Statistical Methods in Medical Research最新文献

Binary endpoints are used widely to evaluate treatment effects during clinical trials. Although clinical trials in many therapeutic areas evaluate a single binary endpoint as the primary endpoint, clinical trials in certain therapeutic areas require two co-primary binary endpoints to evaluate treatment benefit multi-dimensionally. We consider the situation in which evidence of effects on both co-primary endpoints is necessary to conclude that the intervention is effective, which differs from approaches by which significance on at least one endpoint is sufficient for trial success. When designing clinical trials with two co-primary binary endpoints, consideration of correlation between the endpoints can increase trial power and consequently reduce the required sample size, leading to improved efficiency. For clinical trials with two co-primary binary endpoints, methods for calculating power and sample size have been proposed, but they are based on approximations or require Monte Carlo integration. Alternatively, we propose methods for calculating the exact power and sample size in clinical trials with two co-primary binary endpoints. The proposed methods are useful for any statistical test for binary endpoints. Numerical investigation under various scenarios demonstrated that our proposed methods can incorporate consideration of the correlation between two co-primary binary endpoints in sample size calculation, thereby allowing the required sample size to be reduced. We demonstrate that the exact power for the required sample size calculated using our proposed method is approximately equal to target power. Finally, we present application of our proposed methods to a clinical trial of relapsing or refractory eosinophilic granulomatosis with polyangiitis.

This study describes and compares the performance of several semi-parametric statistical modeling approaches to dynamically classify subjects into two groups, based on an irregularly and sparsely sampled curve. The motivating example of this study is the diagnosis of a complication following cardiac surgery, based on repeated measures of a single cardiac biomarker where early detection enables prompt intervention by clinicians. We first simulate data to compare the dynamic predictive performance over time for growth charts, conditional growth charts, a varying-coefficient model, a generalized functional linear model and longitudinal discriminant analysis. Our results demonstrate that functional regression approaches that implicitly incorporate historic information through random effects, provide superior discriminative ability compared to approaches that do not take historic information into account or explicitly model historic information through autoregressive terms. Semi-parametric modeling approaches show a benefit in terms of dynamic discriminative ability compared to the clinical practice of using a fixed threshold on the raw measured value. Under high degrees of sparsity the functional regression approaches are less advantageous compared to varying-coefficient models or quantile regression. The class imbalance of the outcome affects the historic and non-historic approaches in equal measure, with lower event rates reducing performance. Finally, the functional regression and varying-coefficient model were applied to a real-world clinical dataset to demonstrate their performance and application.

In clinical trials, it is often of interest to know whether treatment works differently for some groups than others, known as heterogeneity of treatment effect. Such subgroup analysis is complicated to conduct because trials are typically not powered to find subgroups. Furthermore, it is difficult to identify characteristics of patients pertaining to such subgroups. In this article, we propose a semiparametric mixture model to identify subgroups with time-to-event outcomes. Specifically, we assume a proportional hazards model with subgroup-specific piecewise constant baseline hazards, where the subgroup-specific treatment effect is assumed to be the same within each subgroup. The probability of belonging to a certain subgroup is a function of patient prognostic factors. Adopting a Bayesian approach, classification uncertainty is taken into account. We demonstrate the utility of our approach via simulation and an application to data from a real clinical trial in HIV research.

Although response-adaptive randomisation (RAR) has gained substantial attention in the literature, it still has limited use in clinical trials. Amongst other reasons, the implementation of RAR in real world trials raises important practical questions, often neglected in the technical literature. Motivated by an innovative phase-II stratified RAR rare-disease trial, this paper addresses two challenges: (1) How to ensure that RAR allocations are desirable, that is, both acceptable and faithful to the intended probabilities, particularly in small samples? and (2) What adaptations to trigger after interim analyses in the presence of missing data? To answer (1), we propose a Mapping strategy that discretises the randomisation probabilities into a vector of allocation ratios, resulting in improved frequentist errors. Under the implementation of Mapping, we answer (2) by analysing the impact of missing data on operating characteristics in selected scenarios. Finally, we discuss additional concerns including: pooling data across trial strata, analysing the level of blinding in the trial, and reporting safety results.

The receiver operating characteristic curve is a popular tool for evaluating the discriminative ability of a diagnostic biomarker. Parametric and nonparametric methods exist in the literature for estimation of a receiver operating characteristic curve and its associated summary measures using data usually collected from a case-control study. Since the receiver operating characteristic curve remains unchanged under a monotone transformation, the biomarker data from both cases (diseased subjects) and controls (non-diseased subjects) are often transformed based on a common Box-Cox transformation (or other appropriate transformation) prior to the application of a parametric estimation method. However, careful examination of the data often reveals that the biomarker values in the diseased and non-diseased population can only be normally approximated via different transformations. In this situation, existing estimation methods cannot be directly applied to the heterogeneously-transformed data. In this article, we deal with the situation that biomarker data from both diseased and non-diseased population are normally distributed after being transformed with different Box-Cox transformations. Under this assumption, we show that existing methods based on a common Box-Cox transformation are invalid in that they possess substantial biases. We move on to propose a method to estimate the underlying receiver operating characteristic curve and its area under the curve, and investigate its performance as compared to the nonparametric estimator that ignores any distributional assumptions as well as the estimators based on a common Box-Cox transformation assumptions. The method is exemplified with HIV infection data from the National Health and Nutrition Examination Survey (NHANES).

Driven by evolving Food and Drug Administration recommendations, modern clinical trials demand innovative designs that strike a balance between statistical rigor and ethical considerations. Covariate-adjusted response-adaptive randomization (CARA) designs bridge this gap by utilizing patient attributes and responses to skew treatment allocation in favor of the treatment to be best for an individual patient's profiles. However, existing CARA designs for survival outcomes often rely on specific parametric models, constraining their applicability in clinical practice. To overcome this limitation, we propose a novel CARA method for survival outcomes (called CARAS) based on the Cox model, which improves model flexibility and mitigate risks of model misspecification. Additionally, we introduce a group sequential overlap-weighted log-rank test to preserve the type I error rate in group sequential trials using CARAS. Comprehensive simulation studies and a real-world trial example demonstrate the proposed method's clinical benefit, statistical efficiency, and robustness to model misspecification compared to traditional randomized controlled trial designs and response-adaptive randomization designs.

Despite the widespread use of time-to-event data in precision medicine, existing research has often neglected the presence of the cure fraction, assuming that all individuals will inevitably experience the event of interest. When a cure fraction is present, the cure rate and survival time of uncured patients should be considered in estimating the optimal individualized treatment regimes. In this study, we propose direct methods for estimating the optimal individualized treatment regimes that either maximize the cure rate or mean survival time of uncured patients. Additionally, we propose two optimal individualized treatment regimes that balance the tradeoff between the cure rate and mean survival time of uncured patients based on a constrained estimation framework for a more comprehensive assessment of individualized treatment regimes. This framework allows us to estimate the optimal individualized treatment regime that maximizes the population's cure rate without significantly compromising the mean survival time of those who remain uncured or maximizes the mean survival time of uncured patients while having the cure rate controlled at a desired level. The exterior-point algorithm is adopted to expedite the resolution of the constrained optimization problem and statistical validity is rigorously established. Furthermore, the advantages of the proposed methods are demonstrated via simulations and analysis of esophageal cancer data.

Starting from a Bayesian perspective, this paper proposes a novel response adaptive randomization rule based on the use of the predictive distribution. The intent is to design a randomized mechanism that favors the allocation of the next patient to the "best" treatment, considering the expected future outcomes obtained by combining accrued data with prior information. This predictive rule also stems from a decision-theoretic approach. The method is driven by patients' beneficial motivations, fully debated in this work, but also accounts for essential inferential purposes in clinical trials discussed within the framework of frequentist inference. Some asymptotic properties of the proposed rule are proved and also shown through numerical studies, which compare this method with other competing ones, as the notable Thompson rule for the special case of binary outcomes.

This article discusses regression analysis of interval-censored failure time data in the presence of a cure fraction and nonignorable missing covariates. To address the challenges caused by interval censoring, missing covariates and the existence of a cure subgroup, we propose a joint semiparametric modeling framework that simultaneously models the failure time of interest and the missing covariates. In particular, we present a class of semiparametric nonmixture cure models for the failure time and a semiparametric density ratio model for the missing covariates. A two-step likelihood-based estimation procedure is developed and the large sample properties of the resulting estimators are established. An extensive numerical study demonstrates the good performance of the proposed method in practical settings and the proposed approach is applied to an Alzheimer's disease study that motivated this study.

The optimal designs (ODs) for parallel-arm longitudinal cluster randomized trials, multiple-period cluster randomized crossover (CRXO) trials, and stepped wedge cluster randomized trials (SW-CRTs), including closed-cohort and repeat cross-sectional designs, have been studied separately under a cost-efficiency framework based on generalized estimating equations (GEEs). However, whether a global OD exists across longitudinal designs and randomization schedules remains unknown. Therefore, this research addresses a critical gap by comparing OD feature across complete longitudinal cluster randomized trial designs with two treatment conditions and continuous outcomes. We define the OD as the design with either the lowest cost to obtain a desired level of power or the largest power given a fixed budget. For each of these ODs, we obtain the optimal number of clusters and the optimal cluster-period size (number of participants per cluster per period). To ensure equitable comparisons, we consider the GEE treatment effect estimator with the same block exchangeable correlation structure and develop OD algorithms with the lowest cost for each of six study designs. To obtain OD with the largest power, we summarize the previous and propose new OD algorithms and formulae. We suggest using the number of treatment sequences $L=T-1$, where T is the number of time-periods, in both the optimal closed-cohort and repeated cross-sectional SW-CRTs to have the lowest cost. This is consistent with our previous findings for ODs with the largest power in SW-CRTs. Comparing all six ODs, we conclude that optimal closed-cohort CRXO trials are global ODs, yielding both the lowest cost and largest power.