There is little clinical evidence when AST-120 should be prescribed for subjects with early stage overt diabetic nephropathy. We therefore designed a prospective, randomized, controlled study for subjects with type 2 diabetes (serum creatinine <1.5mg/dl and urinary protein >0.5g/day) in November, 2001. The primary end point was defined as exceeding 2mg/dl of serum creatinine, and the secondary end point was defined as introducing a hemodialysis. Twenty-two subjects were selected, and after excluding 6 drop-out subjects, 16 subjects (10 in the control group; 6 in the KRM group) finally entered the study. Mean follow-up periods were 37 and 34 months in the control and KRM groups, respectively. There was no difference in clinical characteristics including renal dysfunction at baseline between the two groups. There was a significant reduction in urinary indoxyl sulfate at month 12 in the KRM group than in the control group. A significant difference was observed in changes in mean levels of serum creatinine versus time between the two groups. The primary end points were counted in 7 (70%) of the control subjects, while only 1 (17%) of the KRM group, and the Kaplan-Meier analysis was statistically significant. Although 4 (40%) of the control group and 1 (17%) of the KRM group were initiated hemodialysis as the secondary end point, the difference did not reach a statistical significance. Thus, we concluded that administration of AST-120 initiated in early stage overt diabetic nephropathy stunts the progression of renal dysfunction.