Immunoglobulins in sepsis and septic shock

Jessica L. Salinas and Alberto Fica C. Department of Medicine, University of Chile Clinical Hospital. Santiago, Chile. Section of Immunology (JSL). Section of Infectious Diseases (AFC). Address for correspondence

Clinical efficacy of polyspecific immunoglobulins or monoclonal antibodies to treat patients with severe sepsis or septic shock is still under debate after several clinical trials. Only a few of them have been able to demonstrate a direct benefit to reduce mortality or this effect appears after meta-analysis. Evidence sustains that polyspecific immunoglobulin G reduces mortality in these patients, being this effect higher for IgMenriched immunoglobulins. Best indications are postsurgical sepsis or early septic shock patients with high titers of endotoxinemia. The use of intravenous immunoglobulins is also recommended for the treatment of patients with streptococcal toxic shock, as demonstrated by the evidence obtained through case-control studies and one randomized clinical trial with a clear trend toward benefit.Evidence does not sustain a favorable impact on mortality for monoclonal antibodies directed against bacterial lypopolysaccaride, other bacterial antigens or against TNF-. Furthermore, infusion of recombinant IL-1 receptor antagonist or soluble receptors for TNF- that could attenuate the inflammatory response have not demonstrated utility after many clinical trials. These therapeutic tools are characterized by a high acquisition cost and adequate cost-effectiveness analysis has not been yet performed. Key words: intravenous immunoglobulin; sepsis; septic shock; lypopolysaccaride; monoclonal antibodies; antiTNF- antibodies; TNF- soluble receptor fusion proteins; Interleukin 1 receptor antagonist. Keywords: intravenous immunoglobulin, sepsis, shock septic, lipolisacridos; anticuerpos monoclonal; anticuerpos 1anti FNT-, fusion proteins of soluble receptor del FNT-; del receptor antagonist interleukina 1 Summary The clinical efficacy of polyspecific immunoglobulin or monoclonal antibodies to treat patients with severe sepsis or shock septic is still controversial developed after numerous clinical trials. Only some of them could demonstrate a direct benefit to reduce mortality or the effect is evident from a meta-analysis. The evidence supports the polyspecific immunoglobulin G reduces mortality in these patients, this effect being greater for IgM-enriched immunoglobulins. Best indications are postoperative sepsis or patients in shock early septic with high titers of endotoxinemia. It is also recommended to indicate IVIG in the treatment of patients with shock streptococcal toxic, evidence established through case-control studies and one randomized clinical trial that showed a clear trend towards benefit. The evidence supports a favorable impact on mortality for monoclonal antibodies directed against bacterial lipopolysaccharide, bacterial or other antigens against TNF-. Furthermore, infusion of

recombinant receptor antagonists of IL-1 or soluble receptors for TNF- that could attenuate the inflammatory response, have not shown utility after many trials. These therapeutic tools are characterized by a high acquisition cost and still not have been cost-effectiveness.

Introduction
The sepsis syndrome is a major cause of morbidity and mortality, with an average mortality of 40%, which increases beyond 60% in patients with shock septic. In this syndrome, systemic inflammatory response is initiated by the release of microbial components, including endotoxin, peptidoglycan, teichoic acid and several exotoxins. These components stimulate an inflammatory response and concerted cascade mediated by cytokines including tumor necrosis factor alpha (TNF-), interleukin-1 (IL1) and interleukin-6 (IL-6). The first two induce hypotension, the combination of them being more powerful than its isolated action. The three cytokines mentioned are increased concentrations in the serum of patients with shock septic 1 . In this response also involved, naturally synthesized proteins by macrophages, such as the receptor antagonist IL-1 (IL-1ra) anti-inflammatory effect. The advent of purified immunoglobulin from a pool of many donors (polyspecific immunoglobulins) and the ability to produce monoclonal antibodies, have allowed new therapeutic tools in patients with sepsis and shock septic, which have been intended to block the inflammatory cascade in its beginning or its development. In recent decades, there have been numerous studies with these agents in patients with sepsis, without obtaining , however, conflicting results make it necessary to review the real virtues of these interventions. This paper analyzes the pharmaceutical alternatives currently available, their mechanisms of action, side effects and administration of these preparations, and finally the evidence of its efficacy when administered for therapeutic purposes. Alternatives available. Four types of intervention have been explored in these patients, purified immunoglobulin, monoclonal antibodies against bacterial structures, monoclonal antibodies against mediators of immune response and cytokine receptor antagonists ( Table 1 ).

According to the type of preparation, these interventions allow the neutralization of bacterial products, modulation of the immune response or the interruption or limitation of the inflammatory cascade ( Table 1 ).

Characteristics and mechanism of action

Polyspecific immunoglobulins Polyspecific immunoglobulins for intravenous use are products of heterogeneous nature, prepared from a pool of 10,000 to 20,000 human donors. They contain more than 90% immunoglobulin G (IgG) in monomeric, polyspecific, with a normal distribution of subclasses, free of aggregates, small amounts of other immunoglobulins and are free of infectious agents 2 . Polyspecific immunoglobulin preparations differ in the predominant type of immunoglobulin are basically classified into two groups. Those containing IgG (IVIG) and those enriched with IgM and IgA also contain (IGMAIV). The risk of HIV transmission, virus hepatitis B or C or other blood-borne agents is virtually nil with current preparations. This due to the strict selection of seronegative donors, and appropriate physical and chemical viral inactivation later: cold ethanol fractionation, filtration, treatment with solvents and / or detergents (caprylic acid and -propionolactona) Photochemical inactivation, precipitation polyethylene glycol, ionexchange chromatography and pasteurization. Mechanism of action . In the case of sepsis and shock septic, its utility is given by two key mechanisms: antimicrobial action and modulation of the inflammatory response 3 ( Table 2 ).

Commercial IVIG preparations exert a direct antimicrobial action through opsonizing and neutralizing antibodies against bacteria, endotoxins or exotoxins of specific microorganisms, which can enhance your debugging liver or spleen. These include antibodies against Campylobacter jejuni , Yersinia enterocolitica , Staphylococcus aureus (including methicillin-resistant), Enterococcus faecalis (vancomycin-resistant too), Streptococcus hemolytic, Pseudomonas aeruginosa and antibodies to the lipopolysaccharide (LPS), alpha toxin from S. aureus and some superantigens of Streptococcus pyogenes. The antimicrobial action is exerted also by stimulating the production of reactive oxygen species in leukocytes and serum bactericidal activity ( Table 2 ). Mechanisms of modulation of the inflammatory response proposed basically contemplate interfering with the effect of cytokines, modulating the activity of B and T lymphocytes (CD4 + and CD8 +), and prevention of excessive complement activation 4 . The modulation allows a decrease in pro-inflammatory cytokines such as IL-6, IL-1 and TNF- , and increased receptor antagonist IL-1 (IL-1ra), favoring an antiinflammatory state. It has also demonstrated the presence in IVIG preparations, of anti-cytokines such as IL-1, IL-6, TNF- and interferons 4 , 5 . IgM-enriched preparations have a greater inhibition of excessive complement activation compared to IVIG and have been associated in some studies to increased clinical efficacy (see below). Monoclonal Antibodies These immunoglobulins have been generally prepared with a specificity directed to endotoxin, other bacterial compounds or to cytokines involved in the inflammatory response. In the first case the antibodies are E5 and HA-1A directed against the lipid A of LPS, a structure common to gram-negative bacilli clearly involved in the phenomenon of systemic inflammatory response. Have also been designed antibodies directed against antigens common to enterobacteria 6 . In the case of cytokines, these have been directed against tumor necrosis factor (anti-TNF-alpha: infliximab) in July ( Table 3 ).

Cytokine antagonists Sepsis interventions also involve recombinant molecules antagonists of the effect of specific cytokines. One is the natural receptor antagonist IL-1, synthesized during natural infection (IL-1ra: IL-1 "receptor antagonist") and to limit the inflammatory response 1 , 8 . Other molecules include soluble receptors of TNF- (eg etarnecept) also to counteract the effect of this cytokine and limit the immune response. Soluble TNF receptors are released from the cell membrane by proteolysis during infection and to counteract the TNF- . At least two of these receptors with different molecular weights have been described and both have been synthesized as recombinant molecules attached to a human IgG Fc fraction of 9 - 12 ( Table 4 ).

Risks and adverse effects of polyspecific immunoglobulins, monoclonal antibodies and cytokine antagonists
Adverse reactions to these agents are uncommon and rarely severe, allowing these products are considered safe drugs. Adverse reactions associated with polyspecific immunoglobulins . It has been estimated that adverse reactions to these preparations are of very low magnitude and severe or fatal reactions are unlikely. Complications of IVIG therapy can be classified as non-anaphylactic and anaphylactoid. The former include chills, headache, backache, malaise, fever, rash, nausea, paresthesia, hypo-or hypertension and volume overload. Most are related to the infusion rate, occur within the first 30 minutes of administration and are managed by temporarily withholding the infusion, slowing or pre- medication. Rarer complications are aseptic meningitis and acute renal failure. Anaphylactic reactions can be seen in patients with an absolute deficit of IgA, in which administration of IVIG with high levels of IgA can induce the synthesis of specific IgE anti-IgA. In these cases, the reactions do not occur with the first dose, although they can reach the shock anaphylactic with subsequent administrations. For the treatment of these patients should be used commercially prepared with the least possible amount of IgA 2 . Adverse effects of monoclonal antibodies . Available trials, adverse effects observed with these compounds have been infrequent and generally not higher than those observed with placebo. The frequency of these events does not exceed 2 to 3% and include allergic rash, serum sickness, hypotension and occasionally shock anaphylactic

13 - 15

. Anaphylactic reactions do not easily reversed upon discontinuation of therapy but may require anti-inflammatory drugs or corticosteroids. Most patients (90%) received murine monoclonal antibody preparation developed during follow-up 16 . It is unknown whether it can generate a higher frequency of adverse effects during further exposure. Because of the involvement of TNF- in the immune response against intracellular pathogens such as Listeria sp, mycobacteria and certain fungi, blocking TNF- may promote the development or reactivation of infection by intracellular pathogens. In patients treated with molecules designed to this end, either with infliximab or etarnecept not observed an increased frequency of these events in patients with sepsis compared to that observed in the placebo group 10 , 13 . However, this risk is described in patients with Crohn's disease or rheumatoid arthritis undergoing therapy with these antibodies, which have seen an increase in cases of tuberculosis after initiation of therapy, 17 , 18 . The long half-life of infliximab (10 days) and its biological effect (2 months) explains this risk. Adverse effects of cytokine antagonists . The recombinant molecules (eg IL-1ra), are similar to natural molecules and therefore have not been associated with significant adverse effects on 1 , 8 . The fusion molecule p55-IgG has not been associated with an increased frequency of adverse events compared to placebo 9 . There are no reported cases of anaphylaxis, urticaria or bronchospasm prepared for these events although rash 9 . As mentioned above, has not reported an increased frequency of infections by intracellular pathogens in patients treated with the fusion molecule p55-IgG 10 .

Dosing schedules
It has been shown that levels of IgG and IgM in patients with sepsis are severely depressed. The doses used in different trials have been designed to allow a return of their concentrations to physiological values for several days. In Table 5 presents the usual doses used for therapeutic purposes to the currently accepted conditions

The doses of monoclonal antibodies and recombinant molecules have been derived from the results of animal studies and qualifications developed in human trials. Only two immunoglobulin products are currently available in Chile (Pentaglobin Flebogamma and ). Other products have been discontinued (Sandoglobulin , Octagam ). The acquisition cost is lower for the preparation enriched in IgM ( Table 5 ). Differences in activity between preparations . The current standardization fractionation techniques and obtaining IGIV preparations, allowing all possess a similar activity, with a distribution of IgG subclasses similar to that of plasma and free of contaminants, differing in the amounts of IgA in preparations . Only IgM-enriched preparations have a higher content of specific antibodies against a number of pathogens.

Evaluation of clinical efficacy in sepsis and shock septic

Several randomized trials and two meta-analysis have been developed to assess the potential usefulness of preparations polyspecific immunoglobulin (IVIG or IGMAIV), monoclonal antibodies and cytokine antagonists in morbidity or mortality (global or specific) of patients with sepsis or shock septic 19 , 20 . Impact on overall mortality . Most evaluation of immunoglobulins has been reported by Alexandria et al , in a meta-analysis 27 randomized trials appropriate groups, including 11 IVIG or IGMAIV, 8 with antibodies anti-endotoxins, and 8 with anti-cytokine 19 . Most of the works included were performed in adult patients, 4 babies and only one in children. In the subgroup of 11 trials with polyclonal IVIG (6 adults and 5 kids), this meta-analysis failed to demonstrate a reduction in the risk of death with a relative risk (RR) of 0.64 (IC 95 0.51 -0.80), a trend that continued to consider only the best quality work in design (RR 0.56 CI 95 0.40 to 0.79) 19 ( Figure 1 ).

Figure 1. Relative risk (RR) of 28-day mortality in sepsis and shock septic different antibody-based immune interventions recombinant

molecules or cytokine antagonists. Results of a meta-analysis and other studies. From left to right are the values of RR associated with polyspecific immunoglobulin (IgG or IgG-MA), then the values associated with monoclonal antibodies against bacterial structures (E5, HA-1A and enterobacterial common antigen [CEA]), then the associated with antiTNF- and antagonists of cytokines (IL-1 ra and IgG Fc fusion proteins with some of the soluble receptors of TNF- ). The work of Tugrut et al corresponds to the reference 20 . The letters A and B show March 2 consecutive papers published for p55-IgG (refs. 9 and 10 ). The abbreviation MA indicates the RR obtained in the meta-analysis of Alexandria et al, (ref. 19 ). Dotted line indicates a RR equal to 1 ruling out a significant protective effect. Not included in the figure the results of the work SBITS Werder et al (ref. 3 ), the results of the work of Pittet et al with fusion proteins as soluble receptors for TNF- (ref. 12 ) or those associated with the use immunoglobulin in the syndrome of shock toxic. Polyspecific For preparations, the impact on mortality was greater with IgM-enriched preparations compared to those containing IgG with an RR of 0.48 (IC 95 0.30 to 0.76) vs 0.73 (IC 95 0.57 to 0.93) ( Figure 1 ). When considering all types of immunoglobulins (polyspecific, anti-endotoxins or cytokine) showed a significant impact on mortality in adult patients (RR 0.62, with an IC 95 from 0.49 to 0.79) but not in neonates with sepsis, although the latter group only works with polyspecific immunoglobulins. (RR 0.70, with an IC 95 from 0.42 to 1.18). Subsequently, a meta-analysis Olhsson et al 21 in patients with immunoglobulin poliespecficas newborns, demonstrate success in Seven Trabajo selected that it had one overall mortality significantly reduced this cuando fue compuesto used in cases with documented infection (RR 0.55 IC 95 0 0.31-0, 98) but only one reduction of borderline significance cuando fue la immunoglobulin used in a scenario sospe cha without confirmation of sepsis after 21 . Las differences observed between the two meta-analysis pueden reside in it with number of nests included Trabajo (5 in Alexandria he et al 19 and 7 with assessment of mortality in he Olhsson et al 21 ) and only 4 communes Trabajo. For example, lets him in Alexandria Group et al, dismissed from the aquellos included him otro group due to deficient su calidad methodological. Others are concerned the possible inclusion of Tests with different immunoglobulin (poliespecficas, the antiendotoxinas anticitoquinas) in it meta-analysis of adult patients but only in immunoglobulin poliespecficas he meta-analysis of pediatric patients. A recent study developed with polyspecific IVIG and not included in the meta-analysis of Alexandria et al failed to show a better effect than placebo (study SBITS) 3 and contained the number of patients (n = 653) may noticeably affect the results metaanalysis by least significant utility of these agents in adults with sepsis 3 . Similarly, a recent study IGMAIV (not included in the meta-analysis) also demonstrated a favorable impact on a small number of patients with severe sepsis 20 . These new facts weaken the conclusions reached in the meta-analysis of Alexandria et al, on behalf of polyspecific IVIG enriched or not with IgM. ( Figure 1 ). The results with monoclonal antibodies are less clear as the results are directly observed in the original studies, as appreciated in the meta-analysis of Alexandria et al

19

. In fact, failed to demonstrate a favorable impact analyzed separately or jointly studies with monoclonal antibody E5 (n = 4, RR 0.99 with CI 95 0.88 to 1.10) and only one four studies with the antibody HA-1A had a favorable impact on mortality of these patients 19 . Similarly the quality of individual studies separately or bundled with the meta-analysis of Alexandria et al, have failed to demonstrate a favorable impact of anti-cytokine monoclonal antibodies versus placebo 19 . Subsequently published trials with a monoclonal antibody directed toward a common antigen of Enterobacteriaceae (no LPS), also demonstrated the utility of this intervention in patients with gram-negative sepsis 6 . Similarly, four studies conducted with the fusion protein p55-IgG or p80-IgG (soluble receptors for TNF- act as antagonists) and have not been included in metaanalysis have failed to demonstrate favorable effect on mortality overall ity in patients with sepsis or shock septic 9 - 12 ( Figure 1 ). Impact on mortality of sepsis . Only a few studies have directly assessed this issue. In the meta-analysis of Alexandria et al, three studies with IVIG reported a favorable overall effect with a RR of 0.35 (IC 95 0.18 to 0.69). Two of these papers were developed in adults and have served to sustain the usefulness of these preparations in spite of other studies that show no benefit 3 , 22 , 23 . These trials demonstrate the usefulness of a polyspecific immunoglobulin IgG in patients with postoperative sepsis, which meet certain criteria of severity or in patients with shock septic-onset ( 24 hours) with high titers of endotoxinemia ( 12.5 pg / mL) ( Table 6 ). In severe postoperative sepsis has shown a benefit in patients with a high score ( "sepsis score" 20 over a maximum of 24) associated with high mortality ( 90%) 22 , 24 . Unfortunately, measuring endotoxinemia is not readily available and requires standardization. Moreover, the scoring system "sepsis score" is hard to find, has not been universally adopted and has not been correlated with other more widely used. These issues limit the clinical application of these criteria. In the meta-analysis by Ohlsson et al did not evaluate specific mortality in sepsis 21 . Impact on hospitalization and other parameters . In the works of better methodological quality has not been able to demonstrate a favorable effect on the duration of hospitalization in adults or neonates 19 , 21 . Immunoglobulins in toxic shock syndrome. This condition is characterized by hypotension, failure of at least two organ systems and the presence of S. pyogenes in blood or normally sterile fluids. Of occasional occurrence, it has been historically associated with necrotizing fasceitis. It has a high mortality and its sporadic occurrence has hindered the development of controlled studies. Different meta-analysis discussed in general have not included such patients as sporadic appearance and therefore deserve separate analysis. Case series, anecdotal reports and a study against historical controls may have noticed the polyspecific IVIG therapeutic benefit in the recovery of these patients. Several clinical series have reported clinical improvement with its use in patients with invasive infections and have included cases with shock toxic or necrotizing myositis fasceitis 25 . The most important evidence supporting the use of IVIG in this serious condition comes from two studies published in recent years. The first published in 1999, casecontrol, failed to show an increase in survival from approximately 33 to 66% IVIG, is

the treatment with this preparation a significant independent factor for survival of these patients 26 . The second study, a randomized double-blind study published in 2003 showed a noticeable but not significant reduction in mortality and morbidity in patients with syndrome of shock toxic when treated with IVIG 27 . This trial failed, however, recruit a sufficient number of patients after more than two years for the sporadic occurrence of 27 . It is likely that a job with a sufficient number of shows ever possible and why many experts believe that this condition is already sufficient evidence to support its use and you can not raise new trials for ethical reasons 27 . In Chile, it has progressively expanded the use in this condition and are perhaps the most accepted indication 28 ( Table 6 ).

Patients with shock streptococcal toxic, have been evaluated with non-enriched preparation (IVIG) 26.27 and there is no known effective IGMAIV them. Background: the clinical efficacy studies . It has so far been difficult to demonstrate a favorable clinical impact of immunoglobulins in the three parameters analyzed: overall mortality, mortality from sepsis and length of hospitalization. Evidence for the benefit of this therapy on overall mortality of adult patients with sepsis appears to group together to discuss the preparations or adult patients or infants who receive IGIV and especially IGMAIV. This benefit disappears when analyzing anti-cytokine and antiendotoxin or discordant for neonatal patients are obtained by different meta-analysis. In the case of sepsis-specific mortality in adults there is evidence favorable but limited to a small number of patients so far, restricted to patients with severe postoperative sepsis. There is support for using these preparations as a way to shorten hospital stay. Although imperfect, the evidence for polyspecific immunoglobulin use in patients with syndrome of shock toxic, appears consistent and beneficial in analyzing clinical series, case-control studies and one randomized trial available.