METHODS AND SYSTEMS FOR PREDICTING BLEEDING RISK AND DOSE OF PLASMINOGEN ACTIVATOR

Abstract

The present disclosure provides a method and system for estimating the clinical responsiveness of a patient to a dose of a plasminogen activating agent to treat a thrombosis, comprising determining a concentration of α2-antiplasmin in a blood sample of the patient, determining a concentration of activated fibrinolysis inhibitor (“TAFI”) in the blood sample, determining a concentration of plasminogen activator Inhibitor 1 (“PAI-1”) in the blood sample, computing a clot lysis time (“CLT”) based on the concentrations of a2-antiplasmin, TAFI and PAI-1 using the equation CLT=−2,813.6+31.1*a2-antiplasmin (percent activity)+31.1*TAFI (percent activity)+1.49 PAI-1 (ug/L), and determining that the patient is at increased risk of hemorrhage when the computed CLT is less than a first predetermined cutoff time.

Claims

1. A method for estimating the clinical responsiveness of a patient to a dose of a plasminogen activating agent to treat a thrombosis, comprising: determining, using at least one biomarker measurement system, a concentration of α2-antiplasmin in a blood sample of the patient; determining, using the at least one biomarker measurement system, a concentration of activated fibrinolysis inhibitor (“TAFI”) in the blood sample; determining, using the at least one biomarker measurement system, a concentration of plasminogen activator Inhibitor 1 (“PAI-1”) in the blood sample; computing, using a computing device, a clot lysis time (“CLT”) based on the concentrations of α2-antiplasmin, TAFI and PAI-1 using the equation CLT=−2,813.6+31.1*α2-antiplasmin (percent activity)+31.1*TAFI (percent activity)+1.49 PAI-1 (ug/L); and determining, using the computing device, that the patient is at increased risk of hemorrhage when the computed CLT is less than a first predetermined cutoff time.

2. The method of claim 1 wherein the first predetermined cutoff time is approximately 4,926 seconds.

3. The method of claim 1 further including determining, using the computing device, that the patient is at an increased risk of clinical failure with treatment with the plasminogen activating agent when the computed CLT is greater than a second predetermined cutoff time, the second predetermined cutoff time being greater than the first predetermined cutoff time.

4. The method of claim 3 wherein the second predetermined cutoff time is approximately 15,247 seconds.

5. The method of claim 1 wherein the concentrations of α2-antiplasmin and TAFI are percentages of a normative value.

6. The method of claim 1 wherein the at least one biomarker measurement system includes a chromogenic assay for determining the concentration of α2-antiplasmin, TAFI and PAI-1 in the blood sample.

7. The method of claim 1 further including responding to the computed CLT being less than the first predetermined cutoff time by providing a reduced dose plasminogen activator fibrinolytic treatment to the patient.

8. A system for estimating the clinical responsiveness of a patient to administration of a plasminogen activating agent to treat a thrombosis, comprising: at least one biomarker measurement system for determining a concentration of α2-antiplasmin in a blood sample of the patient, a concentration of activated fibrinolysis inhibitor (“TAFI”) in the blood sample, and a concentration of plasminogen activator Inhibitor 1 (“PAI-1”) in the blood sample; a computing device in communication with the at least one biomarker measurement system including a processor and a memory including instructions which when executed by the processor cause the computing device to compute a clot lysis time (“CLT”) based on the concentrations of α2-antiplasmin, TAFI and PAI-1 using the equation CLT=−2,813.6+31.1*α2-antiplasmin (percent activity)+31.1*TAFI (percent activity)+1.49 PAI-1 (ug/L), and to compare the computed CLT to a first predetermined cutoff time; and a user interface configured to provide a user of the computing device information that the patient is at increased risk of hemorrhage when the computed CLT is less than a first predetermined cutoff time.

9. The system of claim 8 wherein the first predetermined cutoff time is approximately 4,926 seconds.

10. The system of claim 8 wherein the user interface is further configured to provide the user information that the patient is at an increased risk of clinical failure with treatment with the plasminogen activating agent when the computed CLT is greater than a second predetermined cutoff time, the second predetermined cutoff time being greater than the first predetermined cutoff time.

11. The system of claim 10 wherein the second predetermined cutoff time is approximately 15,247 seconds.

12. The system of claim 8 wherein the concentrations of α2-antiplasmin and TAFI are percentages of a normative value.

13. The system of claim 8 wherein the at least one biomarker measurement system includes a chromogenic assay for determining the concentration of α2-antiplasmin, TAFI and PAI-1 in the blood sample.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above-mentioned and other features of this disclosure and the manner of obtaining them will become more apparent and the disclosure itself will be better understood by reference to the following description of embodiments of the present disclosure taken in conjunction with the accompanying drawings, wherein:

[0047] FIG. 1 is a chart depicting results of an experimental method for measuring clot lysis time (“CLT”) using turbidimetry;

[0048] FIG. 2 is a chart depicting results of an experimental method for measuring CLT using thromboeslastography (“TEG”);

[0049] FIG. 3 is a chart depicting a characteristic curve demonstrating normal (solid line) and resistance to fibrinolysis (dotted line) using turbidimetry;

[0050] FIG. 4 is a chart depicting a characteristic curve demonstrating normal (solid line) and resistance to fibrinolysis (dotted line) using TEG;

[0051] FIG. 5 is a Dot plot of CLT values for each patient in four groups;

[0052] FIG. 6 is a chart depicting a receiver operating characteristic curve using an equation according to the principles of the present disclosure; and

[0053] FIG. 7 is a block diagram of a system for carrying out the methods of the present disclosure.

[0054] While the present disclosure is amenable to various modifications and alternative forms, specific embodiments have been shown by way of example in the drawings and are described in detail below. The present disclosure, however, is not to limit the particular embodiments described. On the contrary, the present disclosure is intended to cover all modifications, equivalents, and alternatives falling within the scope of the appended claims.

DETAILED DESCRIPTION

[0055] The following detailed description of the embodiment(s) is merely exemplary in nature and is in no way intended to limit the invention, its application, or uses.

[0056] The present disclosure is directed to a computer-based device with a predictive model executable by computer software for estimating the clot lysis time associated with a plasminogen activating agent. The computer-based device typically comprises a processor, computer hardware such as a computer screen or keyboard or disk drive, a computer software application, computer memory, data storage modules and input/output devices. The computer memory comprises instructions executable by the processor to run the predictive model of the present disclosure. The computer device may be communicatively connected to a computer network.

[0057] The predictive model of the present disclosure comprises a multivariate equation. In one embodiment, the following equation that is the basis of the predictive model where the dependent variable is clot lysis time (“CLT”), which is a surrogate marker for effectiveness of plasminogen activation on clot dissolution and determined from thromboelastography:

CLT=−2,813.6+31.1*α2-antiplasmin(percent activity)+31.1*TAFI(percent activity)+1.49 PAI-1(ug/L)  Equation 1:

As a feature of the model, the multivariate equation is a linear regression equation. As another feature of the model, the equation has eight variables or factors. A high CLT measured by thromboelastography has been found to predict clinical failure of tenecteplase in patients with pulmonary embolism. A low CLT has been found to predict an increased risk of hemorrhage.

[0058] As an example of the principles of the present disclosure, work was performed using blood obtained from a prospective, multicenter trial for treatment of submassive pulmonary embolism (TOPCOAT), clinical trials identifier: NCT00680628. The methods for TOPCOAT are detailed in a separate publication (Kline J A, Hernandez-Nino J, Hogg M M et al. Rationale and methodology for a multicenter randomized trial of fibrinolysis for pulmonary embolism that includes quality of life outcomes. Emerg Med Australas. 2013(25):515-526), the entire contents of which hereby being expressly incorporated herein by reference. Patients were randomized to receive a bolus infusion of the plasminogen activator tenecteplase (TNKase®) or volume-matched 0.9% NaCl placebo. To provide negative control data, blood was used from patients who were matched to TOPCOAT patients by age and sex—plasma from apparently healthy patients who were tested for acute PE, but had no clinical evidence of PE within 90 days. To provide positive control data, plasma was used from patients with conditions known to produce hypofibrinolysis via different mechanisms, including patients with diabetes type II, and patients with metabolic syndrome (Alessi M C, Juhan-Vague I. Metabolic syndrome, haemostasis and thrombosis. Thromb Haemost. 2008; 99(6):995-1000; Dunn E J, Philippou H, Ariens R A et al. Molecular mechanisms involved in the resistance of fibrin to clot lysis by plasmin in subjects with type 2 diabetes mellitus. Diabetologia. 2006; 49(5):1071-1080). All patients in the study provided written informed consent. Blood was obtained from an arm vein, either by venipuncture or withdrawal from an indwelling venous catheter prior to treatment into a vacuum tube containing sodium citrate (BD Vacutainer®, 2.7 mL, 0.109 Molar/3.2% sodium citrate). Phlebotomy in TOPCOAT patients was performed prior to administration of study drug. Blood was immediately placed on ice, and centrifuged within 30 minutes at 4° C. at 3,000×g for 20 minutes, which has been shown to deplete platelets (Brookes, K., et al., Issues on fit-for-purpose validation of a panel of ELISAs for application as biomarkers in clinical trials of anti-Angiogenic drugs. Br J Cancer, 2010. 102(10): p. 1524-32). Plasma was immediately aliquoted and frozen at −80° C.

[0059] Two methods to assess CLT were used in order to mimic two distinct in-vivo physiological conditions. First, to represent CLT in stagnant (zero shear) conditions, turbidity was measured using light transmittance and a device marketed as SpectraMax by Molecular Devices of Sunnyvale, Calif. To measure CLT under shear conditions, loss of mechanical stiffness was assessed using TEG and the Haemoscope 5000 (Braintree, Mass.). For both assays, plasma comprised 50% of the total mixture volume. Turbidity was measured at 405 nm at 37° C. Coagulation was initiated by spiking plasma with calcium chloride at final concentration 15 mM, human tissue factor at a final concentration of 0.6 pM (Dade Innovin; Siemens, USA), and phospholipids at a final concentration of 12 μM (Avanti Polar Lipids; Alabaster, Al). To induce fibrinolysis, tissue plasminogen activator (“tPA”) (Alteplase, Genentech; San Francisco, Calif.) was immediately added to the plasma prior to clot formation at a final concentration of 60 ng/mL. Tris-buffered saline (50 mM Tris-HCl, 0.1 M NaCl, pH 7.4) was used as a buffering agent. Calcium chloride, tissue factor, phospholipid mixture, tPA, and the buffer were mixed in disposable TEG cups containing heparinase (Haemonetics Corporation; Braintree, Mass.) prior to the addition of plasma. Following mixing of reagents, human plasma was added to start the reaction. A 100 uL volume was transferred from the TEG cup to a 96 well plate in duplicate, and allowed to run on the spectrophotometer. The remaining reaction volume was run using TEG. Assessment of CLT by measuring turbidity was performed at an absorbance of 405 nm. The CLT is derived from a clot-lysis profile and defined as the time from the midpoint of the baseline turbidity to maximum turbidity, representing clot formation, to the midpoint of maximum turbidity back to baseline turbidity, representing the lysis of the clot (see FIGS. 1 and 2).

[0060] All plasma proteins with the exception of plasminogen activator Inhibitor 1 (“PAI-1”) were measured on the STA Compact coagulation Analyzer® (Diagnostica Stago; Parsippany N.J.) with reagents purchased from the manufacturer and analyzed as follows: Fibrinogen concentration was determined using the Clauss clotting method (STA Fibrinogen 5). α.sub.2-antiplasmin, plasminogen and thrombin activated fibrinolysis inhibitor (“TAFI”) concentrations were determined via chromogenic assays (STA Stachrom α2-antiplasmin, STA Stachrom plasminogen and STA Stachrom TAFI). All assays were performed with the use of a commercial calibration standard; D-dimer levels were measured using a latex agglutination assay (STA Liatest D-DI). PAI-1 was quantified with a commercial ELISA assay (Life Technologies, Grand Island, N.Y.).

[0061] To assess the association of a prolonged CLT with clinical outcomes from the TOPCOAT sample, patients were grouped according to CLT value. Prolonged CLT was defined as those patients with a CLT greater than the 95 percentile from the normal control group compared with patients who had a CLT of less than or equal to the 95 percentile. The a priori outcomes assessed at three months post treatment were the results of psychometric tests for quality of life related to post-thrombotic syndrome (VEINES QoL), overall physical and mental perception of wellness from the Standard Form 36 (SF 36), as well as the SpO2(%), six minute walk distance (m), and echocardiography results.

[0062] Samples were evaluated for normality using the Shapiro-Wilk test. Means were compared using analysis of variance with Dunnett's post-hoc test to determine significance with pairwise comparisons of test groups (PE, diabetes mellitus and metabolic syndrome) versus control with P<0.05 considered significant. To determine which variables explain the change in CLT in the PE samples, a multivariate linear regression was performed with age, sex, body weight, fibrinogen, D-dimer, plasminogen, α2-antiplasmin, thrombin time, TAFI and PAI-1 as independent variables and CLT from turbidity or TEG as the dependent variable using two separate equations. The stepwise removal process was then used to select significant independent variables. Data were analyzed using SPSS (Version 22; IBM, Armonk N.Y.). Graphs were produced with Prism (Version 6.0; GraphPad, San Diego, Calif.) and SigmaPlot (version 12.0; Systat, San Jose, Calif.).

[0063] Table 1 below compares clinical features between the three control groups and the experimental group: normal (n=20), metabolic syndrome (n=10, positive control), diabetes mellitus (n=10, positive control), and intermediate risk PE (TOPCOAT). The groups were similar in age, but patients with metabolic syndrome and TOPCOAT group had a higher body mass index (p<0.05). Table 1 also compares relative concentrations and activities of biomarkers relevant to fibrinolysis in the control and PE groups. As expected, D-dimer concentrations were elevated in patients with PE. Fibrinogen levels were significantly higher in patients with PE when compared to controls. PAI-1 was significantly increased in all three test groups when compared with controls. There were no differences noted between groups in relation to a.sub.2-antiplasmin, plasminogen, or TAFI levels.

TABLE-US-00001 TABLE 1 Comparison of clinical data and plasma proteins between patient groups TOPCOAT Diabetes Metabolic (Intermediate Control Mellitus syndrome risk PE) Variable (n = 20) (n = 10) (n = 10) (n = 76) Age  56.5 ± 14.6  57.8 ± 14.2 65.1 ± 20   55 ± 13.9 Male gender (%) 11 (55%) 5 (50%) 7 (70%) 46 (61%) Body mass index 27.9 ± 8.1 31.3 ± 5.7  34.7 ± 4.8* 33.1 ± 9.1*  (kg/m.sup.2) Diabetes Mellitus 0 10  0 10 Prior venous 0 0 0 17 thromboembolism Active 0 0 0 15 malignancy Thrombin time 18.1 ± 2.sup.  19.6 ± 1.2 19.8 ± 1.7 .sup. 61 ± 47.1* (S) Fibrinogen 320.9 ± 54.7 350.7 ± 52.sup.  .sup. 338 ± 70.7 412.5 ± 148.9* (mg/dL) α.sub.2 antiplasmin† 102.7 ± 12.3 103.2 ± 5.7  106.4 ± 6.7  99.2 ± 18.2  Plasminogen† 98.2 ± 16  102.3 ± 11.3 104.5 ± 13.7 109.8 ± 27.2  TAFI† 107.3 ± 16.9 104.3 ± 18.6 106.9 ± 18.7 99.5 ± 25.7  D-dimer (μg/mL)  0.447 ± 0.429  0.360 ± 0.141  0.429 ± 0.307 6.592 ± 5.102* PAI-1 (pg/mL) 1072.3 ± 780.1    3457 ± 2518.7*  4171.3 ± 2177.7* 2367.5 ± 2212.9* *P < 0.05 from one-way ANOVA with Dunnett's comparison with control. ** Values are listed as mean ± SD unless otherwise indicated. *** Units are expressed as percent activity when compared to standardized controls provided by the manufacturer. Abbreviations: TAFI—thrombolysis activated fibrinolysis inhibitor; PAI—plasminogen activator inhibitor.

[0064] Data in Table 1 were examined using pairwise comparisons of age and BMI between the three test groups versus healthy controls. With equal variances assumed, no significant difference in age was found between groups following Dunnett's post-hoc analysis. Further, no significant difference in BMI was observed between the controls metabolic syndrome group or between the controls and DM patients. There was, however, a significant difference in BMI between the control and TOPCOAT groups (p=0.041).

[0065] Referring now to FIG. 5, a Dot plot of CLT values for each patient in four groups is shown. Horizontal lines in the figure represent the mean of each group of data. Abbreviations used in the figure are as follows: Cont—apparently healthy control patients; TEG—thromboelastography; MtSyn—metabolic syndrome; DM—diabetes mellitus; Spec—spectrophotometry (turbidimetric method); TOP—TOPCOAT. * indicates P<0.05 vs. control, and ** indicates P<0.01 vs. control, ANOVA with Dunnett's post-hoc. The CLTs were measured with both turbidity and TEG. Using the turbidimetric technique, the mean CLT was not significantly prolonged for patients with PE compared with controls, but was prolonged in patients with diabetes mellitus and metabolic syndrome compared with controls (P=0.623, P=0.002 and P=0.003, respectively from Dunnett's). With TEG, the mean CLT was significantly prolonged for patients with PE, diabetes mellitus and metabolic syndrome compared with controls (P=0.03, P=0.0026, and P=0.0005, respectively from Dunnett's). A significantly higher proportion of patients with PE (18%) had a CLT>180 minutes compared with controls (0%) (95% confidence interval for the difference in 18%=0.3 to 27%).

[0066] To determine if a prolonged CLT has clinical significance, Table 3 below compares the mean values for the VEINES QoL score (Kahn S R, Lamping D L, Ducruet T et al. VEINES-QOL/Sym questionnaire was a reliable and valid disease-specific quality of life measure for deep venous thrombosis. Journal of Clinical Epidemiology. 2006; 59(10):1049-1056), pulse oximetry, Body mass index (kg/m2), six minute walk distance at 3 months, and the normalized mental and physical component scores (“PCS”) from the Rand Standard for (SF36) quality of life survey (Hays R D, Sherbourne C D, Mazel R M. The RAND 36-Item Health Survey 1.0. Health Econ. 1993; 2(3):217-227) in patients with and without prolonged CLT as measured by TEG. In patients given tenecteplase, significant differences were found in the VEINEs QoL score, the PCS from the SF36 between those with prolonged CLT and those without prolonged CLT. Additionally, the percentage of patients with right ventricular (“RV”) dysfunction or overload at 3 months were assessed, defined as RV dilation (>43 mm transverse diameter in diastole), RV hypokinesis, or an estimated RV systolic pressure >45 mm Hg. For those treated with tenecteplase, RV dysfunction or overload was found in 36% with prolonged CLT, versus 26% with normal CLT (95% CI for difference,−20.6 to 42.4%, exact two-sided P=0.46), and for placebo, RV dysfunction or overload was found in 54% with prolonged CLT, versus 27% with normal CLT (95% CI for difference of 17%-5.1 to 55.4%, exact two-sided P=0.095).

TABLE-US-00002 TABLE 3 VEINES QoL Distance Mental Health Physical score for post- walked in summary component thrombotic Baseline six score from score from syndrome SpO2 (%) minutes SF36 SF36 mean Tenec prolonged CLT 85.9 96.2 328.5 55.2 42.5 SD Tenec prolonged CLT 12.7 1.7 104.9 11.3 10.7 mean Tenec normal CLT 96.0 97.0 445.7 52.8 49.2 SD Tenec normal CLT 10.5 1.5 79.1 6.9 8.2 P from unpaired t-test 0.021 0.191 0.002 0.453 0.051 mean Placebo prolonged CLT 93.5 97.0 411.6 54.3 41.5 SD Placebo prolonged CLT 14.4 1.3 97.0 8.8 14.1 mean Placebo normal CLT 87.7 97.0 399.4 51.4 41.7 SD Placebo normal CLT 18.7 1.6 122.6 13.9 12.8 P from unpaired t-test 0.326 0.934 0.766 0.494 0.969

[0067] Taken together, data from FIG. 5 indicate the predictivenss of the CLT from TEG. The study then sought to determine predictors of CLT in seconds from TEG in PE patients using multivariate linear regression for each technique. The model included factors that are directly or indirectly known to affect probability of response to lysis and risk of hemorrhage, including age, body mass index (“BMI”), fibrinogen (“FIB”), D-dimer concentration, plasminogen, (“PLG”), α2 antiplasmin (“AP”), thrombin time (“TT”), and thrombin activated thrombolysis inhibitor (“TAFI”), and plasminogen activator inhibitor 1 (“PAI-1”) concentrations. Table 4 below shows the results of a regression analysis done on the CLT data for 76 patients in the original TOPCOAT dataset (Kline J A, Kabrhel C., Courtney D M et al. Treatment of submassive pulmonary embolism with tenecteplase or placebo: cardiopulmonary outcomes at three months (TOPCOAT): Multicenter double-blind, placebo-controlled randomized trial. J Thromb & Haemost. 2014). The analysis was performed with the statistical program StatsDirect (v 10.131). The values denoted by b0 . . . b9 represent the beta coefficients in the equation and the t and P values are the significance tests for each coefficient.

TABLE-US-00003 TABLE 4 Intercept b0 = −4,540.769 t = −1.179 P = 0.243 Age b1 = 14.34 r = 0.059 t = 0.482 P = 0.631 BMI b2 = 34.766 r = 0.095 t = 0.777 P = 0.44 FIB b3 = −0.054 r = −0.002 t = −0.015 P = 0.988 D-Dimer b4 = 27.739 r = 0.041 t = 0.332 P = 0.741 PLG b5 = −19.227 r = −0.102 t = −0.832 P = 0.409 AP b6 = 47.24 r = 0.21 t = 1.741 P = 0.086 TT b7 = −7.076 r = −0.099 t = −0.808 P = 0.422 TAFI avg (calc) b8 = 41.343 r = 0.24 t = 2.007 P = 0.049 PAI-1 (correct b9 = 1.478 r = 0.683 t = 7.603 P < 0.001 analysis) TEG CLT = −4,540.769 +14.34 Age +34.766 BMI −0.054 FIB +27.739 D-Dimer −19.227 PLG +47.24 AP −7.076 TT +41.343 TAFI avg (calc) +1.478 PAI-1 (correct analysis).

[0068] Next an analysis of variance from regression was performed as follows:

TABLE-US-00004 Source of variation Sum Squares DF Mean Square Regression 9.351509E+008 9 1.039057E+008 Residual 7.462891E+008 66 11,307,409.921 Total (corrected) 1.681440E+009 75 Root MSE = 3,362.649 F = 9.189 P < 0.001 Multiple correlation coefficient (R) = 0.746 R.sup.2 = 55.616% Ra.sup.2 = 49.564% Durbin-Watson test statistic = 1.372

Multiple Regression—Best Subset

Selected Variables:

[0069] AP

[0070] TAFI avg (calc)

[0071] PAI-1 (correct analysis)

[0072] F=28.225

[0073] R.sup.2=0.54

[0074] Mallows' Cp=2.336

Multiple Linear Regression

[0075]

TABLE-US-00005 Intercept b0 = −2,813.581 t = −1.228 P = 0.223 AP b1 = 35.346 r = 0.185 t = 1.601 P = 0.114 TAFI avg (calc) b2 = 31.066 r = 0.228 t = 1.985 P = 0.051 PAI-1 (correct b3 = 1.494 r = 0.714 t = 8.653 P < 0.001 analysis) TEG CLT = −2,813.581 +35.346 AP +31.066 TAFI avg (calc) +1.494 PAI-1 (correct analysis).

Multiple Linear Regression—Prediction

[0076] TEG CLT=7,318.816 (least squares mean)

[0077] 95% Confidence interval=6,569.712 to 8,067 92

[0078] 95% Prediction interval=745.453 to 13,892.178

CLT from thromboelastography=−2,813.6+31.1*α2-antiplasmin(percent activity)+31.1*TAFI(percent activity)+1.49 PAI-1(ug/L).  Equation (1):

This equation thus uses the measurement of three plasma proteins to estimate the CLT from thromboelastography, which was previously shown to predict clinically important outcomes. The following data provide descriptive statistical values for the result of Equation 1 in the TOPCOAT population.

TABLE-US-00006 Tenecteplase Placebo Variables All treated treated Valid data 76 36 40 Missing data 7 3 3 Sum 543,143.09 236,929.614 304,103.997 Mean 7,146.62 6,581.378 7,602.6 Variance 11,818,451.154 7,589,679.392 14,784,305.412 Standard deviation 3,437.797 2,754.937 3,845.036 Variance coefficient 0.481 0.419 0.506 Standard error of mean 394.342 459.156 607.954 Upper 95% CL of 7,932.19 7,513.515 8,832.302 mean Lower 95% CL of 6,361.049 5,649.241 6,372.898 mean Geometric mean * * * Skewness 2.009 2.072 1.996 Kurtosis 8.429 7.521 7.933 Maximum 23,043.417 16,995.795 23,043.417 Upper quartile 8,456.127 7,554.542 9,456.283 Median 6,552.58 6,034.587 7,308.727 Lower quartile 4,921.705 4,911.542 5,335.679 Interquartile range 3,534.421 2,642.999 4,120.604 Minimum 2,109.48 2,109.48 3,253.541 Range 20,933.938 14,886.316 19,789.876 Centile 95 15,247.843 16,995.795 19,145.63 Centile 5 3,923.352 4,165.33 3,958.125

[0079] FIG. 6 shows the performance of Equation 1 in terms of its ability to predict patients who had a major or clinically relevant non-major bleed and were treated with tenecteplase. FIG. 6 is a receiver operating characteristic curve using the result of Equation 1 to predict the outcome of hemorrhage that was observed in 8 of 36 patients treated with the plasminogen activator tenecteplase from the TOPCOAT population. To produce the ROC properly, the actual value of the equation was subtracted from the maximal value that was found in the entire TOPCOAT dataset (namely 23,041 seconds). At a cutoff of area under ROC curve by extended trapezoidal rule=0.678309 (result obtained by 23,041-estimated CLT from Equation 1). The Wilcoxon estimate of area under ROC curve=0.607 (95% CI=0.373 to 0.841). The optimum cut-off point selected=18,115 seconds, but because this value must be subtracted from 23,041, the optimal cutoff is 4,926 seconds, which corresponds closely to the mean CLT from the study group without hemorrhage. At values above this number, three patients had clinically relevant but non-major hemorrhage (5/7 or 62.5% sensitivity for detection of those at risk of bleeding) and this included 75% of patients without bleeding (75% specificity). Patients with values below this number are at increased risk of hemorrhage from standard dose plasminogen activator agents. All eight patients who had major or clinically relevant non-major bleeding and who were treated with the standard dose of the plasminogen activator tenecteplase had a TEG CLT time <=4,926 seconds (100% sensitivity) and 19/28 patients who were treated with standard dose tenecteplase had a TEG CLT time <=4,926 seconds, meaning that 11/28 had a value >4,926 seconds (39% specificity).

[0080] From the 36 patients treated with tenecteplase the study then examined the significance of a TEG CLT<4,926 seconds, estimated from Equation 1 to predict a bleeding outcome that could be related to tenecteplase administration to humans. To test for the significance of the TEG CLT>4,926 seconds from Equation 1, a standard odds ratio was performed, represented below.

TABLE-US-00007 5 3 7 21 Observed odds ratio = 5 Conditional maximum likelihood estimate of odds ratio = 4.744 Exact Fisher 95% confidence interval = 0.714 to 38.952 Exact Fisher one sided P = 0.062, two sided P = 0.086 Exact mid-P 95% confidence interval = 0.877 to 29.863 Exact mid-P one sided P = 0.036, two sided P = 0.071

[0081] From the 36 patients treated with tenecteplase the study then examined the significance of a TEG CLT>4,926 seconds, estimated from Equation 1 to predict an adverse outcome that could be related to inadequate clot lysis in the human. These adverse outcomes, assessed at three months, included a low six minute walk distance (<330 m), or a PCS from the SF36<30 points, or right ventricular dilation or hypokinesis or estimated right ventricular systolic pressure >45 mm Hg on echocardiography. To test for the significance of the TEG CLT>4,926 seconds from Equation 1, a standard odds ratio was performed, represented below. This analysis excludes the eight patients with a hemorrhagic outcome, and therefore only includes 28 patients.

Exact Confidence Limits for 2 by 2 Odds

Input Table:

[0082]

TABLE-US-00008 12 3 9 4 Observed odds ratio = 1.778 Conditional maximum likelihood estimate of odds ratio = 1.741 Exact Fisher 95% confidence interval = 0.229 to 15.062 Exact Fisher one sided P = 0.412, two sided P = 0.67 Exact mid-P 95% confidence interval = 0.289 to 11.505 Exact mid-P one sided P = 0.275, two sided P = 0.549

[0083] Thus, a TEG CLT>4,926 seconds, as estimated from Equation 1 had a slight tendancy to predict a worsened outcome in terms of exercise tolerance, quality of life or echocardiographic finding at 3 months for patients treated with tenecteplase. This suggests that patients with a value >4,926 are less likely to benefit from standard dose plasminogen activator treatment delivered by systemic infusion.

[0084] As should be apparent to those skilled in the art, patients with TEG CLT value at or below 4,926 seconds estimated from Equation 1 could benefit from reduced dose plasminogen activator fibrinolytic treatment, whether delivered systemically or with a catheter positioned in close proximity to the thrombus. Moreover, it should also be apparent that patients with a TEG CLT over 4,926 seconds estimated from Equation 1 may benefit from adjunctive treatments including increased or prolonged or repeated dosing of plasminogen activator agent, delivered either systemically by catheter immediately proximal to the thrombus. The finding of a prolonged value from Equation 1 also indicates the need to use of a device that imparts mechanical, ultrasonic or other method of energy transfer to enhance fibrinolysis.

[0085] The finding of more extreme TEG CLT values from Equation 1 could indicate the need for alternative treatment to plasminogen activators. Patients with a value below the 5 percentile (3,923 seconds) could be considered at very high risk of hemorrhage and patients with values above the 95th percentile (15,247 seconds) could be considered at very high risk of clinical failure with treatment with plasminogen activating agents. Therefore, these patients should be considered for treatment with alternative agents, including the so-called direct fibrinolytic agents, plasmin, delta plasmin, miniplasmin or microplasmin, or the use of alternative fibrinolytic agents lumbrokinase or nattokinase, or the use of surgical embolectomy or the use of purely mechanical means of clot removal.

[0086] Referring now to FIG. 7, a system is depicted for carrying out the above-described principles of the present disclosure. System 10 generally includes a biomarker concentration measurement system 12 and a computing device 14. Biomarker concentration measurement system 12 may include a plurality of different hardware and software components. For example, as described above, system 12 may be configured to measure a concentration of α2-antiplasmin in a blood sample of the patient, a concentration of activated fibrinolysis inhibitor (“TAFI”) in the blood sample, and a concentration of plasminogen activator Inhibitor 1 (“PAI-1”) in the blood sample. System 12 may include an STA Compact coagulation Analyzer® (Diagnostica Stago; Parsippany N.J.) with reagents purchased from the manufacturer and analyzed as follows: Fibrinogen concentration was determined using the Clauss clotting method (STA Fibrinogen 5). System 12 may include a chromogenic assay with a commercial calibration standard for determining a concentration α2-antiplasmin, plasminogen and thrombin activated fibrinolysis inhibitor (TAFI). System 12 may further include a commercial ELISA assay (Life Technologies, Grand Island, N.Y.) for determining a plasminogen activator Inhibitor 1 (“PAI-1”) concentration within a sample of the patient's blood. Any and all of these components (and the associated software) are represented by system 12.

[0087] Computing device 14 generally includes an interface 16 which receives data from system 12, a processor 18, a memory 20 and a user interface 22. Computing device 14 may receive data representing biomarker concentrations from system 12 through a wired or wireless connection. While computing device 14 is depicted as including a single processor 18, it should be understood that multiple processors may be used, either as a part of computing device 14 or part of a distributed network of processors. Memory 20 may include non-transient instructions for execution by processor 18 to perform the functions described above, including but not limited to carrying out the computation of CLT for the blood sample and its comparison to the various predetermined cutoff times for predicting or estimating clinical responsiveness of a patient to administration of a plasminogen activating agent as described herein. Memory 20 may also include the predetermined cutoff times and other parameters necessary for performing the various calculations described herein. While memory 20 is depicted as a single component, it should be understood that multiple memory devices may be incorporated (or associated with) computing device 14 according to the principles of the present disclosure. User interface 22 is generically depicted as a single device, but it should be understood that user interface 22 may include a plurality of different devices (and associated software) for receiving user input and providing output to the user of computing device 14, including but not limited to a display, keyboard, mouse, touch-screen, alarm, or audio/visual communication device, which either directly receives and provides information to/from the user or does so indirectly through other intervening devices.

[0088] Various modifications and additions can be made to the exemplary embodiments discussed without departing from the scope of the present disclosure. For example, while the embodiments described above refer to particular features, the scope of this invention also includes embodiments having different combinations of features and embodiments that do not include all of the described features. Accordingly, the scope of the present disclosure is intended to embrace all such alternatives, modifications, and variations as fall within the scope of the claims, together with all equivalents thereof.