Thrombotic microangiopathy following systemic AAV administration is dependent on anti-capsid antibodies

BACKGROUND Systemic administration of adeno-associated virus (AAV) can trigger life-threatening inflammatory responses, including thrombotic microangiopathy (TMA), acute kidney injury due to atypical hemolytic uremic syndrome–like complement activation, immune-mediated myocardial inflammation, and hepatic toxicity. METHODS We describe the kinetics of immune activation following systemic AAV serotype 9 (AAV9) administration in 38 individuals following 2 distinct prophylactic immunomodulation regimens. Group 1 received corticosteroids and Group 2 received rituximab plus sirolimus in addition to steroids to prevent anti-AAV antibody formation. RESULTS Group 1 participants had a rapid increase in immunoglobulin M (IgM) and IgG. Increase in D-dimer, decline in platelet count, and complement activation are indicative of TMA. All Group 1 participants demonstrated activation of both classical and alternative complement pathways, as indicated by depleted C4 and elevated soluble C5b-9, Ba, and Bb antigens. Group 2 patients did not have a significant change in IgM or IgG and had minimal complement activation. CONCLUSIONS This study demonstrates that TMA in the setting of AAV gene therapy is antibody dependent (classical pathway) and amplified by the alternative complement pathway. Critical time points and interventions are identified to allow for management of immune-mediated events that impact the safety and efficacy of systemic gene therapy.


Introduction
Adeno-associated virus (AAV) mediated gene therapy has emerged as a promising treatment approach for a spectrum of monogenic genetic disorders (2)(3)(4).With the approval of the first recombinant AAV vector-based systemic gene therapy, onasemnogene-abeparvovecxioi (Zolgensma) for the treatment of spinal muscular atrophy (SMA), the versatility and clinical success of this approach are evident (3).However, while these therapies offer a new approach to treat devastating diseases, the immunologic responses to AAV vectors pose a unique challenge that affects their safety and efficacy (5)(6)(7).The activation of the innate and adaptive immunologic responses after de novo exposure to AAV, as well as humoral and cellular responses to preexisting host immunity, are some of the primary barriers to the expanded use of these therapies (8)(9)(10).
AAV particles are a potent activator of the innate immune system not only through the physical properties of the capsid particle but also through the response to the high capsid protein load associated with systemic AAV administration (11)(12)(13).Pathogen-associated molecular patterns (PAMPs) in AAV are recognized via adaptor proteins Toll-like receptor (TLR), primarily TLR2 and TLR9, which can trigger an innate immune response and promote the activation of adaptive immunity through activation of cytokines and interferons (IFNs) that in turn activate CD8+T cells (8,12,14,15).It is important to note that cytokine production is influenced by the Fc gamma receptors (FcγRs), the Fc region of immunoglobulin G (IgG).
FcγRIIa is the main cytokine-inducing receptor in humans, which influences activation of TLR9, resulting in the production of IFNα as well as other cytokines and chemokines such as TNFα, IL-1β, IL-6, and IL-8 (16).AAV exposure can trigger the complement system, an innate defensive mechanism that induces rapid destruction of pathogens and acts as a functional sensor of the surface area of invading particles (17)(18)(19).The complement system is comprised of more than 30 proteins that play important roles in recognition and elimination of pathogens (20).Complement activation by AAV is primarily antibody dependent (classical pathway), triggered by anti-capsid IgM and IgG antibodies which can cause complement-mediated cell damage (18).However, studies using human samples both in vivo and in vitro demonstrate that complement can be activated by direct interaction of C3 protein and AAV capsid proteins (alternative pathway) (8,11,21).Zaiss et al demonstrated that AAV capsid particles interact with the complement proteins C3, C3b, iC3b and complement regulatory factor H (11,12,22,23).All pathways result in the formation of the C3 convertases (C4b2b), which cleave C3 into C3a and C3b.C3b binds to C4b2b and creates C5 convertase (C4b2b3b).The C5 convertase produces the most potent small peptide mediator of inflammation, C5a, and the large active fragment, C5b, which initiates the late events of complement activation.C5b binds to C6, C7, C8, and C9 to generate the C5b-9 membrane attack complex (MAC) leading to cell lysis and cell death (17,24).Both C5a and MAC can cause acute hepatic and myocardial injury (25).Liver Injury is reflected in the early elevation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) following systemic AAV exposure.Similarly, vector:antibody complexes are deposited on endothelia and cardiomyocytes.The membrane attack complex's mechanism of cell membrane perforation is similar across affected cells, and may cause contemporaneous troponin leakage and transaminase elevation.
Evidence from ongoing clinical trials (NCT03368742, NCT04281485, NCT03882437) suggests that high doses (5 x 10 13 -2 x 10 14 vg/kg) of AAV significantly increase complement activation.Some subjects from these studies presented with severe and life-threatening inflammatory responses likely secondary to the activation of the complement system (12,(26)(27)(28) , (29,30).In addition to nausea, fever and vomiting likely due to cytokine release, subjects presented with complement-mediated thrombotic microangiopathy (CM-TMA) (31), acute kidney injury due to atypical hemolytic uremic syndrome (aHUS)-like complement activation, thrombocytopenia and immune-mediated myocardial injury (28,(31)(32)(33).Activation of complement is a major safety consideration for gene therapy as growing evidence suggests that high-dose intravenous (IV) AAV infusion or high exposure to AAV empty capsids leads to antibody-dependent activation of the complement system in human plasma (34).Additionally, local tissue activation of complement should be considered a consequence of improved AAV capsid specificity based on capsid development and evolution which may deposit higher levels of AAV in targeted tissue beds.
Cases of life-threatening complement activation have been managed with hemodialysis, platelet transfusion and attempts to blunt the complement-mediated adverse events with C5 inhibition via eculizumab, a monoclonal antibody (mAb) that binds to C5 (28).The clinical findings of severe TMA have also been observed in several subjects who developed complement activation following systemic dosing of AAV after administration of Zolgensma® for the treatment of SMA (30,35).The finding of acquired hemophagocytic lymphohistiocytosis (HLH) following Zolgensma® therapy further demonstrates the broad impact of systemic AAV on immune activation (36,37).
Multiple strategies to mitigate the immunologic response to AAV have been evaluated.These strategies include the administration of 1) high-dose glucocorticoids, 2) rituximab, an anti-CD20 mAb that depletes many B cell populations, resulting in impaired antibody production over time; 3) sirolimus, an mTOR inhibitor that assists in the inhibition of T and B cell activation; 4) plasmapheresis and 5) cleavage of all circulating IgG antibodies (11,34,38).Given the complexity of the processes involved in the safety of systemic gene therapy, there is an urgent need to better understand and manage the mechanisms of complement activation following recombinant AAV administration.
In this manuscript, we present a detailed time course and evaluation of complement activation in subjects receiving a single IV infusion of AAV9 mediated gene therapy.In addition, we characterize the complement profile in subjects who received a systemic dose of AAV9 in conjunction with a targeted immune modulation regimen (IMR).When comparing these findings in Group 1 vs Group 2, we confirm our hypothesis that transient B-cell and T cell immunomodulation (Group 2) prevents the most significant innate and adaptive immune responses following systemic high-dose AAV administration.

Results
The comparative analysis of two groups using distinct immune management regimen was used to evaluate the immunological profile after administration of a therapeutic IV dose of AAV9 in 38 subjects (24 male and 14 females, age range 1 week to 11.7 years).The two groups include those who received only conventional corticosteroid dosing (Group 1, n=23) and includes 16 subjects (Group 1A) with spinal muscular atrophy (SMA) treated with Zolgensma® and 7 subjects (Group 1B) with Duchenne muscular dystrophy (DMD) receiving the investigational product from NCT03368742.Group 2 (n=15) received rituximab plus sirolimus as the IMR in addition to corticosteroids to prevent the formation of anti-capsid antibodies; including 2 subjects (Group 2A) with SMA treated with Zolgensma®, 11 subjects (Group 2B) with GM1 gangliosidosis (GM1) receiving the investigational product from NCT03952637 (National Institute of Health, Bethesda, MD), and 2 subjects (Group 2C) with Danon disease receiving the investigational product RP-A501 (NCT03882437).Table I describes demographics and treatment received in each group.
After AAV9 infusion, both IgM and IgG increased rapidly in Group 1 (prednisone or equivalent).Unlike any natural viral infection in which the first antigenic exposure is at the earliest stage of infection, the antigenic exposure to systemic AAV9 dosing is at a peak level within an hour of completing the infusion.Saturating levels of AAV capsids are found in every tissue compartment and certainly in all lymphoid tissue possibly leading to a more rapid adaptive humoral response (Coleman, unpublished).An increase in IgM level was first detected on day 5 post dosing and peaked on day 14.IgG is detectable by day 7 and continued to increase during the first 30 days post-dosing.Group 2 participants received rituximab plus sirolimus IMR and there was no significant change in either IgM or IgG (Figure 1, Table II).The antibody increase in Group 1 coincides with a marked decline in platelet count and clinically significant thrombocytopenia in some subjects.Thrombocytopenia has been observed in many subjects who receive systemic AAV including serotypes other than AAV9.An additional component of the systemic immune response is the increase in fibrin degradation products represented by the finding of elevated D-dimer in plasma.Extensive platelet depletion or increased D-dimer were not observed in Group 2 (Figure 2, Table III).Importantly, Group 1 showed activation of the complement system as demonstrated by reduction of C3 and C4, as well as increased C3a, C4a, C5a, Ba, Bb, and SC5b-9 (Figure 3).The additional activation of Ba and Bb indicates that both classical and alternative complement pathways are involved.In Group 2, there was no detectable activation of most of these complement components and there was limited C5b-9 activation observed; demonstrating that the activation of complement relies principally on the classical pathway and is antibody dependent (Figure 3, Table IV).
Additionally, of the 38 subjects, 22 subjects (Group 1A, n=13; Group 1B, n=7; Group 2A, n=2) had regular peripheral blood smears performed post gene therapy administration.Of those subjects in Group 1A and 1B, 6 presented with few to moderate schistocytes between Day 7 and 21 and 2 were newborns.Schistocytes were not observed in blood smears for any subjects in Group 2A (Figure 4).

Discussion
The data presented here confirms that complement activation is observed in all subjects receiving systemic AAV9 infusions and results in severe inflammatory responses.A previous in vitro study showed that complement activation is observed with all adenovirus serotypes tested including serotype 1, 3, 4, and 9 (39).In addition, Cichon, et al. found that the presence of preexisting anti-AAV appeared to play an important role in triggering the complement system (39).
Two key factors likely contribute to this response and differentiate therapeutic use of AAV from a natural infection.A unique aspect of systemic AAV gene therapy is the substantial total dosage administered within a brief infusion window.A systemic AAV exposure can effectively overpower the adaptive immune response, which normally mitigates the impact of the majority of viral infections in individuals with a functional immune system.Since vector dose is based on the vector genome titer of the drug product, there has been limited opportunity to evaluate the total protein exposure in systemic AAV dosing when there is no specification in the certificate of analysis (COA) for drug product protein concentration.The BCA assay for total capsid protein concentration is a reliable measure that describes the total capsid antigen exposure.The total protein concentration of an AAV preparation is linked to empty:full capsid ratio when the product purity is known and therefore allows for comparisons across various AAV products.A theoretical product of typical purity at a concentration of 2 x 10 13 vg/ml has a capsid protein concentration of 200-300 ug/ml.Therefore, a systemic infusion of such an AAV9 drug product will deliver a minimum of 5 mg of capsid protein in a newborn and up to 100 mg of capsid protein in an adolescent.Such a high exposure is entirely unlike a natural viral infection, even in the setting of viral sepsis.The second distinguishing feature of therapeutic AAV use is that a systemic dose of AAV9 at doses greater than 1 x 10 13 vg/kg will result in exposure to an enormous surface area of the capsid particles (exceeding 20 m 2 , nearly ½ of the pediatric lung surface area).The innate immune response just from the particle surface area alone is a potent stimulus for immune activation.
We have also demonstrated that the AAV capsid may lead to activation of the alternative complement pathway.The complement antigens Ba and Bb are markers of alternative pathway complement activation and were identified in Group 1 post-infusion and remained elevated through Day 7, suggesting direct interaction of C3 and AAV capsid proteins.The alternative pathway can serve as an amplification loop for classical pathway activation.While there is an antibody response to all AAV serotypes that can lead to IgM and IgG mediated activation of the classical pathway, there is a possibility that unique direct interactions of AAV9 with C3 and the potential for greater amplification via the alternative pathway are part of the pharmacodynamics of AAV9.
These findings confirm our hypothesis that transient B-cell and T cell immunomodulation (Group 2) prevents the most significant innate and adaptive immune responses following systemic high-dose AAV administration.Some study limitations should be acknowledged.First, there is an uneven distribution of AAV product exposure between group 1 and 2 due to the doses utilized, and the subjects' age and size in the respective cohorts.Second, while all products are AAV9, the manufacturing processes differ between the various sponsors, resulting in potential differences in product quality including purity profiles and CpG content.
Other AAV serotypes can also result in complement activation.For example, complement activation was observed in adult subjects with Fabry disease receiving a single IV administration of 4D-310 in combination with prophylactic oral corticosteroids (NCT04519749; NCT05629559).The capsid component of 4D-310 is 4D-C102 which is an AAV2 capsid variant developed through Therapeutic Vector Evolution discovery platform, 4D-C012 was generated by inertion of a unique 10 amino acid peptide which is repeated on the capsid surface.A codon-optimized version of the full-length human GLA gene has been developed by 4DMT for the treatment of Fabry disease.Tests for complement activation markers performed at various time points demonstrated that 5 study participants who developed either aHUS or TMA after receiving 4D-310 in combination with prophylactic corticosteroids.Classical pathway complement activation was confirmed by an increase in SC5b-9 levels.The alternative pathway was also activated confirmed by increase in Bb starting approximately 7 days post-4D-310 dosing.Anti-capsid IgM was significantly elevated within the first 5 days in these subjects.Of note, the single participant who had the most severe aHUS was found to have had alternative complement pathway activation at baseline, a novel finding that may predict increased sensitivity to AAV-mediated induction of aHUS and/or TMA.In vitro complement activation assays demonstrated that the 4D-310 capsid (C102) does not directly activate complement in the absence of anti-capsid antibodies.These findings suggest that complement activation after 4D-310 dosing was mainly driven by IgM binding to C1q.This is consistent with the results of a comparative analysis of clinical trials of systemic administration of AAV9 described in this report.

Immunosurveillance post-AAV gene therapy
This report highlights the importance of frequent immunosurveillance during the first 30 days post-AAV dosing including comprehensive complement and hematology panels, D-dimer and other indicators of endothelial activation.In addition, the total anti-AAV Ab levels should be measured before and after administration of AAV gene therapies since measuring only the titer of neutralizing antibodies provides very little information about the total amount of complementactivating antibodies, such as IgM (40).As a result of the detailed kinetic profile of the innate and adaptive immune responses, we have identified the key timepoints for optimal evaluation of rate of change from baseline and early identification of potential serious adverse events; the interval between days 4-10 post-AAV dosing is really the critical period, and that values at day 7 alone may not enable even the most diligent clinician to conclude that the risk of TMA is low.
We have modeled the immunological event timeline in Figure 5. Tracking the clinical status and predictive laboratory indicators will enable investigators and clinicians using future commercial gene therapy products to monitor and anticipate clinical findings which would allow for time to increase observation, either as an outpatient or in hospital.Importantly, we have identified an approach for pre-treatment using rituximab and sirolimus that protects from early IgM and IgG response that is the key trigger to safety events in the first week following systemic gene therapy.Early recognition of the immunological events described will allow for the opportunity to implement other countermeasures, especially for the prevention and management of cardiovascular adverse events.Systemic gene therapy is one of the few medical therapies devised to date, akin to solid organ transplant, that cannot be reversed (41), therefore access to detailed safety data will help establish best practices and improve the safety of this transformative therapy.

Methods
This study was approved by the University of Florida Institutional Review Board.All samples received from the National Institute of Health for complement and cytokine analysis were de-identified and under a Confidentiality Agreement for Coded Biologic Specimens and Data.Investigations before and after gene therapy were performed as follows: All subjects had baseline labs drawn either on the day before to or on the day of infusion prior to any gene therapy administration (Day 0).All subjects received a single IV infusion of an AAV9 gene therapy product.Following the infusion, labs were drawn at days 3, 5, 7, 14, 21, and 30 postgene therapy administration as the minimal data set.Additional visits were scheduled if needed based on clinical condition or laboratory abnormalities.All labs through Week 1 were drawn through a peripherally inserted central catheter that had been placed prior to gene therapy infusion.

Adjunctive immunomodulation to prevent antibody against AAV
Group 2 received prophylactic adjunctive immunomodulation therapy before AAV dosing to prevent antibody development against AAV.Subjects received a total of 1500 mg/M 2 rituximab intravenously (Rituxan), divided in 2 to 4 doses received up to day -1 prior to AAV9 infusion.
Prior to each rituximab dose, the subjects received pre-medication with oral doses of acetaminophen (15 mg/kg), diphenhydramine (1 mg/kg) and methylprednisolone (1 mg/kg).The subjects also received daily oral sirolimus (Rapamune®, 0.5-1 mg/m2/day, adjusted to maintain a trough level of 3-7 ng/mL) starting the day before AAV9 infusion and continuing to Day 180 post-gene therapy administration.

Corticosteroid treatment and additional immunosuppressive drugs
Subjects in Group 1A (SMA) and 2A (SMA) received prophylactic oral prednisone (1 mg/kg/d) one day prior to AAV gene therapy administration and continued for 4 weeks after infusion as per the drug insert with consecutive tapering per clinician's discretion over 4-8 weeks.. Six of the seven subjects in Group 1B (DMD) received prophylactic oral prednisone (1 mg/kg/d) 1 week prior to AAV administration and continued for 12 weeks post with consecutive tapering.The remaining subject in Group 1B received 2 mg/kg/d oral prednisone one day prior to and continued at a lower dose for 12 weeks post-AAV administration.
In Group 2B, three of the eleven subjects did not receive corticosteroids and the remaining eight received 0.5 mg/kg oral prednisone for three days following gene therapy administration.Along with oral glucocorticoids, Groups 1B and 2 received 1 mg/kg IV methylprednisolone at least 2 hours prior to AAV dosing.A summary of corticosteroid dosing and AAV9 dose is included in Table 1.

Laboratory Assays
Complete blood count, kidney and liver function tests, coagulation panel, troponin I levels, C3, C4 and SC5b-9 levels were measured at the timepoints listed above.Any residual serum or plasma that was left over after all clinical labs had been completed was collected for further complement analysis and biobanking.

Antibody Assay
Total anti-AAV9 IgG and IgM levels in serum were evaluated by enzyme-linked immunosorbent assays (ELISA).Serum samples from the subjects were assayed for circulating antibodies to AAV9 capsid.Briefly, 96-well plates were coated with 1.2  10 9 AAV9 particles per well in sodium bicarbonate buffer, pH 8.4, overnight at 4 °C.Subsequently, the plates were washed with a solution containing phosphate-buffered saline (PBS) and 0.05% Tween-20 (PBS-T) and then blocked with 10% fetal bovine serum (FBS; Cellgro) for 2 hours at 37 °C.After being washed with PBS-T, the samples were serially diluted from 1:10 to 1:10,240 with a known positive human standard and allowed to bind overnight at 4 °C.The plates were washed again, followed by addition of a secondary antibody (goat anti-human IgG or IgM conjugated with horseradish peroxidase [HRP]; Invitrogen) at a dilution of 1:20,000 for 2 hours at 37 °C.Finally, the plates were washed and incubated with 3,3′,5,5′-tetramethylbenzidine (TMB) peroxidase substrate (Seracare Life Sciences) in the dark.Reactions were stopped with 0.1M phosphoric acid.The reaction product was measured by spectrophotometric absorbance at 450 nm using Gen5 Microplate Reader and Imager Software (BioTek Instruments).Sample titers were calculated using the mean absorbance of up to three dilutions that were within the linear region of a 4-parameter logistic standard curve generated by a known positive human standard.

Multiplex Complement Assay
Individual complement proteins Ba, Bb, C3a, C4a, C5a, SC5b-9, Factor H, and Factor I were analyzed simultaneously in duplicate in plasma or serum samples using the MicroVue Complement Multiplex -Standard 8-plex (cat no A900, Quidel, San Diego, CA, USA) according to the manufacturer's instructions.Briefly, human serum (1:100 dilution), plasma (1:100 dilution), high and low controls, or assay calibrators were added to microplate wells arrayed with analyte specific antibodies that captured Ba, Bb, C3a, C4a, C5a, and S C5b-9, thereby immobilizing Ba, Bb, C3a, C4a, C5a, and SC5b-9 to their respective locations within the array.The Factor H and Factor I competitive immunoassay reactions occurred simultaneously with the Ba, Bb, C3a, C4a, C5a, and SC5b-9 sandwich immunoassays.The Factor H and Factor I assays used capture antibodies specific for their respective targets.Human serum (1:100 dilution), plasma samples (1:100 dilution), high and low controls, or assay calibrators were added to microplate wells arrayed with immobilized analyte specific antibodies that capture Factor H and Factor I.During the sample incubation, Factor H present in a sample competed with a fixed amount of biotin-labeled Factor H for sites on the immobilized antibody.In the same step, the Factor I present in sample complexed with binding sites found on its respective immobilized antibody.In the subsequent detection step, biotin-labeled Factor I was added and allowed to fill all available Factor I antibody binding sites.Following wash steps to remove excess biotin-labeled Factor I and Factor H and unbound protein, SHRP was added to the microplate.
After an additional wash, the amount of SHRP remaining on each location of the array was inversely proportional to the amount of Factor H and Factor I initially present in a sample.The amount of conjugated enzyme on each location of the array was measured with the addition of a chemiluminescent substrate, read on a Q-View Imager LS (Quansys, Logan, UT, USA) and analyzed using Q-View Software version 3.12.

Statistical Analysis
Data were summarized using descriptive statistics.Percent change from baseline values for parameters were analyzed using mixed-effects models for repeated measures with treatment group, visit, and treatment-group-by-visit interaction as fixed effects; and baseline values were included as a covariate for analyses on percent change from baseline.A compound symmetry within-subject covariance matrix was assumed.SAS v9.4 software (Cary, NC, USA) was used for the modeling.for Group 1 (dashed black lines, full circles) and 2 (full red lines, full triangles).Subjects in Group 1 presented activation of the complement system as demonstrated by reduction of C3 and C4, as well as increased C3a, C4, C5a, Ba, Bb, and SC5b-9.The activation of Ba and Bb indicates that both classical and alternate pathway are involved.On the other hand, Group 2 did not show an activation of the complement system suggesting that IMR as an adjunctive therapy to AAV prevents the activation of the complement system.Data shown as mean ± SEM of Group 1 and Group 2 baseline % change for complement; p>0.05 (non-significant), p≤0.05(*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****).

Figure 2 .
Figure 2. Hematology. Figure shows the percent change from baseline for plateletcount and D-dimer for Group 1 (dashed black lines, full circles) and 2 (full red lines, full triangles).Data suggest that IMR as an adjunctive therapy to AAV used in Group 2 limits the depletion of platelets and the increase of D-dimer after AAV infusion.Data shown as mean ± SEM of Group 1 and Group 2 baseline % change for hematology; p>0.05 (nonsignificant), p≤0.05(*), p≤0.01 (**).

Figure 3 .
Figure3.Complement system markers.Figure shows C3, C4, C3a, C4a, Ba, Bb, C5a and SC5b-9 for Group 1 (dashed black lines, full circles) and 2 (full red lines, full triangles).Subjects in Group 1 presented activation of the complement system as demonstrated by reduction of C3 and C4, as well as increased C3a, C4, C5a, Ba, Bb, and SC5b-9.The activation of Ba and Bb indicates that both classical and alternate pathway are involved.On the other hand, Group 2 did not show an activation of the complement system suggesting that IMR as an adjunctive therapy to AAV prevents the activation of the complement system.Data shown as mean ± SEM of Group 1 and Group 2 baseline % change for complement; p>0.05 (non-significant), p≤0.05(*), p≤0.01 (**), p≤0.001 (***), p≤0.0001 (****).

Figure 4 .
Figure 4. Peripheral Blood Smear.Sequence of peripheral blood smears post gene therapy administration in patient samples revealing increased schistocytes (blue circles) demonstrating evidence of endothelial damage, burr cells, polychromatic red blood cells and, thrombocytopenia suggesting thrombotic microangiopathy.Blood smear of a patient with DMD (group 1B) that received the investigational drug product in NCT03368742.

Figure 5 .
Figure 5. Immunological Event timeline.Events: 1) AAV dosing and capsid biodistribution; 2) Platelet, C3 and C4 depletion; 3) Asparate aminotransferase (AST) and Alanino transferase (ALT) elevation; 4) D-dimer elevation; 5) IgM elevation; 6) SC5b-9, C5a, C3a, C4a elevation; 7) Bb, Ba, Factor I elevation; 8) IgG elevation.The curve for each parameter was created using the average values obtained in the corticosteroid monotherapy group.The curves of each parameter were overlayed on the graph to show the peaks and trough over time creating a timeline of events.The amplitude of each curve was adjusted to fit all the curves in one graph.The ratio of the peaks' amplitude was maintained from the original data.

Table I . Demographics and Treatment Groups
Subjects received investigational drug (NCT03368742) under approved IRB at the University of Florida, Gainesville, Florida c Subjects received investigational drug (NCT03952637) under approved IRB at National Institutes of Health Clinical Center Bethesda, Maryland, United States d Subjects received investigational drug (NCT03882437) under approved IRB at NCT03882437 e Followed by a taper per clinician's discretion b