Articles Subcutaneous anti-COVID-19 hyperimmune immunoglobulin for prevention of disease in asymptomatic individuals with SARS-CoV-2 infection: a double-blind, placebo-controlled, randomised clinical trial Andrea Alemany,a,b,c,∗ Pere Millat-Martinez,d,z Marc Corbacho-Monné,a,b,c,e,z Clara Suñer,a,b Cristina Galvan-Casas,a,f Caty Carrera,a,g Dan Ouchi,a,h Núria Prat,i Jordi Ara,i Nuria Nadal,j Ricard Riel,j Blanca Funollet,j Carmen Ojeda-Ciurana,k Lluis Esteve Balague,k Betlem Salvador-González,k,l Anna Forcada Arcarons,g Josep Vidal-Alaball,m,n,o María Isabel Del Cura-González,p,q Ricardo Rodríguez Barrientos,p,q Rafel Ramos-Blanes,r Alberto Alum Bou,r Elsa Mondou,s Mireia Torres,s Neus Campins,s Ana Sanz,s Yonggiang Tang,s Miquel Àngel Rodriguez-Arias,a,b Quique Bassat,d,t,u,v,w Bonaventura Clotet,a,b,o,x GC2010 STUDY GROUPaa and Oriol Mitjàa,b,o,y a Fight Infectious Diseases Foundation, Badalona, Spain b Hospital Universitari Germans Trias i Pujol, Badalona, Spain c Facultat de Medicina-Universitat de Barcelona, Barcelona, Spain d ISGlobal, Hospital Clinic - Universitat de Barcelona, Barcelona, Spain e Hospital Universitari Parc Taulí, I3PT, 08028, Sabadell, Spain f Department of Dermatology, Hospital Universitario de Móstoles, Madrid, Spain g Gerència Territorial de la Catalunya Central, Institut Català de la Salut, Barcelona, Spain h Universitat Autònoma de Barcelona, Barcelona, Spain i Gerència Territorial Metropolitana Nord, Institut Català de la Salut, Barcelona, Spain j Gerència Territorial de Barcelona, Institut Català de la Salut, Barcelona, Spain k Gerència Territorial Metropolitana Sud, Institut Català de la Salut, Barcelona, Spain l Unitat de Suport a la Recerca Costa de Ponent, Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), l’Hospitalet de Llobregat, Spain m Unitat de Suport a la Recerca de la Catalunya Central, Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina, Sant Fruitós de Bages, Spain n Health Promotion in Rural Areas Research Group, Gerència Territorial de la Catalunya Central, Institut Català de la Salut, Sant Fruitós de Bages, Spain o Facultat de Medicina, Universitat de Vic - Universitat Central de Catalunya (UVIC-UCC), Vic, Spain p Unidad de Investigación, Gerencia Asistencial de Atención Primaria, Madrid, Spain q Red de Investigación en Cronicidad, Atención Primaria y Promoción de la Salud -RICAPPS- ISCIII, Spain r Unitat de Suport a la Recerca de Girona, Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Girona, Spain s Scientific Innovation Office, Grifols, Barcelona, Spain t Pediatrics Department, Hospital Sant Joan de Déu, Universitat de Barcelona, Esplugues, Barcelona, Spain u Consorcio de Investigación Biomédica en Red de Epidemiología y Salud Pública, Madrid, Spain v ICREA, Pg Lluís Companys 23, Barcelona, Spain w Centro de Investigação em Saúde de Manhiça (CISM), Maputo, Mozambique x IrsiCaixa AIDS Research Institute, Germans Trias i Pujol Research Institute, Can Ruti Campus, Badalona, Spain y Lihir Medical Centre, International SOS, Lihir Island, Papua New Guinea Summary eClinicalMedicine 2023;57: 101898 Background Anti-COVID-19 hyperimmune immunoglobulin (hIG) can provide standardized and controlled antibody content. Data from controlled clinical trials using hIG for the prevention or treatment of COVID-19 outpatients have Published Online xxx https://doi.org/10. not been reported. We assessed the safety and efficacy of subcutaneous anti-COVID-19 hyperimmune 1016/j.eclinm.2023. immunoglobulin 20% (C19-IG20%) compared to placebo in preventing development of symptomatic COVID-19 101898 in asymptomatic individuals with SARS-CoV-2 infection. *Corresponding author. Department of Infectious Diseases and Fight Infectious Diseases Foundation, Hospital Germans Trias Pujol, Badalona, Catalonia, Spain. E-mail address: aalemany@lluita.org (A. Alemany). z Contributed equally. aa Study group are listed in the appendix. www.thelancet.com Vol 57 March, 2023 1 Articles Methods We did a multicentre, randomized, double-blind, placebo-controlled trial, in asymptomatic unvaccinated adults (≥18 years of age) with confirmed SARS-CoV-2 infection within 5 days between April 28 and December 27, 2021. Participants were randomly assigned (1:1:1) to receive a blinded subcutaneous infusion of 10 mL with 1 g or 2 g of C19-IG20%, or an equivalent volume of saline as placebo. The primary endpoint was the proportion of participants who remained asymptomatic through day 14 after infusion. Secondary endpoints included the proportion of individuals who required oxygen supplementation, any medically attended visit, hospitalisation, or ICU, and viral load reduction and viral clearance in nasopharyngeal swabs. Safety was assessed as the proportion of patients with adverse events. The trial was terminated early due to a lack of potential benefit in the target population in a planned interim analysis conducted in December 2021. ClinicalTrials.gov registry: NCT04847141. Findings 461 individuals (mean age 39.6 years [SD 12.8]) were randomized and received the intervention within a mean of 3.1 (SD 1.27) days from a positive SARS-CoV-2 test. In the prespecified modified intention-to-treat analysis that included only participants who received a subcutaneous infusion, the primary outcome occurred in 59.9% (91/ 152) of participants receiving 1 g C19-IG20%, 64.7% (99/153) receiving 2 g, and 63.5% (99/156) receiving placebo (difference in proportions 1 g C19-IG20% vs. placebo, −3.6%; 95% CI -14.6% to 7.3%, p = 0.53; 2 g C19-IG20% vs placebo, 1.1%; −9.6% to 11.9%, p = 0.85). None of the secondary clinical efficacy endpoints or virological endpoints were significantly different between study groups. Adverse event rate was similar between groups, and no severe or life-threatening adverse events related to investigational product infusion were reported. Interpretation Our findings suggested that administration of subcutaneous human hyperimmune immunoglobulin C19-IG20% to asymptomatic individuals with SARS-CoV-2 infection was safe but did not prevent development of symptomatic COVID-19. Funding Grifols. Copyright © 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Keywords: Hyperimmune immunoglobulin; Antibody therapies; COVID-19; SARS-CoV-2; Outpatients; Asymptomatic individuals Research in context Evidence before this study therapy for asymptomatic individuals with confirmed SARS- We searched the PubMed database for articles (including CoV-2 infection. We found that, compared to placebo, preprints) published between April 2020 and October 2022, subcutaneous human hyperimmune immunoglobulin C19- and reporting results from randomised trials evaluating the IG20% at the dose of either 1 g or 2 g did not reduce the risk effect of hyperimmune immunoglobulins (hIG) for the of developing symptomatic COVID-19 when administered to prophylaxis or treatment of SARS-CoV-2 infected individuals. asymptomatic individuals with confirmed SARS-CoV-2 We used various combinations of the terms “COVID-19”, infection within 5 days, regardless of risk factors. There was “COVID”, “SARS-CoV-2”, “Coronavirus”, “hyperimmune no heterogeneity of treatment effect in efficacy among immunoglobulin”, “intravenous immunoglobulin”, “hIG”, or individuals without endogenous antibodies, nor in any of the “hIVIG”, “passive immunotherapy”, “passive immunization”, other subgroup analyses conducted. There were no significant “plasma therapy”, and “clinical trial”. The search retrieved only differences in the safety endpoints, including the proportion three trials (two pilot studies and an international multicentre of treatment-emergent adverse events and severe adverse study funded by the NIH) evaluating the safety and efficacy of events between groups. hIG therapies for COVID-19, all of which included only Implications of all the available evidence hospitalised patients with COVID-19 and administered Our results do not support the use of subcutaneous C19- intravenous infusion of hIG. No trials were found evaluating IG20% in asymptomatic individuals with SARS-CoV-2 the safety and efficacy of hIG therapies in outpatients with infection to prevent symptomatic COVID-19. Our findings SARS-CoV-2 infection. indicate that C19-IG20% is safe and well tolerated if Added value of this study administered at the dose of either 1 g or 2 g. This study is the first placebo-controlled randomised clinical trial to report results of anti-COVID-19 hIG as pre-emptive 2 www.thelancet.com Vol 57 March, 2023 Articles Introduction subjected to robust pathogen reduction rendering it Anti-SARS-CoV-2 antibody products have emerged as virally safe, and it is purified by technologies demon- promising candidates for the treatment and prophylaxis strated to preserve immunoglobulin neutralization ca- of COVID-19 since the beginning of the pandemic. Five pacity and Fc fragment integrity. Third, unlike most anti-SARS-CoV-2 monoclonal antibody (mAb) products mAbs and other hIG evaluated so far, which need to be have shown clinical benefit when used to treat COVID- administered intravenously, C19-IG20% is available for 19 outpatients1–6 and hospitalised patients without subcutaneous infusion, allowing easier and faster detectable antibodies to SARS-CoV-2,7–9 as well as for administration at the primary care level in the outpatient pre- and post-exposure prophylaxis.10–12 However, the setting. To date, data from controlled clinical trials using efficacy of these mAb therapies can be affected by hIG products for prophylaxis or treatment of COVID-19 antigenic shifts of new circulating variants. Currently, outpatients have not been reported. We evaluated the all mAbs for the treatment of COVID-19 have shown to safety and clinical efficacy of the subcutaneous be ineffective in vitro against the Omicron variant and its C19-IG20% in reducing the risk of developing symp- subvariants.13–23 COVID-19 Convalescent Plasma (CCP), tomatic COVID-19 in asymptomatic individuals with an alternative antibody product that contains polyclonal SARS-CoV-2 confirmed infection. antibodies from donors who have recovered from infection, has proven not to reduce mortality in hospi- talised patients.24–27 CCP has been also tested in out- Methods patients with COVID-19 with mixed results. Positive Trial design results were driven by very early administration of CCP The GC2010 trial was a multicentre, double-blinded, (≤5 days after symptoms onset) and high antibody randomised (1:1:1), parallel group study to assess the titers.28–33 safety and efficacy of the anti-COVID-19 hyperimmune A high-titre and high-concentration antibody prepa- immunoglobulin (Human) 20% (C19-IG20%) in pre- ration can be produced by pooling plasma collected venting symptomatic COVID-19 in asymptomatic out- from multiple donors who have recovered from COVID- patients with SARS-CoV-2 infection. The trial was 19, resulting in the so-called anti-COVID-19 hyperim- conducted between April 28, 2021, and December 27, mune immunoglobulin (hIG). The use of hIG 2021, at seven healthcare administrative regions preparations has been established for the treatment and providing universal healthcare to a catchment popula- prophylaxis of several viral infections, including cyto- tion of around 12 M people in Spain (Methods S1, megalovirus, varicella, rubella, and hepatitis B and A.34–36 Supplementary Appendix). However, clinical data on the use of hIG for COVID-19 The study was conducted according to the Helsinki are limited to three clinical trials administering the Declaration of the World Medical Association, and the product intravenously to hospitalised patients. The first study protocol was approved by the Ethics Committee at one, a small single-centre trial of 50 COVID-19 severely Hospital Germans Trias i Pujol (number PI 21-015) and or critically ill patients, reported nonsignificant re- the institutional review boards of the rest of partici- ductions in mortality associated with hIVIG compared pating centres. All patients provided informed consent to the standard of care.37 The second trial (ITAC) was an before enrolling in the study, which was supervised international multicentre study funded by the US Na- by an independent data and safety monitoring board. tional Institute of Health (NIH) that randomized 593 This trial was registered in ClinicalTrials.gov hospitalised COVID-19 patients without end-organ fail- (NCT04847141). The protocol and statistical analysis ure to receive either hIVIG or an equivalent volume of plan are available in the supplementary materials. saline as placebo in addition to standard clinical care. The trial showed no significant improvement of the Participants and recruitment clinical status, measured by a seven-category ordinal We included asymptomatic individuals aged ≥18 years scale.38 The third trial showed a reduction in the risk for with laboratory-confirmed SARS-CoV-2 infection within severe COVID-19 in 18 severely immunocompromised 5 days prior to randomization. SARS-CoV-2 infection hospitalised patients.39 was determined by RT-PCR, rapid antigen test, or C19-IG20% is a subcutaneous formulation contain- transcription-mediated amplification (TMA) test. Can- ing 20% human hIG that consists of purified protein didates were considered to be asymptomatic if they had from pooled plasma donations, with IgG accounting for no fever (oral temperature ≥38 ◦ C), cough, shortness of at least 98% of the protein. C19-IG20% has some ad- breath, fatigue, anorexia, vomiting/diarrhoea, myalgias, vantages over other anti-SARS-CoV-2 antibody products. headache, olfactory disorders, or pneumonia at First, unlike mAbs, the polyclonal nature of its anti- screening. Individuals were excluded from the study if bodies could mitigate the immune evasion of emerging they required hospitalisation for any cause or had an viral variants. Second, it contains a standardized and oxygen saturation level (SpO2) of ≤94% on room air, or controlled high-titre content of neutralizing antibodies, a National Early Warning Score (NEWS) > 2 points at overcoming the inter-unit variability of CCP. It is also the baseline visit. Additionally, individuals were www.thelancet.com Vol 57 March, 2023 3 Articles excluded if they had received a complete or incomplete delta (B.1.617.2) VOC (Methods S4), using a pseudovi- regimen of COVID-19 vaccination, were taking agents rus neutralization assay, as part of a post-hoc analysis. with antiviral activity against SARS-CoV-2 and/or The distribution of VOC during plasma collection and convalescent COVID-19 plasma or had contraindica- recruitment periods are shown in Methods S5. Partici- tions to the investigational product. Female participants pants were all provided with pulse oximeters and ther- who were pregnant, breastfeeding or planning a preg- mometers for daily self-recording of their SpO2 and nancy during the study were also excluded. Further body temperature at home. In-person follow-up visits details on the eligibility criteria are listed in Methods S2. were planned for study days 3, 7, 14, and 29 at the Potential eligible participants were identified by participants’ residences, or in the hospital if they were searching the database systems for SARS-CoV-2 positive hospitalised. Additionally, investigators contacted study individuals nationwide. The investigators contacted participants by telephone on study days 5, 9, and 11 to candidates by phone in order to explain the study, invite assess their clinical status, including the development of them to participate, and obtain their oral consent to symptomatic COVID-19, and to record their daily SpO2 participate in the screening process. Within 24 h, in- and body temperature measurements. We performed a vestigators conducted a baseline visit (day 1) at the home final telephone check on day 60 to assess vital status, of suitable candidates, during which written informed hospital admissions, ICU admissions, requirement for consent was obtained and eligibility was confirmed. invasive mechanical ventilation and adverse events. All collected data were recorded in an electronic case report Randomisation and masking form. Participants were randomly assigned using a central Nasopharyngeal swabs were obtained for quantifica- web-based randomization system to receive either a 1 g tion of SARS-CoV-2 viral load on study days 1, 3, 7, 14, dose of C19-IG20%, a 2 g dose of C19-IG20%, or sterile and 29. Blood samples were obtained on days 1, 7, and 0.9% saline solution (placebo). Randomization was 14 to assess inflammatory biomarkers (D-dimer, stratified by age (<65 years vs. ≥65 years). An unmasked ferritin, and C-reactive protein [CRP]), biochemical and nurse, who was completely independent of the evalu- haematology parameters (creatinine, albumin, ALT, to- ating study team, conducted the randomization after the tal bilirubin, LDH, haemoglobin, haematocrit, platelet investigators had confirmed eligibility. The unmasked count, absolute neutrophil and lymphocyte counts, and nurse prepared and administered the blinded investi- leukocyte counts) and levels of anti-SARS-CoV-2 anti- gational product. All participants and investigators were bodies (IgM and IgG). masked to the treatment allocations, including follow-up Viral load was analysed by real-time quantitative RT- personnel, laboratory personnel, and statisticians, with PCR in two consecutive steps, viral RNA extraction us- the exception of unmasked nurses. The randomization ing QIAmp MinElute Virus Spin kit (Qiagen) and and administration of the investigational product were amplification/detection by TaqPath COVID-19 CE-IVD always conducted on the first day of the study (baseline RT-PCR kit (Thermo Fishe Scientific) at a centralized visit, day 1). laboratory (Progenika Clinical Diagnostics Laboratory, Progenika Biopharma, a Grifols company, Derio, Investigational products and procedures Spain). For absolute quantification, a standard curve was Both, the investigational product and placebo, were built using serial dilutions of a SARS-CoV-2 plasmid administered with a 10 mL subcutaneous infusion over RNA of known concentration (EVAg), run in parallel to a 10–20 min (1–2 min per mL) on day 1. The investiga- set of samples covering all thermal cycles used in the tional product (i.e., C19-IG20%) (prepared and provided analysis (Methods S6). SARS-CoV-2 IgM and IgG anti- by Grifols) was a sterile liquid formulation of immu- bodies were tested using AESKULISA® SARS-CoV-2 S1 noglobulin purified from human plasma with high-anti- IgG and IgM test (AESKU Enzyme Linked Immuno- SARS-CoV-2 antibodies collected from donors recovered sorbent Assay), processed on the SQII Elisa Analyzer from COVID-19 from May 2020 to July 2020. The (AESKU), at Progenika Clinical Diagnostics Laboratory criteria for the selection of convalescent plasma units under specifications described by provider (Methods were anti-SARS-CoV-2 antibody titre corresponding to S7). ≥10.0 using the Ortho-Vitros method or ≥7.0 using the Architect-Abbott method. The highest dose of 2 g was Outcomes selected based on the volume that can be safely The primary outcome was the proportion of participants administered subcutaneously without the need of a who remained asymptomatic through day 14. Symp- peristaltic pump and the maximum lyophilizing capacity tomatic COVID-19 was defined as fulfilling one of the of the manufacturer. Further details on the preparation, following four conditions: (1) developing at least two of manufacturing, and characteristics of the C19-IG20% the following predefined systemic symptoms: fever are provided in Methods S3. The neutralizing activity of ≥38 ◦ C, chills, myalgia, headache, sore throat, cough, C19-IG20% was assessed against the virus lineage fatigue that interfered with daily activities, new olfactory Wuhan-Hu-1, the alpha (B.1.1.7), beta (B.1.351) and or taste disorders, vomiting or diarrhoea; or (2) 4 www.thelancet.com Vol 57 March, 2023 Articles experiencing at least one of the following respiratory The baseline characteristics of the study population signs/symptoms: new or worsening shortness of breath were summarized descriptively using the number of or difficulty breathing; or (3) experiencing SpO2 <94% non-missing observations, mean, standard deviation on room air; or (4) having radiographical evidence of (SD), median and interquartile range (IQR) for the pneumonia. continuous/quantitative data or absolute and relative Prespecified secondary clinical outcomes included frequency counts and percentages for categorical/qual- the proportion of individuals who presented one of the itative data. The primary clinical efficacy endpoint was following non-mutually exclusive events: participants compared between the two doses of C19-IG20% and who remained in an outpatient setting and maintained placebo using the Cochran-Mantel-Haenszel (CMH) test SpO2 ≥94% through day 14, and participants who adjusting for age. Subgroup analyses of the primary required oxygen supplementation, required any medi- clinical efficacy endpoint and analyses of secondary cally attended visit for management or treatment of clinical efficacy endpoints were assessed using the COVID-19, hospitalisation, or ICU admission through Fisher’s exact test or Chi-square test. day 29. Time to the onset of COVID-19 symptoms was The secondary efficacy endpoint of change in SARS- also analysed. CoV-2 viral load (log10 copies/mL) from baseline to day Secondary virological outcomes included viral load 7 and day 14 was assessed by an analysis of covariance reduction in nasopharyngeal swabs on days 7 and 14 (ANCOVA) with treatment and randomization strata as and viral clearance by RT-PCR on days 14 and 29. Other fixed effects and baseline value as covariate. Time-to- secondary outcomes included change in inflammatory event outcomes were assessed using Kaplan–Meier parameters (D-dimer, ferritin, and C-reactive protein estimates. Between-arm analysis for other secondary [CRP]) from baseline to day 14 of follow-up and change outcomes were done using parametric or non- in quantitative anti-SARS-CoV-2 antibodies through parametric methods according to its distributions. day 14. Non-parametric methods were used for non-normal Safety was assessed by the proportion of patients distributed variables with right-skewed distributions experiencing treatment-emergent adverse events (IgG/IgM variables and laboratory markers) and the (TEAEs), defined as the adverse events (AEs) that assessment of distribution was verified visually. One- occurred on or after the time of investigational product way ANOVA or Kruskal–Wallis test for the compari- administration; and the clinically significant change in son between all three treatment arms and Student’s t or key biochemical parameters of organ function/ Dunn’s tests for the pairwise comparison using Holm’s dysfunction (creatinine, albumin, alanine aminotrans- method for p-value correction. All statistical tests were ferase (ALT), total bilirubin, LDH, haemoglobin, hae- performed in the SAS statistical software under a sig- matocrit, platelet count, absolute neutrophil and nificance level of 0.05. lymphocyte counts, and leukocyte counts) from baseline to day 14. Role of the funding source This study was funded by Grifols. Five authors were Statistical analysis employees from Grifols and made substantial contri- We estimated that a sample size of 801 (267 cases per butions to study design (EM), data analysis (YT), and arm) would provide the trial with 80% power to detect manuscript revision (EM, MT, NC, AS, YT). Other au- an increase of 10% in the proportion of asymptomatic thors, independent from the study funder, were also participants remaining asymptomatic after treatment, involved in all of the aforementioned tasks, as described assuming an expected proportion remaining asymp- in the authors’ contributions disclosure. MT had full tomatic of 80%, at a significance level of α = 0.025, and access to the data set, as did OM, AA, and DO. allowing a 10% withdrawal rate. Primary efficacy analyses were performed on the modified intention-to-treat (m-ITT) population, which Results included all the randomized participants who received Study setting and patient characteristics any interventional product infusion. Sensitivity analyses Between 28 April 2021 and 27 December 2021, our team were performed with the intention-to-treat (ITT) popu- identified and contacted approximately 3000 individuals lation (i.e., all randomized participants) and the per- with confirmed SARS-CoV-2 infection, many of whom protocol (PP) population (i.e., participants completing presented with COVID-19 symptoms and were there- the follow-up without major protocol deviations which fore not eligible. We screened 555 asymptomatic in- might have an impact on the primary efficacy endpoints, dividuals with confirmed SARS-CoV-2 infection. Fig. 1 and complete at least 80% of the interventional product). summarizes the recruitment and follow-up of study Safety was assessed in the safety population, which participants. Among 555 individuals screened, 461 included all randomized participants who received at met all the selection criteria and received the allocated least any amount of blinded interventional product intervention, thereby being included in the m-ITT infusion. analysis: 152 received 1 g C19-IG20%, 153 received 2 g www.thelancet.com Vol 57 March, 2023 5 Articles 555 screened 90 screening failure Not meeting eligibility criteria Declined to participate 465 randomized 154 assigned to 1g C19-IG20% 155 assigned to 2g C19-IG20% 156 assigned to placebo (ITT population) (ITT population) (ITT population) 2 didn’t receive allocated 2 didn’t receive allocated 0 didn’t receive allocated intervention intervention intervention 2 consent withdrawal 2 consent withdrawal 152 received allocated 153 received allocated 156 received allocated intervention intervention intervention (m-ITT population) (m-ITT population) (m-ITT population) 3 did not complete follow- 2 did not complete follow- 0 did not complete follow- up up up 1 adverse event 2 consent withdrawal 2 consent withdrawal 149 completed follow-up 151 completed follow-up 156 completed follow-up (PP population) (PP population) (PP population) Fig. 1: Trial profile. ITT, intention to treat; m-ITT, modified intention to treat; PP, per protocol. C19-IG20% and 156 received placebo (Table S1, time from positive SARS-CoV-2 test to random alloca- Supplementary Appendix). All participants included in tion was 3.1 (SD 1.3) days, and the mean time from the m-ITT analysis completed their infusion. exposure to random allocation (among 160 of 461 par- Neutralizing activity of C19-IG20% was evaluated ticipants in which the potential contact with SARS-CoV- using pseudoviral neutralization assay against the orig- 2 could be identified) was 5.6 (SD 2.8) days. inal Wuhan SARS-CoV-2 strain and alpha (B.1.1.7), beta Participants were allocated to a treatment arm and (B.1.351), and delta (B.1.617.2) variants. A 2.4 and 2.9- infused (blinded administration) on the same day. fold decrease in neutralizing antibody titres was Baseline serum antibody results for IgM/IgG were observed against alpha and delta variants, respectively, negative for 345 (80%) of the 431 participants for whom compared with Wuhan-Hu-1 (geometric mean ID50 results were available. Prior to recruitment, all in- 13510 for alpha and 11367 for delta vs ID50 32917 for dividuals had a documented positive SARS-CoV-2 test Wuhan-Hu-1) (Methods S4). result (either by antigen detection or by DNA detection The baseline demographic and clinical characteris- tests), according to inclusion criteria. However, at tics were similar in the three groups (Table 1). Overall, baseline, 119 (26%) of 461 participants had a negative study participants had a mean age of 39.6 (SD 12.8) SARS-CoV-2 RT-PCR result and 342 (74%) participants years, 197 (42.7%) of 461 participants were women, and had a positive RT-PCR test result. In total, 372 (81%) of 101 (21.9%) had at least one comorbidity. The mean the participants had a positive RT-PCR result at any time 6 www.thelancet.com Vol 57 March, 2023 Articles 1 g C19-IG20% 2 g C19-IG20% Placebo N = 152 N = 153 N = 156 Demographics Age, years – mean (SD) 38.8 (12.8) 41.1 (12.4) 38.8 (13.3) Age group – n (%) 18–65 years 146 (96.1) 147 (96.1) 149 (95.5) ≥65 years 6 (3.9) 6 (3.9) 7 (4.5) Women – n (%) 66 (43.4) 62 (40.5) 69 (44.2) Men – n (%) 86 (56.6) 91 (59.5) 87 (55.8) BMI (kg/m2) – mean (SD) 26.0 (4.7) 26.4 (5.1) 25.5 (4.4) SARS-CoV-2 infection characteristics Days from positive testa to random assignmentb, days – mean (SD, N) 3.1 (1.2, 152) 3.1 (1.4, 153) 3.0 (1.2, 156) Days from exposurec to random assignmentb, days – mean (SD, N) 5.8 (2.8, 56) 5.2 (2.5, 58) 5.7 (3.3, 46) Comorbidities – n (%) At least one comorbidity 35 (23.0) 39 (25.5) 27 (17.3) Obesity (BMI ≥30 kg/m2) 29 (19.1) 30 (19.6) 18 (11.5) Diabetes Mellitus 5 (3.3) 11 (7.2) 7 (4.5) Hypertension 11 (7.2) 11 (7.2) 10 (6.4) Heart conditions (i.e. heart failure, coronary artery disease, cardiomyopathies) 2 (1.3) 1 (0.7) 1 (0.6) Chronic Obstructive Pulmonary Disease 0 (0.0) 1 (0.7) 1 (0.6) Asthma 10 (6.6) 7 (4.6) 9 (5.8) Chronic kidney disease 1 (0.7) 2 (1.3) 1 (0.6) History of cancer 3 (2.0) 1 (0.7) 3 (1.9) Immunocompromised state from solid organ transplant 0 (0.0) 0 (0.0) 0 (0.0) Serum IgM and IgG antibody status – n (%) Nd 139 146 146 Negative 114 (82) 118 (81) 113 (77) Positive 25 (18) 28 (19) 33 (23) Study RT-PCR – n (%) Positive at baseline 110 (72) 108 (71) 124 (79) Negative at baseline 42 (28) 45 (29) 32 (21) Positive at any time during the study 120 (79) 118 (77) 134 (86) Positive at baseline and IgM and IgG negative at baseline 77 (51) 80 (52) 83 (53) Viral load Mean Viral Load (SD) in log10 5.8 (2.6) 5.8 (2.6) 6.1 (2.4) Laboratory parameters – mean (SD) Ne 148 152 153 D-dimer (mg/L) 0.6 (1.5) 0.4 (0.4) 0.6 (2.6) Ferritin (ug/L) 111.2 (120.2) 115.2 (138.2) 123.7 (163.4) C-reactive protein (CRP) (mg/dl) 5.8 (10.5) 6.4 (6.7) 5.9 (6.4) BMI = body mass index; SD = standard deviation. Laboratory reference ranges: D-dimer 0-0.50 mg/L; Ferritin 30.0–400.0 ug/L; C-reactive protein 0.0–1.0 mg/dL. aFirst positive PCR (RT-PCR), NAT or other commercial or public health assay result for SARS-CoV-2 infection. bRandom assignment and infusion were always done on the same day. cExposure in terms of first potential contact with virus. d30/461 participants did not have baseline serological test (13/152 in the 1 g C19-IG20% group; 7/153 in the 2 g C19-IG20% group; and 10/156 in the placebo group). Missing data in this variable can be assumed random. e8/461 participants did not have baseline laboratory parameters (4/152 in the 1 g C19-IG20% group; 1/153 in the 2 g C19-IG20% group; and 3/156 in the placebo group). Missing data in this variable can be assumed random. Table 1: Baseline characteristics in the modified intention-to-treat population. during the study. The mean viral load from the naso- presents the final and only report analysis after early pharyngeal swab at baseline was 5.8 (SD 2.6) log10 termination. copies per mL in the 1 g C19-IG20% group, 5.8 (SD 2.6) log10 copies per mL in the 2 g C19-IG20% group and 6.1 Clinical efficacy outcomes (SD 2.4) log10 copies per mL in the placebo group. In the modified ITT population, the primary outcome Trial enrolment was halted on December 27, 2021, analyses (i.e., the proportion of participants who based on the results of a planned interim analysis of all remained asymptomatic on day 14) did not differ available data on primary and secondary efficacy out- significantly between placebo-treated and C19-IG20%- comes, which concluded the lack of potential benefit of treated individuals, irrespective of the dose received the intervention in the target population. This article (Table 2). The primary outcome occurred in 59.9% www.thelancet.com Vol 57 March, 2023 7 Articles 1 g C19-IG20% 2 g C19-IG20% Placebo Difference in p-value Difference in p-value N = 152 N = 153 N = 156 proportions proportions 1 g C19-IG20% - 2 g C19-IG20% - Placebo (95% CI) Placebo (95% CI) Primary clinical efficacy endpoint through day 14 Remained asymptomatic – n(%) 91 (59.9) 99 (64.7) 99 (63.5) −3.6 (−14.6 to 7.3) 0.526 1.1 (−9.6 to 11.9) 0.846 Developed ≥2 systemic COVID-19 58 (38.2) 54 (35.3) 54 (34.6) 3.6 (−7.3 to 14.5) 0.527 0.8 (−9.9 to 11.5) 0.894 symptomsa Experienced ≥1 respiratory symptomsb (new 9 (5.9) 8 (5.2) 15 (9.6) −3.6 (−9.7 to 2.4) 0.245 −4.4 (−10.2 to 1.4) 0.14 or worsening shortness of breath, difficulty breathing) Experienced Sp02 < 94% on room air 5 (3.3) 3 (2.0) 3 (1.9) 1.4 (−2.1 to 5.0) 0.439 0.0 (−3.1 to 3.1) 0.988 Had radiographical evidence of pneumonia 6 (3.9) 7 (4.6) 3 (1.9) 2.1 (−1.7 to 5.9) 0.274 2.8 (−1.2 to 6.8) 0.175 Subgroup analyses of the primary clinical efficacy endpoint: remained asymptomatic through day 14 Positive PCR results at study baseline 59/110 (53.6) 61/108 (56.5) 74/124 (59.7) −6.1 (−18.7 to 6.9) 0.352 −3.2 (−15.9 to 9.8) 0.623 Positive PCR results at any time during the 65/120 (54.2) 68/118 (57.6) 84/134 (62.7) −8.5 (−20.8 to 3.7) 0.169 −5.1 (−17.3 to 7.1) 0.413 study Positive IgM/IgG results at baseline 20/25 (80.0) 26/28 (92.9) 28/33 (84.8) −4.8% (−15% to 24.7%) 0.628 8% (−23.5% to 7.5%) 0.328 Negative IgM/IgG results at baseline 65/114 (57.0) 68/118 (57.6) 66/113 (58.4) −1.4% (−12.3% to 15.1%) 0.938 0.8% (−12.7% to 14.3%) 1.000 Secondary clinical efficacy endpoints Remained in an outpatient setting and 141/148 (95.3) 140/149 (94.0) 148/154 (96.1) −0.8 (−6.1 to 4.2) 0.721 −2.1 (−7.8 to 3.1) 0.390 maintained Sp02 ≥ 94% through day 14 Required oxygen supplementation through 3/152 (2.0) 6/153 (3.9) 2/156 (1.3) 0.7 (−2.9 to 4.6) 0.631 2.6 (−1.2 to 7.3) 0.144 day 29 Required ≥1 related medically attended visit 26/152 (17.1) 29/153 (19.0) 22/156 (14.1) 3 (−5.3 to 11.5) 0.468 4.9 (−3.6 to 13.4) 0.251 through day 29 Required hospitalisation through day 29 3/152 (2.0) 7/153 (4.6) 3/156 (1.9) 0.1 (−3.8 to 4.0) 0.974 2.7 (−1.6 to 7.5) 0.188 Required ICU admission through day 29 1/152 (0.66) 1/153 (0.65) 1/156 (0.64) 0.02 (−3.05 to 3.12) 0.985 0.01 (−2.99 to 3.07) 0.989 Risk difference for the proportions of subjects between groups using CMH (Cochran-Mantel-Haenszel) method adjusted by age for primary efficacy endpoints and Chi-square for secondary clinical efficacy endpoints. aSystemic symptoms: fever (≥38 ◦ C), chills, myalgia, headache, sore throat, cough, fatigue that interfered with daily activities, new olfactory/taste disorder, vomiting/diarrhoea (note that new olfactory/taste disorder and vomiting/diarrhoea only counted as one item of definition). bNew or worsening shortness of breath or difficulty breathing. Table 2: Clinical trial efficacy end points in the modified intention-to-treat population. (91/152) of participants receiving 1 g C19-IG20%, 64.7% Regarding secondary clinical efficacy outcomes, (99/153) receiving 2 g, and 63.5% (99/156) receiving overall, 11 (2.4%) participants required oxygen supple- placebo (difference in proportions: 1 g C19-IG20% vs. mentation at some point during the follow-up, 77 placebo, −3.6%; 95% CI -14.6% to 7.3%, p = 0.53; 2 g (16.7%) required one or more COVID-19 related medi- C19-IG20% vs. placebo, 1.1%; −9.6% to 11.9%, p = 0.85). cal visits, 13 (2.8%) required hospitalisation and 3 The most common presentation among symptomatic (7.2%) required admission to an ICU. No participants in participants was a combination of two or more systemic any treatment group required invasive mechanical COVID-19 symptoms. Only a small proportion of par- ventilation and no participants died during the study. ticipants experienced ≥1 respiratory symptoms, SpO2 None of these secondary endpoints were significantly <94%, or had radiographic evidence of pneumonia. The different between study groups (Table 2). Time to the analysis of the primary outcome in the ITT population onset of COVID-19 symptoms did not show significant (Table S2) and the PP population (Table S3) also differences between groups; 30% of study participants revealed no significant differences between participants developed symptoms within the first three days after treated with placebo versus those treated with C19- infusion (Fig. S1). IG20%. We conducted post-hoc analyses of the pri- mary efficacy endpoint in sub-groups according to the Other efficacy outcomes study RT-PCR test result (baseline and throughout the Fig. 2 shows the SARS-CoV-2 viral load decay follow-up) and serological status of participants at throughout the follow-up period. We found no signifi- baseline. None of the sub analyses revealed significant cant differences between groups regarding the mean differences between groups regarding the primary difference in viral load from baseline to day 7 (absolute endpoint (Table 2). Sensitivity analysis of primary effi- difference 0.10 log10 copies/mL for C19-IG20% 1 g vs. cacy endpoints stratified by age and comorbidities were placebo [95% CI -0.24 to 0.44; p = 0.58]; and −0.17 for also performed, finding no differences between groups C19-IG29% 2 g vs. placebo [95% CI -0.52 to 0.17; (Table S4). p = 0.33]) and from baseline to day 14 (0.11 for 8 www.thelancet.com Vol 57 March, 2023 Articles Fig. 2: Viral load change over 14 days. Figure shows the mean viral load (in log10 copies per millilitre) at baseline, day 7 and day 14 in the overall population and in subgroup of PCR positive at baseline and serum antibody negative status (IgM/IgG negative at baseline). Tables on the figure show difference in least-squares means (LSM) of change from baseline to day 7 and day 14 of viral load (in log10 copies per millilitre) for both doses of C19-IG20% compared to placebo in the overall population and in subgroup of PCR positive at baseline and serum antibody negative status. 95% CI for difference in LSM between each of C19-IG20% dose groups (1 g and 2 g) and placebo and the associated p-value were calculated using an ANCOVA model, including the change from baseline value as a dependent variable; treatment group as a fixed effect; and baseline viral load value, age, and gender as covariates. C19-IG20% 1 g vs. placebo [95% CI -0.12 to 0.34; (46.2%) in the placebo group, with no differences be- p = 0.34] and −0.11 for C19-IG29% 2 g vs. placebo [95% tween treatment groups. Blinded investigators evaluated CI -0.34 to 0.13; p = 0.37]). Likewise, no differences were the 359 TEAEs and determined that 263 (73%) were not observed in time to viral clearance, assessed by RT-PCR related to the investigational product and 10 (2.8%) were up to day 29 (Figs. S2 and S3). definitely related. Regarding the severity of the events, Changes in inflammatory parameters, including 286 (79.7%) were mild (Grade 1), 61 (17.0%) moderate D-dimer, ferritin, and C-reactive protein (CRP), did not (Grade 2), 10 (2.8%) severe (Grade 3), and 2 (0.5%) life- show significant differences between groups from threatening (Grade 4). All TEAEs related to the investi- baseline to day 14 of follow-up (Fig. S4). Likewise, the gational product were mild or moderate in severity. All groups did not differ regarding the change in anti- severe and life-threatening TEAEs were related to SARS-CoV-2 IgM and IgG throughout the follow-up COVID-19. No individuals experienced a TEAE leading (Fig. S5). to death. There were no serious adverse drug reactions reported. Safety Most common TEAEs were related to COVID-19, Table 3 summarizes the TEAEs that occurred from the including gastrointestinal disorders, arthralgia, head- time of administration of the investigational product to ache, cough, and fever. TEAEs related to IP infusion day 14 of follow-up. A total of 359 TEAEs were reported: included injection site pain, puncture site pain and er- 137 TEAEs in 78/152 (51.3%) participants in the C19- ythema, and vasovagal syndrome (Table S5). No severe IG20% 1 g group, 96 TEAEs in 65/153 (42.5%) in the allergic reactions or anaphylaxis and thromboembolic C19-IG20% 2 g group, and 126 TEAEs in 72/156 events were reported. www.thelancet.com Vol 57 March, 2023 9 Articles 1 g C19-IG20% (n = 152) 2 g C19-IG20% (n = 153) Placebo (n = 156) Difference in p-value Difference in p-value proportions proportions Number of Number of Number of Number of Number of Number of 1 g C19-IG20% - 2 g C19-IG20% - subjectsa eventsb subjectsa eventsb subjectsa eventsb Placebo (95% CI) Placebo (95% CI) Treatment Emergent Adverse Events (TEAE) ≥1 TEAE 78 (51.3) 137 65 (42.5) 96 72 (46.2) 126 −5.16% (−7.00% to 0.428 3.67% (−15.00% to 0.593 17.00%) 8.00%) Relationship to investigational product Not related 48 (31.6) 97 43 (28.1) 69 53 (34.0) 97 Possibly related 26 (17.1) 36 15 (9.8) 20 19 (12.2) 29 Definitely related 4 (2.6) 4 6 (3.9) 6 0 (0.0) 0 Severity Mild (Grade 1) 57 (37.5) 112 46 (30.1) 72 54 (34.6) 102 Moderate (Grade2) 17 (11.2) 21 13 (8.5) 18 16 (10.3) 22 Severe (Grade 3) 3 (2.0) 3 5 (3.3) 5 2 (1.3) 2 Life threatening 1 (0.7) 1 1 (0.7) 1 0 (0.0) 0 (Grade 4) Fatal (Grade 5) 0 (0.0) 0 0 (0.0) 0 0 (0.0) 0 TEAE leading to death 0 (0.0) 0 0 (0.0) 0 0 (0.0) 0 TEAE leading to study 1 (0.7) 1 0 (0.0) 0 1 (0.6) 1 withdrawal Treatment Emergent Serious Adverse Events (TE-SAE) ≥1 TE-SAE 3 (2.0) 3 7 (4.6) 7 3 (1.9) 3 −0.05% (−3.00% to 1 −2.65% (−1.00% to 0.319 3.00%) 7.00%) COVID-19 pneumonia 3 (2.0) 7 (4.6) 1 (0.6) Non-COVID-19 0 (0) 0 (0) 2 (1.3) pneumonia AE = adverse event; IP = investigational product; SAE = serious adverse event; TEAE = treatment-emergent adverse event. Note: Treatment-emergent AEs are AEs that occurred on or after the date/time of IP administration. Percentages were based on the total number of safety subjects in each treatment group (N). aAt each level of summation (overall, relationship, severity), subjects reporting more than one AE were counted only once using the strongest relationship to study drug and maximum severity. bNumber of events included all occurrences of AEs. Table 3: Safety end points in the safety population. 13 SAE were reported: 3 (2%) in the C19-IG20% 1 g differ between the treatment groups and were mainly group, 7 (4.6%) in the C19-IG20% 2 g group, and 3 related to COVID-19. No severe or life-threatening events (1.9%) in the placebo group. All SAEs were related to related to interventional product infusion were reported. COVID-19 infection, except for two cases of pneumonia Overall, these findings indicate that C19-IG20% at the not related to COVID-19. dose of 1 g and 2 g is safe and well tolerated but did not Change in biochemical and haematological parame- prevent development of symptomatic COVID-19. ters of organ derangement or systemic inflammatory Passive immunotherapies, including hIG, mAbs and response (i.e., creatinine, albumin, alanine amino- CCP, have demonstrated no clinical benefit in reducing transferase (ALT), total bilirubin, LDH, haemoglobin, the mortality risk in most hospitalised patients with haematocrit, platelet count, absolute neutrophil and COVID-19, except for selected groups (severely immu- lymphocyte counts, and leukocyte counts) from baseline nocompromised and seronegative).7–9,24–27,37–39 As to day 14 did not show clinically relevant differences opposed to these studies, ours focused on outpatients between groups (Table S6 and Fig. S6). with a very recent and asymptomatic infection since antibody products are expected to be more beneficial when administered very early in the course of infection. Discussion For example, early administration of mAbs and CCP has Our findings show that early infusion with 1 g or 2 g of been demonstrated to have clinical benefits in out- C19-IG20%, compared with placebo, did not reduce the patients with mild and moderate COVID-19.1,3–6,30–32 risk of developing symptomatic COVID-19 in individuals Moreover, combinations of mAbs have been shown to diagnosed with asymptomatic SARS-COV-2 infection. reduce the risk of both asymptomatic and symptomatic Furthermore, neither our secondary clinical or virological SARS-CoV-2 infection when administered as pre- and endpoints nor our prespecified subgroup analyses post-exposure prophylaxis (PrEP and PEP).10–12 demonstrated a benefit of this therapy. Safety endpoints, Conversely, CCP (tested only as PEP) failed to prevent including the proportion of treatment-emergent adverse infection in asymptomatic contacts in a clinical trial events (TEAEs) and severe adverse events (SAE) did not conducted in the US.40 10 www.thelancet.com Vol 57 March, 2023 Articles Our results did not reflect previous findings in the enrol in the study, 26% tested negative by PCR at treatment of outpatients with mAbs and CCP, despite baseline and 19% at any time during the study. It is the very early administration of immune therapy. On possible that some of these asymptomatic participants the other hand, our results are consistent with the lack were diagnosed at the end of their infectious period and of benefit found for CCP used as PEP, as opposed to the viral clearance occurred between diagnosis and successful use of mAbs (with a much higher content of randomization. Another explanation might be a false specific antibodies) in the same context. A possible positive diagnostic test, which is more frequent in the explanation is that the amount of specific antibodies context of screening asymptomatic individuals.41,42 contained in 10 mL of C19-IG20% may not be sufficient However, the sensitivity analyses on baseline PCR- to prevent mild COVID-19 symptoms involving the positive participants did not change the trend of the upper respiratory tract, although it may have been able m-ITT analysis. Regarding key clinical endpoints such to prevent progression to severe disease. The combina- as hospitalisation and need for oxygen supplementation, tion of casirivimab/imdevimab contains 1.2 g of SARS- we were unable to draw any strong conclusions due to CoV-2-specific immunoglobulins, a higher amount than the low frequency of these events and the relatively the 200 mg and 400 mg of polyclonal immunoglobulins small sample size, which limited the statistical power of contained in the 1 g (5 mL) and 2 g (10 mL) doses of the analysis. Another limitation is that the study was C19-IG20%, respectively. Our findings showing no conducted mainly in unvaccinated patients, most of change in viral load in nasopharynx, in contrast with the whom were seronegative at baseline. Also, our trial reduction observed with mAb, also supports this hy- included 80% of participants with no comorbidities and pothesis. A maximum volume of 10 mL was adminis- none of them were immunocompromised. It remains tered in our clinical trial based on data from previous unknown whether this intervention could benefit in- studies of other subcutaneous hyperimmune immuno- dividuals at higher risk in the absence of other therapies globulin products that are safe and tolerable. Nonethe- with proven efficacy. Finally, the reduction in neutrali- less, new delivery system designs may enable larger zation activity of the C19-IG20% against different volume subcutaneous infusion viability and tolerability. circulating variants at the time of infusion may have Intravenous administration could also accept higher contributed to lack of efficacy, indicating the importance volumes; for instance, 400 mg/kg body weight and 3.5 g of developing agile production and distribution work- administered in the ITAC38 and OTAC (NCT04910269) flows for hIG therapies, so that they can be adminis- studies, respectively. However, subcutaneous therapies tered timely. are more likely to be successfully deployed in commu- On the other hand, several aspects of the methodol- nity and primary care settings, particularly in countries ogy and study conduct strengthen and increase the with limited healthcare systems. Regarding neutralizing generalizability of our findings. First, this is the first activity, plasma for C19-IG20% was collected in the controlled clinical trial to report results of anti-COVID- United States from the second half of 2020, before the 19 hIG as treatment of ambulatory SARS-CoV-2 infec- emergence of alpha variant, while the trial enrolled ted individuals. The trial included a large and diverse participants in Spain from April to December 2021, trial population that was enrolled at different sites during alpha and delta variants dominance periods. throughout Spain. In addition, intervention was double- Analyses conducted using pseudoviral neutralization blinded, with a very high percentage of participants assays identified a two to three fold reduction in receiving the infusion and completing the follow-up. neutralizing antibody titres for the circulating variants. Finally, the main results of our trial regarding clinical In light of this results, we cannot rule out the possibility efficacy are supported by virological and laboratory of clinical efficacy if higher doses of C19-IG20% and/or endpoints, contributing to more robust conclusions higher neutralization capacity had been administered. about the potential effect of C19-IG20%. Our clinical trial has several limitations. Firstly, the The results of this trial do not support the use of the trial was terminated early by the data safety monitoring subcutaneous human hyperimmune immunoglobulin board based on an interim analysis showing no signs of C19-IG20% at either 1 g or 2 g dose regimen for the potential benefit to the target population; hence the prevention of symptomatic COVID-19 in asymptomatic target sample size was not achieved. However, based on individuals with confirmed SARS-CoV-2 infection. Our the interim analysis results, including the lack of dif- findings indicate that C19-IG20% at the dose of 1 g and ferences in all pre-defined efficacy endpoints, we do not 2 g is safe and well tolerated. Future studies shall expect the analysis with the target sample size to yield investigate the potential benefits of C19-IG20% and different conclusions. Second, recruitment was condi- other hIG therapies with higher antibody dose and tioned by the widespread availability of vaccines during neutralizing activity in other scenarios, such as the study period, as vaccinated individuals were not prevention of disease progression in outpatients eligible to participate. Third, although all participants with COVID-19, particularly those that are were required to have a positive test for SARS-CoV-2 to immunocompromised. www.thelancet.com Vol 57 March, 2023 11 Articles Contributors 6 Dougan M, Azizad M, Mocherla B, et al. A randomized, placebo- OM, EM, AA, PMM, MCM conceived and designed the study. All author controlled clinical trial of bamlanivimab and etesevimab together acquired, analysed, and interpreted the data. DO, YT did the statistical in high-risk ambulatory patients with COVID-19 and validation of analysis. AA and OM drafted the manuscript. All authors critically the prognostic value of persistently high viral load. Clin Infect Dis. 2022;75:e440–e449. revised the manuscript for important intellectual content. All authors 7 Abani O, Abbas A, Abbas F, et al. Casirivimab and imdevimab in were responsible for the final decision to submit the manuscript for patients admitted to hospital with COVID-19 (RECOVERY): a publication. All authors have seen and approved the manuscript. OM, randomised, controlled, open-label, platform trial. Lancet. AA, DO, MT had full access to and verified the data. 2022;399:665–676. 8 Somersan-Karakaya S, Mylonakis E, Menon VP, et al. Casirivimab Data sharing statement and imdevimab for the treatment of hospitalized patients with Individual participant data that underlie the results reported in this COVID-19. J Infect Dis. 2022;227(1):23–34. article, after de-identification (text, tables, figures, and appendices) are 9 Mylonakis E, Somersan-Karakaya S, Sivapalasingam S, et al. LB4. Casirivimab and imdevimab for treatment of hospitalized patients available from the corresponding author on reasonable request. with COVID-19 receiving low flow or No supplemental oxygen. Open Forum Infect Dis. 2021;8:S809–S810. Declaration of interests 10 O’Brien MP, Forleo-Neto E, Musser BJ, et al. Subcutaneous EM, MT, NC, AS, and YT were employees from Grifols. The rest of REGEN-COV antibody combination to prevent COVID-19. N Engl J authors declared no conflict of interest. Med. 2021;385:1184–1195. 11 al Jrdi A, Morena L, Cote M, Bethea E, Azzi J, Riella LV. Tix- agevimab/cilgavimab pre-exposure prophylaxis is associated with Acknowledgements lower breakthrough infection risk in vaccinated solid organ trans- This study was funded by Grifols. We thank all the trial participants who plant recipients during the omicron wave. Am J Transplant. 2022. volunteered for this trial; all the investigational site members involved in 12 Levin MJ, Ustianowski A, de Wit S, et al. Intramuscular AZD7442 this trial (listed in the Supplementary Appendix) and the personnel from (Tixagevimab-Cilgavimab) for prevention of COVID-19. N Engl J the Fight Aids and Infectious Diseases Foundation, ISGlobal, Gerència Med. 2022;386:2188–2200. 13 Takashita E, Kinoshita N, Yamayoshi S, et al. Efficacy of antibodies Territorial Metropolitana Nord (ICS), Gerència Territorial de Barcelona and antiviral drugs against COVID-19 omicron variant. N Engl J (ICS), Gerència Territorial Metropolitana Sud (ICS), Gerència Territorial Med. 2022;386:995–998. de la Catalunya Central (ICS), Gerència Territorial de Girona (ICS) and 14 Takashita E, Kinoshita N, Yamayoshi S, et al. Efficacy of antiviral Gerencia Asistencial de Atención Primaria de Madrid for their support agents against the SARS-CoV-2 omicron subvariant BA.2. N Engl J and contributions. We thank Gerard Carot-Sans for providing medical Med. 2022;386:1475–1477. writing support with manuscript preparation. We also thank Laia Ber- 15 VanBlargan LA, Errico JM, Halfmann PJ, et al. An infectious SARS- tran, Mireia Clua, Jordi Mitjà, Sergi Gavilan and Joan Mercado for the CoV-2 B.1.1.529 Omicron virus escapes neutralization by thera- operational management of the project. We thank Lydia Martinez, Marta peutic monoclonal antibodies. Nat Med. 2022;28:490–495. 16 Cameroni E, Bowen JE, Rosen LE, et al. Broadly neutralizing an- José, Jordi Bozzo and Sandra Camprubí from Grifols. We thank the tibodies overcome SARS-CoV-2 Omicron antigenic shift. Nature. members of the external data monitoring board for their time and 2022;602:664–670. dedication. We thank Patricia Armenteros, Ainoa Rueda, Jordi Ruiz 17 Planas D, Saunders N, Maes P, et al. Considerable escape of SARS- Camp, Diego Vergara, Ana Álvarez, Ramón Mascaró, Nuria Navarro, CoV-2 Omicron to antibody neutralization. Nature. 2022;602: Paula Campello, María Galán, Cristina Aragüés. ISGlobal receives 671–675. support from the Spanish Ministry of Science and Innovation through 18 Westendorf K, Žentelis S, Wang L, et al. LY-CoV1404 (bebt- the Centro de Excelencia Severo Ochoa 2019–2023 Program (CEX2018- elovimab) potently neutralizes SARS-CoV-2 variants. Cell Rep. 000806-S), and support from the Generalitat de Catalunya through the 2022;39. 19 Iketani S, Liu L, Guo Y, et al. Antibody evasion properties of SARS- Centres de Recerca de Catalunya Program. Centro de Investigación en CoV-2 Omicron sublineages. Nature. 2022;604:553–556. Salud de Manhiça is supported by the Government of Mozambique and 20 Yamasoba D, Kosugi Y, Kimura I, et al. Neutralisation sensitivity of the Spanish Agency for International Development. OM was supported SARS-CoV-2 omicron subvariants to therapeutic monoclonal anti- by the European Research Council under grant agreement number bodies. Lancet Infect Dis. 2022;22:942–943. 850450 (European Union’s Horizon 2020 Research and Innovation 21 Wang Q, Guo Y, Iketani S, et al. Antibody evasion by SARS-CoV-2 Program, ERC-2019-STG funding scheme). Omicron subvariants BA.2.12.1, BA.4 and BA.5. Nature. 2022;608: 603–608. 22 Cao Y, Yisimayi A, Jian F, et al. BA.2.12.1, BA.4 and BA.5 escape Appendix A. Supplementary data antibodies elicited by Omicron infection. Nature. 2022;608:593–602. Supplementary data related to this article can be found at https://doi. 23 FDA Announces Bebtelovimab is Not Currently Authorized in Any US org/10.1016/j.eclinm.2023.101898. Region. FDA; 2023. https://www.fda.gov/drugs/drug-safety-and- availability/fda-announces-bebtelovimab-not-currently-authorized- any-us-region. Accessed February 3, 2023. 24 Janiaud P, Axfors C, Schmitt AM, et al. Association of convalescent References plasma treatment with clinical outcomes in patients with COVID- 1 Verderese JP, Stepanova M, Lam B, et al. Neutralizing monoclonal 19: a systematic review and meta-analysis. JAMA. 2021;325: antibody treatment reduces hospitalization for mild and moderate 1185–1195. coronavirus disease 2019 (COVID-19): a real-world experience. Clin 25 Estcourt LJ, Turgeon AF, McQuilten ZK, et al. Effect of convales- Infect Dis. 2022;74:1063–1069. cent plasma on organ support-free days in critically ill patients with 2 U.S. Food and Drug Administration (FDA). Fact Sheet for Health- COVID-19: a randomized clinical trial. JAMA. 2021;326:1690–1702. care providers: Emergency Use Authorization for Bebtelovimab; 2022. 26 Bégin P, Callum J, Jamula E, et al. Convalescent plasma for hos- https://www.fda.gov/emergency-preparedness-and-. Accessed pitalized patients with COVID-19: an open-label, randomized October 11, 2022. controlled trial. Nat Med. 2021;27:2012–2024. 3 Gupta A, Gonzalez-Rojas Y, Juarez E, et al. Effect of sotrovimab on 27 Abani O, Abbas A, Abbas F, et al. Convalescent plasma in patients hospitalization or death among high-risk patients with mild to admitted to hospital with COVID-19 (RECOVERY): a randomised moderate COVID-19: a randomized clinical trial. JAMA. 2022;327: controlled, open-label, platform trial. Lancet. 2021;397:2049–2059. 1236–1246. 28 Alemany A, Millat-Martinez P, Corbacho-Monné M, et al. High- 4 Weinreich DM, Sivapalasingam S, Norton T, et al. REGEN-COV titre methylene blue-treated convalescent plasma as an early antibody combination and outcomes in outpatients with COVID- treatment for outpatients with COVID-19: a randomised, placebo- 19. N Engl J Med. 2021;385:e81. controlled trial. Lancet Respir Med. 2022;10:278–288. 5 Dougan M, Nirula A, Azizad M, et al. Bamlanivimab plus etesevi- 29 Korley FK, Durkalski-Mauldin V, Yeatts SD, et al. Early convales- mab in mild or moderate COVID-19. N Engl J Med. 2021;385: cent plasma for high-risk outpatients with COVID-19. N Engl J Med. 1382–1392. 2021;385:1951–1960. 12 www.thelancet.com Vol 57 March, 2023 Articles 30 Libster R, Pérez Marc G, Wappner D, et al. Early high-titer plasma intravenous immunoglobulin preparations. Transplantation. therapy to prevent severe COVID-19 in older adults. N Engl J Med. 2011;92:267–270. 2021;384:610–618. 37 Ali S, Uddin SM, Shalim E, et al. Hyperimmune anti-COVID-19 31 Sullivan DJ, Gebo KA, Shoham S, et al. Early outpatient treatment IVIG (C-IVIG) treatment in severe and critical COVID-19 pa- for COVID-19 with convalescent plasma. N Engl J Med. tients: a phase I/II randomized control trial. eClinicalMedicine. 2022;386:1700–1711. 2021;36:100926. 32 Gharbharan A, Jordans C, Zwaginga L, et al. Outpatient convales- 38 Polizzotto MN, Nordwall J, Babiker AG, et al. Hyperimmune cent plasma therapy for high-risk patients with early COVID-19: a immunoglobulin for hospitalised patients with COVID-19 (ITAC): randomized placebo-controlled trial. Clin Microbiol Infect. a double-blind, placebo-controlled, phase 3, randomised trial. Lan- 2022;29(2):208–214. cet. 2022;399:530. 33 Millat-Martinez P, Gharbharan A, Alemany A, et al. Prospective 39 Huygens S, Hofsink Q, Nijhof IS, et al. Hyperimmune globulin for individual patient data meta-analysis of two randomized trials on severely immunocompromised patients hospitalized with COVID-19: a convalescent plasma for COVID-19 outpatients. Nat Commun. randomized, controlled trial. J Infect Dis. 2022;227(2):206–210. 2022;13:1–9. 40 Shoham S, Bloch EM, Casadevall A, et al. Transfusing convalescent 34 Young MK. The indications and safety of polyvalent immuno- plasma as post-exposure prophylaxis against severe acute respira- globulin for post-exposure prophylaxis of hepatitis A, rubella and tory syndrome coronavirus 2 (SARS-CoV-2) infection: a double- measles. Hum Vaccin Immunother. 2019;15:2060. blinded, phase 2 randomized, controlled trial. Clin Infect Dis. 35 Nelson NP, Link-Gelles R, Hofmeister MG, et al. Update: recom- 2023;76(3):e477–e486. mendations of the advisory committee on immunization practices 41 Dinnes J, Sharma P, Berhane S, et al. Rapid, point-of-care antigen for use of hepatitis A vaccine for postexposure prophylaxis and for tests for diagnosis of SARS-CoV-2 infection. Cochrane Database Syst preexposure prophylaxis for international travel. MMWR Morb Rev. 2022:2022. Mortal Wkly Rep. 2018;67:1216–1220. 42 Healy B, Khan A, Metezai H, Blyth I, Asad H. The impact of false 36 Planitzer CB, Saemann MD, Gajek H, Farcet MR, Kreil TR. Cyto- positive COVID-19 results in an area of low prevalence. Clin Med. megalovirus neutralization by hyperimmune and standard 2021;21:e54. www.thelancet.com Vol 57 March, 2023 13