gms | German Medical Science

64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

08. - 11.09.2019, Dortmund

Optimal designs for phase II/III drug development programs facilitated by R Shiny applications

Meeting Abstract

  • Stella Preussler - Institut für Medizinische Biometrie und Informatik, Abteilung Medizinische Biometrie, Universität Heidelberg, Heidelberg, Germany
  • Marietta Kirchner - Institut für Medizinische Biometrie und Informatik, Abteilung Medizinische Biometrie, Universität Heidelberg, Heidelberg, Germany
  • Heiko Götte - Merck KGaA, Darmstadt, Germany
  • Meinhard Kieser - Institut für Medizinische Biometrie und Informatik, Abteilung Medizinische Biometrie, Universität Heidelberg, Heidelberg, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 64. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Dortmund, 08.-11.09.2019. Düsseldorf: German Medical Science GMS Publishing House; 2019. DocAbstr. 168

doi: 10.3205/19gmds006, urn:nbn:de:0183-19gmds0064

Published: September 6, 2019

© 2019 Preussler et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Introduction: Recently, a utility-based approach was proposed to optimize planning of phase II/III drug development programs in terms of sample size allocation and go/no-go decision rules (whether to stop or to proceed to phase III) [1]. The utility function takes into account program costs and gains after a potential successful launch. We aimed at expanding the basic framework of [1] addressing several aspects of practical relevance and developing user friendly R Shiny applications to facilitate utilization.

Methods: The main extensions we implemented into the framework are:

1.
Optimal designs for drug development programs including methods for discounting overoptimistic phase II results [2]. Here, two bias adjustment strategies (“additive” [3] and “multiplicative” [4]) were adapted.
2.
Optimal designs for drug development programs, where several phase III trials can be conducted [5]. Different cases (number of successful phase III trials needed) and scenarios (number of conducted phase III trials) can be investigated, leading to more complex success criteria as compared to conducting one phase III trial.
3.
Optimal designs for drug development programs with multiple arms [6]. This is of interest for example for dose selection trials. Potential decisions are whether to conduct the phase III trial with a single (the most promising) treatment only, or with multiple treatments (if sufficiently promising).

The planning of optimized phase II/III drug development programs with utility functions require a lot of input parameters from many different functions (Medical, Operations, Commercial, ...). In order to facilitate the practical application of our approaches we implemented R Shiny applications for the “basic” setting as well as for the three extensions. The Apps will be presented during the talk (https://web.imbi.uni-heidelberg.de/drugdevelopR/).

Results: In general, discounting overoptimistic phase II results by multiplicative adjustment is the preferable option. However, there is no one-fits-all design. The optimal amount of adjustment, the optimal strategy (number of conducted phase III trials) and the optimal number of arms in phase III depend on various input parameters and the scenario at hand. Therefore, program-wise drug development planning is indispensable.

The R Shiny applications do not require programming skills for usage. They are of practical relevance for pharmaceutical companies as they allow convenient real-time planning with all function providing input.

In the user interfaces the drug development planning characteristics (e.g. fixed/ variable costs for phase II/III, assumed true treatment effect(s), significance level etc.) can be entered. Results of the optimization (e.g. maximal expected utility, optimal sample size allocation, optimal threshold value for the decision rule) and further program characteristics, such as the probability of a successful program, are listed tabularly and shown graphically.

Conclusion: Program-wise drug development planning is essential for saving resources. The proposed expanded utility framework enables to determine optimal phase II/III programs in various scenarios encountered in drug development. The R Shiny applications facilitate the utilization of our approaches. By means of illustrative examples, it could be shown that it is a flexible tool to incorporate different features and options so that one can tailor the approach to the specific scenario at hand.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.


References

1.
Kirchner M, Kieser M, Götte H, Schüler A. Utility-based optimization of phase II/III programs. Statistics in medicine. 2016 Jan 30;35(2):305-16.
2.
Preussler S, Kirchner M, Götte H, Kieser M. Optimal designs for phase II/III drug development programs including methods for discounting of phase II results. 2019. Submitted
3.
Wang SJ, Hung HJ, O'Neill RT. Adapting the sample size planning of a phase III trial based on phase II data. Pharmaceutical Statistics: The Journal of Applied Statistics in the Pharmaceutical Industry. 2006 Apr;5(2):85-97.
4.
Kirby S, Burke J, Chuang-Stein C, Sin C. Discounting phase 2 results when planning phase 3 clinical trials. Pharmaceutical Statistics. 2012 Sep;11(5):373-85.
5.
Preussler S, Kieser M, Kirchner M. Optimal sample size allocation and go/no-go decision rules for phase II/III programs where several phase III trials are performed. Biometrical Journal. 2019 Mar;61(2):357-78.
6.
Preussler S, Kirchner M, Götte H, Kieser M. Optimal Designs for Multi-Arm Phase II/III Drug Development Programs. 2019. Submitted