Advancing psychiatric pharmacogenomics using drug development paradigms

Machine Learning and Pharmacogenomics at the Time of Precision Psychiatry

Traditional medicine and biomedical sciences are reaching a turning point because of the constantly growing impact and volume of Big Data. Machine Learning (ML) techniques and related algorithms play a central role as diagnostic, prognostic, and decision-making tools in this field. Another promising area becoming part of everyday clinical practice is personalized therapy and pharmacogenomics. Applying ML to pharmacogenomics opens new frontiers to tailored therapeutical strategies to help clinicians choose drugs with the best response and fewer side effects, operating with genetic information and combining it with the clinical profile. This systematic review aims to draw up the state-of-the-art ML applied to pharmacogenomics in psychiatry. Our research yielded fourteen papers; most were published in the last three years. The sample comprises 9,180 patients diagnosed with mood disorders, psychoses, or autism spectrum disorders. Prediction of drug response and prediction of side effects are the most frequently considered domains with the supervised ML technique, which first requires training and then testing. The random forest is the most used algorithm; it comprises several decision trees, reduces the training set's overfitting, and makes precise predictions. ML proved effective and reliable, especially when genetic and biodemographic information were integrated into the algorithm. Even though ML and pharmacogenomics are not part of everyday clinical practice yet, they will gain a unique role in the next future in improving personalized treatments in psychiatry.

Genetic Variations Associated with Long-Term Treatment Response in Bipolar Depression

Several pharmacogenetic-based decision support tools for psychoactive medication selection are available. However, the scientific evidence of the gene-drug pairs analyzed is mainly based on pharmacogenetic studies in patients with major depression or schizophrenia, and their clinical utility is mostly assessed in major depression. This study aimed at evaluating the impact of individual genes, with pharmacogenetic relevance in other psychiatric conditions, in the response to treatment in bipolar depression. Seventy-six patients diagnosed with bipolar disorder and an index major depressive episode were included in an observational retrospective study. Sociodemographic and clinical data were collected, and all patients were genotyped using a commercial multigene pharmacogenomic-based tool (Neuropharmagen^®, AB-Biotics S.A., Barcelona, Spain). Multiple linear regression was used to identify pharmacogenetic and clinical predictors of efficacy and tolerability of medications. The pharmacogenetic variables response to serotonin-norepinephrine reuptake inhibitors (SNRIs) (ABCB1) and reduced metabolism of quetiapine (CYP3A4) predicted patient response to these medications, respectively. ABCB1 was also linked to the tolerability of SNRIs. An mTOR-related multigenic predictor was also associated with a lower number of adverse effects when including switch and autolytical ideation. Our results suggest that the predictors identified could be useful to guide the pharmacological treatment in bipolar disorder. Additional clinical studies are necessary to confirm these findings.

Pharmacogenomic Characterization in Bipolar Spectrum Disorders

The holistic approach of personalized medicine, merging clinical and molecular characteristics to tailor the diagnostic and therapeutic path to each individual, is steadily spreading in clinical practice. Psychiatric disorders represent one of the most difficult diagnostic challenges, given their frequent mixed nature and intrinsic variability, as in bipolar disorders and depression. Patients misdiagnosed as depressed are often initially prescribed serotonergic antidepressants, a treatment that can exacerbate a previously unrecognized bipolar condition. Thanks to the use of the patient's genomic profile, it is possible to recognize such risk and at the same time characterize specific genetic assets specifically associated with bipolar spectrum disorder, as well as with the individual response to the various therapeutic options. This provides the basis for molecular diagnosis and the definition of pharmacogenomic profiles, thus guiding therapeutic choices and allowing a safer and more effective use of psychotropic drugs. Here, we report the pharmacogenomics state of the art in bipolar disorders and suggest an algorithm for therapeutic regimen choice.

Pharmacogenetics in the clinical analysis laboratory: clinical practice, research, and drug development pipeline

Introduction: Over the last decade, the spread of next-generation sequencing technology along with the rising cost in health management in national health systems has led to widespread use/abuse of pharmacogenetic tests (PGx) in the practice of many clinical disciplines. However, given their clinical significance, it is important to standardize these tests for having an interaction with the clinical analysis laboratory (CAL), in which a PGx service can meet these requirements. Areas covered: A diagnostic test must meet the criteria of reproducibility and validity for its utility in the clinical routine. This present review mainly describes the utility of introducing PGx tests in the CAL routine to produce correct results useful for setting up personalized drug treatments. Expert opinion: With a PGx service, CALs can provide the right tool to help clinicians to make better choices about different categories of drugs and their dosage and to manage the economic impact both in hospital-based settings and in National Health Services, throughout electronic health records. Advances in PGx also allow a new approach for pharmaceutical companies in order to improve drug development and clinical trials. As a result, CALs can achieve a powerful source of epidemiological, clinical, and research findings from PGx tests.

Deep learning in pharmacogenomics: from gene regulation to patient stratification

This Perspective provides examples of current and future applications of deep learning in pharmacogenomics, including: identification of novel regulatory variants located in noncoding domains of the genome and their function as applied to pharmacoepigenomics; patient stratification from medical records; and the mechanistic prediction of drug response, targets and their interactions. Deep learning encapsulates a family of machine learning algorithms that has transformed many important subfields of artificial intelligence over the last decade, and has demonstrated breakthrough performance improvements on a wide range of tasks in biomedicine. We anticipate that in the future, deep learning will be widely used to predict personalized drug response and optimize medication selection and dosing, using knowledge extracted from large and complex molecular, epidemiological, clinical and demographic datasets.