Statistical Methods in Medical Research最新文献

Sequential trial emulation (STE) is an approach to estimating causal treatment effects by emulating a sequence of target trials from observational data. In STE, inverse probability weighting is commonly utilised to address time-varying confounding and/or dependent censoring. Then structural models for potential outcomes are applied to the weighted data to estimate treatment effects. For inference, the simple sandwich variance estimator is popular but conservative, while nonparametric bootstrap is computationally expensive, and a more efficient alternative, linearised estimating function (LEF) bootstrap, has not been adapted to STE. We evaluated the performance of various methods for constructing confidence intervals (CIs) of marginal risk differences in STE with survival outcomes by comparing the coverage of CIs based on nonparametric/LEF bootstrap, jackknife, and the sandwich variance estimator through simulations. LEF bootstrap CIs demonstrated better coverage than nonparametric bootstrap CIs and sandwich-variance-estimator-based CIs with small/moderate sample sizes, low event rates and low treatment prevalence, which were the motivating scenarios for STE. They were less affected by treatment group imbalance and faster to compute than nonparametric bootstrap CIs. With large sample sizes and medium/high event rates, the sandwich-variance-estimator-based CIs had the best coverage and were the fastest to compute. These findings offer guidance in constructing CIs in causal survival analysis using STE.

In pharmacokinetic (PK) bioequivalence (BE) analysis, the recommended approach is the two one-sided tests (TOSTs) on non-compartmental analysis (NCA) estimates of area under the plasma drug concentration versus time curve and $C_{max}$ (NCA-TOST). Sample size estimation for a BE study requires assumptions on between/within subject variability (B/WSV). When little prior information is available, interim analysis using two-stage group sequential (GS) or adaptive designs (ADs) may be beneficial. GS fixes the second stage size, while AD requires sample re-estimation based on first-stage results. Recent research has proposed model-based (MB) TOST, using nonlinear mixed effects models, as an alternative to NCA-TOST. This work extends GS and AD approaches to MB-TOST. We evaluated these approaches on simulated parallel and two-way crossover designs for a one-compartment PK model, considering three variability levels for initial sample size calculation. We compared final sample size, type I error, and power estimates from one-stage, GS, and AD designs using NCA-TOST and MB-TOST. Results showed both NCA-TOST and MB-TOST reasonably controlled type I error while maintaining adequate power in two-stage GS and AD approaches, based on our limited computation power. Two-stage designs reduced sample size compared to traditional designs, especially for highly variable drugs, with many trials stopping at Stage 1 in AD designs. Our findings suggest MB-TOST may serve as a viable alternative to NCA-TOST for BE assessment in two-stage designs, especially when B/WSV impacts BE results.

Many randomized trials have used overall survival as the primary endpoint for establishing non-inferiority of one treatment compared to another. However, if a treatment is non-inferior to another treatment in terms of overall survival, clinicians may be interested in further exploring which treatment results in better health utility scores for patients. Examining health utility in a secondary analysis is feasible, however, since health utility is not the primary endpoint, it is usually not considered in the sample size calculation, hence the power to detect a difference of health utility is not guaranteed. Furthermore, often the premise of non-inferiority trials is to test the assumption that an intervention provides superior quality of life or toxicity profile without compromising survival when compared to the existing standard. Based on this consideration, it may be beneficial to consider both survival and utility when designing a trial. There have been methods that can combine survival and quality of life into a single measure, but they either have strong restrictions or lack theoretical frameworks. In this manuscript, we propose a method called health utility adjusted survival, which can combine survival outcome and longitudinal utility measures for treatment comparison. We propose an innovative statistical framework as well as procedures to conduct power analysis and sample size calculation. By comprehensive simulation studies involving summary statistics from the PET-NECK trial, we demonstrate that our new approach can achieve superior power performance using relatively small sample sizes, and our composite endpoint can be considered as an alternative to overall survival in future clinical trial design and analysis where both survival and health utility are of interest.

The use of external data in clinical trials offers numerous advantages, such as reducing enrollment, increasing study power, and shortening trial duration. In Bayesian inference, information in external data can be transferred into an informative prior for future borrowing (i.e. prior synthesis). However, multisource external data often exhibits heterogeneity, which can cause information distortion during the prior synthesizing. Clustering helps identifying the heterogeneity, enhancing the congruence between synthesized prior and external data. Obtaining optimal clustering is challenging due to the trade-off between congruence with external data and robustness to future data. We introduce two overlapping indices: the overlapping clustering index and the overlapping evidence index . Using these indices alongside a K-means algorithm, the optimal clustering result can be identified by balancing this trade-off and applied to construct a prior synthesis framework to effectively borrow information from multisource external data. By incorporating the (robust) meta-analytic predictive (MAP) prior within this framework, we develop (robust) Bayesian clustering MAP priors. Simulation studies and real-data analysis demonstrate their advantages over commonly used priors in the presence of heterogeneity. Since the Bayesian clustering priors are constructed without needing the data from prospective study, they can be applied to both study design and data analysis in clinical trials.

The integration of backfill cohorts into Phase I clinical trials has garnered increasing interest within the clinical community, particularly following the "Project Optimus" initiative by the U.S. Food and Drug Administration, as detailed in their final guidance of August 2024. This approach allows for the collection of additional clinical data to assess safety and activity before initiating trials that compare multiple dosages. For novel cancer treatments such as targeted therapies, immunotherapies, antibody-drug conjugates, and chimeric antigen receptor T-cell therapies, the efficacy of a drug may not necessarily increase with dose levels. Backfill strategies are especially beneficial as they enable the continuation of patient enrollment at lower doses while higher doses are being explored. We propose a robust Bayesian design framework that borrows information across dose levels without imposing stringent parametric assumptions on dose-response curves. This framework minimizes the risk of administering subtherapeutic doses by jointly evaluating toxicity and efficacy, and by effectively addressing the challenge of delayed outcomes. Simulation studies demonstrate that our design not only generates additional data for late stage studies but also enhances the accuracy of optimal dose selection, improves patient safety, reduces the number of patients receiving subtherapeutic doses, and shortens trial duration across various realistic trial settings.

Covariate-adjusted response adaptive (CARA) designs are effective in increasing the expected number of patients receiving superior treatment in an ongoing clinical trial, given a patient's covariate profile. There has recently been extensive research on CARA designs with parametric distributional assumptions on patient responses. However, the range of applications for such designs becomes limited in real clinical trials. Sverdlov et al. have pointed out that irrespective of a specific parametric form of the survival outcomes, their proposed CARA designs based on the exponential model provide valid statistical inference, provided the final analysis is performed using the appropriate accelerated failure time (AFT) model. In real survival trials, however, the planned primary analysis is rarely conducted using an AFT model. The proposed CARA designs are developed obviating any distributional assumptions about the survival responses, relying only on the proportional hazards assumption between the two treatment arms. To meet the multiple experimental objectives of a clinical trial, the proposed designs are developed based on an optimal allocation approach. The covariate-adjusted doubly adaptive biased coin design and the covariate-adjusted efficient-randomized adaptive design are used to randomize the patients to achieve the derived targets on expectation. These expected targets are functions of the Cox regression coefficients that are estimated sequentially with the arrival of every new patient into the trial. The merits of the proposed designs are assessed using extensive simulation studies of their operating characteristics and then have been implemented to re-design a real-life confirmatory clinical trial.

We investigate the optimal allocation design for response adaptive clinical trials, under the average reward criterion. The treatment randomization process is formatted as a Markov decision process and the Bayesian method is used to summarize the information on treatment effects. A span-contraction operator is introduced and the average reward generated by the policy identified by the operator is shown to converge to the optimal value. We propose an algorithm to approximate the optimal treatment allocation using the Thompson sampling and the contraction operator. For the scenario of two treatments with binary responses and a sample size of 200 patients, simulation results demonstrate efficient learning features of the proposed method. It allocates a high proportion of patients to the better treatment while retaining a good statistical power and having a small probability for a trial going in the undesired direction. When the difference in success probability to detect is 0.2, the probability for a trial going in the unfavorable direction is < 1.5%, which decreases further to < 0.9% when the difference to detect is 0.3. For normally distribution responses, with a sample size of 100 patients, the proposed method assigns 13% more patients to the better treatment than the traditional complete randomization in detecting an effect size of difference 0.8, with a good statistical power and a < 0.7% probability for the trial to go in the undesired direction.

Competing risks models can involve more than one time scale. A relevant example is the study of mortality after a cancer diagnosis, where time since diagnosis but also age may jointly determine the hazards of death due to different causes. Multiple time scales have rarely been explored in the context of competing events. Here, we propose a model in which the cause-specific hazards vary smoothly over two times scales. It is estimated by two-dimensional $P$-splines, exploiting the equivalence between hazard smoothing and Poisson regression. The data are arranged on a grid so that we can make use of generalised linear array models for efficient computations. The R-package TwoTimeScales implements the model. As a motivating example we analyse mortality after diagnosis of breast cancer and we distinguish between death due to breast cancer and all other causes of death. The time scales are age and time since diagnosis. We use data from the Surveillance, Epidemiology and End Results (SEER) program. In the SEER data, age at diagnosis is provided with a last open-ended category, leading to coarsely grouped data. We use the two-dimensional penalised composite link model to ungroup the data before applying the competing risks model with two time scales.

Balancing influential covariates is crucial for valid treatment comparisons in clinical studies. While covariate-adaptive randomization is commonly used to achieve balance, its performance can be inadequate when the number of baseline covariates is large. It is, therefore, essential to identify the influential factors associated with the outcome and ensure balance among these critical covariates. In this article, we propose a novel adaptive randomization approach that integrates the patients' responses and covariates information to select sequentially significant covariates and maintain their balance. We establish theoretically the consistency of our covariate selection method and demonstrate that the improved covariate balancing, as evidenced by a faster convergence rate of the imbalance measure, leads to higher efficiency in estimating treatment effects. Furthermore, we provide extensive numerical and empirical studies to illustrate the benefits of our proposed method across various settings.

Background: There is emerging evidence of the increasing uptake of response-adaptive randomization (RAR) in clinical trials. However, a systematic review of RAR trials, their context of use, characteristics, and stakeholder acceptance has been lacking. Methods: We performed a systematic review of clinical trials that utilized elements of RAR, identified via the Cortellis Regulatory Intelligence database following a pre-specified selection process. We report a summary of relevant characteristics of the identified trials. Results: Out of 170 records, 39 RAR trials were identified (22 completed, 17 ongoing as of October 2024). The majority were Phase 2-focused studies (phases 1/2, 2, 2b, and 2/3), academically sponsored, and concentrated in oncology, neurology, and infectious diseases. Small molecules and biologics were the most common investigational products. Among the 22 completed trials, seven reported positive outcomes. Notably, two of these trials provided pivotal data that informed the further development and subsequent regulatory approval of the investigational compounds. Conclusion: Over the past two decades, RAR has been increasingly utilized in complex adaptive trials across diverse therapeutic areas and clinical research phases. This systematic review provides a critical "baseline" for tracing the dynamics of RAR applications and should help the clinical research community recognize RAR as a valuable methodology for optimizing future trial designs.