Estimating personalized drug responses from real world evidence

Patent metadata

Field	Value
Publication number	US-11183308-B2
Application number	US-201816209196-A
Country	US
Kind code	B2
Filing date	Dec 4, 2018
Priority date	Dec 27, 2017
Publication date	Nov 23, 2021
Grant date	Nov 23, 2021

Abstract

A mechanism is provided in a data processing system comprising at least one processor and at least one memory, the at least one memory comprising instructions executed by the at least one processor to cause the at least one processor to implement a drug response estimation engine. The drug response estimation engine receives real-world evidence for a plurality of patients. A patient similarity network builder component executing within the drug response estimation engine builds a patient similarity network. A regression analysis component executing within the drug response estimation engine builds a network localized regression analysis approach. A patient clustering component executing within the drug response estimation engine groups patients based on demographics and comorbidities to form a plurality of patient clusters. The drug response estimation engine estimates drug responses for a given patient based on the patient similarity network, the network localized regression analysis approach, and the plurality of patient clusters. The drug response estimation engine outputs the drug responses for the given patient.

First claim

What is claimed is:

1. A method, in a data processing system comprising at least one processor and at least one memory, the at least one memory comprising instructions executed by the at least one processor to cause the at least one processor to implement a drug response estimation engine, wherein the drug response estimation engine operates to:
receive, by the drug response estimation engine, real-world evidence for a plurality of patients;
build a drug response matrix, W, representing an individual response of each patient on each drug, vectorized into a column vector, w, a column matrix of laboratory test measurements, y, and a block diagonal matrix, Z, of drug exposures based on information in the real-world evidence;
build, by the regression analysis component, a statistical model using a network localized regression analysis approach, wherein building the statistical model comprises using a difference matrix to determine a difference in test measurements over time and a difference in drug exposures over time based on the column matrix of laboratory test measurements, the block diagonal matrix of drug exposures, and the drug response matrix, wherein the statistical model learns parameters that predict drug responses for each patient and for each time point of the real-world evidence;
group, by a patient clustering component executing within the drug response estimation engine, patients based on demographics and comorbidities to form a plurality of patient groups, wherein the patient clustering component groups the patients using the learned parameters that predict drug responses such that each patient group within the plurality of patient groups represents patients that have similar drug responses and background information;
estimate, by the drug response estimation engine, drug responses for a given patient based on the statistical model and the plurality of patient groups; and
output, by the drug response estimation engine, the drug responses for the given patient.

2. The method of claim 1 , wherein the real-world evidence comprises at least one of patient demographics, lab tests, diagnoses, or medication history.

3. The method of claim 1 , wherein estimating drug responses comprises estimating significant drug responses for indication and adverse drug reactions.

4. The method of claim 1 , wherein estimating drug responses comprises estimating specific drug responses in each patient group within the plurality of patient groups.

5. The method of claim 4 , wherein each patient group comprises patients with similar drug responses for a particular laboratory test measurement.

6. The method of claim 1 , wherein estimating drug responses comprises recording average drug responses for each group within the plurality of patient groups.

7. The method of claim 1 , wherein estimating drug responses comprises generating predicted associations including at least one of drug responses for each patient for each specific time, potential hypotheses about new therapeutic effects and adverse drug reactions, or associations between the given patients' characteristics and demographics with the obtained drug responses.

8. The method of claim 1 , wherein the statistical model optimizes the following objective function:

arg
⁢

min
W

⁢



Dy
-
DZw



2
2

+

λ
1

⁢

∑

i
=
1

N

⁢



w
i



1
2

+

λ
2

⁢

∑

i
>

i
′

N

⁢

∑

i
′

=
1

N
-
1

⁢

r

ii
′

⁢



w
i

-

w

i
′



2

wherein D is a difference matrix, y is a column matrix of laboratory test measurements, Z is a block diagonal matrix of drug exposures, w is a drug response vector, wherein λ 1 is a hyper-parameter that controls an exclusive lasso penalty and λ 2 is a hyper-parameter that controls a network lasso penalty, wherein w i is a vector that denotes a drug response for each individual patient i, and wherein r ii′ represents a patient similarity between the i th and i′ th patients.

9. A computer program product comprising a non-transitory computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on at least one processor of a data processing system, causes the data processing system to implement a drug response estimation engine, wherein the computer readable program causes the data processing system to:
receive, by the drug response estimation engine, real-world evidence for a plurality of patients;
build a drug response matrix, W, representing an individual response of each patient on each drug, vectorized into a column vector, w, a column matrix of laboratory test measurements, y, and a block diagonal matrix, Z, of drug exposures based on information in the real-world evidence;
build, by the regression analysis component, a

Assignees

• IBM

Inventors

• Dey Sanjoy
• Zhang Ping

Classifications

• G16H70/40 Primary

  relating to drugs, e.g. their side effects or intended usage · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G16H20/10

  relating to drugs or medications, e.g. for ensuring correct administration to patients · CPC title

• G06F16/93

  Document management systems · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title