WO2022076835A1

WO2022076835A1 - Methods and apparatuses for modeling adamts13 and von willebrand factor interactions

Info

Publication number: WO2022076835A1
Application number: PCT/US2021/054195
Authority: WO
Inventors: Hoa Q. NGUYEN
Original assignee: Dyax Corp.
Priority date: 2020-10-09
Filing date: 2021-10-08
Publication date: 2022-04-14
Also published as: AR123769A1; TW202230389A

Abstract

Aspects of the present application provide for methods and apparatuses for simulating interactions between von Willebrand factors and AD AMTS 13 in endogenous or recombinant form. Some aspects provide for a computer-implemented method for modeling ADAMTS13 and VWF interactions, comprising obtaining a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors, assigning the disease predictive descriptors to a virtual patient population, and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. The biomarker may include ULVWF multimers, cleaved VWF fragments, lactate dehydrogenase and/or platelet cells. The QSP model may represent AD AMTS 13 interactions with stretched and globular VWF multimers. The QSP model may simulate a conversion of stretched VWF multimers to globular VWF multimers.

Description

METHODS AND APPARATUSES FOR MODELING ADAMTS13 AND VON WILLEBRAND FACTOR INTERACTIONS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit under 35 U.S.C. under § 119(e) of U.S. Provisional Application Serial No. 63/089,935 titled “METHODS AND APPARATUSES FOR MODELING AD AMTS 13 AND VON WILLEBRAND FACTOR INTERACTIONS” filed on October 9, 2020 under Attorney Docket No. D0617.70137US00, which is incorporated by reference in its entirety herein.

BACKGROUND

[0002] Sickle cell disease (SCD) is an inherited blood disorder characterized by sickle hemoglobin formation, leading to rigid and deformed sickle shaped red blood cells. These sickle cells cannot traverse the microcirculation, creating blockages leading to tissue hypoxia and excruciating pain during events of vaso-occlusive crises (VOC). [0003] Von Willebrand factor (VWF) is an adhesive and multimeric glycoprotein that plays an essential role in maintaining hemostatic balance. VWF promotes platelet aggregation and clot formation at sites of endothelial injury. VOC events in patients with SCD are triggered due to high concentrations of uncleaved ultra-large von Willebrand factor (ULVWF) multimers accumulated in plasma of patients to which extracellular hemoglobin (Hb) bonds, causing excessive clotting.

[0004] In order to prevent unnecessary clotting, VWF is negatively regulated by a catalytic enzyme ‘A Disintegrin and Mettaloproteinase with Thrombospondin Type 1 Motifs 13’ (AD AMTS 13). Typically, AD AMTS 13 functions to prevent extracellular Hb from bonding with ULVWF by cleaving ULVWF to form smaller VWF fragments. However, this function is underperformed in patients with an ADAMTS13 deficiency or acquired auto-inhibition of AD AMTS 13, leading to high concentrations of uncleaved ULVWF in plasma.

[0005] Thrombotic thrombocytopenic purpura (TTP) is a disorder that causes blood clots (thrombi) to form in small blood vessels throughout the body. These clots can cause serious medical problems if they block vessels and restrict blood flow to organs such as the brain, kidneys, and heart. For patients with TTP, the formation of thrombi may also be caused by high concentrations of uncleaved VWF multimers in the plasma. Thrombotic thrombocytopenic purpura includes immune-mediated thrombotic thrombocytopenic purpura (iTTP) and congenital thrombotic thrombocytopenic purpura (eTTP).

BRIEF SUMMARY

[0006] Some aspects provide for a computer-implemented method for modeling ADAMTS13 and von Willebrand factor (VWF) interactions, comprising: obtaining a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker.

[0007] Some aspects provide for a system, comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. [0008] Some aspects provide for at one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. [0009] Some aspects provide for a computer-implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultralarge von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which ADAMTS13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers. [0010] Some aspects provide for a system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which AD AMTS 13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

[0011] Some aspects provide for at least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which AD AMTS 13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

[0012] Some aspects provide for a computer-implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non-adherence on reducing the concentration of uncleaved ULVWF multimers.

[0013] Some aspects provide for at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer-implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of rADAMTS13 for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non- adherence on reducing the concentration of uncleaved ULVWF multimers.

[0014] Some aspects provide for at least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of rADAMTS13 for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of nonadherence on reducing the concentration of uncleaved ULVWF multimers.

[0015] Some aspects provide for a computer-implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

BRIEF DESCRIPTION OF DRAWINGS

[0016] Various aspects and embodiments of the application will be described with reference to the following figures. It should be appreciated that the figures are not necessarily drawn to scale. Items appearing in multiple figures are indicated by the same reference number in all the figures in which they appear. For purposes of clarity, not every component may be labeled in every drawing.

[0017] FIG. 1 a schematic diagram of ADAMTS13 cleaving ULVWF to smaller VWF polymers and the inhibition of this action by Hemoglobin (Hb).

[0018] FIG. 2 illustrate a biological process map representing the molecular interplay of components in ADAMTS13-VWF interactions.

[0019] FIG. 3 illustrates a model diagram illustrating a QSP model for simulating interactions between AD AMTS 13 and VWF, according to some non-limiting embodiments. [0020] FIGS. 4A-4C illustrate comparisons of cleaved VWF concentrations predicted by the QSP model of FIG. 3 to cleaved VWF concentrations from in vitro data over a range of rADAMTS13 and Hb levels, according to some non-limiting embodiments.

[0021] FIG. 4D illustrates a sensitivity of a binding constant of Hb to VWF reflected in the QSP model of FIGG, according to some non-limiting embodiments.

[0022] FIG. 4E illustrates a model simulation of percentage VWF cleavage for different binding constants of Hb to VWF, according to some non-limiting embodiments.

[0023] FIG. 5 illustrates a comparison of percent VWF cleavage between model prediction and in vitro pre-incubation data, according to some non-limiting embodiments.

[0024] FIGS. 6A-6B illustrate comparisons of fraction VWF in active form between model prediction and in vivo data, according to some non-limiting embodiments.

[0025] FIG. 7 illustrates a graph comparing pharmacokinetic (PK) data of rADAMTS 13 between model prediction and in vivo data, according to some non-limiting embodiments.

[0026] FIGS. 8A-8B illustrate a simulation of a vaso-occlusive crisis (VOC) event, according to some non-limiting embodiments.

[0027] FIG. 9 is a graph illustrating a range of Hb levels for patients in remission, according to some non-limiting embodiments.

[0028] FIG. 10 is a graph showing a proportion of subjects in each treatment group with detectable rADAMTS13-mediated VWF cleavage product as provided by TTP Phase 1 study data, according to some non-limiting embodiments.

[0029] FIGS. 11 A- 11C illustrate graphs showing pharmacokinetic profiles predicted by the QSP model of FIG. 3 for virtual patients in different treatment groups, according to some non-limiting embodiments.

[0030] FIGS. 12A-12B illustrate graphs comparing model output and clinical data of detectable rADAMTS13-mediated VWF cleavage, according to some non-limiting embodiments.

[0031] FIGS. 13A-13B illustrate comparisons of model output and clinical data of total VWF concentration in a patient, according to some non-limiting embodiments.

[0032] FIGS. 14A-14B illustrate comparisons of model output and clinical data of active VWF amount in a patient, according to some non-limiting embodiments. [0033] FIG. 15A is a flow chart illustrating a computer implemented system and method for simulating interactions of ADAMTS13 and VWF, according to some non-limiting embodiments.

[0034] FIGS. 15B illustrates an example method for modeling ADAMTS13 and VWF interactions, according to some non-limiting embodiments.

[0035] FIG. 15C illustrates an example method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large VWF multimers, according to some non-limiting embodiments.

[0036] FIG. 15D illustrates an example method for determining an effect of nonadherence to a dosing regimen of recombinant ADAMTS13 in reducing a concentration of uncleaved ultra-large VWF multimers, according to some non-limiting embodiments. [0037] FIG. 15E illustrates an example method for determining a concentration of VWF multimers in response to administration of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0038] FIGS. 16A-16B illustrate model results of pharmacokinetic parameters and active VWF amount for various doses and dose schedules of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0039] FIG. 17 is a graph comparing the number of days with active VWF under remission level for three different dose regimens, according to some non-limiting embodiments.

[0040] FIGS. 18A-18B illustrate graphs shown percentages of VWF cleavage over time for different doses of rADAMTS13, according to some non-limiting embodiments.

[0041] FIG. 19 illustrates graphs comparing rADAMTS13 concentration and active VWF concentration over time for a dose of rADAMTS13, according to some nonlimiting embodiments.

[0042] FIG. 20A illustrates a graph showing time profiles of rADAMTS13 in different types of patients and for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0043] FIG. 20B illustrates concentrations of active VWF in different types of patients for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments. [0044] FIG. 21 illustrates a duration in which VWF concentration is under remission levels for various doses of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0045] FIG. 22 illustrates a graph showing the effect of baseline total VWF on a duration in which active VWF is under remission level for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0046] FIG. 23 illustrates a graph showing an effect of cell-free Hb level on a duration that active VWF is under remission level, according to some non-limiting embodiments. [0047] FIG. 24 illustrates an effect of Hemoglobin binding affinity with VWF on a duration in which active VWF is under remission levels for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0048] FIGS. 25A-25C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to the binding affinity of hemoglobin to VWF constant used in the model, according to some non-limiting embodiments.

[0049] FIGS. 26A-26C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to total VWF level, according to some nonlimiting embodiments.

[0050] FIGS. 27A-27C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to endogenous AD AMTS 13 activity, according to some non-limiting embodiments.

[0051] FIGS. 28A-28G are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to remission Hb and vaso-occlusive crises levels, according to some non-limiting embodiments.

[0052] FIG. 28H is a graph comparing the results of the graphs shown in FIGS. 28 A- 28G, according to some non-limiting embodiments.

[0053] FIGS. 29A-29B are graphs illustrating a dose response of hemoglobin-bound VWF amount and active VWF amount ten hours after a dosage of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0054] FIG. 30 is a graph illustrating sensitivity of active VWF amounts predicted by the QSP model to variations of different model parameters, according to some non-limiting embodiments. [0055] FIG. 31 depicts, schematically, an illustrative computing device on which any aspect of the present disclosure may be implemented, according to some non-limiting embodiments.

DETAILED DESCRIPTION

[0056] INTRODUCTION

[0057] Aspects of the present application provide for methods and apparatuses for modeling ADAMTS13 and VWF interactions. In particular, aspects of the present application provide for a quantitative systems pharmacology (QSP) model for simulating AD AMTS 13 and VWF interactions. In some embodiments, the QSP model simulates the mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and the inhibition of this action by extracellular hemoglobin (Hb). The QSP model may comprise pharmacokinetic (PK) and pharmacodynamic (PD) components.

[0058] Use of the QSP model described in the present application may provide various types of information regarding VWF and ADAMTS13 interactions which may be impractical or impossible to clinically obtain. For example, the QSP model may provide, as output, levels of biomarkers of a patient (e.g., concentration ULVWF, concentration cleaved VWF fragments, platelet cell count, or concentration of lactate dehydrogenase (LDH)). The QSP model may receive, as input, a particular dose and/or dose regimen of a therapeutic intervention for treating ADAMTS13 inhibition and/or deficiency. Accordingly, the QSP model may provide a quantitative relationship between a therapeutic intervention (e.g., a dosage or dose regiment of the therapeutic intervention) and a biomarker that provides a useful clinical target in treating patients (e.g., SCD patients experiencing VOC, patients with eTTP or iTTP). This quantitative relationship may be used to assist in determining in-human dosages of therapeutic interventions for SCD, eTTP, iTTP.

[0059] In some embodiments, the QSP model may be used for evaluating the efficacy of a therapeutic intervention for patients with decreased or underperforming AD AMTS 13. In some embodiments, the therapeutic intervention comprises administration of a recombinant form of AD AMTS 13 (rADAMTS13). In some embodiments, the therapeutic intervention comprises a plasma exchange of donor plasma having healthy levels of endogenous AD AMTS 13. In some embodiments, the therapeutic intervention comprises administration of frozen plasma having healthy levels of endogenous AD AMTS 13. Exchange of donor plasma and administration of frozen plasma increases the concentration of endogenous ADAMTS13 in the recipient patient. For example, the donor plasma comprises variable amounts of endogenous AD AMTS 13. Plasma exchange and administration of frozen plasma are currently the standard of care for eTTP and iTTP respectively. The QSP model may simulate the effects on VWF-ADAMTS13 interactions of any of these therapeutic interventions.

[0060] In some embodiments, the QSP model may be used to determine an appropriate in-human dosage or dose regiment of a therapeutic intervention. The therapeutic intervention may be used to regulate homeostasis in VOC through improved VWF- AD AMTS 13 interaction. The QSP model enables these determinations without requiring further human testing thereby providing information which may be impractical or impossible to clinically obtain. In some embodiments, the QSP model may be implemented with a virtual population to execute a virtual clinical trial to evaluate the effects of a therapeutic intervention. The inventors have recognized that such techniques may facilitate development of new and more effective treatment modalities for AD AMTS 13 inhibition and/or deficiency.

[0061] Accordingly, some aspects provide for a computer-implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (UEVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker (e.g., UEVWF multimers including stretched or globular ULVWF multimers, cleaved VWF fragments including stretched or globular VWF fragments, lactate dehydrogenase and/or platelet cells). In some embodiments, the method further comprises displaying the processed data.

[0062] In some embodiments, the method further comprises determining pharmacokinetic parameters; assigning the pharmacokinetic parameters to the virtual patient population; determining therapeutic intervention data based on administration of an administered drug; and processing the therapeutic intervention data and the virtual patient population with the QSP model to determine effectiveness of the administered drug. In some embodiments, the administration of the administered drug comprises administration of endogenous and/or recombinant AD AMTS 13. In some embodiments, administration of the endogenous ADAMTS13 comprises a plasma exchange with plasma having endogenous AD AMTS 13.

[0063] In some embodiments, the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers. The QSP model may simulate a conversion of globular VWF multimers to stretched VWF multimers.

[0064] In some embodiments, the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

[0065] In some embodiments, the method further comprises using the processed data to determine whether the concentration of the at least one biomarker is below a first threshold or above a second threshold. In some embodiments, the method further comprises using the processed data to determine a duration in which the concentration of the at least one biomarker is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold. In some embodiments, the method further comprises using the processed data to determine a change in the concentration of the at least one biomarker over time.

[0066] In some embodiments, the QSP model comprises a plurality of differential equations representing one or more biological reactions.

[0067] In some embodiments, the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics (e.g., at least one of height, weight, age, or gender) of a patient to whom the administered drug is administered.

[0068] In some embodiments, the disease predictive descriptors comprise one or more parameters characterizing a concentration of AD AMTS 13 in a patient. The patient may comprise a patient having sickle cell disease, a patient having a congenital thrombotic thrombocytopenic purpura (eTTP), or a patient comprises a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP).

[0069] In some embodiments, the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient. The pharmacokinetic parameters and disease predictive descriptors may be assigned to the one or more variables of each data set.

[0070] Some aspects provide for a system, comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. [0071] Some aspects provide for at one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. [0072] Some aspects provide for a computer-implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultralarge von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which ADAMTS13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker (e.g., uncleaved ULVWF multimers, cleaved VWF fragments, lactate dehydrogenase and/or platelet cells); and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers. In some embodiments, the method further comprises displaying the processed data.

[0073] In some embodiments, the administered drug comprises endogenous and/or recombinant AD AMTS 13. In some embodiments, the administered drug comprises plasma (e.g., frozen plasma) of a donor patient.

[0074] In some embodiments, the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers. The QSP model may simulate a conversion of globular VWF multimers to stretched VWF multimers. In some embodiments, the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

[0075] In some embodiments, the indicator of the effectiveness of the administered drug is obtained at least in part by comparing the processed data to known data indicating a threshold concentration of the at least one biomarker. The known data may comprise biomarker amounts of an untreated subject with sickle cell disease congenital thrombotic thrombocytopenic purpura, and/or immune mediated thrombotic thrombocytopenic purpura, biomarker amounts of a subject without sickle cell disease, congenital thrombotic thrombocytopenic purpura, or immune mediated thrombotic thrombocytopenic purpura, and/or biomarker amounts of a subject with sickle cell disease in remission.

[0076] In some embodiments, the method further comprises using the processed data to determine whether the concentration of the at least one biomarker is below a first threshold or above a second threshold. In some embodiments, the method further comprises using the processed data to determine a duration in which the concentration of the at least one biomarker is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold. In some embodiments, the method further comprises using the processed data to determine a change in the concentration of the at least one biomarker over time. [0077] In some embodiments, the QSP model comprises a plurality of differential equations representing one or more biological reactions.

[0078] In some embodiments, the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics (at least one of height, weight, age, or gender) of a patient to whom the administered drug is administered.

[0079] In some embodiments, the disease predictive descriptors comprise one or more parameters characterizing a concentration of AD AMTS 13 in a patient (e.g., a patient having sickle cell disease, a patient having a congenital thrombotic thrombocytopenic purpura (eTTP), and/or a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP)).

[0080] In some embodiments, the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient. The pharmacokinetic parameters and disease predictive descriptors may be assigned to the one or more variables of each data set.

[0081] Some aspects provide for a system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which AD AMTS 13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers. [0082] Some aspects provide for at least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which AD AMTS 13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

[0083] Some aspects provide for a computer-implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non-adherence on reducing the concentration of uncleaved ULVWF multimers. The method may further comprise displaying the processed data. [0084] In some embodiments, processed data includes a frequency in which the amount of uncleaved ULVWF fragments exceeds a threshold. In some embodiments, the processed data includes a percentage by which the concentration of uncleaved ULVWF fragments exceeds a threshold. In some embodiments, using the processed data to determine the effect of the frequency of non-adherence includes comparing the processed data to known data.

[0085] In some embodiments, the administered drug comprises endogenous and/or recombinant AD AMTS 13. In some embodiments, the administered drug comprises plasma (e.g., frozen plasma) of a donor patient.

[0086] In some embodiments, the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers. In some embodiments, QSP model simulates a conversion of globular VWF multimers to stretched VWF multimers. In some embodiments, QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

[0087] In some embodiments, the method further comprises determining whether the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is below a first threshold or above a second threshold.

In some embodiments, the method further comprises determining a duration in which the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is below the first threshold or the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is above the second threshold. In some embodiments, the method further comprises determining a change in the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments

[0088] In some embodiments, the QSP model comprises a plurality of differential equations representing one or more biological reactions.

[0089] In some embodiments, the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics (at least one of height, weight, age, or gender) of a patient to whom the administered drug is administered.

[0090] In some embodiments, the disease predictive descriptors comprise one or more parameters characterizing a concentration of AD AMTS 13 in a patient (e.g., a patient having sickle cell disease, a patient having a congenital thrombotic thrombocytopenic purpura (eTTP) and/or a patient immune mediated thrombotic thrombocytopenic purpura (iTTP)).

[0091] In some embodiments, the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient. The pharmacokinetic parameters and disease predictive descriptors may be assigned to the one or more variables of each data set.

[0092] Some aspects provide for at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer-implemented method for determining an effect of non- adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of nonadherence on reducing the concentration of uncleaved ULVWF multimers.

[0093] Some aspects provide for at least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non-adherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of nonadherence on reducing the concentration of uncleaved ULVWF multimers.

[0094] Some aspects provide for a computer-implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

[0095] Ins some embodiments, the administration of AD AMTS 13 comprises administration of recombinant AD AMTS 13. In some embodiments, the administration of AD AMTS 13 comprises administration of plasma (e.g., frozen plasma) of a donor patient. [0096] In some embodiments, wherein the VWF multimers comprise one of uncleaved ultra-large VWF multimers or cleaved VWF fragments. In some embodiments, the QSP model represents ADAMTS13 interactions with stretched and globular VWF multimers. The QSP model may simulate a conversion of globular VWF multimers to stretched VWF multimers. In some embodiments, the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers. [0097] In some embodiments, the method further comprises determining whether the concentration of VWF multimers is below a first threshold or above a second threshold. The method may comprise determining a duration in which the concentration of VWF multimers is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold. In some embodiments, the method further comprises using the QSP model to determine a change in the concentration of VWF multimers over time.

[0098] In some embodiments, the QSP model comprises a plurality of differential equations representing one or more biological reactions.

[0099] In some embodiments, the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics (at least one of height, weight, age, or gender) of a patient to whom the administered drug is administered. In some embodiments, the disease predictive descriptors comprise one or more parameters characterizing a concentration of AD AMTS 13 in a patient (e.g., a patient having sickle cell disease, a patient having a congenital thrombotic thrombocytopenic purpura (eTTP), and/or a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP)).

[0100] In some embodiments, the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient. The pharmacokinetic parameters and disease predictive descriptors may be assigned to the one or more variables of each data set.

[0101] Some aspects provide for a system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer-implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

[0102] Some aspects provide for at least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing AD AMTS 13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

[0103] BIOLOGICAL OVERVIEW

[0104] According to some aspects of the present application, the apparatuses and methods described herein may be used to simulate interactions between VWF and AD AMTS 13. That is, the QSP model described herein may represent the mechanism by which ADAMTS13 cleaves ULVWF and inhibition thereof by extracellular hemoglobin. [0105] Von Willebrand factor (VWF) is an adhesive and multimeric glycoprotein that plays an essential role in maintaining hemostatic balance. VWF promotes platelet aggregation and clot formation at sites of endothelial injury. However, VOC events in patients with SCD are triggered due to high concentrations of uncleaved ultra-large von Willebrand factor (ULVWF) multimers accumulated in plasma of patients to which extracellular hemoglobin (Hb) bonds, causing excessive clotting.

[0106] In order to prevent unnecessary clotting, VWF is negatively regulated by a catalytic enzyme ‘A Disintegrin and Mettaloproteinase with Thrombospondin Type 1 Motifs 13’ (AD AMTS 13). The VWF-cleaving protease AD AMTS 13 is a plasma zinc metalloprotease that cleaves VWF in the A2 domain. ADAMTS13 plays a role in primary hemostasis by modulating the platelet-tethering and hemostatic capacity of VWF. Thus, the ADAMST13-mediated VWF cleavage facilitates to maintain a balance between normal hemostatic function and abnormal platelet agglutination leading to thrombosis. A severe deficiency of AD AMTS 13 activity (<5% of normal) leads to the persistence and accumulation of hyperactive UL VWF multimers in the circulation. ULVWF multimers accumulate in plasma and can lead to undesirable platelet aggregation and widespread microvascular thrombosis.

[0107] Cleavage of ULVWF by AD AMTS 13 is inhibited by extracellular Hb, which may also be represented by the QSP model. Extracellular hemoglobin interacts with VWF, binds to AD AMTS 13 cleavage site on the A2 domain of VWF, and block VWF cleavage by AD AMTS 13.

[0108] FIG. 1 a schematic diagram 100 of AD AMTS 13 cleaving ULVWF to smaller VWF polymer fragments and the inhibition of this action by Hemoglobin (Hb) at the vascular endothelium. The schematic diagram 100 illustrates the mechanism of action employed in the pharmacodynamic model described herein.

[0109] ULVWF multimers are secreted from endothelial cells at the vascular endothelium. These VWF multimers play a significant role in cell adhesion and prothrombotic complications. The diagram 100 illustrates stretched and globular ULVWF as it exists in the body. Shear forces (e.g., due to blood flow) induce the globular ULVWF to unfold into stretched ULVWF which can be cut into VWF fragments by AD AMTS 13. Active VWF, as used herein, refers to uncleaved VWF (that is, ULVWF) in stretched form. In stretched form, binding sites of the ULVWF are exposed for adhesion to platelets, causing clotting, as well as to binding of Hb and AD AMTS 13.

[0110] As shown in diagram 100, AD AMTS 13 may cleave the stretched ULVWF into VWF fragments (plasma VWF). However, extracellular Hb or a lack or inhibition of ADAMTS13 limits the cleavage of stretched ULVWF by ADAMTS13. Extracellular Hb (cell-free Hb in diagram 100) binds to the stretched ULVWF preventing any AD AMTS 13 present from performing cleavage. The Hb-bound ULVWF accumulates leading to clotting. Accumulation of the ULVWF may lead to creation of thrombi and/or blockages in the microcirculation caused by sickle cell build-up, which may lead to voc. [0111] Sickle cell disease (SCD) is a hereditary hemoglobinopathy caused by an autosomal recessive single point mutation in the P-globin chain of adult hemoglobin. During hypoxic conditions, deoxygenation triggers sickling of the red blood cells causing the release of excessive extracellular hemoglobin. SCD is characterized by chronic hemolytic anemia and episodes of vaso-occlusive painful events leading to progressive tissue ischemia and multi-organ damage.

[0112] An important pathophysiologic factor is the presence of high concentrations of uncleaved VWF multimers in the plasma from SCD patients. Plasma of SCD patients (both clinically asymptomatic and with acute painful crises) revealed very mild or no deficiency in ADAMTS13 activity compared to healthy individuals, but higher concentrations of VWF and particularly ULVWF multimers and therefore a relative deficiency of ADAMTS13 activity to its substrate.

[0113] Thrombotic thrombocytopenic purpura (TTP) is a disorder that causes blood clots (thrombi) to form in small blood vessels throughout the body. These clots can cause serious medical problems if they block vessels and restrict blood flow to organs such as the brain, kidneys, and heart. For patients with TTP, the formation of thrombi may also be caused by high concentrations of uncleaved VWF multimers in the plasma. Thrombotic thrombocytopenic purpura includes immune-mediated thrombotic thrombocytopenic purpura (iTTP) and congenital thrombotic thrombocytopenic purpura (eTTP).

[0114] A recombinant (“r”) form of AD AMTS 13 (“rADAMTS13”) may be used to help regulate the production and/or function of AD AMTS 13 in the body. That is, rADAMTS13 may be administered to a patient having irregular production of

AD AMTS 13 or extracellular Hb, or inhibition of AD AMTS 13 functions. Additionally or alternatively, levels of endogenous ADAMTS13 in a patient may be increased by plasma exchange of donor plasma and/or administration of frozen plasma having healthy levels of naturally occurring AD AMTS 13. Plasma exchange is generally performed in an intensive care unit. As will be described further herein, the QSP model represents the molecular interplay of components in ADAMTS13-VWF interactions, for example as shown in the diagram 100, in which AD AMTS 13 cleaves ULVWF multimers into smaller VWF fragment and the inhibition of this action by extracellular Hemoglobin (Hb) binding to the ULVWF multimers. [0115] QUANTITATIVE SYSTEMS PHARMACOLOGY MODEL DEVELOPMENT [0116] As described herein, the inventors have developed a QSP model for simulating interactions between AD AMTS 13 and VWF. The QSP model may comprise a plurality of differential equations with parameters that reflect interactions between ADAMTS13 and VWF. The parameters may be parameterized and calibrated with biological data in literature as well as clinical data from one or more clinical trials. The QSP model may be verified by comparison of QSP model output with known data.

[0117] Development of the QSP model may comprise a number of steps starting with development of a model diagram. The model diagram may be developed based on investigation into the biological mechanism of AD AMTS 13 in the scission of ULVWF to smaller VWF fragments and inhibition thereof by Hb.

[0118] Subsequent to developing the model diagram, the QSP model may be formulated by determining a series of mathematical equations representing the model diagram. Specifically, the series of mathematical equations may comprise a plurality of differential equations that represent the cleavage of ULVWF by AD AMTS 13 and inhibition thereof by Hb.

[0119] The formulated model may then be parameterized. For example, values for model parameters, further described herein, may be estimated based on literature and clinical data. The parameterized model may be calibrated with one or more data sets. For example, the parameterized values of the model may be refit based on in vitro and/or in vivo data used to calibrate the model. Subsequently, the calibrated model may be verified against additional data not used in the calibration.

[0120] The verified model may be tested via simulation. For example, a simulation of a clinical trial in which a therapeutic intervention is administered to a patient may be performed and an effect of the therapeutic intervention on ADAMTS13 and VWF interactions may be observed. For example, a test dosage and/or dose regimen may be input into the QSP model to evaluate the efficacy of the dosage and/or dose regimen in treating a patient, such as a SCD patient under VOC.

[0121] Biological Process Map

[0122] Development of the QSP model may begin with development of a model diagram. FIG. 2 illustrate a biological process map representing the molecular interplay of components in ADAMTS13-VWF interactions. The model diagram may be developed based on investigation into the biological mechanism of AD AMTS 13 in the scission of ULVWF to smaller VWF fragments and inhibition thereof by Hb. Hemoglobin (Hb) used throughout this disclosure is defined as the extracellular Hb. AD AMTS 13 and rADAMTS13 refer to the endogenous and recombinant forms, respectively. VWF starts off as ULVWF when it is first created by the vascular endothelium. Once it has been cleaved to form a single VWF monomer, it is called a VWF fragment.

[0123] As shown in FIG. 2, the process map 200 includes ULVWF in both globular state (VG) and stretched state (VS). Shear stress exhibited on the globular ULVWF (e.g., due to blood flow) causes the globular ULVWF to unfold into stretched ULVWF.

[0124] Stretched ULVWF may be acted on by three different products (AD AMTS 13 in either endogenous or recombinant form, extracellular Hb, or thrombospondin- 1). First, stretched ULVWF may be cleaved into smaller fragments by AD AMTS 13 which is in either endogenous or recombinant form. Cleavage of the ULVWF creates VWF fragments, as shown in FIG. 2. The rate at which this occurs depends on the binding constant between stretched ULVWF and ADAMTS13 (KdvwF-ADAM), which may be parameterized, calibrated, and verified during model development. Cleavage of the ULVWF by the AD AMTS 13 (either in endogenous or recombinant form) may also be based on the turnover number (or the maximal number of molecules of substrate converted to product per active site per unit time) of AD AMTS 13 for VWF (kcat_s), which may be parameterized, calibrated, and verified during model development.

[0125] Binding sites of the ULVWF are exposed when the ULVWF unfolds into stretched form. Accordingly, extracellular hemoglobin (f-Hb) and thrombospondin (TSP- 1) may additionally act on the stretched ULVWF. Specifically, the extracellular hemoglobin and thrombospondin inhibit cleavage of the ULVWF by binding to the binding cites of the stretched ULVWF. Binding of extracellular hemoglobin to the stretched VWF may be based on the binding constant between stretched ULVWF and the extracellular hemoglobin (KdvwF-HB), which may be parameterized, calibrated, and verified during model development. Binding of thrombospondin to the stretched VWF may be based on the binding constant between stretched ULVWF and the thrombospondin (KdvwF-rsp), which may be parameterized, calibrated, and verified during model development.

[0126] The process map 200 further represents input of endogenous or recombinant ADAMSTS13 into the region of a patient’s body containing the stretched ULVWF. The input concentration of AD AMTS 13 for a particular patient may be output from a pharmacokinetic model of the QSP model, further described herein, including a central volume (V_c) and a peripheral volume (V_p).

[0127] FIG. 3 illustrates a model diagram 300 illustrating a QSP model for simulating interactions between AD AMTS 13 and VWF, according to some non-limiting embodiments. As shown in FIG. 3, the QSP model may include multiple individual models, including a pharmacokinetic (PK) model 310, a pharmacodynamic (PD) model 320, and a clinical outcome model 330. The PK model may provide PK parameters for use in one or more PD models, for example, describing how characteristics of a patient (e.g., height, weight, gender, age, etc.) affect a drug administered to the patient (for example, affecting the concentration of the drug in the patient’s bloodstream). The one or more PD models may illustrate interactions between VWF and AD AMTS 13 among other components, such as thrombospondin and extracellular hemoglobin, as described herein.

[0128] The QSP model shown in FIG. 3 further includes a clinical outcome model 330. The clinical outcome model may receive output from the PD model 320 (e.g., concentrations of one or more biomarkers, such as active VWF, VWF fragments, platelet count, and/or LDH concentration). Measured clinical outcomes may include level (e.g., concentrations) of active VWF, level (e.g., concentration) if VWF fragments, a determination of whether active VWF level is under a threshold and a duration thereof, and/or a determination of whether VWF fragment level is above a threshold and a duration thereof.

[0129] In some embodiments, the QSP model may be configured to model interactions between VWF and ADAMTS13 (either in endogenous or recombinant form). The QSP model may also account for inhibition of VWF cleavage by ADAMTS13 by extracellular hemoglobin and/or thrombospondin. For example, the model may include parameters, as described herein, representing binding constants between VWF and AD AMTS 13, thrombospondin, and hemoglobin.

[0130] In some embodiments, the QSP model may represent interactions of AD AMTS 13 with both stretched and globular forms of ULVWF. The QSP model may simulate a conversion of globular VWF multimers to stretched VWF multimers. For example, the QSP model may include one or more equations representing the conversion of globular VWF to stretched VWF. [0131] In some embodiments, the QSP model is utilized in computer-implemented methods for simulating treatment of SCD, eTTP, and/or iTTP. For example, the various PK, PD, and clinical outcome models described herein may be used to evaluate the effectiveness of a new or existing treatment modality for SCD, eTTP, and/or iTTP. In some embodiments, only some of the individual models may be utilized when implementing the QSP model in a computer-implemented method. For example, in some embodiments, the QSP model may be implemented without using the PK model(s) to better understand biomarker behavior in the absence of any therapeutic intervention.

Therefore, as used herein, the quantitative systems pharmacology (QSP) model should be understood to encompass any combination of the PK, PD, and clinical outcome models described herein.

[0132] PK Model

[0133] According to some aspects, the QSP model includes a PK model for providing PK parameters to the PD model. An example PK model 310 is shown in FIG. 3. The PK model may describe how a drug is absorbed and distributed by a particular patient, more particularly, the rate and extent of the distribution of the drug to different tissues and the rate of elimination of the drug. The PK model may be modeled as a series of differential equations describing the transit of the drug throughout the body.

[0134] As shown in FIG. 3, the PK model 300 is a two-compartment PK model with a subcutaneous (SC) depot. In particular, the PK model may be divided into central and peripheral compartments. The central compartment consists of plasma and tissues where distribution of the drug occurs more rapidly, whereas the peripheral compartment consists of tissues and plasma where the distribution of the drug occurs more slowly. The inventors have appreciated that use of a PK model having multiple compartments may account for non-homogeneities in the distribution of the drug.

[0135] The PK model may be used to model the PK behavior of a drug in a patient. For example, in some embodiments, the PK model is used to model the PK behavior of existing treatment modalities, such as rADAMTS13 and/plasma exchange and/or administration of frozen plasma for increased level of endogenous AD AMTS 13. In some embodiments, the PK model may be used to model the PK behavior of a new and/or previously untested drug. For example, absorption rate (ka) and bioavailability (F) for a drug to be modeled may be input into the PK model and the predicted concentration of the drug in the patient may be output for inputting into the PD model. [0136] While in the illustrated embodiment of FIG. 3 a two-compartment PK model is used, in other embodiments a single-compartment PK model with a subcutaneous (SC) depot or a non-compartmental PK model may be used.

[0137] PD Model(s)

[0138] According to some aspects, the QSP model comprises one or more PD models for simulating interactions between ADAMTS13 and VWF. In particular, the PD model reflects the cleavage of ULVWF by endogenous or recombinant AD AMTS 13 and the inhibition thereof by extracellular hemoglobin. Table 2 gives a list of variables used in the QSP model.

[0139] Table 1: List of species in the QSP model

[0140] As described herein, AD AMTS 13 in endogenous or recombinant form cleaves ULVWF into VWF fragments to prevent accumulation of ULVWF which may lead to sickled red blood cells (RBC) occurring due to hypoxia accumulating with the ULVWF and creating blockages in the microcirculation of a patient. Cleavage is inhibited, however, by a deficiency or inhibition of AD AMTS 13 and/or by an excess of extracellular hemoglobin which binds to the stretched ULVWF. These relationships and reactions are reflected in tables 2a and 2b below. Table 2a gives a lists of reactions represented by the QSP model.

[0141] Table 2a: List of reactions in model

[0142] Table 2b: List of governing equations in model

[0143] The QSP model is limited to considering VWF, Hb and ADAMTS13 (both endogenous and recombinant forms). Other proteins that may affect the levels of these proteins such as thrombospondin 1 and haptoglobin are not included. The inventors have recognized that inclusion of certain proteins in the QSP model, such as thrombospondin 1 and haptoglobin, may make the model unstable and inaccurate. Excluding such proteins from the QSP model may improve the overall accuracy of the model.

[0144] The QSP model is based on an assumption that the recombinant rADAMTS13 is has the same behavior as the endogenous AD AMTS 13 and the same model parameter values are applied for the two types of AD AMTS 13. The microvessels of the circulatory system are most susceptible to thrombi and vascular occlusion. The model assumes this region as the dominant site of action.

[0145] The QSP model accounts for ULVWF in both stretched and globular form. The QSP model simulates, and therefore includes parameters for, conversion of ULVWF from globular form to stretched form by unfolding due to shear forces. A steady state conversion is applied to model conversion of ULVWF from globular to stretched form. The same degradation rate parameter is assumed for VWF and their complexes in both stretched and globular form.

[0146] The inventors have recognized that the underlying molecular mechanism represented by the QSP model, that is the competitive binding of AD AMTS 13 and Hb to VWF may be applicable to simulating patients and treatment thereof with either SCD or TTP.

[0147] Equation El 1 in Table 2b above represents the binding affinity between Hb and VWF (kon vs_Hb ■ VS. Hb + koffvG Hb ■ VG_Hb). The QSP model represents binding between Hb and VWF as a one-to-one relationship. The inventors have ruled out possibilities where the QSP model accounts for cooperativity of multiple hemoglobin to one VWF as leading to a less accurate model.

[0148] The model tracks the state of individual monomer units of VWF as unbound, ADAMTS13 bound, Hb bound and cleaved product. The model distinguishes between VWF in stretched and globular forms. The blood flow (velocity gradient in the blood vessel) provides the shear force to induce the stretching of VWF. Microvessels of the circulatory system encounter the highest shear rate.

[0149] All binding/unbinding and proteolysis reactions are mass-action based, i.e., first- order in the reactant(s). The same binding/unbinding reactions occur in both stretched and globular forms but with different reaction parameters (Table 2a). A VWF monomer unit that is bound to ADAMTS13 can undergo proteolysis producing a cleaved product. However, the proteolysis rate of VWF in globular form is reported to be negligible, as the cleavage site in the A2 domain is in folded form and inaccessible to AD AMTS 13. [0150] Clinical Outcome Model

[0151] As described herein, the QSP model may further include a clinical outcome model. Literature data is available comparing ADAMTS13 and VWF levels in healthy control subjects and in SCD patients in the asymptomatic state and those with painful crises. The study found that an VWF Ristocetin Cofactor Activity (VWF:RCo) was markedly higher for SCD patients: 73 U/mL, 143 U/mL and 172 U/mL for healthy controls, SCD patients in asymptomatic state, and SCD patients with painful crises, respectively. VWF:RCo is a measure of the ability of VWF to bind platelet glycoprotein lb (GPIb). This binding requires the Al domain of the VWF to be in stretched form. Additional literature data also reported a significant correlation between the percentage of high molecular weight VWF multimers and VWF:RCo.

[0152] The model assumes that VWF:RCo represents “active VWF” that leads to thrombosis and vascular occlusion and uses it as a measure of clinical outcome. In the model, the VWF:RCo amount is represented by the VWF amount in stretched form with and without bound Hb (active VWF = VS + VS_Hb, names defined in Table 1). The VWF bound to ADAMTS13 was excluded since it is expected to be negligible due to rapid proteolysis once bound. As another measurement of clinical outcome, the model prediction of the number of days that a patient in VOC can maintain the active VWF level below the level found in the SCD remission condition was used.

[0153] Accordingly, the clinical outcome model defines “active VWF” as the uncleaved VWF in stretched form where the binding sites are exposed for adhesion to platelets as well as binding to Hb and AD AMTS 13. Active VWF is one biomarker that can be output by the clinical outcome model.

[0154] The clinical outcome model may also output a determination that a concentration of active VWF is under a threshold level and/or a duration for which a concentration of active VWF is under the threshold level. In some embodiments, the threshold level of active VWF may be the active VWF concentration for an SCD patient in remission (e.g., as available from clinical data). In some embodiments, the clinical outcome model may output a determination that cleaved VWF fragments concentration is above a threshold level and/or a duration for which the concentration of cleaved VWF fragments is above the threshold level. The threshold level of cleaved VWF fragments may be determined based on a concentration of cleaved VWF fragments in a SCD patient in remission (e.g., as available from clinical data).

[0155] QUANTITATIVE SYSTEMS PHARMACOLOGY MODEL PARAMETERIZATION

[0156] Subsequent to developing the model diagram of the QSP model and set of differential equations representing reactions between parameters reflected in the model, the QSP model may be parameterized. For example, parameters of the model, defined in Table 3 below, may be set to initial values based on either literature data, clinical data, known mathematical relationships between other parameters, or obtained via further calibration steps described herein. Table 3 below gives the model parameters, their initial values, and the source for the initial values.

[0157] Table 3: List of model parameters

[0158] Binding between AD AMTS 13 and VWF has been described as interactions between multiple domains of the two proteins in a “molecular zipper” fashion that leads to the Tyrl605 and Metl606 cleavage site in the A2 domain. The dissociation constant between AD AMTS 13 and VWF, KDADAMTSi3-vwrhas been reported to be 80nM when the VWF is in globular conformation (between AD AMTS 13 TSP5-CUB domains and the VWF D4-CK domains) and to be 10 nM when the VWF is in a stretched conformation (between A2 domain of VWF and the spacer domain of AD AMTS 13). [0159] The following data, obtained from literature, was utilized for the binding between Hb and VWF: Hb binds directly to VWF via Al domain with KD (Hb-VWF) with a dissociation constant of ~ 15,000 nM and Hb binds directly to VWF via A2 domain with KD (Hb-VWF) with a dissociation constant of ~ 183 nM. Verification of the model, described herein, demonstrated the ability of the model to accurately reproduce both in vitro and in vivo data representing outcomes of healthy individuals, SCD patients in remission, and SCD patients experiencing VOC.

[0160] As shown in Table 3, some parameters were estimated from literature values. However, binding affinity between VWF in stretched form (active VWF) with Hb, binding affinity between VWF in globular form with Hb, and the fraction of VWF in stretched form (active VWF) were not obtained directly from literature but rather were adjusted in further calibration steps based on clinical data. In some instances, binding affinity between Hb and stretched VWF (active VWF) was the only model parameter for which the in vitro value needed to be changed to match in vivo data.

[0161] QUANTITATIVE SYSTEMS PHARMACOLOGY MODEL CALIBRATION AND VERIFICATION

[0162] As described herein, subsequent to parameterizing the developed model, the QSP model may be calibrated and verified. For example, the QSP model may be calibrated using known data (e.g., in vitro data, in vivo data). The initial values of the parameters that were set when parameterizing the QSP model may be adjusted based on the calibration. The calibrated parameters may be subsequently verified against data not used in the calibration.

[0163] Calibration of QSP model with in vitro data

[0164] First, the QSP model was calibrated using in vitro data obtained according to an assay, described herein. Table 7 shows an overview of the assay setup to measure proteolytic activity of rADAMTS 13 in the presence of Hb with and without preincubation. The study utilized a flow-based assay, to create the shear conditions to unfold the full-length VWF substrate to allow AD AMTS 13 binding and ADAMTS 13 -mediated cleavage. The study assumed that the full-length VWF is fully stretched - i.e., all the monomer units are available for reaction. Literature data has shown that the cleavage level in the flow-based assay closely matched that using pre-denatured full-length VWF and small peptides derived from the VWF-A2 domain.

[0165] Table 4: Overview of assay setup to measure proteolytic activity of rADAMTS 13 in the presence of hemoglobin with and without pre-incubation

[0166] FIGS. 4A-4C illustrate comparisons of cleaved VWF concentrations predicted by the QSP model of FIG. 3 to cleaved VWF concentrations from in vitro data over a range of rADAMTS13 and Hb levels under the direct incubation set up. The cleaved product data is reported as % relative to the cleaved product amount at Hb=0. The cleaved product is measured as the amount of dimeric 176 kDa VWF cleavage fragment. The data shows monotonic reduction in % cleaved product with increasing Hb amount in all cases and with decreasing amount of rADATMTS13 for almost all levels. As shown in FIGS. 4A, the data point at r AD ATMTS 13=0.5, Hb=0.1 (indicated by 1 in FIG. 4A) deviated from this trend and the data point at rAD AMTS 13=0.25, Hb=0.1 (indicated by 2), showed a considerably greater reduction compared to other data sets at the same Hb level.

[0167] FIG. 4B compares the model prediction to the data using the lower of the literature values for the binding constant of Hb to VWF: KDvs Hb of 183nM. This value was applied to binding of the fully- stretched VWF (VS) for the in vitro simulation. The Hb sensitivity from the model results was considerably greater than that shown in the direct incubation data suggesting that the binding constant may be overly strong. Increasing the KDvs Hb 5-fold to 915 nM best matched the direct incubation data, as shown in FIG. 4C, and the sensitivity of rAD ATMTS 13 and Hb on % cleaved product from the model matched the direct incubation data well.

[0168] FIG. 4D shows the sensitivity of the KDvs Hb in fitting to the data obtained under the direct incubation set-up. In particular, illustrates a sensitivity of a binding constant of Hb to VWF (KDvs Hb) reflected in the QSP model of FIGG, according to some nonlimiting embodiments. FIG. 4E illustrates a model simulation of percentage VWF cleavage for different binding constants of Hb to VWF, according to some non-limiting embodiments. [0169] Verification of QSP model calibrated with in vitro data

[0170] The parameterized model may be verified against data not used in the calibration to determine that model results substantially match known data to ensure that the QSP model may accurately model interactions between AD AMTS 13 and VWF; and provide effective evaluation of new existing treatment modalities. For example, subsequent to calibrating the QSP model with in vitro data, the calibrated model may be verified against additional data.

[0171] The verification step described herein was performed using pre-incubation in vitro data. The cleaved VWF product amount is measured in the same way described for the in vitro data used to calibrate the QSP model. The in vitro data used for verification shows monotonic reduction in % cleaved product with increasing Hb amount in all cases. However, the effect of rADATMTS13 level on % cleaved amount showed considerable scatter and no clear trend could be observed.

[0172] Table 5: In vitro data using pre-incubation used for verification

[0173] FIG. 5 illustrates a comparison of percent VWF cleavage between model prediction and in vitro pre-incubation data, according to some non-limiting embodiments. In particular, FIG. 5 compares the pre-incubation data to the model using the calibrated KDvs Hb = 915 nM. The comparison was made using % cleaved product averaged over the range of rADATMTS13 for each Hb level. The trend with Hb amount resulted in % cleaved product predictions consistent with the data but the absolute % cleaved product levels were lower for all Hb levels. Some of this difference may be attributed to the considerable uncertainty in the measurement. Given that, the model may proceed with KDvs Hb = 915 nM as a starting point in comparing to in vivo data.

[0174] Calibration of QSP model with in vivo data

[0175] The QSP model may further be calibrated with one or more additional sets of data. For example, the QSP model may be calibrated using in vivo data reflecting ADAMTS13 and VWF interactions.

[0176] Table 6 summarizes the levels of AD AMTS 13, Hb and VWF in healthy subjects, SCD patients in remission and SCD patients in VOC. These levels were compiled from the literature sources. Using the degradation rates obtained from literature, each of these protein levels was used to estimate the corresponding synthesis rate constant (i.e., ksynADAM, ksynHb and ksynVWF) under steady state conditions. These calibrated rate constants are marked “calibrated” in Table 3.

[0177] Table 6: Levels of ADAMTS13, Hb, and VWF in healthy, SCD patients in remission, SCD patients in VOC, and TTP patients

[0178] To perform in vivo simulation, it is needed to estimate what fraction of the total VWF is in stretched and globular form. Literature data reporting the amount of VWF bound with Hb for healthy individuals and SCD patients in remission, the level of “active VWF” from healthy individuals and SCD patients in VOC, measured with an enzyme linked immunosorbent assay (ELISA) with an antibody directed against the Al domain of VWF, may be used to estimate this. Given that the Al domain is accessible in stretched form only, it may be assumed that this measurement obtained in literature is comparable to the model’s definition of active VWF described herein.

[0179] FIGS. 6A-6B illustrate comparisons of fraction VWF in active form between model prediction and in vivo data, according to some non-limiting embodiments. In particular, FIG. 6A illustrates a comparison of amount of VWF bound with Hb for normal individuals and SCD patients in remission from in vivo data and model prediction. FIG. 6B illustrates a comparison of amount of active VWF in normal individuals and SCD patients in VOC from in vivo data and model prediction.

[0180] In order to match the data, the amount of VWF in stretched form was set to be 2.5% of the total VWF for all conditions (normal individuals, SCD patients, TTP patients). Measured volume fraction of microvessels in humans ranged from 3 to 5%. All the other model parameters were the same as those obtained from the in vitro calibration/verification except for the binding affinity between Hb and VWF. To match the in vivo data, the KDvs Hb was reduced to half of the calibrated value in in vitro, 457.5 nM. One possible explanation for this reduction, representing a greater apparent inhibition by Hb in vivo compared to in vitro, may be that in vivo condition presents effects of other proteins known to have inhibitory effect such as Thrombospondin 1 (TSP-1). The calibrated value of KDvG_Hb = 20,588 nM is close to 15,000 nM reported in literature.

[0181] Verification of QSP model calibrated with in vivo data

[0182] The calibrated model may once again be verified using data not used in the calibration. Verification of the QSP model calibrated with in vivo data was performed using data from a Phase 1 TTP study. The study investigated the first-in-human PK and safety of rADAMTS13 in patients with congenital AD AMTS 13 deficiency. The Simulation Results of this verification step are reported in the results section, described below.

[0183] PARAMETERIZATION OF PK MODEL

[0184] FIG. 7 illustrates a graph comparing pharmacokinetic (PK) data for different doses of rADAMTS13 (5 U/kg, 20 U/kg, and 40 U/kg) between model prediction and in vivo data, according to some non-limiting embodiments. FIG. 7 shows how the two- compartment PK model results compared to the TTP data. The fitted model parameters are shown in Table 3. The data displays dose proportionality, and this is captured closely by the model.

[0185] VIRTUAL POPULATION DEVELOPMENT AND SIMULATIONS

[0186] As described herein, the QSP model may be used to simulate interactions between VWF and ADAMTS13 to obtain information useful in evaluating new and existing treatments for SCD, iTTP, and eTTP. In order to simulate such interactions without the need for a clinical trial, a virtual population and treatment simulation may be developed.

[0187] Virtual Population Development

[0188] The virtual population, on which a therapeutic intervention is tested, may comprise a virtual data set comprising a plurality of “patients”. Each patient may comprise a subsequent data set (e.g., Patienti) and may represent an individual virtual patient of the virtual population.

[0189] Each patient in the virtual patient population may be assigned a set of PK parameters representing variability in the drug disposition for a particular patient (e.g., parameters indicating how a therapeutic intervention is impacted by biographical characteristics of the patient). In some embodiments, PK parameters are randomly assigned to virtual population, and may, in some embodiments, be based on clinical data or synthetic data. Example PK parameters may include body weight, age, sex, height, race, and/or SCD status (e.g., in remission, healthy, under attack).

[0190] In some embodiments, each of the virtual patients in the virtual population may be assigned disease predictive descriptors. Example disease predictive descriptors may include a virtual patient’s propensity to VOC in the absence of therapeutic intervention, for example. For example, the disease predictive descriptors may include a concentration of endogenous AD AMTS 13 and/or a concentration of extracellular Hb in the patient prior to administration of any therapeutic intervention. In some embodiments, the disease predictive descriptors are determined at least in part by simulation from a Poisson distribution informed by known data regarding the disease predictive descriptors.

[0191] In some embodiments, a constant disease predictive descriptor may be applied to each patient in a virtual patient population. For example, in some embodiments, a baseline characteristic may be applied equally to all patients in the virtual patient population. [0192] Simulation Development

[0193] A simulation of a clinical trial using the QSP model may be performed. The simulation design is aimed at making the first human dose prediction and investigating the sensitivity to the model inputs based on variabilities in literature data as well as uncertainties in the model parameters. The overall approach for the simulation is described herein.

[0194] The QSP model described herein was used to represent 3 virtual patients: a healthy individual, a SCD patient under remission, and a SCD patient in VOC. These patients were created using the same model input parameters shown in Table 3 for all the patients and steady-state conditions specific to each patient shown in Table 6.

[0195] A VOC event was created by simulation prior to starting the treatment as shown in FIG. 8A. Starting from the SCD patient in remission at steady-state, a step change was made in the Hb level from the SCD patient in remission to the level for an SCD patient in VOC at 20 hours prior to starting the treatment as shown in FIG. 8 A. FIG. 8B shows the model’s response to this change in terms of active (i.e. stretched) VWF.

[0196] From the patient in VOC (at time=0 in FIG. 8A), the simulation introduces rADAMTS13 intravenously with the following dose regimens: single dose, single dose plus a booster at 2 days at half of the single dose level, and single dose plus a booster at 3 days at half of the single dose level.

[0197] The clinical outcome from the QSP model included (a) the reduction in the amount of active VWF over time and (b) the duration of maintaining the active VWF below a target level. The target level for active VWF was that of the patient in the remission condition.

[0198] Table 7 outlines the simulation design studies that have been performed. The range of the parameter values studied corresponds to the range reported in the literature.

[0199] Table 7: Outline of simulation design studies

[0200] Case 1 represented the Base Case. The three dose regimens described above were simulated for the VOC patient.

[0201] Case 2 investigated the binding affinity between Hb and VWF in stretched form. The lower binding affinity represents a greater inhibition effect by Hb.

[0202] Case 3 investigated variability in the VWF level of patients in remission. [0203] Case 4 investigated the variability in the endogenous AD AMTS 13 activity among the SCD patients. The activity of the rADAMTS13 was unchanged from the nominal value of 1.5 U/pg.

[0204] Case 5 studied the variability in level Hb levels in remission and VOC. FIG. 9 is a graph illustrating a range of Hb levels for patients in remission, according to some nonlimiting embodiments. FIG. 9 shows the range of Hb levels studied. Case 5F is from the Base Case and it represents a mid-point in the Hb level in remission and upper range in the Hb level in VOC. The sensitivity of Hb level in remission was investigated by Cases 5 C-F-G, and the sensitivity of Hb level in VOC was investigated by Cases 5 A-B-C and 5 D-E-F.

[0205] Simulation Results

[0206] The results from the simulation described herein may be verified. In this case, the simulation results were verified using data from a Phase 1 TTP clinical study. The study investigated the first-in-human PK and safety of rADAMTS13 in patients with congenital AD AMTS 13 deficiency. Table 8 shows the key plasma AD AMTS 13 PK parameters for the non-compartment model reported in the TTP Study.

[0207] Table 8. Summary of key plasma ADAMTS13 PK parameters in Shire Phase 1 TTP Study.

[0208] FIG. 10 is a graph showing a proportion of subjects in each treatment group with detectable rADAMTS13-mediated VWF cleavage product as provided by the TTP Phase 1 study data. When comparing the model prediction to the data, using the small number of patient samples used in the Study would create a large variability in the simulation output depending on the values for the PK parameters chosen from within the reported standard deviation for each treatment and hence make it difficult to compare. To overcome this, the simulation may be iteratively performed for increasing virtual patient sample sizes until the simulation output result averages stopped changing. It was found that a sample size of 50 virtual patients was enough to meet this target.

[0209] FIGS. 11 A- 11C illustrate graphs showing pharmacokinetic profiles predicted by the QSP model of FIG. 3 for virtual patients in different treatment groups, according to some non-limiting embodiments. Grey lines 1102A-C are predictions for individual virtual patients and the black lines 1100A-C are the measurements.

[0210] The model prediction on the % of patients with detectable cleavage product was made with the 50 virtual cohort patients. The model is the same as that developed for use with SCD patients, except for the use of TTP-specific input conditions (protein levels, Table 8). FIGS. 12A-12B show comparisons of model output to the TTP Phase 1 study data using PK parameters from each specific treatment group. The PK profile shown in FIGS. 12A-12B was used, and a detectable limit of cleaved products was adjusted to best match the data (17pM).

[0211] A good agreement was found except for the 20 U/kg treatment group. Some of this discrepancy may arise from errors in the PK parameters obtained from the data given the small number of patients in the clinical study (e.g., 3 patients in the 20 U/kg treatment group). This hypothesis was tested by performing the same simulation but using the PK parameters obtained with the 40 U/kg data since this data had the most patients, FIG. 12B illustrates that comparison.

[0212] FIGS. 13A-13B illustrate comparisons of model output and clinical data of total VWF concentration in a patient, according to some non-limiting embodiments. FIG. 13 A illustrates data as measured by VWF Antigen (VWF: Ag). FIG. 13B shows model prediction represented by total amount of VWF. From the same TTP Phase 1 study data, FIG. 13 A shows little change in the VWF:Ag level at the three different rADAMTS13 dose levels. Taking VWF:Ag as a measure of total VWF, FIG. 13B shows that the model predicts a similar behavior. This behavior can be rationalized with the model: most of the VWF is inactive (globular) form and thus the total amount is relatively insensitive to rADAMTS13 dosage.

[0213] FIGS. 14A-14B illustrate comparisons of model output and clinical data of active VWF amount in a patient, according to some non-limiting embodiments. In particular, FIG. 14A illustrates, for a dosage of 40 U/kg of rADAMTS13, data as measured by VWF Ristocetin cofactor activity (VWF: RCo). FIG. 14B illustrates model prediction represented by active amount of VWF.

[0214] It was observed that for almost all individuals, the VWF:RCo level was reduced in the first 5 to 10 hours followed by slight recovery in some and remaining at the reduced level in others. The prediction reproduces the behavior of the initial reduction followed by a recovery and plateau as shown in FIG. 14B. This behavior can also be rationalized with the model: the high initial level of rADAMTS13 leads to greater proteolysis of active VWF followed by gradual reduction in the proteolysis rate due to the degradation of rADAMTS13.

[0215] The model’s underlying molecular mechanism and calibrated model parameters may be applied to both TTP and SCD. The above results support this assumption and the applicability of the model in making a quantitative relationship between the dose level and a biomarker that could be a useful clinical endpoint in treating SCD patients in VOC using a mechanistic PKPD model.

[0216] The model produces a “Base Case” quantitative relationship between dose level of rADAMTS13 and active VWF output from the pharmacodynamic model. The simulation was able to predict the dose effect on cleaved products consistent with clinical data for a virtual patient under a typical TTP condition vs. a population of patient data. [0217] The model shows that a goal of maintaining active VWF level below remission levels for five days can be achieved with a single dose of 120 U/kg rADAMTS13, a single dose of 80 U/kg rADAMTS13 plus a booster does of 40 U/kg at 3 days, or a single dose of 80 U/kg plus a booster dose of 40 U/kg at 2 days.

[0218] EXAMPLE QUANTITATIVE SYSTEMS PHARMACOLOGY MODEL APPLICATIONS

[0219] In some embodiments, the QSP model and/or virtual population described herein may be implemented to conduct a virtual clinical trial. FIG. 15A is a flow chart illustrating a computer implemented system and method for simulating interactions of ADAMTS13 and VWF, according to some non-limiting embodiments.

[0220] At act 101, a QSP model for modeling interactions between AD AMTS 13 and VWF may be established. For example, the QSP model may comprise one or more PK models, one or more PD models, and one or more clinical outcome models as shown in FIG. 3. At act 102, the QSP model may be described with appropriate mathematical equations (e.g., a plurality of ordinary differential equations). In some embodiments, the mathematical equations may describe reactions governing the ADAMTS13-VWF interactions modeled by the QSP model, for example, as shown in Tables 2a- 2b.

[0221] At act 104, parameter estimates for parameterizing the QSP model may be acquired from literature and/or clinical data, as described herein. The parameter estimates may be applied to the QSP model to parameterize the model.

[0222] At act 106, the QSP model may be verified by comparing simulation output from the model to literature and/or clinical data. For example, the QSP model may be applied to obtain output for one or more biomarkers (e.g., cleaved VWF fragments, stretched ULVWF, platelet count, LDH, etc.), and the output may be compared to biomarker values from clinical data to verify the accuracy of the QSP model. As described herein, prior to verification, the QSP model may be calibrated with one or more sets of data (e.g., in vitro data, in vivo data).

[0223] At act 108, virtual population development may begin by establishing a total number of virtual patients and duration of a virtual clinical trial. For example, in some embodiments, the total number of virtual patients is 1000. The duration of the virtual clinical trial may refer to the length of time the ADAMTS13-VWF interactions of a patient population is observed, including a time period during which a therapeutic intervention is applied to the virtual patient population.

[0224] At acts 110-112, PK parameters and disease predictive descriptors and their associated variabilities may be obtained from real patient data. For example, in some embodiments, clinical data may be used to inform the PK parameters and disease predictive descriptors that are to be applied to the virtual patient population. At act 114, virtual PK parameters and virtual disease predictive descriptors may be obtained, for example, based on the PK parameters and disease predictive descriptors obtained from clinical data. At acts 116-118, the virtual PK parameters and disease predictive descriptors may be randomly assigned to virtual patients in the virtual patient population. [0225] At act 120, the QSP model may be used to simulate disease occurrence in virtual patients. For example, in some embodiments, the QSP model may be used to simulate occurrence a VOC in virtual patients and to reflect the resulting protein levels of the attack. At act 122, the virtual patient disease data may be compared to disease profiles of real subjects with SCD and/or TTP.

[0226] At act 124, the QSP model may be used to evaluate the effectiveness of a therapeutic intervention (e.g., for treating AD AMTS 13 inhibition or deficiency and/or excess Hb or thrombospondin). For example, parameters indicating the virtual patient population is being administered a dosage of a drug (e.g., rADAMTS13) according to a dosage regimen or has received a plasma exchange or administration of frozen plasma of healthy donor plasma may be input into the QSP model.

[0227] At act 126, the virtual clinical trial may be executed. For example, the resulting effect of administration of the drug applied in act 124 to the virtual patient population may be observed. In some embodiments, biomarker levels may be evaluated, to determine a relative change in biomarker levels resulting from administration of the therapeutic intervention. In some embodiments, duration of time in which biomarker levels are above or below a threshold may be observed. In some embodiments, the virtual clinical trial data may be compared with data from real subjects.

[0228] In some embodiments, the QSP model may be used to evaluate AD AMTS 13 and VWF interactions, as shown in FIG. 15B. FIGS. 15B illustrates an example method 1500 for modeling AD AMTS 13 and VWF interactions, according to some non-limiting embodiments.

[0229] Method 1500 begins at act 1502 where a QSP model representing ADAMTS13 and VWF interactions is obtained, for example, using any of the techniques for developing, parameterizing, calibrating and/or verifying a QSP model described herein. The QSP model may comprise one or more PK models, one or more PD models, and/or one or more clinical outcome models, as shown in FIG. 2. In some embodiments, QSP model may comprise a plurality of ordinary differential equations. In some embodiments, the mathematical equations may describe interactions between ADAMTS13 and VWF, in endogenous and/or recombinant form, and/or other associated biomarkers, such as Hb and/or thrombospondin, for example, as shown in Tables 2a-2b.

[0230] At act 1504, disease predictive descriptors may be obtained. For example, disease predictive descriptors may include a virtual patient’s propensity to experience VOC or experience thrombosis. In particular, the disease predictive descriptors may include a concentration of endogenous AD AMTS 13, Hb, and/or thrombospondin. In some embodiments, the disease predictive descriptors are determined at least in part by a Poisson process informed by known data regarding the disease predictive descriptors. [0231] At act 1506, the disease predictive descriptors may be assigned to a data set. For example, the data set may represent a virtual patient population for which the QSP model is applied. The virtual population may comprise a plurality of data sets. Each data set (e.g., Patienti) may represent an individual virtual patient of the virtual population and may have one or more variables (e.g., for assigning PK parameters and/or disease predictive descriptors) defining one or more characteristics of the virtual patient.

[0232] At act 1508, the data set may be processed using the QSP model (e.g., by inputting the data set to the QSP model) to obtain processed data. The processed data may include, for example, biomarker concentrations (e.g., cleaved VWF fragments, stretched ULVWF, platelet count, LDH) for a virtual patient. In some embodiments, the method further comprises displaying the processed data. [0233] Evaluating Effectiveness of New or Existing Drugs for Treating ADAMTS13 Deficiency or Inhibition

[0234] In some embodiments, the QSP model may be used to evaluate the effectiveness of a therapeutic intervention (e.g., rADAMTS13, plasma exchange of donor plasma or administration of frozen plasma including healthy levels of endogenous AD AMTS 13). For example, the inventors have recognized that the QSP model provides data that may be impractical or impossible to clinically obtain. The QSP model, therefore, provides for evaluation of therapeutic interventions in a cheaper and faster manner without the need for testing on human subjects.

[0235] FIG. 15C illustrates an example method 1520 for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large VWF multimers, according to some non-limiting embodiments.

[0236] Method 1520 begins at act 1522 where PK parameters for a virtual data set may be obtained. As described herein, the PK parameters may be used to describe the disposition of a drug in a patient. The virtual data set may reflect a virtual patient population on which the virtual clinical trial executed by the QSP model is run. The dosage and characteristics of the drug administered to each virtual patient may be reflected by the PK parameters.

[0237] At act 1524, disease predictive descriptors may be determined for the virtual data set. In some embodiments, the disease predictive descriptors may be informed by clinical data.

[0238] At act 1526, the PK parameters and disease predictive descriptors are assigned to the virtual data set. In some embodiments, the disease predictive descriptors may be assigned using a Poisson process.

[0239] At act 1528, the virtual data set may be processed by a QSP model to obtain processed data. At act 1530, an indicator of the effectiveness of the administered drug may be obtained. In some embodiments, the processed data output by the QSP model may include one or more biomarker concentrations (e.g., uncleaved stretched ULVWF, cleaved VWF fragments, LDH, platelet count). The biomarker concentrations may be used to determine the effectiveness of the administered drug in reducing the concentration of uncleaved stretched ULVWF. For example, reduced levels of uncleaved stretched ULVWF or increased levels of cleaved VWF fragments may indicate the drug is effectively regulating uncleaved ULVWF levels and inhibiting VOC or thrombosis. In some embodiments, the biomarker levels obtained from the QSP model are compared to a threshold (e.g., a biomarker level of a healthy patient or a patient in remission).

[0240] In some embodiments, the administered drug is AD AMTS 13 in either endogenous or recombinant form. The ADAMTS13 may be administered in any suitable manner . For example, in some embodiments, administration may comprise a plasma exchange of a patient’s plasma with a healthy donor’s plasma to increase the amount of naturally occurring (endogenous) AD AMTS 13 in the patient. In some embodiments, administration may include administration of frozen donor plasma to the patient. For example the donor plasma may contain endogenous ADAMTS13 such that the exchange increases the concentration of endogenous AD AMTS 13 in the recipient patient. In some embodiments, a recombinant form of AD AMTS 13 (rADAMTS 13) may be administered to the patient.

[0241] Evaluating Efficacy of Combination Therapies

[0242] In some embodiments, the QSP model may be used to evaluate the effectiveness of combination therapies for treating ADAMTS13 inhibition or deficiency and/or excess Hb. For example, in some embodiments, a virtual patient may be administered two or more drugs for reducing a concentration of uncleaved stretched ULVWF and the QSP model may be used to evaluate the effectiveness of the combination therapy based on output from the clinical outcome model (e.g., biomarker concentrations).

[0243] Evaluating Effectiveness of Dosages

[0244] In some embodiments, the QSP model may be used to evaluate the effectiveness of a particular dosage of an administered drug, such as administration of rADAMTS 13 or a plasma exchange where the donor plasma includes a particular concentration of endogenous AD AMTS 13. The methods described herein for using the QSP model to evaluate the effectiveness of a drug may likewise be applied to evaluate the effectiveness of a particular dosage of a drug (e.g., a dose of AD AMTS 13 in either recombinant or endogenous form).

[0245] Evaluating Effectiveness of Dosage Frequencies and/or Dosage Regimens

[0246] In some embodiments, the QSP model may be used to evaluate the effectiveness of a particular dosage frequency and/or dosage regimen (for example, evaluating the manner or frequency in which a dose is applied). The methods described herein for using the QSP model to evaluate the effectiveness of a drug may likewise be applied to evaluate the effectiveness of a dosage frequency and/or dosage regimen. [0247] Evaluating the Effect of Non-adherence to a Dosage Schedule

[0248] In some embodiments, the QSP model may be used to evaluate the effect of nonadherence to a dosage schedule (e.g., missing one or more scheduled dosages). FIG. 15D illustrates an example method 1540 for determining an effect of non-adherence to a dosing regimen of a drug (ADAMTS13 in either recombinant or endogenous form) in reducing a concentration of uncleaved ultra-large VWF multimers, according to some non-limiting embodiments.

[0249] Method 1540 begins at act 1542 where PK parameters for a virtual data set may be obtained. As described herein, the PK parameters may be used to describe the disposition of a drug in a patient. The virtual data set may reflect a virtual patient population on which the virtual clinical trial executed by the QSP model is run. The dosage and characteristics of the drug administered to each virtual patient may be reflected by the PK parameters. In particular, the PK parameters may reflect one or more missed dosages according to the method 1540.

[0250] At act 1544, disease predictive descriptors may be determined for the virtual data set. In some embodiments, the disease predictive descriptors may be informed by clinical data.

[0251] At act 1546, the PK parameters and disease predictive descriptors are assigned to the virtual data set. In some embodiments, the disease predictive descriptors may be assigned using a Poisson process.

[0252] At act 1548, the virtual data set may be processed by a QSP model to obtain processed data. At act 1550, an effect of non-adherence (including non-adherence frequency) may be determined. For example, the simulation output may provide levels of one or more biomarkers, including changes in biomarker level over time. The simulation output may be used as described herein for determining the effect of missing one or more scheduled dosages. In some embodiments, the effects of different frequencies of non- adherence (e.g., full adherence, 15% missed dose, 20% missed dose, etc.) may be compared to determine the effects of non-adherence on reducing concentration of cleaved ULVWF.

[0253] RESULTS

[0254] The inventors have recognized that the QSP model described herein may assist clinicians in determining first-in-human dosages of rADAMTS13 by providing a quantitative relationship between dose level and a biomarker that may provide a useful clinical target in treating SCD patients experiencing VOC. The QSP model may, in some embodiments, be used to evaluate administration of a plasma exchange or frozen plasma to the patient. Based on information regarding the molecular interactions of certain biomarkers that play a role in ADAMTS13-VWF interactions, and their roles in SCD and TTP diseases, a PD representation of the mechanism of action by which ADAMTS13 cleaves ULVWF to smaller VWF fragments as well as the inhibition of this action by extracellular hemoglobin was incorporated into the QSP model. The uncleaved Hb-bound ULVWF multimers accumulate in the plasma leading to cell adhesion and events such as thrombosis and vascular occlusion. Active VWF, or the uncleaved VWF in stretched form where the binding sites are exposed for adhesion to platelets as well as binding to Hb and ADAMTS13, is provided as an output of the clinical outcome model and may provide a useful data point for clinicians in determining appropriate dosages of AD AMTS 13 according to any therapeutic intervention described herein.

[0255] As described herein, the QSP model may be used to obtain information regarding an impact of a dose of ADAMTS13 on VWF multimer activity. VWF multimer concentrations may provide a useful biomarker in a number of applications. FIG. 15E illustrates an example method 1560 for determining a concentration of VWF multimers in response to administration of AD AMTS 13, according to some non-limiting embodiments.

[0256] Method 1560 begins at act 1562 where PK parameters for a virtual data set may be obtained. As described herein, the PK parameters may be used to describe the disposition of a drug in a patient. The virtual data set may reflect a virtual patient population on which the virtual clinical trial executed by the QSP model is run. The dosage and characteristics of the drug administered to each virtual patient may be reflected by the PK parameters.

[0257] At act 1564, disease predictive descriptors may be determined for the virtual data set. In some embodiments, the disease predictive descriptors may be informed by clinical data.

[0258] At act 1566, the PK parameters and disease predictive descriptors are assigned to the virtual data set. In some embodiments, the disease predictive descriptors may be assigned using a Poisson process. [0259] At act 1568, the virtual data set may be processed by a QSP model to obtain processed data. At act 1570, a concentration of VWF multimers based on the processed data may be obtained. As described herein, in some embodiments, the processed data output by the QSP model may include concentrations of one or more biomarkers related to VWF interactions, such as concentrations of uncleaved stretched ULVWF, cleaved VWF fragments, platelet count, and/or LDH. Accordingly, the model output may be used at act 1570 to obtain the concentration of VWF multimers (e.g., cleaved or uncleaved VWF multimers, stretched or globular ULVWF multimers or a total thereof).

[0260] The QSP model described herein was applied to make first-in-human dose predictions for treating SCD patients in VOC. The amount of active VWF predicted by the model was used as a clinical outcome indicator: specifically, the reduction in the amount of active VWF with the treatment and duration of maintaining the active VWF below the level in remission. The simulation conditions are shown in Table 7.

[0261] Table 9: AUC at different doses and duration for single dose administration, drug only or drug + endogenous ADAMTS13

[0262] FIGS. 16A-16B illustrate model results of pharmacokinetic parameters and active VWF amount for various doses and dose schedules of recombinant AD AMTS 13, according to some non-limiting embodiments.

[0263] FIG. 16A shows graphs 1610, 1620, 1630 illustrating PK simulation results under the Base Case conditions for single dose, single dose plus a booster dose at 3 days, and single dose plus a booster at 2 days. The booster dose was half of the single dose amount.

Table 9 shows the drug exposure (drug only or with endogenous ADAMTS13) at different doses and durations.

[0264] FIG. 16B show graphs 1640, 1650, 1660 illustrating model results for the active VWF amount over time from the VOC occurrence (-20 hours) to the response of the treatment (starting at 0 hour) for the 3 dose regimens. For all the cases, the reduction in the active VWF level upon the start of the treatment reaches minimum level in less than 12 hours and the minimum is reached sooner with higher doses. The horizontal dotted line 1645 indicates the active VWF amount in remission, 221 ng/ml, in this Base Case. It was assumed that this level represents a threshold for the VOC patient recovering to the remission state (i.e., the patient has recovered to the remission state when the active VWF amount falls below this threshold). Applying this assumption, a single dose of 120 U/kg enabled the VOC patient to keep the active VWF level below that of the remission level throughout 4 days since the start of the treatment, whereas in both single plus booster at 3 days regimen, 80 U/kg enabled the patient to maintain the active VWF level below that of the remission level for nearly 5 days. Since the booster dose used half of the single dose amount, the total amount of rADAMTS13 used in these regimens is the same, 120 U/kg.

[0265] Based on the above results, FIG. 17 compares the number of days of active VWF under the remission level for the three dose regimens. From 80 U/kg single dose amount, adding the booster approximately doubled the number of days under the remission level. When compared on a total dosage basis, the booster regimen prolonged the duration less than a day (c.f. 80 U/kg single + booster at 2 or 3 days vs. 120 U/kg single dose regimens).

[0266] FIGS. 18A-24 illustrate example results of applying the QSP model on a virtual population receiving different doses of rADAMTS13. FIGS. 18A-18B illustrate graphs shown percentages of VWF cleavage over time for different doses of rADAMTS 13, according to some non-limiting embodiments. Based on the simulation results, 40 U/kg was predicted to result in a 23% maximum reduction of active VWF and was determined to be the minimum dose required to reduce active VWF level to remission level. An 80 U/kg dose was shown to achieve 32% reduction in active VWF while a 160 U/kg dose was shown to achieve a 50% reduction in active VWF. As shown in FIG. 18B, active VWF trough level was expected at between 5-15 hours from dose administration.

[0267] FIG. 19 illustrates graphs comparing rADAMTS 13 concentration and active VWF concentration over time for a dose of rADAMTS 13, according to some nonlimiting embodiments. In particular, graphs 1910 and 1920 illustrate simulation results for a virtual patient population receiving a 40 lU/kg dose of rADAMTS 13. Graph 1910 illustrates rADAMTS13 concentration over time for the virtual patient population. Graph 1920 illustrates active VWF over time for the virtual patient population.

[0268] FIG. 20A illustrates a graph showing time profiles of rADAMTS13 in different types of patients and for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments. In particular, FIG. 20A compares simulation results for plasma concentration of rADATMS13 over time in virtual patients receiving 40 U/kg, 80 U/kg, and 160 U/kg of r AD AMTS 13.

[0269] FIG. 20B illustrates concentrations of active VWF in different types of patients for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments. In particular, FIG. 20B compares simulation results for level of active VWF in various patients, including healthy patients, SCD patients in remission, and SCD patients receiving 0 U/kg r AD AMTS 13, 40 U/kg r AD AMTS 13, 80 U/kg r AD AMTS 13, and 160 U/kg rADAMTS13 doses, respectively. Using the level of active VWF in SCD patients in remission as a threshold, FIG. 20B illustrates the effectiveness of the various doses in reducing amounts of active VWF in SCD patients.

[0270] FIG. 21 illustrates a duration in which VWF concentration is under remission levels for various doses of recombinant AD AMTS 13, according to some non-limiting embodiments. In particular, FIG. 21 illustrates simulation results for a number of days active VWF concentration was below a threshold for various doses of rADAMTS13. Again, the level of active VWF in an SCD patient under remission was used as a threshold. FIG. 21 shows that in a dose range of 40-160 U/kg, a single dose of 120 U/kg was predicated to enable a patient with a VOC event to keep active VWF level below that of the remission level throughout 3-6 days. FIG. 21 further shows that the dose of 40 U/kg appears to provide marginal impact on reduction of active VWF level to below remission threshold after a single treatment.

[0271] FIG. 22 illustrates a graph showing the effect of baseline total VWF on a duration in which active VWF is under remission level for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments. In particular, FIG. 22 compares simulation results for active VWF level duration under threshold with baseline total VWF amount in a virtual patient receiving various doses of rADAMTS13. FIG. 22 suggests that the level of baseline VWF has a negligible impact on the number of days active VWF level is below remission level. [0272] FIG. 23 illustrates a graph showing an effect of cell-free Hb level on a duration that active VWF is under remission level, according to some non-limiting embodiments. In particular, FIG. 23 illustrates simulation results for active VWF level duration under threshold with cell-free Hb concentration in virtual patients receiving various doses of rADAMTS13. Free hemoglobin is elevated in patients with SCD. FIG. 23 illustrates that with higher levels of free hemoglobin present during VOC, a high dose of rADAMTS13 is needed to maintain active VWF level under threshold for a longer duration.

[0273] FIG. 24 illustrates an effect of Hemoglobin binding affinity with VWF on a duration in which active VWF is under remission levels for different doses of recombinant AD AMTS 13, according to some non-limiting embodiments. In particular, FIG. 24 illustrates the effect of altering the parameter in the QSP model representing Hb binding affinity with VWF, KDvs Hb, for various doses of rADAMTS13. As described herein, the binding affinity between stretched VWF and hemoglobin was the only model parameter to be recalibrated from literature data to match in vivo data. FIG. 24 illustrates a range in which the KDvs Hb constant may be varied with negligible effect on the QSP model.

[0274] EVALUATING MODEL SENSITIVITY AND OUTCOMES

[0275] As described herein, the QSP model may be parameterized with a number of parameters that play a role in VWF-ADAMTS13 interactions. The parameters used in the QSP model described herein are provided in Table 3. In some embodiments, investigation of the sensitivity of model output to variation in one or more of these parameters may be performed.

[0276] FIGS. 25A-25C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to the binding affinity of hemoglobin to VWF constant KDvs_Hb Used in the model, according to some non-limiting embodiments. FIGS. 25A-C shows that changing the binding affinity between Hb and active VWF (KDvs i ih) has a negligible effect on the number of days of active VWF under the remission level for all the dosage regimens studied.

[0277] FIGS. 26A-26C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to total VWF level, according to some nonlimiting embodiments. Similarly, FIGS. 26A-26C shows that total VWF level also has an insignificant effect on the number of days below the remission level for active VWF for all the dosage regimens studied.

[0278] FIGS. 27A-27C are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to endogenous AD AMTS 13 activity, according to some non-limiting embodiments. FIGS. 27A-27C shows the sensitivity to the activity of the endogenous AD AMTS 13 on the number of days of active VWF under the remission level. The results indicate that a higher activity for the endogenous AD AMTS 13 requires a greater dose. For example, to maintain active VWF concentration under the threshold for 4 days or more, 40, 80, 120 U/kg single dose levels were required as the ADAMTS13 activity increased from 0.5, 1, 1.5 U/pg, respectively. The activity rADAMTS13 was fixed at the Base Case value of 1.5 U/ pg.

[0279] FIGS. 28A-28G are graphs illustrating a sensitivity of the number of days that active VWF is under remission levels to remission Hb and vaso-occlusive crises levels, according to some non-limiting embodiments. FIGS. 28A-G shows sensitivity to remission Hb and VOC Hb levels on the number of days of active VWF under the remission level. Case 5F is the Base Case described above. Relative to the Base Case, the dose requirement increases considerably at the lowest Hb level in remission (cases 5 A, 5B, 5C). Among these, the dose requirement is the highest at the highest Hb level in VOC, Case 5C. Overall the dose requirement increased with the increasing difference between the Hb level in VOC and remission, in the diagonal direction from Case 5D to 5C. FIG. 28H is a graph comparing the results of the graphs shown in FIGS. 28A-28G, according to some non-limiting embodiments.

[0280] FIGS. 29A-29B are graphs illustrating a dose response of hemoglobin-bound VWF amount and active VWF amount ten hours after a dosage of recombinant AD AMTS 13, according to some non-limiting embodiments. Two other potential efficacious targets were also investigated including 1) hemoglobin bound VWF and 2) active VWF after 10 hours of dose. Ten hours was selected as model results showed that active VWF had the highest reduction after 10 hours of dose. FIGS. 29A-29B showed each target against different dosages (single dose administration). Table 10 summarizes the dose required to reach remission level and the range if 10-40% reduction comparing with no drug condition is assumed to be acceptable. [0281] Table 10: Required does to reach remission level and dose range using hemoglobin bound VWF and active VWF as efficacious targets

[0282] In summary, the model results suggest use of dose range of 3O-13OU/kg (e.g., the common range between the two target options), which is consistent with the efficacious range 40-200 U/kg using number of days of active VWF under the remission level as the efficacious target.

[0283] FIG. 30 is a graph illustrating sensitivity of active VWF amounts predicted by the QSP model to variations of different model parameters, according to some non-limiting embodiments.

[0284] EXAMPLE COMPUTING SYSTEMS

[0285] FIG. 31 depicts, schematically, an illustrative computing device 1000 on which any aspect of the present disclosure may be implemented, according to some nonlimiting embodiments.

[0286] FIG. 31 shows, schematically, an illustrative computer 1000 on which any aspect of the present disclosure may be implemented. In the embodiment shown in FIG. 31, the computer 1000 includes a processing unit 1001 having one or more computer hardware processors and one or more articles of manufacture that comprise non-transitory computer-readable storage media (e.g., system memory 1002) that may include, for example, volatile and/or non-volatile memory. The memory 1002 may store one or more instructions to program the processing unit 1001 to perform any of the functions described herein. The computer 1000 may also include other types of non-transitory computer-readable media, such as storage 1005 (e.g., one or more disk drives) in addition to the system memory 1002. The storage 1005 may also store one or more application programs and/or external components used by application programs (e.g., software libraries), which may be loaded into the memory 1002. To perform any of the functionality described herein, processing unit 1001 may execute one or more processorexecutable instructions stored in the one or more non-transitory computer-readable storage media (e.g., memory 1002, storage 1005), which may serve as non-transitory computer-readable storage media storing processor-executable instructions for execution by the processing unit 1001.

[0287] The computer 1000 may have one or more input devices and/or output devices, such as devices 1006 and 1007 illustrated in FIG. 31. These devices can be used, among other things, to present a user interface. Examples of output devices that can be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, the input devices 1007 may include a microphone for capturing audio signals, and the output devices 1006 may include a display screen for visually rendering, and/or a speaker for audibly rendering, recognized text.

[0288] As shown in FIG. 31, the computer 1000 may also comprise one or more network interfaces (e.g., the network interface 10010) to enable communication via various networks (e.g., the network 10020). Examples of networks include a local area network or a wide area network, such as an enterprise network or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks, wired networks or fiber optic networks.

[0289] CONCLUSION

[0290] Having thus described several aspects of at least one embodiment, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be within the spirit and scope of the present disclosure. Accordingly, the foregoing description and drawings are by way of example only.

[0291] The above-described embodiments of the present disclosure can be implemented in any of numerous ways. For example, the embodiments may be implemented using hardware, software or a combination thereof. When implemented in software, the software code can be executed on any suitable processor or collection of processors, whether provided in a single computer or distributed among multiple computers.

[0292] Also, the various methods or processes outlined herein may be coded as software that is executable on one or more processors that employ any one of a variety of operating systems or platforms. Additionally, such software may be written using any of a number of suitable programming languages and/or programming or scripting tools, and also may be compiled as executable machine language code or intermediate code that is executed on a framework or virtual machine.

[0293] In this respect, the concepts disclosed herein may be embodied as a non- transitory computer-readable medium (or multiple computer-readable media) (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other non-transitory, tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement the various embodiments of the present disclosure discussed above. The computer-readable medium or media can be transportable, such that the program or programs stored thereon can be loaded onto one or more different computers or other processors to implement various aspects of the present disclosure as discussed above.

[0294] The terms “program” or “software” are used herein to refer to any type of computer code or set of computer-executable instructions that can be employed to program a computer or other processor to implement various aspects of the present disclosure as discussed above. Additionally, it should be appreciated that according to one aspect of this embodiment, one or more computer programs that when executed perform methods of the present disclosure need not reside on a single computer or processor, but may be distributed in a modular fashion amongst a number of different computers or processors to implement various aspects of the present disclosure.

[0295] Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Typically, the functionality of the program modules may be combined or distributed as desired in various embodiments.

[0296] Also, data structures may be stored in computer-readable media in any suitable form. For simplicity of illustration, data structures may be shown to have fields that are related through location in the data structure. Such relationships may likewise be achieved by assigning storage for the fields with locations in a computer-readable medium that conveys relationship between the fields. However, any suitable mechanism may be used to establish a relationship between information in fields of a data structure, including through the use of pointers, tags or other mechanisms that establish relationship between data elements.

[0297] Various features and aspects of the present disclosure may be used alone, in any combination of two or more, or in a variety of arrangements not specifically discussed in the embodiments described in the foregoing and is therefore not limited in its application to the details and arrangement of components set forth in the foregoing description or illustrated in the drawings. For example, aspects described in one embodiment may be combined in any manner with aspects described in other embodiments.

[0298] Also, the concepts disclosed herein may be embodied as a method, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0299] The terms “substantially”, “approximately”, and “about” may be used to mean within ±20% of a target value in some embodiments, within ±10% of a target value in some embodiments, within ±5% of a target value in some embodiments, within ±2% of a target value in some embodiments. The terms “approximately” and “about” may include the target value.

[0300] Use of ordinal terms such as “first,” “second,” “third,” etc. in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed, but are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term) to distinguish the claim elements.

[0301] Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," "having," “containing,” “involving,” and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

Claims

- 62 -CLAIMS What is claimed is:

1. A computer-implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker.

2. The computer- implemented method of claim 1, further comprising displaying the processed data.

3. The computer- implemented method of claim 1 or any other preceding claim, further comprising: determining pharmacokinetic parameters; assigning the pharmacokinetic parameters to the virtual patient population; determining therapeutic intervention data based on administration of an administered drug; and processing the therapeutic intervention data and the virtual patient population with the QSP model to determine effectiveness of the administered drug.

4. The computer- implemented method of claim 3, wherein the administered drug comprises administration of endogenous and/or recombinant AD AMTS 13.

5. The computer- implemented method of claim 4, wherein administration of the endogenous AD AMTS 13 comprises a plasma exchange with plasma having endogenous AD AMTS 13. - 63 -

6. The computer- implemented method of claim 1 or any other preceding claim, wherein the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers.

7. The computer- implemented method of claim 6, wherein the QSP model simulates a conversion of globular VWF multimers to stretched VWF multimers.

8. The computer- implemented method of claim 1 or any other preceding claim, wherein the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

9. The computer- implemented method of claim 1 or any other preceding claim, wherein the at least one biomarker comprises the ULVWF multimers, cleaved VWF fragments, lactate dehydrogenase and/or platelet cells.

10. The computer-implemented method of claim 6, wherein the ULVWF multimers comprises stretched ULVWF multimers and globular ULVWF multimers.

11. The computer- implemented method of claim 6, wherein the cleaved VWF fragments comprise stretched cleaved VWF fragments and globular cleaved VWF fragments.

12. The computer- implemented method of claim 1 or any other preceding claim, further comprising using the processed data to determine whether the concentration of the at least one biomarker is below a first threshold or above a second threshold.

13. The computer-implemented method of claim 12, further comprising using the processed data to determine a duration in which the concentration of the at least one biomarker is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold.

14. The computer- implemented method of claim 1 or any other preceding claim, further comprising using the processed data to determine a change in the concentration of the at least one biomarker over time. - 64 -

15. The computer- implemented method of claim 1 or any other preceding claim, wherein the QSP model comprises a plurality of differential equations representing one or more biological reactions.

16. The computer- implemented method of claim 3, wherein the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics of a patient to whom the administered drug is administered.

17. The computer-implemented method of claim 16, wherein the one or more biographical characteristics comprise at least one of height, weight, age, or gender.

18. The computer- implemented method of claim 1 or any other preceding claim, wherein the disease predictive descriptors comprise one or more parameters characterizing a concentration of ADAMTS13 in a patient.

19. The computer- implemented method of claim 18, wherein the patient comprises a patient having sickle cell disease.

20. The computer- implemented method of claim 18, wherein the patient comprises a patient having a congenital thrombotic thrombocytopenic purpura (eTTP).

21. The computer implemented method of claim 18, wherein the patient comprises a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP).

22. The computer- implemented method of claim 1 or any other preceding claim, wherein the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient.

23. The computer- implemented method of claim 22, wherein the pharmacokinetic parameters and disease predictive descriptors are assigned to the one or more variables of each data set.

24. A system, comprising: - 65 - at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker.

25. At least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for modeling AD AMTS 13 and von Willebrand factor (VWF) interactions, the method comprising: obtaining a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions including a mechanism by which ADAMTS13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; determining disease predictive descriptors; assigning the disease predictive descriptors to a virtual patient population; and processing the virtual patient population using the QSP model to provide processed data, wherein the processed data comprises a concentration of at least one biomarker. - 66 -

26. A computer-implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which ADAMTS13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

27. The computer- implemented method of claim 26, further comprising displaying the processed data.

28. The computer- implemented method of claim 26 or any other preceding claim, wherein the administered drug comprises endogenous and/or recombinant AD AMTS 13.

29. The computer-implemented method of claim 28, wherein the administered drug comprises plasma of a donor patient.

30. The computer-implemented method of claim 29, wherein the plasma comprises frozen plasma.

31. The computer- implemented method of claim 26 or any other preceding claim, wherein the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers.

32. The computer- implemented method of claim 31, wherein the QSP model simulates a conversion of globular VWF multimers to stretched VWF multimers.

33. The computer- implemented method of claim 26 or any other preceding claim, wherein the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

34. The computer-implemented method of claim 26 or any other preceding claim, wherein the at least one biomarker comprises the uncleaved ULVWF multimers, cleaved VWF fragments, lactate dehydrogenase and/or platelet cells.

35. The computer- implemented method of claim 26 or any other preceding claim, wherein the indicator of the effectiveness of the administered drug is obtained at least in part by comparing the processed data to known data indicating a threshold concentration of the at least one biomarker.

36. The computer- implemented method of claim 35, wherein the known data comprises biomarker amounts of an untreated subject with sickle cell disease congenital thrombotic thrombocytopenic purpura, and/or immune mediated thrombotic thrombocytopenic purpura.

37. The computer-implemented method of claim 35, wherein the known data comprises biomarker amounts of a subject without sickle cell disease, congenital thrombotic thrombocytopenic purpura, or immune mediated thrombotic thrombocytopenic purpura.

38. The computer- implemented method of claim 35, wherein the known data comprises biomarker amounts of a subject with sickle cell disease in remission.

39. The computer- implemented method of claim 26 or any other preceding claim, further comprising using the processed data to determine whether the concentration of the at least one biomarker is below a first threshold or above a second threshold.

40. The computer-implemented method of claim 39, further comprising using the processed data to determine a duration in which the concentration of the at least one biomarker is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold.

41. The computer- implemented method of claim 26 or any other preceding claim, further comprising using the processed data to determine a change in the concentration of the at least one biomarker over time.

42. The computer- implemented method of claim 26 or any other preceding claim, wherein the QSP model comprises a plurality of differential equations representing one or more biological reactions.

43. The computer- implemented method of claim 26 or any other preceding claim, wherein the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics of a patient to whom the administered drug is administered.

44. The computer- implemented method of claim 43, wherein the one or more biographical characteristics comprise at least one of height, weight, age, or gender.

45. The computer- implemented method of claim 26 or any other preceding claim, wherein the disease predictive descriptors comprise one or more parameters characterizing a concentration of ADAMTS13 in a patient.

46. The computer- implemented method of claim 45, wherein the patient comprises a patient having sickle cell disease.

47. The computer- implemented method of claim 45, wherein the patient comprises a patient having a congenital thrombotic thrombocytopenic purpura (eTTP).

48. The computer implemented method of claim 45, wherein the patient comprises a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP).

49. The computer- implemented method of claim 26 or any other preceding claim, wherein the virtual patient population comprises a plurality of data sets, each data set of - 69 - the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient.

50. The computer-implemented method of claim 49, wherein the pharmacokinetic parameters and disease predictive descriptors are assigned to the one or more variables of each data set.

51. A system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for determining an effectiveness of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which ADAMTS13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

52. At least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effectiveness of an administered drug in reducing a - 70 - concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model to obtain processed data, wherein the QSP model simulates von Willebrand factor (VWF) interactions with AD AMTS 13 including a mechanism by which ADAMTS13 cleaves the uncleaved ULVWF multimers and inhibition thereof by extracellular hemoglobin, and the processed data comprises a concentration of at least one biomarker; and using the processed data to obtain an indicator of the effectiveness of the administered drug in reducing the concentration of uncleaved ULVWF multimers.

53. A computer-implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultralarge von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of nonadherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non-adherence on reducing the concentration of uncleaved ULVWF multimers. - 71 -

54. The computer-implemented method of claim 53, wherein the processed data includes a frequency in which the amount of uncleaved ULVWF fragments exceeds a threshold.

55. The computer- implemented method of claim 53 or any other preceding claim, wherein the processed data includes a percentage by which the concentration of uncleaved ULVWF fragments exceeds a threshold.

56. The computer- implemented method of claim 53 or any other preceding claim, wherein using the processed data to determine the effect of the frequency of nonadherence includes comparing the processed data to known data.

57. The computer-implemented method of claim 53 or any other preceding claim, further comprising displaying the processed data.

58. The computer- implemented method of claim 53 or any other preceding claim, wherein the administered drug comprises endogenous and/or recombinant AD AMTS 13.

59. The computer- implemented method of claim 58, wherein the administered drug comprises plasma of a donor patient.

60. The computer-implemented method of claim 59, wherein the plasma comprises frozen plasma.

61. The computer- implemented method of claim 53 or any other preceding claim, wherein the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers.

62. The computer- implemented method of claim 61, wherein the QSP model simulates a conversion of globular VWF multimers to stretched VWF multimers. - 72 -

63. The computer- implemented method of claim 53 or any other preceding claim, wherein the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

64. The computer- implemented method of claim 53 or any other preceding claim, wherein the processed data comprises the concentration of the uncleaved ULVWF multimers.

65. The computer- implemented method of claim 53 or any other preceding claim, further comprising using the processed data to determine whether the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is below a first threshold or above a second threshold.

66. The computer-implemented method of claim 65, further comprising using the processed data to determine a duration in which the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is below the first threshold or the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments is above the second threshold.

67. The computer- implemented method of claim 53 or any other preceding claim, further comprising using the processed data to determine a change in the one of the concentration of the uncleaved ULVWF multimers or the concentration of cleaved von Willebrand factor fragments over time.

68. The computer-implemented method of claim 53 or any other preceding claim, wherein the QSP model comprises a plurality of differential equations representing one or more biological reactions. - 73 -

69. The computer- implemented method of claim 53 or any other preceding claim, wherein the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics of a patient to whom the administered drug is administered.

70. The computer-implemented method of claim 69 or any other preceding claim, wherein the one or more biographical characteristics comprise at least one of height, weight, age, or gender.

71. The computer- implemented method of claim 53 or any other preceding claim, wherein the disease predictive descriptors comprise one or more parameters characterizing a concentration of ADAMTS13 in a patient.

72. The computer- implemented method of claim 71, wherein the patient comprises a patient having sickle cell disease.

73. The computer- implemented method of claim 71, wherein the patient comprises a patient having a congenital thrombotic thrombocytopenic purpura (eTTP).

74. The computer implemented method of claim 71, wherein the patient comprises a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP).

75. The computer- implemented method of claim 53 or any other preceding claim, wherein the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient.

76. The computer- implemented method of claim 75, wherein the pharmacokinetic parameters and disease predictive descriptors are assigned to the one or more variables of each data set. - 74 -

77. A system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of nonadherence to the dosing regimen; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non-adherence on reducing the concentration of uncleaved ULVWF multimers.

78. At least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining an effect of non-adherence to a dosing regimen of an administered drug in reducing a concentration of uncleaved ultra-large von Willebrand factor (ULVWF) multimers, the method comprising: determining pharmacokinetic parameters of the administered drug for a virtual patient population, wherein the pharmacokinetic parameters include a frequency of non- adherence to the dosing regimen; - 75 - determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin and the processed data comprises one of the concentration of the uncleaved ULVWF multimers or a concentration of cleaved von Willebrand factor fragments; and using the processed data to determine an effect of the frequency of non-adherence on reducing the concentration of uncleaved ULVWF multimers.

79. A computer-implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

80. The computer-implemented method of claim 79, wherein the administration of AD AMTS 13 comprises administration of recombinant AD AMTS 13.

81. The computer- implemented method of claim 79 or any other preceding claim, wherein the administration of ADAMTS13 comprises administration of plasma of a donor patient. - 76 -

82. The computer- implemented method of claim 81, wherein the plasma comprises frozen plasma.

83. The computer-implemented method of claim 79 or any other preceding claim, wherein the VWF multimers comprise one of uncleaved ultra-large VWF multimers or cleaved VWF fragments.

84. The computer- implemented method of claim 79 or any other preceding claim, wherein the QSP model represents AD AMTS 13 interactions with stretched and globular VWF multimers.

85. The computer- implemented method of claim 84, wherein the QSP model simulates a conversion of globular VWF multimers to stretched VWF multimers.

86. The computer-implemented method of claim 79 or any other preceding claim, wherein the QSP model includes a binding affinity of the extracellular hemoglobin to the ULVWF multimers.

87. The computer- implemented method of claim 79 or any other preceding claim, further comprising determining whether the concentration of VWF multimers is below a first threshold or above a second threshold.

88. The computer- implemented method of claim 87, determining a duration in which the concentration of VWF multimers is below the first threshold or a duration in which the concentration of the at least one biomarker is above the second threshold.

89. The computer-implemented method of claim 79 or any other preceding claim, further comprising using the QSP model to determine a change in the concentration of VWF multimers over time. - 77 -

90. The computer-implemented method of claim 79 or any other preceding claim, wherein the QSP model comprises a plurality of differential equations representing one or more biological reactions.

91. The computer- implemented method of claim 79 or any other preceding claim, wherein the pharmacokinetic parameters comprise one or more parameters indicating how the administered drug is impacted by one or more biographical characteristics of a patient to whom the administered drug is administered.

92. The computer- implemented method of claim 91, wherein the one or more biographical characteristics comprise at least one of height, weight, age, or gender.

93. The computer- implemented method of claim 79 or any other preceding claim, wherein the disease predictive descriptors comprise one or more parameters characterizing a concentration of ADAMTS13 in a patient.

94. The computer- implemented method of claim 93, wherein the patient comprises a patient having sickle cell disease.

95. The computer- implemented method of claim 93, wherein the patient comprises a patient having a congenital thrombotic thrombocytopenic purpura (eTTP).

96. The computer implemented method of claim 93, wherein the patient comprises a patient having immune mediated thrombotic thrombocytopenic purpura (iTTP).

97. The computer- implemented method of claim 79 or any other preceding claim, wherein the virtual patient population comprises a plurality of data sets, each data set of the plurality of data sets representing a virtual patient and having one or more variables defining one or more characteristics of the virtual patient. - 78 -

98. The computer-implemented method of claim 97, wherein the pharmacokinetic parameters and disease predictive descriptors are assigned to the one or more variables of each data set.

99. A system comprising: at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer- implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.

100. At least one non-transitory computer-readable medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a computer implemented method for determining a concentration of von Willebrand factor (VWF) multimers in response to administration of AD AMTS 13, the method comprising: determining pharmacokinetic parameters of the administered ADAMTS13 for a virtual patient population; determining disease predictive descriptors for the virtual patient population; assigning the pharmacokinetic parameters and disease predictive descriptors to the virtual patient population; - 79 - processing the virtual patient population using a quantitative systems pharmacology (QSP) model representing ADAMTS13 and VWF interactions to obtain processed data, wherein the QSP model includes a mechanism by which AD AMTS 13 cleaves ultra-large VWF (ULVWF) multimers and inhibition thereof by extracellular hemoglobin; and determining the concentration of VWF multimers based on the processed data.