An evaluation of inferential procedures for adaptive clinical trial designs with pre-specified rules for modifying the sample size

Point estimation for adaptive trial designs I: A methodological review

Recent FDA guidance on adaptive clinical trial designs defines bias as "a systematic tendency for the estimate of treatment effect to deviate from its true value," and states that it is desirable to obtain and report estimates of treatment effects that reduce or remove this bias. The conventional end-of-trial point estimates of the treatment effects are prone to bias in many adaptive designs, because they do not take into account the potential and realized trial adaptations. While much of the methodological developments on adaptive designs have tended to focus on control of type I error rates and power considerations, in contrast the question of biased estimation has received relatively less attention. This article is the first in a two-part series that studies the issue of potential bias in point estimation for adaptive trials. Part I provides a comprehensive review of the methods to remove or reduce the potential bias in point estimation of treatment effects for adaptive designs, while part II illustrates how to implement these in practice and proposes a set of guidelines for trial statisticians. The methods reviewed in this article can be broadly classified into unbiased and bias-reduced estimation, and we also provide a classification of estimators by the type of adaptive design. We compare the proposed methods, highlight available software and code, and discuss potential methodological gaps in the literature.

Evaluating group-sequential non-inferiority clinical trials following interim stopping: The HIV Prevention Trials Network 083 trial

BACKGROUND/AIMS

The HIV Prevention Trials Network 083 trial was a group-sequential non-inferiority trial designed to compare HIV incidence under a novel experimental regimen for HIV prevention, long-acting injectable cabotegravir, with an active-control regimen of daily oral tenofovir disoproxil fumarate/emtricitabine (brand name Truvada). In March of 2020, just as the trial had completed enrollment, the COVID-19 pandemic threatened to prevent trial participants from attending study visits and obtaining study medication, motivating the study team to update the interim monitoring plan. The Data and Safety Monitoring Board subsequently stopped the trial at the first interim review due to strong early evidence of efficacy.

METHODS

Here we describe some unique aspects of the trial's design, monitoring, analysis, and interpretation. We illustrate the importance of computing point estimates, confidence intervals, and p values based on the sampling distribution induced by sequential monitoring.

RESULTS

Accurate analysis, decision-making and interpretation of trial results rely on pre-specification of a stopping boundary, including the scale on which the stopping rule will be implemented, the specific test statistics to be calculated, and how the boundary will be adjusted if the available information fraction at interim review is different from planned. After appropriate adjustment for the sampling distribution and overrun, the HIV Prevention Trials Network 083 trial provided strong evidence that the experimental regimen was superior to the active control.

CONCLUSIONS

For the HIV Prevention Trials Network 083 trial, the difference between corrected inferential statistics and naive results was quite small-as will often be the case-nevertheless, it is appropriate to report and publish the most accurate and unbiased statistical results.

Das Data Monitoring Committee – State of the Art und Perspektiven für die Zukunft

Hintergrund

Die klassische randomisierte und kontrollierte klinische Prüfung steht mit komplexen Studiendesigns und Disease-Interception-Konzepten vor neuen Herausforderungen. Dabei können Data Monitoring Committees (Datenüberwachungskomitees [DMCs]) eine zentrale Funktion übernehmen, wenn eine fachgerechte Integration in den methodischen Ablauf der klinischen Prüfungen erfolgt. Vor diesem Hintergrund haben die zuständige Bundesoberbehörde und die zuständige Ethikkommission im impliziten/expliziten Genehmigungsverfahren zu überprüfen, ob eine substanzielle Charta die Arbeitsweise des unabhängigen Gremiums abbildet.

Ziel

Es wurde der Frage nachgegangen, mit welchen Häufigkeiten und Gegebenheiten DMCs in klinischen Prüfungen eingesetzt werden.

Methode

Die Datenbank des Bundesinstituts für Arzneimittel und Medizinprodukte (BfArM) bildet die Grundlage für eine Datenanalyse hinsichtlich Häufigkeiten der Implementierung von Data Monitoring Committees unter verschiedenen Kriterien über einen Beobachtungszeitraum von mehr als 15 Jahren.

Ergebnisse

Bei insgesamt 14.135 klinischen Prüfungen mit Arzneimitteln wurde 4152 DMCs eingesetzt. Das unabhängige Expertengremium wurde überwiegend von kommerziellen Sponsoren in der Phase III der klinischen Entwicklung integriert. Die Ethikkommissionen waren mit unterschiedlichen absoluten Häufigkeiten eingebunden.

Diskussion

Die Sponsoren zeigen eine zunehmende Bereitschaft, DMCs in die methodische Durchführung von klinischen Prüfungen, insbesondere bei neuartigen Studiendesigns, zu integrieren. DMCs können eine wichtige wissenschaftliche Unterstützung sein, um die Implikationen des Coronavirus SARS-CoV‑2 auf klinische Prüfungen zu beurteilen.

Sample size re-estimation for pivotal clinical trials

It is well known that if the hypothesis test is left unchanged, the Type I error rate may be inflated for sample size re-estimation (SSR) designs. To address this issue, three main approaches have been proposed in the literature: combination test, conditional error and conventional test with sample size increase in the allowable region (AR) only. These three seemingly different approaches are in fact connected. For each combination test, there is a corresponding conditional error function and AR. Designing adaptation rules in this AR with conventional test guarantees the Type I error rate control but at the same time always leads to smaller power comparing to the corresponding combination test (or conditional error) approach. In cases where conventional test is still preferable, step-wise type adaptation rules that do not fully reside in the AR can be alternatively considered. We believe controversies in the statistical community on the efficiency comparisons between group sequential (GS) and SSR design stem partially from the misalignment of performance metrics and conditional versus unconditional evaluations. We advocate summary metrics, such as median, variance or tail probabilities of the sample size in addition to expectation and personalizing efficiency definition for each trial sponsor. Conditional metrics by favorable, promising and unfavorable zones of the interim results provide additional insights and should always be incorporated into the decision-making process.