Abstract
Background Evidence suggests that early, robust type 1 interferon responses to SARS-CoV-2 are critical determinants for COVID-19 disease outcomes, accelerating viral clearance and limiting viral shedding. Accordingly, we undertook a ring prophylaxis study to determine whether pegylated IFNß-1α could reduce SARS-CoV-2 household transmission.
Methods A household cluster randomized controlled study of IFNß-1α administered to non-hospitalized, symptomatic COVID-19 index cases and treatment-eligible household contacts aged 18-70 years compared to standard care, was conducted. Following randomization participants received IFNß-1α on days 1, 6, and 11 or standard care. Viral shedding was determined by sequential salivary polymerase chain reaction measurements until day 29 in both study arms. A post-hoc ‘at risk population’ was defined as households where the index case was positive at the start of the study and there was at least one treatment eligible contact in a household who tested negative for SARS-CoV-2. Frequentist and Bayesian analyses were undertaken to determine the effects of treatment on (i) reducing viral shedding in index cases and (ii) reducing viral transmission to post-exposure household contacts.
Results In total, 1172 participants in 341 households underwent randomization, with 607 assigned to receive IFNß-1α and 565 to standard care. Based on intention to treat and per protocol analyses, IFNß-1α treatment was ineffective. However, in the ‘at risk’ population, the relative risk of infection was reduced by 23% in treated individuals and that there was a 95% probability that IFNß-1α reduced household transmission.
Conclusion Ring prophylaxis with IFNß-1α reduces the probability of SARS-CoV-2 transmission within a household.
Introduction
The SARS-CoV-2 pandemic has claimed over six million lives. Despite the rapid development and deployment of vaccines in many countries, the number of new cases worldwide still exceeds 500,000 daily (https://covid19.who.int). With each wave of the pandemic, health systems have been challenged and the likelihood of emergence of mutant strains of the virus remains. Mutated strains may be more transmissible 1,2, cause more severe disease than the original pandemic strain of SARS-CoV-2 3 and have the potential to evade available vaccines 4,5. Whilst widespread vaccination has had success limiting the trajectory of the pandemic, the emergence of the omicron variants demonstrated that even with mutations that appear to cause less severe disease 6 , high transmission despite immunization could still result in significant pressures on health services 7.
The solution to halting any pandemic is ending community transmission. During the current pandemic, measures such as healthy hygiene, self-isolation when sick, physical-distancing and use of facemasks have all been effective 8. Moreover, expedited public health responses such as extensive contact-tracing, testing for infection and community lockdowns have all been effective methods to curb transmission 9. International and local border closures plus strict quarantine measures have reduced community transmission to zero for periods in countries such as Australia and New Zealand 10,11. However, in these countries and elsewhere, as restrictions are relaxed, localized outbreaks have occurred 12 that have required rapid, community-wide responses to again supress transmission. Importantly, these community constraints cause unprecedented civil disruption and come at enormous economic 13,14 and social costs 15,16.
Since the evolution of dominant SARS-CoV-2 virus cannot be easily predicted 17, there remains a need to identify interventions that can be rapidly deployed should highly pathogenic strains emerge despite high levels of community immunization. Furthermore, preparations for the next pandemic must include strategies to limit the potential for infection and transmission on first contact with pathogenic respiratory viruses.
One of the many therapeutic approaches investigated that appeared clinically useful early in the course of the pandemic was treatment with type 1 interferons (IFNs). Randomized, controlled studies suggest that IFNs offer clinical benefits in moderate 18 and severe disease 19,20 }, reduce the duration of viral shedding 21 and prevent infection in front-line hospital workers 22. Nonetheless, IFNs are not generally recommended for treatment of proven cases of Covid-19 23.
Since IFNs are sentinel innate immune signalling molecules produced early after first contact with viral pathogens 24–26, we postulated that prophylactic IFN treatment might reduce susceptibility to infection of uninfected contacts of individuals with SARS-CoV-2 infection. Such post-exposure prophylaxis could provide non-specific, antiviral protection to curb episodic viral outbreaks 27, help suppress community transmission, even in vaccinated populations, and therefore reduce the risk of emergence of dangerous mutations 4,28.
We therefore undertook a cluster, randomized, controlled study of sub-cutaneous pegylated IFNß-1α (Plegridy. Biogen Inc, Cambridge MA) administration to determine whether IFNß-1α therapy given to index cases and household contacts can reduce transmission of SARS-CoV-2.
Methods
The Containing Coronavirus Disease-19 (ConCorD-19) trial was a prospective, cluster, randomized trial of subcutaneous (sc) administration of pegylated IFNß-1α (Plegridy, Biogen Inc, Switzerland) versus standard care (control) 29. Each household of an index case was randomized to either the IFNß-1α treatment or the standard care control arm. The study was approved by the Institutional Review Board of the Pontificia Universidad Católica de Chile and was registered with clinicaltrials.gov (NCT04552379). All participants provided written informed consent.
Trial population
Index cases were identified from databases of those with confirmed SARS-CoV-2 from COVID-19 clinics and emergency room visits in Santiago, Chile. Households were contacted by telephone to determine eligibility prior to enrolment. Individuals aged between 18 and 80 years who met inclusion and exclusion criteria were deemed as ‘eligible’ contacts with households only included if there was at least one eligible contact 29. Enrolment and the first doses of IFNß-1α - if in the intervention arm- had to be within 72 hours of the positive identification of SARS-CoV-2 by polymerase chain reaction (PCR) in index cases. A household was excluded if the index case had been in complete self-quarantine from other household members during the 48 hours prior to the diagnosis of SARS-CoV-2 infection 29. Household characteristics were captured consistent with recommendations of the World Health Organization for assessing household transmission 30. All participants implemented quarantine measures as mandated by local authorities and maintained a daily symptom diary which was collected and reviewed at each study visit by the study team (see online supplementary material). At any time during the trial, if a contact participant developed symptoms suggestive of COVID-19, a nasopharyngeal swab was taken by a member of a mobile health team for virus PCR testing in an accredited SARS-CoV-2 diagnostic laboratory. Index cases were instructed to remain in isolation / quarantine for 11 days from onset of symptoms or, if asymptomatic, 11 days from the sample collection date that resulted in the COVID-19 diagnosis. Household contacts remained in isolation / quarantine for 11 days from date of the sample resulting in the diagnosis of the index case or of a newly diagnosed household member as per recommendations/rulings by local authorities.
Intervention
A mobile health team conducted home visits of all participant households on study days 1 (enrolment), 6, 11, 16 and 29. Index cases and eligible household contacts in the IFNß-1α arm received three subcutaneous doses of pegylated IFNß-1α (125µg/0.5ml x 0.5ml) on study days 1, 6 and 11. Ineligible contacts in the IFNß-1α arm, as well as index cases and all household contacts in the SOC arm, received standard care. All participating households received information regarding hygiene, isolation, social distancing and wearing of facemasks as per public health advice at the time of enrolment. The IFNß-1α injection was given by a trained member of a mobile health team and participants were recommended to take paracetamol (1000mg, 6 hourly) commencing at the same time as the IFNß-1α for up to twenty-four hours, in order to mitigate predictable flu-like symptoms31.
Outcomes
The primary outcomes were (i) the proportion of index cases shedding SARS-CoV-2 at study day 11 in the IFNß-1α compared to control arm and (ii) the proportion of household contacts shedding SARS-CoV-2 at day 11, in the IFNß-1α compared to the control arm. Secondary and exploratory outcomes are listed in Supplementary Methods (Supplementary Table 1). Shedding was determined by the presence of SARS-CoV-2 by PCR in saliva collected on days 1, 6, 11, 16, 21, and 29 (see Supplementary Methods). A SARS-CoV-2 PCR Ct value ≥40 was considered negative. Viral load was estimated from the Ct value using a standard algorithm (Supplementary Figure 1
Mean cycle threshold (Ct) values are plotted on the y-axis, and log10 SARS-CoV-2 copies per mL (cp/mL) are plotted on the x-axis. Linear regression resulted in an R2 value of 0.9972, to fit the equation y = – 3.719x + 48.83.
Biospecimen collections
The full schedule of biospecimen collection is provided in the supplementary methods (Supplementary Tables 2 and 3. All consenting, non-eligible household contacts also provided biospecimens according to the schedule collection for non-eligible household contacts.
Adverse events
These were classified in accordance with Good Clinical Practice 32, as non-serious or serious and as related or unrelated to the trial medication.
Statistical considerations
The study was designed at the start of the pandemic and there were few data to guide sample size calculations. We therefore planned both frequentist inference and Bayesian analyses based on assumptions from the available data.
Households were randomized as individual clusters using a minimization technique (biased coin, P=0.7) in order to achieve balance between treatment arms, stratified by the number of people within the household 33.
Sample size calculations assumed, α=0.025 and power 90%, a zero correlation between outcome measures and that the household-wise, type 1 error rate would be below 0.05 and power above 80%. The estimated average household size was 4, based on census data 34.
Primary outcomes:
SARS-CoV-2 shedding at study day 11 by index cases: Data from Wuhan suggested that the proportion of untreated index cases still shedding virus on study day 11 would be ∼85% 35. Based on a two tailed Fisher’s exact test, a sample size of 278 index cases (139 per arm) would have 90% power to detect a difference in the proportion of individuals shedding virus at study day 11, if the proportion in the IFNß-1α arm was 65%.
SARS-CoV-2 shedding at study day 11 by household contacts: We estimated the secondary infection rate (transmission within the household) where there is an untreated index case would be 28% 36,37. Based on a stratified Cochran-Mantel-Haenszel test (two-tailed with intra-class correlation of 0.15), we estimated that a sample of 278 households (834 eligible household contacts) would have >90% power, at alpha 0.025, to detect an odds ratio of 0.5 for a reduction in transmission to a treated household contact.
Based on these assumptions and estimates, we aimed to recruit 310 households (allowing for a 10% drop-out rate) and 930 eligible household contacts.
We undertook 3 separate sets of analyses:
Standard frequentist analyses based on the primary outcomes as stated in the protocol and described in the Statistical Analysis Plan (available online) and included intention to treat and per protocol approaches .
Exploratory frequentist analyses based on a modified dataset that accounted for biological implausibility and effects that might only be associated with active treatment (Figure 1 and Supplementary Table 1 for further explanations).
Both frequentist analyses fitted linear mixed effects models using Ime4 38 (see Statistical Analysis Plan).
Exploratory Bayesian analyses using the same dataset as the exploratory frequentist analysis. Detailed methods are provided in the supplementary methods. Briefly, a generalized linear model with mixed effects was developed to estimate the probability of infection that is influenced by explanatory variables for each contact case. A stan_glmer with binomial logit function in rstanarm R package was used 39,40.
The frequentist approach allowed us to estimate the effects of IFNß-1α treatment on the risk of an individual becoming infected. Whereas the Bayesian analysis allowed us to determine whether the ring prophylaxis strategy using IFNß-1α reduces the probability of transmission within the household of an infected index case.
With regard to the filtered datasets, we observed, after locking the database, that a number of index cases had negative PCR results on the day of recruitment despite a previous positive diagnostic PCR test. In the exploratory analyses we therefore excluded households where the index case was unlikely to transmit virus because they were not shedding virus on days 1 and day 6 of the study. We also excluded households where an eligible contact was already positive at recruitment and there were no other eligible negative cases in the household. We refer to the modified dataset as the population “at risk”. We used the population at risk data to better understand the effects of IFNß-1α treatment on transmission of virus within households.
When undertaking these exploratory analyses, we considered effects during two time periods. Days 1-11 (Period 1) i.e., the treatment and isolation period when eligible contacts in the IFNß-1α arm were likely to have elevated IFN levels based on the known pharmacokinetics of pegylated IFNß-1α and days 12-29 (Period 2) the post-treatment and isolation period when biological effects of pegylated IFNß-1α were likely to be waning in treated individuals. Although samples were not obtained on days 12-15, we assumed that anyone who became infected during this period would likely still to have a positive PCR on day 16.
Results
Recruitment, participation, and completion data are shown in Figure 1. Participant demographic and basic household characteristics data are shown in Table 1.
Between December 2020 and May 2021, three-hundred and forty-one households were enrolled and randomized, of which 137 (IFNß-1α arm) and 151 (control arm) completed the study. Of the 1172 individuals randomized (IFNß-1α arm = 607; Controls = 565), 53 individuals withdrew from the study, of which 15 were index cases; 35 (14 index cases) in the IFNß-1α arm and 18 (1 index case) in the control arm. The reasons for withdrawal are summarized in the supplementary results (Supplementary Table 4). One index case withdrew before being allocated to a treatment arm. Eighty-two households where the index case had a negative salivary PCR on Day 1 and 6, or where there were no eligible contacts who tested negative at recruitment, were excluded from the exploratory analyses. The remaining households were considered as the ‘at risk’ population. (Figure 1).
IC: index case, EHC: Treatment eligible household contacts, CP-HC: COVID-19 positive household contacts, IHC: Treatment ineligible household contacts, IC-INF: IC & CP-HC, IC-ITT: Index case – intention to treat population, IC-PP: Index case – per protocol population, EHC-ITT: Treatment eligible household contacts – intention to treat population, HC: EHC and IHC, EHC-PP: Treatment eligible household contact – per protocol population.
Data is the mean, median, and 95% CrI difference in probability of infection in each study period, computed as probability of infection in the SOC arm minus the probability of infection in the IFN arm.
Data is the mean, median, and 95% CrI difference in probability of infection in each study period, computed as probability of infection in the SOC arm minus the probability of infection in the IFN arm.
(a) Primary analyses
The outcomes for the intention to treat (ITT) and per protocol (PP) frequentist analyses were similar (Supplementary Results). There was no effect of IFNβ-1α treatment on duration of viral shedding in those treated (index cases or infected contacts) vs. controls (IC-INF: treatment OR = 0.979, 95%CI = 0.647 to 1.479); IC-ITT: treatment OR = 1.106, 95%CI = 0.657 to 1.863); IC-PP: treatment OR = 1.062, 95%CI = 0.622 to 1.810); Figure 2A). The observed reduction in transmission to household contacts associated with IFNbeta-1alpha by day 11 was not signif icant (EHC-ITT: treatment OR = 0.582, 95%CI = 0.271 to 1.247; HC: treatment OR = 0.577, 95%CI = 0.287 to 1.160; EHC-PP: treatment OR = 0.589, 95%CI = 0.272to 1.276; Figure 2B). Vaccination status did not affect these outcomes (Supplementary results).
(A) odds ratios for index case populations in Primary Analysis 1 and associated sensitivity analyses, to compare the probability of being SARS-COV-2 negative vs positive on study day 11 in either treatment or standard of care arms. (B) odds ratios for household contact populations in Primary Analysis 2 and associated sensitivity analyses, to compare the probability of household contacts being SARS-CoV-2 negative vs positive on study day 11 in either treatment of standard of care arms. (C) odds ratios for the exploratory frequentist and Bayesian analysis, to determine if IFN is associated with a reduction in the number of positive COVID-19 tests in household contacts across study period 1 (days 1-11) and study period 2 (days 12-29). Treatment with IFN is associated with a significant reduction in the odds of a positive COVID-19 test for all household contacts compared to standard of care; frequentist analysis (p = 0.033; OR = 0.550, 95% CI = 0.364 to 0.989), Bayesian analysis (OR = 0.32, 95% CrI = 0.11 to 0.83). IC-INF = Infected case population (main analysis), IC-ITT = Index case – intention to treat population (sensitivity analysis 7.1.3.2), IC-PP = Index case – per protocol population (sensitivity analysis 7.1.3.3). EHC-ITT = Treatment-eligible household contact – intention to treat population (main analysis), HC = Household contact (eligible + non-eligible contacts) population (sensitivity analysis 7.2.3.2), EHC-PP = Treatment-eligible household contact = per protocol population (sensitivity analysis 7.2.3.3). * p < 0.05.
(b) Exploratory frequentist analysis of ‘At Risk Population.’
Of the at risk population, 33/164 (20%) in the IFNß-1α arm became infected by day 29 of the study compared to 37/142 (26%) in the control arm (relative risk reduction in the IFN arm = 23%). The majority of the risk reduction occurred during the active treatment phase of the study (days 1 to 11). Treatment with IFNß-1α was associated with a significant reduction in the odds of a positive saliva PCR for all household contacts compared to standard of care on Day 11 (p = 0.033; OR = 0.550, 95% CI = 0.364 to 0.989). The treatment effect was not significant in the second period (days 12-29) was not significant (OR = 0.968, 95% CI = 0.405 to 2.840) (Figure 2C)
(c) Exploratory Bayesian analysis:
Overall, there was a 95% probability of infection reduction within a household by IFNß-1α treatment and the credible interval for the reduction in transmission probability was in the order of 0.9 to 15.9%. During the active treatment period (days 1-11), there was a significant reduction in the odds of transmission (Bayesian analysis (OR = 0.32, 95% Credible Interval = 0.11 to 0.83). In contrast, in Period 2 (day 12-29), the 95% credible interval of Beta coefficients for the treatment group includes zero and therefore treatment was not effective in this period (Figure 2C, Supplementary Figure 2). The effect of IFNß-1α on transmission was independent of household size (Supplementary Figure 3)
(A) Distribution in study period 1, and (B) Distribution in study period 2. Vertical black line represents zero difference in probability of infection between SOC and IFN arms.
(A) Distribution in study period 1, and (B) Distribution in study period 2. Vertical dashed black line represents zero difference in probability of infection between SOC and IFN arms.
To determine the effect of IFNß-1α treatment on viral shedding in those who became infected, we examined the viral load trajectories in household contacts who became PCR positive after day 1. We did not observe any difference in viral load trajectory in those treated with IFNß-1α compared to controls (Figure 3A). Furthermore, there was no difference in peak viral load between treated household contacts and controls who became infected (Figure 3B).
(A) No significant difference in median viral load at each study visit between SOC and IFN arms. (B) No significant difference in peak viral load between SOC and IFN arms. SOC = standard of care arm, IFN = interferon arm, cp/mL = copies/mL. n = 57 (IFN) & n = 61 (SOC) participants per group.
There were fifty-eight serious adverse events in index cases, twenty-seven hospitalizations due to COVID-19 (25 in the treatment and 32 in the control arm) and was one death in the standard care arm due to COVID-19 in an individual with significant co-morbidities. Twenty-six household contacts of index cases were hospitalized, 10 in the treatment arm and 16 in the control arm, the difference was not statistically significant.
Discussion
This prospective, household cluster randomized, ring prophylaxis trial demonstrated that pegylated IFNß-1α may reduce household transmission of SARS-CoV-19. Pegylated IFNß-1α is an FDA approved therapy for hepatitis C and multiple sclerosis for which the pharmacokinetics and safety profile are well-characterized 41. The use of this formulation allowed us to predict the likely duration of activity of the three-dose regimen in order to cover the period of peak transmissibility of the virus. Using a mobile medical team to administer doses at home optimized adherence to therapy and allowed reasons for non-adherence and withdrawal to be accurately documented.
Although the intention to treat and per protocol analyses failed to demonstrate a significant effect on primary outcomes, these analyses failed to account for index cases not shedding virus at randomization nor an absence of eligible contacts within a household. Using biologically plausible filters to define an at-risk population provided a better understanding of the effects of IFNß-1α than intention to treat or per protocol analyses. These exploratory analyses that took account of the likelihood that transmission could occur within a household provided a pragmatic assessment of the effects of IFNß-1α therapy when given to household contacts as prophylaxis.
The frequentist analysis of the at-risk populations indicated that individuals treated with IFN were less likely to become infected than untreated individuals when exposed to virus with a relative risk reduction of 23%. The Bayesian analysis revealed that the probability was 95% that transmission was reduced in households randomized to IFNß-1α therapy. This effect of IFNß-1α on household transmission was only significant during the active treatment period. This highlights the physiological importance of IFNß-1α as a sentinel molecule 24 and provides further evidence for post-exposure prophylaxis with IFNß-1α as an antiviral strategy.
To our knowledge, only one previous study assessed ring prophylaxis in COVID-19. Labhardt et al observed that a combination of lopinavir/ritonavir for 5 days as post-exposure prophylaxis was not effective at preventing infection in close contacts of index cases 42.
The overall viral load trajectory we observed was very similar to that observed in a laboratory human challenge experiment using wild-type SARS-CoV-2 virus in healthy volunteers 43. Treatment with IFNß-1α had no effect on viral load trajectory; therefore, reduced viral shedding by infected individuals is unlikely to explain the protective effect on household transmission of IFNß-1α treatment that we report . Given that treatment affected the probability of transmission only in the active treatment phase and appears unrelated to viral load, we speculate that the observed effects of IFNß-1α were direct through protection of the at-risk, exposed individual rather than indirectly though effects on the index case. The ConCoRD-19 biorepository will allow further examination of the mechanisms of action of IFNß-1α on resistance to infection.
The study was undertaken prior to the emergence of the omicron strain of virus. We calculated the sample size based upon known household transmission characteristics of the alpha strain which was the dominant strain of the virus early in the pandemic. Simple hygiene measures and quarantine of affected individuals within households could have contributed to lower rates of transmission than expected. Together with a smaller number of eligible household contacts than anticipated from census data, these factors may have reduced the power of the study for the primary outcomes. The effect size observed in this study could be greater for variants with higher transmissibility such as the Omicron strains that are now the dominant worldwide.
Nonetheless, the beneficial biological effect of IFNß-1α treatment was evident in the at-risk population and therefore, post-exposure or ring prophylaxis with IFNß-1α warrants further investigation, particularly if additional virulent strains of SARS-CoV-2 emerge and for future pandemic virus preparedness.
In summary, although intention to treat and per protocol analyses failed to demonstrate significant effects on primary outcomes, in a biologically defined, at risk population, the relative risk of infection for an individual treated with pegylated IFNß-1α was reduced by 23.0%. Moreover, this is the first study to demonstrate ring prophylaxis with pegylated IFNß-1α significantly reduces the probability of household transmission of SARS-CoV-2. The effect of pegylated IFNß-1α was most obvious during the active phase of treatment. In situations where there is emergence of a highly pathogenic and transmissible mutant strain of SARS-CoV-2 despite high vaccination rates, or when there is a new viral pandemic, IFNß-1α ring prophylaxis could be considered to reduce transmission in households or amongst critical workers and vulnerable communities. Ring prophylaxis with therapies that can interrupt transmission appears to be an effective strategy for highly contagious respiratory viruses.
Funding
Biogen PTY Ltd. Supply of interferon as “Plegridy”.
The study was substantially funded by BHP Holdings Pty Ltd
Authorship
Authors, Castro-Rodriguez; Fish; Kollmann; Iturriaga, Karpievitch and Shannon contributed equally to the conduct of the study and manuscript as first authors.
Authors, Tebbutt, Diego García-Huidobro, Perret, Borzutsky and Stick contributed equally to the conduct of the study and manuscript as senior authors.
Data Availability
All data produced in the present study are available upon reasonable request to the authors
List of Study Investigators
Jose A. Castro-Rodriguez
Department of Pediatric Pulmonology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Eleanor N. Fish
Department of Immunology, University of Toronto, Toronto, Canada
Toronto General Hospital Research Institute, University Health Network, Toronto, Canada
Tobias Kollman
Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Carolina Iturriaga
Department of Pediatric Infectious Diseases and Immunology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Millennium Institute on Immunology and Immunotherapy, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Yuliya Karpievitch
Wal-yan Respiratory Research Centre, Telethon Kids Institute, University of Western Australia, Nedlands, Australia
School of Biomedical Sciences, University of Western Australia, Nedlands, Australia
Casey Shannon
Prevention of Organ Failure (PROOF) Centre of Excellence, Vancouver, Canada
Centre for Heart Lung Innovation, Providence Health, St Paul’s Hospital, Vancouver, Canada
Robert Balshaw
Prevention of Organ Failure (PROOF) Centre of Excellence, Vancouver, Canada
Centre for Heart Lung Innovation, Providence Health, St Paul’s Hospital, Vancouver, Canada
Virginia Chen
Prevention of Organ Failure (PROOF) Centre of Excellence, Vancouver, Canada
Centre for Heart Lung Innovation, Providence Health, St Paul’s Hospital, Vancouver, Canada
Samuel T. Montgomery
Wal-yan Respiratory Research Centre, Telethon Kids Institute, University of Western Australia, Nedlands, Australia
School of Population Health, Curtin University, Bentley, Australia
Joseph Ho
Wal-yan Respiratory Research Centre, Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Rad Aniba
Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Rym Ben Othman
Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Francisca Gidi-Yunge
School of Dentistry, Faculty of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Lucy Hartnell
Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Guillermo Pérez-Mateluna
Department of Pediatric Infectious Diseases and Immunology, School of Medicine, Pontificia Universidad Católica De Chile, Santiago, Chile
Marcela Urzúa
Department of Pediatric Infectious Diseases and Immunology, School of Medicine, Pontificia Universidad Católica De Chile, Santiago, Chile
Scott Tebbit
Prevention of Organ Failure (PROOF) Centre of Excellence, Vancouver, Canada
Centre for Heart Lung Innovation, Providence Health, St Paul’s Hospital, Vancouver, Canada
Diego García-Huidobro
Department of Family Medicine, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Department of Family Medicine and Community Health, University of Minnesota, United States of America
Cecilia Perret
Department of Pediatric Infectious Diseases and Immunology , School of Medicine, Pontificia Universidad Catolica de Chile, Santiago, Chile
Arturo Borzutzky
Department of Pediatric Infectious Diseases and Immunology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Millennium Institute on Immunology and Immunotherapy, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile
Stephen M. Stick
Wal-yan Respiratory Research Centre, Telethon Kids Institute, University of Western Australia, Nedlands, Australia
Perth Children’s Hospital, Nedlands, Australia
Supplementary Methods
Real-time Quantitative Polymerase Chain Reaction (RT-qPCR) for SARS-CoV-2
RNA extraction and RT-qPCR for SARS-CoV-2 detection was performed at the Infectiology and Molecular Virology Laboratory, Red Salud UC Christus, Santiago, Chile. RNA was extracted in parallel from saliva using the MagBind RNA extraction kit (#GN7101907, Maccura Biotechnology, Rockville, MD, USA) with the Auto-Pure 32A Nucleic Purification System (Maccura Biotechnology, Rockville, MD, USA), and using the RNA/DNA Purification Magnetic Bead Kit (#DA0630, DaAnGene, Guangzhou, CN) with the Smart32 Nucleic Acid Extraction Instrument (DaAnGene, Guangzhou, CN) according to manufacturer specifications. Real-time PCR amplification was performed using the LightMix Modular Wuhan CoV RdRP-gene kit (#53-0777-96, TIB MolBiol, Roche, Berlin, DE) targeting the RNA-dependent RNA polymerase (RdRP) gene, targeting SARS-COV-2, but not other SARS-like viruses (CoVNL63, CoV229E, HKE, OC43, MERS), with a positive control containing three separate diagnostic targets (E gene, N gene, RdRP). The reverse transcription and PCR processes occur in a single step in a real-time thermocycler (Lightcycler 480 II, Roche, Berlin, DE) with an analytical sensitivity of 10.6 copies per reaction. A cycle threshold (Ct) value of ≤ 37 was considered positive for SARS-CoV-2.
Viral load calculation
The viral load of SARS-CoV-2 detected in saliva was calculated using a standard curve. A commercial SARS-CoV-2 standard (#COV019, Exact Diagnostics, Fort Worth, TX, USA) at a concentration of 200,000 copies per mL (cp/mL) was extracted in triplicate using the RNA/DNA Purification Magnetic Bead Kit (#DA0630, DaAnGene, Guangzhou, CN) with the Smart32 Nucleic Acid Extraction Instrument (DaAnGene, Guangzhou, CN) according to manufacturer specifications. Following extraction, serial dilutions (1:10; 200,000cp/mL to 20cp/mL & 1:5; 200,000cp/mL to 64cp/mL) were used in triplicate for reverse transcription and RT-qPCR using the LightMix Modular Wuhan CoV RdRP-gene kit (#53-0777-96, TIB MolBiol, Roche, Berlin, DE) with the Lightcycler 480 II thermocycler (Roche, Berlin, DE), resulting in 9 replicates at each concentration. SARS-CoV-2 was detectable in all nine replicates at 200,000cp/mL, 40,000cp/mL, 20,000cp/mL, 8,000cp/mL, 2000cp/mL, and 1,600cp/mL, and a standard curve was generated using the mean Ct values for these concentrations (Supplementary Figure 1) to calculate SARS-CoV-2 viral load in samples of unknown titre, resulting in a lower limit of quantification of 1,600cp/mL.
Populations used for analyses
Participants were excluded for analysis as outlined in the statistical analysis plan, or as outlined in the exploratory analysis (Supplementary Appendix B). Total households and participants excluded and used in each population are described in Supplementary Table 6. Briefly, participants were excluded from the populations used in the primary analyses (IC-INF, IC-ITT, EHC-ITT, HC) if the study visit was not performed, or was performed outside the +/– 1 day window either side of the scheduled visit date (Statistical Analysis Plan 6.5). Participants who didn’t complete the full course of the treatment were excluded from the Per Protocol populations used in the primary analyses (IC-PP, EHC-PP). In the exploratory analysis, whole households were excluded if the index case was SARS-CoV-2 negative via salivary PCR at study days 1 & 6, and if there were no SARS-CoV-2 negative eligible or ineligible contacts in the household at study day 1. Participants were then excluded from the exploratory analysis if they were SARS-CoV-2 positive via salivary PCR at study day 1. Participants were further excluded from the Bayesian analysis where household size was greater than 7, as these household sizes were unable to be modelled using Bayesian modelling.
Bayesian statistical analysis
Bayesian analysis was included in the original statistical analysis plan. The goal was to test the null hypothesis that treatment of household contacts with IFNβ-1α does not reduce the probability of transmission from an infected index case to household contacts. As outlined in the analysis plan, we used data to develop a data generating process governed by probabilities that a household contact of an index case would test positive to COVID-19, and non-informative prior was used.
Bayesian model
A generalized linear model with mixed effects was developed to estimate the probability of infection for each contact. A stan_glmer with binomial logit function from the rstanarm package (1) was run using R (version 4.1.1) to obtain Beta coefficients for 4,000 simulated samples to estimate the probability of infection using the following model:
pcr ∼ (1 | index) + vacc + viral + infected + eligible + sex + age + arm
where: pcr: binary COVID-19 test result, index: index case in the household – treated as a random effect, vacc: number of vaccinated household members, viral: log10 viral load of the index case at the beginning of the period, infected: number of COVID-19 positive household members at the beginning of the period, eligible: whether the household contact was eligible for treatment, sex: sex of the household contact, age: age of the household contact, and arm: study arm (i.e. SOC/IFN) of the household were the fixed explanatory variables.
The features of the model were the same as the exploratory frequentist analysis, with the exception of adding the treatment period as a feature. Instead, the Bayesian analysis split the data into two periods (Period 1: study days 1–11, Period 2: study days 12–29) with analysis conducted separately. For period 2, COVID-19 positivity at day 11 was used to determine COVID-19 positive household members at the beginning of the period. The population used for analysis was the same as the “At-risk” population used in the exploratory frequentist analysis (Supplementary Table 6), with the additional exclusion of participants in households with a size larger than 8.
Computation of difference in probability of infection in IFN vs SOC arms
In each simulation, we obtained a PCR outcome for each contact from the Bayesian model by applying the posterior_predict function in the rstanarm package to the observed data (1). We further computed the percent of infections in each treatment arm irrespective of household size (Supplementary Table 8) and for each household size (Supplementary Table 9). Furthermore, we computed the difference in probability of infection by subtracting the probability of infection in IFN arm from the probability of infection in SOC arm. As a results, we obtained 4000 differences (irrespective of household size) and 4000 differences for each household size
Significance of the treatment effect was determined by 95% credible interval (CrI) for the treatment Beta coefficient. If 0 was included in the CrI, the effect of treatment group was considered not significant. Average Beta coefficient and odds ratio for the effect of the IFN treatment as compared to SOC were computed in each study period. The average odds ratio of being infected with SARS-CoV-2 on IFN treatment vs. SOC was calculated by exponentiating the values of the average of the coefficients (Supplementary Table 7).
Supplementary Results
Bayesian analysis of infection reduction
Treatment with IFN had a significant effect in study period 1 as the 95% CrI for treatment coefficient did not include zero (Supplementary Table 7). However, IFN treatment did not have a significant effect in period 2 as the 95% CrI did include zero (Supplementary Table 7).
In study period 1, there was 95% probability of infection reduction between 0.9% to 15.9% due to the IFN treatment, with 8.5% median and mean infection reduction (Supplementary Table 8). In study period 2, there are 95% probability of infection reduction -6.7% to 4.8% due to the IFN treatment, with -9% of median and mean of infection reduction (Supplementary Table 8).
When we stratified the infection reduction by household size, only households’ size 3 and 4 in period 1 had a 95% CrI without zero (Supplementary Table 9).
As the distribution of Bayesian estimated differences in probabilities of infection in study period 1 is centred to the right of 0 (Supplementary Figure 3A), IFN treatment had an effect on reducing SARS-CoV-2 transmission within households. However, as the distribution in study period 2 is centred around 0, it suggests IFN has no effect on SARS-CoV-2 transmission within households (Supplementary Figure 3B).
Supplementary Appendices
Acknowledgements
Matthew Cooper (Telethon Kids Institute). Advice regarding protocol and establishment of randomization schedule.
Alexia Foti (Telethon Kids Institute). Collating, reviewing, and editing documents.
Zsuzsanna Hollander (PROOF Centre, Vancouver). Statistical advice and statistical analysis plan review.
Nat Eiffler (Telethon Kids Institute). Project management, protocol development, data management. Jessica Meyer (University of Rochester). Assisting collation and formatting of tables and figures.
References
Supplementary References
- 1.
