| Sign In to gain access to subscriptions and/or personal tools. |
A Comparative Analysis of Acute-phase Proteins as Inflammatory Biomarkers in Preclinical Toxicology Studies: Implications for Preclinical to Clinical Translation
1 AstraZeneca, Alderley Park, United Kingdom Correspondence: Anne Lanevschi. 19F58 Mereside, Alderley Park, Macclesfield, Cheshire, SK10 4TG, United Kingdom; e-mail: anne .pietersma{at}astrazeneca.com.
Recently, in early clinical development, a few biologics and small molecules intended as antitumor or anti-inflammatory agents have caused a severe adverse pro-inflammatory systemic reaction also known as systemic inflammatory response syndrome (SIRS). This toxicity could result from expected pharmacological effects of a therapeutic antibody and/or from interaction with antigens expressed on cells/tissues other than the intended target. Clinical monitoring of SIRS is challenging because of the narrow diagnostic window to institute a successful intervening therapeutic strategy prior to acute circulatory collapse. Furthermore, for these classes of therapeutic agents, studies in animals have low predictive ability to identify potential human hazards. In vitro screens with human cells, though promising, need further development. Therefore, identification of improved preclinical diagnostic markers of SIRS will enable clinicians to select applicable markers for clinical testing and avoid potentially catastrophic events. There is limited preclinical toxicology data describing the interspecies performance of acute-phase proteins because the response time, type, and duration of major acute-phase proteins vary significantly between species. This review will attempt to address this intellectual gap, as well as the use and applicability of acute-phase proteins as preclinical to clinical translational biomarkers of SIRS.
Key Words: biomarkers • preclinical safety assessment–risk management • inflammation Abbreviations: A1AG, alpha-1-acid glycoprotein • A2MG, alpha-2-macroglobulin • AP, Alderley Park • APP, acute-phase protein • CARS, counter-inflammatory response syndrome • CRP, C-reactive protein • DIC, disseminated intravascular coagulation • FDPs, fibrin or fibrinogen degradation products • FIB, fibrinogen • HAPT, haptoglobin • HMGB1, high-mobility group B1 protein • ICAM-1, intercellular adhesion molecule-1 • ICH, International Conference for Harmonization • IFN
In recent years, generalized systemic inflammation (SIRS) has become a growing concern in the pharmaceutical industry largely because early in 2006, this was observed as a clinically adverse event in a phase 1 study with TeGenero’s TGN1412. In this trial, this type of toxicity resulted in serious adverse events involving all six healthy volunteers (Gribble et al. 2007; St. Clair 2008). Another trigger for the pharmaceutical industry to explore immunomodulatory effects of drugs more closely has been recent International Conference for Harmonization (ICH) guidance that recommends nonclinical testing approaches to identify compounds unintended to be immunomodulatory, but that have the potential to be immunotoxic (ICH 2006). SIRS is a potentially adverse effect of some biologics and small molecules that modulate the immune system and inflammatory response. These compounds may cause target-or nontarget–related agonist and/or superagonist effects that result in an adverse inflammatory response in humans. Because of the intimate relationship between the inflammatory and hemostatic pathways, profound cardiovascular effects and circulatory collapse are possible, which leads to multi-organ dysfunction.
Vascular leak syndrome, characterized by increased vascular permeability and fluid extravasation, is a dose-limiting toxicity of IL-2, a pro-inflammatory agent. Additionally, lymphocytic and systemic endothelial activation affecting hemostatic function (Gribble et al. 2007) is also observed with IL-2. Other drugs that can trigger immune or inflammatory system activation, as well as coagulopathy, include monoclonal antibodies such as recombinant human IFN In SIRS, the time frame for successful therapy is very short, and the patient may be unresponsive because irreversible changes may already have occurred by the time of diagnosis (Herzum and Renz 2008). Although there are biomarkers that correlate well with disease severity, only a few correlate with prognosis (Herzum and Renz 2008), and characterizing the inflammatory response at a late stage is unacceptable in clinical drug development. SIRS and sepsis carry with them an economic (approximately $17 billion USD yearly) and social impact, constituting the chief cause of death in intensive care units in the United States, with a high mortality of 30%–70% (Riedemann et al. 2003). Furthermore, the incidence of deaths from SIRS is rising by 1.5% yearly in the United States in part owing to an aging and immune-suppressed patient population (Riedemann et al. 2003). The purpose of this review is to (1) discuss the relevance, value, application, and use of acute-phase proteins (APPs) as biomarkers in preclinical toxicology studies to identify compounds with an inherent potential to cause a pro-inflammatory syndrome regardless of mechanism of action; (2) determine species differences that may influence the selection of the APP evaluated in preclinical toxicology studies; (3) determine the acute-phase proteins that have preclinical to clinical translational application; and (4) explore emerging concepts in hemostasis that could be applied in combination with APP to improve pre-clinical and clinical assessment of SIRS.
One aim of preclinical toxicology is to identify candidate drugs that may pose a potential hazard to humans clinically and, where possible, determine the appropriate biomarkers that will enable clinical monitoring to assess human risk. Prediction of human responses is difficult because species differences, genetics, and other factors limit preclinical-to-clinical translation. An understanding of these differences is important to account for them in study design and for accurate data interpretation in preclinical and clinical studies. Furthermore, a sound understanding of interspecies differences allows the selection of the most relevant biomarkers. An ideal biomarker for inflammation will be specific for this process, predict an inflammatory event before it becomes adverse or irreversible, and indicate a favorable or unfavorable prognosis. This would allow timely drug discontinuation and appropriate therapeutic intervention. In clinical drug development, it is particularly desirable to determine predictive versus reporter biomarkers, because signals of an inflammatory response before the development of clinical signs is a vital part of that narrow window in the therapeutic strategy for SIRS. Furthermore, a suitable biomarker(s) must be accurate and reproducible, allow high analytical throughput and short turnaround time, and be widely available for inclusion in a clinical setting.
Candidate Biomarkers for the Inflammatory Response
Cytokines The cytokine IL-6 is one of the most reliable bio-markers, and it correlates well with severity of SIRS and other inflammatory processes (Conway et al. 2004; Herzum et al. 2008; Mokart et al. 2005; Oda et al. 2005; Ridker 2001). Currently there are bedside analytical methods available for IL-6 that allow rapid, reliable, and easy testing (Oda et al. 2005). Other cytokines such as IL-1 or TNF  , although they play an important role in mediating the early inflammatory response (Cavaillon and Annane 2006), are less suitable, given their more transient and variable increases as well as less available and technically more difficult assay platforms.
Acute-phase Proteins In our clinical pathology laboratory, we tested several commercially available APP assays to compare and determine inter-species variability (Wistar rat, AP mouse, Beagle dog) when candidate drugs induced a severe pro-inflammatory response preclinically. Additionally we assessed the preclinical-to-clinical translation of these APPs in phase 1 clinical drug development. Marked interindividual, intraindividual, and species variation were noted for most of the different APPs measured. For example, in the dog, the magnitude of change for SAA (up to 160-fold), CRP (up to 63-fold) and HAPT (up to 21-fold) was greater than for A1AG (up to threefold). These indicate different responses in the magnitude, duration, and transient nature of these proteins in circulation following an inflammatory trigger (Figure 2). Translationally, marked increases in SAA (up to 600-fold) and CRP (up to 60-fold) were also observed in clinical trials when a pro-inflammatory response was a suspected human hazard (Figure 3). A similar pattern and magnitude of change in both these proteins were noted in dog toxicology studies. In contrast, in the rat, the inflammatory response was characterized primarily by increases in A2MG (up to ninefold) and HAPT (up to 4.5-fold) and exceeded the changes observed in fibrinogen and A1AG (up to 2.5-fold, Figure 4). Interestingly, when samples with elevated fibrinogen were tested for CRP to explore the concordance of these two well-established markers of inflammation, the concentrations of CRP were variable and did not support CRP as a sensitive APP in rat (data not shown). In mice administered LPS, marked increases in HAPT (threefold) were observed; however, no salient changes were noted in the A2MG concentration (Figure 5). The elevations in plasma of APP in preclinical toxicology studies correlated with histological evidence of inflammation in several organs, or increased neutrophil count and mild elevations in body temperature in human subjects.
In this safety evaluation setting, most of the APPs tested increased in circulation when there was a confirmed inflammatory trigger, although some species differences were noted. In our experience, the preferred markers for the acute-phase response were considered to be A2MG in the rat, and SAA and CRP for both dog and man. Changes of a similar pattern and magnitude for SAA in man and dog during an acute-phase response support this APP as a valuable diagnostic preclinical-to-clinical translational biomarker for SIRS. In addition, these assessments indicated that CRP in rats was not a sensitive marker for the acute-phase response. Structurally, rat CRP differs from dog or human CRP, as it has a unique, covalently linked dimer in its pentameric structure, whereas human CRP has noncovalently linked subunits (Gentry 1999). This structure may result in different biological activity in the rat. Moreover, in rodents, pregnancy, age, and gender greatly influence circulating concentrations of CRP and can contribute to a variable normal basal serum concentration (Gentry 1999). Healthy rodents also exhibit a higher baseline level of CRP compared to nonrodent species and humans (Gentry 1999). All these factors may affect this protein’s suitability as an APP biomarker in Wistarderived rats. In nonhuman primates, there is support in the literature for CRP as a valuable biomarker of the acute-phase response, with marked increases following an inflammatory trigger, similar to that observed in dog and man. In one study, three strains of monkey (M. irus, M. fusacata, and M. mulatta) were used to evaluate CRP (Jinbo et al. 1998). In this study, CRP levels were associated with increases comparable to that observed in dog and man, of approximately 500-fold, twenty-four hours after administration of turpentine oil (Jinbo et al. 1998). In rhesus monkeys (M. mulatta), increases in FIB and SAA were noted following administration of IL-6 (20 µg/kg/day) for up to thirty-one days, and these findings support the use of these proteins as suitable markers of the acute-phase response in this species (Myers et al. 1995). In nonrodent species, CRP and SAA increase early after the onset of inflammation, and this increased level is clearly different from low baseline levels. Reviews of the literature on early markers of SIRS and other inflammatory processes suggest that CRP and IL-6 are the most widely used markers for early detection of the inflammatory response (Conway et al. 2004; Herzum et al. 2008). However, interspecies differences in the biological behavior of APPs exist, and these differences must be taken into account in preclinical studies. For this reason, CRP and SAA are not markers of choice in the rat, but A2MG can be used as a substitute marker. Acute-phase proteins demonstrate species-related differences in their responses to inflammatory stimuli. Each APP is classified as a major, moderate, or minor APP based on the varying degree of increase following an inflammatory trigger. For example, A2MG is reported as the major APP in rats but not in other species; CRP is a major APP in dogs and humans (Table 1). Interestingly, SAA has been identified as a moderate APP in dogs; however, in our experience the circulating values suggest that this could be considered a major APP in this species. In addition, CRP in rats is reported as a minor APP; however, our experience calls into question its suitability as a biomarker of the acute-phase response.
Hemostatic Markers There is an intimate link between the inflammatory and coagulation pathways at the endothelial interface that provides additional potential biomarkers for early detection of the inflammatory response (Cicala and Cirino 1998; Conway et al. 2004). Following an inflammatory trigger (e.g., LPS), macrophage activation is associated with HMGB1 release. This leads to IL-6 release, which in turn stimulates liver synthesis of APP, including CRP (Riedemann et al. 2003). Interleukin-6 and other mediators also provoke endothelium and blood monocytes to release TF, which initiates the coagulation system. Furthermore, the endothelium is activated by HMGB1, allowing upregulation of adhesion molecules VCAM-1 and ICAM-1. This process favors granulocyte adhesion and extravascular migration, enabling these cells to penetrate affected tissues (where the seat of tissue damage may reside). Endothelial cells also release proinflammatory cytokines, tPA and its inhibitor, PAI-1 (Gribble et al. 2007). Although the two latter molecules are involved in activation and neutralization of fibrinolysis, a hypofibrinolytic effect is thought to predominate, possibly owing to the release of other procoagulant mediators, thus favoring a net hypercoagulable effect. Once coagulation is initiated, thrombin generation is a key process that enables effective clotting (Mann et al. 2003). There are several recent technologies that provide a functional measure of the blood’s ability to generate thrombin and assess whether a net hypercoagulable or hypocoagulable state exists (Aledort et al. 2003; Hemker et al. 1993). These technologies take into account multiple components involved in clotting and may provide a more accurate measure of, if not an important complement to, the measures provided by testing multiple individual plasma components of hemostasis. The latter include but are not restricted to coagulation, endothelial activation markers, or by-products of clotting. They include, among others, TAT complex, tPA, PAI-1, TAFI, protein C, prothrombin fragments, and breakdown products of fibrin or fibrinogen (FDPs or D-dimers). These are the most frequently used markers for monitoring activation of the clotting system and are implicated in the pathogenesis of organ dysfunction as a consequence of SIRS. Most, but not all (e.g., prothrombin fragments) of these commercially available assays cross-react with at least one preclinical species (Ravanat et al. 1995). Some markers, such as TAT and D-dimer, have been shown to assist in the diagnosis of DIC and correlate well with inflammatory states in man and animals (Asakura et al. 2006; Gando et al. 1998; Garcia-Fernandez et al. 2000; Iba et al. 2007; Lanevschi et al. 2003; Watanabe et al. 2001). In SIRS, antithrombin III is one of the most useful predictors of progression, with similar value to IL-6, and protein C has been shown to correlate inversely with mortality (Iba et al. 2007; Fisher and Yan 2000). However, the predictive potential of these markers is manifested in the presence of a clinical syndrome and reflects a state of consumptive coagulopathy that is more advanced than the initial subclinical procoagulant state triggered early in the inflammatory response. In the context of clinical drug development, markers that would be the most desirable are those that describe the net state and possible progression of an activated hemostatic system prior to development of a clinical syndrome such as SIRS. These markers would provide a wider time frame for therapeutic intervention or stopping compound administration, allowing resolution of the response before an adverse reaction occurs. Potential candidates for such markers include kinetic hemostatic assays such as whole blood TEG, or TG assays that use platelet-poor or platelet-rich plasma (Collins et al. 2006). These assays involve multiple components of the hemostatic system, including cellular mediators, and therefore provide a more accurate ex vivo assessment of the net hemostatic balance (hypo- or hypercoagulability or increased fibrinolytic states). In a clinical context, the wide subject-to-subject variation of net hemostatic balance (some people are more "hypercoagulable" than others) presents a challenge for the diagnosis of spontaneous diseases with these technologies and has been a limiting factor for clinical utility (van Veen et al. 2008). Without information on an individual’s hemostatic baseline during health, an accurate measure of deviations in hemostatic balance in individual patients relative to their healthy state cannot be achieved. Measurements must be based on deviations from historical data, with a corresponding loss of diagnostic sensitivity and accuracy. This may have limited the wider use of these assays and limited publications on their applications in a clinical setting. However, these precise technologies are better suited for clinical drug development, where baseline levels can be assessed before administration of a candidate drug and fine deviations monitored with serial measurements over time. There is currently a paucity of data in the literature surrounding these technologies, and it is not yet clear how they play a role in the early detection of hemostatic imbalance, as early markers of inflammation, or of their potential to predict outcome or progression of the inflammatory response. The effective use of hemostatic or inflammatory markers requires a departure from conventional preclinical study designs in toxicity testing, in that additional time points are needed to detect the early inflammatory or procoagulant response. This monitoring, however, more closely mirrors the serial monitoring that occurs in a clinical drug trial, thus providing a better background on a compound’s possible effects prior to initiating the phase 1 clinical trial.
In summary, most markers of the inflammatory response center around inflammatory mediators, including cytokines and acute-phase proteins, although markers of hemostasis also play a role in monitoring a response or predicting outcome. Data generated in our laboratory suggest that CRP and SAA are the major APPs in dog and humans in which an inflammatory response was monitored. Evidence from the literature supports that these are equally useful markers in nonhuman primates. In the rat, A2MG was a reliable marker of the early inflammatory response. The combination of hemostatic markers, cytokines, and APPs may be superior for clinical monitoring when SIRS is a potentially adverse event in humans exposed to immunomodulatory agents. However, further work is needed on emerging technologies such as thrombin generation potential and thromboelastography to adequately improve our assessment of the net state of the hemostatic system ex vivo. These technologies may provide valuable complementary markers for monitoring the early biological responses that occur following an inflammatory trigger, given the concurrent activation of coagulation and inflammation in this biological process. This approach will require changes in conventional preclinical study designs that will better mirror the phase 1 clinical program. The additional preclinical information thus obtained provides a better background on a compound’s possible effects prior to initiating a clinical development program.
Aledort, LM. (2003). Why thrombin generation? From bench to bedside. Pathophysiol Hemost Thromb, 33, 2-3[CrossRef] Asakura, H, Wada, H, Okamoto, K, Iba, T, Uchiyama, T, Eguchi, Y, Kawasugi, K, Koga, S, Mayumi, T, Koike, K, & Gando, S. (2006). Evaluation of hemostatic molecular markers for diagnosis of disseminated intravascular coagulation in patients with infections. Thromb Haem, 95, 282-87[Web of Science][Medline] [Order article via Infotrieve] Cavaillon, JM, & Annane, D. (2006). Compartmentalisation of the inflammatory response in sepsis and SIRS. Review. J Endotoxin Res, 12, 151-70[CrossRef][Medline] [Order article via Infotrieve] Cicala, C, & Cirino, G. (1998). Linkage between inflammation and coagulation: An update on the molecular basis of the crosstalk. Life Sci, 62, 1817-24[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Collins, PW, Macchiavello, LI, Lewis, S, Macartney, N, Saayman, A, Luddington, R, Baglin, T, & Findlay, G. (2006). Global tests of hemostasis in criticially ill patients with severe sepsis syndrome compared to controls. Br J Haematol, 135, 220-27[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Conway, DSG, Buggins, P, Hughes, E, & Lip, EYH. (2004). Relationship of interleukin-6 and C-reactive protein to the prothrombotic state in chronic atrial fibrillation. J Am Coll Cardiol, 43, 2075-82 Fang, XM, Schroder, S, Hoeft, A, & Stuber, F. (1999). Comparison of two polymorphisms of the interleukin-1 gene family: interleukin-1 receptor antagonist polymorphism contributes to susceptibility to sever sepsis. Crit Care Med, 27, 1330-34[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Fisher, CJ., Jr, & Yan, SB. (2000). Protein C levels as a prognostic indicator of outcome in sepsis and related diseases. Crit Care Med, 28, S49-56[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Gando, S, Nanzaki, S, Sasaki, S, & Kemmotsu, O. (1998). Significant correlations between tissue factor and thrombin markers in trauma and septic patients with disseminated intravascular coagulation. Thromb Hemost, 79, 1111-15[Web of Science][Medline] [Order article via Infotrieve] Garcia-Fernandez, N, Montes, R, Purroy, A, & Rocha, E. (2000). Hemostatic disturbances in patients with systemic inflammatory response syndrome and associated acute renal failure. Thromb Res, 100, 19-25[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Gentry, PA. In Loeb, WF, & Quimby, FW (Eds.). (1999). Acute phase proteins. The Clinical Chemistry of Laboratory Animals (pp.336-98). Ann Arbor, MI: Taylor and Francis Gribble, EJ, Sivakumar, PV, Ponce, RA, & Hughes, SD. (2007). Toxicity as a result of immunostimulation by biologics. Expert Opin Drug Metab Toxicol, 3, 209-34[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Hemker, HC, Wielders, S, Kessels, H, & Beguin, S. (1993). Continuous registration of thrombin generation in plasma, its use for the determination of the thrombin potential. Thromb Hemost, 70, 617-24[Web of Science][Medline] [Order article via Infotrieve] Herzum, I, & Renz, H. (2008). Inflammatory markers in SIRS, sepsis and septic shock. Review. Curr Clin Chem, 15, 581-87 Iba, T, Gando, S, Murata, A, Kushimoto, S, Saitoh, D, Eguchi, Y, Ohtomo, Y, Okamoto, K, Koseki, K, Mayumi, T, Ikeda, T, Ishhikura, H, Ueyama, M, Ogura, Y, Endo, S, & Shimazaki, S. (2007). Predicitng the severity of systemic inflammatory response syndrome (SIRS)-associated coagulopathy with hemostatic molecular markers and vascular endothelial injury markers. J Trauma, 63, 1093-98[Web of Science][Medline] [Order article via Infotrieve] International Conference for Harmonization (ICH). (2006). Guidance for industry. S8. Immunotoxic studies for human pharmaceuticals. Center for Drug Evaluation and Research, Food and Drug Administration, Office of Training and Communication. Division of Drug Information, HFD-240: Rockville, MD, USA Jinbo, T, Hayashi, S, Iguchi, K, Shimizu, M, Matsumoto, T, Naiki, M, & Yamamoto, S. (1998). Development of monkey C-reactive protein (CRP) assay methods. Vet Immunol Immunopathol, 61, 195-202[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Kox, WJ, Volk, T, Kox, SN, & Volk, HD. (2004). Immunomodulatory therapies in sepsis. Intensive Care Med, 26, S124-28[CrossRef] Lanevschi-Pietersma, A, Bedard, C, & Dunn, M. (2003). D-dimer, TAT complex and FDPs measured in dogs with different systemic diseases. Vet Clin Pathol, 32, 225 Mann, KG, Butenas, S, & Brummel, K. (2003). The dynamics of thrombin formation. Arterioscler Thromb Vasc Biol, 23, 17-25 Mokart, D, Merlin, M, Sannini, A, Brun, JP, Delpero, JR, Houvenaeghel, G, Moutardier, V, & Blache, JL. (2005). Procalcitonin, interleukin 6 and systemic inflammatory response syndrome (SIRS): early markers of postoperative sepsis after major surgery. Br J Anaesth, 94, 767-73 Murata, H, Shimada, N, & Yoshioka, M. (2004). Current research on acute phase proteins in veterinary diagnosis: an overview. Vet J, 168, 28-40[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Myers, LA, Boyce, JT, & Robison, RL. (1995). The tolerability and pharmacology of interleukin-6 administered in combination with GM-CSF or G-CSF in the rhesus monkey. Toxicol, 101, 156-66 Oda, S, Hirasawa, H, Shiga, H, Nakanishi, K, Matsuda, K, & Nakammua, M. (2005). Sequential measurement of IL-6 blood levels in patients with systemic inflammatory response syndrome (SIRS)/sepsis. Cytokine, 29, 169-75[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Ravanat, C, Freund, M, Dol, F, Cadroy, Y, Roussi, J, Incardona, F, Maffrand, JP, Boneu, B, Drouet, L, Legrand, C, Herbert, JM, & Cazenave, JP. (1995). Cross-reactivity of human molecular markers for detection of prethrombotic states in various animal species. Blood Coagul Fibrinolysis, 6, 446-55[Web of Science][Medline] [Order article via Infotrieve] Ridker, PM. (2001). Role of inflammatory biomarkers in prediction of coronary heart disease. Lancet, 358, 946-48[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Riedemann, NC, Guo, RF, & Ward, PA. (2003). Novel strategies for the treatment of sepsis. Nat Med, 9, 517-24[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Sherwood, E, & Toliver-Kinsky, T. (2004). Mechanisms of the inflammatory response. Best Pract Anaesthesiol, 18, 385-405[CrossRef] St Clair, EW. (2008). The calm after the storm: lessons from the TGN1412 trial. J Clin Invest, 118, 1344-47[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Van Veen, JJ, Gatt, M, & Makris, M. (2008). Thrombin generation testing in routine clinical practice: are we there yet? Br J Haematol, 142, 889-903[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Watanabe, R, Wada, H, Watanabe, Y, Sakakura, M, Nakasaki, T, Mori, Y, Nishikawa, M, Gabazza, EC, Nobori, T, & Shiku, H. (2001). Activity and antigen levels of thrombin-activatable fibrinolysis inhibitor in plasma of patients with disseminated intravascular coagulation. Thromb Res, 104, 1-6[CrossRef][Web of Science][Medline] [Order article via Infotrieve] Wattanathum, A, Manocha, S, Groshaus, H, Russell, JA, & Walley, KR. (2005). Interleukin-10 haplotype associated with increased mortality in critically ill patients with sepsis from pneumonia but not in patients with extrapulmonary sepsis. Chest, 128, 1690-98
This version was published on January
1, 2009 Toxicologic Pathology, Vol. 37, No. 1,
28-33 (2009)
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
Top
Abstract
Introduction
