Gamma’ fibrinogen levels as a biomarker of COVID-19 respiratory disease severity

Coronavirus disease 2019 (COVID-19) is characterized by a pro-inflammatory state associated with organ failure, thrombosis, and death. We investigated a novel inflammatory biomarker, γ’ fibrinogen (GPF), in 103 hospitalized patients with COVID-19 and 19 healthy controls. We found significant associations between GPF levels and the severity of COVID-19 as judged by blood oxygen saturation (SpO2). The mean level of GPF in the patients with COVID-19 was significantly higher than in controls (69.8 (95% CI 64.8–74.8) mg/dL compared with 36.9 (95% CI 31.4–42.4) mg/dL, p < 0.0001), whereas C-reactive protein (CRP), lactate dehydrogenase (LDH), and total fibrinogen levels were not significantly different between groups. Mean GPF levels were significantly highest in patients with severe COVID-19 (SpO2 ≤ 93%, GPF 75.2 (95% CI 68.7–81.8) mg/dL), compared to mild/moderate COVID-19 (SpO2 >93%, GPF 62.5 (95% CI 55.0–70.0) mg/dL, p = 0.01, AUC of 0.68, 95% CI 0.57–0.78; Youden’s index cutpoint 62.9 mg/dL, sensitivity 0.64, specificity 0.63). In contrast, CRP interleukin-6, ferritin, LDH, D-dimers, and total fibrinogen had weaker associations with COVID-19 disease severity (all ROC curves with lower AUCs). Thus, GPF may be a useful inflammatory marker of COVID-19 respiratory disease severity.


Introduction
Coronavirus disease 2019 (COVID-19) is a viral respiratory disease that results in high levels of proin ammatory and pro-thrombotic biomarkers that are associated with worse outcomes, including death 1 .
Elevated levels of C-reactive protein (CRP), interleukin-6 (IL-6), ferritin, lactate dehydrogenase (LDH), Ddimer, and brinogen have suggested various potential mechanisms in the pathogenesis of poor outcomes in COVID-19 2 . For example, elevated levels of CRP, ferritin, and brinogen indicate activation of the acute phase response, while elevated circulating D-dimer levels suggest that brinolytic pathways are likely intact and dissolving brin 3 . Despite this, brin deposits in lungs and other organs suggests dysregulation of balance in brin-forming (i.e., thrombin generation) and dissolving (i.e., plasmin generation) pathways, and autopsies of patients with COVID-19 show the presence of thrombosis in up to 60% of the patients 4 .
γ' brinogen (GPF) is a novel in ammatory biomarker that has been associated with cardiovascular 5 and in ammatory 6 diseases. We have previously shown that extraordinarily high levels of GPF in a small population of patients with COVID-19 were associated with the need for extracorporeal membrane oxygenation and death 7 . We therefore tested the hypothesis that GPF levels can stratify COVID-19 respiratory disease severity in a larger population of patients.

Study Participants
This study was approved by the Institutional Ethical Committee at the Micro Therapeutics Laboratory, Chennai, India and Saveetha Medical College, Chennai, India. All participants provided written informed consent.

Patients with COVID-19
Patients were recruited from Saveetha Hospital. Inclusion criteria were hospitalized patients aged 18-65 years and laboratory con rmation of infection with SARS-CoV-2 by positive reverse transcription polymerase chain reaction (RT-PCR). Exclusion criteria were pregnant or lactating women, participation in any interventional drug clinical study during the study, or a known in ammatory condition. Patients were enrolled between June 14, 2021 and February 21, 2022. Blood samples were obtained and clinical characteristics were determined at admission.

Healthy controls
Inclusion criteria were healthy volunteers age 18-65 years with laboratory con rmation of non-infection with SARS-CoV-2 by negative RT-PCR. Exclusion criteria were known bleeding disorders, liver or kidney disease, cancer, history of surgery or thrombotic event within the past 3 months, previous history of viral infection and other diseases, or participation in any other study.

Laboratory Testing
Subjects had 2 ml of whole blood collected in 3.2% sodium citrate at admission, which was processed into plasma and banked at -80°C. Ferritin, D-dimer, and IL-6 were measured using chemiluminescence (Maglumi), CRP using tubidimetrics (Turbilatex), LDH using a latex reagent (Mindray), and total brinogen using a clot-based auto chemistry analyzer (Beacon Diagnostics PVT LTD). Cycle threshold (Ct) values for (RT-PCR) were obtained from the Micro Therapeutics Laboratory and Saveetha Medical College. GPF was analyzed using a commercial enzyme-linked immunoassay (Gamma Coeur ELISA Kit -GCEK001, Zeus Scienti c).

Statistical Analysis
Severity of COVID-19 was de ned as mild/moderate (SpO 2 > 93%) or severe (SpO 2 ≤ 93%) on room air measured at admission 8 . Differences in GPF levels among patients with mild/moderate or severe COVID-19 were analyzed using ANOVA with multiple comparisons (insert statistical package here -I used Prism 9, GraphPad Software, LLC). We examined the relationship between COVID-19 severity and other biomarkers using a Wilcoxon rank sum test for skewed biomarkers (IL-6, LDH, D-dimer, ferritin, and CRP), and a t-test for normally-distributed biomarkers (total brinogen). These methods were applied to examine the differences between patients with COVID-19 and controls as well. Prior to running t-tests, an F-test for the equality of variances was run. If unequal variances were found, unequal variance t-tests were run. Otherwise, equal variance t-tests were used. ROC curves were used to determine the ability of GPF to differentiate between patients with COVID-19 and controls, as well as between patients with severe COVID-19 and mild/moderate COVID-19. Area under the curve (AUC) and 95% con dence intervals (CI) were calculated. For each of these curves, Youden's index was calculated to determine the GPF cutpoint that maximized sensitivity and speci city. AUC and CI were also calculated for the other biomarkers. Within patients with COVID-19, correlation between GPF and other biomarkers was calculated using Spearman's rho due to the skewed distribution of most of the biomarkers. All statistical analyses were performed using Stata 16 (StataCorp, College Stations, TX).

Discussion
We found that GPF levels were increased signi cantly over controls in hospitalized patients with COVID-19, and had stronger associations with severity of COVID-19 respiratory disease compared to several other in ammatory biomarkers, including CRP, IL-6, ferritin, LDH, D-dimer, and total brinogen. Hence these ndings have potential clinical implications suggesting that elevated GPF levels may be useful as a biomarker for respiratory disease severity in COVID-19. Furthermore, GPF levels did not signi cantly correlate with IL-6, ferritin, D-dimer, or total brinogen. These ndings suggest that the associations between COVID-19 and COVID-19 severity with GPF levels cannot simply be due to total brinogen levels.
These ndings have potential clinical implications regarding prophylactic anticoagulation of COVID-19 patients. The resistance of GPF-bound thrombin (Fig. 4) to heparin suggests that heparin prophylaxis may be less effective than treatment with other anticoagulants, particularly direct thrombin inhibitors. In point of fact, in critically ill patients with COVID-19, therapeutic-dose heparin did not result in a greater probability of survival to hospital discharge or a greater number of days free of cardiovascular or respiratory organ support 9 . It is possible that inhibition of factor Xa by current DOACs may also reduce the levels of active thrombin and thereby prevent activation of thrombin substrates by GPF-bound thrombin, including factor V, factor VIII, factor XI, factor XIII, and brinogen, as well as platelet substrates such as PAR-1 and PAR-4. In addition, it is also possible that warfarin anticoagulation may be effective at preventing thrombosis due to GPF-bound thrombin by reducing the levels of active vitamin K-dependent coagulation factors, and therefore, generation of thrombin.
Our study has limitations, in that this was a small observational study without outcome data. It is known that comorbidities including heart disease, hypertension, and diabetes in patients with COVID-19 increase the risk for severe disease and mortality [10][11][12][13] . These could confound the identi ed association of COVID-19 severity with levels of GPF. GPF (and indeed the other markers) was measured by a single assay, and thus the ndings may not hold true for other assay methods. GPF assays are not available in most labs, unlike the other assays, but a commercial ELISA is now available.
In conclusion, our study suggests that GPF may be a valuable biomarker for assessing COVID-19 respiratory disease severity. Evidence is lacking as to the causal mechanisms or whether the associations are speci c to effects of COVID-19 infection or the consequences of a systemic in ammatory response. However, recent in vitro evidence demonstrates that high GPF levels in patient plasma increase clot formation at both arterial and venous shear conditions in vitro 14 . Future investigations of GPF as a driver of thromboin ammation and poor outcomes in COVID-19, and as a biomarker for other in ammatory diseases, both infectious and non-infectious, should be pursued. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.   Pairwise comparisons between biomarkers in healthy controls, mild/moderate COVID-19 patients, and severe COVID-19 patients. We examined the relationship between COVID-19 severity and other biomarkers using a Wilcoxon rank sum test for skewed biomarkers (GPF, IL-6, LDH, D-dimer, ferritin, and CRP), and a ttest for normally-distributed biomarkers (total brinogen).

Figure 4
Model of GPF mechanism of thrombosis. The model starts with free thrombin cleaving brinopeptides A and B from all forms of brinogen via its active site (cutout triangle). This converts the brinogen to brin, which polymerizes into an insoluble clot. The free thrombin can then bind to GPF via thrombin anion-binding exosite II (cutout square), concentrating active thrombin on the growing brin clot. This