Seeking ethically attractive dose-finding designs for narrow therapeutic index drugs
Rocío Lledó-García, Stefanie Hennig, Joakim Nyberg, Andrew C. Hooker & Mats O. Karlsson
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Objectives: Recently, a simulation-based comparison study on the relative merits of dose control-trials (DCT) with exposure-response analysis vs. concentration control-trials (CCT) for drugs with expected narrow therapeutic index was performed1. Contrary to what had been suggested2, 3, it shown that when learning about the exposure-response relationship, a DCT design is more informative than a CCT.
Herein, we revisit the question employing optimal design methodology, and propose strategies for designing ethically attractive trials for these drugs, which are balancing between individual-collective risk and informativeness.
Methods: A D-optimal study was performed using PopED 2.0, considering a hypothetical immunosuppressant agent with two clinical endpoints (rejections and infections). The drug exposure was described by a one-compartment model at steady state with clearances sampled from a normal distribution and the PD-relationship with two independent regression logistic models. Different scenarios were optimized applying cost-based designs (unwanted events vs. number of subjects/trial or maximal individual risk) both for DCT and CCT. For each design, two target doses (DCT) or exposures (CCT) were considered. The variables in the design which were simultaneously optimized were dose/exposure targets and number of subjects to include in the trial or in each arm. In order to decrease the costs of gaining new knowledge, the inclusion of prior information on baseline risks was also evaluated. .
Results: DCTs are more informative, needing smaller studies to provide the same information compare to CCTs. Using number of unwanted events, rather than subjects, as cost resulted in ethically more attractive designs. The inclusion of prior baseline risk information allowed substantial reduction in number of subject/events as well as utilization of targets closer to the optimal therapy.
Conclusions: Designing dose finding trials for narrow therapeutic index drugs can be substantially improved by using DCT with exposure response analysis, cost-based designs, use of prior information and optimal design analysis providing information on the ethical trade-off between individual risk and information gain..
References:
[1] Lledó-García R, Hennig S and Karlsson MO. Comparison of dose-finding designs for narrow therapeutic index drugs. Concentration-controlled versus dose-controlled trials. Clin Pharm Therapeutics. 2009 (in press).
[2] Endrenyi L and Zha J. Comparative efficiencies of randomized concentration- and dose-controlled clinical trials. Clin Pharmacol Ther. 1994; 56(3): 331-338.
[3] Sanathanan LP and Peck CC. The randomized concentration-controlled trial: an evaluation of its sample size efficiency. Control Clin Trials. 1991; 12(6): 780-794.
