Performance of a neuro-fuzzy model in predicting weight changes of chronic schizophrenic patients exposed to antipsychotics

Machine learning for prediction of schizophrenia using genetic and demographic factors in the UK biobank

Machine learning (ML) holds promise for precision psychiatry, but its predictive performance is unclear. We assessed whether ML provided added value over logistic regression for prediction of schizophrenia, and compared models built using polygenic risk scores (PRS) or clinical/demographic factors. LASSO and ridge-penalised logistic regression, support vector machines (SVM), random forests, boosting, neural networks and stacked models were trained to predict schizophrenia, using PRS for schizophrenia (PRS_SZ), sex, parental depression, educational attainment, winter birth, handedness and number of siblings as predictors. Models were evaluated for discrimination using area under the receiver operator characteristic curve (AUROC) and relative importance of predictors using permutation feature importance (PFI). In a secondary analysis, fitted models were tested for association with schizophrenia-related traits which had not been used in model development. Following learning curve analysis, 738 cases and 3690 randomly sampled controls were selected from the UK Biobank. ML models combining all predictors showed the highest discrimination (linear SVM, AUROC = 0.71), but did not significantly outperform logistic regression. AUROC was robust over 100 random resamples of controls. PFI identified PRS_SZ as the most important predictor. Highest variance in fitted models was explained by schizophrenia-related traits including fluid intelligence (most associated: linear SVM), digit symbol substitution (RBF SVM), BMI (XGBoost), smoking status (XGBoost) and deprivation (linear SVM). In conclusion, ML approaches did not provide substantial added value for prediction of schizophrenia over logistic regression, as indexed by AUROC; however, risk scores derived with different ML approaches differ with respect to association with schizophrenia-related traits.

Key use cases for artificial intelligence to reduce the frequency of adverse drug events: a scoping review

Adverse drug events (ADEs) represent one of the most prevalent types of health-care-related harm, and there is substantial room for improvement in the way that they are currently predicted and detected. We conducted a scoping review to identify key use cases in which artificial intelligence (AI) could be leveraged to reduce the frequency of ADEs. We focused on modern machine learning techniques and natural language processing. 78 articles were included in the scoping review. Studies were heterogeneous and applied various AI techniques covering a wide range of medications and ADEs. We identified several key use cases in which AI could contribute to reducing the frequency and consequences of ADEs, through prediction to prevent ADEs and early detection to mitigate the effects. Most studies (73 [94%] of 78) assessed technical algorithm performance, and few studies evaluated the use of AI in clinical settings. Most articles (58 [74%] of 78) were published within the past 5 years, highlighting an emerging area of study. Availability of new types of data, such as genetic information, and access to unstructured clinical notes might further advance the field.

Metabolic disturbances associated with antipsychotic drug treatment in patients with schizophrenia: State-of-the-art and future perspectives

Metabolic disturbances and obesity are major cardiovascular risk factors in patients with schizophrenia, resulting in a higher mortality rate and shorter life expectancy compared with those in the general population. Although schizophrenia and metabolic disturbances may share certain genetic or pathobiological risks, antipsychotics, particularly those of second generation, may further increase the risk of weight gain and metabolic disturbances in patients with schizophrenia. This review included articles on weight gain and metabolic disturbances related to antipsychotics and their mechanisms, monitoring guidelines, and interventions. Nearly all antipsychotics are associated with weight gain, but the degree of the weight gain varies considerably. Although certain neurotransmitter receptor-binding affinities and hormones are correlated with weight gain and specific metabolic abnormalities, the precise mechanisms underlying antipsychotic-induced weight gain and metabolic disturbances remain unclear. Emerging evidence indicates the role of genetic polymorphisms associated with antipsychotic-induced weight gain and antipsychotic-induced metabolic disturbances. Although many guidelines for screening and monitoring antipsychotic-induced metabolic disturbances have been developed, they are not routinely implemented in clinical care. Numerous studies have also investigated strategies for managing antipsychotic-induced metabolic disturbances. Thus, patients and their caregivers must be educated and motivated to pursue a healthier life through smoking cessation and dietary and physical activity programs. If lifestyle intervention fails, switching to another antipsychotic drug with a lower metabolic risk or adding adjunctive medication to mitigate weight gain should be considered. Antipsychotic medications are essential for schizophrenia treatment, hence clinicians should monitor and manage the resulting weight gain and metabolic disturbances.

Genetics of Common Antipsychotic-Induced Adverse Effects

The effectiveness of antipsychotic drugs is limited due to accompanying adverse effects which can pose considerable health risks and lead to patient noncompliance. Pharmacogenetics (PGx) offers a means to identify genetic biomarkers that can predict individual susceptibility to antipsychotic-induced adverse effects (AAEs), thereby improving clinical outcomes. We reviewed the literature on the PGx of common AAEs from 2010 to 2015, placing emphasis on findings that have been independently replicated and which have additionally been listed to be of interest by PGx expert panels. Gene-drug associations meeting these criteria primarily pertain to metabolic dysregulation, extrapyramidal symptoms (EPS), and tardive dyskinesia (TD). Regarding metabolic dysregulation, results have reaffirmed HTR2C as a strong candidate with potential clinical utility, while MC4R and OGFR1 gene loci have emerged as new and promising biomarkers for the prediction of weight gain. As for EPS and TD, additional evidence has accumulated in support of an association with CYP2D6 metabolizer status. Furthermore, HSPG2 and DPP6 have been identified as candidate genes with the potential to predict differential susceptibility to TD. Overall, considerable progress has been made within the field of psychiatric PGx, with inroads toward the development of clinical tools that can mitigate AAEs. Going forward, studies placing a greater emphasis on multilocus effects will need to be conducted.

Genetics of antipsychotic-induced weight gain: update and current perspectives

Antipsychotic medications are used to effectively treat various symptoms for different psychiatric conditions. Unfortunately, antipsychotic-induced weight gain (AIWG) is a common side effect that frequently results in obesity and secondary medical conditions. Twin and sibling studies have indicated that genetic factors are likely to be highly involved in AIWG. Over recent years, there has been considerable progress in this area, with several consistently replicated findings, as well as the identification of new genes and implicated pathways. Here, we will review the most recent genetic studies related to AIWG using the Medline database (PubMed) and Google Scholar. Among the steadiest findings associated with AIWG are serotonin 2C receptors (HTR2C) and leptin promoter gene variants, with more recent studies implicating MTHFR and, in particular, MC4R genes. Additional support was reported for the HRH1, BDNF, NPY, CNR1, GHRL, FTO and AMPK genes. Notably, some of the reported variants appear to have relatively large effect sizes. These findings have provided insights into the mechanisms involved in AIWG and will help to develop predictive genetic tests in the near future.

Effects of sports participation on psychiatric symptoms and brain activations during sports observation in schizophrenia

Weight gain has been identified as being responsible for increased morbidity and mortality rates of schizophrenia patients. For the management of weight gain, exercise is one of the most acknowledged interventions. At the same time, exercise and sports have been recognized for their positive impact on psychiatric symptoms of schizophrenia. However, the neurobiological basis for this remains poorly understood. We aimed to examine the effect of sports participation on weight gain, psychiatric symptoms and brain activation during sports observation in schizophrenia patients. Thirteen schizophrenia patients who participated in a 3-month program, including sports participation and 10 control schizophrenia patients were studied. In both groups, body mass index (BMI), Positive and Negative Syndrome Scale (PANSS), and brain activation during observation of sports-related actions measured by functional magnetic resonance imaging were accessed before and after a 3-month interval. BMI and general psychopathology scale of PANSS were significantly reduced in the program group but not in the control group after a 3-month interval. Compared with baseline, activation of the body-selective extrastriate body area (EBA) in the posterior temporal-occipital cortex during observation of sports-related actions was increased in the program group. In this group, increase in EBA activation was associated with improvement in the general psychopathology scale of PANSS. Sports participation had a positive effect not only on weight gain but also on psychiatric symptoms in schizophrenia. EBA might mediate these beneficial effects of sports participation. Our findings merit further investigation of neurobiological mechanisms underlying the therapeutic effect of sports for schizophrenia.

Pharmacogenetics of antipsychotic-induced weight gain: review and clinical implications

Second-generation antipsychotics (SGAs), such as risperidone, clozapine and olanzapine, are the most common drug treatments for schizophrenia. SGAs presented an advantage over first-generation antipsychotics (FGAs), particularly regarding avoidance of extrapyramidal symptoms. However, most SGAs, and to a lesser degree FGAs, are linked to substantial weight gain. This substantial weight gain is a leading factor in patient non-compliance and poses significant risk of diabetes, lipid abnormalities (that is, metabolic syndrome) and cardiovascular events including sudden death. The purpose of this article is to review the advances made in the field of pharmacogenetics of antipsychotic-induced weight gain (AIWG). We included all published association studies in AIWG from December 2006 to date using the Medline and ISI web of knowledge databases. There has been considerable progress reaffirming previous findings and discovery of novel genetic factors. The HTR2C and leptin genes are among the most promising, and new evidence suggests that the DRD2, TNF, SNAP-25 and MC4R genes are also prominent risk factors. Further promising findings have been reported in novel susceptibility genes, such as CNR1, MDR1, ADRA1A and INSIG2. More research is required before genetically informed, personalized medicine can be applied to antipsychotic treatment; nevertheless, inroads have been made towards assessing genetic liability and plausible clinical application.

Fuzzy logic: A "simple" solution for complexities in neurosciences?

Background:

Fuzzy logic is a multi-valued logic which is similar to human thinking and interpretation. It has the potential of combining human heuristics into computer-assisted decision making, which is applicable to individual patients as it takes into account all the factors and complexities of individuals. Fuzzy logic has been applied in all disciplines of medicine in some form and recently its applicability in neurosciences has also gained momentum.

Methods:

This review focuses on the use of this concept in various branches of neurosciences including basic neuroscience, neurology, neurosurgery, psychiatry and psychology.

Results:

The applicability of fuzzy logic is not limited to research related to neuroanatomy, imaging nerve fibers and understanding neurophysiology, but it is also a sensitive and specific tool for interpretation of EEGs, EMGs and MRIs and an effective controller device in intensive care units. It has been used for risk stratification of stroke, diagnosis of different psychiatric illnesses and even planning neurosurgical procedures.

Conclusions:

In the future, fuzzy logic has the potential of becoming the basis of all clinical decision making and our understanding of neurosciences.

C825T polymorphism of the GNB3 gene on valproate-related metabolic abnormalities in bipolar disorder patients

BACKGROUND

Valproate (VPA) is a mood stabilizer for treating patients with bipolar disorder (BD). It may cause metabolic abnormalities in certain bipolar patients. However, the genetic factors that influence the susceptibility remain unclear. Genetic polymorphism of the G-protein β3 subunit (GNB3) is reported to be associated with metabolic phenotypes. In the current study, we investigated the possible associations between the GNB3 variation and VPA-induced metabolic abnormalities.

METHODS

Subjects (n = 96) who met the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria for BD were recruited from the National Cheng Kung University Hospital. Their metabolic indices were measured.

RESULTS

The variation of GNB3 C825T showed an association with higher plasma total cholesterol (P = 0.037), triglyceride (P = 0.014), and leptin (P < 0.001) levels in BD patients treated with VPA. After adjusting for age, sex, types of BDs, and serum concentration of VPA, the variation of GNB3 C825T remained significantly associated with the levels of serum leptin and body mass index (BMI; P < 0.001 and P = 0.030, respectively). In addition, the GNB3 C825T showed significant drug-single-nucleotide polymorphism interactions with insulin levels (P = 0.033), triglyceride levels (P = 0.013), leptin levels (P = 0.013), and BMI (P = 0.018). These results indicated that the T allele may be associated with lower serum leptin levels and BMI in BD patients treated with VPA.

CONCLUSIONS

The current study provides evidence that BD patients who are T allele carriers of the GNB3 C825T polymorphism have a lower risk for VPA-induced metabolic abnormalities. Further studies about the underlying mechanisms of G protein in VPA-induced metabolic abnormalities are warranted.

Association of HTR2C, but not LEP or INSIG2, genes with antipsychotic-induced weight gain in a German sample

Background: Drug-induced bodyweight gain (BWG) is a serious concern in pharmacotherapy with second-generation antipsychotics. The interindividual variability is likely to be modulated by genetic factors. In the past, pharmacogenetic studies yielded conflicting results, and none of the identified genetic alterations exerts sufficient predictive value for this severe side effect of psychopharmacotherapy. Aim: We aimed to contribute to the replication and extension of prior association findings and investigated the genes encoding serotonin 2C receptor (HTR2C), insulin-induced gene 2 (INSIG2) and leptin (LEP). Patients & methods: We investigated the association of HTR2C, LEP and INSIG2 SNPs with antipsychotic-induced BWG in 128 German schizophrenic patients. Genotyping was performed for nine SNPs (HTR2C: rs498207, rs3813928, rs6318 and rs3813929; INSIG2: rs17587100, rs10490624, rs17047764 and rs7566605; LEP: rs7799039). Association analysis included logistic regression analysis and Pearson´s χ2 tests. Results: We report a significant association of three HTR2C SNPs (rs498207, rs3813928 and rs3813929) and of the respective haplotype with antipsychotic-induced BWG. Regarding the X-chromosomal SNP rs498207, individuals with AA/A genotype gained more weight than those with GG/G genotype. The association observed with the SNP rs498207 was also significant after correcting for multiple testing (p = 0.0196). No association was found for INSIG2 and LEP SNPs. Conclusion: The results contribute to the accumulating evidence for an association of the X-chromosomal HTR2C gene with antipsychotic-induced BWG. The proposed underlying mechanisms include decreased HTR2C gene expression with reduced 5-HT-modulated activation of hypothalamic proopiomelanocortin-neurons, and inverse 5-HT2C agonism in the presence of D2 receptor antagonism.

The intersection of pharmacology, imaging, and genetics in the development of personalized medicine

We currently rely on large randomized trials and meta-analyses to make clinical decisions; this places us at a risk of discarding subgroup or individually specific treatment options owing to their failure to prove efficacious across entire populations. There is a new era emerging in personalized medicine that will focus on individual differences that are not evident phenomenologically. Much research is directed towards identifying genes, endophenotypes, and biomarkers of disease that will facilitate diagnosis and predict treatment outcome. We are at the threshold of being able to predict treatment response, primarily through genetics and neuroimaging. In this review we discuss the most promising markers of treatment response and adverse effects emerging from the areas of pharmacogenetics and neuroimaging in depression and schizophrenia.

Association of the α2A adrenergic receptor -1291C/G polymorphism and antipsychotic-induced weight gain in European–Americans

Aims: To investigate the -1291 C/G promoter polymorphism (rs1800544) of the adrenergic α-2A receptor (ADRA2A) with clozapine-/olanzapine-induced weight gain in European–Americans and African–Americans. The α-adrenergic receptors inhibit lipolysis in the adipose tissue and are involved in weight gain regulation. Moreover, two previous studies indicated an association with antipsychotic-induced weight gain with the same polymorphism in Asian populations. Materials & methods: We analyzed a relatively large (n = 129) and well-characterized group of patients and monitored them for a period of 6–14 weeks. Our refined sample consisted of 60 European–Americans and 39 African–Americans on clozapine or olanzapine, prospectively. Results: In European–Americans, we observed a significant difference in weight gain across the genotypic categories (p = 0.046). The carriers of the C allele gained more weight compared with the subjects homozygous for the GG allele (CC + CG vs GG; 3.73 ± 4.13 kg vs 0.23 ± 2.92 kg; p = 0.013). We did not find a significant association in African–Americans, although the sample size was probably too small. Conclusion: Our observations suggest a possible role of ADRA2A polymorphisms in clozapine-/olanzpaine-induced weight gain in subjects of European descent.

Cellular cation transport studied by 6/7Li and 23Na NMR in a porous Mo132 Keplerate type nano-capsule as model system

Li+ ions can interplay with other cations intrinsically present in the intra- and extra-cellular space (i.e. Na+, K+, Mg2+ and Ca2+) have therapeutic effects (e.g. in the treatment of bipolar disorder) or toxic effects (at higher doses), likely because Li+ interferes with the intra-/extra-cellular concentration gradients of the mentioned physiologically relevant cations. The cellular transmembrane transport can be modelled by molybdenum-oxide-based Keplerates, i.e. nano-sized porous capsules containing 132 Mo centres, monitored through 6/7Li as well as 23Na NMR spectroscopy. The effects on the transport of Li+ cations through the 'ion channels' of these model cells, caused by variations in water amount, temperature, and by the addition of organic cationic 'plugs' and the shift reagent [Dy(PPP)2](7-) are reported. In the investigated solvent systems, water acts as a transport mediator for Li+. Likewise, the counter-transport (Li+/Na+, Li+/K+, Li+/Cs+ and Li+/Ca2+) has been investigated by 7Li NMR and, in the case of Li+/Na+ exchange, by 23Na NMR, and it has been shown that most (in the case of Na+ and K+, all (Ca2+) or almost none (Cs+) of the Li cations is extruded from the internal sites of the artificial cell to the extra-cellular medium, while Na+, K+ and Ca2+ are partially incorporated.