Sivarajah A, Collins S, Sutton M R, Regan N, West H, Holbrook M, Edmunds N
Global Safety Pharmacology, Pfizer Global Research and Development, Sandwich, Kent, CT13 9NJ, UK.
J Pharmacol Toxicol Methods. 2010 Jul-Aug;62(1):12-9. doi: 10.1016/j.vascn.2010.05.011. Epub 2010 Jun 4.
ICH S7A and S7B guidelines recommend the use of conscious animals for assessment of non-clinical cardiovascular safety of new chemical entities prior to testing in humans. Protocol design and data analysis techniques can affect the quality of the data produced and can therefore ultimately influence the clinical management of cardiovascular risk. It is therefore essential to have an understanding of the magnitude of changes detectable and the clinical relevance of these changes. This paper describes the utilisation of "super-intervals" to analyse and interpret data obtained from our conscious telemetered dog cardiovascular safety protocol and reports the statistical power achieved to detect changes in various cardiovascular parameters.
Cardiovascular data from 18 dog telemetry studies were used to calculate the statistical power to detect changes in cardiovascular parameters. Each study followed a test compound versus vehicle cross-over experimental design with 24h monitoring (n=4). 1 min mean raw data from each individual animal was compressed into 15 min mean data for each dose group for visualisation. Larger summary periods, or "super-intervals", were then selected to best represent any observed cardiovascular effects whilst taking into account the pharmacokinetic profile of the drug e.g. intervals of 1 to 6, 7 to 14 and 14 to 22h post-dose.
With this methodology and study design we predict, using the median percentile that our studies have 80% power to detect the following changes: HR (+/-10bpm), LV +dP/dt max (+/-375mmHg/s), MBP (+/-5mmHg) and QTc (+/-4ms).
Super-intervals are a simple way to handle the high degree of natural variability seen with any ambulatory cardiovascular assessment and, in our hands, result in highly statistically powered studies. The ability of this model to detect cardiovascular changes of small, but biologically relevant, magnitude enables confident decision making around the cardiovascular safety of new chemical entities.
国际协调会议（ICH）的S7A和S7B指南建议，在对新化学实体进行人体试验之前，使用清醒动物评估其非临床心血管安全性。试验方案设计和数据分析技术会影响所产生数据的质量，进而最终影响心血管风险的临床管理。因此，了解可检测到的变化幅度以及这些变化的临床相关性至关重要。本文描述了利用“超级间隔”来分析和解释从我们的清醒遥测犬心血管安全试验方案中获得的数据，并报告了检测各种心血管参数变化所达到的统计效能。
使用来自18项犬遥测研究的心血管数据来计算检测心血管参数变化的统计效能。每项研究均采用受试化合物与赋形剂交叉试验设计，并进行24小时监测（n = 4）。将每只动物的1分钟原始平均数据压缩为每个剂量组的15分钟平均数据以进行可视化。然后选择更大的汇总时间段或“超级间隔”，以在考虑药物药代动力学特征的同时，最好地体现任何观察到的心血管效应，例如给药后1至6小时、7至14小时和14至22小时的间隔。
采用这种方法和研究设计，我们使用中位数百分位数预测，我们的研究有80%的效能检测到以下变化：心率（±10次/分钟）、左心室压力最大上升速率（±375毫米汞柱/秒）、平均动脉压（±5毫米汞柱）和校正QT间期（±4毫秒）。
超级间隔是处理任何动态心血管评估中所见高度自然变异性问题的一种简单方法，在我们的研究中，能得出具有高度统计学效能的研究结果。该模型检测微小但具有生物学相关性的心血管变化的能力，有助于围绕新化学实体的心血管安全性做出可靠决策。
