Bayesian pairwise meta-analysis of time-to-event outcomes in the presence of non-proportional hazards: A simulation study of flexible parametric, piecewise exponential and fractional polynomial models

IF 5 2区 生物学 Q1 MATHEMATICAL & COMPUTATIONAL BIOLOGY
Suzanne C. Freeman, Alex J. Sutton, Nicola J. Cooper, Alessandro Gasparini, Michael J. Crowther, Neil Hawkins
{"title":"Bayesian pairwise meta-analysis of time-to-event outcomes in the presence of non-proportional hazards: A simulation study of flexible parametric, piecewise exponential and fractional polynomial models","authors":"Suzanne C. Freeman,&nbsp;Alex J. Sutton,&nbsp;Nicola J. Cooper,&nbsp;Alessandro Gasparini,&nbsp;Michael J. Crowther,&nbsp;Neil Hawkins","doi":"10.1002/jrsm.1722","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Background</h3>\n \n <p>Traditionally, meta-analysis of time-to-event outcomes reports a single pooled hazard ratio assuming proportional hazards (PH). For health technology assessment evaluations, hazard ratios are frequently extrapolated across a lifetime horizon. However, when treatment effects vary over time, an assumption of PH is not always valid. The Royston-Parmar (RP), piecewise exponential (PE), and fractional polynomial (FP) models can accommodate non-PH and provide plausible extrapolations of survival curves beyond observed data.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>Simulation study to assess and compare the performance of RP, PE, and FP models in a Bayesian framework estimating restricted mean survival time difference (RMSTD) at 50 years from a pairwise meta-analysis with evidence of non-PH. Individual patient data were generated from a mixture Weibull distribution. Twelve scenarios were considered varying the amount of follow-up data, number of trials in a meta-analysis, non-PH interaction coefficient, and prior distributions. Performance was assessed through bias and mean squared error. Models were applied to a metastatic breast cancer example.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>FP models performed best when the non-PH interaction coefficient was 0.2. RP models performed best in scenarios with complete follow-up data. PE models performed well on average across all scenarios. In the metastatic breast cancer example, RMSTD at 50-years ranged from −14.6 to 8.48 months.</p>\n </section>\n \n <section>\n \n <h3> Conclusions</h3>\n \n <p>Synthesis of time-to-event outcomes and estimation of RMSTD in the presence of non-PH can be challenging and computationally intensive. Different approaches make different assumptions regarding extrapolation and sensitivity analyses varying key assumptions are essential to check the robustness of conclusions to different assumptions for the underlying survival function.</p>\n </section>\n </div>","PeriodicalId":226,"journal":{"name":"Research Synthesis Methods","volume":null,"pages":null},"PeriodicalIF":5.0000,"publicationDate":"2024-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/jrsm.1722","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Research Synthesis Methods","FirstCategoryId":"99","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/jrsm.1722","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICAL & COMPUTATIONAL BIOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Background

Traditionally, meta-analysis of time-to-event outcomes reports a single pooled hazard ratio assuming proportional hazards (PH). For health technology assessment evaluations, hazard ratios are frequently extrapolated across a lifetime horizon. However, when treatment effects vary over time, an assumption of PH is not always valid. The Royston-Parmar (RP), piecewise exponential (PE), and fractional polynomial (FP) models can accommodate non-PH and provide plausible extrapolations of survival curves beyond observed data.

Methods

Simulation study to assess and compare the performance of RP, PE, and FP models in a Bayesian framework estimating restricted mean survival time difference (RMSTD) at 50 years from a pairwise meta-analysis with evidence of non-PH. Individual patient data were generated from a mixture Weibull distribution. Twelve scenarios were considered varying the amount of follow-up data, number of trials in a meta-analysis, non-PH interaction coefficient, and prior distributions. Performance was assessed through bias and mean squared error. Models were applied to a metastatic breast cancer example.

Results

FP models performed best when the non-PH interaction coefficient was 0.2. RP models performed best in scenarios with complete follow-up data. PE models performed well on average across all scenarios. In the metastatic breast cancer example, RMSTD at 50-years ranged from −14.6 to 8.48 months.

Conclusions

Synthesis of time-to-event outcomes and estimation of RMSTD in the presence of non-PH can be challenging and computationally intensive. Different approaches make different assumptions regarding extrapolation and sensitivity analyses varying key assumptions are essential to check the robustness of conclusions to different assumptions for the underlying survival function.

Abstract Image

存在非比例危害的时间到事件结果的贝叶斯成对荟萃分析:灵活参数模型、分指数模型和分数多项式模型的模拟研究。
背景:传统上,时间到事件结果的荟萃分析报告的是假设比例危险(PH)的单一汇总危险比。在卫生技术评估评价中,危害比经常被推断到整个生命周期。然而,当治疗效果随时间变化时,PH 假设并不总是有效的。罗伊斯顿-帕尔马模型(RP)、分项指数模型(PE)和分数多项式模型(FP)可以适应非PH值,并在观察数据之外提供可信的生存曲线外推:模拟研究:在贝叶斯框架中评估和比较 RP、PE 和 FP 模型的性能,从有证据表明非 PH 的配对荟萃分析中估计 50 岁时的受限平均生存时间差 (RMSTD)。单个患者数据由混合 Weibull 分布生成。考虑了 12 种不同的情况,包括随访数据量、荟萃分析中的试验数量、非 PH 交互系数和先验分布。通过偏差和均方误差评估其性能。模型被应用于转移性乳腺癌的实例中:当非 PH 交互系数为 0.2 时,FP 模型表现最佳。RP 模型在具有完整随访数据的情况下表现最佳。PE 模型在所有情况下平均表现良好。以转移性乳腺癌为例,50 年的 RMSTD 为 -14.6 到 8.48 个月不等:时间到事件结果的综合以及在非 PH 情况下 RMSTD 的估算具有挑战性,而且计算量很大。不同的方法会对外推法做出不同的假设,因此必须对不同的关键假设进行敏感性分析,以检查结论在基础生存函数的不同假设下的稳健性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Research Synthesis Methods
Research Synthesis Methods MATHEMATICAL & COMPUTATIONAL BIOLOGYMULTID-MULTIDISCIPLINARY SCIENCES
CiteScore
16.90
自引率
3.10%
发文量
75
期刊介绍: Research Synthesis Methods is a reputable, peer-reviewed journal that focuses on the development and dissemination of methods for conducting systematic research synthesis. Our aim is to advance the knowledge and application of research synthesis methods across various disciplines. Our journal provides a platform for the exchange of ideas and knowledge related to designing, conducting, analyzing, interpreting, reporting, and applying research synthesis. While research synthesis is commonly practiced in the health and social sciences, our journal also welcomes contributions from other fields to enrich the methodologies employed in research synthesis across scientific disciplines. By bridging different disciplines, we aim to foster collaboration and cross-fertilization of ideas, ultimately enhancing the quality and effectiveness of research synthesis methods. Whether you are a researcher, practitioner, or stakeholder involved in research synthesis, our journal strives to offer valuable insights and practical guidance for your work.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信