arXiv subjects

Brian D. Segal

Publications and source records attributed to Brian D. Segal.

4 recordsLinked to original sources

arXiv2021

Augmenting control arms with Real-World Data for cancer trials: Hybrid control arm methods and considerations

Randomized controlled trials (RCTs) are the gold standard for assessing drug safety and efficacy. However, RCTs have some drawbacks which have led to the use of single-arm studies to make certain internal drug development and regulatory decisions, particularly in oncology. Hybrid controlled trials with real-world data (RWD), in which the control arm is composed of both trial and real-world patients, have the potential to help address some of the shortcomings of both RCTs and single-arm studies in particular situations, such as when a disease has low prevalence or when the standard of care to be used in the control arm is ineffective or highly toxic and an experimental therapy shows early promise. This paper discusses why it may be beneficial to consider hybrid controlled trials with RWD, what such a design entails, when it may be appropriate, and how to conduct the analyses. We propose a novel two-step borrowing method for the construction of hybrid control arms. We use simulations to demonstrate the operating characteristics of dynamic and static borrowing methods, and highlight the trade-offs and analytic decisions that study teams will need to address when designing a hybrid study.

stat.ME↗

arXiv2020

A note on the amount of information borrowed from external data in hybrid controlled trials with time-to-event outcomes

In situations where it is difficult to enroll patients in randomized controlled trials, external data can improve efficiency and feasibility. In such cases, adaptive trial designs could be used to decrease enrollment in the control arm of the trial by updating the randomization ratio at the interim analysis. Updating the randomization ratio requires an estimate of the amount of information effectively borrowed from external data, which is typically done with a linear approximation. However, this linear approximation is not always a reliable estimate, which could potentially lead to sub-optimal randomization ratio updates. In this note, we highlight this issue through simulations for exponential time-to-event outcomes, because in this simple setting there is an exact solution available for comparison. We also propose a potential generalization that could complement the linear approximation in more complex settings, discuss challenges for this generalization, and recommend best practices for computing and interpreting estimates of the effective number of events borrowed.

stat.ME↗

arXiv2019

Towards replicability with confidence intervals for the exceedance probability

Several scientific fields including psychology are undergoing a replication crisis. There are many reasons for this problem, one of which is a misuse of p-values. There are several alternatives to p-values, and in this paper we describe a complement that is geared towards replication. In particular, we focus on confidence intervals for the probability that a parameter estimate will exceed a specified value in an exact replication study. These intervals convey uncertainty in a way that p-values and standard confidence intervals do not, and can help researchers to draw sounder scientific conclusions. After briefly reviewing background on p-values and a few alternatives, we describe our approach and provide examples with simulated and real data. For linear models, we also describe how confidence intervals for the exceedance probability are related to p-values and confidence intervals for parameters.

stat.ME↗

arXiv2018

P-splines with an l1 penalty for repeated measures

P-splines are penalized B-splines, in which finite order differences in coefficients are typically penalized with an $\ell_2$ norm. P-splines can be used for semiparametric regression and can include random effects to account for within-subject variability. In addition to $\ell_2$ penalties, $\ell_1$-type penalties have been used in nonparametric and semiparametric regression to achieve greater flexibility, such as in locally adaptive regression splines, $\ell_1$ trend filtering, and the fused lasso additive model. However, there has been less focus on using $\ell_1$ penalties in P-splines, particularly for estimating conditional means. In this paper, we demonstrate the potential benefits of using an $\ell_1$ penalty in P-splines with an emphasis on fitting non-smooth functions. We propose an estimation procedure using the alternating direction method of multipliers and cross validation, and provide degrees of freedom and approximate confidence bands based on a ridge approximation to the $\ell_1$ penalized fit. We also demonstrate potential uses through simulations and an application to electrodermal activity data collected as part of a stress study.

stat.ME↗