Depression is the most common psychiatric disorder with a lifetime prevalence of over 10% and is a disorder associated with a high suicide rate. The method of first choice of pharmacological treatment within the group of antidepressants is selective serotonin reuptake inhibitors (SSRI). The therapeutic response is usually observed only after 4-6 weeks. No reliable biological markers for early recognition or correlation with clinical therapeutic response have been established. The most commonly used rating scales in clinical practice are the Montgomery-Åsberg Depression Rating Scale (MADRS) and the Hamilton Depression Rating Scale (HAM-D). Conservatively, a positive therapeutic response is considered a reduction of the total sum on the MADRS of at least 50%. Proton magnetic resonance spectroscopy (1H-MRS) is a non-ionizing radiological method that enables multiple assessments in a shorter period of time and is the only method that allows in vivo recognition of metabolite levels in the brain. The aim of the research was to identify the correlations of clinical therapeutic response in depression with parameters measured by magnetic resonance spectroscopy, and especially choline (Cho) levels in the amygdala (AMYG). The research also included other neurometabolites: N-acetyl-aspartate (NAA), the glutamine/glutamate/GABA complex which appears on 1H-MRS as a Glx peak, myoinositol (MI) and creatine (Cr), which is considered a reference for determining the level of an individual metabolite due to its relatively stable level parameter. In addition to the amygdala, other brain regions of interest (ROIs) included the dorsolateral prefrontal cortex (DLPFC), basal ganglia (BG), anterior cingulate cortex (ACC), and hippocampus (HIPPO). The research was conducted on a group of 59 subjects (36 women and 23 men) who, at the beginning of the study, met the clinical criteria for a depressive disorder according to DSM-IV, with an average age of 29.4 years (SD = 7.4), in the range of 18.4 years to 42.5 years. Neurometabolite levels were measured by 1H-MRS at 3T at the beginning of the study, before starting antidepressant therapy, and after 8 weeks of treatment. No significant differences were found in the initial levels of Cho in AMYG between the group of subjects with a positive therapeutic response and the group in which the expected therapeutic response was not achieved. A linear mixed models (LMM) analysis revealed a negative correlation of Cho level with the degree of depressive symptoms after 8 weeks of SSRI antidepressant therapy. LMM analysis indicates a significant association of improvement in symptoms measured by the change in the total score on the MADRS with the effects of Cho (1.48, p = 0.036), treatment time (-11.53, p < 0.001) and the change in Cho expressed by the interaction of Cho and research time (-1.95, p = 0.044). The level of Cho at the end of the study was significantly higher in subjects who achieved a positive therapeutic response. Research results suggest that Cho level measured by 1H-MRS is an objective correlate of clinical assessment of therapeutic response measured by MADRS, with significant inter-individual variability in Cho levels. No gender differences were observed. The predictive validity of Cho has not been confirmed. Repeated results on larger groups of subjects, with more frequent objective evaluations by 1H-MRS and longer longitudinal follow-up, would make it possible to more reliably determine the correlation potential of Cho measured by 1H-MRS as an objective parameter in monitoring the therapeutic response.