SYSTEMS, METHODS, AND APPARATUS FOR MODELING AND OPTIMIZING DIALYSIS EFFECTS

Abstract

Systems, methods, and devices model, identify, and predict effects of dialysis on drugs and chemical substances in patients. The systems estimate and determine effects of dialysis on elutes and drugs and solve problems with prior systems in determining effects of dialysis on drugs and dosages, especially the removal of target substances from the blood of patients during dialysis. Drug and medicine dosage adjustments for patients undergoing dialysis are made by considering the dialysis systems, patients, and drug variables and the extent to which they affect drug removal. Patients receive proper dosing by accounting for the effects of their dialysis. Systems and methods include a combination of ex vivo simulated treatments and in-silico modeling and simulation. The effects of dialysis are estimated and determined for various drugs and dosages. A reliable and effective surrogate for performing studies on patients provides guidance for use of the drugs in dialysis patients.

Claims

1. A method for determining efficacy of a dialysis therapy comprising: a) performing a dialysis treatment simulation for determining an amount of a drug removed by a dialysis treatment; wherein the dialysis treatment simulation includes: taking blood or plasma samples from an arterial port, a venous port, an effluent port, and a blood or plasma reservoir of a dialysis treatment simulation system; and determining at least one of a drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed at the ports; b) generating in silico a plurality of virtual patients for modeling a health condition based on data collected from a population of previously treated patients; wherein the collected data represents at least one measured biological response of the previously treated patients to a previously administered therapeutic regimen of the drug, wherein each virtual patient includes at least one mathematical model representing a physiological system and exhibiting a one-to-one correspondence with one of the population of previously treated patients, and wherein the at least one mathematical model for each virtual patient is fitted to data of the corresponding previously treated patient; c) applying the dialysis treatment simulation to the virtual patients over a simulated time period; d) determining one or more physiological parameters in the virtual patients in response to the dialysis treatment simulation over the simulated time period; e) applying at least one adjusted simulated dialysis therapy to the virtual patients over the same simulated time period based on the determined one or more physiological parameters of the virtual patients; f) determining one or more physiological parameters in the virtual patients in response to the at least one adjusted simulated dialysis therapy over the same simulated period of time; g) determining an optimal simulated dialysis therapy from among the simulated dialysis therapy and the at least one adjusted simulated dialysis therapy, wherein determining the optimal simulated dialysis therapy includes repeating steps e) and f) until the one or more determined physiological parameters in the virtual patients are within a desired range; and h) recommending the optimal simulated dialysis therapy to a plurality of actual patients.

2. A dialysis treatment simulation system for determining an amount of a drug removed by a dialysis treatment comprising: a dialysis filter 155 including a blood side (arterial) chamber 156 and a dialysate side (venous) chamber 157; a plasma container 102 containing the plasma or blood spiked with the drug; a sampling port 103 for removing a sample of the plasma or blood spiked with the drug and measuring at least one of the group of drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed; a plasma or blood pump 110 for moving the plasma or blood to an input of the blood side (arterial) chamber 156 of the dialysis filter 155; a sampling port 104 at the input of the blood side (arterial) chamber 156 of the dialysis filter 155 for removing a sample of the plasma or blood spiked with the drug and measuring at least one of the group of drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed; a membrane for separating the blood side (arterial) chamber 156 of the dialysis filter 155 from the dialysate side (venous) chamber 157 of the dialysis filter 155 and for removing waste products from the plasma or blood spiked with the drug; a sampling port 105 at the output of the blood side (arterial) chamber 156 of the dialysis filter 155 for removing a sample of the plasma or blood spiked with the drug and measuring at least one of the group of drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed; a dialysate reservoir container 108 containing dialysate; a dialysate pump 111 for moving dialysate from the dialysate container 108 to an input of the dialysate side (venous) chamber 157 where the dialysate carries the waste products from the plasma or blood spiked with the drug to an output of the dialysate side (venous) chamber 157; an effluent pump 112 for receiving the dialysate with the waste products (effluent fluid) and for moving effluent fluid from the output of the dialysate side (venous) chamber 157 to an effluent filter container 107; and a sampling port 106 for removing a sample of effluent fluid from the effluent filter container 107 for removing a sample of the plasma or blood spiked with the drug and measuring at least one of the group of drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed; and comparing the removed samples from the ports to determine the amount of a drug removed by the dialysis treatment.

3. A computer system for determining an efficacy of a therapy comprising: a processor configured to: a) perform a dialysis treatment simulation for determining an amount of a drug removed by a dialysis treatment; wherein the dialysis treatment simulation includes: taking blood or plasma samples from an arterial port, a venous port, an effluent port, and a blood or plasma reservoir of a dialysis treatment simulation system; and determining at least one of a drug concentration, drug rate removal, drug ratio to blood or plasma, and an amount of the drug removed at the ports; b) generate in silico a plurality of virtual patients for modeling a health condition based on data collected from a population of previously treated patients having the health condition; wherein the collected data represents at least one measured biological response of the previously treated patients to a previously administered therapeutic regimen of the drug, wherein each virtual patient includes at least one mathematical model representing a physiological system and exhibiting a one-to-one correspondence with one of the population of previously treated patients, and wherein the at least one mathematical model for each virtual patient is fitted to data of the corresponding previously treated patient; c) apply the dialysis treatment simulation to the virtual patients over a simulated time period; d) determine one or more physiological parameters in the virtual patients in response to the dialysis treatment simulation over the simulated time period; e) apply at least one adjusted simulated dialysis therapy to the virtual patients over the same simulated time period based on the determined one or more physiological parameters of the virtual patients; f) determine an optimal simulated dialysis therapy from among the simulated dialysis therapy and at least one adjusted simulated dialysis therapy by iteratively performing the following steps until the one or more determined physiological parameters in the virtual patients are with a desired range: applying at least one adjusted simulated dialysis therapy to the virtual patients over the same simulated time period based on the determined one or more physiological parameters of the virtual patients; determining one or more physiological parameters in the virtual patients in response to the at least one adjusted simulated dialysis therapy over the same simulated period of time; and g) recommend the optimal simulated therapy for application to a plurality of actual patients.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG. 1 shows an exemplary system configured to conduct simulated treatments in accordance with the invention to test a drug or analyte.

[0017] FIG. 2 shows an exemplary embodiment of a Cybernetic Dialysis™ system in accordance with the invention.

[0018] FIGS. 3A-3C show an exemplary implementation of a Cybernetic Dialysis modeling algorithm to predict and extrapolate dialysis effects in accordance with the invention.

[0019] FIG. 4 shows example arterial, venous, and plasma reservoir concentrations of creatinine for an M150 AN69 membrane in accordance with the invention and indicative of a steady reduction in the concentration of creatinine over time.

[0020] FIG. 5 shows a percentage change in creatinine concentration on a venous side vs. an arterial side of an M150 AN69 membrane (at each timepoint) in an exemplary embodiment of the invention.

[0021] FIG. 6 shows creatinine concentrations over time at the arterial and venous sides of membrane HF1400 and in the plasma reservoir in an exemplary embodiment of the invention.

[0022] FIG. 7 shows percentage changes in creatinine concentration on a venous side vs. an arterial side of an HF1400 membrane (at each timepoint) in an exemplary embodiment of the invention.

[0023] FIG. 8 shows a cumulative amount of creatinine removed from plasma for an M150 AN69 membrane that was collected in effluent in an exemplary embodiment of the invention.

[0024] FIG. 9 shows a cumulative amount of creatinine removed from plasma for an HF1400 membrane that was collected in effluent in an exemplary embodiment of the invention.

[0025] FIG. 10 is an exemplary schematic diagram of a pharmacokinetic model in accordance with the invention that was used for creatinine in treatment simulations.

[0026] FIG. 11 is an exemplary schematic diagram of an additional pharmacokinetic model in accordance with the invention used with other drugs in treatment simulations in accordance with the invention.

[0027] FIG. 12 is a flow diagram of a computer system in an exemplary embodiment of the invention.

[0028] FIG. 13 is a schematic view of a computer network in an exemplary embodiment of the invention.

[0029] FIG. 14 is a block diagram of a computer of a network of FIG. 13.

DETAILED DESCRIPTION

Simulated Treatment Configuration

[0030] FIG. 1 illustrates an exemplary system 100 and configuration for a simulated treatment on a drug or analyte of interest. The exemplary system 100 of FIG. 1 was designed and built to mimic Continuous Renal Replacement Therapy (CRRT). CRRT is a method of slower, continuous dialysis to allow solute and fluid homeostasis. CRRT includes intermittent hemodialysis and peritoneal dialysis. It is typically performed through pump-driven venovenous extracorporeal circuits and acts as renal support through blood purification to allow solute and fluid homeostasis. CRRT is often chosen for renal replacement in hemodynamically unstable patients, and it is often employed in the intensive care unit. With a slower rate of fluid removal, CRRT theoretically causes less hypotension than other dialysis techniques. Another advantage of CRRT for these patients is that they often require a large volume of fluid administration, including medications and parenteral nutrition, and CRRT can prevent an overloaded state.

[0031] With some modification, the exemplary system 100 configuration in FIG. 1 also mimics Intermittent Hemodialysis (IHD). IHD is a more traditional type of dialysis that is still widely used and is highly effective in achieving solute removal by solute clearance and fluid removal by ultrafiltration. In contrast to the slower rate of fluid removal, IHD achieves solute removal by solute clearance and fluid removal over a short period of time, typically 3-5 hours. Depending on the type of dialysis simulated, some of the components shown in FIG. 1 may not be used. For example, in some embodiments simulating some types of CRRT, if a dialysate reservoir 108 and a peristaltic pump dialysate are used, then a replacement fluid reservoir 109 and a peristaltic pump effluent 113 are not used, and vice versa.

[0032] FIG. 1 shows an exemplary embodiment where bovine plasma or blood containing the analyte (e.g., spiked with a drug) is pumped through a system 100 and passed through a filter 155 that has two chambers (blood side chamber 156 and dialysate side chamber 157) separated by a membrane 158. The blood or plasma is introduced to the blood side 156 of the filter. Dialysate fluid that has no drug (nor analyte of interest) is pumped through the other side 157 of the filter. This process creates a movement of the analyte of interest from the blood side 156 of the filter to the dialysate side 157 and ultimately excretion into an effluent container 107 (simulating drug removal). The pumps 110, 111, 112, 113 are set at typical clinical levels (e.g., 300-500 mL/min) for pump speed and are optimized based on performance criteria (and by trial and error) for the specific treatment simulation.

[0033] In some embodiments, the system 100 is run for 1 hour, and during this time samples (500 μL each) are taken at the respective sample points 103, 104, 105, 106 at time 0, 10, 20, 30, 40 and 60 minutes after starting the pumps 110, 111, 112, 113. These samples are assayed for the analyte of interest. These are the numerical treatment simulation results that are used in silico (described below) to scale and to predict human dialysis situations (i.e., create a virtual patient) and also to extend the predictions to many other dialysis settings (e.g., different filter types, flow rates, type of fluids pumped in, etc.) that are used in different clinical conditions.

Virtual Patient

[0034] In the Cybernetic Dialysis™ methods, devices, and systems in accordance with the invention, the results of the concentration of analyte (obtained from assay) in the simulated treatments are used to estimate an amount and a rate of analyte/medicine removal from plasma in the experiment, and these results are scaled to virtual patients with physiological values for blood volume and body fluid volume, and are also interpolated or extrapolated to other types of filters, flow rates, and dialysis types.

[0035] The Cybernetic Dialysis™ systems, devices, and methods in accordance with the invention create virtual patients to account for the effects of dialysis on medicines and analytes. The Cybernetic Dialysis systems, devices, and methods in accordance with the invention incorporate many methods and configurations for performing dialysis in clinical practice. These include the two primary families of dialysis by filter, namely CRRT and IHD, as outlined above. Additionally, the invention accounts for polyacrylonitrile, polymethylmethacrylate, polysulfone, polycarbonate, polyethylene, polyamide, and other filter types for each dialysis family. Systems, devices, and methods in accordance with the invention create virtual patients based on different flow rates and duration of dialysis, including type of vascular access, filter type, device used, and dose and rout of erythropoietin stimulation agents used as well as different dilution methods (e.g., pre- or post-blood). Likewise, the systems, devices, and methods in accordance with the invention can be configured to simulate different types of dialysates (or replacement) fluid, including high-molecular-weight agents such as glucose polymer-containing solutions (e.g., icodextrin), polypeptides, and dextrans, as well as low-molecular-weight agents such as glucose-containing solutions, amino acid-containing solutions, Xylitol-containing solutions, glycerol-containing solutions, and other dialysate solutions. Also, the systems, devices, and methods in accordance with the invention can be configured to simulate different types of anticoagulants (e.g., citrate, heparin, and the like) as well as different concentrations (e.g., high and low) of the analyte.

[0036] Even with this limited list of configurations, there are 2304 different permutations of the elements used in dialysis. This alone makes it very difficult to investigate all these conditions and has contributed to the dearth of systems to create virtual patients. The Cybernetic Dialysis™ systems, devices, and methods in accordance with the invention account for the extremes of a given condition, and by doing so are able to limit the number and type of elements and configurations used to create virtual patients. In silico methods are then applied to interpolate or extrapolate to additional treatment configurations between the extremes.

In-Silico Methods

[0037] The blood concentration-time course of medicine is routinely characterized during the course of drug development. The blood concentration-time course is often described and summarized by mathematical models that are published or by a variation of those models. These existing models represent rates of absorption as well as distribution and elimination of a drug in the body where elimination may be by liver or by kidney. The in-silico methods used in accordance with the Cybernetic Dialysis™ invention utilize similar base models but add a dialysis component that is turned on for a period of time to represent dialysis. The rate of drug removal by this dialysis component is informed by the rates and concentration data collected from the simulated treatment stage of the invention described above.

[0038] In one example implementation of the Cybernetic Dialysis™ invention, a mathematical model is implemented in R (software). The model algorithms are coded and then used to predict and extrapolate dialysis in humans with conditions used in the clinical practice. One example modeling algorithm to predict and extrapolate dialysis in accordance with the invention is shown in FIGS. 3A-3C.

[0039] As shown in an exemplary algorithm and schematic of the invention in FIGS. 3A-3C, the treatment simulation and in-silico modeling, simulations, and extrapolation are used to determine the effects of dialysis on a drug or analyte of interest. In blocks A and B of FIG. 3A, bovine or human plasma or blood is spiked with a drug or an exogenous analyte existing in plasma or blood. In block C, pumps, tubing, flow rates, filters, temperature, pH, dialysis or replacement solutions and apparatus are configured to mimic a type of CRRT (Continuous Renal Replacement Therapy) or IHD (Intermittent Hemodialysis). In block D, the system is run for approximately 1-2 hours. In block E, multiple samples from different ports 103, 104, 105, 106 (arterial, venous, effluent, and blood/plasma reservoir) are taken during the treatment simulation and the sampling information is recorded. In block F, the samples for the drug/analyte collected are assayed. In block G, the amount of the drug that is removed by the treatment simulation is determined, and rate and ratio parameters for the drug removal from plasma or blood are calculated.

[0040] Referring to FIG. 3B, in block H, the in-silico modeling, simulations, and extrapolation are performed to create a virtual patient that can be used to predict and determine the effects of dialysis on the drugs/analytes. Specifically, pharmacokinetic model in a human for the drug is developed by using the collected pharmacokinetic data from clinical research and development human trials. This modeled pharmacokinetic model describes the concentration-time profile of the drug in humans without dialysis. In block I, the model is adapted, and dialysis components are added. The dialysis component acts like a(n additional) kidney that can be switched on and off to mimic CRRT types or IHD duration.

[0041] In block J, the parameters from block G are scaled up and incorporated into the adapted model from block I. This simulates dialysis in humans with the full duration of dialysis (CRRTs or IHD) that is used in clinical practice. In block K, normal doses that are given to patients without dialysis and with dialysis are simulated concomitantly and the extent of drug/analyte removal is determined. If the drug/analyte removal by dialysis is not significant (i.e., clinically relevant), no dose adjustment is needed in block L, and the process continues to block R where guidance is provided for administration of the drug/analyte relative to CRRT types and IHD in clinical studies during clinical development. Blocks A-Q are reiterated and updated with new human data. As pharmacokinetic data in patients with or without dialysis become available, the model and simulations can be updated, reassessed, and if necessary, doses and recommendations can be refined. Ultimately guidance is provided for product labels and NDA submissions of the drug/analyte for administration concomitantly with CRRT dialysis types or IHD.

[0042] If, however, after block K, the drug or analyte removal is significant (i.e., clinically relevant), the process follows block M to block N where scenarios are simulated with a range of higher doses of the drug. The process then moves to block P, where a dose is identified that compensates for the drug/analyte removal by dialysis. Simultaneous to block N, in block O, different timing of the drug/analyte administration relative to dialysis start/stop times is simulated and in block Q, an administration time window relative to dialysis start/stop times is determined that avoids or minimizes the drug/analyte removal. When the activities in blocks P and Q are complete, the process continues to block R illustrated in FIG. 3C as described above.

[0043] The dialysis effect on drug removal is predicted for different configurations of dialysis and their effects, and a virtual patient is created. Then, the models are interrogated for different conditions of interest. For example, a drug researcher may be interested in the effect of a particular dialysis type or configuration on a particular drug and receive guidance on dosing the drug to patients. The models created by the systems and methods in accordance with the Cybernetic Dialysis™ invention provide such capabilities and guidance.

[0044] One example implementation of a Cybernetic Dialysis system in accordance with the invention is shown in FIG. 2. FIG. 2 shows an exemplary system 200 used to run a study that had two analytes of interest. One analyte was creatinine used as control, and the other was drug X (a drug under development). The Cybernetic Dialysis™ system 200 measures how much of drug X and how much creatinine is removed by dialysis and (referring back to FIG. 1) includes water bath 101 with shaking tray with thermal control set at 37° C. (101), and a beaker 102 containing plasma or blood with the analyte of interest (for example, spiked with known amount of medicine X. The sampling point from plasma or blood 103, a sampling point from plasma or blood prior to filter 104, a sampling point from plasma or blood post filter 105, a sampling point from effluent fluid 106, an effluent collection container 107, a dialysate reservoir container 108, a replacement fluid reservoir 109, a peristaltic pump for plasma or blood 110, a peristaltic pump for dialysate 111, a peristaltic pump for effluent 112, and a peristaltic pump for effluent 113 are also shown. Additionally, a deaeration chamber 114 is also shown as well as the filter 155 with blood side chamber 156 and dialysate side chamber 157. As outlined above, the exemplary configuration shown in FIGS. 1 and 2 can be modified depending upon the drug being evaluated and the dialysis configuration assessed.

Experimental Results and Application of Ex-Vivo Experiment Treatment Simulations

[0045] Removal of creatinine in bovine plasma is used in some example embodiments of the invention to demonstrate the systems, methods, and devices in accordance with Cybernetic Dialysis™. Creatinine represents a traceable endogenous analyte, a drug, or an exogenous analyte.

[0046] As shown in FIG. 1, methods in accordance with the Cybernetic Dialysis™ invention were applied to bovine plasma containing creatinine, and samples were taken at several points (including at a sampling point from plasma or blood 103, a sampling point from plasma or blood prior to filter 104, a sampling point from a sampling point from plasma or blood post filter 105, and from a sampling point from effluent fluid 106) up to 60 minutes after start of ex-vivo dialysis. The treatment simulation was replicated with two membrane types (M150 AN69 and HF1400) that represent a spectrum of membranes.

[0047] The M150 AN69 membrane provides continuous fluid management and renal replacement therapy. The M150 AN69 membrane is typically used for patients who have acute renal failure, fluid overload, or both. The M150 AN69 membrane is used in veno-venous therapies, including SCUF (slow continuous ultra-filtration), CVVH (Continuous Veno-Venous Hemofiltration), CVVHD (Continuous Veno-Venous Hemodialysis), and CVVHDF (continuous venovenous hemodiafiltration).

[0048] The HF1400 membrane is a polyarylethysulfone (PAES) membrane and can be used to perform all CRRT therapies (SCUF, CVVH, CVVHD and CVVHDF). The HF1400 membrane is designed with a neutral charge to enhance ultrafiltration of solutes with minimal protein adsorption.

[0049] Table 1 Table 1 shows creatinine concentrations for each of the membranes at sampling time points over one hour. For both membranes, there was appreciable removal of creatinine from plasma over time. The creatinine removal from the bovine plasma was largest at 60 min post start of dialysis.

[0050] Starting creatinine concentration at time zero (prior to starting the ex-vivo portion of a Cybernetic Dialysis™ method in accordance with the invention) was 8.366 and 8.581 μg/mL for the AN69 M150 and for the HF1400 membranes, respectively. By the end of the treatment simulation (at 60 minutes), there was a 60% and a 56% reduction in creatinine on the arterial side of the AN69 M150 and the HF1400 membrane, respectively. There was a 66% and a 63% reduction in creatinine on the venous side of the M150 AN69 and the HF1400 membrane, respectively. This represents a significant reduction in creatinine amount (concentration) in the system over time.

[0051] FIG. 4 shows arterial, venous, and plasma reservoir concentrations of creatinine for M150 AN69 membrane taken at the indicated ports that indicates a steady reduction in the concentration of creatinine over time. The solid horizontal black line at the top of the graph is the concentration of creatinine in plasma reservoir at time zero (i.e., prior to starting dialysis). FIG. 4 also shows venous concentrations of creatinine are consistently lower than the arterial concentration with a mean of −16.83 μg/mL (range −21.77 to −10.59) as shown in FIG. 5. That is, FIG. 5 shows a percentage change in creatinine concentration on a venous side vs. an arterial side of an M150 AN69 membrane (at each timepoint) in an exemplary embodiment in accordance with the invention. These results support the conclusion that the membrane and the system remove creatinine continuously at each cycle of plasma through the membrane.

[0052] FIG. 6 shows creatinine concentrations over time at the arterial and venous sides of membrane HF1400 and in the plasma reservoir in an exemplary embodiment in accordance with the invention. The plasma reservoir concentrations of creatinine for an HF1400 membrane indicate a steady reduction in concentration of creatinine over time. In FIG. 6, the solid horizontal black line is the concentration of creatinine in the plasma reservoir at time point zero (i.e., prior to starting dialysis). FIG. 6 also shows the venous concentration of creatinine is consistently lower at each timepoint (and port) than the arterial concentration with a mean of −25.37 μg/mL (range −37.64 to −16.75) See FIG. 7. These results support the conclusion that the membrane and system remove creatinine continuously at each cycle of plasma through the membrane.

[0053] FIG. 8 shows a cumulative amount of creatinine removed from plasma for an M150 AN69 membrane that was collected in effluent in an exemplary embodiment in accordance with the invention. In a treatment simulation with an M150 AN69 membrane, FIG. 8 shows the total amount of creatinine removed (that ends up in effluent) continuously increased over the period of dialysis. At the end of the dialysis period (at 60 minutes), the total amount of creatinine in the effluent was 68.0% of total amount of creatinine in the reservoir at time zero.

[0054] FIG. 9 shows simulated treatment results with an HF1400 membrane where the cumulative amount of creatinine removed from plasma (and ending up in effluent) continuously increased over the period of dialysis. The solid horizontal black line is the total amount of creatinine in the plasma reservoir at time zero (i.e., prior to starting dialysis). At the end of dialysis period (at 60 minutes), the total amount of creatinine in effluent is 66.3% of total amount of creatinine in the reservoir at time zero.

[0055] For both membrane types (M150 AN69 and HF1400), the results indicate that about 5% of the total amount of creatinine (at time zero) remains in the membrane cartridge and the associated tubing. This also contributes to creatinine removal.

TABLE-US-00001 TABLE 1 Creatinine concentration (μg/mL) measured in ex- vivo part of Cybernetic Dialysis ™ Membrane Time Point AN69 M150 HF1400 (min) Arterial Venous Arterial Venous 0 8.581 NA 8.581 NA 5 ND ND 8.675 5.41 10 7.088 5.545 8.306 6.355 20 6.53 5.216 6.174 5.052 30 5.191 4.353 5.857 4.214 40 4.145 3.706 5.261 3.782 60 3.444 2.909 3.82 3.18 Note: AN69 M150 and HF1400 are Baxter membranes; ND: note done; NA: not applicable; Time is time lapsed after start of ex-vivo dialysis

Exemplary Computer System

[0056] In some embodiments, the invention relates to a computer system for modeling, identifying, and predicting the effects of dialysis on drugs, medications, and chemical substances in patients. An exemplary computer system is shown schematically in FIG. 12 as computer system 1210. The computer system 1210 includes a generation module 1211, a simulation module 1221, a processor 1231, a random-access memory (RAM) 1241, an input/output interface 1242, and one or more databases, such as database 1251 and database 1252. Although two databases are shown in this exemplary embodiment, in other embodiments a single database, or more than two databases, can be employed. A bus 1232, which can include a set of hardware lines, allows transferring data and instructions among various components of the computer system. The input/output interface 1242 connects the computer system to an input/output devices, such as keyboard, mouse, display, printers, speakers, etc. Further, network interface 1244 allows the computer system 1210 to communicate with other network-enabled devices. RAM 1241 provides volatile storage for computer software instructions and data used to implement systems and methods in accordance with the invention.

[0057] The generation module 1211 generates in silico a plurality of virtual patients based on data collected from a population of previously treated patients, such as patients suffering from a health condition. The data collected from actual patients is inputted into the computer system 1210 via one or more input/out devices and stored in database 1251 and/or database 1252. The generation module 1211 accesses the stored data to generate virtual patients as mathematical constructs based on the actual patients' data. The generated virtual patients can also be stored in database 1251 and/or database 1252 and can be accessed by the simulation module 1221.

[0058] The simulation module 1221 is configured to apply a simulated therapy to the virtual patients to determine one or more physiological parameters of the virtual patients in response to the simulated therapy. The determined physiological parameters can be stored in database 1251 and/or database 1252. For example, virtual patients providing a mathematical model of the Peritoneal Dialysis Adequacy Test can be employed to determine how much urea is removed during dialysis.

[0059] The simulation module 1221 can be configured to adjust the simulated therapy based on the one or more physiological parameters to create a modified simulated therapy. The simulation module 1221 can apply the modified simulated therapy to the virtual patients and determine one or more physiological parameters of the virtual patients in response to the modified simulated therapy. The simulation module 1221 can iteratively repeat the process of modifying the therapy and determining one or more physiological parameters until an optimal therapy is obtained. For example, if an initial simulated dialysis treatment was to result in inadequate Kt/V, the duration of the dialysis and/or its frequency of administration can be increased and the Kt/V in response to this modified therapy can be determined, to decide whether additional modifications to the dialysis regimen is needed.

[0060] In some embodiments, the simulation modulation can be configured to recognize one or more physiological parameters determined in response to an applied simulated therapy that are indicative of an adverse effect.

[0061] FIG. 13 shows an exemplary computer network in accordance with the invention. Computer(s) and/or devices 1350 and server computer(s) 1360 provide processing, storage, and input/output devices executing application programs and similar instructions. Computer(s)/devices 1350 can also be linked through communications network 1370 to other computing devices, including other devices and/or processes 1350, digital processor dialysis machines 1350A, including machines with integrated modular drug delivery devices, and server computer(s) 1360. Communications network 1370 can be part of a remote access network, a global network (e.g., the Internet), a worldwide collection of computers, Local area or Wide area networks, and gateways that currently use respective protocols (TCP/IP, Bluetooth, etc.) to communicate with one another. Other electronic device/computer network architectures are suitable.

[0062] FIG. 14 is a diagram of the internal structure of a computer (e.g., processor/device 1350, digital processor dialysis machines 1350A, or server computers 1360) in the computer system of FIG. 13. Each computer 1350, 1350A, 1360 contains a system bus 1479, where a bus is a set of hardware lines used for data transfer among the components of a computer or processing system. Bus 1479 is essentially a shared conduit that connects different elements of a computer system (e.g., processor, disk storage, memory, input/output ports, network ports, etc.) that enables the transfer of information between the elements which could be data collected during a treatment, but also instructions to a treatment simulation system and/or apparatus to follow a drug sampling arrangement provided by a central modeling engine. Attached to system bus 1479 is I/O device interface 1482 for connecting various input and output devices (e.g., keyboard, mouse, displays, printers, speakers, etc.) to the computer 1350, 1360. Network interface 1486 allows the computer to connect to various other devices attached to a network (e.g., network 1370 of FIG. 13). Memory 1490 provides volatile storage for computer software instructions 1492 and data 1494 used to implement an embodiment of the present invention. Disk storage 1495 provides non-volatile storage for computer software instructions 1492 and data 1494 used to implement an embodiment of the present invention. Central processor unit 1484 is also attached to system bus 1479 and provides for the execution of computer instructions. In one embodiment, the processor routines 1492 and data 1494 are a computer program product (generally referenced 1492), including a computer readable medium (e.g., a removable storage medium such as one or more DVD-ROM's, CDROM's, diskettes, tapes, etc.) that provides at least a portion of the software instructions for the system. Computer program product 1492 can be installed by any suitable software installation procedure, as is well known in the art. In another embodiment, at least a portion of the software instructions can also be downloaded over a cable, communication and/or wireless connection. In other embodiments, the programs are a computer program propagated signal product embodied on a propagated signal on a propagation medium (e.g., a radio wave, an infrared wave, a laser wave, a sound wave, or an electrical wave propagated over a global network such as the Internet, or other network(s)). Such carrier medium or signals provide at least a portion of the software instructions for the present routines/program 1492.

[0063] In alternate embodiments, the propagated signal is an analog carrier wave or digital signal carried on the propagated medium. For example, the propagated signal may be a digitized signal propagated over a global network (e.g., the Internet), a telecommunications network, or other network. In one embodiment, the propagated signal is a signal that is transmitted over the propagation medium over a period of time, such as the instructions for a software application sent in packets over a network over a period of milliseconds, seconds, minutes, or longer. In another embodiment, the computer readable medium of computer program product 1492 is a propagation medium that the computer system 1350 can receive and read, such as by receiving the propagation medium and identifying a propagated signal embodied in the propagation medium, as described above for computer program propagated signal product.

Overview of Results and Application of Modeling and Simulation (Virtual Patient)

[0064] FIG. 10 a schematic diagram of a pharmacokinetic model in an exemplary embodiment in accordance with the invention that was used for creatinine in treatment simulations. In the virtual patient creation (modeling and simulation) methods in accordance with the invention, the model shown in FIG. 10 was implemented in population PK (pharmacokinetic model) using NONMEM (Non-linear Mixed Effects Modeling) to simulate pharmacokinetics of creatinine as a traceable entity (or an example drug or analyte) that is administered at the given amount 200 mg (Input), assuming a 40 L volume of distribution (V.sub.c), and a clearance (K.sub.el.V.sub.c) of 10 mL/min. Population parameters for volume and clearance were assigned a 20% CV between-subject variability. The simulations were done with and without dialysis shown as switch on/off in FIG. 10) with a dialysis rate of elimination (K.sub.df) that is scaled up from the ex-vivo simulated treatment using volume of distribution and blood flow. The simulations show for 24-hour continuous dialysis for both an M150 AN69 or an HF1400 membrane, more than 74% (range 60% to 88%) of creatinine is removed compared to the situation without dialysis where only 10% (range 7.5% to 12%) of creatinine is removed. This is an increase of more than 7-fold in drug removal rate per day. Clearly, this is a large dialysis effect on clearance. Such information on any analyte or drug can be generated during the research and development process and approval cycle. This knowledge can guide further design of appropriate clinical studies to allow inclusion of patients on dialysis early in research and development clinical trials. The inclusion/exclusion of patients can be individualized and decided very specifically based on individual characteristics and types of dialysis. This approach provides guidance on product labels and registration and for approval for drug administration in dialysis patients.

[0065] The same process described above for pharmacokinetic models and simulations can be implemented in physiologically based pharmacokinetic (PBPK) models or implemented as deterministic models where a single typical individual profile is simulated. The ex-vivo simulated treatment part of Cybernetic Dialysis™ can be scaled up to inform pharmacokinetic models and to enable predictions of dialysis effects and their magnitude in clinical scenarios. Exemplary embodiments of the invention used creatinine data as an example of an analyte; systems and methods in accordance with the invention (e.g., the same concept with ex-vivo dialysis, pharmacokinetic models, scaling up and simulated treatments (i.e., Cybernetic Dialysis™) can be applied to other drugs in research and development prior to registration and approval and to predict the effect of dialysis on such drugs.

[0066] The exemplary embodiments using creatinine above provide a straightforward example demonstrating the systems, methods, and efficacy of Cybernetic Dialysis™ in accordance with the invention. Creatinine has low adhesion and adsorption properties for the membrane and tubing. Other entities such as biological products (e.g., therapeutic proteins or peptides) have high adhesion and uptake in the membrane which results in large portions of the drug that stay unavailable (not returned) to plasma. Therefore, the results and scale up parameters from the ex-vivo simulated treatments in accordance with the invention are very important to ensure a compound (entity) specific prediction. For example, in the case of a therapeutic peptide, ex-vivo simulated treatments are likely to show that a large amount of drug remains in the membrane and/or tubing irreversibly, although effluent can contribute to small (or moderate) amounts of drug removal. Such predictions and confirmations the systems and methods in accordance with the invention provide guidance on dose adjustments or times of dialysis for such drugs in patients who receive dialysis.

[0067] There are hundreds of different permutations of types of dialysis, types of filters and physico-chemical properties of drugs. Therefore, there are a huge number of possible scenarios for new drugs undergoing research and development to be used in dialysis patients. There was no practical way to do clinical trials for such large numbers of possible scenarios because there are simply too many. The systems and methods in accordance with the invention can be used to study all scenarios. By using a limited number of simulated treatments ex-vivo, combined with modeling and simulation that represents extremes of dialysis membrane characteristics and interpolating or extrapolating other scenarios, and provide predictions for each possible scenario reliably.

[0068] The virtual patient model used for pharmacokinetics of creatinine was selected based on general data on creatinine and its physico-chemical characteristics. Other drugs may behave according to more complex distribution, absorption or pharmacokinetic principles that can be addressed and accounted for accordingly in modified pharmacokinetic models. FIG. 11 is a schematic diagram of an additional pharmacokinetic model in an exemplary embodiment of the invention that was used with other drugs in virtual patient models. For example, FIG. 11 follows a two compartmental distribution and can be used to create virtual patients for other drugs with more extensive distribution in the human body and to apply the same techniques of the invention to predict dialysis effects for creatinine.