DXDong Xi

Dong Xi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Dong Xi has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Comparison of Methods for Treatment Selection in Seamless Phase II/III Design

    by Jialuo Liu, Lulu Wang, Dong Xi

    Seamless phase II/III design aims to integrate a phase II trial for treatment selection and a phase III confirmatory trial. It offers valuable flexibility through mid‐trial modifications, potentially optimizing resource utilization and reducing patient burden. In the first stage, multiple experimental treatments are evaluated to select promising ones, which enter the second stage for the confirmatory analysis against the control using data from both stages. Proper statistical methods are needed for the final analysis to control the overall Type I error rate at a prespecified level, regardless of the treatment selection rule used at the interim analysis. In this article, we provide a comprehensive evaluation of four classes of statistical approaches in the literature, which use different ways to integrate combination functions (or conditional error functions) and multiple testing methods (including Bonferroni, Simes, and Dunnett adjustments). Extensive simulation studies are performed to evaluate both the Type I error control and power. In addition, we illustrate the practical implementation of these approaches in real clinical trial settings.

  • Futility Analyses for the MCP ‐Mod Methodology Based on Longitudinal Models

    by Björn Bornkamp, Jie Zhou, Dong Xi, Weihua Cao

    This article discusses futility analyses for the MCP‐Mod methodology. Formulas are derived for calculating predictive and conditional power for MCP‐Mod, which also cover the case when longitudinal models are used allowing to utilize incomplete data from patients at interim. A simulation study is conducted to evaluate the repeated sampling properties of the proposed decision rules and to assess the benefit of using a longitudinal versus a completer only model for decision making at interim. The results suggest that the proposed methods perform adequately and a longitudinal analysis outperforms a completer only analysis, particularly when the recruitment speed is higher and the correlation over time is larger. The proposed methodology is illustrated using real data from a dose‐finding study for severe uncontrolled asthma.

  • Open Access

    Closed MCP‐Mod for Pairwise Comparisons of Several Doses With a Control

    by Franz Koenig, Sergey Krasnozhon, Bjoern Bornkamp, Frank Bretz, Ekkehard Glimm, Alexandra Graf, Dong Xi

    The MCP‐Mod approach by Bretz et al. is commonly applied for dose–response testing and estimation in clinical trials. The MCP part of MCP‐Mod was originally developed to detect a dose–response signal using a multiple contrast test, but it is not appropriate to make a specific claim that the drug has a positive effect at an individual dose. In this paper, we extend the MCP‐Mod approach to obtain confirmatory p‐values for detecting a dose–response signal as well as for the pairwise comparisons of the individual doses against placebo. We apply the closed test principle from Marcus et al. to the optimal contrast tests based on a candidate set of plausible dose–response shapes available at the planning stage of a clinical trial. We show that the contrast coefficients have to be optimized under suitable constraints to guarantee strong Type 1 error rate control at a pre‐specified significance level. Motivated by a recent clinical trial, we evaluate the operating characteristics of the proposed methods in a comprehensive simulation study.


  • Success and Futility Criteria for Accelerated Approval of Oncology Drugs

    by Dong Xi, Jiangtao Gou

    Project FrontRunner encourages development of cancer drugs for advanced or metastatic disease in an earlier clinical setting by promoting regulatory approaches such as the accelerated approval pathway. The FDA draft guideline proposes a one‐trial approach to combine accelerated approval and regular approval in a single trial to maintain efficiency. This article describes our idea of controlling Type I error for accelerated and regular approvals in the one‐trial approach. We introduce success and futility boundaries on p‐values for accelerated approval to create three outcomes: success, RA, and futility. If success, accelerated approval can be claimed for; for RA, only regular approval (RA) is considered; if futility, we stop the trial early for futility. For both success and RA, the endpoint for regular approval can be tested with no penalty on its significance level. The proposed approach is robust to all possible values of correlation between test statistics of the endpoints for accelerated and regular approvals. This framework is flexible to allow clinical trial teams to tailor success and futility boundaries to meet clinical and regulatory needs, while maintaining the overall Type I error control in the strong sense.


  • Open Access

    Testing One Primary and Two Secondary Endpoints in a Two‐Stage Group Sequential Trial With Extensions

    by Ajit C. Tamhane, Dong Xi, Cyrus R. Mehta, Alexander Romanenko, Jiangtao Gou

    We study the problem of testing multiple secondary endpoints conditional on a primary endpoint being significant in a two‐stage group sequential procedure, focusing on two secondary endpoints. This extends our previous work with one secondary endpoint. The test for the secondary null hypotheses is a closed procedure. Application of the Bonferroni test for testing the intersection of the secondary hypotheses results in the Holm procedure while application of the Simes test results in the Hochberg procedure. The focus of the present paper is on developing normal theory analogs of the abovementioned ‐value based tests that take into account (i) the gatekeeping effect of the test on the primary endpoint and (ii) correlations between the endpoints. The normal theory boundaries are determined by finding the least favorable configuration of the correlations and so their knowledge is not needed to apply these procedures. The ‐value based procedures are easy to apply but they are less powerful than their normal theory analogs because they do not take into account the correlations between the endpoints and the gatekeeping effect referred to above. On the other hand, the normal theory procedures are restricted to two secondary endpoints and two stages mainly because of computational difficulties with more than two secondary endpoints and stages. Comparisons between the two types of procedures are given in terms of secondary powers. The sensitivity of the secondary type I error rate and power to unequal information times is studied. Numerical examples and a real case study illustrate the procedures.


  • Optimal weighted Bonferroni tests and their graphical extensions

    by Dong Xi, Yao Chen

    Regulatory guidelines mandate the strong control of the familywise error rate in confirmatory clinical trials with primary and secondary objectives. Bonferroni tests are one of the popular choices for multiple comparison procedures and are building blocks of more advanced procedures. It is usually of interest to find the optimal weighted Bonferroni split for multiple hypotheses. We consider two popular quantities as the optimization objectives, which are the disjunctive power and the conjunctive power. The former is the probability to reject at least one false hypothesis and the latter is the probability to reject all false hypotheses. We investigate the behavior of each of them as a function of different Bonferroni splits, given assumptions about the alternative hypotheses and correlations between test statistics. Under independent tests, unique optimal Bonferroni weights exist; under dependence, optimal Bonferroni weights may not be unique based on a fine grid search. In general, we propose an optimization algorithm based on constrained nonlinear optimization and multiple starting points. The proposed algorithm efficiently identifies optimal Bonferroni weights to maximize the disjunctive or conjunctive power. In addition, we apply the proposed algorithm to graphical approaches, which include many Bonferroni‐based multiple comparison procedures. Utilizing the closed testing principle, we adopt a two‐step approach to find optimal graphs using the disjunctive power. We also identify a class of closed test procedures that optimize the conjunctive power. We apply the proposed algorithm to a case study to illustrate the utility of optimal graphical approaches that reflect study objectives.


  • Causal discoveries for high dimensional mixed data

    by Zhanrui Cai, Dong Xi, Xuan Zhu, Runze Li

    Causal relationships are of crucial importance for biological and medical research. Algorithms have been proposed for causal structure learning with graphical visualizations. While much of the literature focuses on biological studies where data often follow the same distribution, for example, the normal distribution for all variables, challenges emerge from epidemiological and clinical studies where data are often mixed with continuous, binary, and ordinal variables. We propose to use a mixed latent Gaussian copula model to estimate the underlying correlation structure via the rank correlation for mixed data. This correlation structure is then incorporated into a popular causal discovery algorithm, the PC algorithm, to identify causal structures. The proposed algorithm, called the latent-PC algorithm, is able to discover the true causal structure consistently under mild conditions in high dimensional settings. From simulation studies, the latent-PC algorithm delivers a competitive performance in terms of a similar or higher true positive rate and a similar or lower false positive rate, compared with other variants of the PC algorithm. In the high dimensional settings where the number of variables is more than the number of observations, the causal graphs identified by the latent-PC algorithm are closer to the true causal structures, compared to other competing algorithms. Further, we demonstrate the utility of the latent-PC algorithm in a real dataset for hepatocellular carcinoma. Causal structures for patient survival are visualized and connected with clinical interpretations in the literature.


  • Challenges in Assessing the Impact of the COVID-19 Pandemic on the Integrity and Interpretability of Clinical Trials

    by Mouna Akacha, Janice Branson, Frank Bretz, Bharani Dharan, Paul Gallo, Insa Gathmann, Robert Hemmings, Julie Jones, Dong Xi, Emmanuel Zuber

    The COVID-19 pandemic has a global impact on the conduct of clinical trials of medical products. This paper discusses implications of the COVID-19 pandemic on clinical research methodology aspects and provides points to consider in order to assess and mitigate the risk of seriously compromising the integrity and interpretability of clinical trials. The information in this paper will support discussions that need to occur cross-functionally on an ongoing basis in order to “integrate all available knowledge from the ethical, the medical, and the methodological perspective into decision making” (EMA 2020b). This paper aims at facilitating: (i) risk assessments of the impact of the pandemic on trial integrity and interpretability; (ii) identification of the relevant data and information related to the impact of the pandemic on the trial that needs to be collected; (iii) short-term decision making impacting ongoing trial operations; (iv) ongoing monitoring of the trial conduct until completion, including the possible involvement of Data Monitoring Committees, and adequately documenting all measures taken to secure trial integrity throughout and after the pandemic, and (v) proper analysis and interpretation of the eventual interim or final trial data.


  • Symmetric graphs for equally weighted tests, with application to the Hochberg procedure

    by Dong Xi, Frank Bretz

    The graphical approach to multiple testing provides a convenient tool for designing, visualizing, and performing multiplicity adjustments in confirmatory clinical trials while controlling the familywise error rate. It assigns a set of weights to each intersection null hypothesis within the closed test framework. These weights form the basis for intersection tests using weighted individual p‐values, such as the weighted Bonferroni test. In this paper, we extend the graphical approach to intersection tests that assume equal weights for the elementary null hypotheses associated with any intersection hypothesis, including the Hochberg procedure as well as omnibus tests such as Fisher's combination, O'Brien's, and F tests. More specifically, we introduce symmetric graphs that generate sets of equal weights so that the aforementioned tests can be applied with the graphical approach. In addition, we visualize the Hochberg and the truncated Hochberg procedures in serial and parallel gatekeeping settings using symmetric component graphs. We illustrate the method with two clinical trial examples.