The study was aimed at investigating the association between MDR-1 genetic polymorphisms [C1236T, G2677T(A), C3435T] and parameters describing the clinical course and treatment response of childhood steroid-responsive nephrotic syndrome (SRNS). Three MDR-1 genetic markers were analyzed in 108 children diagnosed with SRNS and in 135 healthy controls with neither allergic nor renal disease. All subjects were genotyped by PCR-restriction fragment length polymorphism (RFLP) analysis, and an EM algorithm-based analysis was utilized to estimate haplotype frequencies. As expected, there was no difference in genotypic and allelic distribution between and among SRNS patients and healthy children. However, all individual polymorphisms were strongly associated with time to response to initial prednisone therapy. The frequencies of the mutated alleles were higher in late responders (time to remission: >7 days) to oral prednisone (0.53, 0.52,0.66) than in early responders (time to remission: <7 days; 0.24, 0.19, 0.32), with all p values <0.001 for positions 1236, 2677 and 3435, respectively). Odds ratios (ORs) reflecting the strength of the associations were as follows: 6.79 (95% CI:1.96- 23.54) for 1236 T/T, 13.7 (95% CI:2.78-67) for 2677 T/T and 9.92 (95% CI: 3.01-32.71) for 3435 T/T as compared to the respective-wild type homozygotes. The TTT haplotype was similarly found to be significantly associated with late oral steroid response (0.49 vs. 0.19, p=0.0003). Variants 1236T, 2677TA and 3435T identify patients that respond slower to oral prednisone. Although the functional properties of the substitutions investigated here are still to be determined, our findings may be a small step toward the optimization of immunosuppressive therapy in SRNS children.