Prentice's Approach and the Meta‐Analytic Paradigm: A Reflection on the Role of Statistics in the Evaluation of Surrogate Endpoints

Alonso, Ariel; Molenberghs, Geert; Burzykowski, Tomasz; Renard, Didier; Geys, Helena; Shkedy, Ziv; Tibaldi, Fabián; Abrahantes, José Cortiñas; Buyse, Marc

doi:10.1111/j.0006-341x.2004.00222.x

Cited by 51 publications

(76 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Prentice criteria are stringent and the evaluation of surrogate endpoints cannot be assessed using only statistical evidence, but should also consider clinical and biological outcomes. 11 One of the greatest advances in cancer research in the last 2 decades has been the demonstration that infection with certain types of HPV is a necessary cause of cervical cancer. 12,13 Using surrogate endpoints in the development of prophylactic vaccines is necessary, as the use of a cervical cancer as endpoint is unethical.…”

Section: Discussionmentioning

confidence: 99%

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Radley

Saah

Stanley

2015

Human Vaccines & Immunotherapeutics

View full text Add to dashboard Cite

We investigated the relationship between high-grade cervical disease (cervical intraepithelial neoplasia [CIN] 2, CIN3 or adenocarcinoma in situ) and persistent infection with HPV16 and/or HPV18 (HPV16/18) among 3970 women who received placebo in 3 clinical trials of a quadrivalent HPV vaccine. Statistical analysis (odds ratios, sensitivity, specificity, negative and positive predictive values, negative and positive likelihood ratios) showed that patients with a persistent infection with HPV16/18 had a much greater risk of HPV16/18-related high-grade cervical disease. Furthermore, subjects without a persistent infection with HPV16/18 were unlikely to have HPV16/18-related high-grade cervical disease. These results suggest that persistent infection with HPV16/18 meets the criteria for a surrogate endpoint for HPV16/18-related highgrade cervical disease and may be used as such in future clinical studies with prophylactic HPV vaccines and in natural history studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Radley

Saah

Stanley

2015

Human Vaccines & Immunotherapeutics

View full text Add to dashboard Cite

show abstract

“…Consequently, the estimates of the mean structure parameters are affected, which in turn influences the trial-specific treatment effects and the estimation of the trial-level association in MFE approach. It is also important to note that Cortiñas et al (2004) have also observed similar problems, when they fitted a hierarchical model ignoring a level, and the variance associated to the level was of the same magnitude or larger than the variance at the higher level, the association at the higher level was affected. Here a similar pattern can be observed.…”

Section: Mfe Srte Ritementioning

confidence: 85%

“…The adjusted association carries over when data are available on several randomized trials, while the RE can be extended to a trial-level measure of agreement between the effects of treatment of both endpoints. Molenberghs et al (2002) and Alonso et al (2004) pointed out serious issues surrounding the Prentice-Freedman framework. It has been asserted that the criteria set out by Prentice are too stringent (Fleming and DeMets, 1996) and neither necessary nor sufficient for his definition to be fulfilled, except in the special case of binary outcomes (Buyse and Molenberghs, 1998).…”

Section: Introductionmentioning

confidence: 99%

Simplified modeling strategies for surrogate validation with multivariate failure-time data

Abrahantes

Burzykowski

2010

Computational Statistics & Data Analysis

Self Cite

View full text Add to dashboard Cite

The linear mixed effects model has become a standard tool for the analysis of continuous hierarchical data such as, for example, repeated measures or data from meta-analyses. However, in certain situations the model does pose unavoidable computational problems. In the context of surrogate markers, this problem has appeared when using an estimation and prediction-based approach for evaluation of surrogate endpoints. Convergence problems can occur mainly due to small between-trial variability or small number of trials. A number of alternative strategies has been proposed and studied for normally distributed data, but not such study have been conducted for other type of endpoints. The idea is to study if such simplified strategies, which always ignore individual level surrogacy, can also be applied when both surrogate and true endpoints are of failure-time types. It is shown via simulations that the 3 simplified strategies produced biased estimates, especially for the cases in which the strength of individual-level association is different from the strength of trial-level association. For this reason, it is recommended not to use simplified strategies when dealing with failure time data, in contrast to the case of normally distributed data, for which simplified strategies are recommended. Possible reasons for this discrepancy might be that, in this case, ignoring the individual level association influences estimates of the mean structure parameters, what results in distorted estimates of the trial level association.

show abstract

“…Alonso and colleagues [28] introduced the likelihood reduction factor (LRF) for general endpoints, to replace R 2 -type measures specific to normal, survival, or discrete endpoints at the patient level. Later, Alonso and Molenberghs [29] proposed an information-theoretic approach to surrogate endpoint evaluation at both individual and trial levels within a meta-analytic setting, designed to accommodate general endpoint types.…”

Section: Discussionmentioning

confidence: 99%

Bayesian adjusted R² for the meta‐analytic evaluation of surrogate time‐to‐event endpoints in clinical trials

Renfro

Sargent

Carlin

2011

Statistics in Medicine

View full text Add to dashboard Cite

Burzykowski et al. introduced a two-stage model to evaluate both trial and patient level surrogacy of correlated timeto-event endpoints using patient level data when multiple clinical trials are available. However, their maximum likelihood approach often suffers from numerical problems when different baseline hazards among trials and imperfect estimation of treatment effects are assumed. We propose performing the second-stage, trial-level evaluation of potential surrogates within a Bayesian framework, where we may naturally borrow information across trials while maintaining these realistic assumptions. Posterior distributions on surrogacy measures of interest may then be used to compare measures or make decisions regarding the candidacy of a specific endpoint. We perform a simulation study to investigate differences in estimation performance between traditional maximum likelihood and new Bayesian representations of common metaanalytic surrogacy measures, while assessing sensitivity to data characteristics such as number of trials, trial size, and amount of censoring. Furthermore, we present both frequentist and Bayesian trial-level surrogacy evaluations of time-torecurrence for overall survival in two meta-analyses of adjuvant therapy trials in colon cancer. Based on these results, we recommend hierarchical Bayesian methods as an attractive alternative in the multi-trial evaluation of potential surrogate endpoints.

show abstract

Prentice's Approach and the Meta‐Analytic Paradigm: A Reflection on the Role of Statistics in the Evaluation of Surrogate Endpoints

Cited by 51 publications

References 14 publications

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Simplified modeling strategies for surrogate validation with multivariate failure-time data

Bayesian adjusted R² for the meta‐analytic evaluation of surrogate time‐to‐event endpoints in clinical trials

Contact Info

Product

Resources

About

Prentice's Approach and the Meta‐Analytic Paradigm: A Reflection on the Role of Statistics in the Evaluation of Surrogate Endpoints

Cited by 51 publications

References 14 publications

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Persistent infection with human papillomavirus 16 or 18 is strongly linked with high-grade cervical disease

Simplified modeling strategies for surrogate validation with multivariate failure-time data

Bayesian adjusted R2 for the meta‐analytic evaluation of surrogate time‐to‐event endpoints in clinical trials

Contact Info

Product

Resources

About

Bayesian adjusted R² for the meta‐analytic evaluation of surrogate time‐to‐event endpoints in clinical trials