Semiparametric copula-based analysis for treatment effects in the presence of treatment switching.

In controlled trials, "treatment switching" occurs when patients in one treatment group switch to alternative treatments during the trial, and poses challenges to treatment effect evaluation owing to crossover of the treatments groups. In this work, we assume that treatment switching can occur after some disease progression event and view the progression and death events as two semicompeting risks. The proposed model consists of a copula model for the joint distribution of time-to-progression (TTP) and overall survival (OS) up to the earlier of the two events, as well as a conditional hazard model for OS subsequent to progression. The copula model facilitates assessing the marginal distributions of TTP and OS separately from the association between the two events, and, in particular, the treatment effect on OS in the absence of treatment switching. The proposed conditional hazard model for death subsequent to progression allows us to assess the treatment switching (crossover) effect on OS given occurrence of progression and covariates. Semiparametric proportional hazards models are employed in the marginal models for TTP and OS. A nonparametric maximum likelihood procedure is developed for model inference, which is verified through asymptotic theory and simulation studies. The proposed analysis is applied to a lung cancer dataset to illustrate its real utility.

Human papillomavirus infections as a marker to predict overall survival in lung adenocarcinoma.

Human papillomavirus (HPV) has been implicated in multiple cancers, but its significance in lung cancer has remained controversial. As the prevalence of HPV 16/18 infection was higher in lung adenocarcinoma among Taiwanese females, the aim of our study was to evaluate the clinical impact of HPV infections in lung adenocarcinoma. Two hundred and ten patients were enrolled to investigate the associations of HPV status in tumors with clinical characteristics as well as its impact on overall survival. The methods to assess HPV status were by immunohistochemistry for HPV L1 capsid protein and E6 protein and by nested polymerase chain reaction for HPV 16 and HPV 18. HPV infections were identified in 35.2% of patients, and associated with localized and smaller sized tumors (p = 0.022 and p = 0.002, respectively). Patients with HPV infections had a significantly better survival (p = 0.023, by log-rank test) and a significantly reduced mortality risk after adjustments of age, tumor extent, epidermal growth factor receptor (EGFR) mutations status and treatments [adjusted hazard ratio = 0.68, 95% confidence interval (CI) = 0.49-0.96, p = 0.026, by multivariate Cox proportional hazards models]. Specifically, patients with both HPV infections and EGFR mutations had the best survival outcome [1-year survival rate, 68.5% (95% CI = 52.2-4.8%)]. Our findings indicate that HPV infections represent an independent prognostic factor for overall survival in patients with lung adenocarcinoma.

Assessing dose-region profile of drug efficacy: a multiregional trial strategy.

In this article, we try to address the optimal dosing issue using a multiregional trial. In this trial, in addition to the treatment group using a globally promising dose of study drug, we have another treatment group using a lower dose treatment with which the drug also shows substantial treatment effect in some of the regions, and both treatment groups share the same placebo (control) group. The incorporation of an additional low-dose treatment group in the multiregional trial can provide the following advantages: (i) The multiregional trial can establish the whole treatment effect profile over different regions as well as drug doses; (ii) the multiregional trial allows for sharing drug efficacy information across different regions; and (iii) the use of a common placebo (control) group for the high- and low-dose treatment groups in the multiregional trial results in logistical convenience. We also examine the regional sample size allocation in the proposed multiregional trial for ensuring a desired power in a local region, using the Bayesian approach we proposed previously for analyzing the multiregional trial data.

Proposals of statistical consideration to evaluation of results for a specific region in multi-regional trials--Asian perspective.

In recent years, global collaboration has become a conventional strategy for new drug development. To accelerate the development process and to shorten approval time, the design of multi-regional trials incorporates subjects from many countries around the world under the same protocol. After showing the overall efficacy of a drug in all global regions, one can also simultaneously evaluate the possibility of applying the overall trial results to all regions and subsequently support drug registration in each of them. Recently, the trend for simultaneous clinical development in Asian countries being undertaken simultaneously with clinical trials conducted in Europe and the United States has been rapidly rising. In this paper, proposals of statistical consideration to multi-regional trials are provided. More specifically, three aspects are addressed: the definition of the 'Asian region,' the consistency criterion between the 'Asian region' and the overall regions, and the sample size determination for the multi-regional trial.

A Bayesian approach to evaluating regional treatment effect in a multiregional trial.

A multiregional trial, conducted in more than one region under a common protocol, is a promising strategy making valuable medicines available to patients globally without time lag. When evaluating the treatment effect for each local region, one may wish to utilize information from other regions to enhance the statistical power. This work proposes a Bayesian approach to bridging data across different regions in a multiregional trial to get an improved analysis of treatment effect for a local region. The new proposal has the following distinct features: (1) It performs internal bridging in a multiregional trial, with the degree of bridging automatically determined by the interregional variability of the treatment effect across different regions; (2) it usually ensures the consistency of the conclusions from local and global inference when the treatment effect is virtually homogeneous across regions and is found nonsignificant globally; (3) it generally protects against overbridging of the global information for evaluating the treatment effect in a very small region. Formulas for statistical power of the proposed method are provided. We illustrate the utility of the proposed method by two numerical examples reflecting typical issues we may encounter in evaluating regional treatment effect in a multiregional trial.

A regulatory view of adaptive trial design.

Developing a new medicine is an expensive and time-consuming process. Researchers are interested in applying better designs to expedite the approval of potential medicinal products. Adaptive designs, which allow for some types of prospectively planned mid-study change, can improve the efficiency of a trial and maximize the chance of success. Possible design adaptations of clinical trials include sample size re-estimation, change in primary endpoint, interim dropping of treatment arms, change in statistical hypothesis, and change in the primary analysis. In this article, the regulatory considerations of the methodological issues with respect to adaptive design are discussed. Several examples of design adaptation that the Center for Drug Evaluation has encountered during the past 3 years are presented.

The use of weighted Z-tests in medical research.

Traditionally the un-weighted Z-tests, which follow the one-patient-one-vote principle, are standard for comparisons of treatment effects. We discuss two types of weighted Z-tests in this manuscript to incorporate data collected in two (or more) stages or in two (or more) regions. We use the type A weighted Z-test to exemplify the variance spending approach in the first part of this manuscript. This approach has been applied to sample size re-estimation. In the second part of the manuscript, we introduce the type B weighted Z-tests and apply them to the design of bridging studies. The weights in the type A weighted Z-tests are pre-determined, independent of the prior observed data, and controls alpha at the desired level. To the contrary, the weights in the type B weighted Z-tests may depend on the prior observed data; and the type I error rate for the bridging study is usually inflated to a level higher than that of a full-scale study. The choice of the weights provides a simple statistical framework for communication between the regulatory agency and the sponsor. The negotiation process may involve practical constrains and some characteristics of prior studies.

Sample size and optimal designs in stratified comparative trials to establish the equivalence of treatment effects among two ethnic groups.

When a new investigational medicine is intended to be applied to populations with different ethnic backgrounds, a stratified comparative phase III trial using ethnic groups as strata may be conducted to assess the influence of ethnic factors on clinical outcomes of this new medicine. In this paper, based on a binomial model with odds ratio as the measure of the treatment effect, we derive the score test and the associated sample size formula for establishing the equivalence/noninferiority of the treatment effects of a medicine among two ethnic groups. A simplified test together with its sample size formula are also given. Taking into account the sample size, cost, and power of testing, respectively, we derive the optimal design parameters, i.e., the allocation among treatment groups and ethnic groups, based on the simplified test.