Imaging genetics: implications for research on variable antidepressant drug response

Machine Learning and Deep Learning for the Pharmacogenomics of Antidepressant Treatments

A growing body of evidence now proposes that machine learning and deep learning techniques can serve as a vital foundation for the pharmacogenomics of antidepressant treatments in patients with major depressive disorder (MDD). In this review, we focus on the latest developments for pharmacogenomics research using machine learning and deep learning approaches together with neuroimaging and multi-omics data. First, we review relevant pharmacogenomics studies that leverage numerous machine learning and deep learning techniques to determine treatment prediction and potential biomarkers for antidepressant treatments in MDD. In addition, we depict some neuroimaging pharmacogenomics studies that utilize various machine learning approaches to predict antidepressant treatment outcomes in MDD based on the integration of research on pharmacogenomics and neuroimaging. Moreover, we summarize the limitations in regard to the past pharmacogenomics studies of antidepressant treatments in MDD. Finally, we outline a discussion of challenges and directions for future research. In light of latest advancements in neuroimaging and multi-omics, various genomic variants and biomarkers associated with antidepressant treatments in MDD are being identified in pharmacogenomics research by employing machine learning and deep learning algorithms.

Deep Learning with Neuroimaging and Genomics in Alzheimer's Disease

A growing body of evidence currently proposes that deep learning approaches can serve as an essential cornerstone for the diagnosis and prediction of Alzheimer's disease (AD). In light of the latest advancements in neuroimaging and genomics, numerous deep learning models are being exploited to distinguish AD from normal controls and/or to distinguish AD from mild cognitive impairment in recent research studies. In this review, we focus on the latest developments for AD prediction using deep learning techniques in cooperation with the principles of neuroimaging and genomics. First, we narrate various investigations that make use of deep learning algorithms to establish AD prediction using genomics or neuroimaging data. Particularly, we delineate relevant integrative neuroimaging genomics investigations that leverage deep learning methods to forecast AD on the basis of incorporating both neuroimaging and genomics data. Moreover, we outline the limitations as regards to the recent AD investigations of deep learning with neuroimaging and genomics. Finally, we depict a discussion of challenges and directions for future research. The main novelty of this work is that we summarize the major points of these investigations and scrutinize the similarities and differences among these investigations.

Pharmacogenetics and Imaging-Pharmacogenetics of Antidepressant Response: Towards Translational Strategies

Genetic variation underlies both the response to antidepressant treatment and the occurrence of side effects. Over the past two decades, a number of pharmacogenetic variants, among these the SCL6A4, BDNF, FKBP5, GNB3, GRIK4, and ABCB1 genes, have come to the forefront in this regard. However, small effects sizes, mixed results in independent samples, and conflicting meta-analyses results led to inherent difficulties in the field of pharmacogenetics translating these findings into clinical practice. Nearly all antidepressant pharmacogenetic variants have potentially pleiotropic effects in which they are associated with major depressive disorder, intermediate phenotypes involved in emotional processes, and brain areas affected by antidepressant treatment. The purpose of this article is to provide a comprehensive review of the advances made in the field of pharmacogenetics of antidepressant efficacy and side effects, imaging findings of antidepressant response, and the latest results in the expanding field of imaging-pharmacogenetics studies. We suggest there is mounting evidence that genetic factors exert their impact on treatment response by influencing brain structural and functional changes during antidepressant treatment, and combining neuroimaging and genetic methods may be a more powerful way to detect biological mechanisms of response than either method alone. The most promising imaging-pharmacogenetics findings exist for the SCL6A4 gene, with converging associations with antidepressant response, frontolimbic predictors of affective symptoms, and normalization of frontolimbic activity following antidepressant treatment. More research is required before imaging-pharmacogenetics informed personalized medicine can be applied to antidepressant treatment; nevertheless, inroads have been made towards assessing genetic and neuroanatomical liability and potential clinical application.

Allelic variation in 5-HTTLPR and the effects of citalopram on the emotional neural network

BACKGROUND

Selective serotonin reuptake inhibitors (SSRIs), such as citalopram, which selectively block serotonin transporter (5-HTT) activity, are widely used in the treatment of depression and anxiety disorders. Numerous neuroimaging studies have examined the effects of SSRIs on emotional processes. However, there are considerable inter-individual differences in SSRI effect, and a recent meta-analysis further revealed discrepant effects of acute SSRI administration on neural responses to negative emotions in healthy adults.

AIMS

We examined how a variant of the serotonin-transporter polymorphism (5-HTTLPR), which affects the expression and function of 5-HTT, influenced the acute effects of an SSRI (citalopram) on emotion-related brain activity in healthy adults.

METHOD

Combining genetic neuroimaging, pharmacological technique and a psychological paradigm of emotion recognition, we scanned the short/short (s/s) and long/long (l/l) variants of 5-HTTLPR during perception of fearful, happy and neutral facial expressions after the acute administration of an SSRI (i.e. 30 mg citalopram administered orally) or placebo administration.

RESULTS

We found that 5-HTTLPR modulated the acute effects of citalopram on neural responses to negative emotions. Specifically, relative to placebo, citalopram increased amygdala and insula activity in l/l but not s/s homozygotes during perception of fearful faces. Similar analyses of brain activity in response to happy faces did not show any significant effects.

CONCLUSIONS

Our combined pharmacogenetic and functional imaging results provide a neurogenetic mechanism for discrepant acute effects of SSRIs.

A multivariate approach linking reported side effects of clinical antidepressant and antipsychotic trials to in vitro binding affinities

The vast majority of approved antidepressants and antipsychotics exhibit a complex pharmacology. The mechanistic understanding of how these psychotropic medications are related to adverse drug reactions (ADRs) is crucial for the development of novel drug candidates and patient adherence. This study aims to associate in vitro assessed binding affinity profiles (39 compounds, 24 molecular drug targets) and ADRs (n=22) reported in clinical trials of antidepressants and antipsychotics (n>59.000 patients) by the use of robust multivariate statistics. Orthogonal projection to latent structures (O-PLS) regression models with reasonable predictability were found for several frequent ADRs such as nausea, diarrhea, hypotension, dizziness, headache, insomnia, sedation, sleepiness, increased sweating, and weight gain. Results of the present study support many well-known pharmacological principles such as the association of hypotension and dizziness with α1-receptor or sedation with H1-receptor antagonism. Moreover, the analyses revealed novel or hardly investigated mechanisms for common ADRs including the potential involvement of 5-HT6-antagonism in weight gain, muscarinic receptor antagonism in dizziness, or 5-HT7-antagonism in sedation. To summarize, the presented study underlines the feasibility and value of a multivariate data mining approach in psychopharmacological development of antidepressants and antipsychotics.

Use of magnetic resonance imaging in pharmacogenomics

Because of the large variation in the response to psychoactive medication, many studies have attempted to uncover genetic factors that determine response. While considerable knowledge exists on the large effects of genetic polymorphisms on pharmacokinetics and plasma concentrations of drugs, effects of the concentration at the target site and pharmacodynamic effects on brain functions in disease are much less known. This article reviews the role of magnetic resonance imaging (MRI) to visualize response to medication in brain behaviour circuits in vivo in humans and assess the influence of pharmacogenetic factors. Two types of studies have been used to characterize effects of medication and genetic variation. In task-related activation studies the focus is on changes in the activity of a neural circuit associated with a specific psychological process. The second type of study investigates resting state perfusion. These studies provide an assessment of vascular changes associated with bioavailability of drugs in the brain, but may also assess changes in neural activity after binding of centrally active agents. Task-related pharmacogenetic studies of cognitive function have characterized the effects in the prefrontal cortex of genetic polymorphisms of dopamine receptors (DRD2), metabolic enzymes (COMT) and in the post-synaptic signalling cascade under the administration of dopamine agonists and antagonists. In contrast, pharmacogenetic imaging with resting state perfusion is still in its infancy. However, the quantitative nature of perfusion imaging, its non-invasive character and its repeatability might be crucial assets in visualizing the effects of medication in vivo in man during therapy.

Is there a personalized medicine for mood disorders?

Major Depressive Disorder (MDD) and antidepressant therapy response are largely based on behavioral criteria, which are known to correlate at best modestly with biological measures. Therefore, it is not surprising that the search for peripheral biological markers (biomarkers) being assessed in distant biological systems such as body fluids has not yet resulted in clinically convincing measures for MDD diagnostics or treatment evaluation. Imaging genetics studies, however, have been successful in the search for intermediate imaging phenotypes of MDD and treatment response that are directly related to the neurobiological underpinnings of MDD, but are not suitable for a broad clinical use today. Hence, we argue that intermediate phenotypes derived from imaging genetics studies should be utilized as substitutes of behaviorally assessed psychiatric diagnoses or therapy response in the search for easily accessible peripheral biomarkers. This article will further cover the current state of peripheral and neural biomarker research.