Convalescent Plasma Treatment and Mortality in COVID-19 Patients: an International Collaborative Meta-Analysis of Randomized Controlled Trials

Registered protocol

Convalescent plasma treatment and mortality in COVID-19 patients: an international collaborative meta-analysis of randomized controlled trials

Cathrine Axfors MD, PhDa,1,2, Perrine Janiaud PhDa,3, Andreas M. Schmitt MD3,4, Janneke van 't Hooft MD, PhD1,5, Emily R. Smith, ScD, MPH6, Noah A. Haber, ScD1, David Moher PhD7, Steven N. Goodman MD, PhD1,8,9, John P.A. Ioannidis MD, DSc1,8,9,10,11, Lars G. Hemkens MD, MPH1,3,11

All investigators of eligible trials will be invited as co-authors of the planned publication aEqual contributions, 1Meta-Research Innovation Center at Stanford (METRICS), Stanford University, Stanford, CA, USA, 2Department for Women’s and Children’s Health, Uppsala University, Uppsala, Sweden, 3Department of Clinical Research, University Hospital Basel, University of Basel, Basel, Switzerland, 4Department of Medical Oncology, University of Basel, Basel, Switzerland, 5Amsterdam University Medical Center, Amsterdam University, Amsterdam, the Netherlands, 6Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC, USA, 7Centre for Journalology, Clinical Epidemiology Program, Ottawa Hospital Research Institute, 501 Smyth Road, Ottawa K1H 8L6, Canada, 8Stanford University School of Medicine, Stanford, CA, USA, 9Department of Epidemiology and Population Health, Stanford University School of Medicine, Stanford, CA, USA, 10Stanford Prevention Research Center, Department of Medicine, Stanford University, Stanford, CA, USA, 11Meta-Research Innovation

Center Berlin (METRIC-B), Berlin Institute of Health, Berlin, Germany

Corresponding author: Lars G. Hemkens, MD, MPH Department of Clinical Research, University Hospital Basel, Spitalstrasse 12, CH-4031 Basel, Switzerland Phone: +41 61 265 3100 Email: [email protected]

Summary

Background: Convalescent plasma treatment to increase survival in COVID-19 patients is widely adopted in many settings; however, supporting evidence from randomized clinical trials (RCTs) is lacking. Although many RCTs are ongoing, recruitment is often slow and conclusive results based on individual trials will not arrive in time to benefit tens of thousands of patients during the fall and winter of 2020. Aim: To estimate the effect of convalescent plasma treatment on all-cause mortality in COVID-19 patients. Design: Collaborative meta-analysis of RCTs. Data sources: Mortality data from published and unpublished RCTs will be systematically identified in trial registries, literature databases and other repositories, including ClinicalTrials.gov, WHO International Clinical Trials Registry Platform, PubMed, and the Cochrane COVID-19 register. Aggregated all-cause mortality data from unpublished trials will be requested from trial investigators. Eligibility criteria for selecting studies: All ongoing, discontinued and completed RCTs that investigate convalescent plasma compared with placebo or no treatment in any patients with confirmed or suspected SARS-CoV-2 infection, regardless of patient setting, treatment schedule, and whether trials have mortality as a primary outcome or not. Outcomes: All-cause mortality. Statistical methods: Random-effects meta-analysis to combine the all-cause mortality effects from all trials as risk ratio with 95% confidence intervals using intention-to-treat data. Subgroup analyses will be performed for, e.g., patient setting, antibody level in the donated plasma, and control type. Implications: This rapid collaborative effort will serve patients and clinicians with all available RCT evidence to date. Funding: Swiss National Science Foundation (COVID-evidence; LGH), Laura and John Arnold Foundation (Meta-Research Innovation Center at Stanford [METRICS]), Uppsala University, Swedish Society of Medicine, Blanceflor Foundation, Sweden-America Foundation (CA). Keywords: Meta-analysis, SARS-CoV-2, COVID-19, Convalescent plasma

Introduction

Identifying treatments that increase the chance of survival in patients with coronavirus disease 2019 (COVID-19) is the focus of a global research agenda, spearheaded by randomized clinical trials (RCTs) such as the British RECOVERY megatrial.1 COVID-19, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been reported in more than 27 million people worldwide by the beginning of September, 2020.2 For some, especially elderly and people with pre-existing conditions, COVID-19 becomes severe enough to warrant supportive treatment (e.g., supplementary oxygen or mechanical ventilation) in a hospital or intensive care unit (ICU). The only specific treatment showing survival benefits so far is dexamethasone, through results from the RECOVERY trial.1,3

The transfer of plasma from a recovered (convalescent) COVID-19 patient to a person currently suffering the disease aims to achieve a transient, passive immunization that helps combat the ongoing infection. A measurable plasma component is the antibody level, commonly categorized as high-titer (hyperimmune) or low-titer (standard). From the preliminary evidence available, early administration of high-titer plasma is believed to be most effective.4 Convalescent plasma treatment has previously been used in, e.g., SARS-CoV-1, MERS, and H1N1 influenza,4–6 and preliminary reports suggest low incidence of adverse events in COVID-19 patients receiving plasma treatment.7 However, quoting a statement by the US Food and Drug Administration (FDA) in March, 2020, “although promising, convalescent plasma has not been shown to be effective in every disease studied”.8 Convalescent plasma treatment for COVID-19 in the US was approved through an Emergency Use Authorization (EUA) on August 23 by the FDA.9

Thousands of COVID-19 patients in the US and worldwide have received convalescent plasma outside clinical trials under the single-patient emergency investigational new drug and the National Expanded Access Protocol 7,10 and many more are expected to receive the treatment following the EUA. As commented by the US National Institute of Health,11 currently available evidence is limited to three RCTs 3,12,13 and several non-randomized studies (e.g., results from the National Expanded Access Protocol14). Although many trials on convalescent plasma are ongoing, the current recruitment rate 15,16 could mean several months and tens of thousands of treatments before enough RCT evidence is disseminated to inform on the benefits or harms. An unintended consequence of the EUA may be that trial recruitment will be further stalled 16 if convalescent plasma treatment is embraced as preferential to its control. Regarding mortality, an outcome warranting a large sample size at least when including patients at early stages of the disease, it is unlikely for any individual trial to present conclusive results before the winter.16 In this situation, a collaboration of investigators to pool together available data on all-cause mortality is expected to shorten the waiting by months.

We systematically searched PubMed, the Cochrane Library, and PROSPERO on September 28, 2020, for published and registered systematic reviews on RCTs investigating convalescent plasma treatment in COVID-19. There were four published systematic reviews that included two RCTs (in addition to non-randomized studies).17,18 Of thirteen systematic reviews registered at PROSPERO, only four planned to include preprints and none described any collaboration to pool non-disseminated data.

Aim

We aim to identify and combine all RCTs that investigate the effects of convalescent plasma treatment on all-cause mortality in any COVID-19 patients compared to placebo or no intervention.

Methods

This protocol is registered at the Open Science Framework before data collection and reported according to the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P).19 We will extract information from publications and preprints as well as invite the core investigator teams of previously non-disseminated trials to provide data and take part as co-authors in the planned publication. The majority of eligible trials are expected to be ongoing.

Eligibility criteria

1. Is described as a randomized controlled trial. 2. Includes persons with confirmed or suspected SARS-CoV-2 infection/COVID-19 in any healthcare setting (e.g., inpatient, outpatient, intensive care unit); i.e., we will not include prevention trials. 3. Participants in at least one study arm (experimental) are intended to receive convalescent plasma under any treatment schedule and regardless of definition of convalescent cases. 4. Participants in at least one study arm (control) are intended to not receive convalescent plasma treatment. Beyond plasma treatment or placebo infusion, there are no intended differences between the experimental and control arms in treatment (which may include any interventions, such as dexamethasone). 5. At least one participant has been enrolled in the experimental group, and at least one participant has been enrolled in the control group. 6. Data are available for all-cause mortality at any time-point.

No exclusions will be made based on trial status (e.g., ongoing, completed, or discontinued), language, geographical location, or primary outcome.

Information sources and search strategy

Eligible trials will be identified at ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform [ICTRP] as of September 28, 2020, through the COVID-evidence database. We will also search PubMed and the Cochrane COVID-19 trial registry (preprints, trial registries and literature databases) as of September 28, 2020, using a PRESS peer-reviewed search strategy with terms related to convalescent plasma and COVID-19 with a standard RCT filter.20 For a detailed search strategy see Supplement 1. Finally, we will compare our results with other published or registered systematic searches (e.g.,21).

Data extraction

Two reviewers will independently extract the following information from all included RCTs to be verified/complemented by collaborating trial investigators: experimental treatment (including e.g., antibody titers and volume administered), control treatment, treatment schedule, patient setting, eligibility criteria for recipients, eligibility criteria for donors (e.g., sex, and disease severity), study location, blinding, target sample size, and trial status. Trials will be categorized as published (in a peer-reviewed journal or posted on a preprint server), or unpublished.

We will contact the investigator teams of all eligible trials of which the results are not previously disseminated (Supplement 2, email invitation). We will ask for mortality data (the number of randomized patients and the number of deceased patients per treatment arm, as intention-to-treat) and confirmation of the trial characteristics as per above.

Outcome

We will only focus on all-cause mortality.

Risk of bias

This review of publish and unpublish RCTs aims at providing a rapid overview on all-cause mortality, we do not intend to conduct a risk of bias assessment for individual studies. However, we will address factors related to risk of bias across studies (publication status, blinding [i.e., placebo control or not] and loss to follow-up) in our subgroup analyses.

Statistical analysis

Absolute numbers and proportions will be reported, in addition to the treatment effect estimate (risk ratio = risk of death in the experimental arm divided by the risk of death in the control arm) with 95% confidence intervals (CIs). We will perform a meta-analysis to combine risk ratios across all trials and

statistical heterogeneity will be described with the I2-statistic. In multi-arm studies, if several eligible comparisons are available we will consider them separately. The main meta-analysis will be run using R (‘meta’ package).

We expect a large variation in sample size and in the number of outcome events across trials, with a proportion of trials presenting with zero events in one or both arms. Therefore, we expect a Hartung- Knapp-Sidik-Jonkman (HKSJ) random-effects model with Paule and Mandel (PM) estimator to perform well in terms of equality of weights between trials. The following variants will be calculated, including the main model and sensitivity analyses.

Model:

1. HKSJ random-effects model (main model). 2. Mantel-Haenszel model. 3. Peto model. 4. Profile likelihood model. 5. DerSimonian-Laird model.

Between-study heterogeneity estimator (tau-square):

1. PM estimator (main model). 2. DerSimonian-Laird estimator. 3. Sidik-Jonkman estimator.

Zero events in one study arm:

1. Correction by adding the reciprocal of the size of the contrasting arm (main model). 2. Arcsine difference.

Subgroup analyses

We will stratify trials by the following characteristics:4,21 - Publications/preprints versus unpublished trials - Patient setting (as proxy to COVID-19 severity: [1] outpatients, [2] ICU patients, [3] inpatients with supplemental oxygen, [4] inpatients without supplemental oxygen) - Antibody titer level (confirmed high-titer versus low-titer or unconfirmed). We define high- titer as S-protein receptor binding domain (RBD)-specific IgG antibody titer of 1:640 or higher, or serum neutralization titer of 1:40 or higher.13 - Control type (placebo control versus other)

- Timing of treatment relative to onset of symptoms (enrollment maximally 14 days after symptom onset versus more than 14 days after symptom onset or no exclusion based on timing of symptom onset). - Diagnostic confirmation for recipients (confirmed SARS-CoV-2 versus suspected cases) - Donor pregnancy history (excluding women with prior pregnancy versus not excluding women with prior pregnancy) - Donor severity criteria (severe COVID [e.g. hospitalization] versus mild or not specified) - Loss to follow-up (more than 10% of randomized participants versus 10% or less) If the main model and the subgroup analyses yield different conclusions, or show the same trend with different certainty, the results will be interpreted with more caution. Additional analyses may be added, which will be marked as exploratory.

Declaration of interest

We have no conflicts of interest to declare.

Funding

This collaborative meta-analysis is supported by the Swiss National Science Foundation (grant supporting COVID-evidence, 31CA30_196190) and the Laura and John Arnold Foundation (grant supporting the post-doctoral fellowship at the Meta-Research Innovation Center at Stanford (METRICS), Stanford University). Funding also includes postdoctoral grants from Uppsala University, the Swedish Society of Medicine, the Blanceflor Foundation, and the Sweden-America Foundation (C. Axfors). The funders have had no role in the design of this collaborative meta-analysis, and will have no role in the collection, analysis, and interpretation of data, or in the report writing.

Planned contributions

Lars G. Hemkens, Cathrine Axfors, and Perrine Janiaud will have full access to all data in this study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Lars G. Hemkens, John P. A. Ioannidis, Steven N. Goodman, David Moher, Cathrine Axfors, Perrine Janiaud, Andreas M. Schmitt, Emily R. Smith, Noah A. Haber Acquisition, analysis, or interpretation of data: All authors Drafting of the manuscript: Cathrine Axfors, Perrine Janiaud, Andreas M. Schmitt, Lars G. Hemkens Critical revision of the manuscript for important intellectual content: All authors

Statistical analysis: Perrine Janiaud, Andreas M. Schmitt, Lars G. Hemkens, John P. A. Ioannidis, Steven N. Goodman Approval of the final manuscript: All authors Obtained funding: Lars G. Hemkens, Cathrine Axfors, John P. A. Ioannidis Administrative, technical, or material support: Cathrine Axfors, Perrine Janiaud, Andreas M. Schmitt, Lars G. Hemkens, Janneke van ‘t Hooft, Emily R. Smith, Noah A. Haber Supervision: Lars G. Hemkens

Data sharing

All data and R code used for the meta-analysis will be shared alongside the planned publication.

References

1. RECOVERY Collaborative Group, Horby P, Lim WS, et al. Dexamethasone in Hospitalized Patients with Covid-19 - Preliminary Report. N Engl J Med. Published online July 17, 2020. doi:10.1056/NEJMoa2021436

2. COVID-19 Map - Johns Hopkins Coronavirus Resource Center. Accessed September 9, 2020. https://coronavirus.jhu.edu/map.html

3. WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group, Sterne JAC, Murthy S, et al. Association Between Administration of Systemic Corticosteroids and Mortality Among Critically Ill Patients With COVID-19: A Meta-analysis. JAMA. Published online September 2, 2020. doi:10.1001/jama.2020.17023

4. Barone P, DeSimone RA. Convalescent plasma to treat coronavirus disease 2019 (COVID-19): considerations for clinical trial design. Transfusion . 2020;60(6):1123-1127.

5. Arabi YM, Hajeer AH, Luke T, et al. Feasibility of Using Convalescent Plasma Immunotherapy for MERS-CoV Infection, Saudi Arabia. Emerg Infect Dis. 2016;22(9):1554-1561.

6. Mair-Jenkins J, Saavedra-Campos M, Baillie JK, et al. The effectiveness of convalescent plasma and hyperimmune immunoglobulin for the treatment of severe acute respiratory infections of viral etiology: a systematic review and exploratory meta-analysis. J Infect Dis. 2015;211(1):80-90.

7. Joyner M, Wright RS, Fairweather D, et al. Early Safety Indicators of COVID-19 Convalescent Plasma in 5,000 Patients. medRxiv. Published online May 14, 2020. doi:10.1101/2020.05.12.20099879

8. Tanne JH. Covid-19: FDA approves use of convalescent plasma to treat critically ill patients. BMJ. 2020;368:m1256.

9. Office of the Commissioner. FDA Issues Emergency Use Authorization for Convalescent Plasma as Potential Promising COVID–19 Treatment, Another Achievement in Administration’s Fight Against Pandemic. Published 2020. Accessed September 9, 2020. https://www.fda.gov/news-events/press- announcements/fda-issues-emergency-use-authorization-convalescent-plasma-potential-promising-covid- 19-treatment

10. Budhai A, Wu AA, Hall L, et al. How did we rapidly implement a convalescent plasma program? Transfusion . 2020;60(7):1348-1355.

11. Statement on Convalescent Plasma EUA. Accessed September 9, 2020. https://www.covid19treatmentguidelines.nih.gov/statement-on-convalescent-plasma-eua/

12. Gharbharan A, Jordans CCE, GeurtsvanKessel C, et al. Convalescent Plasma for COVID-19. A randomized clinical trial. MEDRxiv. Published online 2020. doi:10.1101/2020.07.01.20139857

13. Li L, Zhang W, Hu Y, et al. Effect of Convalescent Plasma Therapy on Time to Clinical Improvement in Patients With Severe and Life-threatening COVID-19: A Randomized Clinical Trial. JAMA. 2020;324(5):460-470.

14. Joyner MJ, Senefeld JW, Klassen SA, et al. Effect of Convalescent Plasma on Mortality among Hospitalized Patients with COVID-19: Initial Three-Month Experience. medRxiv. Published online August 12, 2020. doi:10.1101/2020.08.12.20169359

15. Angus DC. Optimizing the Trade-off Between Learning and Doing in a Pandemic. JAMA. Published online March 30, 2020. doi:10.1001/jama.2020.4984

16. Ledford H. US WIDENS ACCESS TO COVID-19 PLASMA — DESPITE LACK OF DATA. Nature. https://media.nature.com/original/magazine-assets/d41586-020-02324-2/d41586-020-02324-2.pdf. Published August 21, 2020. Accessed September 9, 2020.

17. Sarkar S, Soni KD, Khanna P. Convalescent plasma is a clutch at straws in COVID-19 management! A

systematic review and meta-analysis. J Med Virol. Published online August 10, 2020. doi:10.1002/jmv.26408

18. Piechotta V, Chai KL, Valk SJ, et al. Convalescent plasma or hyperimmune immunoglobulin for people with COVID‐19: a living systematic review. Cochrane Database Syst Rev. 2020;(7). doi:10.1002/14651858.CD013600.pub2

19. Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1.

20. Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons; 2019.

21. Murphy M, Estcourt L, Grant-Casey J, Dzik S. International Survey of Trials of Convalescent Plasma to Treat COVID-19 Infection. Transfus Med Rev. Published online June 27, 2020. doi:10.1016/j.tmrv.2020.06.003

Supplements

1. Search strategy 2. Email invitation 3. Identified potentially eligible trials

Supplement 1. Search strategy

The COVID-evidence database includes trials registered on ClinicalTrials.gov or the WHO International Clinical Trials Registry Platform up to September 28, 2020, as well as trials posted or published at the following sites up to April 9, 2020: PubMed, medRxiv, bioRxiv, the WHO COVID-19 literature database, and a listing of all trials with ethical approval in Switzerland (for details please see the COVID-evidence study protocol on the Open Science Framework: http://dx.doi.org/10.17605/OSF.IO/GEHFX). This Supplement describes the search strategy used to complement the COVID-evidence database with trials registered, posted or published after April 9, 2020.

1. PubMed

PubMed will be searched from inception to September 28, 2020. Search terms for PubMed include extensive controlled vocabulary and Medical Subject Headings (MeSH):

#1 corona[ti] OR covid*[ti] OR sars[ti] OR severe acute respiratory syndrome[ti] OR ncov*[ti] OR "severe acute respiratory syndrome coronavirus 2" [Supplementary Concept] OR "COVID-19" [Supplementary Concept] OR (wuhan[tiab] AND coronavirus[tiab]) OR (wuhan[tiab] AND pneumonia virus[tiab]) OR COVID19[tiab] OR COVID-19[tiab] OR coronavirus 2019[tiab] OR SARS-CoV-2[tiab] OR SARS2[tiab] OR SARS-2[tiab] OR "severe acute respiratory syndrome 2"[tiab] OR 2019-nCoV[tiab] OR (novel coronavirus[tiab] AND 2019[tiab]) NOT (animals[mesh] NOT humans[mesh]) AND ("2019/12/01"[EDAT] : "3000/12/31"[EDAT])

#2 ((((plasma[MeSH Terms]) OR (serum[MeSH Terms])) OR (plasma[Title/Abstract])) OR (serum[Title/Abstract])) OR (convalescen*[Title/Abstract])

# 3 (randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR placebo[tiab] OR clinical trials as topic[mesh:noexp] OR randomly[tiab] OR trial[ti] NOT (animals[mh] NOT humans [mh]))

#1 AND #2 AND #3

2. Cochrane COVID-19 trial registry

The Cochrane COVID-19 trial registry will be searched from inception to September 28, 2020. Search terms will be “convalescent OR plasma”, and we will use the filter categories “Intervention assignment” (“randomised”), “Study aim” (“Treatment and management”), and “Study type” (“Intervention”).

3. L·OVE Platform - Epistemonikos

The L·OVE Platform will be searched from inception to September 28, 2020. We will use the filter categories “Prevention and treatment”, “Procedures - convalescent plasma”, and within the primary studies we will select the filters “by reported data - has data” and “by type of study - RCT”.

4. Other sources

Finally, we will complement our results with trials identified by other published or registered systematic searches as well as personal knowledge.

Supplement 2. Email invitation

Subject: Co-author invitation, international collaborative meta-analysis on convalescent plasma and mortality in COVID-19

Send to:

«Contact_email»; «Contact_email_2»

Dear «Title» «Contact_last_name» «and_» «Contact_last_name_2»,

We would like to invite your trial investigation team to collaborate as co-authors on a rapid international meta-analysis on the effect of convalescent plasma on mortality in COVID-19 in randomized controlled trials (RCTs), together with our collaborative research group coordinated at Stanford University, USA, and the University of Basel, Switzerland. Your study «Acronym_or_title» is highly relevant for this project.

As you are aware, the US Food and Drug Administration (FDA) issued an emergency use authorization for convalescent plasma treatment in COVID-19 on August 23, although randomized evidence is very limited. Our searching suggests that almost 90 RCTs are ongoing; however, no single trial is projected to yield definitive results regarding mortality in 2020. Therefore, we are sending this invitation to investigators of all ongoing, discontinued or completed RCTs that evaluate convalescent plasma treatment for COVID-19 against placebo or no treatment. With your help, we are hoping to provide a rapid summary of the current evidence using a meta-analysis of group-level, aggregated, all-cause mortality data (NOT individual patient data). We aim to rapidly publish the results in a peer-reviewed journal and your core team is invited to co-author the publication.

We have no commercial interest with this work, and our protocol can be found attached as well as registered on the Open Science Framework: [link]. Trials that are eligible for this project are listed in the protocol (Supplement 3, last pages).

As collaborators, we kindly ask you to answer the following questions before October 25. If you are interested in collaborating but uncertain whether the data may be shared before October 25, please respond as soon as possible.

Question 1: Could you please confirm that these criteria apply to your trial? a) The trial is randomized b) The trial has at least one group of patients who receive convalescent plasma c) The trial has at least one control group that does NOT receive convalescent plasma

Question 2: For each of your study arms, a) What intervention did this group receive? b) How many patients were randomized to this group? c) Of these patients, how many have died? d) Of these patients, for how many it is unknown if they are dead or alive?

Once we have completed the required data analyses, a draft manuscript will be shared with all co- authors for comments. We will also ask you to confirm or complement the trial design characteristics listed in the protocol for potential subgroup analyses (attached). The finalized version will be submitted to a peer-reviewed medical journal.

From our perspective, this initiative does of course NOT stand in conflict with taking part in other collaborations or publishing your trial results individually. The focus of this meta-analysis is limited to all-cause mortality. It does not include other outcomes, and it will not use individual patient data. With this publication we aim to make sure that all clinical trial data hitherto collected (unpublished or published) will be of use, regardless of whether the target sample size of each trial was reached or not.

If you know of other collaborations, we would be thankful for contact information so that we may coordinate and optimize our efforts.

Thank you for considering our request and please do not hesitate to ask for clarifications.

Kind regards,

Lars Hemkens, Steve Goodman, David Moher, John Ioannidis, Cathrine Axfors, Perrine Janiaud, and Andreas Schmitt for the COVID-evidence team

Our COVID-evidence database (www.covid-evidence.org) is used for this work, supported by the Swiss National Science Foundation (Project ID 196190) and a large collaboration of researchers from Switzerland, USA, China, Canada, UK, France, Germany, Austria, Sweden, the Netherlands, and other countries.

Registered protocol

Supplement 3. Identified potentially eligible trials

Acronym, registration Experimental Control Setting Location Arms Target sample Contact last name number or title treatment treatment size RECOVERY Convalescent plasma Standard of care Inpatient United 8 15000 Haynes Kingdom REMAP-CAP Convalescent plasma + Standard of care ICU International >8 7100 Green standard of care ASCOT Convalescent plasma Standard of care Inpatient Australia 5 2400 Mora CCAP NA Placebo + Inpatient Denmark 2 1500 Benfield standard of care CSSC-004 Convalescent plasma Placebo Outpatient USA 2 1344 Sullivan CONCOR-1 Convalescent plasma Standard of care Inpatient Canada and 2 1200 Arnold USA VA CURES-1 Convalescent plasma Placebo Inpatient USA 2 702 Janoff PROTECT-Patient study Convalescent plasma + Placebo + Inpatient South Africa 2 600 Nyoni standard of care standard of care C3PO Convalescent plasma Placebo Outpatient United 2 600 Schulman States PassItOnII Convalescent plasma Placebo Inpatient United 2 500 Bistran-Hall States CONFIDENT NA NA NA Belgium 2 500 Misset DAWN-Plasma Convalescent plasma + Standard of care Inpatient Belgium 2 483 Meyfroidt standard of care PLATINA TRIAL Convalescent plasma Standard of care Inpatient India 2 472 Meshram PLACID Convalescent plasma + Standard of care Inpatient India 2 452 NA standard of care COV-PLAS Convalescent plasma Placebo Inpatient Netherlands 2 430 Zwaginga registered as NCT04342182 Convalescent plasma + Standard of care Inpatient Netherlands 2 426 Rijnders standard of care EPCOvid-1 Convalescent plasma Placebo Inpatient Mexico 2 410 Sierra-Madero ILBS-COVID-04 Convalescent plasma + Standard of care Inpatient India 2 400 Bajpai standard of care

COV2-CP Convalescent plasma + Standard of care Inpatient Italy 2 400 Talarico standard of care PLASM-AR Convalescent plasma + Placebo Inpatient Argentina 2 333 Belloso standard of care registered as NCT04456413 Convalescent plasma Standard of care Outpatient United 2 306 Vendivil States CONTAIN COVID-19 Convalescent plasma Placebo Inpatient United 2 300 Ortigoza States ConPlas-19 Convalescent plasma + Standard of care Inpatient Spain 2 278 Avendaño standard of care PLASMA COVID-19 Convalescent plasma + Standard of care Inpatient Colombia 2 236 Quintero standard of care registered as NCT04391101 Convalescent plasma + Standard of care ICU Colombia 2 231 Perilla standard of care ESCAPE Convalescent plasma Placebo Inpatient United 2 220 Kaufman States PLAVID-TRIAL Convalescent plasma + Standard of care Inpatient Brazil 2 220 NA standard of care registered as Convalescent plasma + Standard of care Inpatient Kenya 2 206 Adembesa PACTR202007653923168 standard of care registered as Convalescent plasma + Standard of care Inpatient China 2 200 Cao Bin ChiCTR2000029757 standard of care registered as NCT04390503 Convalescent plasma Placebo Outpatient United 4 200 Justman States IMPACT Convalescent plasma + Placebo + Inpatient Ecuador 2 200 Baldeon standard of care standard of care LIFESAVER Convalescent plasma + Standard of care Unclear Italy 2 182 Landolfi standard of care PLACO-COVID Convalescent plasma + Placebo + Inpatient Italy 3 180 NA standard of care standard of care; standard of care RECOVER Convalescent plasma Standard of care Inpatient Germany 2 174 Müller-Tidow PLACOVID NA NA NA Brazil 2 160 Sekine

MIDC-CCP Convalescent plasma Standard of care Outpatient United 2 150 Van Hise States PC-COVID-HCM NA NA NA Mexico 2 150 Torres Co-CLARITY NA NA NA Philippines 2 136 Gauiran COVIDIT NA NA NA Uganda 2 136 NA TSUNAMI Convalescent plasma + Standard of care Inpatient Italy 2 126 Falcone standard of care COOPCOVID-19 Convalescent plasma + Standard of care Inpatient Brazil 3 120 Nakagawa standard of care CORIPLASM Convalescent plasma Standard of care Unclear France 2 120 Lacombe registered as Convalescent plasma Standard of care Inpatient Iran 3 120 Asghari IRCT20200501047258N1 registered as NCT04333251 Convalescent plasma Standard of care Inpatient United 2 115 Baylor Research States Institute team CAPSID Convalescent plasma Standard of care Inpatient Germany 2 106 Körper registered as NCT04359810 Convalescent plasma Placebo Inpatient United 2 105 O'Donnell States registered as Convalescent plasma + Standard of care Inpatient China 2 100 Le Aiping ChiCTR2000030179 standard of care registered as Convalescent plasma Ordinary plasma Inpatient China 2 100 Zhang Dingyu ChiCTR2000030010 PERUCONPLASMA Convalescent plasma + Standard of care Inpatient Peru 2 100 Krapp standard of care CONCOR-KIDS Convalescent plasma + Standard of care Inpatient Canada 2 100 Upton standard of care MHC-COVID-19-CP Convalescent plasma Standard of care Inpatient India 2 100 Pathak registered as Convalescent plasma + Standard of care Inpatient Iran 2 100 Ghelichkhani IRCT20120215009014N353 standard of care LACCPT Convalescent plasma Placebo Inpatient Nigeria 2 100 Abayomi COVID PLASMA STUDY Convalescent plasma Standard of care Inpatient India 2 100 Chawla registered as NCT04547127 NA NA NA Spain 2 100 Torres registered as NCT04467151 Convalescent plasma Placebo Inpatient United 2 96 McIntee States

registered as NCT04332835 Convalescent plasma + Standard of care Inpatient Colombia 2 90 Anaya standard of care PennCCP-02 Convalescent plasma + Standard of care Inpatient United 2 80 Bar standard of care States CPC-SARS Convalescent plasma + Placebo + Inpatient Mexico 4 80 Perez-Calatayud standard of care standard of care PLASCOSSA NA Placebo Inpatient France 2 80 Martinaud PICP19 Convalescent plasma Standard of care Inpatient India 2 80 Ray LFCOLCOVID-19-001 Convalescent plasma + Standard of care Inpatient Colombia 3 75 Jaramillo standard of care PC/COVID-19 Convalescent plasma Standard of care Inpatient Spain 2 72 Cardesa registered as Convalescent plasma Ordinary plasma Inpatient China 2 60 Binghong Zhang ChiCTR2000030929 registered as NCT04442958 Convalescent plasma + Standard of care ICU Turkey 2 60 NA standard of care registered as NCT04528368 Convalescent plasma + Standard of care Inpatient Brazil 2 60 Rego standard of care registered as NCT04380935 Convalescent plasma + Standard of care Inpatient Indonesia 2 60 Sinto standard of care registered as NCT04385186 Convalescent plasma + Standard of care Inpatient Colombia 2 60 Zuluaga standard of care registered as NCT04392414 Convalescent plasma Placebo Unclear Russian 2 60 Konoplyannikov Federation registered as Convalescent plasma + Standard of care Inpatient Iran 2 60 Jam IRCT20150808023559N21 standard of care registered as Convalescent plasma + Standard of care Inpatient Iran 2 60 Farahani IRCT20200404046948N1 standard of care "Randomized pilot study for Convalescent plasma Standard of care ICU Spain NA 60 Cid the use of convalescent plasma in patients with COVID-19 disease" registered as Convalescent plasma + Standard of care Inpatient India 2 60 Kuzhali CTRI/2020/05/025346 Standard of care

registered as Convalescent plasma + Standard of care Inpatient China 2 50 Cao Bin ChiCTR2000030702 standard of care registered as NCT04442191 Convalescent plasma Placebo Inpatient United 2 50 Herrick States registered as NCT04421404 Convalescent plasma Placebo Inpatient United 2 50 Park States CPT-COVID-19 Convalescent plasma Standard of care Inpatient Iraq 2 49 NA registered as Convalescent plasma + Standard of care Inpatient Iran 3 45 Puladzadeh IRCT20200310046736N1 standard of care CPCP Convalescent plasma + Standard of care Inpatient Vietnam 2 44 Phuong Hoang standard of care Nguyen registered as NCT04356534 Convalescent plasma Standard of care Inpatient Bahrain 2 40 Al Qahtani registered as Convalescent plasma + Standard of care Inpatient Iran 2 40 Hadinedoushan IRCT20200503047281N1 standard of care Not registered Convalescent plasma Standard of care Outpatient Germany NA 40 Meybohm and inpatient registered as NCT04468009 Convalescent plasma Standard of care ICU Argentina 2 36 Gonzalez registered as Convalescent plasma + Standard of care Inpatient China 2 30 Guojun Zhang ChiCTR2000030627 standard of care registered as NCT04530370 Convalescent plasma Placebo Inpatient Egypt 2 30 Salman registered as NCT04385199 Convalescent plasma Standard of care Inpatient United 2 30 Tatem States COP-COVID-19 Convalescent plasma + Standard of care Inpatient Mexico 2 30 Perez standard of care ILBS-COVID-02 Convalescent plasma + Placebo + Inpatient India 2 29 Bajpai standard of care standard of care registered as NCT04403477 Convalescent plasma + Standard of care Inpatient Bangladesh 3 20 Rahman standard of care NA Convalescent plasma Standard of care Inpatient India 2 20 Pardiwalla PROMETEO Convalescent plasma + Placebo + ICU Mexico 2 15 Martinez standard of care standard of care registered as Convalescent plasma + Standard of care Inpatient Iran 3 15 Zangoue IRCT20200413047056N1 standard of care

Not registered NA NA NA Norway NA NA Hervig "Konvalescent plasma för NA NA NA Sweden NA NA Dillner behandling av COVID-19: En randomiserad kontrollerad studie"