Identifying patients for intensive hyperglycemia management
Systems and methods are provided for assigning a treatment to a patient is provided. A set of genetic data representing a patient. A polygenic score representing the likelihood that a patient will benefit from intensive glycemia treatment is generated from the set of genetic data. A parameter representing a response of the patient to intensive glycemia treatment is assigned according to the polygenic score.
1. A method for assigning a treatment to a patient, the method comprising:
receiving a set of genetic data representing a patient at a system comprising a processor and a computer readable medium storing machine-readable instructions executable by the processor via a network interface;
generating a polygenic score representing the likelihood that a patient will benefit from intensive glycemia treatment from the set of genetic data at a feature extractor implemented via the machine-readable instructions; and
assigning a parameter representing a response of the patient to intensive glycemia treatment according to the polygenic score at a predictive model implemented via the machine-readable instructions;
wherein the parameter representing the response of the patient to intensive glycemia treatment is a categorical parameter representing one of a plurality of treatment classes, each representing clusters of trajectories of HbA1c values for patients in response to intensive glycemia treatment determined via an unsupervised clustering process.
2. The method of claim 1 , wherein assigning the parameter representing the response of the patient to intensive glycemia treatment comprises assigning the parameter representing the response of the patient to intensive glycemia treatment according to the polygenic score and a set of clinical parameters representing the patient.
3. The method of claim 2 , wherein the set of clinical parameters includes a clinical parameter representing a blood pressure of the patient.
4. The method of claim 1 , wherein the set of genetic data includes a parameter representing a single nucleotide polymorphism in one of Mas1 proto-oncogene (MAS1), Neural EGFL Like 1 (NELL1), and Supervillin (SVIL), the polygenic score being generated from at least the parameter representing the single nucleotide polymorphism.
5. The method of claim 1 , wherein assigning the parameter representing the response of the patient to intensive glycemia treatment according to the polygenic score and a set of clinical parameters representing the patient comprises providing each of the polygenic score and the set of clinical parameters to a regression model.
6. The method of claim 1 , further comprising treating the patient with one of metformin, sulfonylureas, thiazolidinediones, and insulin products if the parameter representing the response of the patient to intensive glycemia treatment indicates that the patient will respond well to intensive glycemia treatment.
7. The method of claim 1 , further comprising treating the patient with one of GLP-1 receptor agonists and SGLT-2 inhibitors if the parameter representing the response of the patient to intensive glycemia treatment indicates that the patient will not respond well to intensive glycemia treatment.
8. The method of claim 1 , wherein the set of genetic data includes a parameter representing one of a set of single nucleotide polymorphisms including rs220721, rs1793004, and rs1270874, the polygenic score being generated from at least the parameter representing the one of the set of single nucleotide polymorphisms.
9. The method of claim 1 , wherein the set of generic data includes parameters representing the presence or absence of each of a plurality of single nucleotide polymorphisms in the patient and each of the plurality of single nucleotide polymorphisms in the set of genetic data having an associated weight, and generating the polygenic score comprises computing the sum of the weights associated with the single nucleotide polymorphisms present in the patient.
10. A system comprising:
a processor; and
a computer readable medium storing executable instructions for assigning a treatment to a patient, the executable instructions comprising:
a network interface that receives a set of genetic data representing a patient, the set of genetic data including a parameter representing one of a set of single nucleotide polymorphisms including rs220721, rs1793004, and rs1270874;
a feature extractor that generates a polygenic score representing the likelihood that a patient belongs to a treatment class likely to benefit from intensive glycemia treatment given the set of genetic data, the feature extractor generating the polygenic score from at least the parameter representing the single nucleotide polymorphism; and
a predictive model that assigns a parameter to the patient representing the response of the patient to intensive glycemia treatment as a weighted sum of the polygenic score and a set of clinical parameters representing the patient.
11. The system of claim 10 , wherein the set of clinical parameters includes a clinical parameter representing a measurement of blood concentration of thyroid stimulating hormone for the patient.
12. The system of claim 10 , wherein the set of genetic data includes a parameter representing a single nucleotide polymorphism in one of Mas1 proto-oncogene (MAS1), Neural EGFL Like 1 (NELL1), and Supervillin (SVIL), the feature extractor generating the polygenic score.
13. A method for assigning a treatment to a patient, the method comprising:
receiving a set of genetic data representing a patient at a system comprising a processor and a computer readable medium storing machine-readable instructions executable by the processor via a network interface;
generating a polygenic score representing the likelihood that a patient belongs to a treatment class likely to benefit from intensive glycemia treatment given the set of genetic data at a feature extractor implemented via the machine-readable instructions;
assigning the patient to one of a plurality of treatment classes according to the polygenic score and a set of clinical parameters representing the patient at a predictive model implemented via the machine-readable instructions, the plurality of treatment classes including the treatment class likely to benefit from intensive glycemia treatment;
providing intensive glycemia treatment to the patient if the patient is assigned to the treatment class likely to benefit from intensive glycemia treatment; and
treating the patient with one of GLP-1 receptor agonists and SGLT-2 inhibitors if the patient is not assigned to the treatment class likely to benefit from intensive glycemia treatment.
14. The method of claim 13 , wherein each of the plurality of treatment classes represent clusters of trajectories of HbA1c values for patients in response to intensive glycemia treatment determined via an unsupervised clustering process.
15. The method of claim 13 , wherein the set of clinical parameters includes at least a clinical parameter representing alcohol consumption by the patient.
16. The method of claim 13 , wherein the set of genetic data includes a parameter representing a single nucleotide polymorphism in one of Mas1 proto-oncogene (MAS1), Neural EGFL Like 1 (NELL1), and Supervillin (SVIL), the polygenic score being generated from at least the parameter representing the single nucleotide polymorphism.
17. The method of claim 13 , further comprising treating the patient with one of GLP-1 receptor agonists and SGLT-2 inhibitors if the patient is not assigned to the treatment class likely to benefit from intensive glycemia treatment.
18. A system comprising:
a processor; and
a computer readable medium storing executable instructions for assigning a treatment to a patient, the executable instructions comprising:
a network interface that receives a set of genetic data representing a patient, the set of generic data including parameters representing the presence or absence of each of a plurality of single nucleotide polymorphisms in the patient;
a feature extractor that generates a polygenic score representing the likelihood that a patient belongs to a treatment class likely to benefit from intensive glycemia treatment given the set of genetic data, the feature extractor computing the sum of weights associated with the single nucleotide polymorphisms present in the patient to provide the polygenic score; and
a predictive model, trained on training data representing patients for whom an associated treatment class is known, that assigns a parameter to the patient representing the response of the patient to intensive glycemia treatment according to the polygenic score and a set of clinical parameters representing the patient.
19. A system comprising:
a processor; and
a computer readable medium storing executable instructions for assigning a treatment to a patient, the executable instructions comprising:
a network interface that receives a set of genetic data representing a patient, the set of genetic data including a parameter representing a single nucleotide polymorphism in one of Mas1 proto-oncogene (MAS1), Neural EGFL Like 1 (NELL1), and Supervillin (SVIL);
a feature extractor that generates a polygenic score representing the likelihood that a patient belongs to a treatment class likely to benefit from intensive glycemia treatment given the set of genetic data, the feature extractor generating the polygenic score from at least the parameter representing the single nucleotide polymorphism; and
a predictive model, trained on training data representing patients for whom an associated treatment class is known, that assigns a parameter to the patient representing the response of the patient to intensive glycemia treatment according to the polygenic score and a set of clinical parameters representing the patient.