Association of genetic risk scores with body mass index in Swiss psychiatric cohorts.

TLDR: The present study replicated in psychiatric cohorts previously identified BMI risk variants obtained in GWAS analyses from population-based samples previously identified in candidate-gene approach and Genome-Wide Association Studies.

Abstract: Weight gain is associated with psychiatric disorders and/or with psychotropic drug treatments. We analyzed in three psychiatric cohorts under psychotropic treatment the association of weighted genetic risk scores (w-GRSs) with BMI by integrating BMI-related polymorphisms from the candidate-gene approach and Genome-Wide Association Studies (GWAS).

Figures

Table 4. Description of 4 quartiles of GRS for 32 SNP in the combined sample.

Fig 1. Evolution of BMI over time between Genetic Risk Score extreme percentiles

Table 2. Significant results obtained from individual SNP association with BMI in the psychiatric sample 1 at baseline and at 12 months of follow-up treatment.

Fig 1. Evolution of BMI over time between Genetic Risk Score extreme percentiles

Table 3. Weighted GRS association with BMI obtained from 32 Genome Wide Association Studies SNPs.

Table 3. Weighted GRS association with BMI obtained from 32 Genome Wide Association Studies SNPs.

Full text

Serveur Académique Lausannois SERVAL serval.unil.ch
Author Manuscript
Faculty of Biology and Medicine Publication
This paper has been peer-reviewed but dos not include the final publisher
proof-corrections or journal pagination.
Published in final edited form as:
Title: Association of genetic risk scores with body mass index in Swiss
psychiatric cohorts.
Authors: Saigi-Morgui N, Vandenberghe F, Delacrétaz A, Quteineh L,
Gholamrezaee M, Aubry JM, von Gunten A, Kutalik Z, Conus P, Eap CB
Journal: Pharmacogenetics and genomics
Year: 2016 May
Volume: 26
Issue: 5
Pages: 208-17
DOI: 10.1097/FPC.0000000000000210

1
Association of Genetic Risk Scores (GRS) with Body Mass Index in
Swiss Psychiatric Cohorts
Núria Saigi-Morgui (1); Frederik Vandenberghe (1); Aurélie Delacrétaz (1); Lina Quteineh (1); Mehdi
Gholamrezaee (2); Jean-Michel Aubry (3); Armin von Gunten (4); Zoltán Kutalik (5, 6); Philippe Conus (7);
Chin B Eap (1,8)*
1
Unit of Pharmacogenetics and Clinical Psychopharmacology, Centre for Psychiatric Neuroscience,
Department of Psychiatry, Lausanne University Hospital, Prilly, Switzerland.
2
Centre of Psychiatric Epidemiology and Psychopathology, Department of Psychiatry, Lausanne
University Hospital, Prilly, Switzerland.
3
Department of Mental Health and Psychiatry, University Hospital of Geneva, Geneva, Switzerland.
4
Service of Old Age Psychiatry, Department of Psychiatry, Lausanne University Hospital, Prilly
Switzerland.
5
Institute of Social and Preventive Medicine (IUMSP), Lausanne University Hospital,
Lausanne, Switzerland.
6
Swiss Institute of Bioinformatics, Lausanne, Switzerland.
7
Service of General Psychiatry, Department of Psychiatry, Lausanne University Hospital, Prilly
Switzerland.
8
School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Geneva,
Switzerland.
*Corresponding author:
E-mail: Chin.Eap@chuv.ch (CBE)

2
Funding:
This work has been funded in part by the Swiss National Research Foundation (CBE and PC:
320030-120686 and 324730-144064). ZK received support from the Leenaards Foundation and
the Swiss National Science Foundation (31003A-143914). The funding sources had no role in
the writing of the manuscript or in the decision to submit it for publication.
Author disclosure information
CBE received research support from Takeda and from the Roche Organ Transplantation
Research Foundation in the past 3 years. He received honoraria for conferences or teaching
CME courses from Advisis, Astra Zeneca, Janssen-Cilag, Lundbeck, Merck Sharp & Dohme,
Otsuka, Sandoz, Servier and Vifor-Pharma in the past 3 years. AvG received honoraria for a
conference and a workshop participation unrelated to the topic of this study from Vifor and
Bayer Sheringer in the past 3 years. All other authors declare no conflict of interest in relation to
the content of the paper.

3
ABSTRACT
OBJECTIVE: Weight gain is associated with psychiatric disorders and/or with psychotropic drug
treatments. We analyzed in three psychiatric cohorts under psychotropic treatment the
association of weighted genetic risk scores (wGRS) with Body Mass Index (BMI) by integrating
BMI-related polymorphisms from Candidate Gene approach and Genome Wide Association
Studies (GWAS).
MATERIALS AND METHODS: wGRS of 32 polymorphisms previously associated with BMI in
general population GWAS and 20 polymorphisms associated with antipsychotics induced weight
gain were investigated in three independent psychiatric samples.
RESULTS: wGRS of 32 polymorphisms were significantly associated with BMI in the psychiatric
sample 1 (n=425) and were replicated in another sample (n=177). Those at the percentile 95
(p95) of the score had 2.26 and 2.99 kg/m
2
higher predicted BMI compared to individuals at the
percentile 5 (p5) in the Sample 1 and in the Sample 3 (p=0.009, p=0.04, respectively). When
combining all samples together (n=750), a significant difference of 1.89 kg/m
2
predicted BMI
was found between p95 and p5 individuals at 12 months of treatment. Stronger associations
were found among men (difference: 2.91 kg/m
2
of predicted BMI between p95 and p5,
p=0.0002) whereas no association was found among women. wGRS of 20 polymorphisms was
not associated with BMI. The wGRS of 52 polymorphisms and the clinical variables (age, sex,
treatment) explained 1.99% and 3.15%, respectively of BMI variability.
CONCLUSION: The present study replicated in psychiatric cohorts previously identified BMI risk
variants obtained in GWAS analyses from population-based samples. Gender specific analysis
should be considered in further analysis.

4
Key words: Psychotropic drugs, Genetic Risk Score, Psychiatry, Body Mass Index

Frequently asked questions

Q1. What is the appropriate link function for the Gamma family?

The appropriate link function the authors chose for theBMI variable is the inverse function which is the canonical link function for the Gamma family. 

Q2. What is the effect of a deletion of CRTC2 on the glyce?

In addition, a deletion of CRTC2 impairs the expression of thegluconeogenic genes and the ability of glucagon to stimulate glucose production in hepatocytes[68]. 

Q3. what is the inverse link function for the grs?

The choice of Generalized Linear Mixed Models which is a general form of Linear Mixed Modelshas been made to take into account the canonical behavior of BMI values; exploratory analysisof the data strongly suggested an inverse link function which is the canonical link function for theGamma family. 

Q4. What are the advantages of second generation antipsychotics?

Since their introduction onto the market, second generation antipsychotics (SGA) have beenwidely used over first generation antipsychotics (FGA), as they clearly show an advantage interms of reduced risks of extrapyramidal side-effects, as well as some advantages for thetreatment of negative symptoms. 

Q5. What is the way to compare the effects of a GLMM on BMI?

GLMMs combine both linear mixed models (which incorporate random effects) and generalizedlinear models (which deal with non-normal data by using link functions and exponential family)12[59]. 

Q6. What is the method to build a GRS?

Among the several existing methods to build a GRS, it has been shown in disease risk modelingthat weighted GRS methods are preferred to the simple count method when relative risks varyamong SNPs [56]. 

Q7. What is the significance of the present study?

In conclusion, the present study replicated in psychiatric cohorts previously identified BMI riskvariants obtained in GWAS analyses from population-based samples. 

Q8. What are the common genes associated with obesity?

With the emergence of genome wide association studies (GWAS), thousands of newpolymorphisms associated with obesity and metabolic phenotypes have been elucidated.