Genetic variability of GLRB impact cognitive behavioral therapy response in panic disorder

Introduction A previous GWAS (N = 1370) on the Agoraphobic Cognition Questionnaire (ACQ) identified the genome-wide significant GLRB locus and evaluated its influence on agoraphobic behavior in two independent samples (N = 2547 and N = 3845). Functional analyses of GLRB risk variants showed modulated gene expression, increased startle reactivity and fear network activation. Since deletions in GLRB cause the neurological disorder hyperekplexia, characterized by generalized startle reaction and agoraphobic behavior, genetic variability in GLRB may predispose to panic disorder (PD) by increased startle response and fear network activation.

Methods Doing first steps towards genotype-based therapy, we examined the impact of 3 GLRB risk variants (rs17035816, rs191260602, rs7688285) on cognitive behavioral therapy (CBT) response in 411 PD/AG patients obtained from the multicenter Panic-Net. Treatment response was defined as baseline to post treatment percentage change and linear regressions were run, with percentage change of different anxiety relevant traits (HAMA, PAS, ASI, ACQ, MI) as dependent and genotypes, age and sex as independent variables.

Results In accordance with GWAS results two variants affected CBT response predominantly in regard to agoraphobic behavior, captured by the ACQ and MI. A nominal significant reduction of agoraphobic symptoms post CBT was found for rs17035816 (pACQ = 0.023; pMI = 0.026) which was mainly driven by females (Females: pACQ = 0.037; pMI = 0.091) and still present after 6 month follow-up (Total: pACQ = 0.067; pMI = 0.040; Females: pACQ = 0.058; pMI = 0.041). A significant reduction of agoraphobic symptoms for rs7688285 was restricted to the ACQ after 6 month follow-up (Total: pACQ = 0.063; Females: pACQ = 0.021).

Conclusion Altogether, our findings propose that genetic variability of GLRB impact CBT response in regard to agoraphobic behavior. As GLRB can be subjected to pharmacological interventions, its modulation may open new perspectives for personalized pharmacological and behavioral therapeutic approaches in PD/AG. Due to small sample size an independent replication to support the validity of the observed results is necessary.