Myxovirus resistance protein A in peripheral blood predicts supplemental oxygen need in COVID-19.

To the Editor,

We read with great interest the report on the usefulness of FebriDx  (Lumos diagnostics, Sarasota, Florida, US) in the diagnosis of COVID-19 published in October in this journal. FebriDx, which detects Myxovirus resistance protein A (MxA) and C reactive protein (CRP), is not available in Japan, but we also have considered that MxA would be useful for Coronavirus disease 2019 (COVID-19) practice. Therefore, we would like to introduce our study.

MxA induced exclusively by type I and III interferons (IFNs) as a specific response to viral infections, has activity against a wide range of viruses. Although few studies have clarified the relationship between coronavirus and MxA, there is a report that symptomatic patients have higher MxA levels than asymptomatic patients in respiratory viral infections. Therefore, we hypothesized that severely ill patients with COVID-19 had high blood MxA levels and that there was a difference in MxA levels between patients who would need oxygen and those would not. Since the need for supplemental oxygen is a major factor in determining whether a patient requires hospitalization, we examined the relationship between MxA levels and the need for supplemental oxygen and clarified whether MxA is a useful predictor of oxygen demand.

This study investigated 48 patients with COVID-19 who were admitted to Japan Self-Defense Forces Central Hospital in Tokyo, Japan. All patients were SARS-Cov-2 positive Japanese adults who were confirmed by real-time polymerase chain reaction (PCR) using a nasopharyngeal swab specimen and confirmed to have pneumonia by chest computed tomography. We also included 14 non-infected outpatients with metabolic syndrome. We excluded participants had lung disease or had used antivirals or steroids before admission. The whole blood MxA levels were quantified by sandwich enzyme-linked immunosorbent assay (ELISA) using a MxA Protein Human ELISA kit (BioVendor GmbH, Heidelberg, Germany). The obtained values are shown by as median [interquartile range, IQR]. Statistical analyses were performed using SPSS v19 software for Windows (IBM Corp., Armonk, NY).This study was reviewed and approved by Japan Self-Defense Forces Central Hospital (approval number 02–025).

We compared the patient oxygen demand during hospitalization, and categorized the patients into four groups as follows: non-infected individuals (non-infected, n = 14), patients who did not require oxygen (mild, n = 23), patients who required oxygen by conventional nasal cannula (moderate, n = 15), and patients who required oxygen by high-flow nasal cannula or ventilator (severe; n = 10). The three groups of patients with COVID-19 had significantly higher MxA levels than the non-infected group (0.386 [0.371–0.395]; p < 0.001). Severely ill patients showed the highest MxA levels (mild, 3.715 [2.560–9.600]; moderate, 6.079 [3.922–12.084]; severe, 11.777 [5.216–25.183]) (Fig. 1 A). This result highlights the efficacy of MxA in the diagnosis of COVID-19, and agrees with the report by Clark et al. Furthermore, MxA was also associated with the severity of COVID-19.

Fig. 1

Relationship between COVID-19 severity and MxA levels. Blood MxA levels in COVID-19 patients. Boxes show the median (line) and interquartile range (IQR) Whiskers show the highest and lowest values that are no greater than 1.5 times the IQR. A) Comparison of COVID-19 patients with non-infected individuals. Patients were divided according to severity as follows: mild, patients who did not require oxygen supplementation; moderate, patients who required oxygen by conventional nasal cannula; severe, patients who required oxygen by high-flow nasal cannula or ventilator. The days after onset when sample was taken were as follows: mild, 5 days [4–6]; moderate, 7 days [5–8.5]; severe, 7.5 days [5.25–8], respectively). The clear circle indicates an outlier case with values 1.5–3.0 times the IQR. The black circle indicates an extreme case with values >3 times the IQR. *p < 0.001 by Kruskal–Wallis test. B) MxA levels in patients who did not require supplemental oxygen at admission. Patients who did not require oxygen supplementation at the time of admission were divided stable patients who did not require oxygen supplementation throughout hospitalization (stable) and progressive patients who did (progressive). The blood samples of two groups were taken at approximately the same days after onset (5 days [4–6] and 5 days [4–7], respectively). There was a significant difference between the two groups (p < 0.05 by Mann–Whitney U test).

Next, we divided the 35 patients who did not require oxygen at the time of admission into two groups: those who did not require oxygen supplementation during hospitalization (stable patients, n = 23) and those that did (progressive patients, n = 12). The progressive patients had significantly higher MxA levels than the stable patients (6.203 [4.237–23.350] vs 3.715 [2.560–9.600]; p < 0.05) (Fig. 1B). This result shows that MxA predicts the need for supplemental oxygen in COVID-19.

Additionally, we investigated the dynamics of MxA levels in 10 patients using blood specimens drawn at admission and several days thereafter. MxA levels decreased after admission in all cases (4.729 [2.666–15.466] vs 1.688 [0.665–4.157]; p < 0.05), suggesting that they peak before onset or in the early stage of disease (Fig. 2 ).

Fig. 2

Dynamics of MxA. Blood MxA levels in the same individuals at admission and several days thereafter. The median and interquartile range are shown above the groups; p < 0.05 by Wilcoxon signed-rank test.

There were reported that the impaired response of IFN type I and III in patients with COVID-19 wasassociated with a persistent blood viral load and an exacerbated inflammatory response, leading to severe illness4, 5, 6, 7. Based on these reports and taking into consideration that MxA expression is strictly controlled by type I and III IFNs , , MxA expression should be suppressed in COVID-19 patients, especially those who are critically ill. In contrast, all patients in our study, including those critically ill on mechanical ventilation, had significantly higher levels of MxA than the non-infected individuals. Additionally, we found that MxA levels were higher in critically ill patients and patients requiring oxygen supplementation (Fig. 1). Impaired IFNs response can be due to either reduced or delayed IFNs production. Our results support the delayed IFNs production theory , rather than that of reduced IFNs production , . Delayed IFNs may be associated with a high blood viral load, which increase IFNs production and leads high MxA production.

Our finding that MxA declines from the early stage of onset (Fig. 2) is similar to the levels of type I IFN over time in patients with COVID-19 , . When taking into consideration that MxA has fast induction time, up to 10-fold higher MxA protein levels were observed at 24–48 h after IFN induction in vitro, we think that MxA has a peak in the early stage of infection. It suggests may be useful as an auxiliary test instead of real-time PCR, which is insufficiently sensitive to diagnose SARS-CoV-2 infection, not only in the early stage of onset but also in the incubation period. When applied as a screening test for immigration, it may significantly reduce the number of people who need to be quarantined for 14 days.

However, the rapid change in MxA over time also means that care must be taken when interpreting MxA level. In other words, when comparing data, samples with the same days after onset should be compared. Detailed data on changes over time will be needed for clinical application.

In conclusion, we revealed that MxA in peripheral blood predicts the need for supplemental oxygen in patients with COVID-19. We consider that our results are beneficial from a clinical viewpoint. We hope that research on the relationship between MxA and COVID-19 will progress and subsequently contribute to clinical practice.

Declaration of Competing Interest

The authors declare no conflicts of interest associated with this manuscript.