Method to predict clinical outcome of burn patients

Abstract

The present invention provides a method for predicting a clinical outcome in a burn patient through analysis of gene expression in whole blood. The method includes obtaining a whole blood sample from the patient under conditions suitable for preserving mRNA transcripts and determining the mRNA expression levels of a subset of genes selected from a specified panel of biomarkers. Based on these expression levels, the method computes four distinct probability scores corresponding to the likelihoods of (i) early mortality, (ii) later mortality, (iii) prolonged recovery, and (iv) early discharge. Methods of treatment and kits are also disclosed.

Description

METHOD TO PREDICT CLINICAL OUTCOME OF BURN PATIENTS BACKGROUND

[0001] The immune response is vital for hemostasis and wound healing after thermal injury However, a dysregulated immune response can be detrimental. Severe thermal injury triggers a systemic hyperinflammatory response leading to cellular shock, cardiovascular and endothelial dysfunction, and ultimately increased vascular permeability, hypovolemia, decreased cardiac output, and hypoperfusion. This is defined as burn shock and is the leading cause of early mortality (< 48 hours) after thermal injury. Patients who survive this acute period, remain at an increased risk of mortality due to multiorgan failure and infection. Systemic inflammatory mediators also contribute to microvascular thrombosis and endothelial disruption resulting in multiorgan failure, while the development of immunosuppression and disruption of the skin barrier leaves burn patients at risk for infections. Even after discharge, burn survivors face long-term mortality and morbidity, including persistent susceptibility to infections, hypermetabolism, and hypertrophic scarring.

[0002] Currently, biochemical markers characterize the immune response after thermal injury and attempt to explain its association to coagulopathy and endothelial dysfunction. These biochemical markers have also been used to predict poor clinical outcomes, including mortality, multiorgan dysfunction, and sepsis, in burn-injured patients. However, biochemical markers, like cytokine and acute phase proteins, are byproducts of a dysregulated immune response, and not necessarily the etiology.

[0003] Transcriptomics, the study of genome-wide RNA transcript activity, has the potential to elucidate the underlying molecular mechanisms leading to immune dysregulation and associated poor clinical outcomes after thermal injury. Relative RNA transcript abundances from whole blood are quantified using microarray technologies and when compared across patient groups or time can identify differences in biological pathways and processes. Previous transcriptome studies have shown that thermal injury results in a “genomic storm” followed by shutdown of immune-related pathways within 24 hours. However, these transcriptomic studies do not extend beyond 24 hours and provide little insight into the molecular mechanisms associated with clinical outcomes. SUMMARY OF THE INVENTION

[0004] To address the issues noted above, we were able to characterize longitudinal changes to the whole blood transcriptome following thermal injury and identify transcriptome differences between patients with distinct clinical outcomes related to mortality and prolonged recovery. We discovered that distinct clinical outcomes exhibit different quantitative and qualitative transcriptome trajectories.

[0005] In one embodiment, the invention entails a method to predict a clinical outcome of a burn patient, comprising the steps of: a) drawing an initial blood or serum sample from the burn patient as soon as possible after a burn event; b) conducting transcriptomic analysis on the blood or serum sample, using standard techniques, known in the art; and c) assessing the results of b) based on differentially expressed genes listed in Table 3, wherein transcriptomics abundances of significantly differentially expressed genes are used to generate three probability scores that are predictive of the following three potential clinical outcomes for the burn patient, (i) early mortality, (ii) later mortality, and (iii) prolonged recovery, which can include longer (e.g., at least seven days) hospital stays, wherein probability scores are a weighted sum of gene expression levels, and the probability score will range from 0 to 1, and the combined sum of the three probability scores will be 1, wherein weights to calculate the weighted sum are obtained from training / optimization based on log-fold change statistics, wherein the highest probability score is predictive of the likely clinical outcome of the burn patient.

[0006] Preferably, after the initial draw, additional draws are taken at hours 2, 4, 8, 12, 24 and then every 12 hours until 7 days, and steps b) and c) are followed for each draw, and the results of c) for each additional draw are combined to refine an assessment of the clinical outcome.

[0007] Also described herein is a method of predicting a clinical outcome of a burn patient, comprising the steps of: a) determining the mRNA expression levels of at least 65 genes of a set of biomarkers from a whole blood sample obtained from the burn patient under conditions to preserve mRNAtranscripts, wherein the at least 65 genes is selected from a set of biomarkers consisting of: ACSM3, AKR1C3, ANAPC10, B3GNT5, BATF, BCL2A1, C1orf162, CD93, CLDN14, CLEC2B, CREM, CSTA, DHCR7, DIP2A, DNAJB1, DGKQ, DOCK11, DYNLT3, FAM107A, FAM24A, FCN2, FOS, GLI1, G0S2, GPR84, HBE1, HSP90AA1, HSPA1A, HSPA1B, HSPA9, HSPH1, HSPD1, HSPE1, IL18R1, IRAK3, JOSD1, KBTBD11, KIR2DL5A, KIR2DS4, KLF3, KLRF1, METRNL, MRPS30, NIBAN2, NR4A2, OR4D6, PFKFB3, PIK3R1, PIP5K1A, PLA2G7, PLEK, PRPSAP2, PRRG2, RAMP1, RETN, RGS1, RHOXF2, RNASE2, SH2D2A, SHOC1, SLA2, SLC25A16, SLC45A3, SOCS3, SYAP1, TAMALIN, TBC1D19, THBS1, TLR4, TNFAIP3, TNFSF13B, TOP1MT, TRIB1, USP30, and ZRSR2; b) generating four probability scores for a likelihood of the clinical outcome for (i) early mortality (G1), (ii) later mortality (G2), (iii) prolonged recovery (G3), and (iv) early discharge (G4), wherein each of the four probability scores are calculated using the formula: logit(P) =C0+ C1X1+ C2X2+ ··· + CnXnwhere logit() is the log odds of the probability value P, which is the probability of one of the outcome group, C0 is the intercept value, C1 through Cn are weighted coefficients for each of the 65 genes for each clinical outcome as defined in Table 3, and X1through Xnare the mRNA expression levels of the at least 65 genes, and wherein the highest probability score of G1, G2, G3 and G4 is assigned as the predicted clinical outcome of the burn patient. This method can further generate a report describing the predicted clinical outcome.

[0008] The method can also predict a longer hospital stay, such as, for example, at least 7 days in the hospital for the burn patient.

[0009] mRNA expression can be determined by any applicable method, such as, for example, a nucleic acid sequencing assay, a next generation nucleic acid sequencing (NGS) assay, a Sanger sequencing assay, a PCR assay, a quantitative PCR (qPCR) assay, a reverse transcription PCR(RT- PCR) assay, a mRNA assay, a microarray assay, a Northern blot assay, a Southern blot assay, a luciferase assay, a fluorescence immunoassay.

[0010] Timing is also important in assessing clinical outcome. Preferably, the method is performed within the first 24 hours after a burn injury (i.e., “initial draw”). The method can then be performed every 2 hours for 72 hours after the initial draw or is repeated 2, 4, 8, 12, and 24 hours after the initial draw. In other embodiments, the method is repeated every 12 hours for 72 hours. The method described herein can be repeated for each draw, refining the assessment of clinical outcome. In preferred embodiments, the highest probability scores are combined to refine an assessment of clinical outcome.

[0011] The method described herein can be performed by determining the mRNA expression levels of at least 66, 67, 68, 69, 70, 71, 72, 73, 74, or 75 genes of the set of biomarkers.

[0012] Additionally, the method can further comprise that the logit(P) empirically determines the area under the curve (AUC) of Receiver operating characteristic (ROC) curve.

[0013] In other embodiments, each of the four binary classifier models has a performance of the ROC curve with a sensitivity value of at least 0.8. In even further embodiments, each of the four classifier models has a performance of the ROC curve with a specificity value of at least 0.65.

[0014] The method can further comprise communicating a report to a healthcare provider. The method can also include the administration of an antibiotic, intravenous hydration, transfusion of blood products, a vasopressor, ventilator assistance, a non-steroidal anti- inflammatory agent, or an anti-pyretic agent.

[0015] Computer-implemented systems can be utilized to implement the method. This computer- implemented system comprises at least one processor and at least one memory. In some embodiments, the memory comprises instructions executed by the at least one processor to cause the processor to implement a predictive model that predicts the clinical outcome of the burn subject. This computer-implemented system can be in the form of a hand-held device, and can even be configured to allow for the method to be carried out at the burn patient’s bedside.

[0016] Kits can also be configured to perform this method. For example, kits can comprise polynucleotides for carrying out the method of any one of the previous claims, wherein each polynucleotide specifically detects a gene in the biomarker set. Examples of polynucleotides areprobes that specifically bind to a gene in the biomarker set or primers that specifically amplify a gene in the biomarker set. The polynucleotides can be provided on a solid substrate. BRIEF DESCRIPTION OF THE FIGURES

[0017] FIG. 1A-C. Defining and characterizing patient groups. FIG. 1A. Two clinical outcome variables, days to discharge and days to death, were used to define four patient groups. FIG. 1B. Study participants are hospitalized patients and genome-wide gene expression profiling from blood draws are taken at set timepoints during hospital stay. For each patient, timepoint for with available gene expression data is shown with vertical cyan bar. FIG. 1C. Sample sizes of transcriptomic data for patient groups at each timepoint.

[0018] FIG.2A-D. Hospital admission differential expression analysis between G1 and G4 (FIG. 2A), G2 and G4 (FIG. 2B) and G3 and G4 (FIG 2C). Differentially expressed genes (DEG) are identified using two criteria: fold-change between the mean values of two groups (x-axis) and P- value (y-axis). FIG. 2D. Number of overlapping and unique DEGs in the three comparisons at hospital admission.

[0019] FIG. 3A-F. Genes and pathways associated with early mortality at hospital admission. FIG.3A. Relative abundance of top 50 most significant genes in G1 vs G4 comparison. The color intensity encodes the normalized relative abundance of genes (rows) in each sample (columns), where red-color corresponds to higher values and blue color corresponds to smaller abundance. FIG. 3B-E. Four genes are selected from the DEG set to show varied expression patterns of individual genes in the four patient groups. FIG.3F. Most significantly overrepresented GO terms in the gene sets up-regulated at admission. GO terms can belong to one of three categories: BP (biological processes), CC (cellular components) and MF (molecular functions).

[0020] FIG.4A-B. Between-group comparison of transcriptomic profiles at each timepoint in the acute resuscitative phase (first 48-hours). The three least favorable outcome groups (G1, G2, G3) are compared to the most favorable outcome group (G4). FIG. 4A. Number of significantly up- and down-regulated genes for each comparison and timepoint. FIG. 4B. Significantly overrepresented GOBP (Gene Ontology Biological processes) terms in the up-regulated gene sets (P<1e-6).

[0021] FIG. 5A-B. Intra-group transcriptomics analysis in the four patient groups. FIG. 5A. Number of significantly up- and down-regulated genes from admission to subsequent timepoints. FIG. 5B. Pathway activity patterns for patient group with the longest hospital stay (G3).

[0022] FIG. 6. Longitudinal pathway activation patterns in three distinct outcome groups. Significantly enriched GOBP (Gene Ontology Biological Processes) terms include both common between the groups as well as unique to one of the groups. Pathways with P < 1e-5 in at least one of the groups are depicted.

[0023] FIG. 7. Principal component analysis among batches.

[0024] FIG. 8. Longitudinal pathway activity patterns for patients with early mortality (G1).

[0025] FIG. 9. Longitudinal pathway activity patterns for patients with late mortality (G2).

[0026] FIG. 10. Longitudinal pathway activation patterns in G2, the group with later mortality, shows a number of pathways are persistently activated during the first 48 hours from the admission timepoint. DESCRIPTION OF THE PREFERRED EMBODIMENTS OF THE INVENTION

[0027] A dysregulated immune response after severe burn injury is associated with detrimental short and long-term clinical outcomes. Key changes to gene expression within the first 24 hours after burn injury have been identified, but longitudinal data has been lacking. We were able to characterize gene expression during the first 3 weeks after burn injury and identify genes / pathways associated with distinct clinical outcomes. Patients presenting within 4 hours of thermal injury had RNA isolated for microarray gene expression at admission and set timepoints to 21 days. Inter- and intra-group comparisons were performed between 4 groups (G1 died within 7 days; G2 died after 7 days; G3 discharged after 7 days; G4 discharged within 7 days). A total of 17,289 transcripts were quantified from 116 patients. At admission, there were 110, 80, and 31 differentially expressed genes (DEG) in G1, G2 and G3, respectively, compared to G4. DEGs were largely non-overlapping between groups. Longitudinal intra-group analyses showed distinct group- and time-dependent patterns. G4 had the least pronounced transcriptomic response without noticeable change over time, G3 mounted a response that increased gradually over time, and G1 had the most rapid and pronounced response. Pathway analysis identified persistent upregulation of platelet aggregation and shutdown of immune-related pathways after an initial upregulation among patients with a prolonged recovery. Overall, diverging transcriptome signatures wereassociated with distinct clinical outcomes after burn injury and longitudinal pathway analyses provided insight into the molecular mechanisms underlying the long-term outcomes associated with burn injury.

[0028] This is the most comprehensive and longitudinal study that characterizes the blood transcriptome after thermal injury. We identified distinct, time-dependent gene expression patterns among burn-injured patients with different clinical outcomes based on mortality and prolonged recovery. Differences were seen as early as hospital admission, with worse clinical outcomes being associated with a greater magnitude of DEGs. This magnitude difference persists for 48 hours and pathways that predominate are related to the protein re-folding and the innate immune response. Longitudinal intra-group analyses identified distinct transcriptome signatures among patients with different clinical outcomes. Patients with early mortality experience a rapid increase in DEGs followed by a steep decline at early time-points. Comparatively, all other groups had minimal DEGs at early time points, with diverging transcriptome patterns after 48 hours. Patients with late mortality had a delayed, but dramatic increase in gene regulation followed by a steady decline. Survivors with a prolonged recovery / hospitalization had a gradual increase in gene regulation over time without any decline, and survivors with a short hospitalization had a minimal transcriptome response. These gene expression differences have the potential to identify burn- injured patients who will develop poor clinical outcomes.

[0029] This study re-demonstrated that thermal injury, like endotoxemia and non-burn trauma, results in a rapid and profound transcriptomic response in circulating blood, and that magnitude of this transcriptomic response is somewhat associated with burn size. Previous research has shown that patients with large burns (>20% TBSA) mount a greater transcriptome response compared to patients with smaller burns (<20% TBSA). However, this magnitude difference does not continue to increase in a dose-dependent manner among patients with larger burn sizes (20-40% vs >40% TBSA). In our study, a magnitude difference in the transcriptome response between small and large burn sizes was identified. Patients with minimal injury (median TBSA of 4.5%) and a short hospitalization (G4) had a minimal transcriptome response, while patients with larger burns (G1, G2, G3) had significant alternations in gene expression. However, contrary to the previous study, we identified a magnitude difference in the transcriptome response among patients with large burns when comparing clinical outcomes. Patients with worse outcomes experienced a more pronounced transcriptome response that was not qualitatively different across groups. This comports with aprior observation that the most important differentiating factor between complicated and uncomplicated outcomes after severe blunt trauma is the intensity of the transcriptomic response, rather than qualitative differences.

[0030] Next, we explored specific genes / pathways to better understand the development of immune dysregulation and endotheliopathy after burn injury. Toll-like receptor 4 (TLR-4) is the principal receptor for exogenous molecules from pathogens, including lipopolysaccharide (LPS) and endogenous molecules released by injured tissues or cells. When these molecules are recognized activation of nuclear factor- κB (NF-κB) occurs leading to the synthesis of pro- inflammatory cytokines by blood monocytes and tissue macrophages. Overactivation can lead to a detrimental hyperinflammatory response. Airway epithelia and endothelial cells lining vasculature also express TLR-4 and function to recruit neutrophils when activated. However, overactivation is thought to contribute to endotheliopathy. We identified a significant upregulation of TLR-4 during the first 8 hours following thermal injury, which has previously been shown in other disease states including sepsis and non-burn trauma. Worse clinical outcomes were associated with a greater magnitude of TLR-4 upregulation. This difference helps explain the association between poor clinical outcomes and elevations of serum biomarkers representing a hyperinflammatory response and endotheliopathy, including pro-inflammatory cytokine proteins, syndecan-1, and soluble thrombomodulin.

[0031] After a systemic hyperinflammatory response, burn patients develop immunosuppression leaving them at an increased susceptibility to infections. However, the exact cause of immunosuppression is unclear. Previous transcriptomic research has shown shutdown of immune pathways, including T-cell differentiation through PKCθ signaling and necroptosis, as early as 12 hours after injury. Our study confirms shutdown of these pathways, as well as many others. The downregulation of major histocompatibility complex class II (MHC-II) may also contribute to acquired immunosuppression following burn injury. MHC-II molecules are found on all antigen presenting cells and their major function is to present antigens to CD4+T-lymphocytes stimulating the adaptive immune response Previous in vitro studies have shown that downregulation of MHC- II molecules is associated with impaired interferon (IFN)-ϒ production, decreased T-cell proliferation and acquired immune suppression

[0046] . Our analysis demonstrated that there is downregulation of MHC-II protein complex assembly, antigen presenting, and a decreased response to INF-ϒ among all groups.

[0032] In addition to magnitude differences between patients based on clinical outcomes, there is time-dependent differences in the downregulation of immune-related pathways. Patients who died quickly had downregulation of immune pathways as early as hour 2, while patients who survived but had a prolonged recovery did not have down regulation until hour 12. This time-dependent difference may represent a maladaptive response that leads to poor clinical outcomes.

[0033] Burn injury is now considered a chronic condition. The hypermetabolic state persists for up to three years post-injury and patients remain at risk for infections and thromboembolism even after discharge from the hospital. By analyzing longitudinal transcriptomic data in burn patients who survived but had a prolonged recovery, we identified underlying molecular mechanisms that are associated with these poor outcomes following injury. Previous studies have shown upregulation of prothrombotic genes like platelet factor 4 (PF4) and prostaglandin-endoperoxides within the first 7 days following thermal injury, and we showed that upregulation of prothrombotic genes and platelet aggregation pathways persisted for at least three weeks after injury. Another longitudinal discovery was the persistent shutdown of immune-related pathways, well beyond 24 hours as previously shown. Our data helps explain the mechanisms associated with long-term complications in burn-injured patients and emphasizes the need for further transcriptomic research beyond hospital discharge.

[0034] A potential limitation of this study includes the lack of uninjured controls. However, previous transcriptome research has shown that there is a minimal amount of differentially expressed genes in patients with a TBSA <20% compared to healthy controls. Therefore, our study we used patients with the most optimal outcome (discharge within 7 days) as a “control” group as these patients had minimal injury (TBSA <10%) and insignificant alterations to their transcriptome across time.

[0035] To better understand the long-term impact of burn injury on gene expression we analyzed transcriptomic data up to 21 days after injury. However, complex management of burn-injured patients, including resuscitation, transfusions, and procedures, likely change the transcriptome of a patient. Based on our current study, the cause of the alterations in gene expression identified at later time points could be from burn injury itself, interventions, infection, or likely a combination of multiple factors. Future studies are needed to elucidate the impact of clinical interventions and complications, like sepsis, on gene expression in burn-injured patients.

[0036] Lastly, the RNA used to run microarrays was extracted using whole blood. Complete blood count data was available for 54 patients within a similar timeframe (±30 minutes), and there was no significant difference in any cell composition tested. The lymphocyte counts between groups was not significantly different (p=0.1125). To validate transcriptome findings, blood plasma was stored for future analysis. In this study, no confirmatory analyses were performed at the mRNA or protein level.

[0037] We demonstrated that there are distinct, time-dependent transcriptomic signatures associated with mortality and prolonged recovery after burn injury. Molecular mechanisms associated with long-term complications following burn injury were also identified and include persistent shutdown of immune-related pathways and activation of coagulation-related pathways for at least three weeks. Future development of therapeutic interventions that target differently expressed genes and pathways could potentially alter a patient’s transcriptome trajectory to improve clinical outcomes. Table 1: Clinical and demographic characteristics of study population in the four patient groups. Overall G1 G2 G3 G4 P-value Sample size 116 9 6 70 31les, mean (SD) or median (IQR) are shown. Letter superscripts indicate groups that are significantly different from one another.eett ,h naiC eera2.9 1 2 2 0 - 2- 7 3 1 1- 9- 8- 8-e7-e7 6-e6-e6- 5 wviGtnI- eFe e1 4e-e e e e3 3 -e7 3e-e3 59 8 2 6 3 3 9 1 7 0 0 4 7 3 86 93 bu0 0 4 7 3 4 1 4 5 9 s nidnla 3 6 4 0 0 6 8 3 2 6 7 68 63 7 47 63 9 6 6 4 0 2 3 2 ecdea V.45 10 0 3 0 4 6 83 02 0 2 1 1 84 58 35 5 2 3 3 75 n B P 0 4 4 9 8 4 8 1 9 4 ermrefo2.jff .d 07 8 1 2 3 4 9 8 8 9 37 94 2 6 9 0 5 67 0 9 6 7 7 02 87 9 9 3 7 39 irdepGIa 32 6 3 9 0 8 4 5 0 52 18 77 4 1 2 29 77 1 29 92 12 2 6 8 4 5 35 18 i F.9 6 5 93 9 0 5 15 83 3 61 51 5 0(G a olsedn B n 2reo 1 eab b B 1 A A L ANI2,m m 1BJ1 2 A 2 1 A 1 3S N 1 2 M L g G A uysA A S S G 4 0 A 0 D 9 S C R HP D E A de ,2n_cPS N R R G PS POO T E S 2 R M ss1el ebhtn N g H D H SJS M H H S C A erG(a H T psTsh x p e u ylolr.sw a glo e h ds4 4 i 0 20 40 4 9 4 4 1 7 1 4 9 4 5 teo D2 Gdi_ 82 02 1 3 0 2 8 3 6 2 3 7 2 7 8 2 5 3 9 6 9 3 0 20 54 05 60 87 75 8 n m rm. s eefc3 3 9 5 2 1 e o t nGuoi Iv3 8 0 8 7 2 2 9 6 F 1ne2 g 0 1 0 0 0 1 1 2 0 1 1 2 1 0 0 1 0 00 00 00 0 0 0 0 0 0 0 0 0 fioss .G _l0 0 0 0 0 00 0 0 0 0 0 0 0 0 De-sy rer0 0 0 0 0 0 00 00 00 00 00 00 00 00 gle–b m 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 oeerv l writAecsG G G G G G G G G G G G G G e2elnS S S S S S S S S S S S S S beaeraenpsbeN N N N N N N N N N N N N N E E E E E E E E E E E E E E e a Thtil rTob m 1 1 1 3 P 3 A5 A7 4S 1 1 1 A 1 8 2 0 ysE R DIC1 L 2 K 1 1 1 A 01 D F1IP P 4B3 D FE M GE _ P 3 c S P A D S F R 2 L 2 RL L A M D2IR P R G D R n HKIg P H N K T ARIS M ARIK G Y S A R H T A D H R A K F K S M T R P h A di14 57 18 30 93 67 28 90 58 54 78 59 92 11 43 86 90 46 75 12 _e55 6 1 5 3 4 4 5 4 8 1 6 0 3 8 0 0 8 3 6 3 9 8 9 6 3 1 69 47 10 86 67 05 11 9 2 4 3 7 3 4 7 9 ne10 10 10 10 10 20 10 10 20 10 1 61 31 01 71 01 71 70 50 01 g 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS G G G G G G G G G G G G G n S S S S S S S S S S S S S e N N N N N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E E E E Elob 1 m L 2 R B 2 1 2 A 4 L 4 1 2 3 1 R 4 1 3IB ysD 1 4 L C 2 D N B8 D R S D APA I 6 1 8 C X B 4 _ 3 c nRIC V X A M P P 2 D B N 4 3 S R R R R R T TRIW H CSP N F G E T A N B LIG L K U T D P g K H F P A K F K R Z S h di97 48 77 14 24 32 13 39 10 18 80 29 37 41 27 54 33 80 88 _e36 1 7 3 0 4 5 5 3 4 2 9 7 1 8 3 5 4 7 0 5 5 6 7 5 1 43 18 87 06 73 87 91 6 0 6 9 4 0 0 4 ne20 10 10 10 10 00 20 10 10 10 1 31 81 11 31 31 20 21 81 g 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS G G G G G G G G G G G G G n S S S S S S S S S S S S S e N N N N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E E E Eob 1 3 m P A 1 1 S 1 L 5 7 y C 5 s N 5 R A L 2 4 6 2 7 3 01 8 2 D M 2TIR R D A 5 V N S N R T R 2fr_cR 4 S C 1 L G S D C T O AGIP1 PE C D C P C O C L F T A C T X F N o7 n C LLIP T S F A S C Z 1 g S B M D C h di55 5 4 9 7 5 9 7 9 0 8 2 6 6 5 2 8 5 0 6 17 06 73 1 8 0 4 3 9 5 5 6 4 4 1 7 1 4 01 _e44 85 51 0 3 4 7 9 0 9 1 9 4 8 6 1 7 8 0 6 8 1 8 3 0 3 3 8 3 2 8 7 2 67 69 60 84 07 88 8 2 7 5 ne1 1 1 1 1 0 1 1 1 1 2 1 1 1 1 0 31 71 61 02 g 00 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 0 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS GS G G G G G G G G G G G G n S S S S S S S S S S S S e N N N N N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E E E E Eob B 2 m 5 2 1 2 A 8 8 6 y N 3 4 P 2 1 A 1 0 5 M 5 E G 4 s T L D A R R 7 C P 3 S 6 2 1 3 N G A R S L S G _ E C D G G AP X LIM A E D A A C L C K A S G O c R C H E O R A X N A R N n A R L C M C M C N C E R F g h G A T C di81 7 0 5 7 1 1 7 0 4 1 0 9 2 9 7 3 0 2 1 9 65 68 08 78 2 8 2 9 5 8 0 7 3 6 2 1 6 5 9 _e40 87 99 82 2 1 2 8 2 9 1 4 6 0 4 0 4 9 2 2 2 7 7 3 8 1 4 3 4 5 1 6 1 11 46 35 42 58 96 7 7 3 ne10 2 0 1 1 1 1 1 1 0 1 1 1 1 1 2 1 11 11 81 g 0 00 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 0 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 0 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS GS G G G G G G G G G G G G n S S S S S S S S S S S S e N N N N N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E E E E E05 74 4 64 14 0 21 93 49 70 49 8 99 58 0 9 1 7 2 0 0.2 7 9 9 6 0 3 8 1 8 4 42.0.0 64 1 4 514 1 4 1.-11-5.1 -3.1 -1- .1.12.15.16.11.15.1.-26.12.17.1 lob m A 5 2 B S 4 E y B 9 7 G 2 1 O 1 B 1 s E 1 3 5 6 H L 7 T S DAJM FIC K _cFID D CTIM R G D n E C P K A DNI2 H O R F L M A D T P O N S L B N A ERIg F C L S K E D G h di71 84 15 03 05 01 28 47 04 93 80 30 82 32 52 71 67 _e48 8 e 5 0 8 1 9 1 6 9 5 4 6 5 9 3 4 2 4 58 75 54 53 40 95 22 41 61 30 n 10 10 1 7 9 2 3 6 4 0 1 0 7 9 0 00 10 20 10 10 10 10 10 10 10 00 1 1 0 g 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _l0 0 0 0 0 0 0 0 0 0 0 0 0 b 00 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS GS G G G G G G G G G n S S S S S S S S S e N N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E E5 e 5- 4 4 4 e 4 4 4 4 4 9 4 8 8 --e- 6e- 1e4 4 4 4 5 0 2 9 7 -e-e-e-e-e-e-e-e-e19 16 9 6 8 8 8 4 6 8 8 8 9 9 1 9 9 6 8 9ob 2 0 T 3 61 9 B 3 2 m P 7 1 2 ysA R C R M T L A Q 9 1 A S 3 1 D 1 B A42 N A 03 _ S C P S 1 5 c P H A R P N 2 K F C G P B S H L 1 A F P K M A 2 PIPS n R D N Z O Y L D H T K C B S F ABID U g P A T D S T H P F N h di72 3 2 9 8 9 2 4 3 1 7 0 4 5 5 0 5 3 1 98 61 42 24 61 19 12 10 08 8 8 0 2 9 3 0 9 _e1 e 4 2 4 9 4 5 2 5 3 7 79 69 82 50 73 86 30 05 n 1 7 0 1 6 0 1 6 0 1 8 0 1 6 0 1 2 0 1 4 0 1 11 31 01 01 32 71 02 31 61 31 g 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _l0 0 0 00 00 00 00 00 00 00 00 00 00 0 0 0 0 0 b 00 00 00 00 00 0 0 0 0 0 0 0 0 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS GS GS GS GS G G G G G G G n S S S S S S S e N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N Eob m 2 A 3 2 A 3 1 6 41 03 85 A 41 4 4 A 2 B7 2 ysG 1 2 K S 5 S C D N 2 N Sfr3 D L F NL S B7 T _ R c R 5 0 4 C C O 4 R D C P o HMI7 S L S R C G G 1 C n PPIG L O S H S O L F R X D M C L M S O A D P A A GLID g P S C A T L h di06 89 98 51 75 18 48 16 93 69 28 01 38 76 66 16 01 20 34 45 _e46 3 2 3 6 4 3 7 2 8 5 5 4 1 8 2 3 9 1 2 6 6 3 4 7 9 7 3 8 56 66 95 06 21 56 27 3 4 8 7 8 2 7 3 ne10 10 10 10 10 10 10 10 10 10 1 0 61 71 60 91 51 82 21 61 g 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS G G G G G G G G G G G G G n S S S S S S S S S S S S S e N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N Eob m 8 0 1 y 3 1 1 B E 1 01 1 5 97 2E 7 0 4 2 1 5 4 B s S 1 XNI8 C M M R M C 2S 6 4 2 CS 2 28 N2 N _cS C R D OS P R E M T O H E 2 M BLX B F E A O F 5 C ng P C C Q R E R P L SIF Y F I M C M S BIT N T G R X E F N Z R O A C h S di88 19 06 55 71 89 81 50 98 27 13 42 63 54 23 69 88 48 90 98 _e85 4 8 8 2 6 6 3 1 1 0 5 6 6 3 4 8 1 5 6 8 8 3 9 0 3 2 17 48 18 10 43 36 65 5 1 4 9 8 0 2 1 2 ne10 10 10 00 10 20 20 10 10 10 1 21 11 61 31 71 70 01 81 81 g 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS G G G G G G G G G G G G G n S S S S S S S S S S S S S e N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N Eob 3 m L 1 B ys1 C 8 0 0 F 1S 1 B M 1 4 1L 1 A14 L 41 1 0J1 1 D 1 C 1 X C R LB87 2 6 R 3 7 X C P L _ N C OD- D CIE H c B T F 1 T P B n P G N R M A 3 Q P Y g E G Z C E B P A O F L M A X B H L T C O T A B P X A D M H P C F U C h T di79 53 84 12 33 16 14 29 29 97 66 83 43 28 33 27 92 27 99 03 _e32 4 8 9 8 4 3 2 6 9 8 1 4 2 0 0 9 4 3 8 8 3 3 2 3 5 6 92 38 53 00 24 85 12 2 3 6 8 7 8 5 8 8 ne00 10 10 10 20 10 10 10 20 10 1 80 71 21 91 00 61 61 41 61 g 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 00 00 00 00 00 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 esGS GS GS GS GS GS GS G G G G G G G G G G G G G n S S S S S S S S S S S S S e N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N EV.18 2.8 P 82 65 0 5 5 5 1 1 8 0 0 0 4 4 86 92 9 9 9 9 7 Pj7.8 5j0 8 8 3 3 3 4 4 2 2 2 2 4 . 2 d 8 1 3 3 3 9 9 0 2 2 2 2 2 6 d 0 2 a 1 9 23 5 a 8 9 59 59 57 5 03 40 40 40 40 40 12 7 29 52 07 88 02 02 0 7 2 0 0 0 0 0 3 2 0 0 5 7 7 7 7 7 7o o b b B 3 3 1 2 1 11 6 3 B 1 5 m 4 1 mS 1 N A A E 4 1 D 2 A F T ys1F Y _ SysC FS T T 2 S R S K K C 2 T N S L A L O C B A E 2 A c G n F H_CcO F E n S N R C C T F O T L C G3 B N BRIC L B B g g T R D K S h h di66 37 4di75 42 81 25 97 58 96 5 1 5 6 2 9 7 7 e 2 1 5 5 9 4 5 4 7 5 0 2 9 _ 4 8 Gs_e4 5 3 3 8 3 2 6 3 8 8 1 5 8 20 40 12 04 96 63 07 74 37 09 01 40 65 6 ne01 31v ne10 10 10 10 10 10 10 10 10 2 0 1 0 1 71 g 0 0 3 g 0 0 0 0 0 0 0 0 0 0 0 0 _l0 0_l0 0 0 0 0 0 0 0 0 0 0 0 b 00 00Gb 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 m 0 0–0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 esG G C mesGS GS GS GS GS GS GS GS GS G G G G G G n S S e N2E N EelnS S S S S S N N N N N N beN N N N N N N N N a E E E E E E E E E E E E E E E TFg 8 ol4 2 1 0 7 3 9 4 0 6 28 54 39 1 9 8 7 49 6 5 1 93 35 1 73 5 7 0 5 8 0 21 01 07 2 33 3 75 27 42 87 59 7 4 6 9 5 5 1 6 2 4 4 3 3 4 0 3 1 8 7 5 3 2 8 7 4 0 3 3 2 5 9 2 3 . 5 9 8 8 0 6 6 6 6 7 7 7 8 6 8 119.14.16.1 -8.91.0.0 7 4 1 -0.15.13.10.11- 10.13.10.10.1 -0.1 -lob 2 2 7 3 1 m2 3 4 1 F 6 K 1 G 6 2 1 1 B L ysS 9 8 _c0 E X 1 R 2 M 3 M N X 8 F D R B G OfrE 8 S D P R P C C P o L 1 A C C R N 2 T T n G H H 1 P CLIL P A T V P A C g R h di98 0 e 6 1 2 8 7 1 5 3 1 9 2 0 7 1 6 1 5 4 9 0 0 6 7 7 0 8 8 1 1 5 2 8 99 50 32 29 _ 3 5 9 3 1 3 5 5 6 5 1 2 4 9 0 e 2 2 3 1 3 4 1 1 4 5 2 2 8 1 4 n 10 10 10 20 10 10 10 1 1 00 31 41 11 01 80 21 g 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _lb 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 0 00 00 00 00 00 00 00 00 00 00 00 m 0 0 0 0 0 0 0 0 0 0 0 0 esGS G G G G G G G G G G G G G G G n S S S S S S S S S S S S S S S e N N N N N N N N N N N N N N N N E E E E E E E E E E E E E E E E,6 2 0 2 5 1 9 0 6 1 1 46 2 77 2 7 21 3 1 09 2 42 5 3 0 9 5 7 62 2 7 Ge,1ht19 35 1 5 6 7 8 6 7 1 3 4 9 2 6 1 4 9 2 8 1 3 0 5 3 81 5 0 43 Gt( u 4 6 G 5 4 7 1 3 7 0 0 4 6 3 7 8 5 6 5 9 4 1 8 5 8 56 2 02 5 3 5 2 poyr_f2 0 3 6 5 e 0 1 0 1 6 4 3 6 4 87 1 0 53 0 29 9 6 2 1 1 7 22 8 5ifsfaew6. .0 - 012.2. . .0 3 4 111. .73. .01.9.-51- 21- 7 47.5.214 osi- cdsyletlhao n b B 1 L A gieam lay 1 1 1 2 s ABJS A 2 A P A S 1 A A A 1 3 0 D S N 1 2NIG 4 0 P 9 S C R H T P D M 2 E L R A wit ._ S N R R G S POO E S H C M e n n hit er ec r n H D N H SJS M H S A T seg H h eefzifir dhaedebidi4 2 4 4 9 4 4 1 7 1 4 9 4 5 _ 0 0 0 3 8 2 2 2 5 3 9 6 9 3 m h Trce 82 0 3 2 1 2 6 7 8 2 5 0 6 8 7 8 3 09 35 3 7 0 0 4 5 0 7 5 1 msen 2 1 0 1 21 3 8 0 8 7 2 2 9 6 u.)e 0 0 0 0 0 20 0 1 1 2 1 0 0 1 S.4 d Gsg a_l0 b 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 3 d d 0 0 0 0 0 0 0 0 0 0 00 00 0 0 eln baao m 0 0 h e 3tes G n S GS GS GS GS GS GS GS GS GS GS GS GS GS T G m e N E N E N E N E N E N E N E N E N E N E N E N E N E N E03 2 2 54.7 9 50 1 7 3 9 7 4 9 3 7 9.9 4.5 3.1. .6 91.4.9.7. . .5.7.4.6 57.53- -1- 83- 0 93-7-4- 75- 2 13-lob m 1 1 y E R 1 3 DPI3 A C 5 A 1 L 2 7 4S 1 1 1 2 7 A 0 D F1IP P 1 P R _BA s P 3 P A D 1 R L A MIS C c SKIS F R 2 L 2 N K S MRIL K G Y R S A T P H n H P H T ARIA K R R D g K F P h di14 57 18 30 93 6 2 9 5 5 7 5 9 4 7 3 _ 5 6 7 8 0 8 4 8 9 2 3 2 9 e 5 5 34 58 16 64 01 38 86 00 01 89 3 3 1 8 ne 1 g 1 41 41 11 91 72 0 6 7 5 1 6 23 37 14 27 _l00 0 0 0 0 0 10 10 20 10 10 10 10 10 10 10 0 00 00 00 00 00 00 00 00 0 0 0 0 0 0 0 b 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 m 0 0 0 0 0 0 0 0 0 0 0 0 0 es GS GS GS GS GS GS GS GS G G G G G G G G n S S S S S S S S e N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E7.0 1 021.-4-6.5 0 05.31.6 9 5 7 7 0.4.5.4.3.76 4 - 23-6-4- 3 65.9 3 8 5 33.4.4.7.02-2- 3 2 T 3 61 M T A Q 9 1 9 3 1 3 A 2 B 4 N A 0 2 RS 1 L R P N 52 K A G P S S B F D L 1 F 2 K M A 2 3 BIPIP G S R Z O Y C D H K C F A D R T D L H T B T P F N U P S 94 82 96 21 41 31 10 78 0 5 5 0 5 3 0 2 4 1 9 2 8 2 9 3 0 9 6 9 4 5 2 5 03 87 79 69 50 73 86 30 0 4 6 8 6 2 4 1 3 0 0 7 5 6 1 1 1 1 1 1 1 0 3 6 3 2 0 0 0 0 1 1 1 2 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 0 0 0 0 0 0 0 0 00 00 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 GS GS GS GS GS GS GS GS GS GS GS GS GS GS GS N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E4 0.2 77.5 8 33.6 7 0 2 -6.5.5.5 4 7.7.5 9.0.7 6 08.8.3 6.0 0.2.0 1 05-2-5- 61- 68- 81-1- 3A 1 6 4 0 B3 1 2 1 1 C 1 2 3 1 N A A E 4 1 1 54 D C O 4 N N SP F T TS 2L S S K D A R O B L H R D C S S O L C F R SF E R C C N L T F C M N O TB T B R D K 51 18 48 16 93 69 42 81 25 97 5 9 5 1 5 7 1 8 2 3 9 5 8 6 4 5 4 85 5 6 9 0 2 2 94 51 30 39 86 30 27 63 1 61 61 51 61 11 01 01 21 4 6 3 7 4 7 0 0 0 0 0 1 1 1 1 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 0 0 0 0 0 0 0 0 00 00 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 GS GS GS GS GS GS GS GS GS GS GS GS GS GS GS N E N E N E N E N E N E N E N E N E N E N E N E N E N E N E4 8 4 0 7 9 5 3 8 8 1 0 7 8 6 16 0 9 8 8 0 8 7 1 6 0 1 2 4 5 0 4 9 0 9 9 4 5 5 9 0 0 3 7 56 9 2 8 7 3 9 6 5 9 2 2 9 2 4 5 8 8 9 3 8 7 9 0 0 3 0 38 4 2 3 0 2 7 3 3 2 5 6 3 2 53 0 3 9 8 1 6 3 2 9 2 8 4 9 94 5 2 8 4 6 9 6 8 5 9 0 4 1 9 9 3 5 35.5 9 2 42547 0 2 4.1 -1.4 0 1 -9.7.3.13- 11. .2 018. .0 13.6.6.16-3- 61 B2 5 4 2 2 7 3)tp C F T T N 39 8 1 F 6 E A G D R E X 1 P B OfrK 1 E R G e o L 8 2 M 1 A ScrL B 3 C B C G H H R 1 P CLIL C P AetnI(25 72 79 01 27 1 1 0 6 4 0 7 8 1 5 8 5 39 2 1 5 0 7 8 0 6 6 5 9 3 71 13 9 6 0 1 1 5 7 2 3 1 3 4 51 51 6 5 10 1 1 1 1 2 1 1 1 1 41 00 0 0 0 0 0 0 0 0 0 0 0 0 A 0 00 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 N 0 0 0 00 00 00 00 00 00 00 00 00 GS GS GS GS GS GS GS GS GS GS GS GS N E N E N E N E N E N E N E N E N E N E N E N EEXAMPLES

[0038] The invention is further described by the following non-limiting examples.

[0039] Methods. The Institutional Review Board of MedStar Health Research Institute (IRB # 2012-029) and the Human Research Protections Office of the US Army Medical Research and Development Command approved this research. This analysis is part of a larger multicenter Systems Biology Coagulopathy of Trauma (SYSCOT) Research Program.

[0040] Study Population. Patients who presented to a regional burn center within 4 hours of injury were screened for enrollment between October 2012 and March 2017. Patients aged 18 years and older presenting with a flash, flame, hot contact, scald, or electrical injury were included. Patients who presented with chemical injuries, were not fluent in either English or Spanish, had a history of coagulopathy or anticoagulant use, or were otherwise not thought fit for inclusion based on pre-existing conditions were excluded from the study.

[0041] Clinical Data and Sample Collection. Patient demographics, burn injury characteristics, laboratory values, clinical events / outcomes, and interventions with associated timing were prospectively collected from the medical record and stored in REDCap. Blood draws were performed at hospital admission, at hours 2, 4, 8, 12, 24, and then every 12 hours until 7 days. Additional samples were collected on days 14 and 21, or until discharge or death. For transcriptomic analysis, samples were collected in PAXgene™ reagent (Qiagen / BD, CA) and BD Vacutainer™ tubes (Qiagen / BD, CA).

[0042] mRNA Isolation and Transcriptome Profiling. Total RNA was isolated using the PAXgene Blood RNA Kit (Qiagen / BD, CA) following manufacture directions, and global transcript levels were measured over time using the Agilent SurePrint G3 Human Gene Expression v3 Microarray Kit (Agilent Technologies, Inc., CA).

[0043] Bioinformatics Analysis. RNA hybridization was done in two batches and principal component analysis confirmed that the samples largely clustered together without significant differences based on the batch. Therefore, all samples were analyzed together. Next, the Limma package in R statistical software was used to fit a linear model to each probe and calculate differential gene expression (DEGs) between groups. Groups were determined based on mortality status and hospital length of stay, G1- death within 7 days, G2- death after 7 days, G3- survivors with a prolonged recovery (hospital length of stay > 7 days), G4- survivors with a quick recovery (hospital length of stay <7 days). Differentially expressed genes (DEGs) were defined with two-part criteria, P-value and fold-change thresholds (False Discovery Rate (FDR) adjusted P < 0.05 and magnitude of log fold-change > 1) and the Benjamini-Hochbery method was used to control for multiple testing.

[0044] Two complementary pathway enrichment analysis methods as implemented in clusterProfiler R package were utilized: over-representation analysis (ORA) and gene set enrichment analysis (GSEA). We implemented ORA for between-group analyses and GSEA for intra-group analyses. Gene ontology (GO) terms of three categories (BP: biological processes, CC: cellular components and MF: molecular functions) were examined. Differentially expressed probes were then mapped to their responsive pathways using Ingenuity Pathway Analysis (IPA). IPA’s Canonical Pathways and Upstream Regulator features were used to predict the differential activity of cellular pathways and regulatory networks. All plots and visualizations are generated using the ggplot2 R package unless otherwise stated.

[0045] Statistical Analysis. Descriptive statistics characterized the demographics and injury characteristics among groups. Categorical variables were presented as frequencies and percentages and tested using the χ2. Continuous variables were tested for normality using the Shapiro-Wilk test and descriptive plots. Normally distributed data were expressed as mean and standard deviations (SD) and tested for differences using the one-way ANOVA followed by Turkey correction for post hoc pairwise comparisons. Otherwise, data were presented as median and interquartile ranges (IQR) and tested using the Kruskal-Wallis test followed by Dunn’s test for pairwise comparisons. Statistical significance was determined at the p<0.05 level (2-sided).

[0046] Data Availability. The raw gene expression data has been uploaded to GEO (accession number: GSE182616) Results Patient Demographics, Burn Injury Characteristics, and Overall Transcripts Admission Transcriptomic Signatures Associated with Distinct Clinical Outcomes.

[0047] Admission transcriptomic signatures were analyzed first to minimize the impact of medications, resuscitation, and interventions. A total of 106 whole-genome transcriptomic profiles were obtained, consisting of 9, 5, 65, and 27 samples for G1, G2, G3, and G4, respectively (FIG. 1C). G4, the group with minimal injury and most favorable outcome, was compared against all other groups (G1 vs G4, G2 vs G4, G3 vs G4). There were 110 (80 up and 30 down), 80 (53 upand 27 down) and 31 (26 up and 5 down) genes differentially expressed in G1, G2 and G3, respectively (FDR-corrected P < 0.05 and magnitude of log fold-change > 1) (FIG. 2A-C). DEG sets were largely nonoverlapping among these three comparisons, with only a few immune-related genes being significantly altered across all three groups (FIG.2D). There was also a larger number of genes up-regulated than down-regulated across all three groups.

[0048] The top 50 most significantly regulated (least P-value) genes between G1 vs G4 are shown in FIG.3A. Notably, the most significantly up-regulated gene list includes several genes encoding heat shock proteins (HSPA1B, DNAJB1, HSP90AA1, HSPA1A, and HSPH1) and important immune-related genes (G0S2, SOCS3, TNFAIP3, and RGS1). There was a varying activation pattern when analyzing these significant genes across all groups. Abundances of select representative genes at hospital admission are shown in FIG. 3B-E. Genes related to heat shock (HSPA1B) show a dose-dependent relationship with poorer prognosis (FIG. 3B), while other genes related to immune activation (TLR-4) seem to be activated similarly in all three patient groups compared to G4 (FIG. 3E).

[0049] Next, enrichment analysis of gene ontology (GO) terms was conducted to gain insight into the biological pathways and mechanisms. After multiple testing corrections (FDR < 0.05), 196 and 47 GO terms were significantly overrepresented in the significant gene sets of G1 vs G4 and G3 vs G4 analyses, respectively. No GO term survived the FDR-corrected significance level of enrichment for the G2 vs G4 gene set due to sample size. The top 15 significant GO terms (by P- value) are shown in FIG.3F. In the early mortality group (G1), the most significantly upregulated genes included the response to unfolded proteins (GO:0006986), activation of innate immune response (GO:0002218), and tumor necrosis factor production (GO:0032640). The most significant GO terms in the group with a prolonged recovery and long hospital length of stay (G3) included temperature homeostasis (GO:0001659), regulation of the response to oxidative stress (GO:1902882) and macrophage cytokine production (GO:0010934). Inter-Group Transcriptomics Differences during the First 48-hours

[0050] To explore the effect of temporal proximity on injury, between-group differential expression analysis was performed during the first 48 hours after admission, consisting of 8 time points (0, 2, 4, 6, 12, 24, 36, and 48 hours post admission). This time was chosen to capture the acute resuscitation period after burn injury. Compared to G4, all other groups (G1, G2, and G3)had significantly more transcriptomic alterations across all time points. The largest magnitude of transcriptomic alterations is observed in the G1 group, with markedly more up-regulated genes than down-regulated ones (FIG. 4A).

[0051] GO term enrichment analysis was next performed on the up-regulated set of significant genes in groups G1, G2 and G3 compared to G4. Protein refolding-related pathways in the G1 were limited to hospital admission (FIG. 4B). G1 also had significant activation of the innate immune response (GO:0002218) and tumor necrosis factor production (GO:0032640) during the first 12 hours, but not at later time points. GO terms uniquely activated in G3 include platelet activation (GO:0030168) and regulation of phagocytosis (GO:0050764) across most time points. Enrichment analysis with the G2 vs G4 significant gene sets did not reach the same level of FDR- corrected statistically significant threshold due to sample size in G2. Longitudinal Intra-Group Transcriptomic Trajectories

[0052] To understand the progression of mRNA regulation and the dynamics of transcriptomic trajectories within groups over time, we compared intra-group changes from baseline (hospital admission, 0-hr) abundances. Overall, alterations in gene expression show distinct group- and time-dependent patterns (FIG. 5A). The most favorable outcome group (G4) showed the least pronounced transcriptomic response without noticeable change over time. Patients who died quickly (G1) mount the largest transcriptome response. Upregulated DEGs in G1 peak at hour 12 (250 DEGs) and then steadily decline, while downregulated DEGs peak at hour 24 (190 DEGs) and then steadily decline before death. Patients who died later (G2), did not mount a large transcriptome response during early timepoints. However, at later timepoints there was significant upregulation of DEGs peaking at hour 108 (305 DEGs) followed by a steady decline before death. Lastly, patients who survive but have a prolonged recovery (G3) mount a transcriptomic response that gradually increases over time without a peak or dramatic decline.

[0053] To probe biological processes and functions, gene set enrichment analysis (GSEA) was performed on DEG sets from G3, the group with the most longitudinal data. GO terms persistently activated, beginning at early timepoints until day 21, include blood coagulation (GO:0007596), hemostasis (GO:0007599), and platelet aggregation (GO: 0070527). Cytokine production (GO:0001817) and leukocyte-mediated immunity (GO:0002443) pathways are initially upregulated, then downregulated by hour 24. These immune-related pathways remaindownregulated until day 21. GO terms that are never upregulated and are downregulated by hour 48 until day 21 include response to type II interferon (GO: 0034341), MHC class II protein assembly (GO: 0002399), and NF-kappaβ signaling (GO: 0038061). (See FIGS. 6-10 for longitudinal GSEA data for G1, G2, and G4).

[0054] Concordance / relationship with findings from other molecular layers. To better visualize the relationship between complement activation’s role in immune dysregulation we constructed a network from significant DEGs at hour 0 between groups. All references cited herein are incorporated by reference in their entirety. References 1. Qu, Z. and E.L. Chaikof, Interface between hemostasis and adaptive immunity. Curr Opin Immunol, 2010. 22(5): p. 634-42. 2. Ellis, S., E.J. Lin, and D. Tartar, Immunology of Wound Healing. Curr Dermatol Rep, 2018. 7(4): p. 350-358. 3. Al-Benna, S., Inflammatory and coagulative pathophysiology for the management of burn patients with COVID-19: systematic review of the evidence. Ann Burns Fire Disasters, 2021. 34(1): p. 3-9. 4. Burgess, M., et al., The Immune and Regenerative Response to Burn Injury. Cells, 2022. 11(19). 5. Kuypers, F.A., Hyperinflammation, apoptosis, and organ damage. Exp Biol Med (Maywood), 2022. 247(13): p. 1112-1123. 6. Pantalone, D., et al., The Role of DAMPS in Burns and Hemorrhagic Shock Immune Response: Pathophysiology and Clinical Issues. Review. Int J Mol Sci, 2021. 22(13). 7. DeJesus, J.E., J.J. Wen, and R. Radhakrishnan, Cytokine Pathways in Cardiac Dysfunction following Burn Injury and Changes in Genome Expression. J Pers Med, 2022. 12(11). 8. Vigiola Cruz, M., et al., Plasma Ameliorates Endothelial Dysfunction in Burn Injury. J Surg Res, 2019. 233: p. 459-466. 9. Sahib, A.S., F.H. Al-Jawad, and A.A. Al-Kaisy, Burns, endothelial dysfunction, and oxidative stress: the role of antioxidants. Ann Burns Fire Disasters, 2009. 22(1): p. 6-11. 10. Jeschke, M.G., et al., Burn injury. Nat Rev Dis Primers, 2020. 6(1): p. 11. 11. Kaddoura, I., et al., Burn injury: review of pathophysiology and therapeutic modalities in major burns. Ann Burns Fire Disasters, 2017. 30(2): p. 95-102. 12. Latenser, B.A., et al., National Burn Repository 2006: a ten-year review. J Burn Care Res, 2007. 28(5): p. 635-58. 13. Brusselaers, N., et al., Severe burn injury in Europe: a systematic review of the incidence, etiology, morbidity, and mortality. Crit Care, 2010. 14(5): p. R188. 14. Swanson, J.W., et al., Trajectories to death in patients with burn injury. J Trauma Acute Care Surg, 2013. 74(1): p. 282-8. 15. Hellenthal, K.E.M., L. Brabenec, and N.M. Wagner, Regulation and Dysregulation of Endothelial Permeability during Systemic Inflammation. Cells, 2022. 11(12).Gando, S., Microvascular thrombosis and multiple organ dysfunction syndrome. Crit Care Med, 2010. 38(2 Suppl): p. S35-42. Escandon-Vargas, K., et al., Healthcare-associated infections in burn patients: Timeline and risk factors. Burns, 2020. 46(8): p. 1775-1786. Tejiram, S. and J.W. Shupp, Sepsis in surgical patients: Burn sepsis. Surgery, 2023. Wibbenmeyer, L., et al., Prospective analysis of nosocomial infection rates, antibiotic use, and patterns of resistance in a burn population. J Burn Care Res, 2006. 27(2): p. 152-60. Kankam, H.K.N., et al., Acute burn injuries associated with long-term mortality: a systematic review and meta-analysis. Burns, 2022. 48(8): p. 1783-1793. Williams, F.N., D.N. Herndon, and M.G. Jeschke, The hypermetabolic response to burn injury and interventions to modify this response. Clin Plast Surg, 2009. 36(4): p. 583-96. Osborne, T., et al., Current understanding of the chronic stress response to burn injury from human studies. Burns Trauma, 2023. 11: p. tkad007. Bryarly, J. and K. Kowalske, Long-Term Outcomes in Burn Patients. Surg Clin North Am, 2023. 103(3): p. 505-513. Mulder, P.P.G., et al., Burn-Induced Local and Systemic Immune Response: Systematic Review and Meta-Analysis of Animal Studies. J Invest Dermatol, 2022. 142(11): p. 3093- 3109 e15. Gibson, B.H.Y., et al., Plasmin drives burn-induced systemic inflammatory response syndrome. JCI Insight, 2021. 6(23). Pileri, D., et al., Concentrations of cytokines IL-6 and IL-10 in plasma of burn patients: their relationship to sepsis and outcome. Ann Burns Fire Disasters, 2008. 21(4): p. 182-5. Sapan, H.B., et al., Pattern of cytokine (IL-6 and IL-10) level as inflammation and anti- inflammation mediator of multiple organ dysfunction syndrome (MODS) in polytrauma. Int J Burns Trauma, 2016. 6(2): p. 37-43. Hur, J., et al., Inflammatory cytokines and their prognostic ability in cases of major burn injury. Ann Lab Med, 2015. 35(1): p. 105-10. Bergquist, M., et al., The time-course of the inflammatory response to major burn injury and its relation to organ failure and outcome. Burns, 2019. 45(2): p. 354-363. Lowe, R., et al., Transcriptomics technologies. PLoS Comput Biol, 2017. 13(5): p. e1005457. Schena, M., et al., Microarrays: biotechnology's discovery platform for functional genomics. Trends Biotechnol, 1998. 16(7): p. 301-6. Xiao, W., et al., A genomic storm in critically injured humans. J Exp Med, 2011. 208(13): p. 2581-90. Keyloun, J.W., et al., Early Transcriptomic Response to Burn Injury: Severe Burns Are Associated With Immune Pathway Shutdown. J Burn Care Res, 2022. 43(2): p. 306-314. Shupp, J.W., et al., Assessment of Coagulation Homeostasis in Blunt, Penetrating, and Thermal Trauma: Guidance for a Multicenter Systems Biology Approach. Shock, 2019. 52(1S Suppl 1): p. 84-91. Ritchie, M.E., et al., limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res, 2015. 43(7): p. e47. Wu, T., et al., clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb), 2021. 2(3): p. 100141.Boyle, E.I., et al., GO::TermFinder--open source software for accessing Gene Ontology information and finding significantly enriched Gene Ontology terms associated with a list of genes. Bioinformatics, 2004. 20(18): p. 3710-5. Subramanian, A., et al., Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A, 2005. 102(43): p. 15545-50. Calvano, S.E., et al., A network-based analysis of systemic inflammation in humans. Nature, 2005. 437(7061): p. 1032-7. Breslin, J.W., et al., Toll-like receptor 4 contributes to microvascular inflammation and barrier dysfunction in thermal injury. Shock, 2008. 29(3): p. 349-55. Tsujimoto, H., et al., Neutrophil elastase, MIP-2, and TLR-4 expression during human and experimental sepsis. Shock, 2005. 23(1): p. 39-44. Keyloun, J.W., et al., Circulating Syndecan-1 and Tissue Factor Pathway Inhibitor, Biomarkers of Endothelial Dysfunction, Predict Mortality in Burn Patients. Shock, 2021. 56(2): p. 237-244. Ball, R.L., et al., Burn-Induced Coagulopathies: a Comprehensive Review. Shock, 2020. 54(2): p. 154-167. Roche, P.A. and K. Furuta, The ins and outs of MHC class II-mediated antigen processing and presentation. Nat Rev Immunol, 2015. 15(4): p. 203-16. Wieczorek, M., et al., Major Histocompatibility Complex (MHC) Class I and MHC Class II Proteins: Conformational Plasticity in Antigen Presentation. Front Immunol, 2017. 8: p. 292. Wolk, K., et al., Comparison of monocyte functions after LPS- or IL-10-induced reorientation: importance in clinical immunoparalysis. Pathobiology, 1999. 67(5-6): p. 253-6. Barrett, L.W., et al., Understanding acute burn injury as a chronic disease. Burns Trauma, 2019. 7: p. 23. Nunez, J.H. and A.T. Clark, Burn Patient Metabolism and Nutrition. Phys Med Rehabil Clin N Am, 2023. 34(4): p. 717-731. Duke, J.M., et al., A retrospective cohort study to compare post-injury admissions for infectious diseases in burn patients, non-burn trauma patients and uninjured people. Burns Trauma, 2018. 6: p. 17. Meizoso, J.P., et al., Hypercoagulability and venous thromboembolism in burn patients. Semin Thromb Hemost, 2015. 41(1): p. 43-8. Sood, R.F., et al., Early leukocyte gene expression associated with age, burn size, and inhalation injury in severely burned adults. J Trauma Acute Care Surg, 2016. 80(2): p. 250-7.

Claims

CLAIMS:

1. A method of predicting a clinical outcome of a burn patient, comprising the steps of: a) determining the mRNA expression levels of at least 65 genes of a set of biomarkers from a whole blood sample obtained from the burn patient under conditions to preserve mRNA transcripts, wherein the at least 65 genes is selected from a set of biomarkers consisting of: ACSM3, AKR1C3, ANAPC10, B3GNT5, BATF, BCL2A1, C1orf162, CD93, CLDN14, CLEC2B, CREM, CSTA, DHCR7, DIP2A, DNAJB1, DGKQ, DOCK11, DYNLT3, FAM107A, FAM24A, FCN2, FOS, GLI1, G0S2, GPR84, HBE1, HSP90AA1, HSPA1A, HSPA1B, HSPA9, HSPH1, HSPD1, HSPE1, IL18R1, IRAK3, JOSD1, KBTBD11, KIR2DL5A, KIR2DS4, KLF3, KLRF1, METRNL, MRPS30, NIBAN2, NR4A2, OR4D6, PFKFB3, PIK3R1, PIP5K1A, PLA2G7, PLEK, PRPSAP2, PRRG2, RAMP1, RETN, RGS1, RHOXF2, RNASE2, SH2D2A, SHOC1, SLA2, SLC25A16, SLC45A3, SOCS3, SYAP1, TAMALIN, TBC1D19, THBS1, TLR4, TNFAIP3, TNFSF13B, TOP1MT, TRIB1, USP30, and ZRSR2; b) generating four probability scores for a likelihood of the clinical outcome for (i) early mortality (G1), (ii) later mortality (G2), (iii) prolonged recovery (G3), and (iv) early discharge (G4), wherein each of the four probability scores are calculated using the formula: logit(P) =C0+ C1X1+ C2X2+ ··· + CnXnwhere logit() is the log odds of the probability value P, which is the probability of one of the outcome group, C0 is the intercept value, C1 through Cn are weighted coefficients for each of the 65 genes for each clinical outcome as defined in Table 3, and X1through Xnare the mRNA expression levels of the at least 65 genes, and wherein the highest probability score of G1, G2, G3 and G4 is assigned as the predicted clinical outcome of the burn patient.

2. The method of the previous claim, wherein the method generates a report describing the predicted clinical outcome.

3. The method of either of the previous claims, wherein the method predicts a longer hospital stay for the burn patient.

4. The method of any one of the previous claims, wherein the mRNA expression is determined by a nucleic acid sequencing assay, a next generation nucleic acid sequencing (NGS) assay, a Sanger sequencing assay, a PCR assay, a quantitative PCR (qPCR) assay, a reverse transcription PCR (RT- PCR) assay, a mRNA assay, a microarray assay, a Northern blot assay, a Southern blot assay, a luciferase assay, a fluorescence immunoassay.

5. The method of any one of the previous claims, wherein the method is performed within the first 24 hours after a burn injury (“initial draw”).

6. The method of the previous claim, wherein said method is performed every 2 hours for 72 hours after the initial draw.

7. The method of either one of the two previous claims, wherein the method is repeated 2, 4, 8, 12, and 24 hours after the initial draw.

8. The method of any one of the previous claims, wherein said method is repeated every 12 hours for 72 hours.

9. The method of any one of the previous three claims, wherein said highest probability scores are combined to refine an assessment of clinical outcome.

10. The method of any one of the previous claims, wherein the method comprises determining the mRNA expression levels of at least 66, 67, 68, 69, 70, 71, 72, 73, 74, or 75 genes of the set of biomarkers.

11. The method of any one of the previous claims, wherein logit(P) empirically determines the area under the curve (AUC) of Receiver operating characteristic (ROC) curve.

12. The method of claim 12, wherein each of the four binary classifier models has a performance of the ROC curve with a sensitivity value of at least 0.8.

13. The method of either one of the two previous claims, wherein each of the four classifier models has a performance of the ROC curve with a specificity value of at least 0.

65.

14. The method of any of the previous claims, wherein said report is communicated to a health care provider.

15. The method of any of the previous claims wherein the method further comprises administering an antibiotic, intravenous hydration, transfusion of blood products, a vasopressor, ventilator assistance, a non-steroidal anti-inflammatory agent, or an anti- pyretic agent.

16. The method of any one of the previous claims, wherein said method is carried out by a computer-implemented system comprising at least one processor and at least one memory.

17. The method of the previous claim, wherein said memory comprises instructions executed by the at least one processor to cause the at least one processor to implement a predictive model to predict the clinical outcome of the burn subject.

18. The method of either one of the previous claims, wherein said computer-implemented system is a hand-held device.

19. The method of any one of the previous claims, wherein the method is carried out at the burn patient’s bedside.

20. A kit comprising polynucleotides for carrying out the method of any one of the previous claims, wherein each polynucleotide specifically detects a gene in the biomarker set.

21. The kit of the previous claim, wherein said polynucleotide is a probe that specifically binds to a gene in the biomarker set.

22. The kit of either claim 20 or 21, wherein said polynucleotide is a primer that specifically amplifies a gene in the biomarker set.

23. The kit of any one of claims 20-22, wherein the probes or primers are provided on a solid substrate.

Images