文献速递临床试验与 GCP

数据安全监查委员会(DSMB)分享期中结果的职业观点调查

Survey of professional views on sharing interim results by the Data Safety Monitoring Board (DSMB): what to share, with whom and why.
PubMed 文献 · Trials · 2018-05-21

该研究于2015年对SCT和ISCB成员进行在线调查,了解专业人士对DSMB期中结果分享内容的看法。应答率12%(371/3136)。多数受访者认为期中对照事件率(IControlER)、适应性条件功效(ACP)和无条件条件功效(UCP)不应与非DSMB成员分享;多数认为期中合并事件率(ICombinedER)应与指导委员会分享,但作者不建议分享ICombinedER,因其含义模糊可能引发猜测。

要点

  • IControlER和ACP会揭示期中结果,不应与非DSMB成员分享。
  • UCP技术性指标可能误导,也不应分享。
  • 多数受访者认为ICombinedER应与指导委员会分享,但作者不推荐分享。

核心结论

  • IControlER和ACP会揭示期中结果,不应分享;UCP可能误导,也不应分享。
  • 多数受访者认为ICombinedER应与指导委员会分享,但作者不推荐分享,因其含义模糊可引发对组间表现的猜测。

研究设计与样本

  • 2015年对SCT和ISCB成员进行在线调查,邀请3136人(936名SCT成员+2200名ISCB成员),应答率12%(371/3136)。

主要发现

  • 多数受访者报告IControlER不应与非DSMB成员分享(149/237;62.9% [95% CI, 56.7-69.0%])。
  • 多数受访者报告ACP不应与非DSMB成员分享(144/224;64.3% [95% CI, 58.0%-70.6%])。
  • 多数受访者报告UCP不应与非DSMB成员分享(126/208;60.6% [95% CI, 53.9-67.2%])。
  • 多数受访者报告ICombinedER应与非DSMB成员分享,尤其是指导委员会(168/262;64.1% [95% CI, 58.0-69.9%])。

临床怎么用

  • DSMB分享期中结果时,应避免分享IControlER、ACP和UCP,以防揭示期中结果或误导。
  • 即使多数受访者认为ICombinedER可与指导委员会分享,作者建议不分享,因其可能引发对组间表现的猜测。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Sharing interim results by the Data Safety Monitoring Board (DSMB) with non-DSMB members is an issue that can affect trial integrity. It is unclear what should be shared. This study assesses the views of professionals to understand what interim information should be shared at interim, with whom and why. METHODS: Conducted an online survey of members of the Society of Clinical Trials (SCT) and International Society of Clinical Biostatistics (ISCB) in 2015 asking their professional views on sharing interim results. Email was used to advertise the survey and a link in the email was provided to the online survey. RESULTS: Approximately 3136 (936 SCT members + 2200 ISCB members) members were invited. The response rate was 12% (371/3136). The majority reported the Interim Control Event Rate (IControlER) (149/237; 62.9% [95% CI, 56.7-69.0%]), Adaptive Conditional Power (ACP) (144/224; 64.3% [95% CI, 58.0%-70.6%]) and the Unconditional Conditional Power (UCP) (126/208; 60.6% [95% CI, 53.9-67.2%]) should not be shared with non-DSMB members. The majority reported that the Interim Combined Event Rate (ICombinedER) (168/262; 64.1% [95% CI, 58.0-69.9%]) should be shared with non-DSMB members particularly the steering committee (SC) because it does not unmask interim results and helps with monitoring trial progress, safety, and design assumptions. CONCLUSION: The IControlER and ACP are unmasking of interim results and should not be shared. The UCP is a technical measure that is potentially misleading and also should not be shared. The ICombinedER is usually known by the SC and sponsor making it easy to determine group rates if the IControlER is known. Though most respondents thought the ICombinedER should be shared with the SC as it does not unmask relative effects between groups, we do not recommend sharing the ICombinedER as it is flawed measure that can have multiple interpretations possibly suggesting that one group is performing better, worse or the same as a comparator group, leading to guesses about how groups are doing relative to one another.

来源:Trials,查看原文。版权归原作者所有。

引用

  1. Borg Debono V, Mbuagbaw L, Paul J, et al. Survey of professional views on sharing interim results by the Data Safety Monitoring Board (DSMB): what to share, with whom and why. Trials. 2018. PMID: 29784030. doi:10.1186/s13063-018-2655-y

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。