Clinical pharmacology studies in patients with renal impairment: past experience and regulatory perspectives

The objective of this report is to provide a regulatory perspective on the quality of pharmacokinetic studies in renal impairment (RI) studies submitted in support of new drug applications (NDAs) or supplements to NDAs (sNDAs) submitted to the Food and Drug Administration (FDA). Fifty-one NDA and 20 sNDA submissions reviewed between 1996 and 1997 by the Office of Clinical Pharmacology and Biopharmaceutics were evaluated for the following: (1) whether an RI study was conducted; (2) contribution of the renal clearance to the overall clearance in subjects without renal impairment; (3) degree of plasma protein binding (%PB) in subjects without renal impairment; (4) dose proportionality of single and multiple doses; (5) study design, including dosing regimen; (6) definition of renal impairment; (7) stratification of renal functions; (8) number of subjects/group; (9) data analysis and interpretation; and (10) impact on labeling. Results of the analysis indicated that 67% of the NDAs and 30% of supplemental NDAs contained RI-studies (34/51 for NDAs and 6/20 for sNDAs). No obvious differences in the pharmacokinetic characteristics (e.g., percentage excreted unchanged in urine and %PB) were observed between drugs for which RI studies were conducted versus those not conducted. Most studies conducted were designed as single dose (70%). Seventy-five percent of the studies used doses within the therapeutic dosage range of the drug. The measured 24-hour creatinine clearance was most often used to assess the renal function. Stratification of renal function ranged from one to five groups, with 6 to 8 subjects enrolled per group. In most studies conducted (38/40), data were analyzed by point estimate using ANOVA. Results of RI studies were adequately reflected in the labeling. The survey reveals that RI study design can be improved for regulatory review purposes. In part based on this analysis, the FDA has prepared a guidance that provides recommendations on the design, analysis, and impact on dosing and labeling for RI studies to include recommendations on when RI studies do not need to be performed. The guidance proposes an equivalence approach with confidence intervals, as opposed to a point estimate approach, to assess the impact of RI on systemic exposure measures.