WO2018223124A2

WO2018223124A2 - Predictive factors for successful wound closure

Info

Publication number: WO2018223124A2
Application number: PCT/US2018/035841
Authority: WO
Inventors: Eric A. ELSTER; Seth A. SCHOBEL-MCHUGH; Beverly J. GAUCHER
Original assignee: The Henry M. Jackson Foundation For The Advancement Of Military Medicine, Inc.
Priority date: 2017-06-02
Filing date: 2018-06-04
Publication date: 2018-12-06
Also published as: WO2018223124A3

Abstract

The present invention relates to methods of determining if a subject has an increased risk of developing unsuccessful wound closure (UWC) prior to the onset of any detectable symptoms thereof. The methods comprise analyzing at least one sample from the subject to determine a value of the subject's risk profile and comparing the value of the subject's risk profile with the value of a normal risk profile. A change in the value of the subject's risk profile, over or under normal values is indicative that the subject has an increased risk of having or developing symptoms associated with UWC prior to the onset of any detectable symptoms thereof.

Description

PREDICTIVE FACTORS FOR SUCCESSFUL WOUND CLOSURE

Statement Regarding Federally Sponsored Research or Development

[0001] This invention was made with government support under HT9404-13-1 and HU0001-15-2- 0001 awarded by The Department of Defense. The government has certain rights in the invention.

Background of the Invention

Field of the Invention

[0002] The present invention relates to methods of determining if a subject has an increased risk of developing unsuccessful wound closure (UWC) prior to the onset of any detectable symptoms thereof. The methods comprise analyzing at least one sample from the subject to determine a value of the subject's risk profile and comparing the value of the subject's risk profile with the value of a normal risk profile. A change in the value of the subject's risk profile, over or under normal values is indicative that the subject has an increased risk of having or developing symptoms associated with UWC prior to the onset of any detectable symptoms thereof.

Background of the Invention

[0003] The time of closure of combat-related extremity wounds is planned based on the extent of injury and assessment of local signs of healing such as the absence of tissue necrosis and resolution of infection after multiple surgical debridements (SD). Approximately 80% of these wounds are successfully closed without subsequent complications such as wound failure. Successful wound healing is associated with a distinct local and systemic cytokine response. The present disclosure investigates if the time of unsuccessful wound closure (UWC) could be estimated based on the cytokine response, traumatic wound characteristics, and occurrence of critical colonization (CC) associated with these injuries.

Summary of the Invention

[0004] The present invention relates to methods of determining if a subject has an increased risk of developing unsuccessful wound closure (UWC) prior to the onset of any detectable symptoms thereof. The methods comprise analyzing at least one sample from the subject to determine a value of the subject's risk profile and comparing the value of the subject's risk profile with the value of a normal risk profile. A change in the value of the subject's risk profile, over or under normal values is indicative that the subject has an increased risk of having or developing symptoms associated with UWC prior to the onset of any detectable symptoms thereof. Brief Description of the Drawings

[0005] Figure 1 depicts Debridement 2 Reduced Model ROC

[0006] Figure 2 depicts Debridement 2 Reduced Model Missing Values

[0007] Figure 3 depicts Debridement 2 Comparative Decision Curves

[0008] Figure 4 depicts Debridement 2 Example Kaplan-Meier Plot

[0009] Figure 5 depicts Debridement 3 Reduced Model ROC

[0010] Figure 6 depicts Debridement 3 Reduced Model Missing Values

[0011] Figure 7 depicts Debridement 3 Comparative Decision Curves

[0012] Figure 8 depicts Debridement 3 Example Kaplan-Meier Plot

[0013] Figure 9 depicts Debridement 4 Reduced Model ROC

[0014] Figure 10 depicts Debridement 4 Reduced Model Missing Values

[0015] Figure 11 depicts Debridement 4 Comparative Decision Curves

[0016] Figure 12 depicts Debridement 4 Example Kaplan-Meier Plot

Detailed Description of the Invention

[0017] The present invention relates to methods of determining if a su bject has an increased risk of developing unsuccessful wound closure (UWC) prior to the onset of any detectable symptoms thereof. The methods comprise analyzing at least one sample from the subject to determine a value of the subject's risk profile and comparing the value of the subject's risk profile with the value of a normal risk profile. A change in the value of the subject's risk profile, over or under normal values is indicative that the subject has an increased risk of having or developing symptoms associated with UWC prior to the onset of any detectable symptoms thereof.

[0018] The present invention also relates to methods of detecting elevated levels of a specific collection of analytes in one or more samples obtained from a subject. In one embodiment, the collection of analytes comprises wound effluent levels FGF-Basic, serum levels VEG F, wound effluent levels I FNy, serum levels RANTES, wound effluent levels I P10, serum levels MCP1, serum levels IFNy, wound effluent levels I LIB, serum levels I L2R and wound effluent levels I L8. [0019] As used herein, the term "subject" or "test subject" indicates a mammal, in particular a human or non-human primate. The test subject is in need of an assessment of susceptibility of UWC. For example, the test subject may have no symptoms that UWC may occur.

[0020] The term unsuccessful wound closure, or UWC, is used herein to mean either evidence of complete or partial wound disruption after wound closure, and/or greater than 10% loss of skin graft.

[0021] As used herein, the term means "increased risk" is used to mean that the test subject has an increased chance of developing UWC compared to a normal individual. The increased risk may be relative or absolute and may be expressed qualitatively or quantitatively. For example, an increased risk may be expressed as simply determining the subject's risk profile and placing the patient in an "increased risk" category, based upon previous population studies. Alternatively, a numerical expression of the subject's increased risk may be determined based upon the risk profile. As used herein, examples of expressions of an increased risk include but are not limited to, odds, probability, odds ratio, p-values, attributable risk, biomarker index score, relative frequency, positive predictive value, negative predictive value, and relative risk.

[0022] For example, the correlation between a subject's risk profile and the likelihood of developing UWC may be measured by an odds ratio (OR) and by the relative risk ( RR). If P(R⁺) is the probability of developing UWC for individuals with the risk profile (R) and P(R^") is the probability of developing UWC for individuals without the risk profile, then the relative risk is the ratio of the two probabilities: RR=P(R⁺)/P(R ).

[0023] In case-control studies, however, direct measures of the relative risk often cannot be obtained because of sampling design. The odds ratio allows for an approximation of the relative risk for low-incidence diseases and can be calculated: OR=(F⁺/( l-F⁺))/(F^"/(l-F^")), where F⁺ is the frequency of a risk profile in cases studies and F^" is the frequency of risk profile in controls. F⁺ and F^" can be calculated using the risk profile frequencies of the study.

[0024] The attributable risk (AR) can a lso be used to express an increased risk. The AR describes the proportion of individuals in a population exhibiting UWC to a specific member of the risk profile. AR may also be important in quantifying the role of individual components (specific member) in condition etiology and in terms of the public health impact of the individual risk factor. The public health relevance of the AR measurement lies in estimating the proportion of cases of UWC in a population of wounded subjects that could be prevented if the profile or individual factor were absent. AR may be determined as follows: AR=P_e(RR-1)/(PE(RR-1)+1), where AR is the risk attributable to a profile or individual factor of the profile, and PE is the frequency of exposure to a profile or individual component of the profile within the population at large. RR is the relative risk, which can be approximated with the odds ratio when the profile or individual factor of the profile under study has a relatively low incidence in the general population.

[0025] In one embodiment, the increased risk of a patient can be determined from p-values that are derived from association studies. Specifically, associations with specific profiles can be performed using regression analysis by regressing the risk profile with the presence or absence of UWC. In addition, the regression may or may not be corrected or adjusted for one or more factors. The factors for which the analyses may be adjusted include, but are not limited to age, sex, weight, ethnicity, type of wound, number of wounds, number of days from injury, number of previous surgical debridements, geographic location, fasting state, state of pregnancy or post-pregnancy, menstrual cycle, general health of the subject, alcohol or drug consumption, caffeine or nicotine intake and circadian rhythms, to name a few.

[0026] Increased risk can also be determined from p-values that are derived using logistic regression. Binomial (or binary) logistic regression is a form of regression which is used when the dependent is a dichotomy and the independents are of any type. Logistic regression can be used to predict a dependent variable on the basis of continuous or categorical or both (continuous and categorical) independents and to determine the percent of variance in the dependent variable explained by the independents; to rank the relative importance of independents; to assess interaction effects; and to understand the impact of covariate control variables. Logistic regression applies maximum likelihood estimation after transforming the dependent into a "logit" variable (the natural log of the odds of the dependent occurring or not). In this way, logistic regression estimates the probability of a certain event occurring. These analyses may be conducted with virtually any statistics program, such as not limited to SAS, R package available through CRAN repository.

[0027] SAS ("statistical analysis software") is a general purpose package (similar to Stata and SPSS) created by Jim Goodnight and N .C. State University colleagues. Ready-to-use procedures handle a wide range of statistical analyses, including but not limited to, analysis of variance, regression, categorical data analysis, multivariate analysis, survival analysis, psychometric analysis, cluster analysis, and nonparametric analysis. R package is free, general purpose package that complies with and runs on a variety of U N IX platforms. [0028] Accordingly, select embodiments of the present invention comprise the use of a computer comprising a processor and the computer is configured or programmed to generate one or more risk profiles and/or to determine statistical risk. The methods may also comprise displaying the one or risk profiles on a screen that is communicatively connected to the computer. In another embodiment, two different computers can be used: one computer configured or programmed to generate one or more risk profiles and a second computer configured or programmed to determine statistical risk. Each of these separate computers can be communicatively linked to its own display or to the same display.

[0029] As used herein, the phrase "risk profile" means the combination of a subject's risk factors analyzed or observed. The terms "factor" and/or "component" are used to mean the individual constituents that are assessed when generating the profile. The risk profile is a collection of measurements, such as but not limited to a quantity or concentration, for individual factors taken from a test sample of the subject. Examples of test samples or sources of components for the risk profile include, but are not limited to, biological fluids, which can be tested by the methods of the present invention described herein, and include but are not limited to whole blood, such as but not limited to peripheral blood, serum, plasma, cerebrospinal fluid, wound effluent, urine, amniotic fluid, lymph fluids, various external secretions of the respiratory, intestinal and genitourinary tracts, tears, saliva, white blood cells, myelomas and the like.

[0030] The risk profile can include a "biological effector" aspect and/or a non-biological effector aspect. As used herein, the term "biological effector" is used to mean a molecule, such as but not limited to, a protein, peptide, a carbohydrate, a fatty acid, a nucleic acid, a glycoprotein, a proteoglycan, etc. that can be assayed. Specific examples of biological effectors can include, cytokines, growth factors, antibodies, hormones, cell surface receptors, cell surface proteins, carbohydrates, etc. More specific examples of biological effectors include interleukins (I Ls) such as I L-la, IL-Ιβ, I L-1 receptor antagonist (I L-1RA), I L-2, I L-2 receptor (I L-2R), I L-3, I L-4, I L-5, I L-6, I L-7, I L-8, I L-10, I L-12, I L-13, I L-15, I L-17, as well as growth factors such as tumor necrosis factor alpha (TN Fa), granulocyte colony stimulating factor (G-CSF), granulocyte macrophage colony stimulating factor (GM-CSF), interferon alpha (I N F-a), interferon gamma (I FN-y), epithelial growth factor (EGF), basic endothelial growth factor (bEGF), hepatocyte growth factor (HGF), vascular endothelial growth factor (VEGF), and chemokines such as monocyte chemoattractant protein-1 (CCL2/MCP-1), macrophage inflammatory protein-1 alpha (CCL3/M I P-la), macrophage inflammatory protein-1 beta (Ο- /Μ Ι Ρ-Ιβ), CCL5/RANTES, CCLll/eotaxin, monokine induced by gamma interferon (CXCL9/MIG) and interferon gamma-induced protein-10 (CXCL10/I P10). [0031] Techniques to assay levels of individual components of the risk profile from test samples are well known to the skilled technician, and the invention is not limited by the means by which the components are assessed. In one embodiment, levels of the individual factors in the serum or wound effluent of the risk profile are assessed using mass spectrometry in conjunction with ultra- performance liquid chromatography (U PLC), high-performance liqu id chromatography (H PLC), gas chromatography (GC), gas chromatography/mass spectroscopy (GC/MS), and UPLC to name a few. Other methods of assessing levels of some of the individual components include biological methods, such as but not limited to ELISA assays, Western Blot and multiplexed immunoassays etc. Other techniques may include using quantitative arrays, PCR, Northern Blot analysis. To determine levels of components or factors, it is not necessary that an entire component, e.g., a full length protein or an entire RNA transcript, be present or fully sequenced. In other words, determining levels of, for example, a fragment of protein being analyzed may be sufficient to conclude or assess that an individual component of the risk profile being analyzed is increased or decreased. Similarly, if, for example, arrays or blots are used to determine component levels, the presence/absence/strength of a detectable signal may be sufficient to assess levels of components.

[0032] As used herein, the term non-biological effector is a component that is generally considered not to be a specific molecule. Although not a specific molecule, a non-biological effector may nonetheless still be quantifiable, either through routine measurements or through measurements that stratify the data being assessed. For example, number or concentrate of red blood cells, white blood cells, platelets, coagulation time, blood oxygen content, etc. would be a non-biological effector component of the risk profile. All of these components are measureable or quantifiable using routine methods and equipment. Other non-biological components include data that may not be readily or routinely quantifiable or that may require a practitioner's judgment or opinion. For example, wound severity may be a component of the risk profile. While there may be published guidance on classifying wound severity, stratifying wound severity and, for example, assigning a numerical value to the severity, still involves observation and, to a certain extent, judgment or opinion. In some instances the quantity or measurement assigned to a non-biological effector could be binary, e.g., "0" if absent or "1" if present. In other instances, the non-biological effector aspect of the risk profile may involve qualitative components that cannot or should not be quantified.

[0033] In one embodiment, the mechanism of injury is included in the risk profile. As used herein, the phrase "mechanism of injury" means the manner in which the subject received an injury. For example, the mechanism of injury may be described as a gunshot wound, a vehicle accident, laceration, etc. In another embodiment, data regarding injury type is included in the risk profile. In another embodiment, data on the occurrence of multiple wounds is included in the risk profile. In another embodiment, data on the number of days from injury is included in the risk profile. In another embodiment, data on the number of surgical debridements (SDs).

[0034] Other examples of individual components of the risk profile include but are not limited to ISS score of head, ISS score of chest (thorax) and critical colonization. "ISS score" (injury severity score) is well-known in the art and is used routinely in clinics to assess severity of wounds or injuries.

[0035] As used herein, the term critical colonization (or "CC") is a measure of CFU that the subject has in serum and/or tissue for at least one wound when initially examined by the attending physician. For example, if a subject has CFU of lxlO⁵ per ml of serum, or if at least one wound has CFU of 1x10^s per mg of tissue, the subject is said to be "positive" for CC. If the total serum CFU or no single would has CFU of at least lxlO⁵ the subject is said to be "negative" for CC.

[0036] To determine which of the biological effector or non-biological effector components may be critical in the subjects' risk profiles, routine statistical methods can be employed. For example, rflmpute from the randomForest R package can be used to impute missing data. Up-sampling and predictor rank transformations can be performed on the data set only for variable selection to accommodate class imbalance and non-normality in the data.

[0037] For variable selection, the constraint-based algorithms /asi./amfa, iamb and gs and the constraint-based local discovery learning algorithms mmpc and si.hiton.pc from the "bnlearn" R package can be used to search the input dataset for nodes of Bayesian networks. The nodes can be chosen as the reduced variable sets. Before running the data through variable selection and binary classification algorithms, the variables may or may not randomly re-ordered. The data can be run through the variable selection and binary classification algorithms more than once, for example, 10, 20, 30, 40, 50 or even more times.

[0038] For binary classification and model selection, each variable set can be pulled from the raw data and run in sundry binary classification algorithms using the train function from the R caret package: linear discriminant analysis {Ida), classification and regression trees {cart), k-nearest neighbors {knn), support vector machine {svm), logistic regression {glm), random forest {rf), generalized linear models {glmnet) and naive Bayes {nb). The best variable set and binary classification algorithm combination that first produces the highest kappa and then the highest sensitivity with reasonable specificity can then be chosen. [0039] The resultant models are then examined using accuracy, no information rate, positive predictive value and negative predictive value. Model performance can be further assessed using the plot. roc command to compute the Receiver Operator Characteristic Curves (ROC) and area under curve (AUC). The dca R command from the Memorial Sloan Kettering Cancer Center website, www.mskcc.org, can be used to compute the Decision Curve Analysis (DCA).

[0040] Finally, for univariate analysis, a Wilcoxon rank-sum test can be used used to identify which biomarkers from specific patient groups are were associated with a specific indication.

[0041] The assessment of the levels of the individual components of the risk profile can be expressed as absolute or relative values and may or may not be expressed in relation to another component, a standard an internal standard or another molecule of compound known to be in the sample. If the levels are assessed as relative to a standard or internal standard, the standard may be added to the test sample prior to, during or after sample processing.

[0042] To assess levels of the individual components of the risk profile, a sample may be taken from the subject. The sample may or may not processed prior assaying levels of the components of the risk profile. For example, whole blood may be taken from an individual and the blood sample may be processed, e.g., centrifuged, to isolate plasma or serum from the blood. The sample may or may not be stored, e.g., frozen, prior to processing or analysis.

[0043] In one embodiment, the individual levels of each of the risk factors are higher or lower than those compared to normal levels. In another embodiment, levels of each of the factors are higher or lower than normal levels while others, if any, are lower than or the same as normal levels.

[0044] The levels of depletion of the factors or components compared to normal levels can vary. In one embodiment, the levels of any one or more of the factors or components is at least 1.05, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 higher than normal levels (where, for sake of clarity, a marker a level of "1" would indicate that the component is at the same level in both the subject and normal samples). For the purposes of the present invention, the number of "times" the levels of a factor are higher over normal can be a relative or absolute number of times. In the alternative, the levels of the factors or components may be normalized to a standard and these normalized levels can then be compared to one another to determine if a factor or component is lower, higher or about the same.

[0045] For the purposes of the present invention the risk profile comprises at least wound effluent FGF-Basic, serum VEGF, wound effluent I FNy, serum RANTES, wound effluent I P10, serum MCP1, serum IFNy, wound effluent I LIB, serum I L2R and wound effluent I L8 of factors or components for the prediction of UWC. If one factor or component of the biological effector aspect of the risk profile is used in generating the risk profile for the prediction of UWC, then any one of the listed factors or components can be used to generate the profile. If two factors or components of the biological effector aspect of the risk profile are used in generating the risk profile for the prediction of UWC, any combination of the two listed above can be used. If three factors or components of the biological effector aspect of the risk profile are used in generating the risk profile for the prediction of UWC, any combination of three of the factors or components listed above can be used. If four factors or components of the biological effector aspect of the risk profile are used in generating the risk profile for the prediction of UWC, any combination of four of the factors or components listed above can be used. If five factors or components of the biological effector aspect of the risk profile are used in generating the risk profile for the prediction of UWC, any combination of five of the factors or components listed above can be used. If six factors or components of the biological effector aspect of the risk profile are used in generating the risk profile for the prediction of UWC, any combination of six of the factors or components listed above can be used. Of course all members or components of the biological effector aspect of the risk profile can be used in generating the risk profile for the prediction of UWC.

[0046] The subject's risk profile is compared to the profile that is deemed to be a normal risk profile. To establish the risk profile of a normal individual, an individual or group of individuals may be first assessed to ensure they have no signs, symptoms or diagnostic indicators of UWC. Once established, the risk profile of the individual or group of individuals can then be determined to establish a "normal risk profile." In one embodiment, a normal risk profile can be ascertained from the same subject when the subject is deemed as healthy with no signs, symptoms or diagnostic indicators of UWC. In one embodiment, a risk profile from a "normal subject," e.g., a "normal risk profile," is a subject with a wound but did not exhibit or display UWC. In one specific embodiment, the normal subject has a chest wound, head wound or as extremity (arm, hand, finger(s), leg, foot, toe(s)) wound but did not exhibit UWC.

[0047] In one embodiment, a "normal" risk profile is assessed in the same subject from whom the sample is taken prior to the onset of any signs, symptoms or diagnostic indicators that they may exhibit UWC. That is, the term "normal" with respect to a risk profile can be used to mean the subject's baseline risk profile prior to the onset of any wounds, signs, symptoms or diagnostic indicators of potential UWC. The risk profile can then be reassessed periodically and compared to the subject's baseline risk profile. Thus, the present invention also includes methods of monitoring the progression of UWC in a subject, with the methods comprising determining the subject's risk profile at more than one time point. For example, some embodiments of the methods of the present invention will comprise determining the subject's risk profile at two, three, four, five, six, seven, eight, nine, 10 or even more time points over a period of time, such as a week or more, two weeks or more, three weeks or more, four weeks or more, a month or more, two months or more, three months or more, four months or more, five months or more, six months or more, seven months or more, eight months or more, nine months or more, ten months or more, 11 months or more, a year or more or even two years. The methods of monitoring a subject's risk of developing UWC would also include embodiments in which the subject's risk profile is assessed before and/or during and/or after treatment of UWC. In other words, the present invention also includes methods of monitoring the efficacy of treatment of UWC by assessing the subject's risk profile over the course of the treatment and after the treatment. In specific embodiments, the methods of monitoring the efficacy of treatment of UWC comprise determining the subject's risk profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points prior to the receipt of treatment for UWC and subsequently determining the subject's risk profile at at least one, two, three, four, five, six, seven, eight, nine or 10 or more different time points after beginning of treatment for UWC, and determining the changes, if any, in the risk profile of the subject. The treatment may be any treatment designed to cure, remove or diminish the likelihood of developing UWC.

[0048] In another embodiment, a normal risk profile is assessed in a sample from a different subject or patient (from the subject being analyzed) and this different subject does not have or is not suspected of developing UWC. In still another embodiment, the normal risk profile is assessed in a population of healthy individuals, the constituents of which display no signs, symptoms or diagnostic indicators that they may have or will develop UWC. Thus, the subject's risk profile can be compared to a normal risk profile generated from a single normal sample or a risk profile generated from more than one normal sample.

[0049] Of course, measurements of the individual components, e.g., concentration, ratio, log ratios etc., of the normal risk profile can fall within a range of values, and values that do not fall within this "normal range" are said to be outside the normal range. These measurements may or may not be converted to a value, number, factor or score as compared to measurements in the "normal range." For example, a measurement for a specific factor or component that is below the normal range, may be assigned a value or -1, -2, -3, etc., depending on the scoring system devised. [0050] In another embodiment, the measurements of the individual components themselves are used in the risk profile, and these levels can be used to provide a "binary" value to each component, e.g., "elevated" or "not elevated." Each of the binary values can be converted to a number, e.g., "1" or "0," respectively.

[0051] In one embodiment, the "risk profile value" can be a single value, number, factor or score given as an overall collective value to the individual components of the profile. For example, if each component is assigned a value, such as above, the component value may simply be the overall score of each individual or categorical value. For example, if five components of the risk profile for predicting UWC are used and three of those components are assigned values of "+2" and two are assigned values of the risk profile in this example would be +8, with a normal value being, for example, "0." In this manner, the risk profile value could be a useful single number or score, the actual value or magnitude of which could be an indication of the actual risk of developing UWC, e.g., the "more positive" the value, the greater the risk of developing UWC.

[0052] In another embodiment the "risk profile value" can be a series of values, numbers, factors or scores given to the individual components of the overall profile. In another embodiment, the "risk profile value" may be a combination of values, numbers, factors or scores given to individual components of the profile as well as values, numbers, factors or scores collectively given to a group of components, such as a biological effector portion. In another example, the risk profile value may comprise or consist of individual values, number or scores for specific component as well as values, numbers or scores for a group of components.

[0053] In another embodiment individual values from the risk profile and/or the mechanism of injury can be used to develop a single score, such as a "combined risk index," which may utilize weighted scores from the individual component values reduced to a diagnostic number value. The combined risk index may also be generated using non-weighted scores from the individual component values. When the "combined risk index" exceeds a specific threshold level, determined by a range of values developed similarly from control (normal) subjects, the individual has a high risk, or higher than normal risk, of developing UWC, whereas maintaining a normal range value of the "combined risk index" would indicate a low or minimal risk of developing UWC. In this embodiment, the threshold value would be or could be set by the combined risk index from one or more normal subjects. [0054] In another embodiment, the value of the risk profile can be the collection of data from the individual measurements and need not be converted to a scoring system, such that the "risk profile value" is a collection of the individual measurements of the individual components of the profile.

[0055] In specific embodiments, a subject is diagnosed of having an increased risk of suffering from UWC if the subject's seven, six, five, four, three, two or even one of the components or factors herein are at abnormal levels.

[0056] If it is determined that a subject has an increased risk of developing UWC, the attending health care provider may subsequently prescribe or institute a treatment program. In this manner, the present invention also provides for methods of treating individuals for UWC. The attending healthcare worker may begin treatment, based on the subject's risk profile, before there are perceivable, noticeable or measurable signs of UWC in the individual.

[0057] Accordingly, the invention provides methods of treating UWC in a subject in need thereof. The treatment methods include obtaining a subject's risk profile as defined herein and prescribing a treatment regimen to the subject if the risk profile indicates that the subject is at risk of developing UWC.

[0058] The methods of treatment also include methods of monitoring the effectiveness of a treatment for UWC. Once a treatment regimen has been established, with or without the use of the methods of the present invention to assist in a diagnosis of a risk of developing UWC, the methods of monitoring a subject's risk profile over time can be used to assess the effectiveness of treatments for UWC. Specifically, the subject's risk profile can be assessed over time, including before, during and after treatments for UWC. The risk profile can be monitored, with, for example, the normalization or decline in the values of the profile over time being indicative that the treatment may be showing efficacy of treatment.

[0059] The present invention also provides kits that can be used in the methods of the present invention. Specifically, the present invention provides kits for assessing the increased risk of developing UWC, with the kits comprising one or more sets of antibodies that are immobilized onto a solid substrate and specifically bind to at least one of the factors or components listed herein. In specific embodiments, the kits comprise at least two, three, four, five, six or seven sets of antibodies immobilized onto a solid substrate, with each set corresponding to a factor.

[0060] The antibodies that are immobilized onto the substrate may or may not be labeled. For example, the antibodies may be labeled, e.g., bound to a labeled protein, in such a manner that binding of the specific protein may displace the label and the presence of the marker in the sample is marked by the absence of a signal. In addition, the antibodies that are immobilized onto the substrate may be directly or indirectly immobilized onto the surface. Methods for immobilizing proteins, including antibodies, are well-known in the art, and such methods may be used to immobilize a target protein, e.g., I L-10, or another antibody onto the surface of the substrate to which the antibody directed to the specific factor can then be specifically bound. In this manner, the antibody directed to the specific biomarker is immobilized onto the surface of the substrate for the purposes of the present invention.

[0061] The kits of the present invention may or may not include containers for collecting samples from the subject and one or more reagents, e.g., purified target biomarker for preparing a calibration curve. The kits may or may not include additional reagents such as wash buffers, labeling reagents and reagents that are used to detect the presence (or absence) of the label.

[0062] All patents and publications cited herein are incorporated by reference to the same extent as if each individual publication was specifically and individually indicated as having been incorporated by reference in its entirety.

Examples

[0063] Treatment of 73 patients with 116 combat related extremity wounds was evaluated for which the injuries involved multiple debridement operations before definitive closure of each wound. In each debridement, samples of serum and wound effluent were collected and tested for the level of 32 cytokines associated with wound healing. The method combined three statistical techniques: ( 1) the max-min parents and children algorithm to calculate the nodes of Bayesian Networks as a reduced variable set; (2) a random forest for classification of dehiscence and healing; and (3) the Kaplan-Meier curve for visualization of joint proba bilities of closure and successful or unsuccessful wound healing.

[0064] Random forest models were trained only on debridements with both successful and unsuccessful wound healing. A Kaplan-Meier curve was computed for each debridement. The random forest predictions and the Kaplan-Meier curves were multiplied by each other to produce a scaled Kaplan-Meier curve of each wound's present and future joint probabilities of closure and successful or unsuccessful wound healing.

[0065] Cross-validated models trained using reduced variable sets calculated on each debridement classifying both successful and unsuccessful wound closure had accuracies of 0.79 to 0.9, Kappas of 0.35-0.7, sensitivities of 0.42 to 0.89 and specificities of 0.9 to 0.96. The scaled Kaplan-Meier curve analysis produced a graphic of estimated outcomes. Decision curve analysis confirmed reduced models outperform, or perform similarly to, models trained with the full variable set.

[0066] Combined multivariate analyses may be an effective way to evaluate healing of combat- related extremity wounds. This work highlights a potential approach to developing a clinical decision support tool (CDST) for clinician-driven interactive tools. Prospective developments include other clinical outcomes, more data and external validation sets.

Claims

What is Claimed is:

1. A method of determining the approximate time of wound healing in a subject with a wound, the method comprising a) analyzing at least one sample from the subject to determine a value of the subject's risk profile for developing unsuccessful wound closure (UWC) of the wound, wherein the risk profile comprises wound effluent levels FGF-Basic, serum levels VEGF, wound effluent levels IFNy, serum levels RANTES, wound effluent levels IP10, serum levels MCP1, serum levels IFNy, wound effluent levels I LIB, serum levels IL2R and wound effluent levels IL8. and b) comparing the value of the subject's risk profile a normal risk profile to determine if the subject's risk profile is altered compared to a normal risk profile, wherein an increase in the value of the subject's risk profile is indicative that the subject has an increased risk of developing UWC compared to individuals with a normal risk profile.

2. The method of claim 1, wherein the at least one sample is a serum sample obtained from the subject.

3. The method of claim 1, wherein the at least one sample is a wound effluent sample obtained from the subject.

4. The method of any of claims 1-3, wherein the normal risk profile comprises a risk profile generated from a population of individuals that did not exhibit UWC.

5. A method of detecting elevated levels of biomarkers in a subject with a wound, the method comprising measuring levels of one or more wound effluent levels FGF-Basic, serum levels VEGF, wound effluent levels IFNy, serum levels RANTES, wound effluent levels IP10, serum levels MCP1, serum levels I FNy, wound effluent levels IL1B, serum levels IL2R and wound effluent levels IL8. .

6. A method of treating a subject for unsuccessful wound closure (UWC), the method

comprising a) assessing a risk profile comprising individual risk factors: wherein the risk factors

comprise wound effluent levels FGF-Basic, serum levels VEGF, wound effluent levels IFNy, serum levels RANTES, wound effluent levels IP10, serum levels MCP1, serum levels IFNy, wound effluent levels I LIB, serum levels IL2R and wound effluent levels IL8. and b) administering a treatment for UWC to the subject when the risk profile for the subject is greater than the risk profile of a normal subject.

7. The method of claim 6, wherein the risk profile is assessed from a serum sample from the subject.

8. The method of claim 6, wherein the risk profile is assessed from a wound effluent sample from the subject.

9. The method of claim 7 or 8, wherein the treatment is administered to the subject prior to the onset of any detectable symptoms of the subject exhibiting UWC.

10. A kit for performing the methods of any of the preceding claims.