There are currently no proven effective treatments for COVID-19 (Li and De Clercq, 2020; Sanders et al., 2020). For the treatment of patients, several antiviral drugs are repurposed, based on their in vitro effectiveness against SARS-CoV-2. The SARS-CoV-2 infection causes COVID-19 and other coronavirus strains (Mantlo et al., 2020; Wang et al., 2020).

Interferons (IFN) have demonstrated clinical efficacy in the treatment of various viral infections and are widely used to treat patients with COVID-19 (Hung et al., 2020; Xie et al., 2020; Zuo et al., 2020).

Early administration of IFN slows down virus replication and prevents a cytokine storm

Acute inflammation, often seen in patients with severe COVID-19, is the result of suppressed type I IFN expression and subsequent unbalanced IFN response with excessive cytokine production. This reaction is correcting by introducing a therapeutic IFN. Indeed, animal studies have shown that early IFN treatment protects mice from deadly SARS-CoV and MERS-CoV infections.

The results of this study suggest that early administration of IFN to patients with COVID-19 with a high viral load or a weakened immune system would slow down virus replication and disease progression, and prevent the cytokine storm that causes the most damage.

Interferon and increased ACE2 expression. A potential risk for increased penetration of SARS-CoV-2 into target cells

It has recently been demonstrated that the angiotensin-converting enzyme ACE2, the docking protein for SARS-CoV-2, is an interferon-stimulated (ISG) gene that is regulated by IFN-α. It means that IFN signals can facilitate virus penetration and replication (Su and Jiang, 2020). In this study, type I IFN was injected directly into the airway via an aerosol nebulizer. Introduced in this way, it should have a high ability to enhance the expression of ACE2 in airway cells.

Data from this study did not show any association between early IFN use and hospital stay and time to improve CT in survivors. Moreover, later use of IFN was associated with slower improvement in CT. These results indicate the possibility that during the peak of virus replication in moderately ill patients, the positive and harmful effects of IFN may compensate for each other. Since severely ill patients often have a high viral load that may have peaked in airway cells, IFN-induced ACE2 expression may no longer pose a significant risk, as it does in moderately ill patients.

A retrospective study to test the effectiveness of IFN-α2b and the potential risk of its use

A retrospective multicenter cohort study was conducted to investigate the relationship between α2b use and time of administration and clinical outcomes. The study included 446 patients with a confirmed diagnosis of COVID-19. All patients received antiviral therapy at two regional medical centres in neighbouring cities in Hubei province, China. The purpose of this study was to assess the Association of IFN-α2b use with the progression of COVID-19 disease and identify the potential risk of IFN therapy for patients with COVID-19.

Patients sampling

In the two participating medical centres, 730 patients diagnosed with COVID-19 were registered from January 15 to March 31. The following were excluded from the sample for the study:

  • 185 patients of mild severity who had a significant part of the diagnosis or treatment history missing from their charts. These patients did not need hospital treatment and were transferred to isolation units until they recovered.
  • 99 patients who did not have antiviral treatment, hospitalized for non-COVID-19 reasons, and underwent CT scans less than twice or spent less than 5 days in the hospital.

The remaining 446 patient records were summarized and included in the analysis (picture 1). The final update is May 22, 2020. All patients included in the study were discharged or died before this date.

Picture 1. sampling Strategy for patients with COVID-19

Among the analyzed patients, the age range was from 8 to 96, with a median of 50. 47.1% of patients were women. For all patients, the average duration from onset of symptoms to admission to the clinic was 6 days, and the average length of hospital stay was 19 days. For survivors, the average duration from hospitalization to the improvement of CT was 10 days. The average duration from the onset of the disease to discharge from the hospital was 26 days. Since patients were re-tested for SARS-CoV-2 nucleic acids only when symptoms and CT scans improved, the time from clinic admission to a negative nucleic acid test could not be accurately determined in this study.

The most common risk factors were hypertension (21.1%) and diabetes (7.4%). A high risk of exposure to SARS-CoV-2 was registered in 62.6% of the analyzed patients.

The main symptoms on admission were fever (82.7%), cough (67.7%), fatigue, muscle pain (20.2%), chest pain, or shortness of breath (18.4%). Headache (5.2%) and diarrhoea (2.5%) were not as significant in this sample of patients.

Laboratory tests showed 61.3% eosinopenia, 46.2% lymphopenia, and 36.7% elevated lactate dehydrogenase (LDH) within 24 hours of admission.

During hospitalization, 6.1% of patients needed artificial ventilation, and 1.1% received extracorporeal membrane oxygenation (ECMO).

Care points are not associated with clinical outcomes

During January-March 2020, the Chinese centre for disease prevention and control (CDC) published updated national guidelines for the treatment of patients with COVID-19. The guidelines included the use of antiviral medications such as IFN, lopinavir/ritonavir (LPV/r), and umifenovir (UFV).

During the pandemic, each hospital developed its preferred medication regimen based on both clinical experience and drug availability. Patients were treated according to their preferred regimen before resorting to alternative medications.

Demographic characteristics and clinical features within 24 hours of admission were comparable between the first and second medical centres, except that the first centre had more at-risk patients, fewer patients with eosinopenia on admission, and fewer patients with symptoms on admission such as fever, cough, fatigue, muscle pain, chest pain or shortness of breath, headache, and diarrhoea (table S1).

Among patients admitted to the first centre, only 25% received IFN-α2b, while 72.2% received LPV/r. in Contrast, among patients admitted to the second centre, 73% received IFN-α2b and 36.3% received LPV/r.

To evaluate standard care at the two medical centres, we compared cumulative event curves for hospital discharge and improved computed tomography. Among all survivors, patients admitted to the first centre had a more extended hospitalization (logarithmic criterion, p < 0.001) (Pic. S1 A). Since LPV/r treatment has side effects and often prolongs hospitalization, after excluding patients who received LPV/r, both centres showed a comparable rate of hospital discharge (logarithmic criterion, p = 0.685) and improved computed tomography (logarithmic criterion, p = 0.749) (Pic. S1 B and S1 C).

The analysis of 58 patients who did not receive antiviral therapy also showed no differences in hospital discharge (logarithmic criterion, p = 0.678) and improvement in computed tomography (logarithmic criterion, p = 0.561) between both centres (Pic. S1 D and S1 E).

Two cities of those two medical centres are located about 50 miles apart. Population density, ethnicity, economy, job composition, and urban air pollution levels are comparable.

Given all these facts, the place of care was not associated with clinical outcomes in this study.

Early and late IFN therapy (or its absence)

Since the timing of IFN therapy may be critical to its effectiveness against SARS-CoV-2 infection, the study calculated the time from admission to the clinic to the start of IFN therapy. It appears that 68.2% of patients received the first dose of IFN within the first 2 days after hospitalization, and 89.3% received the first dose within the first 5 days. Along with the fact that the first CT re-evaluation and treatment plan adjustments were also performed around the fifth day of hospitalization, it was empirically determined that starting IFN therapy within the first 5 days of hospitalization was considered early IFN therapy, and Vice versa (picture 2A). According to this criterion, the combined cohort from both medical centres was divided into three groups: 216 patients (48.4%) received early IFN therapy, 26 (5.8%) received late IFN therapy, and 204 (45.7%) did not receive IFN therapy at all. No patient received IFN-α2b therapy before admission.

Picture 2. Early IFN therapy associated with reduced in-hospital mortality, but not with early recovery of COVID-19

Demographic characteristics did not significantly differ between the early, late, and non-interferon groups, except for the place of care and a smaller number of female patients receiving early or late interferon (table 1).

As for the clinical features at admission, the early and non-IFN groups had a comparable prevalence of high respiratory rate and low O2 saturation within 24 hours after admission. The late IFN group had a higher prevalence than the early and non-IFN. More symptoms were reported in the early IFN group, where 19.9% of patients reported three or more symptoms upon admission. In groups without IFN and with late IFN – 12.8% and 0%, respectively. The group without IFN had the lowest prevalence of lymphopenia and eosinopenia on admission, while the group with late IFN had the highest. In General, patients in the late IFN group had a higher degree of disease severity at admission, whereas the severity in the early IFN and non-IFN groups were comparable.

In the groups with IFN, the average time from admission to the first dose of IFN was 2 and 8.5 days in the early and late IFN groups, respectively. The median duration of IFN therapy was 10 and 8.5 days, respectively.

Among all groups, the prevalence of antibiotics and Lianhua Qingwen capsules, a herbal-based drug with potential antiviral efficacy, was 60-90%.

As for the overall assessment of hospitalization, patients in the late IFN group had significantly more extended hospital stays, delayed CT improvement, and a longer course of the disease from the onset of symptoms to discharge from the hospital. The later IFN group also had more severe and critical patients, retrospectively identified under the Chinese CDC guidelines. These patients were more likely to start late IFN therapy after abandoning the non-IFN regimen. The severity of the disease in the early IFN and non-IFN groups was comparable. The group without IFN had a lower prevalence (80.9%) of positive tests for the presence of SARS-CoV-2 nucleic acids than in the other two groups (94.9% and 92.3%), despite the presence of 97.5% of pathological CT results at admission. The prevalence of virus isolation after the improvement of computed tomography was comparable in all 3 groups.

Table 1. Patients ‘ characteristics on the use of IFN

Research results

Main result

Hospital mortality was 3.6% among all patients and 14.4% among seriously and critically ill patients (n = 111). No deaths were reported in mild and moderate patients.

Comparing the Kaplan-Meyer curves, a significant difference in mortality was observed in the early (0.9%), late (15.4%) and IFN-free (4.9%) groups (logarithmic criterion, p < 0.001) (Pic. 2 B). Logistic regression showed that early IFN therapy was uniquely associated with lower mortality (odds ratio [OR] = 0.18, p = 0.029), whereas late IFN therapy (OR = 3.53, p = 0.046), age > 60 years ( OR = 6.87, p < 0.001), hypertension (OR = 6.87, p < 0.001), diabetes (OR = 8.96, p < 0.001), respiratory rate > 22 times per minute on admission (OR = 10.1, p <0.001), O2 saturation of 90% -93% (or = 11.8, p < 0.001) or < 93% (or = 25.2, p <0.001) were clearly associated with higher mortality (Pic S2).

In the initial analysis, both logistic regression models and Cox proportional risk models were tested and adapted to all patients. Factors associated with mortality were included in the model with the following exceptions. Gender and number of symptoms were included due to a significant difference in prevalence between groups and a potential association with mortality (female OR = 0.36, p = 0.083). High respiratory rate was excluded due to the coincidence with low O2 saturation both statistically (Spearman’s p = 0.536, p < 0.001) and clinically. The onset of symptoms before admission > 7 days was included as an indicator of pre-admission severity because patients with severe illness were generally hospitalized within 7 days of onset of symptoms.

After adjusting for gender, age, hypertension, diabetes, O2 saturation at admission, the number of symptoms at admission, and the onset of symptoms before admission > 7 days, early IFN therapy was estimated to have an adjusted OR 0.05 (95% confidence interval [CI], 0.01–0.37) and an adjusted risk ratio (HR) of 0.10 (95% CI, 0.02-0.50) for in-hospital mortality compared to no IFN therapy. Late IFN therapy was estimated to have an adjusted OR 6.82 (95% CI, 1.14–40.8) and an adjusted HR 2.30 (95% CI, 0.64–8.27) for in-hospital mortality compared to no IFN therapy (table 2). The assumption of a proportional hazard was not justified when comparing the groups of late IFN and no IFN.

Table 2. Model-adjusted probabilities of in-hospital death, hospital discharge, and computed tomography improvements

Secondary results

The length of hospital stay of patients who survived COVID-19 treatment varied significantly in all three groups (Kruskal-Wallis test, p = 0.001). There is no significant difference between the cumulative event curves of the early IFN and non-IFN groups (logarithmic criterion, p = 0.335) for hospital discharge, while the late IFN group had more extended hospitalization than the non-IFN group (rough HR score [95% CI], 1.65 [1.15-2.38]; logarithmic criterion, p = 0.016) (Picture 2 C).

All 430 survivors had improved CT scans before discharge from the hospital, including the 6 who were admitted after positive tests for SARS-CoV-2 nucleic acids without any abnormal CT results upon admission. There is no significant difference between the cumulative event curves of early IFN and non-IFN groups (logarithmic criterion, p = 0.970) for improving CT. In the late IFN group, there was a delayed improvement in CT compared to the group without IFN (rough HR [95% CI], 0.53 [0.37–0.75]; logarithmic criterion, p = 0.002) (Pic. 2 D).

Using a generalized gamma model with a logarithmic model, we found that late IFN therapy, age > 60 years, hypertension, respiratory rate > 22 / min on admission, O2 saturation of 90-93% or < 93%, lymphopenia and eosinopenia were uniquely associated with both longer hospitalization and delayed CT improvement. Diabetes and onset of symptoms before admission > 7 days were associated with delayed and early improvement in CT, respectively (Pics S3 B and S3 C). After adjusting for gender, age, hypertension, diabetes, O2 saturation on admission, number of symptoms on admission, and the onset of symptoms on admission > 7 days, early IFN therapy was evaluated as adjusted HR 1.14 (95% CI, 0.93–1.41) for hospital discharge, and 1.00 (95% CI, 0.81– 1.22) for improved computed tomography compared to no IFN therapy. For late IFN therapy, the estimated adjusted HR was 0.69 (95% CI, 0.44–1.08) for hospital discharge and 0.50 (95% CI, 0.32–0.80) for improved CT compared to no IFN therapy (table 2).

Combinations of drugs

Since IFN is regularly used in combination with antiretroviral drugs to treat patients with COVID-19, one of the goals of the study was to find a combination associated with favourable clinical responses.

After excluding patients who received late IFN therapy, 83 patients received IFN + LPV/r upon admission, and 94 received IFN + UFV. Among patients receiving combination therapy, 11 and 4 received LPV/r and UFV on the sixth day of admission or later, respectively. Other patients were transferred to single-agent therapy (table 3).

Demographic characteristics were comparable in 5 groups, except for the IFN + LPV/r group, which had significantly fewer women than the other groups and more older adults among those receiving LPV/r. Clinical features at admission were comparable between all groups, except for the group of one UFV with a lower prevalence of eosinopenia. In General, the IFN + LPV/r, IFN + UFV, and LPV/r groups had similar patient characteristics, clinical features, and sample size.

Table 3. Characteristics of patients by type of treatment

The cumulative event curves of survivors, except those who received late IFN, showed long-term hospitalization of both IFN + LPV/r groups and LPV/r only compared to other groups without LPV/r (logarithmic criterion for all curves, p < 0.001 ) (Picture 3A). Improvement of computed tomography in groups that took IFN + LPV/r and only LPV/r was also delayed to a lesser extent (logarithmic criterion for all curves, p < 0.001) (Pic. 3 B). The combination of IFN with LPV/r or UFV was not associated with variations in hospital discharge or improvement in computed tomography compared to a single LPV/r or UFV (logarithmic criterion, p > 0.100) (Pictures 3A and 3B). After adjusting for gender, age, hypertension, diabetes, O2 saturation on admission, the number of symptoms on admission, and the onset of symptoms on admission > 7 days, the Cox proportional risk model determined that only IFN, UFV, and IFN + UFV were associated with early hospital discharge compared to IFN + LPV/r. In contrast, UFV alone was associated with an early improvement in computed tomography compared to IFN + LPV/r (table 4). LPV/r alone did not show significant differences with IFN + LPV/r in terms of hospital discharge and improved computed tomography. The assumption of proportional risk was not justified in this comparison pair. It is worth noting that LPV/r was associated with delayed hospital discharge (a rough estimate of HR 0.51 [95% CI, 0.42–0.62]; HR adjusted for the Cox model 0.48 [95% CI, 0.3–0.60]) and improved computed tomography (a rough estimate of HR 0.66 [95% CI, 0.55–0.81]; HR adjusted for the Cox model 0.70 (95% CI, 0.57–0.86]) compared to the absence of LPV/r (Pic. 3D).

Picture 3. IFN-based COVID-19 therapy associated with a more favourable clinical response than LPV/r

Table 4. Adjusted according to the model, the probability of discharge from hospital and the improvement of computed tomography

The study assessed the Association of treatment groups with in-hospital mortality. Since the mortality rate is low, there was a significant association between the use of one LPV/r and increased in-hospital mortality compared to other types of treatment (Picture 3 E). Although hospital discharge and CT rates were relatively delayed, the IFN + LPV/r group had significantly lower mortality than the single LPV/r group, which may be due to early IFN use.

Study limitations

This study has several limitations. First, the retrospective design and non-randomized nature of therapy prescribing limit the interpretation of results.

Second, there is a potential for survival bias due to the requirement to receive IFN for at least 3 consecutive days in the early IFN group, which is not required for the group without IFN. However, a study of 16 records of deceased patients found that the shortest duration from hospitalization to inpatient death was 5 days, with an average of 17 days, which can accommodate a 3-day course of IFN or any other therapy. All deceased patients received at least one antiviral therapy, which was among the entry criteria for this study.

Records not included in this study were also checked. No deceased patients were found among them. However, it cannot be ruled out that some patients who received antiviral therapy may have died before the COVID-19 diagnosis was confirmed and thus were excluded from the study.

Third, detailed virological data were not included in the study, which precluded comparison with randomized controlled trials of IFN.

Fourth, the regression models did not include the location of care as a determining factor due to multicollinearity with the choice of therapy between IFN and LPV/r. Although variations in standard care have been studied and it has been concluded that there are no significant differences between the two medical centers, rejection of the choice and rejection of the confirmation may still affect the results.

Finally, assistive and supportive therapies such as corticosteroids, immunoglobulins, immunomodulatory drugs, and phytotherapy were not included in the analyses but could affect the length of hospital stay.

Confirmation

This study was supported by the US Department of Veterans Affairs (5I01BX001353) and the Seventh hospital of the Sun Yat-sen University (393003).

Source

Our Telegram channel: