As of 30/03/2020, 715600 people have been infected with COVID-19 worldwide and 35 500 people
have died, mainly from acute respiratory distress syndrome (ARDS) complicated in 25% of cases
with acute renal failure. No specific pharmacological treatment is available yet. Pulmonary
lesions in these patients are related to both viral infection and an inflammatory reaction.
Patients admitted to intensive care have an important inflammatory response and increased
plasma concentrations of IL2, IL7, IL10, GCSF, IP10, MCP1, MIP1A, and TNFα.

In the blood, the number of peripheral CD4 and CD8 T cells appears to be significantly
reduced, while their status is hyperactivated. This is evidenced by immunoreactive
cytometrics for HLA-DR (CD4 3-47%) and CD38 (CD8 39-4%) or by an increase in the proportion
of highly pro-inflammatory Th 17 CCR6+ lymphocytes. In addition, CD8 T cells would exhibit a
highly cytotoxic profile characterized by high concentrations of cytotoxic granules,
perforin+, granulysin+ or double positive, suggesting associated complement activation.
Because of their immunomodulatory action, which can attenuate the inflammatory response; and
also strengthen the anti-viral defence, it is proposed to evaluate the efficacy and safety of
intravenous immunoglobulin (IGIV) administration in patients developing post-SARS-CoV2 ARDS.

IGIV modifies cell function of dendritic cells, cytokine and chemokine networks and
T-lymphocytes, resulting in the proliferation of regulatory T cells to regulate the activity
of T lymphocytes CD4 or CD8. The action of IGIV induces an activation more particularly of
lymphocytes T regulators that could modulate the effects of the lymphocyte populations
described in the study by Xu et al during COVID-19. In addition, IGIV modulate humoral
acquired immunity, through their effect on the idiotypic network and antibody production.
They also act on innate immunity, through antigen neutralization and modulation of phagocytic
cells. These effects result in a decrease in the production of pro-inflammatory cytokines and
complement activation, key factors in post-SARS-CoV2 ARDS.

IGIV is part of the treatment for a variety of autoimmune and inflammatory diseases. The
standard IGIV as well as polyclonal IGIV significantly reduced mortality in patients with
septic shock and in Kawasaki disease, which is post-viral vasculitis of the child. In
addition, they would not only be beneficial in post-influenza ARDS, but also would also in 3
cases of post-SARS-CoV2 ARDS. IVIG is a treatment option because it is well tolerated,
especially regarding renal function.

These factors are encouraging to quickly conduct a multicentre randomized placebo-controlled
trial testing the benefit of IGIV in post-SARS-CoV2 ARDS.

We hypothesize that the number of days without invasive mechanical ventilation (IMV) is 10
days in the placebo group and 15 days in the experimental group with a standard deviation of
6 days, considering a mortality of 50% and 40% in the placebo and experimental groups
respectively (26, 27). The number of days without IMV in the placebo group is (50% x 10 D) +
(50% x 0 D) or 5 D on average, and following the same calculation for the experimental group
of (60% x 15 D) + (40% x 0 D) or 9 D.

Therefore, a mean value of 5 days without ventilation in the placebo group versus 9 in the
experimental group is assumed, and the 6-day standard deviation is assumed to be stable.
Given the uncertainty regarding the assumption of normality of distributions, the
non-parametric Wilcoxon-Mann-Whitney test (U-test) was used for the estimation of the sample
size. Considering a bilateral alpha risk of 5% and a power of 90% and an effect size of 0.6,
the number of subjects to be included is 138 patients, 69 in each arm.

The primary and secondary analyses will be stratified by age categories, sex and other
clinically relevant factors (comorbidities). Demographic characteristics and parameters
identified at enrolment will be summarized using descriptive statistical methods.

Demographic summaries will include gender, race/ethnicity, and age. For demographic and
categorical background characteristics, a Cochran-Mantel-Haenszel test will be used to
compare treatment groups. For continuous demographic and baseline characteristics, a Wilcoxon
test will be used to compare treatment groups.

The number of days without mechanical ventilation will be presented as a mean with standard
deviation. The groups will be analyzed in terms of intention to treat and the difference
between the two groups will be analyzed by a non-parametric test of comparison of means,
stratified for the primary endpoint. The point estimate of the difference between treatments
and the associated 95% confidence interval will be provided.

A regression model for censored data (Cox model) will explore prognostic factors. The IGIV
immunological and pathological related efficacy endpoints will also be compared according to
their distribution and analyzed using Student, Mann-Whitney and Fisher tests.

Other variables will be presented as means and standard deviations or medians and
interquartile ranges according to their distribution and analyzed by Student, Mann-Whitney
and Fisher tests.

Parameters that are measured on a time scale from randomization or start of administration
will be compared between treatment groups using the Log-Rank test.

The choice of statistical tests and multivariate models (parametric or non-parametric) will
be made for each variable based on observed characteristics (normality of distributions and
residuals, collinearity).

The statistical analyses relating to the main objective will be carried out as intention to
treat. Secondary analyses on the population per protocol may also be carried out.

All tests will be bilateral with a significance threshold of 5%. The software used will be
SPSS v26 (SPSS Inc., Chicago, IL, USA). An interim analysis will be performed after 50
participants are enrolled and another after 100 inclusions.