Statistical Considerations on Subgroup Analysis in Clinical Trials

The objective of subgroup analysis of a clinical trial is to investigate consistency or heterogeneity of the treatment effect across subgroups, defined based on background characteristics. As such, subgroup analysis plays an essential role in the interpretation of the clinical trial findings. Consistency of treatment effect across trial subgroups indicates that the average treatment effect is in general applicable regardless of the specific background characteristics. Substantial heterogeneity in treatment effect may be indicative that treatment benefit pertains only to a subset of the population. However, heterogeneity in the observed treatment effect across subgroups can arise due to chance as a result of partitioning the population into several subgroups. Furthermore, as it is known, clinical trials are generally not powered for detecting heterogeneity, thus statistical tests may miss detection of existing heterogeneity due to low power. In this article, we aim to: (i) outline the major issues underlying subgroup analysis in clinical trials and provide general statistical guidance for interpretation of its findings, (ii) provide statistical perspectives on the design and analysis of a clinical trial that aims for establishing efficacy in a targeted subgroup along with that of its overall population, and (iii) highlight some of the underlying assumptions and issues relevant to Bayesian subgroup analysis, subgroup considerations for noninferiority trials, personalized medicine, subgroup misclassification, and finally, subgroup analysis for safety assessment.     

Rationale for a non-inferiority clinical trial design focused on subpopulations

Although double-blind, placebo-controlled clinical trials are utilized extensively to characterize the efficacy and safety of new treatment options, the characteristics of the trial participants often do not most cases, one or more patient subgroups (whether defined by race, ethnicity, co-morbidity, concomitant medication, age, or gender) will be under-represented. Understanding treatment responses in these subpopulations is a vital component of the overall therapeutic profile of a medication. Several different approaches to subgroup analyses within a single trial have been described. In addition, meta-analytic and data pooling approaches utilize results from multiple clinical trials of similar design to increase the number of patients within a targeted subgroup. If the results from exploratory analyses are suggestive of a clinically relevant difference in treatment response for a particular subgroup, then implementation of a prospectively designed clinical trial may be warranted. In this commentary, we discuss the design and results of various studies that include subgroup analyses. In addition, we describe a novel study design with a non-inferiority subpopulation analysis (NISA) that may provide new insights with respect to subgroup analyses. The NISA study design relies on characterization of the dominant group of patients recruited to date in placebo-controlled trials. In the NISA study, the group of patients with those same characteristics is referred to as the Core group. The other key features of the NISA design include non-inferiority analyses comparing subgroups to the Core group and study conditions closely aligned with routine clinical practice (heterogeneous study population and open-label drug administration without placebo). Limitations of the NISA design include the requirement of previously conducted placebo-controlled trials, the inability to compare treatment response to placebo, and that NISA has yet to be validated in practice. We also describe the implementation of the NISA study design in two ongoing clinical trials. After completion of these two studies, the practical value of the NISA design can be more thoroughly evaluated.     

General guidance on exploratory and confirmatory subgroup analysis in late-stage clinical trials

Abstract

This article focuses on a broad class of statistical and clinical considerations related to the assessment of treatment effects across patient subgroups in late-stage clinical trials. This article begins with a comprehensive review of clinical trial literature and regulatory guidelines to help define scientifically sound approaches to evaluating subgroup effects in clinical trials. All commonly used types of subgroup analysis are considered in the article, including different variations of prospectively defined and post-hoc subgroup investigations. In the context of confirmatory subgroup analysis, key design and analysis options are presented, which includes conventional and innovative trial designs that support multi-population tailoring approaches. A detailed summary of exploratory subgroup analysis (with the purpose of either consistency assessment or subgroup identification) is also provided. The article promotes a more disciplined approach to post-hoc subgroup identification and formulates key principles that support reliable evaluation of subgroup effects in this setting.     

Adaptive Designs for Confirmatory Clinical Trials with Subgroup Selection

Growing interest in stratified medicine is leading to increasing importance of subgroup analyses in confirmatory clinical trials. Conventionally, confirmatory clinical trials either focus on a subgroup identified in advance or assess subgroup effects once the trial is completed. The focus of this article is methodology for adaptive clinical trials that both identify whether a treatment is particularly effective in a predefined subgroup, potentially enabling alteration of recruitment, and assess the effectiveness in the subgroup and/or whole population. Methods for such adaptive trials are described and compared, and the logistical and regulatory issues associated with such approaches are discussed.     

Multiplicity issues in exploratory subgroup analysis

ABSTRACT

The general topic of subgroup identification has attracted much attention in the clinical trial literature due to its important role in the development of tailored therapies and personalized medicine. Subgroup search methods are commonly used in late-phase clinical trials to identify subsets of the trial population with certain desirable characteristics. Post-hoc or exploratory subgroup exploration has been criticized for being extremely unreliable. Principled approaches to exploratory subgroup analysis based on recent advances in machine learning and data mining have been developed to address this criticism. These approaches emphasize fundamental statistical principles, including the importance of performing multiplicity adjustments to account for selection bias inherent in subgroup search.

This article provides a detailed review of multiplicity issues arising in exploratory subgroup analysis. Multiplicity corrections in the context of principled subgroup search will be illustrated using the family of SIDES (subgroup identification based on differential effect search) methods. A case study based on a Phase III oncology trial will be presented to discuss the details of subgroup search algorithms with resampling-based multiplicity adjustment procedures.     

Reliably Basing Conclusions on Subgroups of Randomized Clinical Trials

The classical paradigm of Phase III clinical research is to demonstrate efficacy of a drug in an unselected patient population representative for later clinical practice. The flip side of the coin is that homogeneity of the treatment effect in subpopulations of the patient population cannot be assumed to be trivially given. Close inspection of relevant subgroups is important, as soon as overall efficacy has been demonstrated. This may lead to restrictions regarding the patient population to be treated. Similarly, although subgroup findings may be misleading, it should be possible in rare instances to base valid conclusions on subgroups of trials where this has not been precisely prespecified. Subgroups in multiregional clinical trials are different and deserve special consideration.     

Practical Considerations for Subgroups Quantification,Selection and Adaptive Enrichment in Confirmatory Trials

Abstract

Subgroup analyses (e.g., baseline information, biomarkers measurements) are commonly encountered and conducted in confirmatory clinical trials to ensure the risk–benefit consistency and appropriate interpretation of the study results. However, there are natural methodological complications that come with multiple analyses which can result in incorrect scientific or regulatory conclusions for subgroup analysis. Typical issues that may arise include, but are not limited to (a) How to make sure subgroup results are reliable and reproducible? (b) How to quantify subgroup reversal effects which may be due to chance finding, or a lack of power in subgroup and/or treatment interaction tests? (c) How to design efficient trials to establish treatment effect in subgroups or/and overall population with proper Type I error control? These are challenges statisticians and clinical trialist regularly face in the drug development process. In this article, we discuss and present different approaches corresponding to these general issues of subgroup analysis, and their impacts and implications on the interpretation of clinical trials as well as the innovations and opportunities in the era of precision medicine. We also conduct simulations to illustrate the operating characteristics of different methodological tools and designs, along with further practical recommendations and guidance from a statistical perspective.     

Rationale for a non-inferiority clinical trial design focused on subpopulations

Although double-blind, placebo-controlled clinical trials are utilized extensively to characterize the efficacy and safety of new treatment options, the characteristics of the trial participants often do not reflect those of the wider patient population. In most cases, one or more patient subgroups (whether defined by race, ethnicity, co-morbidity, concomitant medication, age, or gender) will be under-represented. Understanding treatment responses in these subpopulations is a vital component of the overall therapeutic profile of a medication. Several different approaches to subgroup analyses within a single trial have been described. In addition, meta-analytic and data pooling approaches utilize results from multiple clinical trials of similar design to increase the number of patients within a targeted subgroup. If the results from exploratory analyses are suggestive of a clinically relevant difference in treatment response for a particular subgroup, then implementation of a prospectively designed clinical trial may be warranted. In this commentary, we discuss the design and results of various studies that include subgroup analyses. In addition, we describe a novel study design with a non-inferiority subpopulation analysis (NISA) that may provide new insights with respect to subgroup analyses. The NISA study design relies on characterization of the dominant group of patients recruited to date in placebo-controlled trials. In the NISA study, the group of patients with those same characteristics is referred to as the Core group. The other key features of the NISA design include non-inferiority analyses comparing subgroups to the Core group and study conditions closely aligned with routine clinical practice (heterogeneous study population and open-label drug administration without placebo). Limitations of the NISA design include the requirement of previously conducted placebo-controlled trials, the inability to compare treatment response to placebo, and that NISA has yet to be validated in practice. We also describe the implementation of the NISA study design in two ongoing clinical trials. After completion of these two studies, the practical value of the NISA design can be more thoroughly evaluated.     

支持抗肿瘤新药进入临床试验的非临床评价要点及关注问题

新药非临床评价的主要目的是决策新药能否进入临床试验。抗肿瘤新药非临床评价中的利弊权衡需考虑到肿瘤疾病及其药物应用特点。多数抗肿瘤药物常伴有严重的不良反应,其临床受试对象多为临床治疗效果较差或根本无有效治疗方法的病人。若新药具有有效性的优势或特点,潜在毒性风险小于疾病自身危险,此时是可以考虑接受新药进入临床试验。本文主要根据该思路对非临床研究中的药效学、毒性实验、药动学进行技术审评,对临床前研究结果的支持性或存在问题进行评价。此外,本文也较为详细讨论了非临床评价的考虑要点及其关注问题。     

Practical Recommendations for Regional Consistency Evaluation in Multi-Regional Clinical Trials with Different Endpoints

In recent years, there is an increasing trend to conduct multi-regional clinical trials (MRCT) for drug development in pharmaceutical industry. A carefully designed MRCT could be used in supporting the new drug's approval in different regions simultaneously. The primary objective of an MRCT is to investigate the drug's overall efficacy across regions while also assessing the drug's performance in some specific regions. In order to claim the study drug's efficacy and get drug approval in some specific region(s), the local regulatory authority may require the sponsors to provide evidence of consistency in the treatment effect between the overall patient population and the local region. Usually, the regional specific consistency requirement needs to be pre-specified before the study conduct and the consistency in treatment effect between the region(s) of interest and overall population will be evaluated at the final analysis. In this article, we evaluate the consistency requirements in multi-regional clinical trials for different endpoints, that is, continuous, binary, and survival endpoints. We also compare the different consistency requirements of the same endpoint/measurement if multiple consistency requirements are enforced and our recommendations for each endpoint/measurement will be made based on the comprehensive consideration.     

Statistical planning to address strongly correlated endpoints for inferential subgroups: An innovative approach for an illustrative clinical trial with complex multiplicity issues

ABSTRACT

Multiplicity is an important statistical issue that arises in clinical trials when the efficacy of the test treatment is evaluated in multiple ways. The major concern for multiplicity is that uncontrolled multiple assessments lead to inflated family-wise Type I error, and they thereby undermine the integrity of the statistical inferences. Multiplicity comes from different sources, for example, making inferences either on the overall population or some pre-specified sub-populations, while multiple endpoints need to be evaluated for each population. Therefore, a sound statistical strategy that controls the family-wise Type I error rate in a strong sense, without excessive loss of power from over-control, is crucial for the success of the trial. For a recent phase III cardiovascular trial with such complex multiplicity, we illustrate the use of a closed testing strategy that begins with a global test; and subsequent testing only proceeds when the global test is rejected. Also, we discuss a simulation study based on this trial to compare the power of the illustrated closed testing strategy to some well-known alternative approaches. We found that this strategy can comprehensively meet most of the primary objectives of the trial effectively with reasonably high overall power.     

Neuroprotective agents in traumatic brain injury

The role of neuroprotection in traumatic brain injury (TBI) is reviewed. Basic research and experimental investigations have identified many different compounds with potential neuroprotective effect. However, none of the Phase III trials performed in TBI have been successful in convincingly demonstrating efficacy in the overall population. A common misconception is that consequently these agents are ineffective. The negative results as reported in the overall population may in part be caused by specific aspects of the head injury population as well as by aspects of clinical trial design and analysis. The heterogeneity of the TBI population causes specific problems, such as a risk of imbalances between placebo and treated groups but also causes problems when a possible treatment effect is evaluated in relation to the prognostic effect present. Trials of neuroprotective agents should be targeted first of all to a population in which the mechanism at which the agent is directed is likely to be present and secondly to a population in which the chances of demonstrating efficacy are realistic, e.g., to patients with an intermediate prognosis. The possibilities for concomitant or sequential administration of different neuroprotective agents at different times deserve consideration. The potential for neuroprotection in TBI remains high and we should not be discouraged by recent failures obtained up until now. Rather, prior to initiating new trials, careful consideration of experimental evidence is required in order to optimise chances for mechanistic targeting and lessons learned from previous experience need to be taken to heart in the design of future studies.     

Tradeoff-based optimization criteria in clinical trials with multiple objectives and adaptive designs

ABSTRACT

The article discusses clinical trial optimization problems in the context of mid- to late-stage drug development. Using the Clinical Scenario Evaluation approach, main objectives of clinical trial optimization are formulated, including selection of clinically relevant optimization criteria, identification of sets of optimal and nearly optimal values of the parameters of interest, and sensitivity assessments. The paper focuses on a class of optimization criteria arising in clinical trials with several competing goals, termed tradeoff-based optimization criteria, and discusses key considerations in constructing and applying tradeoff-based criteria. The clinical trial optimization framework considered in the paper is illustrated using two case studies based on a clinical trial with multiple objectives and a two-stage clinical trial which utilizes adaptive decision rules.     

Stratification and Partial Ascertainment of Biomarker Value in Biomarker-Driven Clinical Trials

This article examines the role of stratification of treatment assignment with regard to biomarker value in clinical trials that accept biomarker-positive and -negative patients but have a primary objective of evaluating treatment effect separately for the marker-positive subset. It also examines the issue of incomplete ascertainment of biomarker value and how this affects inference about treatment effect for the biomarker-positive subset of patients. I find that stratifying the randomization for the biomarker ensures that all patients will have tissue collected but is not necessary for the validity of inference for the biomarker-positive subset if there is complete ascertainment. If there is not complete ascertainment of biomarker values, it is important to establish that ascertainment is independent of treatment assignment. Having a large proportion of cases with biomarker ascertainment is not necessary for establishing internal validity of the treatment evaluation in biomarker-positive patients; independence of ascertainment and treatment is the important factor. Having a large proportion of cases with biomarker ascertainment makes it more likely that biomarker-positive patients with ascertainment are representative of the biomarker-positive patients in the clinical trial (with and without ascertainment), but since the patients in the clinical trial are a convenience sample of the population of patients potentially eligible for the trial, requiring a large proportion of cases with ascertainment does not facilitate generalizability of conclusions.     

调脂药物注册临床试验的考虑

本文旨在介绍调脂药物注册临床试验的主要考虑.有效性主要在于对血脂参数的影响、血管的保护作用和对于死亡率及心血管事件的影响.安全性集中在肝脏、肌肉和长期的心血管事件的发生上.同时本文也对调脂药物在儿童中开展试验的要求进行了阐述.     

临床试验中AE/SAE判别与处置的现状及趋势

合理完善与规范临床试验中不良事件的判断标准与处置流程意义重大。这也是研究者、临床试验机构和监管部门十分关注的热点问题。通过检索和回顾相关文献,从保证药物临床试验安全性着手,明确不良事件定义,探讨药物临床试验不良反应因果判定关联性、不良事件通用术语评价标准（CTCAE）在健康受试者临床试验中的适用性、以及严重不良反应的上报与处置流程的发展更新。通过借鉴国外标准及追踪最新进展与趋势,提出在我国建立临床试验不良事件的判断策略具有重要的意义,并对今后临床试验的相关工作提出了新的思考和展望。     

Definitions and statistical properties of master protocols for personalized medicine in oncology

ABSTRACT

Within the field of cancer research, discovery of biomarkers and genetic mutations that are potentially predictive of treatment benefit is motivating a paradigm shift in how cancer clinical trials are conducted. In this review, we provide an overview of the class of trials known as “master protocols,” including basket trials, umbrella trials, and platform trials. For each, we describe standardized terminology, provide a motivating example with modeling details and decision rules, and discuss statistical advantages and limitations. We conclude with a discussion of general statistical considerations and challenges encountered across these types of trials.     

Statistical considerations for rare diseases drug development

ABSTRACT

One of the most challenges for rare disease clinical trials is probably the availability of a small patient population. It is then a great concern on how to conduct clinical trials with a small number of subjects available for obtaining substantial evidence regarding safety and effectiveness for approval of the rare disease drug product under investigation. FDA, however, does not have the intention to create a statutory standard for approval of orphan drugs that are different from the standard for approval of drugs in common conditions. Thus, it is suggested that innovative trial designs such as a complete n-of-1 trial design or an adaptive design should be used for an accurate and reliable assessment of rare disease drug products under investigation. In this article, basic considerations, innovative trial designs, and statistical methods for data analysis are discussed. In addition, some innovative thinking for the evaluation of rare disease drug products is proposed.     

Statistical Evaluation and Analysis of Regional Interactions: The PLATO Trial Case Study

Multiregional randomized clinical trials often are conducted to evaluate interventions proposed for treatment or prevention of diseases, especially in clinical settings such as cardiovascular diseases where it is important to enroll a large number of patients in a timely manner. As outlined in the International Conference on harmonization E5 Ethnic Factors in the Acceptability of Foreign Clinical Data, an obvious interest when evaluating the results of such trials is the consistency of effects across regions. The purpose of this article is to review a recent example, a large cardiovascular outcomes trial known as “PLATO,” where substantial evidence of regional heterogeneity emerged during the analysis. We present the statistical thinking and methodology that went into the evaluation of the results and the logic that led to the judgment that, while chance cannot be ruled out entirely, the appearance of the regional heterogeneity was likely a manifestation of an underlying interaction with concurrent aspirin dosage. This example may provide a useful reference point for the statistical evaluation of regional effects in future large outcomes trials. Supplementary materials for this article are available online.