Anxiety disorders (ADs) are common psychiatric disorders, with a lifetime prevalence estimated at 33.7% in epidemiological studies. ADs are associated with serious disability and severe impairment in quality of life. Although several treatments [e.g. selective serotonin reuptake inhibitors (SSRIs), serotonin-noradrenaline reuptake inhibitors (SNRIs), pregabalin, tricyclic antidepressants and benzodiazepines and/or cognitive-behaviour therapy (CBT)] are recommended, a large number of patients (i.e. from 30 to 70%) do not achieve complete remission. According to the novel paradigm of personalized medicine, the search of possible predictors of both disease vulnerability and treatment response might be the best way to prevent more accurately disease risk and to tailor the most effective treatment for each individual. Although a growing body of studies have proposed several endophenotypes/markers (i.e. neurochemical, neuroimaging, physiological, genetic and epigenetic endophenotypes/markers) as possible predictors of ADs susceptibility and/or treatment response, findings are not robust enough to be considered acceptable to incorporate in the clinical practice. In order to obtain more reliable results, larger studies with a multimodal approach, based on a combination of different biomarkers, are needed.