key: cord-0771075-91baxc76
authors: Liu, C.; Wang, Q.; Wang, Y.; Wang, G.; Wang, L.; Chen, H.; Jiao, T.; Hu, C.; Lei, X.; Guo, L.; Ren, L.; Li, M.; Zeng, X.; Zhang, D.; Cao, B.; Wang, J.
title: Augmentation of anti-MDA5 antibody implies severe disease in COVID-19 patients
date: 2020-08-01
journal: nan
DOI: 10.1101/2020.07.29.20164780
sha: 61d312c1621266e40efaca9fa27d172c8316176e
doc_id: 771075
cord_uid: 91baxc76

Recent studies have provided insights into the autoinflammation triggered by severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) infection, which is associated with high mortality of coronavirus disease 2019 (COVID-19). Striking similarities has been noted between COVID-19 and anti-melanoma differentiation-associated gene 5 (MDA5) antibody (Ab)-related dermatomyositis (DM), implying a shared autoinflammatory aberrance. However, it is unclear whether anti-MDA5 Ab is present in COVID-19 and correlates with the severity and adverse outcome of COVID-19 patients. Here, we found that the positive rate of anti-MDA5 Ab in patients with COVID-19 was 48.2% and the anti-MDA5 Ab positive patients tended to develop severe disease (88.6% vs 66.9%, P<0.0001). In particular, the titer of anti-MDA5 Ab was increased in the non-survivals (5.95{+/-}5.16 vs 8.22{+/-}6.64, P=0.030) and the positive rate was also higher than that in the survivals (23.5% vs 12.0%, P=0.012). Regarding to severe COVID-19 patients, we found that high titer of anti-MDA5 Ab ([≥]10.0 U/mL) was more prevalent in the non-survivals (31.2% vs 14.0%, P=0.006). Moreover, early profiling of anti-MDA5 Ab could distinguish severe patients from those with non-severe ones. Overall, our data reveal that anti-MDA5 Ab is prevalent in the COVID-19 patients and high titer of this antibody is correlated with severe disease and unfavorable outcomes.

7 respiratory distress syndrome (ARDS), Innate immunity.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020 . . https://doi.org/10.1101 /2020 Coronavirus Disease 2019 , caused by highly contagious severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), has become a pandemic involving more than 12 million cases globally by July 2020. 1 The average mortality is estimated to be 1%, 2 but can raise up to 62% in critically ill patients, mostly due to acute respiratory distress syndrome (ARDS). 3 Therefore, reducing mortality of severe COVID-19 patients has become an urgent task in this battle. Unfortunately, few antiviral agents have been proved to be effective enough to treat the disease, and whether to use corticosteroids and other immunomodulatory drugs remains controversial. 4 Accumulating evidence has demonstrated that a bunch of cytokines (e.g. IL-2, IL-6, IL-7, IL-10, G-CSF, IP10, MCP1, TNFα, etc.) are elevated in the blood of patients with severe COVID-19, resembling cytokine storm syndrome. The COVID-19 patients with hypercytokinemia and hyperferritinemia were more likely to exhibit extensive lung damage and ARDS. [5] [6] [7] In addition, high prevalence of antinuclear antibodies (35.6%) and lupus anticoagulant (46.6%), along with antineutrophil cytoplasmic antibodies (6.6%) and anti-Ro antibody (4.4%), were also identified in hospitalized patients with Thus, hypothesis that SARS-CoV2 might trigger autoimmune and/or autoinflammatory aberrance in genetically predisposed subjects has been raised. 9 It is reasonable that all patients with severe COVID-19 should be screened for hyperinflammation to identify the subgroup of patients for whom adjunctive immunosuppression therapy would improve mortality. 4 Striking similarities has been noted between multifaceted features of COVID-19 and a rare autoimmune disease, the . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020 . . https://doi.org/10.1101 /2020 anti-melanoma-differentiation-associated gene 5 (MDA5) antibody (Ab)-related dermatomyositis (DM). 10, 11 Both diseases can involve the lungs, skin 12,13 and muscles. 14 Anti-MDA5 Ab-related DM patients often develop rapid progressive interstitial lung disease (RP-ILD), whose prognosis is disappointingly poor with more than half of patients dying within 3 months of disease onset. 15 The initial radiological features of lung in Anti-MDA5 Ab-related DM patients are mainly subpleural ground-glass opacities or mixed with consolidation and signs of ARDS, which resemble severe and critical 16 Besides, similar skin eruptions have been reported in both diseases. 13 Furthermore, serum cytokine profiles are also similar in these two conditions, as serum levels of ferritin, IL-6, IL-8, and IL-10 usually were elevated in patients with RP-ILD secondary to Anti-MDA5 Ab-related DM. 17 The similarity of these two diseases implies a shared underlying autoinflammatory/autoimmune mechanisms. To date, there is no report on whether anti-MDA5 Ab also exists in patients with COVID-19.

MDA5 is a well-known cytoplasmic sensor for viral RNA and its expression is induced by RNA viruses. This activates the expression of antiviral type I and III interferons (IFNs) with inflammatory cytokines. Correspondingly, IFN signaling can induce the expression of MDA5. SARS-CoV2 infection has been reported to trigger the expression of MDA5. 18, 19 In addition, MDA5 is involved in pathogenesis of several autoimmune disorders as well, 20 such as systemic lupus erythematosus 21, 22 , multiple sclerosis 23 , and even type 1 diabetes 24,25 .

Nevertheless, it remains unclear whether the anti-MDA5 Ab plays a role in the pathophysiology of COVID-19 or whether it correlates with the disease severity.

Some researchers have called for screening the anti-MDA5 Ab in severe . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 1, 2020. .

COVID-19 patients. 10, 11 In the present study, we report the presence of anti-MDA5 Ab in patients with SARS-CoV2 infection and address its correlation with the clinical severity and outcomes of COVID-19 by measuring the anti-MDA5 Ab.

A total of 274 patients were included in this study ( Figure 1a ; . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 1, 2020. .

To determine the presence of anti-MDA5 Ab in COVID-19 patients, ELISA

analysis was employed to test the plasma collected from COVID-19 patients.

We found that the titer of anti-MDA5 Ab was increased in these examined samples as compared to normal controls (1.85±0.67 vs 6.60±5.50, P<0.0001) included in the ELISA were also confirmed by Western blots (Figure 1e ). Next, a total of 17 plasma samples of COVID-19 were conducted Western blotting analysis, which included five non-severe and 12 severe COVID-19 patients.

These data showed that the anti-MDA5 Ab were detected in these examined samples as well (Figures 1f and 1g ).

Altogether, our findings indicate that SARS-CoV2 infection leads to an increased anti-MDA5 Ab titer in patients with COVID-19.

COVID-19 patients with positive anti-MDA5 Ab tend to exhibit severe disease.

Anti-MDA5 Ab is first identified in DM and correlated with the status of this . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 1, 2020 . . https://doi.org/10.1101 /2020 disease, which promoted us to further investigate whether the titer of anti-MDA5 Ab was related to the clinical severity of patients with COVID-19. To this end, the confirmed COVID-19 patients were stratified into two groups based on the cutoff value, that is, the anti-MDA5 Ab negative (<5.0 U/mL) and positive (≥5.0 U/mL). We found that the percentage of severe COVID-19 patients was much higher in anti-MDA5 Ab positive group than that in negative (Table 2) .

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. .

Taken together, our findings suggest that anti-MDA5 Ab is positively correlated with the clinical severity of COVID-19 patients.

The correlation between anti-MDA5 Ab and COVID-19 outcomes.

As mentioned above, COVID-19 patients with positive anti-MDA5 Ab tended to develop severe disease. The titer of anti-MDA5 Ab was higher in severe COVID-19 patients as compared to the non-severe diseases ( Figure 3a ). The positive rate of this antibody was also higher in COVID-19 patients with severe performance (Figure 3b ). We also observed that the titer of anti-MDA5 Ab depicted a significant increase in COVID-19 patients suffering from the chronic comorbidities, for instance, hypertension, diabetes, and cardiovascular disease ( Figure 3c ). An augment of this antibody was noticed in the plasma samples of COVID-19 patients suffering from shock, respiratory or other organ failure ( Figure 3d ).

We next investigated whether the titer of anti-MDA5 Ab was correlated with the outcome of COVID-19 patients. To this end, a further comparison of anti-MDA5

Ab was employed in the survival COVID-19 patients and the non-survivals, which indicated that the titer of anti-MDA5 Ab was upregulated in dead COVID-19 patients as compared to the survival ( Figure 3e ). Its positive rate was lower in the survivals with COVID-19 than that in the non-survivals ( Figure   3f ). These data suggested that anti-MDA5 Ab had the potential to serve as a prognostic factor for COVID-19. Consistent with published predictive factors for COVID-19 outcomes, we found that the levels of LDH, ferritin, and CRP were significantly decreased in the non-survivals as compared to that in the . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. .

survivals, and the number of lymphocytes was also markedly reduced in the non-survivals ( Figure 3g ; Table 1) .

We Since the alteration of anti-MDA5 Ab titer is correlated with the activity and outcome of DM, we asked whether the change of anti-MDA5 Ab was associated with the clinical features of COVID-19. To this end, a cross-sectional analysis was employed using the titer of anti-MDA5 Ab . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . We further determined the titer of anti-MDA5 Ab in sequential samples from severe COVID-19 disease as shown in Figure 4a . The titer of anti-MDA5 Ab in patient #1, #2, and #3 depicted a similar alteration compared with that in the . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. .

cross-sectional analysis (Figures 4h and 4i ). Next, the titer of anti-MDA5 Ab was examined in the samples collected at the WFSO-2, that is, the days following symptoms onset (DFSO) 8-14. We found that the titer and positive rate of anti-MDA5 Ab remained substantial (Figures 4j, 4k , and 4l;

Supplementary Figure 2 ).

Collectively, our data indicate that COVID-19 patients with high titer of anti-MDA5 Ab initially tend to develop severe disease.

The present study, for the first time, identified and confirmed the prevalence of anti-MDA5 Ab in patients with COVID-19 by both ELISA and Western blots. We also demonstrated that the positive rate and titer of anti-MDA5 Ab was associated with the clinical severity and outcomes of COVID-19. In severe COVID-19 patients, we found that high titer of anti-MDA5 Ab (≥10.0 U/mL) was more prevalent in non-survival patients. Moreover, early profile of anti-MDA5

Ab could distinguish severe patients from non-severe ones. Our study provides the evidence that early screening of anti-MDA5 Ab might help identify high risk population and predict the outcome of patients with COVID-19.

MDA5 is a intracellular RNA-specific helicase that belongs to a family of retinoic acid-inducible gene I-like receptors. 26 It is triggered to oligomerize upon binding RNA via its helicase domain, which transmits a signal through its homotypic-interacting caspase recruitment domain (CARD) to induce oligomerization of the adaptor mitochondrial antiviral signaling (MAVS) protein and then activates the production of type I and III IFNs, which plays a key role . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. .

in anti-viral immune response. 27,28 Thus, MDA5 is a crucial antiviral factor and has been previously reported to involve in SARS-CoV, MERS-CoV and SARS-CoV2 infections. 18, 29, 30 Interestingly, MDA5 is also involved in several autoimmune disorders such as anti-MDA5 Ab-related DM. Therefore, it is not surprising that COVID-19 and anti-MDA5 Ab-related DM share similar features of hyperinflammation and multi-systemic manifestations, especially rapid progressive interstitial lung disease that results in ARDS and death. In this study, we determined anti-MDA5 Ab in as many as 48.2% patients with COVID-19.

Our study revealed a positive correlation between the anti-MDA5 Ab and the severity of COVID-19, and high titer of anti-MDA5 Ab was associated with higher mortality in severe COVID-19 patients. Similar observation was reported in anti-MDA5 Ab-related DM patients. 31 However, the titer of this antibody is even higher in anti-MDA5 Ab-related DM than that in COVID-19.

This may indicate that high titer of anti-MDA5 Ab probably is related to an uncontrolled autoinflammation and autoimmune response to SARS-CoV2 infection in genetically predisposed hosts. Furthermore, our study also demonstrated that elder age, chronic comorbidities, lymphocytopenia, hypoalbuminemia, hyperferritinemia, increased D-dimer and CRP levels were more prevalent in COVID-19 patients with organ dysfunction and the mortality was comparatively high, which has been reported in previous studies and implies a dysregulation of inflamation. [5] [6] [7] [32] [33] [34] It has been reported that the change of anti-MDA5 Ab titer correlates with disease activity and predicts treatment response and disease outcome in . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . There are several limitations in our study. Firstly, since MDA5 is validated as a general sensor for diverse RNA viruses, no evidence has addressed whether . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. Overall, we, for the first time, revealed that anti-MDA5 Ab is present in patients with COVID-19 and correlates with severe disease and poor outcomes. Early screening and serially monitoring of anti-MDA5 Ab titer has the potential to predict the disease progression of COVID-19.

This retrospective study included three cohorts of adult patients (≥18 years old) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. 

We extracted demographic, clinical, laboratory, treatment, and outcome data from medical and nursing records using standardized data collection forms (a revised version of case record form for severe acute respiratory infection shared by WHO and the International Severe Acute Respiratory and Emerging Infection Consortium). All data were checked by two investigators (QW and CL) and a third researcher (BC) adjudicated any difference in interpretation between the two primary reviewers.

According to the clinical classification of COVID-19 by the Protocol for Prevention and Control of cases of COVID-19 (Edition 7), 39 we divided the . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164780 doi: medRxiv preprint 1 patients into two groups on hospitalization: (1) Severe group, the patients fulfilled the diagnostic criteria of severe and critical cases, who meets any one of follows: i) Respiratory rate > 30/min, ii) Pulse oxygen saturation < 93%, iii)

Oxygenation index < 300 mmHg, or iv) respiratory failure or other organ dysfunction requiring transmission to intensive care unit. (2) Non-severe group, the patients' severity was mild or moderate that didn't meet the above criteria. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. 

For the detection of anti-MDA5 Ab, each experiment was repeated 3 times. 

We thank Dr. Jinmin Peng and Jingjing Bai for providing assistance with . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. 

All data were checked by two investigators (QW and CL) and a third researcher (BC)

All the authors have no disclosures. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. .

Autoimmun Rev 19, 102524, doi:10.1016/j.autrev.2020.102524 (2020 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . -2673 - , doi:10.1158 - /0008-5472.CAN-16-2339 - (2017 . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020 . . https://doi.org/10.1101 /2020 determined by using unpaired, two-sided Mann-Whitney U-test and X2 test.

P<0.05, *; P<0.01, **; P<0.001, ***; P<0.0001, ****. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. χ ² test comparing all subcategories.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164780 doi: medRxiv preprint . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 1, 2020. .

Brain natriuretic peptide, pg/mL (n=135)

Brain natriuretic peptide, pg/mL (n=135)