Safety and efficacy of convalescent plasma for COVID-19 in the pre-vaccine era, and implementation in a large healthcare system: A phase I/II randomized, multicentre clinical trial (PC-COVID)

Abstract

Introduction

A phase 2 trial was conducted to evaluate the safety and efficacy of convalescent plasma (CP) for patients admitted with COVID-19 in the pre-vaccine era.

Methods

CP was compared to standard of care (SoC) in a randomized clinical trial in patients admitted with COVID-19 pneumonia and with either hypoxia and/or some underlying condition from April to December 2020 in 15 hospitals in Andalusia, Spain. The primary endpoints were adverse events and failure (all-cause death and need for mechanical ventilation during the first 21 days).

Result

The procedure for CP was developed and worked properly. Overall, 72 patients of 88 evaluated were recruited (37 and 35) assigned to CP and SoC. Median duration of COVID-19 related symptoms was 7 and 6 days, and median oxygen saturation at randomization was 93% and 92.5% in experimental and control arms, respectively. Overall, 8 (21.6%) patients in the experimental CP group (non related to CP) and 8 (22.9%) in the control group developed grade 3 adverse events (relative risk [RR] 0.94; 95% CI: 0.39–2.24); failure occurred in 3 (8.1%) and 4 (11.4%) patients, respectively ([RR] 0.94, CI: 0.14–6.35 for both events). No differences in secondary endpoints (change in inflammatory parameters, negativization of SARS-CoV-2 PCR or duration of hospital stay) were found.

Conclusions

The results suggested that CP is safe for use in COVID-19 patients, although no significant efficacy signal could be observed. A successful regional system for CP administration was developed.

Resumen

Introducción

Se llevó a cabo un ensayo de fase II para evaluar la seguridad y la eficacia del plasma de convalecientes (PC) en pacientes ingresados con COVID-19 en la era previa a la vacuna.

Métodos

Se desarrolló un procedimiento para la obtención, preparación, transporte y administración de PC en toda Andalucía. El PC se comparó con el tratamiento estándar (TE) en un ensayo clínico aleatorizado en pacientes ingresados con neumonía por COVID-19 y con hipoxia y/o alguna afección subyacente entre abril y diciembre de 2020 en 15 hospitales de Andalucía, España. Los criterios de valoración principales fueron los eventos adversos y el fracaso (muerte por cualquier causa y necesidad de ventilación mecánica durante los primeros 21 días).

Resultados

Se reclutaron 72 pacientes de los 88 evaluados (37 y 35 asignados a PC y TE). En total, 8 (21,6%) pacientes del grupo experimental de PC (no relacionados con la CP) y 8 (22,9%) del grupo de control desarrollaron eventos adversos de grado 3 (riesgo relativo [RR] 0,94; IC del 95%: 0,39-2,24); hubo fracaso en 3 (8,1%) y 4 (11,4%) pacientes, respectivamente (RR 0,94; IC del 95%: 0.14-6.35 para ambos fracasos). No se encontraron diferencias en los objetivos secundarios (cambio en los parámetros inflamatorios, negativización de la PCR del SARS-CoV-2 o duración de la estancia hospitalaria).

Conclusión

Los resultados de este ensayo piloto sugirieron que el PC es seguro para su uso en pacientes con COVID-19, aunque no se pudo demonstrar eficacia significativa. Se desarrolló un sistema exitoso para la administración de PC a nivel autonómico.

Introduction

According to the WHO, the COVID-19 pandemic has caused more than 7 million deaths worldwide [https://data.who.int/dashboards/covid19/deaths visited on 23.10.2025]. While massive vaccination has very significantly reduced the burden of disease, the emergence of new SARS-CoV-2 variants are a threat for high-risk population including the elderly and immunocompromised patients.

Early in the pandemic, convalescent plasma was considered a potentially effective treatment. In fact, some studies, mostly performed in unvaccinated patients, suggested its efficacy.^1,2 However, a recent meta-analysis did not find a beneficial effect,³ and most guidelines do not recommend its routine use for the treatment of COVID-19.^4,5 Nevertheless, convalescent plasma could be reconsidered for specific at-risk populations such as elderly, unvaccinated and immunocompromised patients. Although COVID-19 is now endemic in a largely immune population, the aforementioned populations remain at risk for severe outcome. Results from previous studies suggest that more research is needed to investigate the potential benefits of convalescent plasma (CP) in COVID-19 patients, particularly if administered early enough and with high neutralizing antibody titers; benefits might be more important in immunocompromised patients.^6–8

Convalescent plasma might also play a role in future pandemics before vaccines and specific therapeutic drugs are available and therefore, implementing services capable of distributing convalescent plasma is recommended as part of preparedness for new pandemics.⁶

During the first waves of the COVID-19 pandemic, we conducted a randomized controlled trial with the objective of detecting safety and efficacy signals of convalescent plasma in COVID-19 patients. Here, we present the results of the trial and the implementation of a service capable of providing convalescent plasma to patients in Andalusia, a region in the south of Spain with more than 8.5 million population and 87,000 km².

Methods

Study design

A multicentre, two-arm, controlled, randomized clinical trial was conducted to evaluate the safety of convalescent plasma after pathogen reduction process (Terumo BCT, Zaventem, Belgium; Mirasol® System) added to standard of care compared to a control group receiving standard of care alone. The design was carried out jointly with the Andalusian Network for the Design and Translation of Advanced Therapies, which also acted as sponsor. The study was performed from April to December 2020 in 15 hospitals in Andalusia, Spain (13 public and 2 privates), located in 7 of the 8 provinces of the region.

Study subjects

During the study period, all patients presenting with respiratory symptoms were evaluated for COVID-19 at the participating hospitals. Adult patients hospitalized with pneumonia (i.e., acute respiratory signs/symptoms and pulmonary infiltrates in chest X-ray or CT scan) caused by SARS-CoV-2 as diagnosed by a positive RT-PCR in nasopharyngeal swab or respiratory specimen were eligible. To be included, patients must have at least one of the following criteria: oxygen saturation at room air ≤94% or PaO₂/FiO₂ ≤300 mm Hg; age >65 years; or an underlying risk condition including hypertension, chronic heart failure, chronic pulmonary disease, liver cirrhosis, diabetes mellitus and obesity (body mass index ≥30). Patients were excluded if the diagnostic RT-PCR had been performed >72 h before, were symptomatic for ≥10 days, were already on mechanical ventilation, had already received immunomodulator drugs (tocilizumab, sarilumab, anakinra, baricitinib, interferon), were being previously treated with biological immunosuppressive drugs, incompatibility or allergy to human plasma, grade 4 renal insufficiency or dialysis, pregnancy or lactation.

Intervention

Participants were randomized to receive the investigational product, plasma from convalescent donors of COVID-19 after pathogen reduction process (Terumo BCT, Zaventem, Belgium; Mirasol® System) (experimental arm, see below) or not (control arm). In addition, patients in both arms were treated with what was protocolized at the participating sites as the standard of care at the time. The study was open-label, and therefore no blinding procedures were used. Randomization was performed centrally using a predefined code in 18 blocks of 4 patients each.

Endpoints and variables

The primary safety endpoint was the proportion of patients developing adverse events grade 2 (moderate) or higher according to the Common Terminology Criteria for Adverse Events (CTCAE) scale, V5.0.

The primary efficacy endpoint was “failure” which included all-cause death and need for mechanical ventilation during the first 21 days after randomization. The study protocol also included an increase in SOFA score >3 from baseline as a criterium for failure; however, due to difficulties in measuring the score during the first days of follow-up during the very busy first waves of the pandemic, this criterium was not used.

Secondary and exploratory endpoints included mortality at days 14 and 28, length of hospital stay, need to use immunomodulatory drugs (tocilizumab, sarilumab, baricitinib, anakinra), 1.5-fold increase in biomarker levels (i.e., ferritin, C-reactive protein, IL-6, D-dimer), negative RT-PCR at days 7 and 21, and change in SARS-CoV-2 IgM and IgG antibody titers.

Study visits were performed at randomization (day 0) and on days 1, 3, 7, 10, 14, 21 and 28. Data collected included demographics, underlying diseases, acute signs and symptoms, imaging and laboratory data and treatments received.

Study approval and ethical aspects

The study was approved by the Ethics Committee for Clinical Research of the Virgen Macarena and Virgen del Rocio University Hospitals. All participants gave their consent to participate in the study and the clinical trial was conducted according to the principles of Good Clinical Practice (GCP). The trial is registered on ClinicalTrials.gov with ID NCT04366245.

Statistical analysis

Although the study was considered as a phase I/II trial, an explorative sample size of 72 patients (36 in each arm) was calculated to demonstrate superiority of plasma, based on 25% difference in mortality between the SOC arm (25%) and experimental arm (5%) with 80% power and 5% alfa error. There were no available estimations for mortality with convalescent plasma when the study was being designed; because this study is a phase I/II trial, the mortality difference was established as the hypothetical. The analyses were performed in the intention-to-treat population which included all randomized patients. Categorical variables were compared by Chi square or Fisher test as appropriate, and continuous variables by Mann–Whitney U test. Time until discharge was compared by log-rank test and plotted by a Kaplan–Meier curve. Relative risks with 95% confidence intervals (CI) were calculated for dichotomous outcomes. IBM SPSS Statistics v19.0 was used for the analyses. Due to the limited sample size, we did not perform subgroup analyses. This report was made in accordance with CONSORT guidelines (Supplementary Table S2).

Implementation: plasma preparation and administration

Potential donors were identified by the medical staff of the Red Andaluza de Medicina Transfusional, Tejidos y Células (RAMTCC) (Andalusian Network for Transfusion Medicine, Tissues and Cells) which has multiple centres in each province of the Andalucia region. The medical staff was able to identify donors with the help of the Preventive Medicine unit and from local regional and registries of COVID-19 patients. Ideal patients were healthy subject who had fully recovered from non-severe COVID-19. Donors had the following criteria: adult person weighing >50 kg who had suffered from non-severe microbiologically confirmed COVID-19; were asymptomatic and SARS-CoV-2 RT-PCR negative at the time of donation (performed at least 14 days after cessation of acute symptoms); only donors with SARS-CoV-2 IgG in plasma equivalent to >1/640 (COVID-19 Spike Quantitative VIRCLIA IgG Monotest®, Vircell Microbiologists, Granada, Spain); and fulfilled all criteria for being apheresis plasma donors according to European and Spanish regulation (RD 1088/2005). Exclusion criteria included blood donation in the previous 30 days, plasmapheresis in the previous 7 days, donation of >25 L of plasma in the last year, and pregnancy. All donors had signed an informed consent form.

The serological studies were carried out using COVID-19 Spike Quantitative VIRCLIA IgG Monotest® (Vircell Microbiologists, Granada, Spain) following the manufacturer‘s protocol; serum samples were previously inactivated at 56 °C for 30 min; samples were analyzed immediately after inactivation or stored at 4 °C for no more than 4 days prior to testing.

Candidates were contacted by telephone and informed; if interested, a full questionnaire was carried out and if eligible, a visit was arranged. In the visit, informed consent was signed, answers to the questionnaire were checked, and RT-PCR for SARS-CoV-2 in nasopharyngeal swab and blood samples for usual tests performed in blood donors and for SARS-CoV-2 antibodies were taken. The results were reported to the donor by phone and, if the candidate was still eligible, a new visit was agreed for plasma donation in the following 10 days. In that visit, criteria for donation, blood pressure, heart rate and temperature were checked. A single identification number (SIN) was assigned to label all extraction bags and samples. Using an apheresis machine, 600 mL of de-leukocyted plasma were obtained, separated in 2 bags of 300 mL and processed immediately or preserved at <−20 °C until processing. This process was carried out in the transfusion centre closest to the patients and the samples were then shipped to the Seville centre for inactivation and to the Granada centre for storage.

A validated pathogen reduction process (Terumo BCT, Zaventem, Belgium; Mirasol® System) was applied to convalescent plasma according to the manufacturer's instructions. This method combines the addition of riboflavin to plasma samples for subsequent illumination with UV light, causing irreversible damage to the DNA and RNA of bacteria, viruses, and other possible pathogens, as well as white blood cells.

All processes were performed using closed connection systems to guarantee the sterility of the product. Plasma was frozen in the first 24 h from extraction and kept at ≤−18 °C, a temperature at which it retains its properties for three years. Plasma bags were labelled following the European rules for blood products and specifications about convalescent plasma. All processes were registered in the corporate information system (eProgesa) for traceability using the “reserved donation” procedure.

A blood sample was obtained from all patients assigned to the experimental arm to determine their blood group. Compatible plasma bags were transported from RAMTTC to the transfusion units of the patient's site at a temperature between −18 °C and −25 °C. Upon arrival to the site, the blood group compatibility was again checked. The units of convalescent plasma were administered following all requirements for transfusion of blood products, according to blood group compatibility and under the supervision of a medical specialist in Haematology or expert nurse staff. All procedures were registered in the local informatic systems for traceability. One bag of 250–300 mL of convalescent plasma was administered via a peripheral or central venous catheter (10 mL/min) to patients assigned to the experimental arm.

Convalescent plasma was delivered to the sites and administered to the patients in <24 h after randomization in 34 patients; in one patient it was administered between 24 and 48 h, and in other at 72 h.

Results

Patients features and outcomes

Overall, 88 patients were evaluated for participation in the trial; COVID-19 was not confirmed in 2, 12 had some exclusion criteria and 74 were randomized; 2 patients withdrew consent. Finally, the intention-to-treat population was formed by 37 patients assigned to plasma and 35 to control arm (Fig. 1).

Fig. 1.

Flowchart of patients’ recruitment.

The features of patients are summarized in Table 1. In summary, 43 (59.7%) were males, median age was 60 and 65 in patients assigned to plasma and control arm, respectively. Underlying conditions were frequent and similarly distributed in the two groups, except for hypertension, which was more frequent among control patients. Median duration of COVID-19 related symptoms was 7 and 6 days, and median oxygen saturation at randomization was 93% and 92.5% in experimental and control arms, respectively; more than 80% of patients in both groups had a bilateral pulmonary infiltrate. A numerically higher proportion of patients in the experimental arm had a SOFA ≥2. Levels of inflammatory markers were also similar in both groups. Most patients were treated with steroids and low-molecular weight heparin (prophylactic dosing), and one third with remdesivir. Basal levels of IgM and IgG antiSARS-CoV-2 antibodies were low and similar in both groups (SARS-CoV-2 vaccines were not available when the study was performed); IgM levels were higher at day 3 and 7 in the experimental arm; evolution of IgG levels were similar in both groups (Fig. 2 and Table S3).

Table 1.

Baseline features of patients included in the intention-to-treat population. Data are number of patients (percentage) except where specified. For features with missing data, the denominator is provided.

	Plasma(n = 37)	Control(n = 35)	P
Recruitment period			0.72
April–September 2020	12 (32.4)	10 (28.6)
October–December 2020	25 (67.6)	25 (71.4)
Male gender	21 (56.8)	17 (48.6)	0.48
Age in years, median (IQR)	60 (47–72)	65 (55–72)	0.80
Underlying conditions
Heart disease	11 (29.7)	6 (17.1)	0.20
Hypertension	13 (35.1)	23 (65.7)	0.009
Diabetes mellitus	7 (18.9)	10 (28.6)	0.33
Chronic obstructive pulmonary disease	5 (13.5)	3 (8.6)	0.71
Asthma	3 (8.1)	4 (11.4)	0.70
Autoimmune disease	0	2 (5.7)	0.23
Chronic liver disease	2 (5.4)	4 (11.4)	0.42
Chronic renal insufficiency	2 (5.6)	4 (11.1)	0.67
Obesity	5 (14.3)	6 (16.2)	0.82
Neurologic disease	2 (5.4)	1 (2.9)	>0.99
Immunosuppression	1 (2.7)	4 (11.4)	0.19
Cancer (patient in active treatment)	5 (13.5)	2 (5.7)	0.43
Days of symptoms, median (IQR)	7 (5–8)	6 (4–8)	0.20
Fever	30 (81.1)	25 (71.4)	0.33
Cough	22 (59.5)	24 (68.6)	0.42
Dyspnea	21 (56.8)	21 (60.0)	0.78
Anosmia	7 (18.9)	6 (17.1)	0.84
Respiratory rate per minute, median (IQR)	18 (15–22)	20 (15–21)	0.51
Oxygen saturation, median (IQR)	93 (92–97)	92 (92–96)	0.65
Bilateral pulmonary infiltrate	30 (81.1)	30 (85.7)	0.59
SOFA ≥2 at randomization	5/33 (15.1)	1/32 (3.1)	0.19
Biomarkers
C reactive protein in mg/L, median (IQR)	55 (28–86)	78 (20–164)	0.29
D-dimer in ng/mL, median (IQR)	510 (404–881)	557 (381–1067)	0.61
Ferritin in ng/mL, median (IQR)	461 (185–735)	472 (167–755)	0.56
IL-6 in pg/mL, median (IQR)	19.7 (6.4–36.0)	19.7 (7.1–34.1)	0.89
Concomitant medication for COVID-19
Steroids	29 (78.4)	26 (74.3)	0.84
Remdesivir	13 (35.1)	12 (34.3)	0.93
Anticoagulants	30 (81.1)	26 (74.3)	0.48

Fig. 2.

Levels of IgM and IgG antibodies (ELISA) against SARS-CoV-2.

Safety data are shown in Table 2. The number of patients with moderate or severe adverse events was similar in both groups; these events are described in the supplementary annex (Tables S1 and S2). No adverse event was considered as related to plasma administration. Efficacy data are also shown in Table 2. The primary endpoint was reached by 3 (8.1%) and 4 (11.4%) patients in the experimental and control arm, respectively. There were no differences in secondary and exploratory outcomes.

Table 2.

Outcomes data. Data are number of patients (proportion) except where specified.

Endpoint	Plasma(n = 37)	Control(n = 35)	Relative risk(95% CI)	P
Grade ≥3^a adverse events
Total number	10	17	NA	NA
Patients	8 (21.6)	8 (22.9)	0.94 (0.39–2.24)	0.89
Primary efficacy endpoint (failure)^b	3 (8.1)	4 (11.4)	0.70 (0.17–2.94)	0.93
Mortality at day 21	2 (5.4)	2 (5.7)	0.94 (0.14–6.35)	>0.99
Mechanical ventilation until day 21	2 (5.4)	2 (5.7)	0.94 (0.14–6.35)	>0.99
Secondary and exploratory efficacy endpoints
Mortality at day 14	1 (2.7)	1 (2.9)	0.47 (0.04–4.98)	0.95
Mortality at day 28	2 (5.4)	3 (8.6)	0.63 (0.11–3.55)	0.94
Use of rescue immunomodulatory drugs^c	3 (8.1)	5 (14.3)	0.56 (0.14–2.19)	0.64
≥1.5-Fold increase in ferritin after day 3	12 (32.4)	7 (20.0)	1.62 (0.72–3.64)	0.23
≥1.5-Fold increase in IL-6 after day 3	9 (24.3)	10 (28.6)	0.85 (0.39–1.84)	0.68
≥1.5-Fold increase in CRP after day 3	11 (29.7)	16 (45.7)	0.65 (0.35–1.20)	0.16
≥1.5-Fold increase in D-dimer afterday 3	5 (13.5)	6 (17.1)	0.78 (0.26–2.35)	0.66
Negative PCR for SARS-CoV-2 at day 7	11/29 (37.9)	9/28 (32.1)	1.18 (0.57–2.40)	0.85
Negative PCR for SARS-CoV-2 at day 21	20/31 (64.5)	17/26 (65.3)	0.98 (0.67–1.44)	0.83
Median hospitalization days (IRQ)	7 (5–11)	8 (6–16)	NA	0.46^d

a

Severe, life-threatening or death.

b

Death or mechanical ventilation anytime until day 21.

c

Tocilizumab, sarilumab, baricitinib, tofacitinib, anakinra.

d

Log-rank test.

Discussion

In this pilot randomized trial, the use of CP plasma after pathogen reduction process (Terumo BCT, Zaventem, Belgium; Mirasol® System) for the treatment of COVID-19 carried out during the pre-vaccine era did not show safety concerns; however, no signal of efficacy was evident compared to SOC alone in reducing mortality or improving clinical outcomes. When comparing administration at <6 days from the start of symptoms vs ≥6 days, no significant differences was shown; however, due to the reduced sample size the results may not be generalizable. Of note, CP was administered after 7 days of symptoms to most of our patients. These findings are consistent with other trials, suggesting that the role of convalescent plasma may be more limited than initially expected. An important aspect of the trial is that it helped implementing a procedure for the collection, transport and administration of convalescent plasma within the healthcare system in a large region.

CP was well-tolerated, with no serious adverse events directly attributed to its use. This aligns with data from other studies suggesting that CP can be safely administered to patients with COVID-19,⁹ particularly when rigorous donor screening and plasma preparation processes are followed. This safety profile makes CP a viable option in settings where no other treatments are available, even if its efficacy remains uncertain. The lack of efficacy is consistent with subsequent larger studies and meta-analyses that reported mixed results regarding the clinical benefits of CP.^10,11 However, whether CP may be beneficial in some subgroups of patients is debatable. In a randomized trial, administration of convalescent plasma during the first 7 days in outpatients with COVID-19 reduced the risk of hospitalization by 54%.¹⁰ A Bayesian analysis of RECOVERY trial suggested a modest chance in reduction of mortality among patients who presented within 7 days of symptoms, and also in seronegative patients.¹² This is important because, despite massive vaccination, some immunodepressed patients have a poorer immune response to vaccines.^13,14

A limitation of CP is related to the variability in antibody titers among donors. In our study, we only included plasma from donors with a high level of antibodies as measured by ELISA; although we could not measure binding antibody activity (BAU) in plasma samples, there are available data from our laboratory suggesting the ELISA breakpoint used (>1/640) as a criterium for accepting donors correlates with sufficient neutralizing power.¹⁵ Interestingly, the study found that patients who received CP had significantly higher levels of IgM antibodies at days 7 and 21 compared to those not receiving CP. Higher IgM levels could theoretically provide short-term protection against viral replication. A retrospective cohort study studied 81 immunosuppressed patients (69% receiving antiCD-20 drugs) with severe COVID-19 and viremia who received CP; although 75% were vaccinated, only 17% had positive SARS-CoV-2 serology at diagnosis. On day 7, 82% had a positive serology and 67% a negative PCR in plasma; achieving positive serology was associated with viremia control¹⁶; however, whether CP has a direct effect on viremia or is associated with improved outcomes in this population would need to be studied in a randomized trial.

One of the notable successes of this trial was the effective implementation of the CP collection and distribution process, confirming other experiences.¹⁷ The rapid mobilization of convalescent plasma resources, including donor recruitment, plasma collection, and distribution to clinical sites, demonstrates the feasibility of deploying CP as a treatment option in a public health emergency. The appropriate coordination within the network permitted for the patients to be identified and selected at their local RAMTCC centre, a key aspect for inclusion during a time where inter-province traveling was restricted. As stated before, the following steps of the trials were conducted in Sevilla for pathogen reduction and Granada for central storage and distribution. This is a further testimony to the very strong collaboration inside the network as it required constant contact and coordination between centres. This infrastructure could be repurposed in future pandemics, especially in the early stages when vaccines and targeted therapies may not be available. The success of the implementation process also highlights the importance of establishing flexible, scalable systems for the collection and distribution of biological therapies in response to emerging infectious diseases. CP, despite its limitations in efficacy, may still serve as an important measure in the absence of other treatments.⁸ This experience underscores the need for robust clinical trial frameworks that can rapidly evaluate the safety and efficacy of potential therapies during future public health crises.

Several limitations should be acknowledged. First, the sample size of this pilot trial was small by definition, thus limiting the statistical power to detect subtle differences in efficacy. Second, variability in antibody titers among plasma donors may have influenced the results, as not all plasma units may have contained sufficient levels of neutralizing antibodies. Another major limitation of the study is the timeframe in which it was carried out, the second half of 2020, where vaccines were not yet available and neither were new antivirals such as nirmatrelvir/ritonavir or molnupiravir. These measures caused a major shift in the impact and therapeutical management of COVID-19 that would limit the applicability of this study. In addition, the timing of plasma administration might have been too late in most patients. It is also worth mentioning that besides the already mentioned differences between the two study groups, some numeric differences (albeit not statistically significant) exist in chronic liver disease or kidney insufficiency.

In summary, the results of this pilot trial suggested that CP is safe for use in COVID-19 patients, although no significant efficacy signal could be observed. The successful implementation of the CP collection and distribution system is an important outcome of this study, offering valuable lessons for future public health responses.

CRediT authorship contribution statement

JRB is responsible for the conceptualization, methodology, formal analysis validation, resources, writing – review and editing, supervision and funding acquisition. SO is responsible for methodology, writing – review and editing, investigation, supervision, resources, funding acquisition, funding acquisition. RM is responsible for methodology, data curation, writing – review and editing, supervision, funding acquisition, project administration. MGC is responsible for formal analysis, data curation, validation and writing original draft. JAML and BQR are responsible for investigation, data curation, writing – review and editing, validation, funding acquisition. GCS is responsible for formal analysis, data curation, validation and writing original draft. MMS, MSG, ARP, IMG, JM, FJMM, PRG and SLC were responsible for investigation, writing – review and editing, validation.

Ethical considerations

The study was approved by the Ethics Committee for Clinical Research of the Virgen Macarena and Virgen del Rocio University Hospitals. The study was conducted according to the principles of Good Clinical Practice (GCP).

Informed consent

The informed consent was obtained from all participants.

Declaration of generative AI and AI-assisted technologies in the writing process

No artificial intelligence was used for analysis or writing of the manuscript.

Funding

Andalusian Network for the Design and Translation of Advanced Therapies (Andalusian Public Foundation for Progress and Health); Andalusian Regional Ministry of Health and Families, grant from 2020 (grant number: COVID-001-01-2020).

Conflict of interest

MAG received payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Pfizer and Merck Sharp Dohme of Spain, S.A. JRB received grant from Andalusian Public Foundation for completion to institution for the present manuscript. PRG received payment for lectures from Menarini, support for congress attending from Gilead and has participated in an Advisory panel for Advanz. ARM received grants or contracts from Gilead Science; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events by Gilead Science and VIIV Healthcare; support for attending meetings and/or travel by Gilead Science, Johnson & Johnson and VIIV Healthcare; Participation on a Data Safety Monitoring Board or Advisory Board for Gilead Science, Johnson & Johnson and VIIV Healthcare. RM, JML, MMS's institution provided partial funding for both the clinical trial and the manuscript's preparation and publication.

Acknowledgements

We are very grateful to all healthcare staff implicated in the care of COVID-19 patients in the participating hospitals who collaborated with the development of the trial during the very hard times of the first months of the pandemic. We really appreciate the staff at Red Andaluza de Medicina Transfusional, Tejidos y Células and Unidad de Producción y Reprogramación Celular de Sevilla for their work in obtaining the plasma used for the treatment of patients.