System and method for providing patient-specific dosing as a function of mathematical models updated to account for an observed patient response

Abstract

A system and method for predicting, proposing and/or evaluating suitable medication dosing regimens for a specific individual as a function of individual-specific characteristics and observed responses of the specific individual. Mathematical models of observed patient responses are used in determining an initial dose. The system and method use the patient's observed response to the initial dose to refine the model for use to forecast expected responses to proposed dosing regimens more accurately for a specific patient. More specifically, the system and method uses Bayesian averaging, Bayesian updating and Bayesian forecasting techniques to develop patient-specific dosing regimens as a function of not only generic mathematical models and patient-specific characteristics accounted for in the models as covariate patient factors, but also observed patient-specific responses that are not accounted for within the models themselves, and that reflect variability that distinguishes the specific patient from the typical patient reflected by the model.

1 - 51 . (canceled) 
     
     
         52 . A method for preparing a patient-specific drug dosing regimen for treating a patient with a medication using a computerized medication dosing regimen recommendation system, the system comprising a processor, the method comprising:
 forecasting, based on a mathematical model, a plurality of predicted medication concentration time profiles for a plurality of proposed dosing regimens, each predicted medication concentration time profile being indicative of a patient response to the medication corresponding to a proposed dosing regimen of the plurality of proposed dosing regimens;   selecting, from the plurality of proposed dosing regimens, a dosing regimen for the medication to reach a treatment objective for the patient;   outputting the selected dosing regimen;   receiving data reflecting a response of the patient to the selected dosing regimen;   updating the mathematical model, based on the received data reflecting the response of the patient to the selected dosing regimen, to generate an updated mathematical model;   calculating, based on the updated mathematical model, an updated dosing regimen for the medication to reach the treatment objective; and   outputting the updated dosing regimen for the medication for administration to the patient.   
     
     
         53 . The method of  claim 52 , wherein the plurality of predicted medication concentration time profiles are forecasted by processing the mathematical model using a Bayesian forecasting technique. 
     
     
         54 . The method of  claim 52 , wherein the treatment objective is a target blood level concentration of the medication in the patient. 
     
     
         55 . The method of  claim 54 , wherein the medication is infliximab. 
     
     
         56 . The method of  claim 52 , further comprising:
 automatedly selecting the plurality of proposed dosing regimens according to a predefined logic.   
     
     
         57 . The method of  claim 56 , wherein automatedly selecting the plurality of proposed dosing regimens according to a predefined logic comprises:
 selecting a next proposed dosing regimen as a function of a predication medication concentration time profile of a first proposed dosing regimen.   
     
     
         58 . The method of  claim 57 , wherein the predefined logic provides for selection of the plurality of proposed dosing regimens to provide a corresponding predicted medication concentration time profile best meeting the treatment objective. 
     
     
         59 . The method of  claim 57 , wherein the predefined logic provides for selection of the plurality of proposed dosing regimens to find a minimum of an objective function. 
     
     
         60 . The method of  claim 57 , wherein the predefined logic provides for selection of the plurality of proposed dosing regimens to find a global minimum of an objective function. 
     
     
         61 . The method of  claim 52 , further comprising:
 outputting a plurality of alternative dosing regimens.   
     
     
         62 . The method of  claim 52 , wherein updating the mathematical model comprises adjusting a set of parameters in the selected model to be conditional to the received data. 
     
     
         63 . The method of  claim 62 , wherein updating the mathematical model comprises using a Bayesian updating technique. 
     
     
         64 . The method of  claim 63 , wherein the Bayesian updating technique comprises minimizing an objective function describing a difference between an expectation of the mathematical model and the received data. 
     
     
         65 . The method of  claim 64 , wherein the Bayesian updating technique further comprises using a random function to interject variation into the objective function in order to find a global minimum. 
     
     
         66 . The method of  claim 52 , further comprising:
 receiving patient characteristic data reflecting a set of characteristics of the patient; and   selecting the mathematical model from a plurality of mathematical models to match the set of characteristics of the patient with covariate patient factors of the mathematical model.   
     
     
         67 . The method of  claim 66 , wherein each mathematical model of the plurality of mathematical models describes response profiles for a population of patients treated with the medication and having a set of covariate patient factors, and wherein the response profiles for the population of patients are not specific to any particular patient. 
     
     
         68 . The method of  claim 67 , wherein the characteristics of the patient include at least one of: disease stage, disease status, prior therapy, concomitant diseases, demographic information, laboratory test result information, race, sex, age, and weight. 
     
     
         69 . The method of  claim 66 , further comprising:
 updating the patient characteristic data to reflect at least one updated patient characteristic.   
     
     
         70 . The method of  claim 66 , wherein the mathematical model comprises at least one of: a pharmacokinetic (PK) model, a pharmacodynamic (PD) model, and an exposure/response model. 
     
     
         71 . The method of  claim 52 , wherein the method is executed as a web service via a network computing model, and wherein the network computing model is configured to connect to a client device.