Theory of drug development (Q2864531)

Drug development is a complex process, starting from docking in biochemistry up to testing new molecules in order to show that drus are safe and effective. But in all stages of this process the main basis is biostatistics. The given book presents a theory of drug development that is based on maximizing the efficiency with which drugs that truly provide clinical benefits are identified. The author shows how to optimize the drug development process at its' three main stages (Phases 1, 2, 3), and at some transitional sub-stages, so that the number of molecules that result in a positive final Phase 3 clinical trial per investment is maximized.NEWLINENEWLINE In Phase 2 are conducted screening trials when checks of a lot of candidates have completed Phase 1 (in-vitro). So the author presents the construction of admissible class of Phase 2 trial designs and decision rules. Then it is shown that a clinical development program consisting only of a Phase 3 trial with no accompanying Phase 2 trial maximizes the minimum relative efficiency and then that it maximizes the minimum efficiency. But selection bias can make a drug that is not different than a placebo to have promising activity. So he then presents an approach to using Phase 2 single-arm trials that account for potential selection bias. the author concludes that ``Phase 2 single-arm trials that enroll all eligible subjects will have more power than a randomized Phase 2 trial with the same number of subjects''. Then the author elaborates the case when a Phase 2 trial is based on a sample including surrogates. What is the impact on the efficiency? Using maximum likelihood estimates and the variance-covariance matrix for parameters it is approved that it can improve the efficiency of drug development aimed at discovering clinical settings where a class of drugs may provide a survival benefit.NEWLINENEWLINE Sizing a trial in one phase of the drug development and then deciding whether to go on to the next phase is an issue that permeates all of the drug development (not only). So the author shows how to maximize the efficiency of a drug development through the optimization of the size and type 1 error of Phase 2 trials. Aiming this it an approach was adapted developed for evaluating the Phase 2/3 decision problem to address the optimization of multiple screening trials that precede a definitive trial. But the conclusions from this case seem as trivial. Really, ``the efficiency of a program that includes Phases 1, 2, 3, trials is better than a program that includes only Phases 2 and 3''.NEWLINENEWLINE The next chapter earmarks for an exploration of three important properties that impact the strength of evidence developed by a clinical trial program to support the marketing approval of a new drug: the supporting statistical test, the determination of what is a ``positive'' result, and the degree of confirmation. The finding of the formal conclusions is the preference for simple statistical tests of hypotheses over model-based tests. In quantifying the strenth of evedence were used Wald's likelihood ratio test and Pocock's stopping boundary for group sequential clinical trials.NEWLINENEWLINE There are two issues in clinical trials: the first issue is whether Phase 2 data can be combined with data from Phase 3 trials. In this case the total number of subjects required to achieve a significant p-value with sufficient power that will support regulatory approval can be reduced. The second issue is whether a single Phase 3 trial can be considered confirmatory, i.e., a positive Phase 3 trial is consistent with a global null hypothesis without performing a second Phase 3 trial with the global null hypothesis without performing a second Phase 3 trial. Writing likelihood functions for the collection of Phase 2 and 3 trials the author derives a formula for estimation of the impact on the efficiency of the drug development of scientifically interpreting the data collected in Phase 2 as well as the impact of using the Phase 2 trial as a screen.NEWLINENEWLINE Section 3 of this book is dedicated to additional topics, namely, to: maximizing of the efficiency subject to a constraint on the ratio of true to false positives, minimizing the log rank test, adaptive Phase 2/3 designs, size of the Phase 3 trial, and extending the model of the clinical drug development.