Methods and systems for evaluation of risk of substance use disorders

Abstract

Provided here are systems and methods for predicting risk of a substance use disorder and for providing decision support to healthcare professionals to implement a treatment regimen recommendation and mitigate any potential risk of a substance use disorder.

1 . A non-transitory machine-readable storage medium storing processor-executable instructions that, when executed by at least one processor, cause the at least one processor to:
 in response to receipt of a sample from a patient, determine if the sample contains a specified set of allelic variants;   in response to a determination that the sample does not include the specified set of allelic variants, transmit a response to a user indicating that a score indicative of risk to a substance use disorder is not available;   in response to a determination that the sample includes the specified set of allelic variants,   determine a first plurality of SNP (single-nucleotide polymorphism) profiles for each of the specified set of allelic variants in the sample; and   determine, using a machine learning ensemble model, the score indicative of risk to a substance use disorder for the patient based on the first plurality of SNP profiles for each of the specified set of allelic variants in the sample, wherein the machine learning ensemble model is trained with inputs comprising a second plurality of SNP profiles associated with the specified set of allelic variants from a first plurality of test subjects and a third plurality of SNP profiles associated with the specified set of allelic variants from a second plurality of test subjects and outputs comprising a first plurality of substance use indicators specifying that the first plurality of test subjects have been diagnosed with the substance use disorder and a second plurality of substance use indicators specifying that the second plurality of test subjects have not been diagnosed with the substance use disorder.   
     
     
         2 . The non-transitory machine-readable storage medium of  claim 1 , wherein clinical data of the patient is provided along with the sample. 
     
     
         3 . The non-transitory machine-readable storage medium of  claim 1 , wherein the score indicative of risk to the substance use disorder for the patient from the machine learning ensemble model is a value between 0 and 1. 
     
     
         4 . The non-transitory machine-readable storage medium of  claim 3 , further comprising instructions that cause the at least one processor to:
 in response to the score based on the first plurality of SNP profiles for each of the specified set of allelic variants in the sample being greater than 0.33, return an output indicating a predisposition to develop the substance use disorder.   
     
     
         5 . The non-transitory machine-readable storage medium of  claim 3 , further comprising instructions that cause the at least one processor to:
 in response to the score based on the first plurality of SNP profiles for each of the specified set of allelic variants in the sample is less than or equal to  0 . 33 , return an output indicating a lower likelihood to develop the substance use disorder.   
     
     
         6 . The non-transitory machine-readable storage medium of  claim 1 , wherein the substance use disorder includes one or more of an opioid use disorder, an alcohol use disorder, a cannabinoid use disorder, and a cocaine use disorder. 
     
     
         7 . A method of determining a subject's risk for developing a substance use disorder, the method comprising:
 analyzing a sample of the subject to obtain one or more SNP profiles;   in response to the one or more SNP profiles each including a set of specified allelic variants:
 determining, via a processor of a computing device and a machine learning ensemble model stored in a machine-readable storage medium of the computing device, a score indicating a risk of the subject for developing a substance use disorder based on the one or more SNP profiles of each of the set of specified allelic variants; and 
 in response to the score indicating a high risk for the subject to develop the substance use disorder:
 determining, via the processor, a treatment regimen recommendation for the subject based on the score, and 
 transmitting, via the processor, the score and the treatment regimen recommendation to a user device. 
 
   
     
     
         8 . The method of  claim 7 ,
 wherein the score indicating the subject's risk for developing the substance use disorder is a value from the machine learning ensemble model between 0 and 1,   wherein the machine learning ensemble model includes a pre-determined threshold,   wherein the score being greater than the pre-determined threshold indicates the subject has a higher risk for developing the substance use disorder, and   wherein the score being less than or equal to than the pre-determined threshold indicates the subject has a lower risk for developing the substance use disorder.   
     
     
         9 . The method of  claim 7 , wherein the computing device includes a SNP analyzer to analyze the sample of the subject to obtain the one or more SNP profiles of each of a set of specified allelic variants. 
     
     
         10 . The method of  claim 7 , wherein the method further comprises:
 in response to a determination of the score, generating, via the processor, a report, the report including the treatment regimen recommendation and the score.   
     
     
         11 . The method of  claim 10 , wherein the method further comprises:
 in response to generation of the report, transmitting, via the processor, the report to a state prescription drug monitoring program (PDMP) database.   
     
     
         12 . The method of  claim 7 , wherein the method further comprises:
 prior to receipt of the sample, determining, via the processor, whether a prior determined score indicating the subject's risk for developing a substance use disorder is available for the subject; and   in response to a determination that the prior determined score is not available for the subject, transmitting, via the processor of the computing device, a request for another sample of the subject;   in response to a determination that the prior determined score is available for the subject, determining, via the processor, a treatment regimen recommendation based on the prior determined score.   
     
     
         13 . The method of  claim 12 , wherein the prior determined score is stored in a patient record database containing electronic health records of the subject. 
     
     
         14 . The method of  claim 7 , wherein the set of specified allelic variants includes one or more of allelic variants of genes: serotonin 2A receptor, galanin, ATP binding cassette transporter 1, catechol-O-methyltransferase, dopamine transporter, dopamine D2 receptor, dopamine D1 receptor, methylene tetrahydrofolate reductase, dopamine beta hydroxylase, delta opioid receptor, a first mu opioid receptor (OPRM1), dopamine D4 receptor, gamma-aminobutyric acid, kappa opioid receptor, and a second mu opioid receptor (MUOR). 
     
     
         15 . The method of  claim 7 , wherein the set of specified allelic variants includes allelic variants of genes: serotonin 2A receptor, galanin, ATP binding cassette transporter 1, catechol-O-methyltransferase, dopamine transporter, dopamine D2 receptor, dopamine D1 receptor, methylene tetrahydrofolate reductase, dopamine beta hydroxylase, delta opioid receptor, a first mu opioid receptor (OPRM1), dopamine D4 receptor, gamma-aminobutyric acid, kappa opioid receptor, and a second mu opioid receptor (MUOR). 
     
     
         16 . The method of  claim 7 , wherein a determination of the score is further based on a subject's clinical data. 
     
     
         17 . A non-transitory machine-readable storage medium encoded with instructions executable by a processing resource, the non-transitory machine-readable storage medium comprising instructions to:
 store one or more sets of data, each of the one or more sets of data including one or more SNP profiles for each of a specified set of allelic variants, each of the one or more SNP profiles associated with a subject and value indicating whether the subject is diagnosed with substance use disorder, each of the set of specified allelic variants includes one or more of allelic variants of genes: serotonin 2A receptor, galanin, ATP binding cassette transporter 1, catechol-O-methyltransferase, dopamine transporter, dopamine D2 receptor, dopamine D1 receptor, methylene tetrahydrofolate reductase, dopamine beta hydroxylase, delta opioid receptor, a first mu opioid receptor (OPRM1), dopamine D4 receptor, gamma-aminobutyric acid, kappa opioid receptor, and a second mu opioid receptor (MUOR);   generate one or more subsets of data based on the one or more sets of data, each subset of data including SNP profiles associated with a same value indicating whether subjects are diagnosed with substance use disorder; and   train a machine learning model with training data defined by the one or more subsets of data, such that the training of the machine learning model provides a substance use disorder predictor.   
     
     
         18 . The non-transitory machine-readable storage medium of  claim 17 , wherein the one or more sets of data include clinical data, the clinical data including age, sex, ethnicity, race, and, if a subject has been exposed to a specific substance, a time since first exposure. 
     
     
         19 . The non-transitory machine-readable storage medium of  claim 17 , wherein each subset of data is pre-processed prior to training the machine learning model, the pre-processing including normalizing SNP profiles for each of the specified set of allelic variants based on importance of an alleles variant on substance use disorder predictability. 
     
     
         20 . The non-transitory machine-readable storage medium of  claim 17  wherein the substance use disorder includes opioid use disorder, alcohol abuse disorder, cannabinoid use disorder, cocaine use disorder, and nicotine use disorder.