Randomized trial of the safety and efficacy of anti-SARS-CoV-2 mAb in the treatment of patients with nosocomial COVID-19 (CATCO-NOS)

Abstract

Background:

Patients with nosocomial acquisition of COVID-19 have poor outcomes but have not been included in therapeutic trials to date.

Methods:

A pragmatic open-label randomized controlled trial of anti-SARS-CoV-2 monoclonal antibodies (mAb) was performed in hospitalized patients with nosocomial COVID-19 infection in acute care hospitals spanning a provincial health care network. Participants within 5 days of first positive test or symptom onset were randomized to standard of care (SOC) plus a single dose intravenous mAb treatment (bamlanivimab or casirivimab/imdevimab) or SOC alone on a 2:1 basis. The primary study endpoint was the need for invasive mechanical ventilation (IMV) or inpatient mortality by day 60 after randomization.

Results:

Forty-six participants were enrolled from 13 hospitals between February 14 and October 8, 2021: 31 in the mAb and 15 in the SOC arm. IMV or inpatient mortality up to day 60 occurred in 4 (12.9%) participants in the mAb versus 3 in the SOC arm (20.0%), difference of −7.1% (95% CI −22.5 to 13.4, p = 0.67). The study was terminated early due to lack of equipoise as effectiveness of anti-viral therapies and mAb was published in similar high-risk patient populations.

Conclusions:

The trial was underpowered to detect meaningful differences given its early termination. The study does highlight the feasibility of undertaking trials in this patient population using a pragmatic approach allowing for trial participation and treatment access across a large health care network and may serve as a template for future designs.

Introduction

Nosocomial acquisition of SARS-CoV-2, leading to COVID-19 is a significant concern across acute and long-term care settings. Nosocomial infection can occur from contact with health care providers (1), patient roommates (2,3) or visitors, all more probable during periods of high community prevalence of infection. Nosocomial COVID-19 infections are associated with substantial morbidity and mortality—higher than what is seen in community acquired infection (4–7). A study from the United Kingdom reported that 27% of 196 patients identified with nosocomial acquisition of COVID died (4) while a larger German study reported an 11% mortality rate although outpatient health care acquisitions were also included in this cohort (5). A meta-analysis suggested that mortality rates were 1.3 time greater in nosocomial cases versus community acquired hospitalized patients (6). Given that only a minority of community cases require admission to hospital, the actual relative increase in mortality from nosocomial acquired COVID-19 may be much greater.

Minimal evidence exists as to optimal treatment approaches for nosocomial COVID-19 infection which would be important to establish since interventions readily available in acute care hospitals and early treatment may improve outcomes (8–11). Accordingly, we designed a randomized clinical trial (RCT) to derive preliminary estimations of efficacy and safety of anti-SARS-CoV-2 monoclonal antibodies (mAb) for the treatment of patients with hospital-acquired COVID-19.

Methods

This study was an adaptive, pragmatic, randomized, open-label, controlled clinical trial. Eligible patients were 18 years of age or older, admitted to an acute care hospital in the province of Alberta, Canada for a non-COVID-19 diagnosis, had a laboratory-confirmed SARS-CoV-2 infection as determined by reverse transcriptase polymerase chain reaction (RT-PCR) of throat or nasopharyngeal swab specimens on the third day of admission or later and were within 5 days of this diagnosis based on test collection date or initial development of symptoms, which ever was earliest. Exclusions included a clinician plan for palliation only treatment, World Health Organization (WHO) ordinal scale (12) of 6 or higher (ie, hospitalized, on high flow nasal oxygen or non-invasive ventilation), non-medical/surgical reason for admission (eg, obstetric or psychiatric), pregnancy or breast feeding, or weight less than 40 kg. The trial was approved by the University of Calgary Conjoint Health Research Ethics Board (REB21–0096) and the University of Alberta Health Research Ethics Board (PRO00107664) and registered with clinicaltrials.gov (NCT04748588). The study protocol is provided in the supplemental materials. Specific funding was not obtained for this study, which was supported through in-kind contributions from research groups and Alberta Health Services as part of pandemic mobilization.

Subjects were randomized by study personnel using the REDCap (13) Electronic Data Capture system in a ratio of 1:2 to standard of care treatment or anti-SARS-CoV-2 mAb treatment plus standard of care treatment. Stratification was performed by baseline (pre-hospitalization) Clinical Frailty score (14) (dichotomized as <5 vs. ≥5) in blocks of 12.

Participants randomized to the treatment arm received a single intravenous dose of an anti-SARS-CoV-2 mAb. The mAbs were provided in-kind from Alberta Health Services. As part of a planned adaptive design, the specific antibody used evolved through the course of pandemic as new products were approved by Health Canada and sensitivity patterns of new variants of concern changed (Figure 1). The alpha (B.1.1.7) variant of concern became prevalent by March of 2021 in trial hospital catchment areas, with a small proportion of the gamma (P.1) variant of concern present in April and May of 2021. As of July 2021, until study closure, the delta (B1.617.2) variant of concern was the dominant strain in circulation. The initial product used was bamlanivimab (700 mg IV), which was removed as an option once evidence suggested resistance of the delta (B1.617.2) strain (15,16). Casirivimab/imdevimab (1,200 mg/1,200 mg IV) combination product was then used as of July 26, 2021. Sotrovimab (500 mg) was also added as an option on September 24, 2021 but no patients were randomized to this treatment. Standard of care may have evolved through the course of the study. Province wide recommendations and guidelines for inpatient management of COVID-19 were available and updated throughout the course of the pandemic and were widely disseminated to health care providers (Appendix A). Vaccines initially became available in early January 2021 for high-risk individuals and health care providers.

Figure 1:

Timeline of variants of concerns in Alberta, Canada over the course of the study

Study endpoints

The primary study endpoint was estimation of clinical effectiveness of anti-SARS-CoV-2 mAb treatment relative to standard of care on the need for invasive mechanical ventilation (IMV) or inpatient mortality, both by day 60 after randomization. Secondary endpoints included inpatient mortality, treatment with IMV, intensive care unit (ICU) admission, hospital length of stay (LOS) after randomization, and safety based on clinician’s assessment of adverse events (AE) at 24 hours.

Recruitment and outcome assessment

Patients could be referred to the study by attending clinical teams at any of 86 hospitals with acute care inpatient beds in the province of Alberta in collaboration with the two Provincial Health Care Agencies, either Alberta Health Services (AHS) or Covenant Health. Clinical teams were made aware of the trial through engagement with Provincial Infection Prevention and Control teams as well as through direct contact by research coordinators who could track any declared inpatient outbreaks on a Provincial electronic dashboard. Once patient or designated representatives gave permission to be contacted, care providers submitted basic information on the patient through a Research Electronic Data Capture (REDCap) (13) survey, after which central coordinators contacted the patient or designated representative for additional information, study consent, and enrollment. If randomized to the treatment arm of the study, the mAb product was ordered by the investigators or clinical teams through the local hospital pharmacy using clinical procedures and processes. Given the large number of hospitals and the limited supply of treatment doses, a hub and spoke model was used with stock maintained in regional centres available for rapid transfer to smaller hospitals as needed. An electronic follow-up survey was sent to the enrolling clinician at 24 hours to assess for any adverse events. All other research endpoints were extracted from central AHS databases through the Alberta Support for Patient-Oriented Research (SPOR) support unit. To validate the central database approach for outcomes assessment, a post hoc review of four key items extracted from the AHS databases (date of admission and discharge, need for mechanical ventilation and in-hospital death) was performed via manual review of discharge summaries for all participants.

Statistical analysis

Data were analyzed using SPSS v28 (IBM Corporation, Chicago, IL, USA). Group characteristics and main outcome measures were compared with two-sample independent t-tests, Pearson chi-square tests or Fisher exact tests as appropriate.

Study sample size and early termination

Sample size calculations for the primary endpoint were based on estimates of hospital mortality in other studies—previously reported in the 25%–30% range for nosocomial infection; and, ICU admission rates—variable by region and patient population, but local estimates ranged between 10% and 20%. For an 80% power and a 0.05 alpha, a sample size of 648 patients (216 control, 432 intervention) was selected to detect an absolute between-group difference of 10% for the primary outcome.

In October of 2021, in consultation with the broader Canadian Trial for the Treatment of COVID-19 (CATCO) investigators planning to launch the trial in other Canadian Provinces (17) and the Alberta Health Services (AHS) COVID-19 Therapeutics Committee, trial investigators reviewed evolving evidence supporting early outpatient treatment of COVID-19 in high-risk individuals with mAb. As these treatment options became available for outpatients in Alberta, and with increased access to remdesivir for inpatients, it was no longer felt justifiable to randomize nosocomial inpatients to a standard of care arm since many eligible patients would likely qualify for such therapies on an outpatient basis, and participants randomized to SOC given remdesivir instead. The trial was therefore closed to new enrollment on November 24, 2021.

Results

Forty-six participants were enrolled in the study from 13 hospitals between February 14 and October 8, 2021: 31 in the mAb plus standard of care arm, and 15 in the standard of care arm (Figure 2). Subject demographics and characteristics are outlined in Table 1. Participant mean age was 75 (SD 14.2) years and 67.4% were male. The majority had a high frailty score (60.9% ≥5). They had been hospitalized for a mean of 28.9 (SD 31.4) days at the time of COVID-19 diagnosis and enrolled in the study 1.7 days after initial symptoms or test positivity, with the mAb administered within 24 hours of enrollment in those in the treatment arm. Fifty percent of participants were hospitalized in a regional hospital, while 32.6% were in a tertiary and 17.4% in a local hospital. Infection was determined to be from the native SARS-CoV-2 strain in 54.3%, the B.1.1.7 (alpha) strain in 32.6%, and B.1.617.2 (delta) strain in 13% of participants. The majority of participants (67.4%) did not require supplemental oxygen at the time of enrollment.

Figure 2:

CONSORT diagram

Table 1:

Participant demographics

	No. (%) unless otherwise specified
	Study cohort; n = 46	SOC cohort; n = 15	mAb cohort; n = 31	mAb cohort; n = 31	p-value*
Sex (male)	31 (67.4)	10 (66.7)	21 (67.7)	1
Age, mean (SD), y	75 (14.2)	73.9 (15.1)	75.5 (14.0)	0.717
Race (Caucasian)	40 (87.0)	11 (73.3)	29 (93.5)	0.078
Frailty score				0.045
2	1 (2.2)	0 (0)	1 (2.2)
3	7 (15.2)	4 (26.7)	3 (9.7)
4	10 (21.7)	2 (13.3)	8 (25.8)
5	8 (17.4)	5 (33.3)	3 (9.7)
6	14 (30.4)	1 (6.7)	13 (41.9)
7	6 (30.4)	3 (20)	3 (9.7)
Frailty score (≥5)	28 (60.9)	9 (60.0)	19 (61.3)	0.933
Initial hospitalization type (medical)	44 (95.7)	14 (93.3)	30 (96.8)	1
Time from admission to COVID-19 diagnosis, mean (SD), d	28.9 (31.4)	24.2 (19.8)	31.1 (36.0)	0.491
Time from COVID-19 diagnosis to enrollment, mean (SD), d	1.7 (1.2)	2.0 (1.1)	1.6 (1.2)	0.266
Variant of concern				0.119
Native strain/unknown	25 (54.3)	8 (53.3)	17 (54.8)
B.1.1.7 alpha	15 (32.6)	7 (46.7)	8 (25.8)
B.1.617.2 delta	6 (13.0)	0 (0)	6 (19.4)
Co-morbidities
BMI, mean (SD), kg/m2	25.7 (5.4)	24.6 (4.9)	26.2 (5.6)	0.348
Obesity	10 (22.7)	2 (13.3)	8 (25.8)	0.452
Diabetes	22 (47.9)	5 (33.3)	17 (54.8)	0.346
Chronic kidney disease	12 (26.1)	3 (20.0)	9 (29.0)	0.629
Chronic kidney disease with dialysis	2 (4.3)	0 (0)	2 (6.5)
Respiratory disease	17 (37.0)	5 (29.4)	12 (38.7)	0.723
Hematologic malignancy	3 (6.5)	1 (6.7)	2 (6.5)	1
Cancer	3 (6.5)	0 (0)	3 (9.7)	0.541
Cardiovascular disease	11 (23.9)	1 (6.7)	10 (32.3)	0.074
Cirrhosis	2 (4.3)	1 (6.7)	1 (3.2)	1
Ordinal scale at randomization				0.74
4—no oxygen theraphy	31 (67.4)	11 (73.3)	20 (64.5)
5—oxygen by mask or nasal prongs	15 (32.6)	4 (26.7)	11 (35.5)
Concommitant COVID-19 therapies
Dexamethasone	12 (26.1)	2 (13.3)	10 (32.3)	0.285
Remdesivir	3 (6.5)	2 (13.3)	1 (3.2)	0.244
Tocilizumab	1 (2.2)	1 (6.7)	0 (0)	0.326

^* Fisher exact test, Pearson chi-square or t-test

mAb = Monoclonal antibody; SOC = Standard of care

In the mAb treatment arm, bamlanivimab was administered to 21 patients and casirivimab/imdevimab to 10. One patient randomized to mAb received bamlanivimab for a delta variant infection in June 2021. No patients in the standard of care arm received an anti-SARS-CoV-2 mAb, and three patients (two in mAb arm and one in standard of care arm) received intravenous remdesivir.

Outcomes

Mechanical ventilation or inpatient death by day 60 occurred in 4 (12.9%) participants in the mAb arm versus 3 in the standard of care arm (20.0%), a difference of −7.1% (95% CI −22.5 to 13.4, p = 0.667). In-hospital mortality was similar: 4/31 (12.9%) in the mAb arm; and 2/15 (13.3%) in the standard of care group (p = 1.0).

Post study enrollment length of hospital stay for individuals discharged alive was 29.5 days in the treatment arm versus 26.7 days for the standard of care group (difference 2.8 days [95% CI −17.7 to 23.3], p = 0.782). Three patients in the mAb arm (9.7%) versus one in the standard of care group (6.7%) remained in hospital at day 60 post enrollment (p = 1.0).

Review of discharge summaries confirmed admission and discharge date, need for mechanical ventilation, and in-hospital mortality in all 46 patients except for one file where no discharge summary was located and discharge date could not be corroborated.

Forty-four of 46 (95.6%) follow-up electronic surveys were returned by referring physicians. No infusion reactions requiring discontinuation of the mAb infusion was reported. No significant side effects were reported at 24 hours post infusion (Table 2).

Table 2:

Adverse events reported at 24 hours

	SOC cohort, no. (%); n = 15	mAb cohort, no. (%); n = 31	p-value*
Nausea	1 (6.7)	2 (6.5)	1
Diarrhea	0 (0)	2 (6.5)	1
Headache	2 (13.3)	0 (0)	0.101
Fever	1 (6.7)	1 (3.2)	1

^* Fisher exact test

Discussion

We report the results of a RCT in patients with nosocomial SARS-CoV-2 infection, which we believe is the first such randomized trial in this specific population. Early termination of the trial due to evolving evidence on effective mAb treatments for early COVID-19 in outpatients resulted in an underpowered study for the trial outcomes as well as inappropriately large randomization block sizes for this reduced sample size. The trial did however confirm the feasibility of a decentralized approach to a treatment RCT in a large health care (Provincial) network of hospitals. Collaboration and integration of the trial within the Provincial health care agency (AHS), was also critical to implementation of the study. The mAbs were supplied by AHS in-kind who in turn had procured them from the Public Health Agency of Canada. At the time, there was no evidence supporting inpatient use of these treatments and little capacity for an outpatient administration program given the need to focus resources on vaccination efforts. As such, value was seen by AHS and its COVID-19 therapeutic committee to offer these treatments on a research trial basis to a particularly vulnerable group of patients for whom no proven therapies were available. The trial leveraged local clinicians to refer participants and gather basic clinical information as well as local hospital pharmacies and patient units to dispense and administer the mAb. AHS infection Prevention and Control clinicians were also mobilized to inform local teams of the trial upon declaring and investigating nosocomial COVID-19 outbreaks, and study coordinators could also identify new outbreaks through a Provincial electronic dashboard and send study information to the affected patient units. This approach was important for trial recruitment given the sporadic nature of nosocomial COVID-19 outbreaks. We also demonstrated the feasibility of using centralized Provincial administrative health care databases to determine key trial outcomes. The accuracy in data extraction for key endpoints of hospital length of stay, need for mechanical ventilation and in-hospital mortality was corroborated with manual file review but could be a more efficient method to collect information in a decentralized study particularly in the midst of a pandemic when resources are already limited. Such pragmatic designs may be helpful for rapid and cost-effective implementation of trials across health care systems. This approach could also help bring research out of academic centres and closer to the patient bedside and ensure more equal access to investigational therapies—especially in areas with large rural populations.

Limitations of our approach include limitations of outcomes assessment to reliably collect information in administrative databases. Some information, such as need for supplemental oxygen, could not be obtained beyond the baseline assessment. Other endpoints such as side effects or infusion reactions monitored through an electronic questionnaire could not be validated. Nevertheless, in an open label study, limiting outcome measures to hard endpoints likely reduces the risk of bias. In addition, the pragmatic approach leveraging clinical processes for treatment administration could not be used for non-approved investigational products and would not fulfill requirements for regulatory approval. We defined our population as those with nosocomial acquisition of COVID-19, but it is possible that some patients were already infected and pre-symptomatic at the time of admission but only tested positive or developed symptoms after 72 hours in hospital.

While the decision to close the trial early did not allow us to determine the impact of mAb on outcomes from COVID-19 in these high-risk individuals, other trials in high-risk outpatients have shed light on early antiviral treatments against this disease. Bamlanivimab, which unfortunately is generally ineffective against most variants of concern, was first shown to reduce emergency room visits or hospitalizations by 4%–5% in the BLAZE-1 trial performed prior to vaccine availability or the emergence of variants of concern (18). The casirivimab/imdevimab combination as well as sotrovimab have also been shown to reduce hospitalizations in high-risk unvaccinated patients (10,19). As well, antiviral drugs such as remdesivir (9) and nirmatrelvir and ritonavir (11) have demonstrated reduction in progression to severe illness in this patient population. Patients with nosocomial SARS-CoV-2 infection are typically at higher risk for poor outcomes but may be identified early in the course of illness, unlike the population of patients admitted for COVID-19 (who are often at a later phase of illness). As the demographics of patients with nosocomial infection overlap to a large extent with patients eligible for these outpatient therapies, the decision to end the trial due to loss of equipoise was made.

Reducing the impact of nosocomial transmission of SARS-CoV-2 should begin with preventative measures which should include ensuring high levels of vaccination for both patient and health care providers, careful and routine monitoring for new symptoms and rapid testing should these occur, diligent use of personal protective equipment (PPE) (20), and isolation of symptomatic or positive inpatients. Pre (21) and post (8) exposure prophylaxis could also be considered, particularly in unvaccinated individuals or those who do not mount appropriate responses to vaccines due to illness or medications. Finally, early mAb therapy (if an agent is available for currently circulating variants of concern) or an anti-viral agent should be considered on an inpatient basis. This trial demonstrates the feasibility of rapid therapy for nosocomial infection.

In summary, we report on a RCT of mAb therapy in the treatment of nosocomial SARS-CoV-2 infection, highlighting the feasibility of trials in this patient population using an efficient, pragmatic approach allowing for trial participation and treatment access across a large health care network. Unfortunately, the trial was underpowered to detect meaningful differences given its early termination in the setting of the rapidly emerging evidence on COVID-19 therapies during this phase of the pandemic. Nonetheless organization and provincial implementation of this trial demonstrates the ability to quickly pivot and initiate an adaptive platform that would allow eligibility of any patient with hospital-acquired COVID-19 across an entire health region and serves as a template for future designs.

Appendix A

Evolving standard of care over the study period (February 14 and October 8, 2021)

Standard of care evolved through the course of the study. Province wide recommendations and guidelines for inpatient management of COVID-19 were available and updated throughout the course of the pandemic and were widely disseminated to health care providers through emails, newsletters, reference websites, inpatient unit posters, and continuing medical education events.

In general, dexamethasone was recommended for inpatients with hypoxemia during the entire study period.

Prior to April 2021, Remdesivir was only accessible through another trial (17) which only enrolled hypoxemic patients during this period. Request for clinical Remdesivir access could be made on a case-by-case basis (eg, immunocompromised patients) until October 2021 when access was broadened to any hypoxemic but not mechanically ventilated patient.

Tocilizumab became available for patients with severe respiratory failure in April 2021, defined as requiring more than 6 L/min of oxygen.

Casirivimab/imdevimab became available in September 2021 for hospitalized seronegative patients.

Full dose low-molecular weight heparin was recommended for non-critically ill COVID-19 with low-bleeding risk in October 2021.

No specific recommendations were offered for nosocomial COVID-19 infections. Anecdotally, supportive care alone was provided unless lower respiratory involvement or hypoxemia developed in which case the usual guidelines for admitted COVID-19 treatments were applied.

SARS-CoV-2 vaccines initially became available in early January 2021 for high-risk individuals and health care providers and more broadly thereafter so that by June of 2021, 70% of the adult population in Alberta had received at least one dose of a COVID vaccine.

Funding Statement

Funding: The CATCO trial was supported by a CIHR Grant.

Contributors:

Conceptualization, A Tremblay, S Murthy, R Fowler, W Sligl; Data Curation, A Tremblay, C O’Neil; Formal Analysis, A Tremblay; Investigation, A Tremblay, R Somayaji, H Hoang, C O’Neil, R Fowler, W Sligl; Methodology, A Tremblay, R Somayaji, J Conly, S Murthy, W Sligl; Project Administration, A Tremblay, H Hoang, A Sonpar, R Fowler, W Sligl; Resources, W Sligl; Supervision, A Tremblay; Writing – Original Draft, A Tremblay; Writing – Review & Editing, R Somayaji, H Hoang, C O’Neil, A Sonpar, J Conly, S Murthy, R Fowler, W Sligl.

Ethics Approval:

The Conjoint Health Research Ethics Board, Calgary, Alberta, approved this study.

Informed Consent:

Informed consent was received for this article.

Registry and the Registration No. of the Study/Trial:

clinicaltrials.gov NCT04748588.

Data Accessibility:

Inquiries regarding data sharing can be made through the corresponding author.

Funding:

The CATCO trial was supported by a CIHR Grant.

Disclosures:

J Conly reports grants from Canadian Institutes for Health Research, Pfizer, WHO; support from bioMerieux Canada; and a member and Chair of the WHO Infection Prevention and Control Research and Development Expert Group for COVID-19 and a member of the WHO Health Emergencies Programme (WHE) Ad-hoc COVID-19 IPC Guidance Development Group, both of which provide multidisciplinary advice to the WHO, for which no funding is received and from which no funding recommendations are made for any WHO contracts or grants. He is also a member of the Cochrane Acute Respiratory Infections Group. He also works as an Infectious Diseases Consultant at Alberta Health Services, Calgary, Canada. C O’Neil reports honoraria and support from Gilead. R Somayaji reports grants from Cystic Fibrosis Foundation, CIHR, University of Calgary, Snyder Institute, and Vertex Pharmaceuticals; ad board honoraria from Vertex Pharmaceuticals; and DSMB membership for Oncovir. H Hoang, reports Alberta Innovates PRIHS Grant for a S. aureus bacteremia quality improvement project; she is Medical Director for the Covenant Health Antimicrobial Stewardship Program and member of the Alberta Provincial Antimicrobial stewardship committee. Srin Murthy is an unpaid member of DSMB for other COVID-19 trials.

Peer Review:

This manuscript has been peer reviewed.

Animal Studies:

N/A