System and method for drug-agnostic patient-specific dosing regimens

Abstract

To provide a system and a method for a drug-agnostic patient-specific dosing regimen.SOLUTION: A system and a method described herein determine a patient-specific pharmaceutical dosing regimen for a patient, using a computerized pharmaceutical dosing regimen recommendation system. The system and method herein provide a drug-agnostic model, such that the computerized pharmaceutical dosing regimen recommendation system is configured to evaluate and recommend one or more patient-specific dosing regimens that apply to not only a single drug but a class of drugs. The drug-agnostic model provides greater utility than drug-specific models by accessing a broad range of clinical data gathered for all of the drugs in the class of drugs.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims the benefit of priority under 35 U.S.C. Section 119(e) of U.S. Utility Application No. 16 / 391,950, filed April 23, 2019, and International Application No. PCT / US2019 / 028750 and U.S. Provisional Application No. 62 / 815,825, filed March 8, 2019, the contents of which are incorporated herein by reference in their entireties.

[0002] (Technical field) The present disclosure relates generally to patient-specific medication and treatment recommendations, including, but not limited to, computerized systems and methods that use drug-specific mathematical models specific to a class of drugs and observed patient-specific responses to treatment to predict, suggest, modify and evaluate suitable drug treatment plans for a particular patient. [Background technology]

[0003] (background) The physician's decision to initiate a drug-based treatment regimen for a patient involves the determination of the dosing regimen for the drug to be prescribed. Different dosing regimens are appropriate for different patients with different patient factors. As an example, the dosage amount, dosing interval, duration of treatment, and other variables may vary across different dosing regimens. For example, a patient with a low neutrophil count may require delayed dosing, a smaller dose than a typical patient, or both. Traditionally, the physician prescribes an initial dosing regimen based on the drug's package insert (PI) and the physician's own personal clinical experience. After the initial cycle of treatment, the physician may follow up with the patient, reassess the patient, and reconsider the initial dosing regimen. The PI sometimes provides quantitative indications to increase or decrease the dosage, or increase or decrease the dosing interval. Based on both the physician's assessment of the patient, his clinical experience, and the PI information, the physician will adjust the dosing regimen on an ad hoc basis. Adjusting a medication regimen for a patient is, in large part, a trial and error process informed by the physician's experience and clinical judgment.

[0004] An appropriate dosing regimen can be highly beneficial and therapeutic, while an inappropriate dosing regimen can be ineffective or even harmful to the patient's health. Moreover, both under-dosing and over-dosing generally result in the loss of time, money, and / or other resources, and increase the risk of undesirable outcomes. A computerized dosing recommendation system can assist medical personnel in providing and assessing dosing regimens. Previously disclosed examples of systems and methods are described in U.S. Patent No. 10,083,400, filed October 7, 2013, and issued September 25, 2018, and U.S. Application No. 15 / 094,379, filed April 8, 2016, and published October 13, 2016 as U.S. Patent Application Publication No. 2016 / 0300037 A1, each of which is incorporated herein by reference in its entirety.

[0005] Some known population pharmacokinetic models are developed to describe a particular drug and are specific not only to the drug but also to the specific patient population that will take the drug. Such models are frequently specific to a particular route of administration as well. These models are intended to be as specific as possible with respect to both the drug and the route of administration, and the patients that are expected to take the drug during the course of treatment for a specific disease, so that the models can be used to simulate other dosing regimens, compare the pharmacokinetics of the drug after formulations are changed, and various other applications. This approach of developing a model for a single drug limits its application to that drug and the specific patient population for which data was collected to create the model. Limiting the development and use of the model to a single drug narrows the amount of information and data available. Summary of the Invention [Means for solving the problem]

[0006] (summary) Thus, systems and methods are described herein for determining a patient-specific drug dosing regimen for a patient using a computerized drug dosing regimen recommendation system that is applied to a class or other group of drugs. The systems and methods described herein provide a population model that can be used with respect to a class or set of drugs. In particular, the systems and methods herein can be configured to develop a patient-specific dosing regimen based on data indicative of pharmacokinetic ("PK") and pharmacodynamic ("PD") parameters of multiple drugs. A model for a group of drugs would be able to predict patient response and optimize future dosing, for example, based on historical measurements of physiological parameters from different drugs and / or administration routes, since the model is drug-independent and utilizes patient-specific data based on similarities between drugs. It may also be desirable for the model to be applicable to a set of drugs and drug-independent, so that information about the administered drug can be learned by comparing measured data with model-based simulations of other drugs in the group, which can then be examined against historical results. A model that is applicable to a certain group of drugs may be tailored to a physician who may wish to treat a patient with multiple drugs or vary certain therapeutic factors, such as route of administration.

[0007] A dosing regimen (also referred to as a treatment plan) may include a schedule for dosing, one or more dosage amounts, and / or one or more routes of administration. A dosing regimen is not limited to only one drug, but can include multiple drugs, using the same or different routes of administration. A drug (also referred to as a medicine, drug, agent, biologic, compound, treatment, therapy, or any other similar term) is a substance that exerts a physiological effect when introduced into the body. In some implementations, the system described herein is not specific to a particular drug, but instead applies to a class, or subset, or group of drugs used in a drug-independent model. As used herein, the term "drug" may refer to a single drug or a class or set of drugs.

[0008] A drug class may be a group of more than one drug that exhibits at least one similar pharmacokinetic (PK) and / or pharmacodynamic (PD) behavior, shares a common mechanism of action, or a combination thereof. As examples, a set of drugs may treat the same disease or be used for the same indication, examples of which include systemic inflammatory disease, inflammatory bowel disease (IBD), ulcerative colitis, Crohn's disease, rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, or multiple sclerosis. A set of drugs may have similar chemical structures. For example, a set of drugs may include monoclonal antibodies (mAbs), anti-inflammatory compounds, corticosteroids, immunomodulators, antibiotics, or biologic therapies. A user, such as a physician, clinician, or a user creating a drug-independent model, may define a drug class based on specific criteria, and members of the class may be electronically designated in the database as being part of the class. The database is accessible to the systems and methods disclosed herein for use in determining a class-based dosing regimen that may be used for any drug in the class. In some implementations, the dosing regimen output from the model is not specific to a single drug, but is general to a class of drugs and suitable for any drug in that class. For example, the dosing regimen may include a drug-independent unit measurement (e.g., one unit, two units, three units, etc., where the unit corresponds to a prescribed amount of active agent) and a time or times for administration.

[0009] The development and application of drug-independent models allows for different utility than single drug-specific models. Rather than implementing multiple models, each model corresponding to a single drug in a class, drug-independent models are applicable to all drugs in the class and therefore may be applicable to a wider range of patients than single drug models across a range of administration routes. For example, the models described herein may include a pharmacokinetic drug-independent model that can be used for all biologics used in the treatment of inflammatory diseases. Such models can be used to suggest dosing regimens for fully human monoclonal antibodies (mAbs), chimeric mAbs, fusion proteins, and mAb fragments (i.e., a range of drugs with different pharmacokinetic properties but with other similarities such as molecular weight and indications) using the same model. The models can be used in a wide range of patient populations, including inflammatory bowel disease, rheumatoid arthritis, psoriatic arthritis, psoriasis, multiple sclerosis, and other such diseases resulting from immune dysregulation. The development and application of drug-independent Bayesian models for agents in other broad drug sets (eg, aminoglycoside antibiotics, chemotherapy drugs that cause low white blood cell counts, etc.) is similarly feasible.

[0010] Drugs within a class may be administered through various routes, such as subcutaneous, intravenous, or oral. A drug-agnostic model may take into account the route of administration by accepting the route of administration as a variable input to the system, allowing for greater flexibility for the model. Because the drug-agnostic model applies to a set of drugs, rather than only a single drug, the model may retain patient-specific information when a patient is treated with multiple drugs in the set of drugs. For example, the set of drugs may include infliximab, vedolizumab, adalimumab, and other anti-inflammatory biologics. If a patient is treated with one drug (e.g., infliximab) and then later treated with another drug (e.g., vedolizumab), the model may retain all patient-specific data (drug concentration measurements, clearance rates, weight measurements, etc.) from the patient's infliximab-based treatment when determining the appropriate dosing regimen once the patient is treated with the new drug. Retaining patient-specific data allows the drug-independent model to accurately predict a patient's ability to process drugs, thereby providing more suitable patient-specific dosing regimens when patients change drug therapy.

[0011] One aspect of the present invention relates to a method for generating a patient-specific drug dosing regimen. As discussed in detail below, the method may be implemented on a system, which may be a computer system, including a single computer or multiple computers communicating via any network, such as in a distributed architecture. At least one processor may be stored in one, some, or all of the computers in the computer system, and may communicate with at least one electronic database stored on the same computer or on a different computer in the computer system. The system may include a cloud-based computing system operated by the same, related, or unrelated entity.

[0012] The method includes receiving an input into a processor of the system. The input may include drug data indicating a plurality of drugs and associated routes of administration, where drugs in the plurality of drugs are expected to exhibit similar pharmacokinetic (PK) behavior, similar pharmacodynamic (PD) behavior, or both. For example, the drugs may have similar chemical structures or mechanisms of action leading to similar pharmacokinetic (PK) behavior, similar pharmacodynamic (PD) behavior, or both. The input may include concentration (for a PK model) or response (for a PD model) data indicating a concentration or response level of a specific drug of the plurality of drugs in a sample obtained from the patient. The input may include a target drug exposure or response level for the patient. The target may be determined by a physician based on the drug data and / or the concentration or response. In some implementations, the target may be automatically determined by the system to produce a therapeutic response in the patient. The system may evaluate the plurality of targets input to determine one or more targets that produce a therapeutic response in the patient.

[0013] The method includes selecting a mathematical model from a database stored in a memory of the system. The database may be accessible by a processor and may store a plurality of mathematical models. The selected mathematical model may represent responses by a plurality of patients to one or more drugs among the plurality of drugs. Each response of the responses indicates a patient response to at least one drug among the plurality of drugs. In some cases, the mathematical model is not specific to a particular drug. In other cases, the model may be initially specific to a particular drug and then adapted over time using the methods described herein and applied to a class of similar drugs. In other cases, the user may find that the drug-specific model can be used to predict concentrations or responses for other drugs and that the drug-specific model can be used for drug-independent predictions.

[0014] The method includes estimating a plurality of predicted concentration-time profiles using the selected mathematical model and based on the input. Each predicted concentration-time profile may be indicative of a patient's response to any drug among the plurality of drugs via a particular route of administration. Each predicted concentration-time profile of the plurality of predicted concentration-time profiles may correspond to a dosing regimen among a plurality of dosing regimens output by the system. Each dosing regimen may include at least one dosage and / or a recommended schedule for administering the at least one dosage to the patient. The method may include selecting a first dosing regimen for the plurality of drugs from the plurality of dosing regimens to achieve a therapeutic objective based on a target drug exposure or response level.

[0015] The method may include outputting a first dosing regimen for the patient for the multiple drugs. At least one advantage of the method is that the model may have access to a wider range of clinical data compared to a drug-specific model because the model may be refined to use measured data, for example, from administration of one drug in the set of drugs to provide a better recommendation for dosing any of the drugs in the set of drugs.

[0016] In some implementations, the mathematical model is selected based on one or more criteria, such as the inputs, the error associated with the model, or the commonality between each model and multiple drugs. In some implementations, the mathematical model is selected based on the drug data. If the model is selected based on the drug data, the selection step may involve comparing the parameters or covariates of the model with the drug data. This step may involve maximizing the number of similar parameters between the model and the set or class of drugs. Each model in the database may be associated with an error or confidence interval that indicates the model's performance so far, and the model may be selected based on the error or confidence interval. The model may be selected based on the amount of data or information available for or applicable to the model. It would be advantageous for the selected model to use the widest range of information or data available, so that the model can be refined to make more accurate predictions and provide recommendations that are more likely to achieve targets. For example, the concentration or response data and / or the type of concentration or response data may be analyzed to select a model that is capable of incorporating that type of data into the model, or that is capable of incorporating at least some or the maximum of the data. In some implementations, the selection is performed by an optimization function. Alternatively or in addition to the selection methods discussed above, Bayesian methods may be used to select a model. In some implementations, multiple models may be compared to each other and the model that best represents the patient or input may be selected.

[0017] In some implementations, the drug data excludes information identifying a specific drug belonging to the plurality of drugs. The drug data may include one or more available dosage units corresponding to one or more drugs in the plurality of drugs. In such cases, the dosage may be configured to be a multiple of the available dosage units. In some implementations, the method further includes receiving additional drug data indicating an updated route of administration for the specific drug or one or more drugs in the plurality of drugs. The system may update the mathematical model based on the specific drug or one or more drugs in the plurality of drugs. The updated mathematical model may then be used to calculate at least one updated dosing regimen to reach a therapeutic objective for the patient. The at least one updated dosing regimen may be output for the patient.

[0018] In some implementations, the model is a pharmacokinetic model. The model may alternatively be a pharmacodynamic model. The model may account for both pharmacokinetics and pharmacodynamics. The pharmacokinetic or pharmacodynamic components of the model may represent the concentration-time profiles of multiple drugs. The pharmacokinetic components of the model may be based on clearance parameters, which represent the influx and efflux of drugs into the patient's body. The pharmacodynamic components of the model may be based on synthesis and degradation rates of pharmacodynamic markers, which represent the patient's individual response to multiple drugs. In some implementations where the model includes a pharmacokinetic component and a pharmacodynamic component, the components are interrelated. For example, the clearance of the pharmacokinetic component may be a function of the pharmacodynamic response, and / or vice versa. The model may employ Bayesian methods, such as Bayesian inference, to predict patient response or concentration-time profiles for one or more dosing regimens.

[0019] In some implementations, the method further includes receiving additional patient data indicative of a second response of the patient to administration of the specific drug according to the first dosing regimen or a modified version of the first dosing regimen. The additional patient data may include additional concentration data indicative of one or more concentration levels of the specific drug in one or more samples obtained from the patient. The model can then be updated based on the second response of the patient to administration of the specific drug according to the first dosing regimen or a modified version of the first dosing regimen. At least one updated dosing regimen can be calculated to reach a therapeutic objective for the patient using the updated mathematical model. The at least one updated dosing regimen can be output for the patient.

[0020] The route of administration may be at least one of subcutaneous, intravenous, oral, intramuscular, intrathecal, sublingual, buccal, rectal, vaginal, intraocular, nasal, inhalation, aerosol, dermal, or transdermal. The multiple drugs may be one of monoclonal antibodies and / or antibody constructs, cytokines, drugs used for enzyme replacement therapy, aminoglycoside antibiotics, and chemotherapy drugs that cause leukopenia. In some implementations, each drug in the multiple drugs shares a similar chemical structure. Each drug in the multiple drugs may share a similar mechanism of action. In some implementations, the multiple drugs are used to treat inflammatory diseases, such as inflammatory bowel disease (IBD), rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, and multiple sclerosis. The specific drug may be infliximab or adalimumab. The drug data may include a dosage strength for a particular drug and / or an indicator of whether the particular drug is fully human, chimeric, or fragmented. The system may receive as input input patient data indicating a patient disease to treat.

[0021] In some implementations, the received input includes physiological data indicative of one or more measurements of at least one physiological parameter of the patient. The at least one physiological parameter of the patient may include at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapy, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure reading, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharpe score, and demographic information. The mathematical model may be selected from a set of mathematical models to best fit the received physiological data.

[0022] In some implementations, the method includes generating for display a patient-specific concentration-time profile indicative of the patient response to the multiple drugs in response to the first dosing regimen and an indication of at least a portion of the concentration data and the additional concentration data. An indication of a target drug exposure or response level may be generated for display. In some implementations, the system receives historical data indicative of the patient's response to a second drug other than the specific drug. The computational model may take into account the historical data to generate a prediction of the concentration-time profile of the multiple drugs in the patient. At least one advantage provided herein relates to a model that can be applied to a wider range of data than is typical, such as data collected from one or more patients administered one or more drugs, to generate a prediction regarding the patient response to a class of drugs, including drugs similar to the previously administered one or more drugs.

[0023] Another aspect relates to a method for generating a patient-specific drug dosing regimen using a computerized drug dosing regimen recommendation system without having previously measured concentration data. The method may be implemented on any of the systems described herein. The method involves receiving an input into a processor of the system. The input may be drug data indicating a plurality of drugs and routes of administration. The drugs in the plurality of drugs may be expected to exhibit similar pharmacokinetic behavior effects, similar pharmacodynamic effects, or both. The input may include initial drug concentration input data representing an initial comparative concentration data point. A target drug exposure or response level is included in the input. The method further includes selecting a mathematical model representing a response by a plurality of patients to one or more drugs in the plurality of drugs. Each response may represent a patient response to at least one drug in the plurality of drugs. The mathematical model may be configured to be non-specific to a particular drug.

[0024] The method includes using the selected model and inputs to predict a plurality of predicted concentration-time profiles indicative of a patient's response to any of the plurality of drugs via a route of administration. Each predicted concentration-time profile may correspond to a dosing regimen of the plurality of dosing regimens. Each dosing regimen may include at least one dosage and / or a recommended schedule for administering at least one dosage to the patient. The method includes selecting a first dosing regimen for the predicted plurality of drugs to achieve a therapeutic objective based on the target drug exposure or response level. The first dosing regimen may be output for the patient. At least one advantage of this aspect results from the model providing predicted concentration-time profiles and dosing regimens configured to meet a therapeutic objective using only initial comparative concentration data points, without having measured concentration data from a specific patient.

[0025] In some implementations, the initial comparative concentration data point indicates a concentration level of a specific drug of the multiple drugs in a sample obtained from the patient. Alternatively, the initial comparative concentration data point is calculated based on physiological parameters of the patient such that direct concentration measurements from the patient are not required. The patient's physiological parameters may include at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapy, previous laboratory test information, concomitantly administered drugs, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure reading, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharp score, and demographic information. The drug data may include available dosage strengths for the specific drug and / or an indicator of whether the specific drug is fully human, humanized, chimeric, or fragmented.

[0026] Another aspect relates to a system comprising a controller configured to perform the method of the aforementioned aspect. The system may comprise an interface connection to an electronic medical record (EMR) database. The interface may be operable by software or firmware configured to retrieve one or more indicators of patient information for the patient via the interface connection. An appropriate medication for administration to the patient may be determined using the retrieved patient information. This aspect provides at least one advantage in that a physician may enter a patient name and the EMR for the patient will automatically feed drug and / or physiological data into the system to perform an initial guess. This aspect may be particularly useful when recommending a dosing regimen for a relatively new drug, since the new drug may have structural similarities and similar identified covariates with other drugs that allow a drug-independent model to be used simply by adjusting for the particularities of the new drug, such as dosage strength, to generate a dosing recommendation. For example, a new drug may only have clinical data from FDA review. This clinical trial data can be fed into the drug-independent model system and normalized to other drugs in a similar drug class or group. Without much knowledge of a new drug, especially knowledge that would be used to dose it on an ad hoc basis, a physician may use the systems and methods proposed herein to find appropriate dosages for the new drug based on historical data of similar drugs.

[0027] Another aspect relates to a pharmaceutical preparation comprising a first dosage of a specific drug. The first dosage may be determined by a method described herein. For example, the first dosage of a drug in the pharmaceutical preparation is determined by the system to meet a therapeutic objective based on an input target exposure or response level of the drug after having estimated concentration-time profiles for a plurality of drugs including the specific drug. The specific drug and each drug of the plurality of drugs may have similar pharmacokinetic, pharmacodynamic, and / or structural properties. The pharmaceutical preparation may include a revised first dosage determined by the system using an updated mathematical model, the updated mathematical model taking into account concentration or response data indicating a response by the patient to a dosage of one or more drugs in the plurality of drugs. The present invention provides, for example, the following: (Item 1) 1. A method for generating a patient-specific drug dosing regimen using a computerized drug dosing regimen recommendation system comprising a processor and a memory, the method comprising: (a) receiving an input into a processor of the system, the input comprising: (i) drug data indicating a plurality of drugs and associated routes of administration, the drugs in the plurality of drugs being expected to exhibit similar pharmacokinetic (PK) behavior, similar pharmacodynamic (PD) behavior, or both; (ii) concentration or response data indicating concentration or response levels of specific drugs of the plurality of drugs in a sample obtained from the patient; and (iii) a target drug exposure or response level for said patient; and (b) selecting a mathematical model from a database stored in the memory, the database being accessible by the processor and storing a plurality of mathematical models, the selected mathematical model representing responses by a plurality of patients to a plurality of drugs among the plurality of drugs, each response of the responses being indicative of a patient response to at least one drug among the plurality of drugs, the mathematical model being non-specific to a particular drug; (c) using the selected mathematical model and based on the concentration data and the route of administration, estimating a plurality of predicted concentration-time profiles indicative of the patient's response to any drug among the plurality of drugs via the route of administration, each predicted concentration-time profile of the plurality of predicted concentration-time profiles corresponding to a dosing regimen among a plurality of dosing regimens, each dosing regimen of the plurality of dosing regimens including (i) at least one dosage and (ii) a recommended schedule for administering the at least one dosage to the patient; (d) selecting, from the plurality of dosing regimens, a first dosing regimen for the plurality of drugs predicted to achieve a therapeutic objective based on the target drug exposure or response level; (e) outputting the first dosing regimen for the plurality of drugs for the patient; and A method comprising: (Item 2) 2. The method of claim 1, wherein the drug data excludes information identifying the specific drug among the plurality of drugs. (Item 3) The method further comprises: receiving additional drug data indicating an updated route of administration for said particular drug; updating the mathematical model based on the updated route of administration for the specific drug; calculating at least one updated dosing regimen for reaching the treatment objective for the patient based on the updated mathematical model; outputting the at least one updated medication regimen for the patient; and The method according to any of the preceding items, comprising: (Item 4) 20. The method of claim 19, wherein the drug data further comprises one or more available dosage units corresponding to one or more drugs in the plurality of drugs, and the dosage amount is a multiple of the available dosage units. (Item 5) The method of any of the preceding items, wherein the model is a pharmacokinetic model. (Item 6) The method of any of the preceding items, wherein the model is a pharmacodynamic model. (Item 7) The method further comprises: receiving additional patient data indicative of a second response of the patient to administration of a specific drug of the plurality of drugs or another drug in accordance with the first dosing regimen or a modified version of the first dosing regimen, the additional patient data including additional concentration data indicative of one or more concentration levels of the specific drug of the plurality of drugs or another drug in one or more samples obtained from the patient; updating the mathematical model based on a second response of the patient to administration of a specific drug of the plurality of drugs or another drug according to the first dosing regimen or a modified version of the first dosing regimen; calculating at least one updated dosing regimen for reaching the treatment objective for the patient based on the updated mathematical model; outputting the at least one updated medication regimen for the patient; and The method according to any of the preceding items, comprising: (Item 8) The method of any of the preceding items, wherein the route of administration is at least one of subcutaneous, intravenous, oral, intramuscular, intrathecal, sublingual, buccal, rectal, vaginal, intraocular, nasal, inhalation, aerosol, dermal, and transdermal. (Item 9) The method of any of the preceding items, wherein the multiple drugs are one of monoclonal antibodies and antibody constructs, cytokines, drugs used for enzyme replacement therapy, aminoglycoside antibiotics, and chemotherapy drugs that cause leukopenia. (Item 10) 20. The method of any preceding item, wherein each drug in the plurality of drugs shares a similar chemical structure. (Item 11) 20. The method of any of the preceding items, wherein each drug in the plurality of drugs shares a similar mechanism of action. (Item 12) The method of any of the preceding items, wherein the multiple drugs are used to treat an inflammatory disease. (Item 13) 13. The method of claim 12, wherein the multiple drugs are used to treat at least one of inflammatory bowel disease (IBD), rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, and multiple sclerosis. (Item 14) The method of any of the preceding items, wherein the specific drug is infliximab. (Item 15) The method of any of items 1-13, wherein the specific drug is adalimumab. (Item 16) 20. The method of claim 19, wherein the received input includes physiological data indicative of one or more measurements of at least one physiological parameter of the patient. (Item 17) 17. The method of claim 16, wherein the at least one physiological parameter of the patient comprises at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapies, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure reading, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharp score, and demographic information. (Item 18) 20. The method of claim 17, further comprising selecting the mathematical model from a set of mathematical models that best fits the received physiological data. (Item 19) 2. The method of any preceding claim, wherein the model is a Bayesian model. (Item 20) The method of any of the preceding items, further comprising generating, for display, (i) a patient-specific predicted concentration time profile indicative of the patient response to the plurality of drugs in response to the first dosing regimen, and (ii) an indication of at least a portion of the concentration data and the additional concentration data. (Item 21) 20. The method of any preceding claim, further comprising generating an indication of said target drug exposure or response level for display. (Item 22) 20. The method of any of the preceding items, wherein the model includes pharmacokinetic or pharmacodynamic components that represent concentration or response time profiles of the multiple drugs. (Item 23) 20. The method of any of the preceding items, wherein the model includes a pharmacodynamic component based on synthesis and degradation rates of pharmacodynamic markers indicative of the patient's individual response to the multiple drugs. (Item 24) 20. The method of claim 1, further comprising receiving historical data indicative of the patient's response to a second drug other than the specific drug, wherein the computational model takes into account the historical data to generate a prediction of the concentration-time profile of the multiple drugs in the patient. (Item 25) 20. The method of claim 19, wherein the medication data includes at least one of one or more available dosage strengths corresponding to one or more medications in the plurality of medications, and one or more indicators of whether one or more medications in the plurality of medications are fully human. (Item 26) 13. The method of any preceding claim, wherein the received input includes patient data indicative of a patient disease for treatment. (Item 27) 1. A method for generating a patient-specific drug dosing regimen using a computerized drug dosing regimen recommendation system, the method comprising: (a) receiving an input into a processor of the system, the input comprising: (i) drug data indicating a plurality of drugs and routes of administration, the drugs in the plurality of drugs being expected to exhibit similar pharmacokinetic (PK) effects, similar pharmacodynamic (PD) effects, or both; (ii) initial drug concentration or response input data representing initial comparative concentration or response data points; (iii) a target drug exposure or response level for said patient; and (b) selecting a mathematical model representing responses by a plurality of patients to a plurality of drugs among the plurality of drugs, each response of the responses being indicative of a patient response to at least one drug among the plurality of drugs, the mathematical model being non-specific to a particular drug; (c) using the selected mathematical model and based on the patient data and the route of administration, estimating a plurality of predicted concentration-time profiles indicative of the patient's response to any drug among the plurality of drugs via the route of administration, each predicted concentration-time profile of the plurality of predicted concentration-time profiles corresponding to a dosing regimen among a plurality of dosing regimens, each dosing regimen of the plurality of dosing regimens including (i) at least one dosage and (ii) a recommended schedule for administering the at least one dosage to the patient; (d) selecting, from the plurality of dosing regimens, a first dosing regimen for the plurality of drugs predicted to achieve a therapeutic objective based on the target drug exposure or response level; (e) outputting the first dosing regimen for the plurality of drugs for the patient; and A method comprising: (Item 28) 28. The method of claim 27, wherein the initial comparative concentration data point is indicative of a concentration level of a particular drug of the plurality of drugs in a sample obtained from the patient. (Item 29) 28. The method of claim 27, wherein the drug data includes at least one of one or more available dosage strengths corresponding to one or more drugs in the plurality of drugs, and one or more indicators of whether one or more drugs in the plurality of drugs are fully human. (Item 30) 28. The method of claim 27, wherein the initial comparative concentration data point is calculated based on physiological parameters of the patient. (Item 31) 31. The method of claim 30, wherein the patient's physiological parameters include at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapies, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure readings, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharp score, and demographic information. (Item 32) A system comprising a controller configured to implement the method according to any of items 1-31. (Item 33) A system according to any of the systems described in the preceding items, or a method according to any of the preceding method items, comprising an interface connection to an electronic medical record database, the interface connection being operable by software or firmware configured to read one or more indicators of patient information for a patient via the interface connection and use that information to determine an appropriate medication to administer to the patient. (Item 34) 1. A system for generating a patient specific drug dosing regimen, the system comprising: Memory, At least one processor, the at least one processor comprising: (a) receiving an input into a processor of the system, the input comprising: (i) drug data indicating a plurality of drugs and associated routes of administration, the drugs in the plurality of drugs being expected to exhibit similar pharmacokinetic (PK) behavior, similar pharmacodynamic (PD) behavior, or both; (ii) concentration or response data indicating concentration or response levels of specific drugs of the plurality of drugs in a sample obtained from the patient; and (iii) a target drug exposure or response level for said patient; and (b) selecting a mathematical model from a database stored in the memory, the database being accessible by the one or more processors and storing a plurality of mathematical models, the selected mathematical model representing responses by a plurality of patients to a plurality of drugs among the plurality of drugs, each response being indicative of a patient response to at least one drug among the plurality of drugs, the mathematical model being non-specific to a particular drug; (c) using the selected mathematical model and based on the concentration data and the route of administration, estimating a plurality of predicted concentration-time profiles indicative of the patient's response to any drug among the plurality of drugs via the route of administration, each predicted concentration-time profile of the plurality of predicted concentration-time profiles corresponding to a dosing regimen among a plurality of dosing regimens, each dosing regimen of the plurality of dosing regimens including (i) at least one dosage and (ii) a recommended schedule for administering the at least one dosage to the patient; (d) selecting, from the plurality of dosing regimens, a first dosing regimen for the plurality of drugs predicted to achieve a therapeutic objective based on the target drug exposure or response level; (e) outputting the first dosing regimen for the plurality of drugs for the patient; and at least one processor configured to A system comprising: (Item 35) 35. The system of claim 34, wherein the drug data excludes information identifying the specific drug among the plurality of drugs. (Item 36) The at least one processor further comprises: receiving additional drug data indicating an updated route of administration for said particular drug; updating the mathematical model based on the updated route of administration for the specific drug; calculating at least one updated dosing regimen for reaching the treatment objective for the patient based on the updated mathematical model; outputting the at least one updated medication regimen for the patient; and 36. The system of any of items 34 and 35, configured to perform the following: (Item 37) 37. The system of any of claims 34-36, wherein the drug data further includes one or more available dosage units corresponding to one or more drugs among the plurality of drugs, and the dosage is a multiple of the available dosage units. (Item 38) 38. The system according to any of items 34-37, wherein the model is a pharmacokinetic model. (Item 39) 39. The system according to any of items 34-38, wherein the model is a pharmacodynamic model. (Item 40) The at least one processor further comprises: receiving additional patient data indicative of a second response of the patient to administration of a specific drug of the plurality of drugs or another drug in accordance with the first dosing regimen or a modified version of the first dosing regimen, the additional patient data including additional concentration data indicative of one or more concentration levels of the specific drug of the plurality of drugs or another drug in one or more samples obtained from the patient; updating the mathematical model based on a second response of the patient to administration of a specific drug of the plurality of drugs or another drug according to the first dosing regimen or a modified version of the first dosing regimen; calculating at least one updated dosing regimen for reaching the treatment objective for the patient based on the updated mathematical model; outputting the at least one updated medication regimen for the patient; and 40. The system according to any one of items 34-39, configured to: (Item 41) 41. The system according to any of items 34-40, wherein the route of administration is at least one of subcutaneous, intravenous, oral, intramuscular, intrathecal, sublingual, buccal, rectal, vaginal, intraocular, nasal, inhalation, aerosol, dermal, and transdermal. (Item 42) 42. The system of any of items 34-41, wherein the plurality of drugs is one of monoclonal antibodies and antibody constructs, cytokines, drugs used for enzyme replacement therapy, aminoglycoside antibiotics, and chemotherapy drugs that cause leukopenia. (Item 43) 43. The system of any of items 34-42, wherein each drug in the plurality of drugs shares a similar chemical structure. (Item 44) 44. The system of any of items 34-43, wherein each drug in the plurality of drugs shares a similar mechanism of action. (Item 45) 45. The system of any of items 34-44, wherein the plurality of drugs is used to treat an inflammatory disease. (Item 46) 46. ​​The system of claim 45, wherein the plurality of drugs are used to treat at least one of inflammatory bowel disease (IBD), rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, and multiple sclerosis. (Item 47) 47. The system according to any of items 34-46, wherein the specific drug is infliximab. (Item 48) 47. The system according to any of items 34-46, wherein the specific drug is adalimumab. (Item 49) 49. The system of any of items 34-48, wherein the received input includes physiological data indicative of one or more measurements of at least one physiological parameter of the patient. (Item 50) 50. The system of claim 49, wherein the at least one physiological parameter of the patient includes at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, gender, race, disease stage, disease status, previous therapies, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure reading, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharp score, and demographic information. (Item 51) Item 51. The system of item 50, wherein the method further comprises selecting, from a set of mathematical models, the mathematical model that best fits the received physiological data. (Item 52) 52. The system of any of items 34-51, wherein the model is a Bayesian model. (Item 53) 53. The system of any of claims 34-52, wherein the at least one processor is further configured to generate, for display, (i) a patient-specific predicted concentration-time profile indicative of the patient response to the plurality of drugs in response to the first dosing regimen, and (ii) an indication of at least a portion of the concentration data and the additional concentration data. (Item 54) 54. The system of any of items 34-53, wherein at least one processor is further configured to generate an indication of the target drug exposure or response level for display. (Item 55) 55. The system of any of items 34-54, wherein the model includes a pharmacokinetic or pharmacodynamic component that describes the concentration-time profiles of the multiple drugs. (Item 56) 56. The system of any of items 34-55, wherein the model includes a pharmacodynamic component based on synthesis and degradation rates of pharmacodynamic markers indicative of the patient's individual response to the multiple drugs. (Item 57) 57. The system of any of items 34-56, wherein the at least one processor is further configured to receive historical data indicative of the patient's response to a second drug other than the specific drug, and the computational model takes into account the historical data to generate a prediction of concentration-time profiles of the multiple drugs in the patient. (Item 58) 58. The system of any of items 34-57, wherein the drug data includes at least one of one or more available dosage strengths corresponding to one or more drugs in the plurality of drugs, and one or more indicators indicative of whether one or more drugs in the plurality of drugs are fully human. (Item 59) The system of any of items 34-58, wherein the received input includes patient data indicative of a patient disease to treat. (Item 60) 60. The system of any of items 34-59, further comprising a cloud-based computing system operated by the at least one processor. (Item 61) Item 61. The system of item 60, wherein the cloud-based computing system comprises a network and at least one server, the network configured to couple at least two selected from the group of the at least one processor, the memory, the at least one server, and at least one user device. (Item 62) 62. The system of any of items 34-61, further comprising an interface connection to an electronic medical record database, the interface connection operable by software or firmware configured to read one or more indicators of patient information for the patient via the interface connection. (Item 63) A pharmaceutical drug for administration to a subject patient in a patient-specific dosing regimen determined by a computerized drug dosing regimen recommendation system comprising a processor and a memory, the pharmaceutical drug comprising: Active ingredient in the first dosage having an amount per unit Including, The first dosage amount is determined by a model analysis, the model analysis comprising: (a) receiving an input into a processor of the system, the input comprising: (i) drug data indicating a plurality of drugs and associated routes of administration, the drugs in the plurality of drugs being expected to exhibit similar pharmacokinetic (PK) behavior, similar pharmacodynamic (PD) behavior, or both; (ii) concentration or response data indicating a first response corresponding to an initial dosage of a particular drug of the plurality of drugs in a sample obtained from the patient; (iii) a target drug exposure or response level for said patient; and (b) selecting a mathematical model from a database stored in the memory, the database being accessible by the processor and storing a plurality of mathematical models, the selected mathematical model representing responses by a plurality of patients to a plurality of drugs among the plurality of drugs, each response being indicative of a patient response to at least one drug among the plurality of drugs, the mathematical model being non-specific to a particular drug; (c) using the selected mathematical model and based on the concentration data and the route of administration, estimating a plurality of predicted concentration-time profiles indicative of the patient's response to any drug among the plurality of drugs via the route of administration, each predicted concentration-time profile of the plurality of predicted concentration-time profiles corresponding to a dosing regimen among a plurality of dosing regimens, each dosing regimen of the plurality of dosing regimens including (i) at least one dosage and (ii) a recommended schedule for administering the at least one dosage to the patient; (d) selecting, from the plurality of dosing regimens, a first dosing regimen for the plurality of drugs predicted to achieve a therapeutic objective based on the target drug exposure or response level, the first dosing regimen comprising the first dosage amount; Pharmaceutical preparations, including (Item 64) 64. The pharmaceutical of claim 63, wherein prior to the step of estimating, the selected model is updated based on the concentration or response data. (Item 65) The pharmaceutical comprises the active ingredient in a second dosage amount, and the model analysis further comprises: receiving additional patient data indicative of a second response of the patient to administration of a specific drug of the plurality of drugs or another drug in accordance with the first dosing regimen or a modified version of the first dosing regimen, the additional patient data including additional concentration data indicative of one or more concentration levels of the specific drug of the plurality of drugs or another drug in one or more samples obtained from the patient; updating the mathematical model based on a second response of the patient to administration of a specific drug of the plurality of drugs or another drug according to the first dosing regimen or a modified version of the first dosing regimen; calculating an updated dosing regimen for reaching the treatment objective for the patient based on the updated mathematical model, the updated dosing regimen including the second dosage amount; 65. The pharmaceutical preparation of any of items 63 and 64, comprising: (Item 66) 66. The pharmaceutical composition of any of items 63-65, wherein the updating comprises a Bayesian updating. (Item 67) 67. The pharmaceutical preparation of any of items 63-66, wherein the initial dosage amount is the amount per unit of the specific drug administered to the patient in an initial dosing regimen prior to administration of the first dosing regimen. (Item 68) 68. The pharmaceutical of any of items 63-67, wherein the concentration or response data is obtained upon administration of the initial dosage. (Item 69) 69. The pharmaceutical of any of items 63-68, wherein the received input includes physiological data indicative of one or more measurements of at least one physiological parameter of the patient. (Item 70) 70. The preparation of claim 69, wherein the at least one physiological parameter of the patient comprises at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapies, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure readings, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharp score, and demographic information. (Item 71) 71. The pharmaceutical of any of items 69 and 70, wherein the model analysis further comprises selecting, from a set of mathematical models, the mathematical model that best fits the received physiological data. (Item 72) 72. The pharmaceutical according to any of items 63-71, wherein the model is a Bayesian model. (Item 73) 73. The pharmaceutical of any of items 63-72, wherein the specific drug is infliximab. (Item 74) 74. The pharmaceutical of any of items 63-73, wherein the model analysis comprises a Bayesian analysis. (Item 75) 75. The pharmaceutical composition according to any of items 63-74, wherein the model analysis is performed by a cloud-based computing system. [Brief description of the drawings]

[0028] The foregoing and other objects and advantages will become apparent upon consideration of the following detailed description in conjunction with the accompanying drawings, in which like reference characters refer to like parts throughout.

[0029] [Figure 1] FIG. 1 is a flow chart illustrating a method for determining a patient-specific dosing regimen for a set of medications, according to one illustrative implementation.

[0030] [Diagram 2] FIG. 2 is a block diagram of a system for performing the methods described herein, according to one illustrative implementation.

[0031] [Diagram 3] 3A and 3B are exemplary graphs depicting simulated concentration / time profiles showing predicted patient responses to multiple drugs for a given dosing regimen to meet target levels based on a drug-independent model, according to one illustrative implementation.

[0032] [Figure 4] FIG. 4 illustrates a process for developing a drug-independent model for a set of drugs, according to one illustrative implementation.

[0033] [Diagram 5] FIG. 5 is a flowchart showing a process for modifying an adaptive dosing system, according to one illustrative implementation.

[0034] [Figure 6] FIG. 6 shows a system diagram of a computer network for an adaptive medication dispensing system, according to one illustrative implementation.

[0035] [Figure 7A] 7A and 7B are example displays of a user interface on a clinical portal providing several recommended medication regimens according to one illustrative implementation. [Figure 7B] 7A and 7B are example displays of a user interface on a clinical portal providing several recommended medication regimens according to one illustrative implementation. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0036] (Detailed Description) The systems and methods described herein determine a patient-specific drug dosing regimen for a patient using a computerized drug dosing regimen recommendation system that is applied to a class or other group of drugs. The systems and methods described herein provide a population model that can be used for a class or set of drugs. In particular, the systems and methods herein can be configured to develop a patient-specific dosing regimen based on data indicating pharmacokinetic ("PK") and pharmacodynamic ("PD") parameters of multiple drugs. Examples of such drugs are included in Table 1. The following definitions are used in Table 1: "iv" is intravenous, "sc" is subcutaneous, "RA" is rheumatoid arthritis, "AS" is ankylosing spondylitis, "UC" is ulcerative colitis, "CD" is Crohn's disease, "IBD" is inflammatory bowel disease, which may include ulcerative colitis and / or Crohn's disease, "PSO" is psoriasis, "PSA" is psoriatic arthritis, and "MS" is multiple sclerosis. In some implementations, the systems described herein are not specific to a particular drug, but instead may be applied to a class or other subset or group of drugs (e.g., drugs expected to have similar PK / PD, drugs known to be candidates for treating a particular disease state, or other similarities). The deployment and application of a drug-independent model allows for greater utility for a single model. Rather than implementing multiple models, each model corresponding to a single drug in a class, a drug-independent model is applicable to all drugs in the class and thus may be applicable to a wider range of patients across a range of administration routes than a single drug model. [Table 1-1] [Table 1-2]

[0037] For example, the models described herein may include a pharmacokinetic drug-independent model that can be used for all biologics used in the treatment of inflammatory diseases. Such models can be used to suggest and / or test dosing regimens for a range of drugs, such as those listed in Table 1, including fully human monoclonal antibodies (mAbs), chimeric mAbs, humanized mAbs, fusion proteins, and mAb fragments (i.e., drugs with different pharmacokinetic properties but similar molecular weights and indications) using the same model. The models may be used in a wide range of patient populations, including inflammatory bowel disease, rheumatoid arthritis, psoriatic arthritis, psoriasis, multiple sclerosis, and other such diseases resulting from immune dysregulation. The development and application of drug-independent Bayesian models for agents in other broad drug sets (e.g., aminoglycoside antibiotics, chemotherapy drugs that cause low white blood cell counts, etc.) is similarly feasible. Drugs within a category may be administered through various routes, such as subcutaneous, intravenous, oral, intramuscular, intrathecal, sublingual, buccal, rectal, vaginal, ocular, nasal, inhalation, aerosol, dermal, or transdermal. A drug-independent model may take into account the route of administration by accepting the route as a variable input to the system, allowing for greater flexibility for the model. A computational model, such as a Bayesian model, may be used to determine the dosing regimen recommendation. For example, each iteration of the model may include a calculation or determination of a recommended dosing regimen. When additional data (such as physiological parameter data or drug concentration data obtained from the patient) is made available, another iteration of the model may be performed to determine an updated recommended dosing regimen based on the additional data. This process may be repeated any number of times to reflect any new data describing the patient.

[0038] As used herein, a "dosing regimen" includes at least one dosage of a drug or class of drugs and a recommended schedule for administering at least one dosage of the drug to a patient. The dosage may be a multiple of the available dosage unit for the drug. For example, the available dosage unit may be a tablet or a suitable piece of a tablet that results when easily divided, such as a half of a tablet. In some implementations, the dosage may be an integer multiple of the available dosage unit for the drug. For example, the available dosage unit may be a 10 mg injection or capsule that cannot be divided. For some routes of administration (e.g., IV and subcutaneous), any fraction of the dosage strength can be administered. The recommended schedule includes a recommended time for administering the next dosage of the drug to the patient such that the predicted concentration-time profile of the drug in the patient in response to the first pharmaceutical dosage regimen is at or above the target drug exposure or response level (e.g., target drug concentration trough level) at the recommended time.

[0039] Because the drug-independent model applies to a set of drugs, rather than just a single drug, the model may retain patient-specific information when a patient is treated with multiple drugs in the set of drugs. For example, the set of drugs may include infliximab, vedolizumab, adalimumab, and other anti-inflammatory biologics. If a patient is treated with one drug (e.g., infliximab) and then later treated with another drug (e.g., vedolizumab), the model may retain all patient-specific data (drug concentration measurements, clearance rates, weight measurements, etc.) from the patient's infliximab-based treatment when determining the appropriate dosing regimen once the patient is treated with the new drug. Retaining patient-specific data allows the drug-independent model to accurately predict the patient's ability to process the drug, thereby providing a more suitable patient-specific dosing regimen when the patient changes drug therapy. Because the drug-independent model may fit a wide range of data with multiple routes of administration and a wide range of diseases, the model should learn about the drug and the individual patient (e.g., via Bayesian learning). Such drug-independent pharmacokinetic models, for example, represent a novel application of traditional population pharmacokinetic modeling. The ability to develop such drug-independent pharmacokinetic (PK) models can be based on one or more of several factors, including: 1) a common generic structural PK model for all drugs in a specific class; 2) similar effects of patient factors on PK parameters; and 3) similar indications. Thus, development and application of drug-independent Bayesian models for drugs in other broad drug classes (e.g., aminoglycoside antibiotics) is similarly feasible, allowing for greater utility of a single drug-independent model rather than implementation of multiple models for each drug in a class. The development of specific "classes" for drug-independent modeling purposes is a novel means of classification or categorization for drugs.

[0040] Similarly, drug-independent models can be constructed for drug classes that exhibit commonalities regarding pharmacodynamic effects (the measured response of the drug). For example, many chemotherapy drugs cause neutropenia or low white blood cell counts. This is a delayed response, with the nadir white blood cell count generally occurring 7-9 days after chemotherapy is administered. Although each drug's effect on duration and white blood cell count nadir may differ, the underlying relationship between drug exposure and white blood cell count decline is structurally similar, allowing practical drug-independent pharmacodynamic models to be developed for classes of chemotherapy drugs that cause white blood cell reduction.

[0041] In some implementations, the drug-independent model accounts for pharmacokinetics and pharmacodynamics. The model includes a PK component and a PD component, which may be separate or interrelated within the model. For example, the PK component and the PD component may be interrelated such that the effect of PK on PD and the effect of PD on PK are included within the model. The PK component may include a PK clearance parameter, and the PD component includes a PD response parameter. The interrelationship between the PK component and the PD component may be reflected by a PK clearance parameter that is a function of the PD response, or vice versa. One or more differential equations may be used to account for the patient response and the clearance of the drug in the patient. The PD component of the model may consist of a first differential equation, and the PK component of the model may consist of a second differential equation. The first differential equation may represent the PD response by the patient, and the second differential equation may represent the PK clearance by the patient. The first or second differential equation may include the PD response and / or the PK clearance.

[0042] The systems and methods described herein may output a recommended dosing regimen for a class of drugs without identifying the specific drug to administer to the patient. As used herein, a "dosing regimen" may include a dosage of the drug and a recommended schedule for administering the dosage to the patient. The recommended schedule includes a recommended time for administering a next dosage of the drug to the patient to achieve a predicted concentration-time profile of the drug in the patient in response to a first pharmaceutical drug dosing regimen that is at or above a target, e.g., drug concentration trough level, at the recommended time.

[0043] A drug class refers to a group of more than one drug that exhibits at least one similar PK or PD effect, or shares a common mechanism of action or some other similarity. For example, the similar PK effect may be clearance within a specific range. The similar effect may be a measured concentration within a specific range, such as bioavailability, absorption, white blood cell count, blood concentration level, or any of the biomarkers / measurements discussed herein. The specific range may be within a 10-fold difference, i.e., values ​​between 0.1 and 1 may be considered similar. The specific range may be defined by the user on the system interface. Drugs may be grouped into classes by the disease they treat, such as systemic inflammatory diseases, or more specifically, inflammatory bowel disease (IBD), ulcerative colitis, Crohn's disease, rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, or multiple sclerosis. Drug classes may also be based on drug structure. For example, the classes may include monoclonal antibodies (mAbs), chimeric mAbs, fully human mAbs, humanized mAbs, fusion proteins, and / or mAb fragments. Drug classes may include anti-inflammatory compounds, chemotherapeutic agents, corticosteroids, immunomodulators, antibiotics, or biologic therapies, or any other suitable group. Drug classes may further be determined by patient population, i.e., pediatrics, geriatrics. Drug classes may also be determined by the user based on other criteria, and the members of the class (or group) may be electronically designated as being part of the class (or group) in a database. The database is accessible to the systems and methods disclosed herein for use in determining class (or other group) based dosing regimens. Drug classes or groups may include variations of the same drug, such as the same drug with different routes of administration or different manufacturers. This feature may be particularly useful when a physician needs to compare generic and branded drugs that vary in price, availability, indications, and / or routes. Many of the examples described herein relate to the pharmaceutical agent infliximab.However, implementations described herein may be applied to immunosuppressants, anti-inflammatory, antibiotics, antibacterials, chemotherapy, anticoagulants, procoagulants, antidepressants, antipsychotics, psychostimulants, antidiabetics, anticonvulsants, analgesics, or any other suitable treatments.

[0044] Many of the implementations described herein relate to the treatment of IBD, such as ulcerative colitis or Crohn's disease. Although there is no standard treatment regimen for IBD, the following groups of drugs can be used to treat IBD patients: anti-inflammatory compounds, corticosteroids, immunomodulators, antibiotics, or biological therapies. One recently developed treatment involves biological therapies (e.g., monoclonal antibodies (mAbs) such as infliximab) that target, bind to, and inactivate an inflammatory protein called tumor necrosis factor (TNF). In some cases, a combination of anti-TNF drugs, such as infliximab, can be combined with one or more immunomodulators, such as thiopurines. Such combination therapy can effectively reduce the rate of elimination (thereby increasing the drug concentration level in the patient's blood) and reduce the formation of anti-drug antibodies. The greatest challenge in treating patients with IBD is ensuring that the patient receives sufficient exposure to the treatment. The body presents several routes of "clearance" for drugs. For example, the patient's metabolism may degrade the mAb by proteolysis (degradation of proteins), by cellular uptake, and by additional atypical clearance mechanisms associated with IBD. For example, due to the nature of the disease, patients with conditions such as focal segmental glomerulosclerosis (FSGS) often suffer from excessive drug loss into the urinary or digestive tract. Also, in severe IBD, mAbs are sometimes lost in the stool through ulcerated and denuded mucosa, creating additional routes of clearance. Overall, IBD patients are estimated to have a 40%-50% higher infliximab loss rate than other inflammatory diseases, making IBD particularly difficult to treat. The systems and methods described herein may also develop medication regimens to treat rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, plaque psoriasis, low levels of clotting factor VIII, hemophilia, schizophrenia, bipolar disorder, depression, bipolar disorder, infectious disease, cancer, stroke, transplantation, or any other suitable affliction.

[0045] Inputs into the system may be used to update and refine the model for a specific patient taking a specific drug. Inputs into the systems described herein may include concentration data, physiological data, and target responses. Inputs into the model generally include concentration data, physiological data, and target responses. As discussed above, the concentration data indicates one or more concentration levels of a drug in one or more samples obtained from a patient, such as blood, plasma, urine, hair, saliva, or any other suitable patient sample. The concentration data may reflect a measurement of the concentration level of the drug itself in the patient sample, or another analyte in the patient sample that indicates the amount of the drug in the patient's body. The drug may be part of a treatment plan for treating a patient suffering from a particular health condition, such as a disease or disorder, such as inflammatory bowel disease (IBD, including ulcerative colitis and Crohn's disease), rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, plaque psoriasis, or any other suitable affliction. Drugs used to treat such conditions may include monoclonal antibodies (mAbs), such as infliximab or adalimumab. Although many of the examples described herein refer to using infliximab to treat IBD, it should be understood that the systems and methods of the present disclosure are applicable to any drug or treatment that loses its effectiveness over time in a measurable way, and may be used to treat any number of diseases, including any inflammatory disease, such as IBD.

[0046] Inputs to the system may also include other drug information such as the disease to be treated, the class of drug, the route of administration, available dosage strengths, preferred dosage (e.g., 100 mg vials, 50 mg tablets, etc.), and whether the specific drug is fully human or not (e.g., chimeric). The drug information may be used to determine the available treatment options for the patient, the model to be selected, and the model parameters. For example, patients treated for IBD often have higher clearance rates than patients not suffering from IBD, and drug dosing regimens for treatment with IBD must be adjusted accordingly. The preferred dosage may modify the dosing regimen before the regimen is recommended for the patient. For example, if a drug is only available in 100 mg vials, the recommended dosage may be rounded to the nearest 100 mg increment. In some implementations, the drug information excludes information identifying a drug currently being used to treat the patient. For example, the drug data may be general to the drug class. The physiological data generally indicates one or more measurements of at least one physiological parameter of the patient. This may include at least one of medical record information, markers of inflammation, indicators of drug clearance such as albumin measurements or C-reactive protein (CRP) measurements, anti-drug antibody measurements, hematocrit levels, biomarkers of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapies, previous laboratory test result information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure readings, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharpe / van der Heijde score, and demographic information.

[0047] The target response may be selected by the physician based on the physician's assessment of the patient's tolerance and response to drug therapy. In an embodiment, the target response may include a target drug concentration level (such as a concentration maximum, minimum, or exposure window) of the drug in a sample obtained from the patient and may be used to determine when the patient should receive the next dosage and the next dosage. The target drug concentration level may include a target drug concentration trough level, a target drug concentration maximum, a target drug area under the concentration time curve (AUC), both the target drug concentration maximum and trough, a target pharmacodynamic endpoint such as blood pressure or clotting time, or any suitable metric of drug exposure. The target may be determined by the physician based on the drug data and / or concentrations or responses. In some implementations, the target may be automatically determined by the system to produce a therapeutic response in the patient. The system may evaluate multiple input targets to determine one or more targets that produce a therapeutic response in the patient. The inputs described above (eg, concentration data, physiological data, drug information, and target responses) are used by the systems and methods of the present disclosure to personalize dosing regimen recommendations for the patient.

[0048] Based on the received input, the systems and methods described herein set one or more parameter values ​​for a computational model (such as any of the model parameters described in U.S. Patent Application No. 15 / 094,379 (the '379 application), filed April 8, 2016, and published as U.S. Patent Application Publication No. 2016-0300037, entitled "Systems and Methods for Patient-Specific Dosing," which is incorporated herein by reference in its entirety) that generates a prediction of the concentration-time profile of the drug in the patient. In some implementations, the computational model is a Bayesian model. For example, the computational model may consider historical and / or current patient data and develop a patient-specific targeted dosing regimen. As discussed in the '379 application, the computational model may include a pharmacokinetic component that is indicative of the concentration-time profile of the drug, and a pharmacodynamic component based on synthesis and degradation rates of pharmacodynamic markers that are indicative of the patient's individual response to the drug. The computational model may be selected from a set of computational models that best fit the received physiological data. For example, if the patient is a 45 year old male, the system may select a computational model specific to males between the ages of 30 and 50. The computational model can be individualized to a particular patient by taking into account patient specific measurements, such as additional concentration data and additional physiological parameter data described herein.

[0049] A patient-specific drug dosing regimen may be provided as a function of a mathematical model (e.g., a pharmacokinetic and / or pharmacodynamic model) that is updated to take into account the observed patient response. This is described in detail in the '379 application, which is incorporated herein by reference in its entirety. In particular, the observed response of a specific patient to an initial dosing regimen is used to adjust the dosing regimen. The patient's observed response (e.g., the observed drug concentration in the patient's blood) is used in conjunction with the mathematical model and patient-specific characteristics to account for inter-subject variability (BSV) that may not be accounted for by the mathematical model alone. The observed response of a specific patient can be used to refine the model and associated assumptions, effectively personalizing them so that the model can be used to more accurately predict an expected response to a proposed dosing regimen for a specific patient. In this way, the observed patient-specific response data is used as "feedback" to adapt a general model describing typical patient responses to a patient-specific model capable of accurately predicting a patient-specific response, so that a patient-specific dosing regimen may be predicted, proposed, and / or evaluated on a patient-specific basis. Using observed response data to individualize the model allows the model to be modified to take into account BSV, which is not considered in previous mathematical models that accounted for only typical responses for a patient population, or "covariate-typical" responses for a typical patient with certain characteristics that are considered as covariates in the model.

[0050] The present system and method may rely on Bayesian analysis. For example, Bayesian analysis may be used to determine the appropriate dosage required to achieve a desired outcome, such as maintaining the concentration of a drug in a patient's blood near a particular level. Bayesian analysis may involve Bayesian inference and Bayesian updating. These Bayesian techniques may be used to develop models that are not only a function of patient-specific characteristics considered in the model as covariate patient factors, but also a function of observed patient-specific responses that reflect inter-subject variability (BSV) that is not considered in the model itself and that distinguishes a specific patient from a typical patient reflected by the model. In this way, the present disclosure considers variability between individual patients that is not accounted for and / or considered by traditional mathematical models (e.g., patient responses that would not have been predicted based on dosing regimens and patient factors alone). Furthermore, the present disclosure allows patient factors considered by typical models, such as weight, age, race, clinical test results, etc., to be treated as continuous functions rather than as categorical (cut-off) values. By doing this, the model is tailored to the particular patient such that patient-specific inferences and analyses can be performed to predict, suggest, and / or evaluate a personalized dosing regimen for the particular patient.

[0051] It is noted that the present disclosure can be used not only to retrospectively assess a dosing regimen previously administered to a patient, but also to proactively assess a proposed dosing regimen before administering the proposed dosing regimen to the patient, or to identify a dosing regimen (dosage administered, dosing interval, and route of administration) for a patient that will achieve a desired outcome. The Bayesian inference process may be used to test various dosing regimens for a patient as a function of the patient's specific characteristics, which are considered as patient factor covariates and mathematical models in the model. The inference involves evaluating dosing regimens based on predicted responses for a typical patient with patient-specific characteristics. In general, Bayesian inference involves using mathematical model parameters to infer the likely responses that a specific patient will exhibit with various dosing regimens. It is noted that the inference step allows for the determination of the likely patient response to a proposed dosing regimen before the actual administration of the proposed dosing regimen. Thus, the step of guessing can be used to test multiple different proposed dosing regimens (e.g., different dosage amounts, dosing intervals, and / or routes of administration) to determine how each dosing regimen may affect the patient as predicted by the patient-specific factors and / or data in the model / composite model. The guesses may be compared to generate a set of dosing regimens that are satisfactory or best for achieving the therapeutic objective or target exposure or concentration level. For example, the target may involve maintaining trough blood concentration levels above a therapeutic threshold.

[0052] In some implementations, the recommended dosing regimen is provided with a confidence interval that indicates the likelihood that a particular dosing regimen will be therapeutically effective for the patient. In particular, the confidence interval of the response or concentration estimated from the individual data may be assessed based on the complexity of the model and the amount of individual data (PK data and / or PD data). In particular, the confidence interval may reflect the possible error of the individual predictions from the model. Initially, when no individual measurements have been obtained from the patient, the model predictions have an error associated with them that is approximately equal to the unexplained variability of the PK and PD models. However, as individual measurements are obtained and introduced into these models, the error (or, equivalently, the confidence interval) decreases before eventually approaching the calibration error, which may correspond to the measurement error. The confidence interval may also be provided to a clinical portal to give medical professionals a sense of the amount of error remaining in the model predictions.

[0053] A Bayesian updating process may be used to update the model based on the patient's response to the dosing regimen. In general, Bayesian updating involves Bayesian inference, which is a method in which Bayes' rule is used to update probability estimates for hypotheses as additional evidence is acquired. Bayesian updating is particularly important in dynamic analysis of data collected over time (sequentially). The method as applied here uses a model that not only describes the time course of exposure and / or response, but also includes terms that describe unexplained (random) variability of exposure and response. The result of Bayesian updating is a set of parameters that are conditional on the observed data. The process involves sampling parameters from a previous distribution (e.g., an underlying model) and calculating an expected response based on the underlying model. For each underlying model, the difference between the expected value of the model and the observed data is compared. This difference is referred to as the "objective function". The parameters are then adjusted based on the objective function, and the new parameters are tested against the observed data by comparing the difference between the expected value of the new model and the observed data. This process is carried out iteratively until the objective function is minimized, suggesting that the parameters that minimize the objective function best explain the current data. In some implementations, a random function may be used to inject some variability and ensure that a drug-independent minimum of the objective function is obtained.

[0054] In some cases, the disclosed system and method determines dosing regimen recommendations based on how the patient responded to previous treatment methods from within or across drug classes (or other groups). As an example, when a patient does not respond to treatment with one drug (e.g., a previously administered drug), physicians sometimes switch the patient to a different drug (e.g., a currently administered drug), and drugs may be related to each other (e.g., from the same class that share a common mechanism of action). Lack of response to treatment is sometimes referred to as a "failure" of treatment, which may occur in patients with high elimination rates. In these patients, the drug is often cleared from the body before the full beneficial effect of the drug is realized by the body. Generally speaking, IBD patients with high elimination rates for one drug (previously administered) may also be expected to have high elimination rates for another similar drug (not yet administered), since the mechanisms of clearance for similar drugs are generally similar as well. In this way, the clearance rate of a previously administered drug (as reflected by the measured drug concentration level) may inform the clearance rate of the drug to be administered. The systems and methods described herein may exploit this correlation and use historical patient data with one drug (e.g., a previously administered drug) in determining a recommended dosing regimen for another drug (e.g., a drug currently being used to treat the patient). For example, a physician may prescribe adalimumab for a patient. The patient may have a higher than average rate of clearance for adalimumab. If the patient fails an adalimumab-based therapy, the physician may then attempt therapy with infliximab. The systems and methods described herein then take into account the higher than average clearance of adalimumab when calculating a dosing regimen for infliximab, since the patient is likely to clear infliximab at a higher than average rate as well due to the fact that adalimumab and infliximab are both mAbs with similar clearance mechanisms.

[0055] The system and method described herein uses an iterative approach to determine and provide a recommended dosing regimen. In an embodiment, an initial dosing regimen (determined based on the patient's available information and the physician's experience) is administered to the patient. Data indicating the patient's response or reaction to the initial dosing regimen, such as the patient's physiological and / or concentration data, is provided to the system as feedback on the initial dosing regimen. All, some, or none of the data is then used as input to a computational model that calculates an updated dosing regimen that is provided to the physician as a recommendation. The physician may choose to administer the dosing regimen exactly as recommended, or the physician may choose to slightly modify the recommended dosing regimen before administering it. For example, the recommended dosing regimen may include a specific dosing interval (e.g., 4 weeks) and a specific dosage (e.g., 1.9 vials). The physician may choose to modify the regimen to accommodate the patient's schedule (e.g., if the patient is only able to return to a different dosage after 4 weeks and one day), round up to a certain number of vials (e.g., to 2 vials), or both. This iterative approach is described in more detail below.

[0056] Using the computational model and set parameters described above, the systems and methods described herein determine a first pharmaceutical dosing regimen for the patient. The first pharmaceutical dosing regimen includes at least one dosage of the drug and a recommended schedule for administering the at least one dosage of the drug to the patient. The recommended schedule includes a recommended time for administering a next dosage of the drug to the patient such that a predicted concentration-time profile or pharmacodynamic marker profile of the drug in the patient in response to the first pharmaceutical dosing regimen is at or above a target drug exposure level at the recommended time. In some implementations, the dosing regimen may be displayed (e.g., through a user interface). In some implementations, a physician may receive or view (e.g., through a user interface) the first dosing regimen. The physician may decide to modify the dosing regimen before it is administered to the patient. Once the physician determines the course of treatment, the dosage may be administered to the patient and additional data (indicative of how the patient is responding to the dosage) is received.

[0057] In some implementations, after initiation of administration of a dosing regimen based at least in part on the first pharmaceutical dosing regimen to the patient, the systems and methods described herein receive additional concentration data and additional physiological data obtained from the patient. A medical practitioner may choose to administer the dosing regimen to the patient strictly as recommended by the system. However, the medical practitioner may also choose to modify the first pharmaceutical dosing regimen prior to administration of the drug. For example, the medical practitioner may choose to round up or down the dosage, may choose to modify the dosing time to better fit the patient and physician schedule, may make any suitable modification to the first pharmaceutical dosing regimen, or may make any suitable combination thereof. For example, the first pharmaceutical dosing regimen may prescribe a dosage of 100 mg on April 20th, but the physician may decide to prescribe administration of 90 mg on April 22nd. Details of the administered dosage can then be entered into the system. The system may then receive additional data (eg, concentration and physiological parameter data) indicative of how the patient is responding to the administered dosage.

[0058] In many cases, when a patient is in the early stages of treatment, there is not much patient data. In particular, historical data showing how a patient responded or reacted to different dosages of a drug is typically not available. In this case, when there is little patient data to rely on, the system and method of the present disclosure may limit the number of concentration data points considered in the model iteration to only a single data point, which may be the most recent data point. The most recent data point may be the only concentration data input into the model, since it most accurately reflects the patient's current status and allows the system and method of the present disclosure to heavily weight the data point in determining the recommended dosing regimen. In this way, the input into the model is determined based on whether the cycle of treatment of the first pharmaceutical dosing regimen administered exceeds a certain percentage of the total time of the first pharmaceutical dosing regimen (e.g., whether the patient is in the early stages of treatment).

[0059] The systems and methods described herein may display information to a user. The system may display information through a user interface with which a physician or other user may interact. For example, the user interface may display a dosing regimen (e.g., as shown in FIGS. 7A and 7B and described below) or may display concentration and / or physiological parameter data. In some implementations, a user may switch or otherwise select which data to view on the user interface. In some implementations, a predicted concentration-time profile of a drug in a patient in response to a first pharmaceutical dosing regimen, as generated by a computational model, is displayed. An indication of at least a portion of the concentration data and additional concentration data, physiological data and additional physiological data, target drug exposure level, and a recommended time as a point where the predicted concentration-time profile of the drug intersects with the target drug exposure level may also be displayed. For example, a physician may use a graphical user interface to input a target response for a patient, input patient data (e.g., physiological and concentration data), and view predicted concentration-time profiles for the patient in response to different dosing regimens.

[0060] In some implementations, inputs such as those described above are retrieved from an electronic medical record. For example, a physician may enter a patient's name into a medication system, which may then communicate with an electronic database system (stored locally in the medication system or remotely) that holds a plurality of patient information including health information about the patient. The health information in the electronic database system may include, for example, current medications that the patient is taking. The medication system may then receive information about the patient's health and / or medication information from the electronic database system. In some examples, medication information (such as drug class, route of administration, minimum dosage, or any other suitable information) is retrieved from a separate database or server once the drug name has been identified. Such a system may be advantageous, for example, in reducing the time required for a clinician to manually enter information and reducing the opportunity for input errors by automatically pulling information from these databases. Such EMR applications may be activated or deactivated within a given system depending on the situation and need.

[0061] The systems and methods described herein may be used to predict patient drug clearance for a class (or other group) of drugs. Such models may be standardized to account for differences between drugs within a group of drugs. In some implementations, the models are generated by collecting parameter values ​​from a set of published models that correspond to the class of drugs. The parameter values ​​may be collected in a lookup table. The parameter values ​​may be converted to "standardized values" so that they can be compared or pooled in a drug-independent model. This allows the system to simulate PK characteristics for patient populations for published models with covariate effects such as those published, and for patient populations for extended published models with all measurements and assumed covariate effects. Standardized parameters may include weight, albumin, ADA negativity, presence of immunosuppressants, CRP, glucose, human origin or chimeric, non-IBD disease, gender, non-linear clearance, and CL. The lookup table may be used to normalize the parameters to allow preliminary estimates from the drug-independent model. The lookup table may be manipulated by a user through a user interface or stored in the model database 606D of FIG. 6. The lookup table may be structured such that a subset of the table may be sent to a simulation function in the program so that each drug may be easily simulated in a variety of scenarios. The simulated concentrations from the normalized parameters for each drug in a group of drugs may be compared and analyzed with respect to the pooled data for that group of drugs to fit the simulated concentration data to the pooled data. A drug-independent model for a group of drugs provides a set of parameters that apply to or represent all of the drugs in the group.

[0062] FIG. 1 illustrates a process 100 implemented by a system or method for providing a dosing regimen for multiple drugs (also referred to as a set of drugs) according to one illustrative implementation. In step 102, input to the system is received. The input includes drug data indicating a set of drugs and route of administration, concentration or response data, and target drug exposure or response levels. In some implementations, drugs in the set of drugs are expected to exhibit similar PK behavior, similar PD behavior, or both. In some implementations, the drug data includes, for each drug, an indication, such as an intended patient population or disease. In some implementations, the drug data includes information about the structure or function of the drug, such as a mechanism of action. Information about the structure of the drug may include, for example, if one or more drugs in the set of drugs are antibodies, whether each antibody is chimeric, fragmented, human, or fully human. It is possible that drug data may be provided without identifying a specific drug, allowing the physician to obtain simulated results, such as predicted concentration time profiles, clearance, and / or dosing regimens, without identifying or constraining the results to a specific drug. The concentration or response data may indicate the concentration or response level of a specific drug of a set of drugs in a sample obtained from a patient. The concentration or response data may further indicate the concentration or response level of a specific drug administered using a specific route of administration in a sample obtained from a patient. Patient-specific covariates or measurements, such as weight, albumin, ADA negativity, presence of immunosuppressants, CRP levels, glucose levels, complications, disease status, and gender, may also be included as inputs to the system.

[0063] In step 104, a mathematical model is selected from a database (e.g., model database 606D of FIG. 6) stored in memory. The database is by a processor and stores a plurality of mathematical models. The mathematical models represent responses by a plurality of patients to a plurality of drugs in a set of drugs, each response being indicative of a patient response to at least one drug in the set of drugs, and the mathematical model is not specific to a particular drug. The mathematical model may be selected based on one or more of the inputs to the system.

[0064] In some implementations, the mathematical model is selected based on the drug data. When the model is selected based on the drug data, the selection step may involve comparing the parameters or covariates of the model with the drug data. This step may involve maximizing the number of similar parameters between the model and the set or class of drugs. Each model in the database may be associated with an error or confidence interval that indicates the model's performance so far, and the model may be selected based on the error or confidence interval. The model may be selected based on the amount of data or information available for the model or applicable to it. It would be advantageous to use the most extensive information or data available so that the selected model can be refined so that the model makes more accurate predictions and provides recommendations that are more likely to achieve the target. For example, the concentration or response data and / or the type of concentration or response data may be analyzed to select a model that is capable of incorporating that type of data into the model, or that is capable of incorporating at least a portion or a maximum of that data. In some implementations, the selection may be performed by an optimization function. Instead of or in addition to the selection methods discussed above, a Bayesian method may be used to select a model. In some implementations, multiple models may be compared to each other and the model that best represents the patient or input may be selected. In certain implementations, the model is a Bayesian model, which may be capable of performing Bayesian analysis such as Bayesian inference.

[0065] In some implementations, parameters may be set for a computational model that generates a prediction of the concentration-time profile of the drug in the patient. The parameters are set based on the inputs received in step 102. In some implementations, the computational model is a Bayesian model. In some implementations, the computational model includes a pharmacokinetic component that is indicative of the concentration-time profile of the drug and a pharmacodynamic component that is based on synthesis and degradation rates of pharmacodynamic markers that are indicative of the patient's individual response to the drug. In some implementations, the process 300 includes an additional optional step, where the computational model is selected from a set of computational models that best fits the received physiological data. In some implementations, the computational model considers historical data that is indicative of the patient's response to the drug in the past to generate a prediction of the concentration-time profile of the drug in the patient. In some implementations, the past drug may belong to the same drug class as the current drug. In some implementations, the past drug may belong to a different drug class than the current drug.

[0066] In step 106, a plurality of predicted concentration-time profiles are estimated. The predicted concentration-time profiles are estimated using a selected mathematical model and based on the concentration data and the route of administration. Each predicted concentration-time profile of the plurality of predicted concentration-time profiles corresponds to a dosing regimen within the plurality of dosing regimens. Each dosing regimen of the plurality of dosing regimens may include at least one dosage and a recommended schedule for administering the at least one dosage to the patient. In some implementations, the recommended schedule is determined to maintain a target concentration or exposure or response level based on the at least one dosage or concentration or response data using the model. Step 106 may involve Bayesian inference.

[0067] In step 108, a first dosing regimen for the set of drugs is selected. The first dosing regimen is predicted to achieve the therapeutic objective based on the target drug exposure or response level. The first dosing regimen may be output from the system. In some implementations, the first dosing regimen is output for display on a user device. In some implementations, a plurality of first dosing regimens are output. In some implementations, a physician may select the first dosing regimen from the plurality of first dosing regimens. The selection of the first dosing regimen may be determined by a predetermined criterion or by a score associated with each dosing regimen in the plurality of dosing regimens. For example, a user may select the dosing regimen that best meets the therapeutic objective or best matches the target concentration or concentration-time profile. The score may be a percentage calculated by the system that represents the closeness to which the predicted concentration-time profile for a particular dosing regimen matches the target. The score may be a p-value, as known in the state of the art. The scores may indicate the residual error. For example, the scores may be the average residual error associated with the simulations using a particular dosing regimen, or may be presented as a confidence interval. These scores may be output along with the corresponding dosing regimen.

[0068] In an example, in response to viewing the selected / recommended first dosing regimen via the user interface, the healthcare practitioner may choose to administer the first dosing regimen as recommended, or the healthcare practitioner may choose to slightly modify the recommended dosing regimen, such as by changing one or more dates or times in the recommended schedule to accommodate the patient's or healthcare practitioner's schedule, and / or by changing the amount of the dosage (e.g., by rounding up to the nearest whole number of vials). The dosing regimen (e.g., either the recommended dosing regimen or a modified version of the recommended dosing regimen) is administered by the healthcare practitioner, for example, through an oral drug, intravenous, intramuscular, intrathecal, or subcutaneous injection, rectal or vaginal insertion, injection, topical, nasal, sublingual, or buccal application, inhalation or spray, ocular or otic route, or any other suitable route of administration. The dosage may be a multiple of the available dosage units for the drug. For example, the available dosage unit may be a tablet, or a suitable piece of a tablet that results when easily divided, such as half a tablet. In some implementations, the dosage may be an integer multiple of the available dosage unit for the drug. For example, the available dosage unit may be a 10 mg injection or capsule that cannot be divided. For some routes of administration (e.g., IV and subcutaneous), any fraction of the dosage strength can be administered.

[0069] A medical professional (or another user of the system) may provide the system with data indicative of the actual administered dosing regimen. In an example, if the administered dosing regimen is the same as the recommended dosing regimen, the user may simply select a button on the user interface indicating that the recommended dosing regimen has been selected to be administered. Alternatively, if the administered dosing regimen differs from the recommended dosing regimen, the user provides the system with data indicative of the administered dosing regimen. After initiation of administration of the dosing regimen to the patient (which may be the same as the recommended dosing regimen provided by the model or a modified version thereof), the system receives additional data indicative of the patient's observed response to the dosing regimen. In particular, the system may receive additional concentration data, additional physiological data, or both. The system receives additional concentration data that is concatenated with the concentration data. For example, the additional concentration data may represent the patient's response to the administered dosing regimen and may be a single data point or multiple data points. Additionally or alternatively, the system receives additional physiological parameter data that is concatenated with the concentration data. For example, the additional physiological parameter data may represent a patient's response to an administered dosing regimen and may be a single data point or multiple data points in the form of a vector or matrix. The system may receive the additional concentration data without receiving the additional physiological parameter data. Alternatively, the system may receive the additional physiological parameter data without receiving the additional concentration data. In another example, the system may receive both the additional concentration data and the additional physiological data.

[0070] The method may include additional steps. In some implementations, the additional steps include receiving additional drug data indicating an updated route of administration for the specific drug, updating the mathematical model based on the updated route of administration for the specific drug, calculating at least one updated dosing regimen to reach the therapeutic objective for the patient based on the updated mathematical model, and outputting at least one updated dosing regimen for the patient. In such implementations, these additional steps allow the physician to analyze the entire set of drugs while varying the route of administration to identify one or more dosing regimens that provide the best treatment for the patient, i.e., that best meet the therapeutic objective. In some implementations, the additional steps include receiving additional patient data indicative of the patient's response to administration of the specific drug according to the first dosing regimen or a modified version of the first dosing regimen, the additional patient data including additional concentration data indicative of one or more concentration levels of the specific drug in one or more samples obtained from the patient, updating the mathematical model based on the patient's second response to administration of the specific drug according to the first dosing regimen or a modified version of the first dosing regimen, calculating at least one updated dosing regimen to reach a treatment objective for the patient based on the updated mathematical model, and outputting the at least one updated dosing regimen for the patient. The updating step may involve Bayesian updating.

[0071] In some implementations, the system also receives as input certain drug constraints, such as available dosage strengths, preferred dosage strengths, available dosage units, manufacturing schedule, vial strength, or price. These constraints may be incorporated within the steps described above. For example, a first dosing regimen may be generated to include dosages that are a multiple of the available dosage units. The dosing schedule in the dosing regimen may be configured to be a multiple of the manufacturing schedule. The system may be configured to determine the dosage that best meets the therapeutic objective at the lowest cost to the patient based on the input drug cost. The route of administration may be at least one of subcutaneous, intravenous, oral, intramuscular, intrathecal, sublingual, buccal, rectal, vaginal, intraocular, nasal, inhalation, aerosol, dermal, or transdermal. The multiple drugs may be one of monoclonal antibodies and / or antibody constructs, cytokines, drugs used for enzyme replacement therapy, aminoglycoside antibiotics, and chemotherapy drugs that cause leukopenia.

[0072] In some implementations, each drug in the plurality of drugs shares a similar chemical structure. Each drug in the plurality of drugs may share a similar mechanism of action. In some implementations, the plurality of drugs is used to treat inflammatory diseases, such as inflammatory bowel disease (IBD), rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis, asthma, and multiple sclerosis. In some implementations, the specific drug is one of the drugs listed in Table 1. The drug data may include, for example, a dosage strength for the specific drug, to satisfy a parameter that may be included in the mathematical model or that may account for similarities or differences between drugs in the plurality of drugs, and / or an indicator that represents whether the specific drug is fully human, chimeric, or fragmented. The system may receive as input patient data indicating a patient disease to treat, for example, to classify drugs that are applicable to the patient disease, or to retrieve information related to the patient disease to be used in the model or various steps.

[0073] In some implementations, the received input includes physiological data indicative of one or more measurements of at least one physiological parameter of the patient. The at least one physiological parameter of the patient may include at least one of a marker of inflammation, an albumin measurement, an indicator of drug clearance, a C-reactive protein (CRP) measurement, an anti-drug antibody measurement, a hematocrit level, a biomarker of drug activity, weight, body size, sex, race, disease stage, disease status, previous therapy, previous laboratory test information, concomitantly administered medications, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure reading, Psoriasis Area and Severity Index (PASI) score, Disease Activity Score (DAS), Sharpe score, and demographic information. A mathematical model may be selected from the set of mathematical models to best fit the received physiological data.

[0074] In some implementations, the method includes generating a patient-specific concentration-time profile indicative of the patient response to the multiple drugs in response to the first dosing regimen and an indication of at least a portion of the concentration data and the additional concentration data for display. An indication of a target drug exposure or response level may be generated for display. In some implementations, the system receives historical data indicative of the patient's response to a second drug other than the specific drug. The computational model may take into account the historical data to generate a prediction of the concentration-time profile of the multiple drugs in the patient. Thus, the model may be able to use a broader range of data, such as data collected from administration of drugs other than the specific drug, to generate a prediction regarding the patient response to the multiple drugs. The drug-independent model may complement the lack of drug specificity using observed patient response data or past drug data, e.g., clinical trial data, to predict the patient response to the dosing regimen and recommend an optimized dosing regimen.

[0075] In some implementations, the system receives as input initial drug concentration input data representing an initial comparative concentration data point. In some cases, the user or physician may not have access to measured concentration or response data, and instead the system must estimate or calculate the first point based on physiological data. In some cases, the initial comparative concentration data point indicates a concentration level of a specific drug of the multiple drugs in a sample obtained from the patient. In some cases, the initial comparative concentration data point is calculated based on the patient's physiological parameters. The first dosing regimen may be calculated based on an estimate of a predicted concentration time profile based on the selected mathematical model and input data obtained from the specific patient's physiological parameters.

[0076] 2 is a block diagram of a computing device for performing any of the processes described herein. Each of these system components may be implemented on one or more computing devices 200. In some aspects, multiple components of these systems may be included within one computing device 200. In some implementations, components and storage devices may be implemented across several computing devices 200.

[0077] The computing device 200 includes at least one communication interface unit, an input / output controller 210, a system memory, and one or more data storage devices. The system memory includes at least one random access memory (RAM 202) and at least one read-only memory (ROM 204). All of these elements communicate with a central processing unit (CPU 206) to facilitate the operation of the computing device 200. The computing device 200 may be configured in many different ways. For example, the computing device 200 may be a traditional stand-alone computer, or alternatively, the functions of the computing device 200 may be distributed across multiple computer systems and architectures. In FIG. 2, the computing device 200 is coupled to other servers or systems via a network or a local network.

[0078] The computing device 200 may be configured in a distributed architecture where databases and processors are housed in separate units or locations. Some units perform primary processing functions and contain, at a minimum, a general controller or processor and system memory. In a distributed architecture implementation, these units may each be attached to a communication hub or port (not shown) that serves as a primary communication link with other servers, client or user computers, and other related devices via a communication interface unit 208. The communication hub or port may have minimal processing capabilities on its own, serving primarily as a communication router. A variety of communication protocols may be part of the system, including, but not limited to, Ethernet, SAP, SASTM, ATP, BLUETOOTH™, GSM, and TCP / IP.

[0079] CPU 206 includes a processor, such as one or more conventional microprocessors, and one or more auxiliary coprocessors, such as a mathematical coprocessor, for offloading workload from CPU 206. CPU 206 communicates with a communications interface unit 208 and an input / output controller 210, through which CPU 206 communicates with other devices, such as other servers, user terminals, or devices. Communications interface unit 208 and input / output controller 210 may include multiple communications channels, for example, for simultaneous communication with other processors, servers, or client terminals.

[0080] The CPU 206 also communicates with a data storage device. The data storage device may include any suitable combination of magnetic, optical, or semiconductor memory, and may include, for example, RAM 202, ROM 204, a flash drive, an optical disk such as a compact disk, or a hard disk or drive. The CPU 206 and the data storage device may each be located entirely within a single computer or other computing device, or may be connected to each other by a communication medium, such as a USB port, a serial port cable, a coaxial cable, an Ethernet cable, a telephone line, a radio frequency transceiver, or other similar wireless or wired medium, or a combination of the foregoing. For example, the CPU 206 may be connected to the data storage device via a communication interface unit 208. The CPU 206 may be configured to perform one or more specific processing functions.

[0081] The data storage device may store, for example, (i) an operating system 212 for the computing device 200, (ii) one or more applications 214 (e.g., computer program code or computer program products) adapted to instruct the CPU 206 in accordance with the systems and methods described herein, and in particular, in accordance with the processes described in detail with respect to the CPU 206, or (iii) a database 216 adapted to store information that may be utilized to store information requested by a program.

[0082] The operating system 212 and applications 214 may be stored, for example, in compressed, uncompiled, and encrypted formats and may include computer program code. The instructions of the program may be read into the processor's main memory from ROM 204 or from a computer-readable medium other than a data storage device, such as from RAM 202. Execution of a sequence of instructions in the program causes the CPU 206 to perform the process steps described herein, although hardwired circuitry may be used in place of or in combination with software instructions for implementation of the processes of the present invention. Thus, the described systems and methods are not limited to any specific combination of hardware and software.

[0083] Suitable computer program code may be provided to carry out one or more of the functions described herein. The programs may also include program elements such as an operating system 212, a database management system, and "device drivers" that enable the processor to interface with computer peripheral devices (e.g., video displays, keyboards, computer mice, etc.) via the input / output controllers 210.

[0084] The term "computer-readable medium" as used herein refers to any non-transitory medium that provides or participates in providing instructions to the processor of the computing device 200 (or any other processor of the device described herein) for execution. Such media can take many forms, including, but not limited to, non-volatile media and volatile media. Non-volatile media include, for example, optical, magnetic, or magneto-optical disks, or integrated circuit memories such as flash memory. Volatile media include dynamic random access memory (DRAM), which typically constitutes the main memory. Common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, DVDs, any other optical media, punch cards, paper tapes, any other physical media with patterns of holes, RAM, PROMs, EPROMs or EEPROMs (electronically erasable programmable read-only memory), FLASH-EEPROMs, any other memory chips or cartridges, or any other non-transitory media that a computer can read.

[0085] Various forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to CPU 206 (or any other processor of the devices described herein) for execution. For example, the instructions may initially be borne on a magnetic disk of a remote computer (not shown). The remote computer can load the instructions into its dynamic memory and transmit the instructions over an Ethernet connection, a cable line, or even a telephone line using a modem. A communications device 200 (e.g., a server) local to the computing device may receive the data over a separate communications line and place the data on a system bus for the processor. The system bus carries the data to a main memory, from which the processor reads and executes the instructions. The instructions received by the main memory may optionally be stored in memory either before or after execution by the processor. In addition, the instructions may be received via a communications port as electrical, electromagnetic, or optical signals, which are exemplary forms of wireless communications or data streams carrying various types of information.

[0086] 3A and 3B are graphs of predicted concentration-time profiles for a specific patient in response to a dosing regimen for each drug within a class of drugs. In each graph, the concentration-time profiles for a specific patient are simulated for a class of drugs using a drug-independent model, such that each profile corresponds to a drug within the class of drugs. The classes of drugs simulated here include various mAbs that share some commonalities as explained above. In this illustrative example, the dosing regimen is the standard dose and interval recommended for each drug in the package insert. In particular, FIG. 3A shows predicted concentration-time profiles for a class of drugs. One of the standard dosing regimens applied is shown in the chart above as a regimen including 10 mg of etanercept ("ETAN") administered intravenously in a single dose. If the concentrations are allowed to drop to very low (subtherapeutic) values, for example if the patient does not comply with the dosing regimen, the profiles will become significantly different due to differences between drugs. These differences are reflected in the drug-independent model by standardized parameters and may include certain aspects of the pharmacokinetics and / or pharmacodynamics associated with the drug.

[0087] Figure 3B shows the predicted concentration time profile for the drug class. One of the standard dosing regimens applied is shown in the chart above as a regimen containing 150 mg of saliumab ("SARI") administered subcutaneously every other week. The profiles depicted in Figure 3B are relatively similar, or at least follow similar trends, maintaining the concentration at higher values ​​over time. For example, the simulation may represent a patient who is compliant with the dosing regimen. Thus, differences between drugs and general variability can be minimized by managing patients at reasonable drug concentrations. Variability occurs between drugs within a drug class only when the dosing regimen is not followed or is insufficient.

[0088] FIG. 4 depicts a flow chart illustrating an exemplary method 400 of developing a drug-agnostic model as described herein. Step 402 involves generating a list of marketed drugs for treating a disease or patient population. These diseases or patient populations may be any of those described above or any disease or patient population targeted by the marketed drug. For example, the list of marketed drugs may be sourced from medical journals or literature or regulatory agencies such as the Food and Drug Administration (FDA). Table 1 may be an exemplary list of such drugs. Step 404 involves defining drug information. The drug information may include route of administration, structure, indications, constraints, PK / PD parameters, or other data described herein, as similarly discussed with respect to step 102 of FIG. 1. This information may be sourced from the same source as the list of drugs or may be typed in by a user. Step 406 involves obtaining a clinical pharmacological summary for the marketed drugs. The pharmacological information is used in the drug-agnostic model to take into account similarities or differences between marketed drugs. Step 408 involves extracting population PK model parameters. Step 410 involves simulating concentration-time profiles for each drug. Step 412 involves pooling the simulated data and fitting the data to a model, i.e., a drug-independent model. Step 414 involves comparing the simulation from the model to the original simulation. Step 416 involves further refining the model using drug information from step 404. Step 418 involves comparing the original simulation to the refined model. Step 420 involves further refining the model based on the dosage used, for example, based on concentration or response data as described in connection with step 102 of FIG. 1. In step 422, a final drug-independent model is provided.

[0089] In some implementations, in connection with step 408, a model is generated by collecting parameter values ​​for a set of published models corresponding to a class of drugs. It should be understood that step 408 may alternatively or additionally involve extracting population PD model parameters. In implementations where a PD model is desired, response data and response targets may be used instead of concentration data and concentration targets. The parameter values ​​may be collected in a lookup table or database described below in FIG. 6. The lookup table may be operated by the user or automatically by the system, as desired. The lookup table may be located anywhere in the system infrastructure, such as the systems described in FIGS. 2 and 6. The parameter values ​​may be converted to "normalized values" so that they can be compared or pooled in a drug-independent model. This allows the system to simulate PK and / or PD characteristics in step 410 for patient populations for published models with covariate effects as published, and for patient populations for extended published models with all of the measured and assumed covariate effects. The lookup table may be used to normalize the parameters and allow preliminary estimation from a drug-independent model. The normalization process may occur during steps 412 or 416. The lookup table may be structured such that a subset of the table can be sent to a simulation function within the program so that each drug can be easily simulated in various scenarios during, for example, step 410.

[0090] As described in step 414, the simulated concentrations from the normalized parameters for each drug in a group of drugs may be compared and analyzed with respect to the pooled data for that group of drugs to fit the simulated concentration data to the pooled data. The drug-independent model for a group of drugs provides a set of parameters that apply to or represent all drugs in that group, and this feature represents a significant technical contribution. Another technical effect of the drug-independent model is realized in step 420, where data collected from administered dosages can be used to further refine the model. Because the model is independent of a specific drug, the data from the dosages used can be from any drug in the class of drugs, and because drugs in the class of drugs share similarities that are reflected in the drug-independent model parameters, the model will receive that data and become further refined, for example, through Bayesian updating.

[0091] FIG. 5 shows a process 500 for using data from a patient's response to a previous treatment to inform a recommended dosing regimen for a different treatment, according to an illustrative implementation. The process 500 can be implemented using the computerized system 600 of FIG. 6 or any other suitable computerized system. In step 502, input to the system is received. The input includes previous concentration data indicating one or more previous concentration levels of a previous drug in one or more samples obtained from the patient, physiological data indicating one or more measurements of at least one physiological parameter of the patient, and a target drug exposure or response level of a current drug. The patient may have failed a previous drug-based therapy. Medical information (e.g., concentration data) indicating the patient's response to the previous drug may be useful to help determine how the patient will respond to the current drug. In some implementations, the previous and current drugs may belong to the same class or may have similar effects. For example, the previous and current drugs may be different monoclonal antibodies. If a patient's data indicates that the patient has a high elimination rate for the previous drug (e.g., the patient's body clears the previous drug from the patient's system at a rate higher than average or expected), the patient is likely to have a similarly high elimination rate for the current drug. By retaining information from the previous drug, the systems and methods described herein can rapidly determine an individualized dosing regimen for the current drug, thereby reducing the amount of time the patient is untreated when switching drugs.

[0092] In step 504, parameters are set for a computational model that generates a prediction of the concentration-time profile of the current drug in the patient. The computational model may be a Bayesian model, a PK / PD model, any of the models described above in connection with Figures 1 and 4 and below in connection with Figure 6 (e.g., stored in model database 606D), or any suitable model. The parameters are set based on the inputs received in step 502. In some implementations, the computational model is a Bayesian model. In some implementations, the computational model includes a pharmacokinetic component that is indicative of the concentration-time profile of the drug, and a pharmacodynamic component based on synthesis and degradation rates of pharmacodynamic markers that is indicative of the patient's individual response to the drug. In some implementations, process 1500 includes an additional optional step in which a computational model is selected from a set of computational models that best fits the received physiological data.

[0093] In step 506, a first pharmaceutical dosing regimen for the patient is determined using the computational model and the set parameters. The first pharmaceutical dosing regimen includes (i) at least one dosage of the drug, and (ii) a recommended schedule for administering the at least one dosage of the drug to the patient. The recommended schedule includes a recommended time for administering a next dosage of the drug to the patient, such that a predicted concentration-time profile of the drug in the patient in response to the first pharmaceutical dosing regimen is at a target drug concentration trough level at the recommended time. In step 508, additional concentration data and / or additional physiological data corresponding to the current drug are obtained from the patient. The additional concentration data and additional physiological data may result from administration of the first pharmaceutical dosing regimen to the patient. In some implementations, the process 500 includes an additional step of providing a display of various system inputs and outputs. In some implementations, process 500 provides a predicted concentration time profile of the drug in the patient in response to the first pharmaceutical dosing regimen, as generated by the computational model, and an indication of at least a portion of the concentration data for display (e.g., through user interface 622 of FIG. 6) as well as additional concentration data. In some implementations, an indication of a target drug concentration trough level is provided for display (e.g., through user interface 622 of FIG. 6).

[0094] At step 510, the concentration data is updated to include both the concentration data of step 502 and the additional concentration data of step 508. In some implementations, at step 510, the original concentration data from step 502 is excluded and the additional concentration data becomes the concentration data. At step 512, parameters for the computational model are updated based on the updated inputs, and at step 514, an iteration of the model is performed to determine a second pharmaceutical dosing regimen for the patient based on the updated parameters.

[0095] FIG. 6 shows a block diagram of a computerized system 600 for implementing the systems and methods disclosed herein. In particular, the system 600 uses drug-specific mathematical models and observed patient-specific responses to treatment to predict, suggest, modify, and evaluate suitable drug treatment plans for a specific patient. The system 600 includes a server 604, a clinical portal 614, a pharmacy portal 624, and an electronic database 606, all connected via a network 602. The server 604 includes a processor 605, the clinical portal 614 includes a processor 610 and a user interface 612, and the pharmacy portal 624 includes a processor 620 and a user interface 622. As used herein, the term "processor" or "computing device" refers to one or more computers, microprocessors, logic devices, servers, or other devices configured with hardware, firmware, and software to perform one or more of the computerized techniques described herein. Processors and processing devices may also include inputs, outputs, and one or more memory devices for storing data currently being processed. An illustrative computing device 200 that may be used to implement any of the processors and servers described herein is described in detail above with reference to FIG. 2. As used herein, a "user interface" includes, but is not limited to, any suitable combination of one or more input devices (e.g., a keypad, a touch screen, a trackball, a voice recognition system, etc.) and / or one or more output devices (e.g., a visual display, a speaker, a tactile display, a printing device, etc.). As used herein, a "portal" includes, but is not limited to, any suitable combination of one or more devices configured with hardware, firmware, and software to perform one or more of the computerized techniques described herein.Examples of user devices on which the portal may be implemented include, but are not limited to, personal computers, laptops, and mobile devices (smartphones, Blackberries, PDAs, tablet computers, etc.). For example, the portal may be implemented via a web browser or a mobile application installed on the user device. Although only one server, one clinical portal 614, and one pharmacy portal 624 are shown in FIG. 6 to avoid overcomplicating the drawing, the system 600 can support multiple servers and multiple clinical and pharmacy portals.

[0096] In FIG. 6, a patient 616 is examined by a medical practitioner 618, who has access to a clinical portal 614. The patient may suffer from a disease with a known progression and consults with the medical practitioner 618. The medical practitioner 618 takes measurements from the patient 616 and records these measurements via the clinical portal 614. For example, the medical practitioner 618 may take a sample of the patient's 616 blood and measure the concentration of a biomarker in the blood sample. In general, the medical practitioner 618 may take any suitable measurements of the patient 616, including test results such as concentration measurements from the patient's blood, urine, saliva, or any other fluid sampled from the patient. The measurements may correspond to observations of the patient 616 made by the medical practitioner 618, including any symptoms exhibited by the patient 616. For example, the medical practitioner 618 may perform tests on the patient to collect or measure physiological parameters or determine drug concentrations in the patient. This involves identifying patient characteristics (physiological parameters and concentrations) that are reflected as patient factor covariates in a mathematical model that will be used to predict the patient's response to a drug treatment regimen.

[0097] For example, if a model is constructed to describe typical patient response as a function of weight and gender covariates, the patient's weight and gender characteristics will be identified. Any other characteristics shown to predict response and thus reflected as patient factor covariates in the mathematical model may be identified. By way of example, such patient factor covariates may include markers of inflammation, albumin measurements, indicators of drug clearance, C-reactive protein (CRP) measurements, antibody measurements, hematocrit levels, biomarkers of drug activity, weight, body size, gender, race, disease stage, disease status, previous therapies, previous laboratory test information, concomitantly administered drugs, comorbidities, Mayo score, partial Mayo score, Harvey Bradshaw index, blood pressure readings, Psoriasis Area and Severity Index (PASI) score, and demographic information. Based on the patient's measurement data, the medical professional 618 may perform an assessment of the patient's disease status and may identify a suitable drug or class of drugs to administer to the patient 616 to treat the patient 616. The clinical portal 614 may then transmit the patient's measurements, the patient's disease status (as determined by a medical professional 618), and optionally, a medication identifier, via the network 602 to the server 604, which uses the received data to select one or more appropriate computational models from the model database 606.

[0098] The appropriate computational model is one that is determined to be capable of predicting the patient's response to administration of a drug or class of drugs. One or more selected computational models are used to determine a recommended set of planned dosages of drugs to administer to the patient, and the recommendations are sent back to the clinical portal 614 via the network 602 for viewing by the medical practitioner 618. Alternatively, the medical practitioner 618 may not be able to assess the patient's disease status or identify the drug, and either or both of these steps may be performed by the server 604. In this case, the server 604 receives the patient's measurement data and correlates the patient's measurement data with data of other patients in the patient database 606A. The server 604 may then identify other patients who presented with symptoms or data similar to the patient 616 and determine the disease states, drugs used, and outcomes for the other patients. Based on the data from the other patients, the server 604 may identify the most common disease states and / or drugs used that produced the most favorable outcomes, and provide these results to the clinical portal 614 for consideration by the medical practitioner 618.

[0099] As shown in FIG. 6, the database 606 includes a set of four databases, including a patient database 606A, a disease database 606B, a treatment plan database 606C, and a model database 606D. These databases store individual data regarding patients and their data, diseases, drugs, dosage schedules, and computational models. In particular, the patient database 606A stores measurements obtained by a medical practitioner 618 or symptoms observed by a medical practitioner 618, including physiological parameter data and concentration data. The disease database 606B stores data regarding various diseases and possible symptoms often exhibited by patients infected with the diseases. The treatment plan database 606C stores data regarding possible treatment plans, including drugs and dosage schedules, for a set of patients. The set of patients may include a population with different characteristics, such as weight, height, age, sex, and race.

[0100] The model database 606D stores data on a set of computational models that can be used to describe pharmacokinetic (PK), pharmacodynamic (PD), or both PK and PD changes to the body. Any suitable mathematical model may be stored in the model database 606D, such as in the form of a compiled library module. In particular, a suitable mathematical model is a mathematical function (or set of functions) that describes the relationship between a dosing regimen and observed patient exposure and / or observed patient response (collectively "response") for a specific drug. Thus, the mathematical model describes the response profile for a population of patients. In general, developing a mathematical model involves developing a mathematical function or equation that defines a curve that best "fits" or describes the observed clinical data. A typical model also describes the expected impact of specific patient characteristics on the response, as well as quantifying the amount of unaccounted for variability that cannot be accounted for by patient characteristics alone. In such models, patient characteristics are reflected as patient factor covariates in the mathematical model. Thus, mathematical models are typically mathematical functions that account for underlying clinical data and associated variability seen in patient populations. These mathematical functions include terms that account for the variation of individual patients from the "average" or typical patient, allowing the models to explain or predict various outcomes for a given dosage, making the models not only mathematical functions but also statistical functions, although the models and functions are referred to herein in a general and non-limiting manner as "mathematical" models and functions.

[0101] In particular, the output of the model corresponds to a dosing regimen or schedule that achieves optimal target levels for the physiological parameters of the patient 616. The model provides optimal target levels as recommendations that are specifically designed for the patient 616, verifying that the optimal target levels are expected to produce an effective and therapeutic response in the patient 616. In the example shown, the concentration data corresponds to the concentration of the drug in the patient's blood. In some implementations, the drug may not be known, for example, the drug may be any drug in a class of drugs. The physiological parameter data may correspond to any number of measurements from the patient. For example, when the drug is infliximab, it may be desirable to measure (and use the model to predict) the drug concentration and other measurable units (which may be predicted by the model), such as C-reactive protein, endoscopic disease severity, and fecal calprotectin. Each measurable (e.g., drug concentration, C-reactive protein, endoscopic disease severity, and fecal calprotectin) may be associated with one or more models, such as a PK or PD model. The interplay between PK and PD models may be particularly important for drugs such as infliximab, where patients with more severe disease clear the drug faster (modeled by a higher clearance from the PK model, as explained in detail below). One goal of the drug infliximab may be to normalize C-reactive protein levels, reduce fecal calprotectin levels, and achieve endoscopic remission.

[0102] In one example, the medical practitioner 618 may assess the likelihood that the patient 616 will exhibit a therapeutic response to a particular drug and dosing regimen. In particular, this likelihood may be low when several dosing regimens of the same drug have been administered to the patient, but no measurable response from the patient is detected. In this case, the medical practitioner 618 may determine that the patient is unlikely to respond to further adjustments to the dosage, and other drugs may be considered. Also, a confidence interval may be assessed regarding the predictive model results and the body's predicted response to the presence of the drug. As more data is collected from the patient 616, the confidence interval becomes narrower, indicating more reliable results and recommendations. The systems and methods described herein may identify individualized target levels (e.g., target trough levels) and provide individualized dosing recommendations based on the individualized target levels.

[0103] In many cases, the medical personnel 618 may be members or employees of the medical center. The same patient 616 may meet with multiple members of the same medical center in various roles. In this case, the clinical portal 614 may be configured to run on multiple user devices. The medical center may have its own records for a particular patient. In some implementations, the present disclosure provides an interface between the computational models described herein and the medical center's records. For example, any medical personnel 618, such as a doctor or nurse, may be required to enter credentials (such as a username and password) or scan an employee badge via the user interface 612 to log into the system provided by the clinical portal 614. Once logged in, each medical personnel 618 may have a corresponding set of patient records that the medical personnel is allowed to access. In some implementations, the patient 616 interacts with the clinical portal 614, which may have a patient-specific page or area for interaction with the patient 616. For example, the clinical portal 614 may be configured to monitor the patient's treatment schedule and send appointments and reminders to the patient 616. Also, one or more devices (such as smart mobile devices or sensors) may be used to monitor the patient's ongoing physiological data and report the physiological data to the clinical portal 614 or directly to the server 604 via the network 602. The physiological data is then compared to expectations and deviations from expectations are flagged. Such intermittent monitoring of patient data allows for early detection of possible deviations from expectations in the patient's response to medication and may indicate the need for modification of dosing recommendations.

[0104] As described herein, measurements from the patient 616 provided into the computational model may be determined from a medical practitioner 618, directly from a device monitoring the patient 616, or a combination of both. As the computational model predicts the time progression of diseases and drugs and their effects on the body, these measurements may be used to update the model parameters so that the treatment plan (provided by the model) is refined and corrected to take into account the patient's specific data. In some implementations, it is desirable to separate the patient's personal information from the patient's measurement data required to run the computational model. In particular, the patient's personal information may be protected health information (PHI), and access to an individual's PHI should be limited to authorized users. One way to protect the patient's PHI is to assign each patient an anonymization code when the patient is registered with the server 604. The code may be entered manually by the medical practitioner 618 via the clinical portal 614 or may be entered using an automated but secure process. The server 604 may only be able to identify each patient according to the anonymization code and may not have access to the patient's PHI. In particular, the clinical portal 614 and the server 604 may exchange data regarding the patient 616 without identifying the patient 616 or exposing the patient's PHI. In some implementations, the clinical portal 614 is configured to communicate with the pharmacy portal 624 via the network 602. In particular, after a medication regimen is selected to be administered to the patient 616, the healthcare provider 618 may provide an indication of the selected medication regimen to the clinical portal 614 for transmission of the selected medication regimen to the pharmacy portal 624. In response to receiving the medication regimen, the pharmacy portal 624 may display the medication regimen and an identifier of the healthcare provider 618 via a user interface 622 that interacts with a pharmacist 628 to fulfill the order.

[0105] 6, the server 604 is a device (or set of devices) that is remote from the clinical portal 614. Depending on the computing capabilities of the device housing the clinical portal 614, the clinical portal 614 may simply be an interface that primarily transfers data between the medical practitioner 618 and the server 604. Alternatively, the clinical portal 614 may be configured to locally perform any or all of the steps described as being performed by the server 604, including, but not limited to, receiving patient symptom and measurement data, accessing any of the databases 606, invoking one or more computational models, and providing recommendations regarding dosage schedules based on the patient's specific symptom and measurement data. Also, although FIG. 6 depicts the patient database 606A, disease database 606B, treatment plan database 606C, and model database 606D as being entities separate from the server 604, clinical portal 614, or pharmacy portal 624, those skilled in the art will understand that any or all of the databases 606 may be stored locally on any of the devices or portals described herein without departing from the scope of the present disclosure.

[0106] 7A depicts an exemplary display screen (e.g., user interface 612 of FIG. 6) displaying predicted concentration-time profiles for four different dosing regimens. In particular, each dosing regimen has a corresponding dosing interval (5 weeks, 4 weeks, 3 weeks, and 2 weeks) with decreasing dosage as the dosing interval decreases (as shown by the height of the second peak in each predicted concentration-time profile). In this case, the user has selected to plot IFX concentration versus time and to allow a computational model to be run to identify a recommended dosing regimen and maintain IFX concentration above the critical trough value.

[0107] In FIG. 7B, a user has selected (e.g., on user interface 612 of FIG. 6) to display the results from the plot shown in FIG. 7A in a table format. In particular, the exemplary display screen of FIG. 7B lists the last dosing date (2 January 2015), various dosing intervals, the trough date (corresponding to the first date after the dosing date when the predicted concentration-time profile drops to or is at the clinical trough value), the suggested dosage (in mg), the normalized suggested dosage (in mg / kg), the number of vials used per dose, and the target concentration (in ng / ml). As shown in FIG. 7B, a set of suggested dosing schedules is shown, with the dosing schedules having different dosing intervals ranging from 2 to 8 weeks. Although some of the dosing schedules with longer dosing intervals (6 to 8 weeks) are not recommended, four dosing schedules (with dosing intervals of 2 to 5 weeks) are suggested with increasing dosages as the dosing interval increases. In particular, when interacting with the display screen of FIG. 7B, the medical practitioner 618 may select a dosing regimen based on a specific goal. For example, if it is desired to administer doses to the patient 616 less frequently, a longer dosing interval (e.g., 5 weeks) may be selected. Alternatively, it may be desirable to use as many vials as possible, since the patient is often charged for a full vial even when a partial vial is used. In this case, a 4-week dosing regimen may be selected, since 4.9 vials are used per dose, leading to little drug waste (e.g., only 0.1 vials per dose). Alternatively, a shorter dosing interval (e.g., 2 weeks) may be selected, if it is desirable to administer doses to the patient 616 more frequently, or to charge the patient 616 for only two vials at a time.

[0108] While various illustrative implementations have been described, it should be understood that the foregoing description is illustrative only and does not limit the scope of the invention. Although several examples have been provided in this disclosure, it should be understood that the disclosed systems, components, and manufacturing methods may be embodied in many other specific forms without departing from the scope of the disclosure.

[0109] The disclosed embodiments can be implemented in combination or subcombination with one or more other features described herein. Various devices, systems, and methods can be implemented based on the present disclosure and still fall within the scope of the invention. Also, various features described or illustrated above may be combined or integrated in other systems, or certain features may be omitted or not implemented.

[0110] While various implementations of the present disclosure have been shown and described herein, it will be apparent to those skilled in the art that such implementations are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the present disclosure. It will be understood that various alternatives to the implementations of the present disclosure described herein may be employed in practicing the present disclosure.

[0111] All references cited herein are incorporated by reference in their entirety and made part of this application.

Claims

[Claim 1] The invention described in this specification.

Images