Object : Currently, the alteration of cytokine system has been known to play an important role in regard to depressive symptom. We focused on the relationship between immunological parameters and clinical improvement in major depressive disorder. Method : Data were collected on 26 patients with major depressive disorder using a 8-week prospective follow-up design. After 8-week treatment period with fluoxetine, patients were classified into a response group and a non-response group according to their psychopathological outcome as evaluated by Hamilton Depression Rating Scale. The differences of the immunological parameters between pre-treatment phase and post-treatment phase were compared among patients. The difference of those was also compared within each phase among them. The relationship between socio-demographic variables, depression, cytokine, mononuclear cells was examined by correlation analysis. Multiple regression analyses were performed to explore the predictors of clinical improvement of major depressive disorder. Result : Pre-treatment levels of IL-$1{eta}$ in the response group were significantly higher than those in the non-response group. Pre-treatment levels of IL-$1{eta}$ of all patients and in the response group were positively correlated with pre-treatment monocyte counts. Patients with subsequent remission showed significantly lower IL-6 values at baseline than those with non-response. Post-treatment values of IL-6 did not differ significantly among the patients. The correlation test showed more frequent relations among cytokines and mononuclear cells in the response group than in the non-responder group. Especially, serum level of IL-6 in pre-treatment phase was only significantly correlated with HAMD score after 8-week treatement phase, while other cytokines and mononuclear cells were not. Pretreatment level of IL-6 was of paramount importance in predicting clinical improvement of depressive symptom. Conclusion : The immune system of major depressive disorder patients might dichotomize the patients into subsequent responders and non-responders. Immune system might be of great influence on the clinical improvement of major depressive disorder. The mode of interaction between depression and cellular immune function and the mediators responsible for the cytokine production need to be studied further.