in the body (38). As a feedback response to lymphopenia, serum IL-7 209 concentration was also negatively related to the number of T cells, CD4 + and CD8 + cells, 210 highlighting the link between lymphopenia and higher IL-7 levels in SARS-CoV-2 infection 211 (39). 212 In this study, mild/ moderate COVID-19 cases had considerably greater levels of IL-12/IL-23p40 213 than severe COVID-19 patients and healthy controls. This is in line with the findings of the prior 214 study (40). Induction of IL-12 is essential to sustain NK cell counts during the early stages of 215 SARS-CoV-2 infection and this induction could aid in evasion from virus spreading. The number 216 of peripheral NK cells in patients with severe COVID-19 was much lower than in healthy people 217 (41). Furthermore, IL-17A is a homeostatic proinflammatory cytokine that is also involved in the 218 development of autoimmune disorders. Our findings revealed that IL-17A levels in severe 219 COVID-19 patients were significantly lower than in mild or moderate COVID-19 patients. A 220 meta-analysis comparing IL-17A levels in severe and moderate patients found that severe 221 patients had greater levels (42). The discrepancy could be due to the data limitations that were 222 revealed during the meta analysis.

pathways; neutrophils infiltrating the lungs in large numbers and macrophages migrating to the 230 lungs to phagocytize apoptotic bronchial epithelial cells, pneumocytes, T-lymphocytes and 231 neutrophils (44). The potential of this cytokine to recruit more neutrophils and remodel the 232 airways through modulating processes used by the virus to generate pulmonary fibrosis explains 233 its influence in SARS-CoV-2 infection. 234 The severe group had significantly greater levels of VEGF-D than the mild or moderate group. 235 There was a data shortage for this biomarker, but it was consistent with the prior research, which 236 compared critical and severe group (45). We hypothesized that a high level of VEGF-D is linked 237 to a storm of blood clots in COVID-19 patients, which eventually leads to disease severity. 238 The ROC curve data further revealed that CRP, SAA and IL-6 levels had excellent sensitivity 239 and specificity for COVID-19 severity. This biomarker's likely hood ratio value also accurately 240 predicted illness dynamics, making it easier to recognize and act with underlying severe patients 241 earlier, which was critical for lowering mortality. Binary logistic regression analysis found odds 242 ratio (OR=1.056) for age, implying that the chance of being severe increases by 1.056 times with 243 each unit rise in age. As people become older, their chance of becoming severely ill will increase. 244 The same is true for CRP, with an OR of 1.004; one unit CRP increase has a 1.004 times 245 increased likelihood of being severe. Increases in age and CRP will enhance the likelihood of 246 severity, according to the regression model. These data suggest that both increment of CRP and 247 age has a role in the advancement of COVID-19 disease severity. 248 The limitation of this study might be that it was impossible to get a comparable group because 249 there were no elders in the healthy controls group (for age > 40 years ), though adjustments were 250 tried. Because blood was tapped and severity assessed shortly after positive test results, often at 251 an outpatient setting, the study could not identify any future disease progression after the date of Bold p value emphasizes 0.05 criteria met. Analytes were: Basic fibroblast growth factor (bFGF); 513 C-reactive protein (CRP); eosinophil chemotactic proteins (Eotaxin, Eotaxin- Bold p value emphasizes 0.05 criteria met. Cut off value units for CRP, SAA and VCAM-1 were 539 (mg/L) and IL-6, IL-8, IP-10, TNF-ɑ, IL-10, VEGF-D and MDC were (ng/L). (mg/L) and IL-6, IL-8, IP-10, TNF-ɑ, IL-10, VEGF-D and MDC were (ng/L). 

Bold p value emphasizes 0.05 criteria met. IL-10 R-value P-value