Clin Genet 2014 Printed in Singapore. All rights reserved
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: 10.1111/cge.12368
Review
The beginning of personalized medicine in sepsis: small steps to a bright future Christaki E., Giamarellos-Bourboulis E.J. The beginning of personalized medicine in sepsis: small steps to a bright future. Clin Genet 2014. © John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2014 There is a growing recognition that there is a need for a more personalized approach towards sepsis care. In most clinical trials investigating novel therapeutic interventions against sepsis, patients have been considered a rather homogeneous population. However, there is probably more individual variability between septic patients than previously considered. The pathophysiology of sepsis is a complex and dynamic process that originates from the host immune response to infection and varies according to the genetic predisposition, immune status and co-morbid conditions of the host, the type of pathogen and the site and extent of infection. Until now, efforts to stratify septic patients according to their immune profile were hampered by the lack of specific biomarkers. Recent advances in molecular medicine have made it possible to develop tools that will facilitate a faster and more precise diagnosis of infection. Individual variability between each patient’s responses to infection can assist in tailoring therapeutic interventions to the individual’s disease profile and monitoring treatment response. In this review, we describe those recent advances in genomics and theragnostics, which are slowly entering clinical practice and which will make possible a more personalized approach to each septic patient in the next decade.
E. Christakia,b and E.J. Giamarellos-Bourboulisc,d a First Department of Internal Medicine, AHEPA University Hospital, Thessaloniki, Greece, b Infectious Diseases Division, Alpert School of Medicine of Brown University, Providence, RI, USA, c 4th Department of Internal Medicine, University of Athens Medical School, Athens, Greece, and d Center for Sepsis Control and Care, Jena University Hospital, Jena, Germany
Key words: genomics – individualized medicine – metabolomics – personalized medicine – sepsis – septic shock – single nucleotide polymorphism – transcriptomics Corresponding author: Eirini Christaki, MD, PhD, First Department of Internal Medicine, AHEPA University Hospital, St. Kiriakidi 1, Thessaloniki, Greece. Tel: +30 6973579927; fax: +30 2310993271; email: [email protected]
Conflict of interest
The authors have no conflict of interest to declare.
The decline of severe sepsis mortality that has been noted during the last 20 years is encouraging, given the absence of a specific diagnostic test and a targeted therapeutic intervention. Although improvements in sepsis management such as the use of early goal-directed therapy, timely administration of antibiotics and optimal mechanical ventilation strategies have probably contributed to the decline, mortality from severe sepsis remains high (1). There is a growing recognition that there is a need for a more personalized approach towards the septic patient. Sepsis pathophysiology is traditionally considered to stem from an overwhelming host response to microbial stimuli. Microorganisms bear highly evolutionary conserved molecules known as pathogen-associated molecular patterns (PAMPs) that interact with pattern recognition receptors (PRRs) of the host innate immune cells. This interaction generates a storm of proinflammatory and anti-inflammatory
Received 21 January 2014, revised and accepted for publication 24 February 2014
mediators that orchestrate septic phenomena (2). This concept of sepsis pathogenesis has led the conduct of several randomized clinical trials during the last two decades. Most of these trials studied the efficacy of compounds that block inflammatory mediators, however, most of them failed to show a survival benefit (3). Part of an explanation for these failures may rely on patient heterogeneity. Despite the use of common clinical characteristics of sepsis as inclusion criteria in clinical trials, enrolled patients can be very heterogeneous in terms of their underlying source of infection, their predisposition and their co-morbidities. Moreover, studies from the Hellenic Sepsis Study Group (HSSG) have shown that the innate and adaptive immune responses in septic patients are not ubiquitous and differ substantially based on the underlying type of infection and individual host responses, making sepsis a ‘prototype’ for personalized disease (4, 5).
1
Christaki and Giamarellos-Bourboulis Genetic predisposition
Genetic diversity is one of the factors that account for the individual variability seen between septic patients and is defined as the carriage of single nucleotide polymorphisms (SNPs) of genes that encode for the protein molecules of inflammatory mediators and of PRRs. Many studies have been published investigating the impact of carriage of one (heterozygotes) or two (homozygotes) SNP alleles on susceptibility to a given infection, however less is known on the role of these SNPs in the development of sepsis and their effect on outcome. It should be noted that a genotype that has a protective role in prevention of infection could prove deleterious if systemic infection occurs. For example, a number of studies suggest that increased tumour necrosis factor (TNF) response decreases the risk of infection but increases the risk of death from septic shock should it occur. Hence, the sequelae of an altered host response may have competing effects in population-based studies (6). The most broadly studied SNPs implicated in sepsis pathophysiology are those of TNF, MIF, PAI-1 and of genes encoding for toll-like receptors (TLR). The most widely studied SNP of TNF involves position −308 of the promoter although SNPs at positions −376 and −238 have also been investigated. All these SNPs involve a substitution of guanine (G) by adenine (A). Existing studies are characterized by great heterogeneity. Two studies in 280 and 93 patients, respectively, failed to disclose a role of carriage of A alleles at the position −308 for the development of community-acquired pneumonia (CAP; 7, 8). However, three studies in surgical patients showed that carriers of SNPs were more prone to the development of severe sepsis either post-operatively or after major trauma (9–11). Analysis of all three SNPs in 213 mechanically ventilated patients indicated that carriage of at least one SNP allele is associated with earlier development of ventilator-associated pneumonia (VAP) and sepsis (12). In a recent study of 1057 critically ill patients of Caucasian origin, mortality was 48.7% for AA homozygotes for the TNF −308 SNP compared with 30.5% in the GA heterozygotes and 29.5% in the GG homozygotes for the wild-type allele (13). The shear value of studying TNF is the location of the gene being in close proximity with the major histocompatibility complex. In a study of 490 paediatric septic patients, haplotype analysis was conducted. Haplotypes constituted of SNPs at position +252 of LTA encoding for lymphotoxin A and of SNPs at positions −863 and −308 of TNF. Results revealed that haplotype +252G LTA, −863C TNF and −308A TNF was protective from the development of acute respiratory distress syndrome (ARDS), independently of patient age, gender, race and co-morbidities (14). TLR1 encodes for TLR1, a PRR that primarily recognizes gram-positive cocci. The most widely described SNPs are one adenine to guanine transition at position −7202 (rs5743551), one adenine to guanine transition at position +742 (rs4833095), and one guanine to thymidine transversion at position +1804 (rs5743612).
2
In a cohort of 1498 patients with major trauma, patients homozygous for the minor frequency G allele at position −7202 and for the minor frequency T allele at position +1804 had greater risk for death from sepsis (adjusted odds ratios: 3.16 and 2.48, respectively). These patients were also at greater risk for death when sepsis was due to gram-positive cocci (15). Another study showed that the −7202GG and the +742GG genotypes were associated with a greater risk for septic shock. These patients had also greater circulating levels of C-reactive protein (CRP) and lower levels of interleukin (IL)-10 than the rest of the cohort (16). Results on the impact of TLR1 SNPs on mortality were fully confirmed by an independent cohort of 493 patients with septic shock (17). The adaptor protein myeloid differentiation protein-88 adaptor-like (Mal), otherwise known as TIR domain-containing adaptor protein (TIRAP) is necessary for TLR2 and TLR4 downstream signalling and thus plays an important role in the innate immune response. In a case–control study of 6106 individuals from the UK, Vietnam and several African countries 33 TIRAP SNPs were genotyped. It was found that heterozygous carriage of TIRAPS180L polymorphism was associated with protection against invasive pneumococcal disease, bacteraemia, malaria and tuberculosis in the different study populations (18). Carriage of one SNP leading to the substitution of asparagine with glycine at position +299 of TLR4 leads to enhanced production of TNFα and to attenuated production of IL-10 by monocytes. This SNP is mainly found across African populations providing protection against mortality from malaria but has also been associated with increased mortality from septic shock (6). MIF encodes for migration inhibition factor (MIF) that antagonizes the function of glucocorticosteroids. Two SNPs are known: one at position −173 and one microsatellite region at position −794. Two independent studies have showed a major role of the CC genotype at the −173 position on sepsis outcome. The first study enrolled 169 patients with severe sepsis of Caucasian origin and showed that patients with the GG genotype had increased risk of mortality mainly after infection of non-pulmonary and of non-abdominal origin (odds ratio: 3.818, p = 0.0066; 19). This association of the GG genotype with unfavourable outcome was also confirmed in a cohort of 1739 patients with CAP (20). PAI-1 encodes for plasminogen activator inhibitor-1. One major 4G/5G SNP has been recognized at position −675. The association of this SNP with the development and outcome of sepsis has been studied with contradictory results. A recent meta-analysis of 12 case–control studies and three cohort studies concluded that SNP carriage increased susceptibility for the development of sepsis and for unfavourable outcome (21). A recent randomized trial by Morelli et al. showed that patients with septic shock who received intravenously the short-acting β1-adrenoreceptor blocker, esmolol, had lower 28-day mortality compared with the control group (49.4% vs 80.5%, p < 0.001; 22). The β2-adrenergic receptor gene (ADRB2) is involved in the regulation of responses to adrenergic agonists in cardiovascular
The beginning of personalized medicine in sepsis and inflammatory diseases. In a study performed in two centres, patients with septic shock who carried the AA genotype of ADRB2 rs1042717 G/A polymorphism (CysGlyGln haplotype) exhibited increased heart rate, worse mortality and more organ dysfunction. Interestingly, in patients treated acutely with low dose corticosteroids for septic shock, the differences in mortality conferred by the ADRB2 genotype were eliminated (23). Diagnosis and monitoring – biomarkers
The host immune response in sepsis varies between patients and in the same patient over time (2). Moreover, the two-stage paradigm of sepsis, where an initial hyperactivation of the innate immune response of variable duration is followed by a hypoinflammatory phase, has been challenged by recent genome-wide expression studies that revealed inconsistency in the observed inflammatory pathway changes in early and late sepsis (24). Studies performed by the HSSG have shown that immune response kinetics differ according to the underlying type of infection, therefore the strategy of using the same biomarker in all sepsis cases may not be optimal (4, 5). Hence, monitoring the inflammatory response during the course of sepsis is critical to stratify patients into subgroups according to their predisposition and functional immune status. The optimal immunomodulatory treatment can then be given and their response to therapy can be accurately assessed. Cytokines, soluble receptors, cell-surface molecules, coagulation, endothelial markers and vasoactive hormones have all been evaluated for their diagnostic and prognostic utility in sepsis. Monocyte cell-surface HLA-DR expression is decreased in septic patients and it is used as marker for immunossuppresion (25). However, it is becoming evident that a single biomarker cannot successfully discriminate sepsis from non-infectious inflammatory conditions or reflect the complex pathophysiologic processes that occur in sepsis and combinations of biomarkers are being increasingly investigated in clinical trials (26). Gene expression assays have also been developed in an effort to map those genes that regulate the innate and adaptive response and are either upregulated or downregulated during the course of sepsis (27). Using gene microarray analysis in whole blood, Johnson et al. showed that infectious systemic inflammatory response syndrome (SIRS) had a different gene expression profile from non-infectious SIRS even before sepsis became clinically apparent (28). Other research groups have identified white blood cell gene signatures that could also differentiate SIRS from sepsis (27, 29). In a paediatric study, a 100-gene expression signature (genes corresponding to adaptive immunity, glucocorticoid receptor signalling, and the peroxisome proliferator-activated receptor-α signalling pathway) was validated for the allocation of children with septic shock into three clinically relevant subclasses (30). Continuous progress in proteomics and bioinformatics has allowed the identification in the blood
or urine of proteins differentially expressed in sepsis and their assessment as potential diagnostic or prognostic biomarkers (31, 32). Proteomic analysis determined novel biomarkers that were then used in a prospective study of neonatal sepsis to successfully risk stratify neonates and guide antibiotic de-escalation (33). Also, in a recent study, using liquid chromatography–tandem mass spectrometry (LC-MS/MS) to perform metabolomic analysis of 186 metabolites, the investigators were able to define two markers, acylcarnitines and the glycerophospholipids, which could help discriminate sepsis from non-infectious SIRS (34). However, gene expression profiling has certain limitations such as inter and intraindividual variation and differences in gene expression between target organs, which will need to be addressed before transcriptome analysis can be interpreted and applied in clinical practice as a surrogate marker for sepsis (35). Although genetic signatures provide a snapshot of an individual’s gene expression profile at any given point during the septic process, they cannot characterize all the factors that regulate an evolving inflammatory process or yet predict the response to a given intervention. Systems biology and systems immunology integrate genomic, proteomic and metabolomic data with the use of advanced computational methods to study the molecular and cellular networks during infection and can provide comprehensive and possibly predictive models of host–pathogen interactions (36, 37). Treatment – future perspectives
In order to apply a personalized approach in sepsis management, we need to be able to map the complex host–pathogen interactions that occur in each septic patient and to translate this knowledge into individualized therapy to improve sepsis-related outcomes (Fig. 1). So far, some components of the septic process have been identified where we can intervene and modify outcomes. Early diagnosis
It is widely recognized that antimicrobial therapy improves sepsis outcomes when administered within the first hour of diagnosis (38). However, limitations of conventional diagnostic methods such as the blood culture, often lead to a delay in initiating appropriate antimicrobial treatment and promote widespread use of unnecessary combinations of broad-spectrum antibiotics. Advances in automated molecular and proteomic methods in microbiology have made possible the early identification and quantification of specific pathogens in blood and other biologic specimens, however, some of these methods still need to be validated in clinical trials. Genetic identity
There is no doubt that in sepsis, certain SNPs of host genes are associated with unfavourable prognosis. Early
3
Christaki and Giamarellos-Bourboulis Host Gene expression
(activation of pro-inflammatory vs anti-inflammatory pathways)
Host DNA (SNPs of TNF, MIF, TLR, PAI-1)
Microorganism
(molecular detection of species and antimicrobial susceptibilities)
Cell secreted and circulating proteins (dynamic changes advise of treatment efficacy)
SEPSIS OUTCOME Fig. 1. Schematic representation of the host and pathogen factors that contribute to outcome in sepsis in a given patient. MIF: migration inhibition factor, PAI-1: plasminogen activator inhibitor-1, SNP: single nucleotide polymorphism, TNF: tumour necrosis factor, TLR: Toll-like receptor.
genotyping of septic patients could enable clinicians to triage those with increased risk for death to early aggressive intensive care support and to identify those patients who will benefit from immunomodulatory interventions based on their immune kinetics. It has been shown that when signs of sepsis develop, monocytes of carriers of A SNP alleles of TNF produce greater amounts of proinflammatory cytokines (12), whereas carriers with chronic inflammatory disorders fail to respond favourably to anti-TNF treatment (39). Gene profiling
There are several available agents that can modulate the host response. Some of them attenuate proinflammatory responses such as antibodies targeting TNFα and the soluble receptor antagonist of IL-1β, whereas others stimulate proinflammatory responses such as interferon-gamma (IFNγ). It is evident that selection of the appropriate immunomodulatory agent necessitates knowledge of the patient’s exact immune response status. This could be achieved with the use of transcriptomic and proteomic profiling in order to discriminate which proinflammatory and anti-inflammatory pathways are up or downregulated. Biomarker-guided therapy
Protein biomarkers are most commonly used to monitor clinical response and guide appropriate therapy. Results of a randomized clinical study from our group, in 200 patients with sepsis due to VAP, indicate that signs of clinical severity accompanied by an increased ratio of IL-10/TNFα, reduced production of IL-6 by circulating monocytes and decreased expression of CD86 on circulating monocytes can serve as a marker for the initiation of adjunctive clarithromycin (40). A prospective study
4
by Döcke et al. (41) also showed that IFN-γ 100 μg/day in patients with sepsis and HLA-DR expression on monocytes less than 30%, reverses immunoparalysis of monocytes. Similarly, in another randomized trial it was found that administration of granulocyte-macrophage colony stimulating factor (GM-CSF) for 8 days restored monocytic immunocompetence and was associated with a shorter time of mechanical ventilation and shorter hospital stay (42). Administration of dexamethasone as adjunctive therapy in CAP decreased the time to pneumonia resolution by 1 day. However, most of the mortality benefit was shown only in the subgroup of patients who presented with increased levels of IL-6, IL-8 and MCP-1 and decreased cortisol levels (43). One biomarker widely used to guide de-escalation or modification of antimicrobial treatment in the critical care setting is procalcitonin (PCT) whose usage can decrease unnecessary antibiotic administration, albeit without a clear effect on outcome (44). Clinical guidance
It is true that a personalized treatment in sepsis, guided by molecular profiling and protein biomarkers, is not feasible yet due to the lack of standardization and randomized clinical trials. However, when used selectively after application of certain clinical criteria, two interventions have shown favourable results. First, current guidelines suggest the intravenous administration of 50 mg hydrocortisone every 6 hours for 7 days in patients with refractory shock, followed by gradual tapering (45). Second, the efficacy of clarithromycin was tested as adjuvant to standard treatment in patients with VAP (46) and in patients with suspected Gram-negative sepsis (47). Significant decrease in mortality was found in both studies in patients with septic shock and multiple organ dysfunction.
The beginning of personalized medicine in sepsis Conclusions
A personalized approach in sepsis would entail a rapid identification of the culprit pathogen, early administration of targeted antimicrobial treatment and dynamic immune response modulation according to the patient’s genomic, transcriptomic and proteomic biomarker profile. Although it may not appear possible to achieve such an approach in the near future, small steps have already been made towards the direction of an individualized management of sepsis. References 1. Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis. Crit Care Med 2013. 2. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003: 348 (2): 138–50. 3. Christaki E, Anyfanti P, Opal SM. Immunomodulatory therapy for sepsis: an update. Expert Rev Anti Infect Ther 2011: 9 (11): 1013–33. 4. Gogos C, Kotsaki A, Pelekanou A et al. Early alterations of the innate and adaptive immune statuses in sepsis according to the type of underlying infection. Crit Care 2010: 14 (3): R96. 5. Poukoulidou T, Spyridaki A, Mihailidou I et al. TREM-1 expression on neutrophils and monocytes of septic patients: relation to the underlying infection and the implicated pathogen. BMC Infect Dis 2011: 11: 309. 6. Ferwerda B, McCall MB, Alonso S et al. TLR4 polymorphisms, infectious diseases, and evolutionary pressure during migration of modern humans. Proc Natl Acad Sci U S A 2007: 104 (42): 16645–50. 7. Waterer GW, Quasney MW, Cantor RM, Wunderink RG. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am J Respir Crit Care Med 2001: 163 (7): 1599–604. 8. Gallagher PM, Lowe G, Fitzgerald T et al. Association of IL-10 polymorphism with severity of illness in community acquired pneumonia. Thorax 2003: 58 (2): 154–6. 9. Tang GJ, Huang SL, Yien HW et al. Tumor necrosis factor gene polymorphism and septic shock in surgical infection. Crit Care Med 2000: 28 (8): 2733–6. 10. O’Keefe GE, Hybki DL, Munford RS. The G>A single nucleotide polymorphism at the −308 position in the tumor necrosis factor-alpha promoter increases the risk for severe sepsis after trauma. J Trauma 2002: 52 (5): 817–25 discussion 825–6. 11. Barber RC, Aragaki CC, Rivera-Chavez FA, Purdue GF, Hunt JL, Horton JW. TLR4 and TNF-alpha polymorphisms are associated with an increased risk for severe sepsis following burn injury. J Med Genet 2004: 41 (11): 808–13. 12. Kotsaki A, Raftogiannis M, Routsi C et al. Genetic polymorphisms within tumor necrosis factor gene promoter region: a role for susceptibility to ventilator-associated pneumonia. Cytokine 2012: 59 (2): 358–63. 13. Watanabe E, Zehnbauer BA, Oda S, Sato Y, Hirasawa H, Buchman TG. Tumor necrosis factor −308 polymorphism (rs1800629) is associated with mortality and ventilator duration in 1057 Caucasian patients. Cytokine 2012: 60 (1): 249–56. 14. Azevedo ZM, Moore DB, Lima FC et al. Tumor necrosis factor (TNF) and lymphotoxin-alpha (LTA) single nucleotide polymorphisms: importance in ARDS in septic pediatric critically ill patients. Hum Immunol 2012: 73 (6): 661–7. 15. Thompson CM, Holden TD, Rona G et al. Toll-like receptor 1 polymorphisms and associated outcomes in sepsis after traumatic injury: a candidate gene association study. Ann Surg 2014: 259 (1): 179–85. 16. Pino-Yanes M, Corrales A, Casula M et al. Common variants of TLR1 associate with organ dysfunction and sustained pro-inflammatory responses during sepsis. PLoS One 2010: 5 (10): e13759. 17. Wurfel MM, Gordon AC, Holden TD et al. Toll-like receptor 1 polymorphisms affect innate immune responses and outcomes in sepsis. Am J Respir Crit Care Med 2008: 178 (7): 710–20. 18. Khor CC, Chapman SJ, Vannberg FO et al. A Mal functional variant is associated with protection against invasive pneumococcal disease, bacteremia, malaria and tuberculosis. Nat Genet 2007: 39 (4): 523–8.
19. Lehmann LE, Book M, Hartmann W et al. A MIF haplotype is associated with the outcome of patients with severe sepsis: a case control study. J Transl Med 2009: 7: 100. 20. Yende S, Angus DC, Kong L et al. The influence of macrophage migration inhibitory factor gene polymorphisms on outcome from community-acquired pneumonia. FASEB J 2009: 23 (8): 2403–11. 21. Li L, Nie W, Zhou H, Yuan W, Li W, Huang W. Association between plasminogen activator inhibitor-1-675 4G/5G polymorphism and sepsis: a meta-analysis. PLoS One 2013: 8 (1): e54883. 22. Morelli A, Ertmer C, Westphal M et al. Effect of heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock: a randomized clinical trial. JAMA 2013: 310 (16): 1683–91. 23. Nakada TA, Russell JA, Boyd JH et al. Beta2-Adrenergic receptor gene polymorphism is associated with mortality in septic shock. Am J Respir Crit Care Med 2010: 181 (2): 143–9. 24. Tang BM, Huang SJ, McLean AS. Genome-wide transcription profiling of human sepsis: a systematic review. Crit Care 2010: 14 (6): R237. 25. Reinhart K, Bauer M, Riedemann NC, Hartog CS. New approaches to sepsis: molecular diagnostics and biomarkers. Clin Microbiol Rev 2012: 25 (4): 609–34. 26. Gibot S, Bene MC, Noel R et al. Combination biomarkers to diagnose sepsis in the critically ill patient. Am J Respir Crit Care Med 2012: 186 (1): 65–71. 27. Tang BM, McLean AS, Dawes IW, Huang SJ, Lin RC. Gene-expression profiling of peripheral blood mononuclear cells in sepsis. Crit Care Med 2009: 37 (3): 882–8. 28. Johnson SB, Lissauer M, Bochicchio GV, Moore R, Cross AS, Scalea TM. Gene expression profiles differentiate between sterile SIRS and early sepsis. Ann Surg 2007: 245 (4): 611–21. 29. Sutherland A, Thomas M, Brandon RA et al. Development and validation of a novel molecular biomarker diagnostic test for the early detection of sepsis. Crit Care 2011: 15 (3): R149. 30. Wong HR, Cvijanovich NZ, Allen GL et al. Validation of a gene expression-based subclassification strategy for pediatric septic shock. Crit Care Med 2011: 39 (11): 2511–7. 31. Su L, Cao L, Zhou R et al. Identification of novel biomarkers for sepsis prognosis via urinary proteomic analysis using iTRAQ labeling and 2D-LC-MS/MS. PLoS One 2013: 8 (1): e54237. 32. Paugam-Burtz C, Albuquerque M, Baron G et al. Plasma proteome to look for diagnostic biomarkers of early bacterial sepsis after liver transplantation: a preliminary study. Anesthesiology 2010: 112 (4): 926–35. 33. Ng PC, Ang IL, Chiu RW et al. Host-response biomarkers for diagnosis of late-onset septicemia and necrotizing enterocolitis in preterm infants. J Clin Invest 2010: 120 (8): 2989–3000. 34. Schmerler D, Neugebauer S, Ludewig K, Bremer-Streck S, Brunkhorst FM, Kiehntopf M. Targeted metabolomics for discrimination of systemic inflammatory disorders in critically ill patients. J Lipid Res 2012: 53 (7): 1369–75. 35. LaRosa SP, Opal SM. Biomarkers: the future. Crit Care Clin 2011: 27 (2): 407–19. 36. Fontana JM, Alexander E, Salvatore M. Translational research in infectious disease: current paradigms and challenges ahead. Transl Res 2012: 159 (6): 430–53. 37. Rapin N, Lund O, Bernaschi M, Castiglione F. Computational immunology meets bioinformatics: the use of prediction tools for molecular binding in the simulation of the immune system. PLoS One 2010: 5 (4): e9862. 38. Kumar A, Roberts D, Wood KE et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006: 34 (6): 1589–96. 39. Savva A, Kanni T, Damoraki G et al. Impact of Toll-like receptor-4 and tumour necrosis factor gene polymorphisms in patients with hidradenitis suppurativa. Br J Dermatol 2013: 168 (2): 311–7. 40. Spyridaki A, Raftogiannis M, Antonopoulou A et al. Effect of clarithromycin in inflammatory markers of patients with ventilator-associated pneumonia and sepsis caused by Gram-negative bacteria: results from a randomized clinical study. Antimicrob Agents Chemother 2012: 56 (7): 3819–25. 41. Docke WD, Randow F, Syrbe U et al. Monocyte deactivation in septic patients: restoration by IFN-gamma treatment. Nat Med 1997: 3 (6): 678–81.
5
Christaki and Giamarellos-Bourboulis 42. Meisel C, Schefold JC, Pschowski R et al. Granulocyte-macrophage colony-stimulating factor to reverse sepsis-associated immunosuppression: a double-blind, randomized, placebo-controlled multicenter trial. Am J Respir Crit Care Med 2009: 180 (7): 640–8. 43. Remmelts HH, Meijvis SC, Heijligenberg R et al. Biomarkers define the clinical response to dexamethasone in community-acquired pneumonia. J Infect 2012: 65 (1): 25–31. 44. Bouadma L, Luyt CE, Tubach F et al. Use of procalcitonin to reduce patients’ exposure to antibiotics in intensive care units (PRORATA trial): a multicentre randomised controlled trial. Lancet 2010: 375 (9713): 463–74.
6
45. Dellinger RP, Levy MM, Rhodes A et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013: 41 (2): 580–637. 46. Giamarellos-Bourboulis EJ, Pechere JC, Routsi C et al. Effect of clarithromycin in patients with sepsis and ventilator-associated pneumonia. Clin Infect Dis 2008: 46 (8): 1157–64. 47. Giamarellos-Bourboulis EJ, Mylona V, Antonopoulou A et al. Effect of clarithromycin in patients with suspected Gram-negative sepsis: results of a randomized controlled trial. J Antimicrob Chemother 2013.
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: 10.1111/cge.12368
Review
The beginning of personalized medicine in sepsis: small steps to a bright future Christaki E., Giamarellos-Bourboulis E.J. The beginning of personalized medicine in sepsis: small steps to a bright future. Clin Genet 2014. © John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2014 There is a growing recognition that there is a need for a more personalized approach towards sepsis care. In most clinical trials investigating novel therapeutic interventions against sepsis, patients have been considered a rather homogeneous population. However, there is probably more individual variability between septic patients than previously considered. The pathophysiology of sepsis is a complex and dynamic process that originates from the host immune response to infection and varies according to the genetic predisposition, immune status and co-morbid conditions of the host, the type of pathogen and the site and extent of infection. Until now, efforts to stratify septic patients according to their immune profile were hampered by the lack of specific biomarkers. Recent advances in molecular medicine have made it possible to develop tools that will facilitate a faster and more precise diagnosis of infection. Individual variability between each patient’s responses to infection can assist in tailoring therapeutic interventions to the individual’s disease profile and monitoring treatment response. In this review, we describe those recent advances in genomics and theragnostics, which are slowly entering clinical practice and which will make possible a more personalized approach to each septic patient in the next decade.
E. Christakia,b and E.J. Giamarellos-Bourboulisc,d a First Department of Internal Medicine, AHEPA University Hospital, Thessaloniki, Greece, b Infectious Diseases Division, Alpert School of Medicine of Brown University, Providence, RI, USA, c 4th Department of Internal Medicine, University of Athens Medical School, Athens, Greece, and d Center for Sepsis Control and Care, Jena University Hospital, Jena, Germany
Key words: genomics – individualized medicine – metabolomics – personalized medicine – sepsis – septic shock – single nucleotide polymorphism – transcriptomics Corresponding author: Eirini Christaki, MD, PhD, First Department of Internal Medicine, AHEPA University Hospital, St. Kiriakidi 1, Thessaloniki, Greece. Tel: +30 6973579927; fax: +30 2310993271; email: [email protected]
Conflict of interest
The authors have no conflict of interest to declare.
The decline of severe sepsis mortality that has been noted during the last 20 years is encouraging, given the absence of a specific diagnostic test and a targeted therapeutic intervention. Although improvements in sepsis management such as the use of early goal-directed therapy, timely administration of antibiotics and optimal mechanical ventilation strategies have probably contributed to the decline, mortality from severe sepsis remains high (1). There is a growing recognition that there is a need for a more personalized approach towards the septic patient. Sepsis pathophysiology is traditionally considered to stem from an overwhelming host response to microbial stimuli. Microorganisms bear highly evolutionary conserved molecules known as pathogen-associated molecular patterns (PAMPs) that interact with pattern recognition receptors (PRRs) of the host innate immune cells. This interaction generates a storm of proinflammatory and anti-inflammatory
Received 21 January 2014, revised and accepted for publication 24 February 2014
mediators that orchestrate septic phenomena (2). This concept of sepsis pathogenesis has led the conduct of several randomized clinical trials during the last two decades. Most of these trials studied the efficacy of compounds that block inflammatory mediators, however, most of them failed to show a survival benefit (3). Part of an explanation for these failures may rely on patient heterogeneity. Despite the use of common clinical characteristics of sepsis as inclusion criteria in clinical trials, enrolled patients can be very heterogeneous in terms of their underlying source of infection, their predisposition and their co-morbidities. Moreover, studies from the Hellenic Sepsis Study Group (HSSG) have shown that the innate and adaptive immune responses in septic patients are not ubiquitous and differ substantially based on the underlying type of infection and individual host responses, making sepsis a ‘prototype’ for personalized disease (4, 5).
1
Christaki and Giamarellos-Bourboulis Genetic predisposition
Genetic diversity is one of the factors that account for the individual variability seen between septic patients and is defined as the carriage of single nucleotide polymorphisms (SNPs) of genes that encode for the protein molecules of inflammatory mediators and of PRRs. Many studies have been published investigating the impact of carriage of one (heterozygotes) or two (homozygotes) SNP alleles on susceptibility to a given infection, however less is known on the role of these SNPs in the development of sepsis and their effect on outcome. It should be noted that a genotype that has a protective role in prevention of infection could prove deleterious if systemic infection occurs. For example, a number of studies suggest that increased tumour necrosis factor (TNF) response decreases the risk of infection but increases the risk of death from septic shock should it occur. Hence, the sequelae of an altered host response may have competing effects in population-based studies (6). The most broadly studied SNPs implicated in sepsis pathophysiology are those of TNF, MIF, PAI-1 and of genes encoding for toll-like receptors (TLR). The most widely studied SNP of TNF involves position −308 of the promoter although SNPs at positions −376 and −238 have also been investigated. All these SNPs involve a substitution of guanine (G) by adenine (A). Existing studies are characterized by great heterogeneity. Two studies in 280 and 93 patients, respectively, failed to disclose a role of carriage of A alleles at the position −308 for the development of community-acquired pneumonia (CAP; 7, 8). However, three studies in surgical patients showed that carriers of SNPs were more prone to the development of severe sepsis either post-operatively or after major trauma (9–11). Analysis of all three SNPs in 213 mechanically ventilated patients indicated that carriage of at least one SNP allele is associated with earlier development of ventilator-associated pneumonia (VAP) and sepsis (12). In a recent study of 1057 critically ill patients of Caucasian origin, mortality was 48.7% for AA homozygotes for the TNF −308 SNP compared with 30.5% in the GA heterozygotes and 29.5% in the GG homozygotes for the wild-type allele (13). The shear value of studying TNF is the location of the gene being in close proximity with the major histocompatibility complex. In a study of 490 paediatric septic patients, haplotype analysis was conducted. Haplotypes constituted of SNPs at position +252 of LTA encoding for lymphotoxin A and of SNPs at positions −863 and −308 of TNF. Results revealed that haplotype +252G LTA, −863C TNF and −308A TNF was protective from the development of acute respiratory distress syndrome (ARDS), independently of patient age, gender, race and co-morbidities (14). TLR1 encodes for TLR1, a PRR that primarily recognizes gram-positive cocci. The most widely described SNPs are one adenine to guanine transition at position −7202 (rs5743551), one adenine to guanine transition at position +742 (rs4833095), and one guanine to thymidine transversion at position +1804 (rs5743612).
2
In a cohort of 1498 patients with major trauma, patients homozygous for the minor frequency G allele at position −7202 and for the minor frequency T allele at position +1804 had greater risk for death from sepsis (adjusted odds ratios: 3.16 and 2.48, respectively). These patients were also at greater risk for death when sepsis was due to gram-positive cocci (15). Another study showed that the −7202GG and the +742GG genotypes were associated with a greater risk for septic shock. These patients had also greater circulating levels of C-reactive protein (CRP) and lower levels of interleukin (IL)-10 than the rest of the cohort (16). Results on the impact of TLR1 SNPs on mortality were fully confirmed by an independent cohort of 493 patients with septic shock (17). The adaptor protein myeloid differentiation protein-88 adaptor-like (Mal), otherwise known as TIR domain-containing adaptor protein (TIRAP) is necessary for TLR2 and TLR4 downstream signalling and thus plays an important role in the innate immune response. In a case–control study of 6106 individuals from the UK, Vietnam and several African countries 33 TIRAP SNPs were genotyped. It was found that heterozygous carriage of TIRAPS180L polymorphism was associated with protection against invasive pneumococcal disease, bacteraemia, malaria and tuberculosis in the different study populations (18). Carriage of one SNP leading to the substitution of asparagine with glycine at position +299 of TLR4 leads to enhanced production of TNFα and to attenuated production of IL-10 by monocytes. This SNP is mainly found across African populations providing protection against mortality from malaria but has also been associated with increased mortality from septic shock (6). MIF encodes for migration inhibition factor (MIF) that antagonizes the function of glucocorticosteroids. Two SNPs are known: one at position −173 and one microsatellite region at position −794. Two independent studies have showed a major role of the CC genotype at the −173 position on sepsis outcome. The first study enrolled 169 patients with severe sepsis of Caucasian origin and showed that patients with the GG genotype had increased risk of mortality mainly after infection of non-pulmonary and of non-abdominal origin (odds ratio: 3.818, p = 0.0066; 19). This association of the GG genotype with unfavourable outcome was also confirmed in a cohort of 1739 patients with CAP (20). PAI-1 encodes for plasminogen activator inhibitor-1. One major 4G/5G SNP has been recognized at position −675. The association of this SNP with the development and outcome of sepsis has been studied with contradictory results. A recent meta-analysis of 12 case–control studies and three cohort studies concluded that SNP carriage increased susceptibility for the development of sepsis and for unfavourable outcome (21). A recent randomized trial by Morelli et al. showed that patients with septic shock who received intravenously the short-acting β1-adrenoreceptor blocker, esmolol, had lower 28-day mortality compared with the control group (49.4% vs 80.5%, p < 0.001; 22). The β2-adrenergic receptor gene (ADRB2) is involved in the regulation of responses to adrenergic agonists in cardiovascular
The beginning of personalized medicine in sepsis and inflammatory diseases. In a study performed in two centres, patients with septic shock who carried the AA genotype of ADRB2 rs1042717 G/A polymorphism (CysGlyGln haplotype) exhibited increased heart rate, worse mortality and more organ dysfunction. Interestingly, in patients treated acutely with low dose corticosteroids for septic shock, the differences in mortality conferred by the ADRB2 genotype were eliminated (23). Diagnosis and monitoring – biomarkers
The host immune response in sepsis varies between patients and in the same patient over time (2). Moreover, the two-stage paradigm of sepsis, where an initial hyperactivation of the innate immune response of variable duration is followed by a hypoinflammatory phase, has been challenged by recent genome-wide expression studies that revealed inconsistency in the observed inflammatory pathway changes in early and late sepsis (24). Studies performed by the HSSG have shown that immune response kinetics differ according to the underlying type of infection, therefore the strategy of using the same biomarker in all sepsis cases may not be optimal (4, 5). Hence, monitoring the inflammatory response during the course of sepsis is critical to stratify patients into subgroups according to their predisposition and functional immune status. The optimal immunomodulatory treatment can then be given and their response to therapy can be accurately assessed. Cytokines, soluble receptors, cell-surface molecules, coagulation, endothelial markers and vasoactive hormones have all been evaluated for their diagnostic and prognostic utility in sepsis. Monocyte cell-surface HLA-DR expression is decreased in septic patients and it is used as marker for immunossuppresion (25). However, it is becoming evident that a single biomarker cannot successfully discriminate sepsis from non-infectious inflammatory conditions or reflect the complex pathophysiologic processes that occur in sepsis and combinations of biomarkers are being increasingly investigated in clinical trials (26). Gene expression assays have also been developed in an effort to map those genes that regulate the innate and adaptive response and are either upregulated or downregulated during the course of sepsis (27). Using gene microarray analysis in whole blood, Johnson et al. showed that infectious systemic inflammatory response syndrome (SIRS) had a different gene expression profile from non-infectious SIRS even before sepsis became clinically apparent (28). Other research groups have identified white blood cell gene signatures that could also differentiate SIRS from sepsis (27, 29). In a paediatric study, a 100-gene expression signature (genes corresponding to adaptive immunity, glucocorticoid receptor signalling, and the peroxisome proliferator-activated receptor-α signalling pathway) was validated for the allocation of children with septic shock into three clinically relevant subclasses (30). Continuous progress in proteomics and bioinformatics has allowed the identification in the blood
or urine of proteins differentially expressed in sepsis and their assessment as potential diagnostic or prognostic biomarkers (31, 32). Proteomic analysis determined novel biomarkers that were then used in a prospective study of neonatal sepsis to successfully risk stratify neonates and guide antibiotic de-escalation (33). Also, in a recent study, using liquid chromatography–tandem mass spectrometry (LC-MS/MS) to perform metabolomic analysis of 186 metabolites, the investigators were able to define two markers, acylcarnitines and the glycerophospholipids, which could help discriminate sepsis from non-infectious SIRS (34). However, gene expression profiling has certain limitations such as inter and intraindividual variation and differences in gene expression between target organs, which will need to be addressed before transcriptome analysis can be interpreted and applied in clinical practice as a surrogate marker for sepsis (35). Although genetic signatures provide a snapshot of an individual’s gene expression profile at any given point during the septic process, they cannot characterize all the factors that regulate an evolving inflammatory process or yet predict the response to a given intervention. Systems biology and systems immunology integrate genomic, proteomic and metabolomic data with the use of advanced computational methods to study the molecular and cellular networks during infection and can provide comprehensive and possibly predictive models of host–pathogen interactions (36, 37). Treatment – future perspectives
In order to apply a personalized approach in sepsis management, we need to be able to map the complex host–pathogen interactions that occur in each septic patient and to translate this knowledge into individualized therapy to improve sepsis-related outcomes (Fig. 1). So far, some components of the septic process have been identified where we can intervene and modify outcomes. Early diagnosis
It is widely recognized that antimicrobial therapy improves sepsis outcomes when administered within the first hour of diagnosis (38). However, limitations of conventional diagnostic methods such as the blood culture, often lead to a delay in initiating appropriate antimicrobial treatment and promote widespread use of unnecessary combinations of broad-spectrum antibiotics. Advances in automated molecular and proteomic methods in microbiology have made possible the early identification and quantification of specific pathogens in blood and other biologic specimens, however, some of these methods still need to be validated in clinical trials. Genetic identity
There is no doubt that in sepsis, certain SNPs of host genes are associated with unfavourable prognosis. Early
3
Christaki and Giamarellos-Bourboulis Host Gene expression
(activation of pro-inflammatory vs anti-inflammatory pathways)
Host DNA (SNPs of TNF, MIF, TLR, PAI-1)
Microorganism
(molecular detection of species and antimicrobial susceptibilities)
Cell secreted and circulating proteins (dynamic changes advise of treatment efficacy)
SEPSIS OUTCOME Fig. 1. Schematic representation of the host and pathogen factors that contribute to outcome in sepsis in a given patient. MIF: migration inhibition factor, PAI-1: plasminogen activator inhibitor-1, SNP: single nucleotide polymorphism, TNF: tumour necrosis factor, TLR: Toll-like receptor.
genotyping of septic patients could enable clinicians to triage those with increased risk for death to early aggressive intensive care support and to identify those patients who will benefit from immunomodulatory interventions based on their immune kinetics. It has been shown that when signs of sepsis develop, monocytes of carriers of A SNP alleles of TNF produce greater amounts of proinflammatory cytokines (12), whereas carriers with chronic inflammatory disorders fail to respond favourably to anti-TNF treatment (39). Gene profiling
There are several available agents that can modulate the host response. Some of them attenuate proinflammatory responses such as antibodies targeting TNFα and the soluble receptor antagonist of IL-1β, whereas others stimulate proinflammatory responses such as interferon-gamma (IFNγ). It is evident that selection of the appropriate immunomodulatory agent necessitates knowledge of the patient’s exact immune response status. This could be achieved with the use of transcriptomic and proteomic profiling in order to discriminate which proinflammatory and anti-inflammatory pathways are up or downregulated. Biomarker-guided therapy
Protein biomarkers are most commonly used to monitor clinical response and guide appropriate therapy. Results of a randomized clinical study from our group, in 200 patients with sepsis due to VAP, indicate that signs of clinical severity accompanied by an increased ratio of IL-10/TNFα, reduced production of IL-6 by circulating monocytes and decreased expression of CD86 on circulating monocytes can serve as a marker for the initiation of adjunctive clarithromycin (40). A prospective study
4
by Döcke et al. (41) also showed that IFN-γ 100 μg/day in patients with sepsis and HLA-DR expression on monocytes less than 30%, reverses immunoparalysis of monocytes. Similarly, in another randomized trial it was found that administration of granulocyte-macrophage colony stimulating factor (GM-CSF) for 8 days restored monocytic immunocompetence and was associated with a shorter time of mechanical ventilation and shorter hospital stay (42). Administration of dexamethasone as adjunctive therapy in CAP decreased the time to pneumonia resolution by 1 day. However, most of the mortality benefit was shown only in the subgroup of patients who presented with increased levels of IL-6, IL-8 and MCP-1 and decreased cortisol levels (43). One biomarker widely used to guide de-escalation or modification of antimicrobial treatment in the critical care setting is procalcitonin (PCT) whose usage can decrease unnecessary antibiotic administration, albeit without a clear effect on outcome (44). Clinical guidance
It is true that a personalized treatment in sepsis, guided by molecular profiling and protein biomarkers, is not feasible yet due to the lack of standardization and randomized clinical trials. However, when used selectively after application of certain clinical criteria, two interventions have shown favourable results. First, current guidelines suggest the intravenous administration of 50 mg hydrocortisone every 6 hours for 7 days in patients with refractory shock, followed by gradual tapering (45). Second, the efficacy of clarithromycin was tested as adjuvant to standard treatment in patients with VAP (46) and in patients with suspected Gram-negative sepsis (47). Significant decrease in mortality was found in both studies in patients with septic shock and multiple organ dysfunction.
The beginning of personalized medicine in sepsis Conclusions
A personalized approach in sepsis would entail a rapid identification of the culprit pathogen, early administration of targeted antimicrobial treatment and dynamic immune response modulation according to the patient’s genomic, transcriptomic and proteomic biomarker profile. Although it may not appear possible to achieve such an approach in the near future, small steps have already been made towards the direction of an individualized management of sepsis. References 1. Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis. Crit Care Med 2013. 2. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003: 348 (2): 138–50. 3. Christaki E, Anyfanti P, Opal SM. Immunomodulatory therapy for sepsis: an update. Expert Rev Anti Infect Ther 2011: 9 (11): 1013–33. 4. Gogos C, Kotsaki A, Pelekanou A et al. Early alterations of the innate and adaptive immune statuses in sepsis according to the type of underlying infection. Crit Care 2010: 14 (3): R96. 5. Poukoulidou T, Spyridaki A, Mihailidou I et al. TREM-1 expression on neutrophils and monocytes of septic patients: relation to the underlying infection and the implicated pathogen. BMC Infect Dis 2011: 11: 309. 6. Ferwerda B, McCall MB, Alonso S et al. TLR4 polymorphisms, infectious diseases, and evolutionary pressure during migration of modern humans. Proc Natl Acad Sci U S A 2007: 104 (42): 16645–50. 7. Waterer GW, Quasney MW, Cantor RM, Wunderink RG. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am J Respir Crit Care Med 2001: 163 (7): 1599–604. 8. Gallagher PM, Lowe G, Fitzgerald T et al. Association of IL-10 polymorphism with severity of illness in community acquired pneumonia. Thorax 2003: 58 (2): 154–6. 9. Tang GJ, Huang SL, Yien HW et al. Tumor necrosis factor gene polymorphism and septic shock in surgical infection. Crit Care Med 2000: 28 (8): 2733–6. 10. O’Keefe GE, Hybki DL, Munford RS. The G>A single nucleotide polymorphism at the −308 position in the tumor necrosis factor-alpha promoter increases the risk for severe sepsis after trauma. J Trauma 2002: 52 (5): 817–25 discussion 825–6. 11. Barber RC, Aragaki CC, Rivera-Chavez FA, Purdue GF, Hunt JL, Horton JW. TLR4 and TNF-alpha polymorphisms are associated with an increased risk for severe sepsis following burn injury. J Med Genet 2004: 41 (11): 808–13. 12. Kotsaki A, Raftogiannis M, Routsi C et al. Genetic polymorphisms within tumor necrosis factor gene promoter region: a role for susceptibility to ventilator-associated pneumonia. Cytokine 2012: 59 (2): 358–63. 13. Watanabe E, Zehnbauer BA, Oda S, Sato Y, Hirasawa H, Buchman TG. Tumor necrosis factor −308 polymorphism (rs1800629) is associated with mortality and ventilator duration in 1057 Caucasian patients. Cytokine 2012: 60 (1): 249–56. 14. Azevedo ZM, Moore DB, Lima FC et al. Tumor necrosis factor (TNF) and lymphotoxin-alpha (LTA) single nucleotide polymorphisms: importance in ARDS in septic pediatric critically ill patients. Hum Immunol 2012: 73 (6): 661–7. 15. Thompson CM, Holden TD, Rona G et al. Toll-like receptor 1 polymorphisms and associated outcomes in sepsis after traumatic injury: a candidate gene association study. Ann Surg 2014: 259 (1): 179–85. 16. Pino-Yanes M, Corrales A, Casula M et al. Common variants of TLR1 associate with organ dysfunction and sustained pro-inflammatory responses during sepsis. PLoS One 2010: 5 (10): e13759. 17. Wurfel MM, Gordon AC, Holden TD et al. Toll-like receptor 1 polymorphisms affect innate immune responses and outcomes in sepsis. Am J Respir Crit Care Med 2008: 178 (7): 710–20. 18. Khor CC, Chapman SJ, Vannberg FO et al. A Mal functional variant is associated with protection against invasive pneumococcal disease, bacteremia, malaria and tuberculosis. Nat Genet 2007: 39 (4): 523–8.
19. Lehmann LE, Book M, Hartmann W et al. A MIF haplotype is associated with the outcome of patients with severe sepsis: a case control study. J Transl Med 2009: 7: 100. 20. Yende S, Angus DC, Kong L et al. The influence of macrophage migration inhibitory factor gene polymorphisms on outcome from community-acquired pneumonia. FASEB J 2009: 23 (8): 2403–11. 21. Li L, Nie W, Zhou H, Yuan W, Li W, Huang W. Association between plasminogen activator inhibitor-1-675 4G/5G polymorphism and sepsis: a meta-analysis. PLoS One 2013: 8 (1): e54883. 22. Morelli A, Ertmer C, Westphal M et al. Effect of heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock: a randomized clinical trial. JAMA 2013: 310 (16): 1683–91. 23. Nakada TA, Russell JA, Boyd JH et al. Beta2-Adrenergic receptor gene polymorphism is associated with mortality in septic shock. Am J Respir Crit Care Med 2010: 181 (2): 143–9. 24. Tang BM, Huang SJ, McLean AS. Genome-wide transcription profiling of human sepsis: a systematic review. Crit Care 2010: 14 (6): R237. 25. Reinhart K, Bauer M, Riedemann NC, Hartog CS. New approaches to sepsis: molecular diagnostics and biomarkers. Clin Microbiol Rev 2012: 25 (4): 609–34. 26. Gibot S, Bene MC, Noel R et al. Combination biomarkers to diagnose sepsis in the critically ill patient. Am J Respir Crit Care Med 2012: 186 (1): 65–71. 27. Tang BM, McLean AS, Dawes IW, Huang SJ, Lin RC. Gene-expression profiling of peripheral blood mononuclear cells in sepsis. Crit Care Med 2009: 37 (3): 882–8. 28. Johnson SB, Lissauer M, Bochicchio GV, Moore R, Cross AS, Scalea TM. Gene expression profiles differentiate between sterile SIRS and early sepsis. Ann Surg 2007: 245 (4): 611–21. 29. Sutherland A, Thomas M, Brandon RA et al. Development and validation of a novel molecular biomarker diagnostic test for the early detection of sepsis. Crit Care 2011: 15 (3): R149. 30. Wong HR, Cvijanovich NZ, Allen GL et al. Validation of a gene expression-based subclassification strategy for pediatric septic shock. Crit Care Med 2011: 39 (11): 2511–7. 31. Su L, Cao L, Zhou R et al. Identification of novel biomarkers for sepsis prognosis via urinary proteomic analysis using iTRAQ labeling and 2D-LC-MS/MS. PLoS One 2013: 8 (1): e54237. 32. Paugam-Burtz C, Albuquerque M, Baron G et al. Plasma proteome to look for diagnostic biomarkers of early bacterial sepsis after liver transplantation: a preliminary study. Anesthesiology 2010: 112 (4): 926–35. 33. Ng PC, Ang IL, Chiu RW et al. Host-response biomarkers for diagnosis of late-onset septicemia and necrotizing enterocolitis in preterm infants. J Clin Invest 2010: 120 (8): 2989–3000. 34. Schmerler D, Neugebauer S, Ludewig K, Bremer-Streck S, Brunkhorst FM, Kiehntopf M. Targeted metabolomics for discrimination of systemic inflammatory disorders in critically ill patients. J Lipid Res 2012: 53 (7): 1369–75. 35. LaRosa SP, Opal SM. Biomarkers: the future. Crit Care Clin 2011: 27 (2): 407–19. 36. Fontana JM, Alexander E, Salvatore M. Translational research in infectious disease: current paradigms and challenges ahead. Transl Res 2012: 159 (6): 430–53. 37. Rapin N, Lund O, Bernaschi M, Castiglione F. Computational immunology meets bioinformatics: the use of prediction tools for molecular binding in the simulation of the immune system. PLoS One 2010: 5 (4): e9862. 38. Kumar A, Roberts D, Wood KE et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006: 34 (6): 1589–96. 39. Savva A, Kanni T, Damoraki G et al. Impact of Toll-like receptor-4 and tumour necrosis factor gene polymorphisms in patients with hidradenitis suppurativa. Br J Dermatol 2013: 168 (2): 311–7. 40. Spyridaki A, Raftogiannis M, Antonopoulou A et al. Effect of clarithromycin in inflammatory markers of patients with ventilator-associated pneumonia and sepsis caused by Gram-negative bacteria: results from a randomized clinical study. Antimicrob Agents Chemother 2012: 56 (7): 3819–25. 41. Docke WD, Randow F, Syrbe U et al. Monocyte deactivation in septic patients: restoration by IFN-gamma treatment. Nat Med 1997: 3 (6): 678–81.
5
Christaki and Giamarellos-Bourboulis 42. Meisel C, Schefold JC, Pschowski R et al. Granulocyte-macrophage colony-stimulating factor to reverse sepsis-associated immunosuppression: a double-blind, randomized, placebo-controlled multicenter trial. Am J Respir Crit Care Med 2009: 180 (7): 640–8. 43. Remmelts HH, Meijvis SC, Heijligenberg R et al. Biomarkers define the clinical response to dexamethasone in community-acquired pneumonia. J Infect 2012: 65 (1): 25–31. 44. Bouadma L, Luyt CE, Tubach F et al. Use of procalcitonin to reduce patients’ exposure to antibiotics in intensive care units (PRORATA trial): a multicentre randomised controlled trial. Lancet 2010: 375 (9713): 463–74.
6
45. Dellinger RP, Levy MM, Rhodes A et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013: 41 (2): 580–637. 46. Giamarellos-Bourboulis EJ, Pechere JC, Routsi C et al. Effect of clarithromycin in patients with sepsis and ventilator-associated pneumonia. Clin Infect Dis 2008: 46 (8): 1157–64. 47. Giamarellos-Bourboulis EJ, Mylona V, Antonopoulou A et al. Effect of clarithromycin in patients with suspected Gram-negative sepsis: results of a randomized controlled trial. J Antimicrob Chemother 2013.
