The 5-HTR2A gene is a candidate gene for influencing the clinical response to treatment with antidepressants. The purpose of this study was to determine the relationship between the -1438A/G polymorphism of the 5-HTR2A gene and the response to mirtazapine in a Korean population with major depressive disorder. Mirtazapine was administered for eight weeks to the 101 patients who completed the study, during which we evaluated the clinical outcome using repeated-measures ANCOVA. A main effect of genotype or an effect of genotype-time interactions on the decrease in HAMD score during the eight-week follow-up was not found, which suggests that the 5-HTR2A -1438A/G polymorphism does not affect the clinical outcome to mirtazapine administration. However, significant effects of genotype and allele carriers on the decrease in the sleep score over the eight weeks were found (genotype: F = 4.093, p = 0.017; allele: F = 4.371, p = 0.037), whereas no effect of genotype-time interactions on the decrease in the HAMD score over the eight-week follow-up was found. These observations suggest that the -1438A/G polymorphism on the sleep improvement at each time period revealed significant differences in the sleep scores after two weeks of mirtazapine administration. The sleep scores were lower for carriers of the A+ allele than of the A- allele after two weeks of mirtazapine administration (p = 0.041), which means that the -1438GG genotype is associated with less improvement in sleep, and suggests that the effect of mirtazapine on improving the sleep quality differs with the 5-HTR2A -1438A/G polymorphism within two weeks of mirtazapine treatment. In conclusion, although the -1438A/G polymorphism affects the sleep improvement resulting from the administration of mirtazapine to Korean patients with major depressive disorder, our results do not support the hypothesis that this polymorphism of the 5-HTR2A gene is involved in the therapeutic response to mirtazapine.