SciaNews began in 1990 and is a semi-annual publication. The mandate of SciaNews is to provide a source of information and a forum for discussion of current concepts in biostatistics, clinical data management and study design. Special emphasis is given to the exploration of trends and innovations in study design. Original contributions are invited from the healthcare industry and academic communities.

This publication is distributed worldwide and is complimentary. For your copy of SciaNews please contact us through our web-site or fax us a request with your mailing address and professional affiliation.

acrobat

   Acrobat Reader is required to view our Press Releases. If you do not already have it, please click the icon to download it.
In a Phase II clinical trial, the primary objective is to determine the efficaciousness of a drug, such that decisions to proceed with further studies and development of the drug are warranted. To detect a measurable effect in an investigation product, a two-stage design is often used, whereby an interim look at the data may result in the decision to terminate the trial early, or proceed with the second stage of the study. The Simon’s two-stage design to determine sample size is one that is widely used and accepted in Phase II cancer trials. A number of alternative methods that implement adaptive aspects to the design have also been published and used in trials. This paper will look specifically at Lin and Shih’s adaptive method, which include extensions to Simon’s optimal and minimax designs. Currently, PROC POWER in SAS v9 does not produce sample size calculations for two-stage designs, which led to the motivation behind this paper. The SAS codes for the traditional and adaptive designs have been separated into two macros, SIMON and ADAPTIVE, designed to produce multiple sample size outputs which satisfy userspecified constraints.
Early-phase trials typically focus on the safety and efficacy aspects separately. This article discusses some innovative approaches to clinical trials that are well-suited to early clinical development. The methodology is primarily Bayesian, which allows for the use of prior and historical information. Most traditional clinical trial designs have their roots in Frequentist statistics. When we report results with confidence intervals and p-values, we are describing Frequentist characteristics of the data. Bayesian statistics provides a mathematical method for calculating the likelihood of a future event given knowledge from prior events. In doing so, it provides a means for quantifying uncertainty about an unknown parameter of interest.
TORONTO, Ontario – On May 3-4, 2006, the MaRS Centre was the venue for the “Innovative Proof-of-Concept Designs for Phase I/II and IIIb Studies” seminar hosted by SciAn Services, Inc. The first of a planned annual event, this particular seminar was undertaken in part to address the FDA’s “diagnosis” of a slowed drug development process, and to bring together members of industry and the academia to discuss the challenges of designing clinical trials. Key speakers on statistical and regulatory topics included Dr. Peter Thall (MD Anderson Cancer Centre), Dr. Chyi-Hung Hsu (Novartis US) and Dr. Agnes Klein (Health Canada).
One of the most interesting and challenging aspects of early phase clinical trials is studying the pharmacokinetics of a new drug, in order to determine how a body processes it. For those readers with limited knowledge of the subject, this article is intended to provide an overview of pharmacokinetics, and the pharmacokinetic terms that are commonly used in medical journals. The article will also introduce the concepts of bioavailability, therapeutic range, accumulation and steady state, and dose proportionality.
This article illustrates the use of analysis-of-covariance to explore the association between a covariate and the treatment effect in a 22 cross-over trial. The type I sum of squares is used in the analysis and the results are identical to the 2 separate analyses of covariance presented by Jones and Kenward (1989). An illustration of the analyses based on the example used by Jones and Kenward is presented. Based on the expected mean squares, the appropriateness of the hypothesis tests on the main and interaction effects is discussed.
One of the benefits of a sequential design is smaller sample size. This article presents the results of a Phase III trial which utilized a sequential design and resulted in a modest reduction in sample size compared to that of a fixed sample design.
The Joint Meeting of the International Biometric Society (ENAR) was held in Birmingham, Alabama from March 26-29, 1995. This conference consisted of sessions covering various areas of bio-statistics as well as a programme of several continuing education courses. Some of these presentations will be summarized in this article, as well as an overview of the course "Applied Bayesian Statistics".
Interaction between biologically active agents is of great importance in normal/abnormal phyisiology, toxicology, the control of environmental pollution and in therapeutics. Combination of drugs are often used to achieve better efficacy and/or reduce toxicity in the treatment of abnormal disorders that range from malignancies and severe cardiovascular disease to infections, asthma, diabetes and arthritis. An adequate rationale for treatment with drug combinations is often missing and a trial and error method of screening/evaluating drug combinations is therefore a necessary and potentially rewarding avenue of research. The following paper will introduce our readers to the most frequently used methods in the analysis of drug interactions.
The Continual Reassessment Method (CRM) for determining the MTD in Phase I trials works well when the dose-response curve is the expected “S-shape”. However, in some cases, the dose response curve may deviate from this assumption and reflect a slower rate of toxicity over the dose-range tested. The original CRM underperforms in this scenario. This article introduces a 2-parameter model of the original CRM which overcomes this deficiency.
The standard phase I design for determining the MTD of a drug is typically the 1-in-3 method or variations thereof. In 1990, O’Quigley et al proposed a novel method, the Continual Reassessment Method (CRM) based on Bayesian statistics. This article reviews and compares the traditional methods in Phase I clinical trials with the CRM approach.
Regulatory criteria established for the acceptance of average bioequivalence do not fully satisfy the requirements arising for the interchangeability of drug products. The divergence is illustrated by noting that the regulatory criteria are compatible with large inter-individual variation of within-subject differences or ratios of the relevant kinetic parameters. Regulatory criteria of three jurisdictions are considered: the United States, Canada, and the Province of Ontario. For example, in a crossover trial conducted with 24 subjects, in as many as 60% of the individuals, the difference between the relevant kinetic parameters for the two formations could, on average, be outside the range of 70-130% of the reference product, and still satisfy the criterion of the Food and Drug Administration (FDA) for the acceptance of average bioequivalence. It is suggested that individual bioequivalence can be effectively evaluated in three- and four-way crossover trials.
This article summarizes the plenary sessions of the 1992 BioInternational Conference on Bioavailability, Bioequivalence and Pharmacokinetic Studies.
This article reviews the draft guideline “In vivo criteria to establish equivalence of safety and efficacy of a generic or second entry drug delivered by metered dose inhaler” released by the Division of Endocrinology, Metabolism and Allergy (HPB) in 1992.
For some clinical trials, the outcome variable is dichotomous in nature, i.e. the result is classified into 1 of 2 possibilities (yes/no, recovered/not recovered). In addition, this outcome variable may be measured by 2 different researchers resulting in 2 different ratings. This article explores statistical methods for assessing the reliability of the measurement instrument by looking at the degree of agreement between the 2 ratings.
2x2 cross-over designs can result in smaller sample sizes compared to that of a parallel group design. When there is a positive correlation between the observations in periods 1 and 2, the sample size savings can be even larger. However, the opposite occurs in the presence of a negative correlation. This paper explores one of the causes of negative correlation, namely the effect of cyclic variation in the outcome variable (endpoint).
Sequential designs from clinical trials are often overlooked in favor of ‘tried-and-true’ designs such as the parallel group design. This article puts forward key reasons for using sequential designs and the potential benefits over conventional designs.
The use of the log-transformation on data from bioequivalence studies to satisfy normality assumptions has been a point of argument among statisticians and medical doctors. In this article, we evaluate the effect of erroneously applying (or not applying) the log-transformation using simulation studies.
Why bother using style guides? If you have ever been unsure of the correct way to cite reference material from books, electronic sources or government reports, or whether to use periods in abbreviations, or when to use a a generic drug name instead of a proprietary name, a style guide can provide the answer. Do you really know what an acronym is? See what leading style guides say in the following sections.
Quality Assurance is a well established discipline, not least in the healthcare industry where ethical and regulatory considerations require the application of strict quality criteria to the final product delivered to patients. Less often are Quality Assurance criteria applied to more intangible aspects of human endeavour, such as planning and forecasting. We need look no further than government deficit projections to realize that this is a lamentable omission. And deficit financing is not an option for industry.
As the range of therapeutic products available has expanded, regulatory agency requirements for safety documentation have also grown. Hence, all individuals involved in the process of risk-benefit assessment of investigational and approved drug products need to be familiar with the most frequently encountered drug reactions. This includes investigational site clinical staff as well as clinical research monitors, project managers and safety officers of sponsoring pharmaceutical companies and CRO's working as partners with the pharmaceutical industry.
Regulatory criteria established for the acceptance of average bioequivalence do not fully satisfy the requirements arising for the interchangeability of drug products. The divergence is illustrated by noting that the regulatory criteria are compatible with large inter-individual variation of within-subject differences or ratios of the relevant kinetic parameters. Regulatory criteria of three jurisdictions are considered: the United States, Canada, and the Province of Ontario. For example, in a crossover trial conducted with 24 subjects, in as many as 60% of the individuals, the difference between the relevant kinetic parameters for the two formations could, on average, be outside the range of 70-130% of the reference product, and still satisfy the criterion of the Food and Drug Administration (FDA) for the acceptance of average bioequivalence. It is suggested that individual bioequivalence can be effectively evaluated in three- and four-way crossover trials.
A case report form (CRF) is a data collection tool used in clinical trials to support investigators and coordinators in capturing all protocol-required information. A well-designed CRF facilitates data collection and entry, and directly benefits other facets of data management and statistical analysis. An informative and structured CRF simplifies database design and data validation processes as well as manipulation of data during statistical analysis. This paper explores CRF design techniques that consider efficient data management and statistical analysis.
An increasing number of pharmaceutical companies and CRO's are turning their attention to groupware technologies (a set of applications which allow a group of people to share information simultaneously; Lotus NotesTM or Microsoft ExchangeTM) to address their information management needs, including clinical study management. In a series of articles, we will review the key characteristics of groupware, the specific requirements of a clinical study management system and its possible extension to incorporate data management. We will assess the feasibility of a distributed data management system which creates the necessary environment for data entry and review to be performed at various study locations, e.g., at investigational sites and/or regional CRO offices.
Data exchange, whether in the form of documents, databases, or other media, has become an essential part of daily business. Additionally, there is a growing need for companies to maintain data in a usable form for an extended period of time, particularly in the pharmaceutical industry.
In 1997, FDA released the 21 CFR Part 11 regulation “Electronic Records; Electronic Signatures”. This article summarizes the main points of the regulation and its effect on distributed trial and data management systems.
In the past 25 years, the clinical trial processes for developing new drugs have grown increasingly complex. The nature of the therapies being investigated has increased the difficulty of the procedures being conducted, and competition for trial subjects has resulted in complex multi-centre, multi-regional clinical trials becoming the standard for evaluating medical therapies prior to regulatory approval or as post-registration commitments. In turn, the legal, operational, training, and quality processes have also become more complex due to the increase in the number of sites, site staff, data variables, labs, etc.
The pressure to examine the allocation of health care resources has greatly increased in recent years for several reasons, including escalating health care costs, new health care technologies that are increasingly effective, but increasingly expensive, and health care budgets that have been reduced. The above pressures are forcing decision makers to seek new principles for the adoption of new technologies. Consequently, health economics, which studies the economic aspects of health care interventions, has become a rapidly growing field.
Providing a high level of transparency and penetration into study performance as measured using defined and consistently applied metrics enables the different participants in the study project management framework to grasp a quick view of the study’s progress, get an understanding of the most critical performance issues, and to take action to isolate and eliminate the sources of poor performance quickly and efficiently. This article reviews some of the common failings in developing and implementing effective operational metrics programs.
Many factors make up the budget for a clinical trial. However, whereas a Project Manager may have some idea on the cost for the trial management segment (CTM), they are often less knowledgeable in the cost of the CDM and BioStatistics segments. This article equips the Project Manager with useful information on the services and fees involved in CDM and BioStatistics to make an educated estimate on cost for budget purposes.
In clinical trials, a sample size estimate (the number of patients to be enrolled or randomized) or at least a statistical justification of the selected sample size is required in the protocol. Whereas, the protocol may be written by a Medical Directory, Project Manager, CRA or other trial personnel, the sample size estimate is the responsibility of the BioStatistician. This article prepares clinical trial personnel for typical questions that may be asked by the BioStatistician when a sample size estimation is requested.
Quality assurance is a well established discipline in the healthcare industry. However, it is less applied in intangible situations such as planning and forecasting. This article addresses the place of quality assurance in development planning and provides a list of criteria applicable to the research and development arena.