A combination treatment of IFN-α2b and IFN-γ accelerates viral clearance and control inflammatory response in COVID-19: Preliminary results of a randomized controlled trial

Esquivel-Moynelo Idelsis1, Pérez-Escribano Jésus2, Duncan-Roberts Yaquelin3, Vázquez-Blomquist Dania4, Bequet-Romero Mónica4, BáezRodríguez Lisandra5, Castro-Ríos Jésus6, Cobas-Cervantes Lisbeth7, Pagé-Calvet Ernesto5, Travieso-Pérez Saily5, Martinez-Suarez Claudia3, Campa-Legra Ivan3, Fernandez-Masso Julio4, CamachoRodriguez Hamlet4, Díaz-Gálvez Marisol1, Sin-Mayor Adriana1, GarcíaSánchez Maura1, Martínez-Martín Sara1, Alonso-Valdés Marel3, Hernández-Bernal Francisco3, Nodarse-Cuni Hugo3, Bello-Garcia Dianela8, Canaan-Haden Ayala Camila4, Gonzáles-Moya Isabel4, BeatoCanfuk Abrahan9, Vizcaino-Cesar Tania1, Guillén-Nieto Gerardo4, Muzio-González Verena3, Fish Eleanor10 and Bello-Rivero Iraldo3*


Background
SARS-CoV-2 infection that results in COVID- 19 have spread in 190 countries around the globe, resulting in 171.271 million cases and 3.571 deaths, with devastating social and economic consequences [1]. In the absence of a vaccine, there is an urgent need to develop strategies to limit virus infection. The fi rst cases of COVID-19 in Cuba were confi rmed on March 11, 2020: three tourists from the Italian region of Lombardy, who were immediately hospitalized [2]. In Cuba, 12.0% of people who have been diagnosed with COVID-19 are under the age of 20. 53.9% of patients are asymptomatic, and even in the most vulnerable patients, comprising the 80+ -year-old group, 51.7% of those infected present with no symptoms at virus confi rmation [3].
Type I interferons (IFNs)-/ [4] and IFN- [5,6] exhibit antiviral activities. IF Ns exert both direct antiviral effects on different viruses at different stages of their replicative cycles and also elicit an immune response to clear virus [4]. Notably, the severity of COVID-19 correlates with the failure to initiate an IFN response to SARS-CoV-2 infection [7].
Cognizant that a number of antivirals are under evaluation globally in clinical trials, the Cuban Protocol for Management of COVID-19 [2], includes Heberon Alpha R (IFN-2b) and other potential antiviral treatments including lopinavir/ ritonavir (Kaletra) and chloroquine (CQ), administered during the symptomatic phase of disease. Cuban patients who are symptomatic and their close contacts are isolated in centers established for that purpose, and receive treatment. On confi rmation of a positive SARS-CoV-2 test, cases are hospitalized and continue or start to receive Heberon Alpha R, Kaletra, and CQ, as established by Cuban Health Ministry guidelines. 2 In a cohort of 761 confi rmed SARS-CoV-2 infected individuals who received Heberon Alpha R plus Kaletra and CQ, 95.4% recovered fully from COVID-19, with only a 0.92% case fatality rate [8].
Early studies with combination treatments of a type I IFN and IFN- revealed synergistic inhibition of SARS CoV replication in vitro [9][10][11][12]. IFN- is a critical regulator linking the innate and adaptive immune responses [13] Moreover, combination treatments with antivirals that target different stages of the virus replicative cycle [14][15][16] and IFN-2b and IFN- may synergize further to accelerate viral clearance. Accordingly, we conducted a phase II randomized clinical trial to evaluate whether a combination of IFN-2b and IFN-, in addition to standard of care would offer a therapeutic advantage in COVID-19 cases, promoting faster viral clearance and preventing progression to severe disease.

Study design and patients
Adult (19-82 yrs), PCR(+) confi rmed SARS-CoV-2 infected cases were enrolled in this open-labeled, single center, prospective, randomized and controlled clinical trial at Military Central Hospital Luis Diaz Soto Hospital, Havana, Cuba. Patients were randomly assigned to receive either subcutaneous treatment with a co-lyophilized combination of 3.0 million international units (MIU) IFN-2b and 0.5 MIU IFN- (HeberFERON®, Cuba Center for Genetic Engineering (CIGB), Havana, Cuba), twice a week for two weeks, or an intramuscular injection of 3.0 MIU IFN-2b (Heberon® Alpha R, CIGB, Havana, Cuba), three times a week for 2 weeks. Additionally, all patients received standard of care, which included lopinavir/ritonavir (200/50 mg orally twice daily and chloroquine (250 mg orally twice daily). Patients were randomized individually to one of the two treatment arms by means of random computer-generated lists, with an allocation ratio of 1:1, with block sizes of six patients, based on a power calculation of 80%, and a level of confi dence set at 95%, while also considering a dropout rate of 5%.
Heberon® Alpha R has been produced in Cuba by the CIGB for the past 34 years, with proven antiviral effi cacy and an adequate safety profi le [17]. HeberFERON® (IFN-2b and IFN-, co-lyophilized in the same vial) is produced at CIGB, and has been registered in Cuba since 2016 for the treatment of basal cell carcinoma [18].
Study execution conformed with the ethical principles of the Declaration of Helsinki and the International Council for Harmonization of Good Clinical Practice guidelines. No compensation was provided for enrollment in the trial. Patient personal data were protected. The authors were responsible for designing the trial and for collecting and analyzing the data. The authors assured the completeness and accuracy of the data collection and adherence to the protocol. The details of the trial are provided in the protocol that has been posted in TRIALS [19]. Primary endpoints were the time to viral RNA elimination from the start of treatment and the time to progression to severe COVID-19.

Eligibility criteria
A COVID-19 diagnosis was obtained by a positive realtime reverse transcription-polymerase chain reaction (RT-PCR) amplifi cation of the viral E gene and then confi rmation by amplifi cation of the RdRP gene, from throat swab samples. Adults (≥19yrs), confi rmed PCR(+) for SARS CoV-2, with ECOG functional status ≥2 (Karnofsky ≥ 70%) and who signed informed written consent, were included. Patients with any of the following characteristics were excluded: decompensated chronic diseases at the time of inclusion (severe arterial hypertension, ischemic heart disease, diabetes mellitus, and a history of autoimmune disease), presence of hyper infl ammation syndrome, serious coagulation disorders, known hypersensitivity to any of the components of the formulations under evaluation, pregnancy or lactation, and obvious mental incapacity to issue consent.

Statistical analyses
Quantitative variables were plotted as arithmetic means and standard deviations, and the median with IQR range. Absolute and relative frequencies (%) for qualitative variables were measured. The hypothesis test used was Fisher's exact test.
Time to viral clearance was analyzed using Kaplan-Meier plots and the comparisons used Mantel-Cox Log Rank and Gehan-Breslow-Wilcoxon tests. Changes in laboratory parameters were analyzed using a paired mixed model (which cannot adjust for missing values due to patient release from hospital).
Correlations between time to viral clearance and laboratory parameters were studied using a two-tailed non-parametric Spearman correlation with a 95% confi dence interval (CI).

Results
Between Apr 11, 2020, and May 13, 2020, 144 PCR(+) COVID-19 cases were screened for trial eligibility (Figure 1). 57 patients were ineligible: one with icterus, one with chronic decompensating renal insuffi ciency, two unconfi rmed SARS-CoV-2 PCR results, and 53 PCR(+) cases with persistent viral shedding beyond 21 days. Eight patients did not consent and were therefore excluded. 79 subjects met the inclusion criteria and were randomly assigned (1:1) to either the HeberFERON group (41 patients)   to start of treatment, and one in the Heberon Alpha R group who refused throat swab sampling. 34 patients received standard of care which included Heberon Alpha R (control group) and 33 patients were treated with HeberFERON plus standard of care.
In summary, 30 and 33 patients were analyzed by intention to treat (ITT) in the HeberFERON and Heberon Alpha R groups, respectively. Refer to Table 1 for patient characteristics. In the Heberon Alpha R group the median age was 31.0 years (IQR: , and for the HeberFERON group the median age was 42.0 years (IQR: 27-56) (p=0.023). The Heberon Alpha R cohort comprised 60.6% males (20/33) compared with 46.7% males in the HeberFERON group (14/30) ( Table 1). These sex differences between HeberFERON and Heberon Alpha R groups were not statistically signifi cant. More symptomatic patients (51.5%) were in the Heberon Alpha R group than in the HeberFERON group (40.0%), but this difference was also not statistically signifi cant.
The median age of symptomatic subjects was higher in the HeberFERON group (50yrs [IQR: 28-67)]) than in the Heberon Alpha R group (24yrs (IQR: ). In symptomatic patients, the sex distribution was not statistically signifi cant in the whole cohort: 51.7% males, 48.3% females (p=0.093).
However, 66.7% were females in the HeberFERON treated group compared with 35.3% in the Heberon Alpha R treated group. Patients in the HeberFERON group had symptoms for >7days, compared with 35.3% in the Heberon Alpha R group, but this difference in symptom duration was not statistically signifi cant. In asymptomatic patients a difference in sex distribution (p=0.002) was detected ( Table 1).
The more common symptoms were fever and unproductive cough (16.4%), followed by headache (9.6%), weakness (8.4%), odynophagia and nasal secretions (5.4%), diarrhea, dyspnea, chills and general malaise (4.1%), and sore throat and myalgia  There were no differences between the incidence or duration of any of the adverse events between the treatment groups.
No serious adverse events were reported and no patients died during the study.
We analyzed 63 patients with available throat swab samples.   (Table 3). Notably, no patient in either treatment group progressed to severe COVID-19 (Table  3). No differences in the evolution of vital signs were detected between groups (Table 3).
Close scrutiny of a number of clinical measures reveals differences between those patients that received IFN-2b plus IFN- (HeberFERON) and those who only received IFN-2b (Heberon Alpha R), along with standard of care, beyond faster viral clearance. Lymphocyte concentration was similar during the treatment time course for both cohorts of patients ( Figure  3A). In the fi rst seven days following treatment onset, those patients who received HeberFERON saw signifi cant increases (p=0.0058) in the percentage of their lymphocytes that exceeded the increases observed for Heberon Apha R patients ( Figure 3B). For Heberon Alpha R treated patients, a signifi cant increase in the percentage of lymphocytes was achieved by 2 weeks post treatment onset (p=0.011). For both the HeberFERON and Heberon Alpha R treated patients, treatment decreased circulating levels of neutrophils, signifi cantly (p=0.0019). The data in Figure 3C and 3D suggest that the majority of Heberon Alpha R treated patients became neutropenic.
Additionally, HeberFERON treatment signifi cantly reduced the neutrophil:lymphocyte ratio (NLR) in the fi rst week of treatment (p=0.0264), which was normalized by the end of the second week of treatment (p=0.0148). By contrast, the NLR remained low in Heberon Alpha R treated patients (p=0.0022) ( Figure 4A (Table 4 and fi gure 6).  In a fi nal series of analyses we examined whether the rate of viral clearance correlated with clinical laboratory parameters. This correlation was analysed using a two-tailed non-parametric Spearman correlation with 95% confi dence interval. and PLR index (p=<0.0001) and viral clearance ( Figure 6).

Discussion
Our study had several limitations. This trial was open label, without a placebo group and with unbalanced demographics (age) between treatment arms. In addition, throat swab sampling for detection of SARS CoV-2 does not refl ect infection burden in the lower respiratory airways: one published report provided evidence for lower viral loads in throat swab specimens [20]. Irrespective of these limitations, the fi ndings from this trial are that HeberFERON plus Kaletra and CQ treatment for COVID-19 is safe and results in faster viral clearance than treatment with Heberon Alpha R. Specifi cally; a combination treatment of IFN-2b with IFN- promotes viral clearance more rapidly than IFN-2b alone and exerts antiinfl ammatory effect.
There is accumulating evidence that for COVID-19 there is an asymptomatic incubation period, with or without detectable viral RNA, followed by non-severe symptomatic conditions and detectable virus, which in some individuals progresses to a severe symptomatic stage with high viral load [21]. The  asymptomatic incubation period of 4-7 days is associated with aggressive transmission of the virus [22,23]. Added to this, some infected patients remain asymptomatic and yet are fully capable of transmitting virus [24][25][26]. Asymptomatic individuals pose the greatest threat to viral transmission, not only because they remain undetected in the general population, spreading infection, but also because there is evidence that viral shedding persists longer in asymptomatic versus symptomatic COVID-19 cases [27,28] this despite there being no evidence of higher viral loads in asymptomatic individuals [20]. We provide evidence of rapid viral clearance with HeberFERON treatment for both symptomatic and asymptomatic COVID-19 cases. By 48hrs after onset of treatment with HeberFERON, viral clearance was achieved in 45% of cases, and by day fi ve in 96% of cases. Notably, a combination treatment of lopinavir/ ritonavir, ribavirin and IFN- did not achieve this rate of viral clearance [16], with an average of seven days to viral clearance, and the combination of lopinavir/ritonavir reportedly has no therapeutic advantage for course of disease [29,30]. Our results with Heberon Alpha R in combination with lopnavir/ ritonavir and CQ show viral clearance by day fi ve in 74% of cases. In another exploratory clinical study, inhaled IFN-2b treatment resulted in accelerated viral clearance in COVID-19 cases, on average seven days sooner than in those cases treated with the antiviral arbidol [31]. Recently, fi ndings from a randomized controlled trial of treatment with Novaferon, a novel protein generated by shuffl ing the cDNAs from IFN-s [12] revealed that Novaferon accelerated viral clearance by 3 days when compared with treatment with lopinavir/ritonavir [32]. Systemic viral dissemination is an important determinant in severe disease [33]. HeberFERON treatment, by accelerating viral clearance, may prevent disease progression, as noted in all COVID-19 cases in this trial.
The timing of initiation of antiviral therapy is a critical factor in the treatment of viral infections. With respect to SARS-CoV, no effect of several antiviral drugs was observed when the treatments were started 6-14 days after symptom onset [34] How ever, when SARS-CoV patients were treated with an IFN- between 4-10 days post symptom onset, rapid resolution of disease was observed [35]. Early administration of antiviral medications may improve outcomes for COVID-19 cases [36]. Early IFN treatment was recommended for MERS [37] and late therapy (10-22 days) may contribute to poor   [51,57].
A major concern relating to the use of IFN-/ß for the treatment of COVID-19, given that an exacerbated lung infl ammatory response is associated with severe disease, is the notion that these IFNs will induce a cytokine storm to create an infl ammatory response. In several published reports, where IFNs-/ß were used as antiviral therapeutics for SARS [35] ebola virus infection [58] and more recently, COVID-19 [29] there was no evidence of IFN-induced cytokine storms or exacerbation of infl ammation. Indeed, when COVID-19 cases were treated with an inhaled IFN-2b, evidence was provided for reductions in circulating levels of IL-6 and CRP, both biomarkers of infl ammation [29] As ill and there were no deaths. In a similar Cuban cohort of COVID-19 patients, treated early with IFN-2b and standard of care, mortality rate was reduced to 0.9% [8].
It has been suggested that IFN treatment will only be effective in patients who lack co-morbidities [34,65] and that co-morbidities, such as diabetes, will diminish any response to IFN [64] However, we provide evidence for rapid viral clearance, resolution of disease symptoms, and hospital discharge for all patients treated with HeberFERON, a cohort that included 57% of cases with existing co-morbidities. Indeed two diabetic patients in our cohort cleared virus by day 3 and before day 14 of treatment, respectively.
The favorable pharmacodynamics of HeberFERON treatment [18] are associated with a regimen that involves few doses and lower dosing of IFN-2b and IFN- than is traditionally used when they are administered as monotherapies. The fi ndings presented herein indicate that HeberFERON® was a safe, well-tolerated treatment, and superior to Heberon Alpha R in shortening the time to SARS-CoV-2 viral clearance in a cohort of both symptomatic and asymptomatic COVID-19 patients, ages 19 -82 years. Viral clearance was achieved in 95% of patients within fi ve days of treatment onset. Our fi ndings indicate that this rapid viral clearance resulted in a blunted infl ammatory response, suggested by reduced CRP, CPK, TP levels and NLR index. The combination of IFN-2b plus IFN- while promoting host resistance to viral infection, control the infl ammatory response which prevented disease progression in all patients.

Conclusions
The fi ndings reported herein are the fi rst to suggest therapeutic effi cacy in COVID-19 for a combination treatment of IFN-2b and IFN- with clear evidence of viral clearance, infl ammatory control and no progression to severe disease or death. The use of HeberFERON combined with Kaletra and chloroquine could be a promissory approach to the early treatment of patients positive to SARS-CoV-2, symptomatic or not.

Declarations
The clinical trial protocol was approved by the Ethics Committee on Clinical Investigation of Military Central Hospital Luis Diaz Soto, and the Center for the State Control of Medicines, Equipment and Medical Devices (CECMED) in Cuba. Patients were asked for written consent to participate after having been duly informed about the nature of the trial, objectives, benefi ts and possible risks. They were informed of their rights to participate or not, and to withdraw their consent at any time, without exposing themselves to any limitations on their medical care or other retaliation.
Study execution conformed to the ethical principles of the Declaration of Helsinki and the International Council for Harmonization of Good Clinical Practice guidelines. No compensation was provided for enrollment in the trial. Patient personal data were protected. The authors were responsible for designing the trial and for collecting and analyzing the data.

Availability of data and materials
Qualifi ed individuals may request access to the deidentifi ed participant data, anonymized clinical study reports, informed consent forms, and related documents including the study protocol, through submission of a proposal with a defi ned research question to the corresponding author, Bello-Rivero Iraldo, provided that the necessary data protection and ethical committee approvals are in compliance with the trial registration. An agreement for transfer of these data will be required.

Authors' contributions
LBR, JCR, LCC, EPC, STP, were the principal investigators that conducted the trial, recorded the clinical histories of patients, and supervised each team of health workers that managed COVID-19 cases; IEM, MDG, ASM, MGS, SMM, JPE, ABC, MVT, coordinated, executed and supervised the trial at the hospital; YDR, CMS, ICL were responsible for monitoring the trial and conducted quality checks, and interpretation of clinical data. DVB was responsible, executed and analyzed data from SARS-CoV-2 viral RT-PCR determinations wrote reports and contributed to proofreading the manuscript. HCR and JRFM were responsible for SARS-CoV-2 viral RT-PCR evaluations; MBR, CACH, and IMG evaluated and analyzed clinical laboratory data and contributed to proofreading the manuscript. MAV was responsible for data base and its quality, DBG was responsible for statistical analyses. GGN, HNC and FHB contributed to the design of trial protocol, VLMG supervised all the study steps for GCP. ENF analyzed the clinical and virologic data and signifi cantly contributed to the writing of the fi nal versions of the manuscript. IBR designed and wrote the protocol, supervised the trial, interpreted and discussed all the trial data and results, and contributed to the writing of the manuscript. All authors read and approved the fi nal manuscript.