Interim analyses and sequential designs in phase III studies

A multi-arm multi-stage platform design that allows preplanned addition of arms while still controlling the family-wise error

There is growing interest in platform trials that allow for adding of new treatment arms as the trial progresses as well as being able to stop treatments part way through the trial for either lack of benefit/futility or for superiority. In some situations, platform trials need to guarantee that error rates are controlled. This paper presents a multi-stage design, that allows additional arms to be added in a platform trial in a preplanned fashion, while still controlling the family-wise error rate, under the assumption of known number and timing of treatments to be added, and no time trends. A method is given to compute the sample size required to achieve a desired level of power and we show how the distribution of the sample size and the expected sample size can be found. We focus on power under the least favorable configuration which is the power of finding the treatment with a clinically relevant effect out of a set of treatments while the rest have an uninteresting treatment effect. A motivating trial is presented which focuses on two settings, with the first being a set number of stages per active treatment arm and the second being a set total number of stages, with treatments that are added later getting fewer stages. Compared to Bonferroni, the savings in the total maximum sample size are modest in a trial with three arms, <1% of the total sample size. However, the savings are more substantial in trials with more arms.

A systematic review of randomised controlled trials with adaptive and traditional group sequential designs - applications in cardiovascular clinical trials

BACKGROUND

Trial design plays a key role in clinical trials. Traditional group sequential design has been used in cardiovascular clinical trials over decades as the trials can potentially be stopped early, therefore, it can reduce pre-planned sample size and trial resources. In contrast, trials with adoptive designs provide greater flexibility and are more efficient due to the ability to modify trial design according to the interim analysis results. In this systematic review, we aim to explore characteristics of adaptive and traditional group sequential trials in practice and to gain an understanding how these trial designs are currently being reported in cardiology.

METHODS

PubMed, Embase and Cochrane Central Register of Controlled Trials database were searched from January 1980 to June 2022. Randomised controlled phase 2/3 trials with either adaptive or traditional group sequential design in patients with cardiovascular disease were included. Descriptive statistics were used to present the collected data.

RESULTS

Of 456 articles found in the initial search, 56 were identified including 43 (76.8%) trials with traditional group sequential design and 13 (23.2%) with adaptive. Most trials were large, multicentre, led by the USA (50%) and Europe (28.6%), and were funded by companies (78.6%). For trials with group sequential design, frequency of interim analyses was determined mainly by the number of events (47%). 67% of the trials stopped early, in which 14 (32.6%) were due to efficacy, and 5 (11.6%) for futility. The commonly used stopping rule to terminate trials was O'Brien- Fleming-type alpha spending function (10 (23.3%)). For trials with adaptive designs, 54% of the trials stopped early, in which 4 (30.8%) were due to futility, and 2 (15.4%) for efficacy. Sample size re-estimation was commonly used (8 (61.5%)). In 69% of the trials, simulation including Bayesian approach was used to define the statistical stopping rules. The adaptive designs have been increasingly used (from 0 to 1999 to 38.6% after 2015 amongst adaptive trials). 25% of the trials reported "adaptive" in abstract or title of the studies.

CONCLUSIONS

The application of adaptive trials is increasingly popular in cardiovascular clinical trials. The reporting of adaptive design needs improving.

On the Methodological Aspects of the Clinical Trials for COVID-19 Conducted in the First Year of the Pandemic: A Descriptive Analysis

In 2020, the whole planet was plagued by the extremely deadly COVID-19 pandemic. More than 83 million people had been infected with COVID-19 while more than 1.9 million people around the planet had died from this virus in the first year of the pandemic. From the first moment, the medical community started working to deal with this pandemic. For this reason, many clinical trials have been and continue to be conducted to find a safe and efficient cure for the virus. In this paper, we review the 96 clinical trials, registered in the ClinicalTrials.gov database, that had been completed by the end of the first year of the pandemic. Although the clinical trials contained significant heterogeneity in the main methodological features (enrollment, duration, allocation, intervention model, and masking) they seemed to be conducted based on an appropriate methodological basis.

Effects of dose change on the success of clinical trials

The search for disease modifying therapies in Alzheimers disease (AD) has recently led to promising results but also revealed design issues in clinical trials themselves. Of particular importance is the potential statistical challenges that can arise when dosages change after an interim analysis, which is not uncommon in contemporary AD trials. Following the recent Aducanumab trials, we sought to study the implications of dose changes on the statistical power of an AD trial. We conducted extensive simulations to calculate statistical power when the relationship between treatment effect size and time is linear or non-linear, and the investigated drug has delayed treatment effect or not. Statistical power depends on many design factors including the dose change time, correlation, population homogeneity, and treatment effect time. We recommend that researchers conduct simulation studies at the interim analysis to justify the modified sample size and/or follow-up time modification meanwhile the type I and II error rates are controlled.

Power walking based outpatient cardiac rehabilitation in patients with post-coronary angioplasty: Randomized control trial

PURPOSE

The purpose of this trial was to compare the effectiveness of standardized outpatient cardiac rehabilitation combined with treadmill power walking versus standardized outpatient cardiac rehabilitation alone on health-related quality of life (HQoL), functional exercise capacity (FEC), left ventricular ejection fraction (LVEF) and metabolic equivalent of tasks (METs) in patients who went post coronary angioplasty (CA). Further this study evaluated the association between average numbers of steps taken with above clinical outcomes.

METHODS

In a pragmatic sequential randomized clinical trial, 24 patients were randomized into two groups (n = 12) and participated in a standardized outpatient cardiac rehabilitation program (SOCRP) with treadmill power walking as an intervention group and SOCRP alone in control group. Scores obtained before and after 4 weeks of intervention, that is, after 12 treatment sessions were assessed using a HQoL questionnaire and 6-min walk test (6 MWT). Average number of steps taken throughout the 4 weeks, METs and LVEF values were obtained by pedometer, exercise stress testing and echocardiogram respectively.

RESULTS

Significant improvements were found in intergroup and intragroup comparison after 4 weeks of cardiac rehabilitation (p < 0.05). Scores of 6 MWT and LVEF significantly improved in the intervention group (p < 0.003) compared to the control group (p < 0.032). HQoL components that is, global and physical, MET values and average number of steps were significantly higher in the intervention group compared to the control group (p < 0.001).

CONCLUSION

SOCRP with power walking was more effective in improving HQoL, FEC, LVEF, METs and average numbers of steps than SOCRP alone although both interventions were significant after 4 weeks in patients underwent CA and completed cardiac rehabilitation program. Positive significant associations were found between the average number of steps taken with scores of METs and scores of global and physical domains of HQoL.

Sequential trials in the context of competing risks: Concepts and case study, with R and SAS code

Sequential designs and competing risks methodology are both well established. Their combined use has recently received some attention from a theoretical perspective, but their joint application in practice has been discussed less. The aim of this paper is to provide the applied statistician with a basic understanding of both sequential design theory and competing risks methodology and how to combine them in practice. Relevant references to more detailed theoretical discussions are provided, and all discussions are illustrated using a real case study. Extensive R and SAS code is provided in the online Supplementary Material.

Is Implant Coating With Tyrosol- and Antibiotic-loaded Hydrogel Effective in Reducing Cutibacterium (Propionibacterium) acnes Biofilm Formation? A Preliminary In Vitro Study

BACKGROUND

Studies have suggested that Cutibacterium acnes (formerly known as Propionibacterium) is the most frequently isolated pathogen after shoulder arthroplasty. To address the burden of periprosthetic joint infections associated with this pathogen, new prevention methods are needed. Tyrosol has a promising record of effectiveness in the field of biofilm-associated infections; however, to our knowledge, it has not been tested against C. acnes thus far.

QUESTIONS/PURPOSES

In this in vitro study, we asked: (1) Is tyrosol effective in inhibiting and eradicating C. acnes planktonic growth? (2) Is there synergy between tyrosol and rifampicin? (3) Is supplementation of hydrogel with tyrosol at the minimum inhibitory and subinhibitory concentrations efficacious in reducing free-floating C. acnes growth? (4) Is implant hydrogel coating (either alone or combined with tyrosol, rifampicin, or vancomycin) beneficial in reducing C. acnes biofilm formation? (5) Is the administration of soluble tyrosol an effective measure against C. acnes biofilm formation?

METHODS

We assessed C. acnes planktonic growth and eradication by inspecting visually the results of the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) assays. We also evaluated macroscopically the presence of synergy among tyrosol and rifampicin by means of the MIC checkerboard testing. Thereafter, we addressed colorimetrically the efficacy of tyrosol-loaded Defensive Antibacterial Coating (DAC®) hydrogel against the C. acnes free-floating form by means of the XTT cell proliferation reduction assay. Then, we explored photometrically the effect of hydrogel and soluble tyrosol at reducing C. acnes biofilm formation on titanium alloy disks that simulated orthopaedic implants by using the minimum biofilm inhibition concentration assay. In particular, 16 disks were sequentially allocated to each of the following testing conditions: (1) hydrogel alone; (2) tyrosol-loaded hydrogel; (3) rifampicin-supplemented hydrogel; (4) vancomycin-loaded hydrogel; and (5) soluble tyrosol. Subsequently, implants were sonicated and cell viability was evaluated in terms of the XTT assay.

RESULTS

Tyrosol was effective in inhibiting C. acnes planktonic (free-floating) growth demonstrating MIC values of 63 mM (9 mg/mL) and MBC values of 250 mM (35 mg/mL). Concerning synergy assessment, the checkerboard testing revealed additivity among tyrosol and rifampicin with a fractional inhibitory concentration index of 0.56. In addition, a hydrogel coating with tyrosol at the MIC showed no difference in the inhibition of free-floating C. Acnes form over control (median absorbance [MA] for tyrosol-supplemented hydrogel versus control groups were 0.21 [interquartile range {IQR}, 0.19-0.24] versus 0.26 [IQR, 0.23-0.31], p = 0.066). Furthermore, loaded hydrogel with tyrosol at 597 mg/mL (1 M) was no more effective than control in reducing C. acnes biofilm formation (MAs for tyrosol versus control were 0.12 [IQR, 0.11-0.13] versus 0.14 [IQR, 0.12-0.16], respectively; p = 0.076). This was also the case when we considered hydrogel in conjunction with vancomycin and rifampicin (MAs for vancomycin at 2% and 5% and rifampicin at 1% versus biofilm control were 0.139 [IQR, 0.133-0.143] and 0.141 [IQR, 0.133-0.143] and 0.135 [IQR, 0.128-0.146] versus 0.142 [IQR, 0.136-0.144], correspondingly). In contrast, soluble tyrosol at 597 mg/mL (1 M) inhibited biofilm formation compared to control (MAs for tyrosol and control groups were 0.11 [IQR, 0.09-0.13] versus 0.13 [IQR, 0.12-0.14], p = 0.007).

CONCLUSIONS

Although the implant coating with hydrogel (either pure or supplemented with antimicrobial agents) did not diminish C. acnes biofilm development in vitro, soluble tyrosol at 597 mg/mL (1 M) exceeded the meaningful biofilm inhibition threshold of 80%.

CLINICAL RELEVANCE

The results of the current preclinical investigation did not support the use of a fast, bioresorbable hydrogel as a coating method against C. acnes biofilms. Instead, direct local administration of soluble tyrosol at high concentrations should be further tested in future animal studies.

An introduction to multiplicity issues in clinical trials: the what, why, when and how

In clinical trials it is not uncommon to face a multiple testing problem which can have an impact on both type I and type II error rates, leading to inappropriate interpretation of trial results. Multiplicity issues may need to be considered at the design, analysis and interpretation stages of a trial. The proportion of trial reports not adequately correcting for multiple testing remains substantial. The purpose of this article is to provide an introduction to multiple testing issues in clinical trials, and to reduce confusion around the need for multiplicity adjustments. We use a tutorial, question-and-answer approach to address the key issues of why, when and how to consider multiplicity adjustments in trials. We summarize the relevant circumstances under which multiplicity adjustments ought to be considered, as well as options for carrying out multiplicity adjustments in terms of trial design factors including Population, Intervention/Comparison, Outcome, Time frame and Analysis (PICOTA). Results are presented in an easy-to-use table and flow diagrams. Confusion about multiplicity issues can be reduced or avoided by considering the potential impact of multiplicity on type I and II errors and, if necessary pre-specifying statistical approaches to either avoid or adjust for multiplicity in the trial protocol or analysis plan.

Increasing the efficiency of clinical trials in neurodegenerative disorders using group sequential trial designs

OBJECTIVES

Clinical trials in neurodegenerative disorders are facing high futility rates and rising development costs. We aim to review and exemplify the value of group sequential trial designs (i.e., designs with one or more prospectively planned interim analyses) within the field of amyotrophic lateral sclerosis.

STUDY DESIGN AND SETTING

We reviewed the literature to identify sequentially conducted trials. Subsequently, we reanalyzed the dexpramipexole trial (EMPOWER), a classically designed and conducted trial involving 942 participants, by sequentially monitoring the functional questionnaire and survival endpoint. Finally, we simulated the performance of the sequential methodology under different treatment effects.

RESULTS

Only six (12%) randomized, placebo-controlled trials incorporated stopping rules for both futility and superiority. Despite its high enrollment rate, sequential reanalysis of the EMPOWER study reduced the total trial duration with 140 days (23.4%, 95% confidence interval [CI] 13.2-34.4%), the number of follow-ups with 2,688 visits (23.6%, 95% CI 11.3-38.6%), and the total drug exposure time with 73,377 days (20.6%, 95% CI 9.8-35.9%). The functional questionnaire considerably increased the heterogeneity in the test statistics, which may negatively affect sequential monitoring.

CONCLUSION

Group sequential trials can result in important reductions in the trial duration, which could make clinical trials more ethical by reducing the patients' exposure to noneffective treatments or by limiting their time on placebo.

Medications development for the treatment of alcohol use disorder: insights into the predictive value of animal and human laboratory models

Development of effective treatments for alcohol use disorder (AUD) represents an important public health goal. This review provides a summary of completed preclinical and clinical studies testing pharmacotherapies for the treatment of AUD. We discuss opportunities for improving the translation from preclinical findings to clinical trial outcomes, focusing on the validity and predictive value of animal and human laboratory models of AUD. Specifically, while preclinical studies of medications development have offered important insights into the neurobiology of the disorder and alcohol's molecular targets, limitations include the lack of standardized methods and streamlined processes whereby animal studies can readily inform human studies. Behavioral pharmacology studies provide a less expensive and valuable opportunity to assess the feasibility of a pharmacotherapy prior to initiating larger scale clinical trials by providing insights into the mechanism of the drug, which can then inform recruitment, analyses, and assessments. Summary tables are provided to illustrate the wide range of preclinical, human laboratory, and clinical studies of medications development for alcoholism. Taken together, this review highlights the challenges associated with animal paradigms, human laboratory studies, and clinical trials with the overarching goal of advancing treatment development and highlighting opportunities to bridge the gap between preclinical and clinical research.

Model-Based Assessment of Alternative Study Designs in Pediatric Trials. Part I: Frequentist Approaches

Alternative designs can increase the feasibility of pediatric trials when compared to classical parallel designs (PaD). In this work we present a model-based approach based on clinical trial simulations for the comparison of PaD with the alternative sequential, crossover, and randomized withdrawal (RWD) designs. Study designs were evaluated in terms of: type I and II errors, sample size per arm (SS), trial duration (TD), treatment exposures, and parameter estimate precision (EP). The crossover requires the lowest SS and TD, although it implies higher placebo and no treatment exposures. RWD maximizes exposure to active treatment while minimizing that to placebo, but requires the largest SS. SS of sequential designs can sometimes be smaller than the crossover one, although with poorer EP. This pharmacometric framework allows a multiscale comparison of alternative study designs that can be used for design selection in future pediatric trials.

Simultaneous sequential monitoring of efficacy and safety led to masking of effects

OBJECTIVE

Usually, sequential designs for clinical trials are applied on the primary (=efficacy) outcome. In practice, other outcomes (e.g., safety) will also be monitored and influence the decision whether to stop a trial early. Implications of simultaneous monitoring on trial decision making are yet unclear. This study examines what happens to the type I error, power, and required sample sizes when one efficacy outcome and one correlated safety outcome are monitored simultaneously using sequential designs.

STUDY DESIGN AND SETTING

We conducted a simulation study in the framework of a two-arm parallel clinical trial. Interim analyses on two outcomes were performed independently and simultaneously on the same data sets using four sequential monitoring designs, including O'Brien-Fleming and Triangular Test boundaries. Simulations differed in values for correlations and true effect sizes.

RESULTS

When an effect was present in both outcomes, competition was introduced, which decreased power (e.g., from 80% to 60%). Futility boundaries for the efficacy outcome reduced overall type I errors as well as power for the safety outcome.

CONCLUSION

Monitoring two correlated outcomes, given that both are essential for early trial termination, leads to masking of true effects. Careful consideration of scenarios must be taken into account when designing sequential trials. Simulation results can help guide trial design.

The differential effects of core stabilization exercise regime and conventional physiotherapy regime on postural control parameters during perturbation in patients with movement and control impairment chronic low back pain

BACKGROUND

The purpose of the present study was to examine the differential effect of core stability exercise training and conventional physiotherapy regime on altered postural control parameters in patients with chronic low back pain (CLBP). As heterogeneity in CLBP population moderates the effect of intervention on outcomes, in this study, interventions approaches were used based on sub-groups of CLBP.

METHODS

This was an allocation concealed, blinded, sequential and pragmatic control trial. Three groups of participants were investigated during postural perturbations: 1) CLBP patients with movement impairment (n = 15, MI group) randomized to conventional physiotherapy regime 2) fifteen CLBP patients with control impairment randomized to core stability group (CI group) and 3) fifteen healthy controls (HC).

RESULTS

The MI group did not show any significant changes in postural control parameters after the intervention period however they improved significantly in disability scores and fear avoidance belief questionnaire work score (P < 0.05). The CI group showed significant improvements in Fx, Fz, and My variables (p < 0.013, p < 0.006, and p < 0.002 respectively with larger effect sizes: Hedges's g > 0.8) after 8 weeks of core stability exercises for the adjusted p values. Postural control parameters of HC group were analyzed independently with pre and post postural control parameters of CI and MI group. This revealed the significant improvements in postural control parameters in CI group compared to MI group indicating the specific adaptation to the core stability exercises in CI group. Though the disability scores were reduced significantly in CI and MI groups (p < 0.001), the post intervention scores between groups were not found significant (p < 0.288). Twenty percentage absolute risk reduction in flare-up rates during intervention was found in CI group (95% CI: 0.69-0.98).

CONCLUSIONS

In this study core stability exercise group demonstrated significant improvements after intervention in ground reaction forces (Fz, Mz; g > 0.8) indicating changes in load transfer patterns during perturbation similar to HC group.

TRIAL REGISTRATION

UTRN095032158-06012009423714.

Sequential analysis applied to clinical trials in dentistry: a systematic review

Clinical trials employ sequential analysis for the ethical and economic benefits it brings. In dentistry, as in other fields, resources are scarce and efforts are made to ensure that patients are treated ethically. The objective of this systematic review was to characterise the use of sequential analysis for clinical trials in dentistry. We searched various databases from 1900 through to January 2008. Articles were selected for review if they were clinical trials in the field of dentistry that had applied some form of sequential analysis. Selection was carried out independently by two of the authors. We included 18 trials from various specialties, which involved many different interventions. We conclude that sequential analysis seems to be underused in this field but that there are sufficient methodological resources in place for future applications.Evidence-Based Dentistry (2008) 9, 55-62. doi:10.1038/sj.ebd.6400587.

Assessing Efficacy and Therapeutic Claims in Emerging Indications for Recombinant Factor VIIa: Regulatory Perspectives

When compared with the evidence-based, cost-effectiveness criteria underpinning most government reimbursement schemes in the social market economies, the three regulatory hurdles of safety, quality and efficacy are probably of modest impact in influencing increased usage of recombinant activated factor VII (rFVIIa; NovoSeven, Novo Nordisk, Bagsvaerd, Denmark). Nevertheless, efficacy claims must be supported if regulatory approval is to be granted for the wider range of indications that have been proposed for rFVIIa. With the refinement of clinical trial designs over the past 40 years, the randomized controlled trial (RCT) has assumed the role of gold standard, providing the highest level of evidence for therapeutic efficacy. However, it is incorrect to assume that regulatory authorities give sole credence to RCTs in assessing claims. It is noteworthy that the indications already accepted for rFVIIa by international regulatory authorities--including the treatment of inhibitors to factor VIII and factor IX, substitution for FVII deficiency, and treatment of Glanzmann's thrombasthenia--were supported not by RCTs but by studies conventionally considered to provide modest evidence levels. Therefore, the use of studies other than RCTs for the more recently proposed indications for rFVIIa in a range of conditions requiring hemostatic correction is perfectly feasible. What regulators expect are well-conducted and well-described studies adhering to principles of good clinical practice, which can be scrutinized for evidence of clinical efficacy and which are based on the initially proven principle for the drug. This paper discusses the regulatory history of rFVIIa in the major regulatory authorities and assesses the route needed to support claims being made in the mainstream literature. Recent episodes where post-market events have forced regulators to be more than usually cautious will be used as examples to suggest possible pitfalls to the extension of approved claims for rFVIIa. The major paths for enhancing access for indications in small patient numbers, where RCTs are even more difficult to perform, will be described and their use for possible extension of rFVIIa indications will be discussed.

Stopping clinical trials by design

Before any clinical trial begins, a detailed trial protocol must be prepared. The authority of the trial results will depend on the quality of this document. In many protocols, a key component is a plan for a series of interim analyses of the accumulating trial data, and a 'stopping rule' based on them. Such a rule might be intended to prevent participants from continuing to receive a drug that already seems to be unsafe, or to allow a successful drug to become generally available as soon as sufficient evidence of its advantages has been collected. There has been considerable misunderstanding of these rules, and controversies associated with them. Here, I discuss why this might be, and what can be done to promote their successful and beneficial use in the future.

A randomized sequential trial of creatine in amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis (ALS) is a fatal disease with no cure. In a transgenic mouse model of ALS, creatine monohydrate showed a promising increase in survival. We performed a double-blind, placebo-controlled, sequential clinical trial to assess the effect of creatine monohydrate on survival and disease progression in patients with ALS. Between June 2000 and December 2001, 175 patients with probable, probable-laboratory supported, or definite ALS were randomly assigned to receive either creatine monohydrate or placebo 10 gm daily. A sequential trial design was used with death, persistent assisted ventilation, or tracheostomy as primary end points. Secondary outcome measurements were rate of decline of isometric arm muscle strength, forced vital capacity, functional status, and quality of life. The trial was stopped when the null hypothesis of indifference was accepted. Creatine did not affect survival (cumulative survival probability of 0.70 in the creatine group vs 0.68 in the placebo group at 12 months, and 0.52 in the creatine group vs 0.47 in the placebo group at 16 months), or the rate of decline of functional measurements. Creatine intake did not cause important adverse reactions. This placebo-controlled trial did not find evidence of a beneficial effect of creatine monohydrate on survival or disease progression in patients with ALS.

Planning and monitoring of placebo-controlled survival trials: comparison of the triangular test with usual interim analyses methods

AIMS

For ethical and economic reasons, interim analysis of phase III clinical trials is essential. This study was conducted to compare the efficiency of two interim analysis procedures, which could be used to allow early termination of a clinical trial.

METHODS

We made a post hoc application of two interim analysis methods (Lan & DeMets with O'Brien-Flemming modification, and the triangular test according to Whitehead) by using individual patient data from four published placebo-controlled survival trials. We determined the date the trial would have been stopped had each method been used, and we estimated consequent results in terms of events and patient numbers included in the trial, the duration of the trial, and on treatment effect.

RESULTS

The triangular test provided the lowest number of events required to reach a conclusion of the trials while providing an accurate estimate of experimental treatment effects. The triangular test thus indicated the smallest number of patients that would have been enrolled, and the shortest trial duration. The difference between the methods was most important with a detrimental effect of experimental treatment: the number of required events was reduced by 75% and the trial duration was shortened by 48% with the triangular test compared to the Lan & DeMets method.

CONCLUSIONS

Stopping a trial early must depend on the clinical context. It is most important to stop a placebo-controlled trial as soon as possible when the experimental treatment can be shown deleterious. In such a situation the triangular test is more appropriate than the Lan & DeMets method. When a treatment effect is no different from, or better than, placebo the triangular test is also superior but the importance of premature termination of the trial in such cases has to be balanced against the inevitable reduction of information that the trial can provide.

Learning from previous responses in phase I dose-escalation studies

Dose escalation in phase I studies is generally performed on the basis of clinical experience and judgement. In this paper some of the statistical approaches that have been proposed for the formalization of the procedure are described. Apart from the use of the Continual Reassessment Method in oncology studies, such formal methods have received little implementation. The purpose of presenting them here is to promote their further exploration and appropriate implementation. Certain limitations are discussed, which will be best overcome by collaboration between clinical pharmacologists and statisticians.

Stopping rules for phase II studies

This paper, the second in a series of three papers concerned with the statistical aspects of interim analyses in clinical trials, is concerned with stopping rules in phase II clinical trials. Phase II trials are generally small-scale studies, and may include one or more experimental treatments with or without a control. A common feature is that the results primarily determine the course of further clinical evaluation of a treatment rather than providing definitive evidence of treatment efficacy. This means that there is more flexibility available in the design and analysis of such studies than in phase III trials. This has led to a range of different approaches being taken to the statistical design of stopping rules for such trials. This paper briefly describes and compares the different approaches. In most cases the stopping rules can be described and implemented easily without knowledge of the detailed statistical and computational methods used to obtain the rules.