Jielai Xia

Optimal Caliper Width for Propensity Score Matching of Three Treatment Groups: A Monte Carlo Study

Propensity score matching is a method to reduce bias in non-randomized and observational studies. Propensity score matching is mainly applied to two treatment groups rather than multiple treatment groups, because some key issues affecting its application to multiple treatment groups remain unsolved, such as the matching distance, the assessment of balance in baseline variables, and the choice of optimal caliper width. The primary objective of this study was to compare propensity score matching methods using different calipers and to choose the optimal caliper width for use with three treatment groups. The authors used caliper widths from 0.1 to 0.8 of the pooled standard deviation of the logit of the propensity score, in increments of 0.1. The balance in baseline variables was assessed by standardized difference. The matching ratio, relative bias, and mean squared error (MSE) of the estimate between groups in different propensity score-matched samples were also reported. The results of Monte Carlo simulations indicate that matching using a caliper width of 0.2 of the pooled standard deviation of the logit of the propensity score affords superior performance in the estimation of treatment effects. This study provides practical solutions for the application of propensity score matching of three treatment groups.

Bridging continual reassessment method for phase I clinical trials in different ethnic populations

Accumulating evidence shows that the conventional one-size-fits-all dose-finding paradigm is problematic when applied to different ethnic populations. Because of inter-ethnic heterogeneity, the dosage established in a landmark trial for a certain population may not be generalizable to a different ethnic population, and a follow-up bridge trial is often needed to find the maximum tolerated dose for the new population. We propose the bridging continual reassessment method (B-CRM) to facilitate dose finding for such follow-up bridge trials. The B-CRM borrows information from the landmark trial through a novel estimate of the dose-toxicity curve and accommodates the inter-ethnic heterogeneity using the Bayesian model averaging approach. Extensive simulation studies show that the B-CRM has desirable operating characteristics with a high probability to select the target dose. This article focuses on ethnic heterogeneity, but the proposed method can be directly used to handle other types of patient heterogeneity, for example, patient subgroups defined by prognostic factors or biomarkers. The software to implement the B-CRM design is available for free download at http://odin.mdacc.tmc.edu/~yyuan/.

A web-based quantitative signal detection system on adverse drug reaction in China

ObjectiveTo establish a web-based quantitative signal detection system for adverse drug reactions (ADRs) based on spontaneous reporting to the Guangdong province drug-monitoring database in China.MethodsUsing Microsoft Visual Basic and Active Server Pages programming languages and SQL Server 2000, a web-based system with three software modules was programmed to perform data preparation and association detection, and to generate reports. Information component (IC), the internationally recognized measure of disproportionality for quantitative signal detection, was integrated into the system, and its capacity for signal detection was tested with ADR reports collected from 1 January 2002 to 30 June 2007 in Guangdong.ResultsA total of 2,496 associations including known signals were mined from the test database. Signals (e.g., cefradine-induced hematuria) were found early by using the IC analysis. In addition, 291 drug-ADR associations were alerted for the first time in the second quarter of 2007.ConclusionsThe system can be used for the detection of significant associations from the Guangdong drug-monitoring database and could be an extremely useful adjunct to the expert assessment of very large numbers of spontaneously reported ADRs for the first time in China.

A Practical Simulation Method to Calculate Sample Size of Group Sequential Trials for Time-to-Event Data under Exponential and Weibull Distribution

Group sequential design has been widely applied in clinical trials in the past few decades. The sample size estimation is a vital concern of sponsors and investigators. Especially in the survival group sequential trials, it is a thorny question because of its ambiguous distributional form, censored data and different definition of information time. A practical and easy-to-use simulation-based method is proposed for multi-stage two-arm survival group sequential design in the article and its SAS program is available. Besides the exponential distribution, which is usually assumed for survival data, the Weibull distribution is considered here. The incorporation of the probability of discontinuation in the simulation leads to the more accurate estimate. The assessment indexes calculated in the simulation are helpful to the determination of number and timing of the interim analysis. The use of the method in the survival group sequential trials is illustrated and the effects of the varied shape parameter on the sample size under the Weibull distribution are explored by employing an example. According to the simulation results, a method to estimate the shape parameter of the Weibull distribution is proposed based on the median survival time of the test drug and the hazard ratio, which are prespecified by the investigators and other participants. 10+ simulations are recommended to achieve the robust estimate of the sample size. Furthermore, the method is still applicable in adaptive design if the strategy of sample size scheme determination is adopted when designing or the minor modifications on the program are made.

A calibrated power prior approach to borrow information from historical data with application to biosimilar clinical trials

A biosimilar refers to a follow-on biologic intended to be approved for marketing based on biosimilarity to an existing patented biological product (i.e., the reference product). To develop a biosimilar product, it is essential to demonstrate biosimilarity between the follow-on biologic and the reference product, typically through two-arm randomization trials. We propose a Bayesian adaptive design for trials to evaluate biosimilar products. To take advantage of the abundant historical data on the efficacy of the reference product that is typically available at the time a biosimilar product is developed, we propose the calibrated power prior, which allows our design to adaptively borrow information from the historical data according to the congruence between the historical data and the new data collected from the current trial. We propose a new measure, the Bayesian biosimilarity index, to measure the similarity between the biosimilar and the reference product. During the trial, we evaluate the Bayesian biosimilarity index in a group sequential fashion based on the accumulating interim data, and stop the trial early once there is enough information to conclude or reject the similarity. Extensive simulation studies show that the proposed design has higher power than traditional designs. We applied the proposed design to a biosimilar trial for treating rheumatoid arthritis.

A Bayesian prediction model between a biomarker and the clinical endpoint for dichotomous variables

BackgroundEarly biomarkers are helpful for predicting clinical endpoints and for evaluating efficacy in clinical trials even if the biomarker cannot replace clinical outcome as a surrogate. The building and evaluation of an association model between biomarkers and clinical outcomes are two equally important concerns regarding the prediction of clinical outcome. This paper is to address both issues in a Bayesian framework.MethodsA Bayesian meta-analytic approach is proposed to build a prediction model between the biomarker and clinical endpoint for dichotomous variables. Compared with other Bayesian methods, the proposed model only requires trial-level summary data of historical trials in model building. By using extensive simulations, we evaluate the link function and the application condition of the proposed Bayesian model under scenario (i) equal positive predictive value (PPV) and negative predictive value (NPV) and (ii) higher NPV and lower PPV. In the simulations, the patient-level data is generated to evaluate the meta-analytic model. PPV and NPV are employed to describe the patient-level relationship between the biomarker and the clinical outcome. The minimum number of historical trials to be included in building the model is also considered.ResultsIt is seen from the simulations that the logit link function performs better than the odds and cloglog functions under both scenarios. PPV/NPV ≥0.5 for equal PPV and NPV, and PPVu2009+u2009NPV ≥1 for higher NPV and lower PPV are proposed in order to predict clinical outcome accurately and precisely when the proposed model is considered. Twenty historical trials are required to be included in model building when PPV and NPV are equal. For unequal PPV and NPV, the minimum number of historical trials for model building is proposed to be five. A hypothetical example shows an application of the proposed model in global drug development.ConclusionsThe proposed Bayesian model is able to predict well the clinical endpoint from the observed biomarker data for dichotomous variables as long as the conditions are satisfied. It could be applied in drug development. But the practical problems in applications have to be studied in further research.

Design of adaptive two-stage double-arm clinical trials for dichotomous variables

It is well known that flexibility is one of the major advantages of an adaptive two-stage design, and the intended adaptation should be as preplanned as possible to maintain the integrity of the clinical trial. The design of adaptive two-stage double-arm clinical trials for dichotomous variables was proposed by simulation and forecasting procedure at the planning stage. To further ensure the integrity of the clinical trial, the sample size scheme for each scenario, which was supposed to be based on the first stage, was provided in the protocol by Monte Carlo simulation. In addition, the study parameters were determined by comparing the assessment indexes such as total sample size, expected sample size and the test power at the first stage. Furthermore, Fishers combination test and pooled data analysis were considered and compared through the simulation. The latter, which has the larger overall power and the better overall type I error control, with the same sample size was adopted for further simulation and statistical analysis in the clinical trial.

Nateglinide versus repaglinide for type 2 diabetes mellitus in China

The purpose of this study is to evaluate efficacy and safety of nateglinide tablet administration in comparison with those of repaglinide tablet as control on treating type 2 diabetes mellitus in China. Pooled-analysis with analysis of covariance (ANCOVA) method was applied to assess the efficacy and safety based on original data collected from four independent randomized clinical trials with similar research protocols. However meta-analysis was applied based on the outcomes of the four studies. The results by meta-analysis were comparable to those obtained by pooled-analysis. The means of HbA1c, and fasting blood glucose in both the nateglinide and repaglinide groups were reduced significantly after 12xa0weeks duration but no statistical differences in reduction between the two groups. The adverse reaction rates were 9.89 and 6.51% in the nateglinide and repaglinide groups respectively, with the rate difference showing no statistical significance, and the Odds Ratio of adverse reaction rate (95% confidence interval) was 1.59 (0.99, 2.55). Both nateglinide and repaglinide administration have similarly significant effects on reducing HbA1c and FBG. However, the adverse reaction rate in the nateglinide group is higher than that in the latter using repaglinide but no statistical significance difference as revealed in the four clinical trials detailed below.

Comparison of Immunogenicity and Persistence between Inactivated Hepatitis A Vaccine Healive® and Havrix® among Children: A 5-Year Follow-up Study.

ABSTRACT Background: Inactivated vaccines for hepatitis A virus (HAV) infection are widely used in China. Mass vaccination programs drive the need for data on long-term persistence of vaccine-induced protection. Methods: A prospective, randomized, open-label clinical trial was conducted to compare geometric mean concentrations (GMCs) and seroconversion rates (SRs) of anti-HAV antibody elicited by the inactivated vaccines Healive and Havrix for 5 y post immunization, in which 400 healthy children were randomly assigned in a 3:1 ratio to receive 2 doses of Healive or Havrix at 0 and 6 month. Anti-HAV antibody concentration was detected by microparticle enzyme immunoassay (MEIA) during the study. Furthermore, an attempt was made to predict persistence of protective immunogenicity by using a suitable statistical model. Results: The GMCs were significantly higher after vaccination with Healive than after Havrix as comparator vaccine at 1, 6, 7, 18, 30, 42, 54 and 66 month (P < 0.01) with the peak point at 7 month (3427.2 mIU/ml for Healive and 1441.9 mIU/ml for Comparator). Similarly significant differences of SRs were found between the 2 groups at 1 and 6 month (P < 0.01). Afterwards, the SRs of both groups reached 100% at 7 month and did not decline until 66 month(99.1% for Healive and 97.5% for Comparator). A linear mixed model with a change point at 18 month(Model 3) was found to be suitable to predict persistence of protective immunogenicity induced by vaccines. It was estimated that the duration of protection for Healive was at least 20 y with a lower limit of GMC 95% confidence interval (CI) no less than 20 mIU/mL. Conclusions: Compared with Havrix, the new preservative-free inactivated hepatitis A vaccine (Healive) in 2 doses showed better persistence of antibody concentrations for 5 y after full-course immunization among children and the persistence of protective immunogenicity was estimated for at least 20 y.

The Continual Reassessment Method for Multiple Toxicity Grades: A Bayesian Model Selection Approach

Grade information has been considered in Yuan et al. (2007) wherein they proposed a Quasi-CRM method to incorporate the grade toxicity information in phase I trials. A potential problem with the Quasi-CRM model is that the choice of skeleton may dramatically vary the performance of the CRM model, which results in similar consequences for the Quasi-CRM model. In this paper, we propose a new model by utilizing bayesian model selection approach – Robust Quasi-CRM model – to tackle the above-mentioned pitfall with the Quasi-CRM model. The Robust Quasi-CRM model literally inherits the BMA-CRM model proposed by Yin and Yuan (2009) to consider a parallel of skeletons for Quasi-CRM. The superior performance of Robust Quasi-CRM model was demonstrated by extensive simulation studies. We conclude that the proposed method can be freely used in real practice.