Abstract 4190: A survey of the most common methods of group randomization and distribution in preclinical in vivo studies

Abstract

Abstract To more closely emulate clinical research methodology and improve clinical relevance, researchers are increasingly utilizing clinically-similar group randomization and distribution methods for oncology studies in animals. One common pre-clinical approach, a random number generator in a spreadsheet, is paradoxically the least clinically-relevant, and subject to manual and transcription errors. This method provides no easy way to exclude animals from distribution based on other qualitative or quantitative data parameters. Due to limitations in spreadsheet software, commercially available pre-clinical trial applications can be used to perform the most recommended randomization and distribution methods based on tumor volume or other parameters: deterministic and matched distribution, pure randomization, stratified sampling randomization and block randomization across multiple numeric parameters. Subjects may be excluded initially based on recorded qualitative observations, and then by acceptable value ranges for numeric parameters. Animals can be randomized iteratively in rolling enrollment studies and previously randomized subjects can be re-randomized into new groups as required on drug-resistance studies. ANOVA results are displayed instantly to ensure similarity between groups. Standardizing randomization and distribution using clinically relevant methods improves the efficiency, clinical relevance, and outcome of animal studies relative to spreadsheet-based methods. Citation Format: Neil O'Brien, Jeffrey L. Kumer, Eric M. Ibsen. A survey of the most common methods of group randomization and distribution in preclinical in vivo studies. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4190. doi:10.1158/1538-7445.AM2014-4190