spacer
home > ict > winter 2013 > dose of reality
PUBLICATIONS
International Clinical Trials

Dose of Reality

While there are many examples of well-designed and successful Phase 2 trials, there is also increasing awareness of the inefficiencies and high failure rates of drug development programmes. This has focused attention on the limited ability of existing approaches to efficiently detect efficacy, identify the best-performing dose, and allow the rapid elimination of poorly performing development candidates or dosing strategies (1-3).

During such dose-ranging studies, information is gained about the statistical properties of clinical endpoints, the incidence of common adverse events, and other factors potentially influencing the design of subsequent pivotal trials. A failure in Phase 2 to identify the correct dose, detect dose-limiting toxicity, or discover that a surrogate endpoint ultimately behaves differently than the endpoint required for a Phase 3 registration trial, can lead to a failure in Phase 3.

Due in part to these concerns, there has been growing interest in the utilisation of adaptive approaches for the design and analysis of Phase 2 trials (4,5). The use of adaptive designs in dose-ranging studies increases the efficiency of drug development by improving our ability to efficiently learn about the dose response and better determine whether to take a drug forward into confirmatory phase testing and at what dose.

Efficiency and Ethics

Adaptive designs explicitly address multiple trial goals, adaptively allocate subjects according to ongoing information needs, and allow termination for both early success and futility. This approach can maximise the ability to test a larger number of doses in a single trial, while simultaneously increasing the efficiency of the trial in terms of making better go/no-go decisions about continuing the trial and/or the development of the drug for a specific indication.

Judicious use of adaptive designs in dose-ranging studies may increase the information value per resource unit invested, by avoiding allocation of patients to non-efficacious/unsafe therapies and allowing stopping decisions to be made at the earliest possible time-point.

These designs may also result in more ethical treatment of patients from at least two perspectives. Firstly, a larger number of patients can be randomised to more favourable, effective doses, with fewer patients exposed to less effective doses. Secondly, there is the potential to include fewer total patients, thus reducing their risk of exposure to adverse events.

Study Modications

During frequent interim analyses in a dose-ranging study, many aspects of the trial design can be modified: number of subjects, study duration, endpoint selection, treatment duration, number of treatments and patient population (6). These design changes are an opportunity to calibrate initial assumptions and enhance clinical trial efficiency in learning about the research questions.

However, the adaptations are prospectively defined prior to the start of the trial; these modifications are a design feature aimed at enhancing the trial, not a remedy for inadequate planning. They are not ad hoc study corrections made via protocol amendment (7). The specific adaptations considered, and the basis of their implementation, are carefully defined based on strict rules, justified statistically and scientifically, and are an integral part of the final, pre-recruitment trial protocol. Failure here would compromise the interpretation and acceptance of study results.

Choosing the Right Design

Many approaches exist for the proper design and analysis of these trials. The ultimate choice of the method to be applied depends on the particular settings and goals. Dose-ranging studies should thus be tailored to best fit the needs of the particular drug development programme under consideration.

Response-adaptive designs directly address the goals of ‘learn’ trials and comprise an integrated approach to efficiently learn about the dose-response relationship using a parsimonious working design model.

Another important design component is response-adaptive dose allocation. The objectives of this design are to learn about the most important region of a dose-response curve or to better estimate the target dose, efficiently reduce uncertainty in important predictive probabilities, and establish early stopping rules for efficacy or futility, as well as to define the number of available doses, frequency of the interim analyses, and control of accrual rate (8).

The efficiency of optimal design of experiments, including the so-called MCP-Mod approach when the primary endpoint is immediate, has been well documented (8). Recent simulation studies conducted by the PhRMA Working Group on Adaptive Dose-Ranging Studies show that response-adaptive dose allocation designs are generally superior in performance to conventional pairwise comparison approaches (9,10). Knowledge about the research question – for example, identifying a target dose or estimating the dose-response – is optimised with these designs (11).

Longitudinal Modelling

In some clinical trials there will be a delay in the response of clinical interest. Since few patients will have experienced the endpoint in the early stages of a trial, there may be little information that can be used when making a decision to modify the trial’s course. But almost always, the clinical efficacy endpoint will be measured at early time points, and these measurements might be correlated with, and predict for, the primary long-term endpoint.

The focus of the definitive analysis is still the primary clinical endpoint and not these short-term endpoints. However, the latter may be used as a necessary condition of potential treatment effect and can enhance the interim decision of dropping a treatment arm or changing the treatment allocation. The research questions are: what is the optimal number of measurements per patient; and what are the optimal time intervals between these measurements? A major benefit of modelling relationships between early and late endpoints is that it makes for stronger interim assessments of long-term endpoints, and therefore improves the effi ciency of adaptive designs.

A parametric model for the time-profi le of the repeated measurements per patient was recently considered (12). Locally optimal designs and information matrices for these special non-linear mixed effects models were defi ned. A cost for repeated measurements was proposed and the optimality criterion was maximised, taking into account both the cost for patient recruitment and the cost incurred in taking a single measurement. Additional technical details in implementing the adaptive version of the optimal designs in this situation were also presented.

Successful Implementation

Adaptive designs represent a completely new technology in drug development. A prerequisite for their successful implementation is an understanding of the underlying methodology and how it affects the logistics of the trial, such as data management and monitoring procedures, electronic data capture (EDC), drug supply management and data monitoring committee operating procedures. These designs have an impact on drug development strategies, trial protocols, informed consent forms, data analysis and reporting plans. Adaptive designs will also change the procedures for enrollment, randomisation, data capturing, monitoring and data cleaning.

EDC – at least for the decision-making critical endpoints – is an important enabler for adaptive designs. Very fast timelines for data analysis, decision-making and the implementation of changes are critical, requiring careful coordination. The use of the supporting infrastructure for adaptive designs can also facilitate the implementation of flexible just-in-time drug supply chain management.

In order to make informed decisions, the sponsor must have access to real-time clinical data from all sources, and be able to easily review and assess those data. Sponsors will require a data management system for rapid clinical data access and cleaning, a drug supply system for supplies planning and management, and a system that provides randomisation, medication kit management and emergency unblinding.

Analysis Tools

A potential barrier to wide implementation of adaptive designs relates to the planning of adaptive clinical trials. Specifically, determining the optimal characteristics of the study design can be a complex yet critical decision that may require more planning and time at the design stage.

However, there are already several commercial software packages that are powerful tools for the planning, simulation and analysis of complex adaptive designs for dose-ranging studies. For example, FACTS is comprehensive software for the ‘learn’ phase, and includes early-stage dose-escalation designs, adaptive dose-ranging studies and interim decision rules based on both efficacy and safety responses (13). ADDPLAN is software for adaptive clinical trials in the ‘confirm’ phase, incorporating sample size re-estimation, sequential designs and multiple comparison procedures for multi-armed adaptive trials (14).

New analysis tools such as these will enable drug developers to remove a number of the uncertainties inherent in Phase 2 dose-finding trials. Increased awareness and documentation of the success of these designs in clinical trials will improve critical decisions on dose-selection and directly impact the probability of success in Phase 3.

References
1. O’Neill RT, A perspective on characterising benefits and risks derived from clinical trials: Can we do more? Drug Information Journal 42: pp235-245, 2008
2. Pinheiro J et al, Adaptive and model-based dose-ranging trials: Quantitative evaluation and recommendation, Statistics in Biopharmaceutical Research 2: pp435-454, 2010
3. Woodcock J and Woosley R, The FDA Critical Path Initiative and its influence on new drug development, Annual Review of Medicine 59: pp1-12, 2008 4. Berry DA, Bayesian clinical trials, Nature Reviews Drug Discovery 5: pp27-36, 2006
5. Dragalin V, Hsuan F and Padmanabhan SK, Adaptive designs for dose-finding studies based on Sigmoid Emax Model, Journal of Biopharmaceutical Statistics 17: pp1,051-1,070, 2007
6. Dragalin V, Adaptive designs: terminology and classification, Drug Information Journal 40(4): pp425-436, 2006
7. Gallo P et al, Executive summary of the PhRMA Working Group on Adaptive Designs in Clinical Drug Development, Journal of Biopharmaceutical Statistics 16: pp275-283, 2006
8. Dragalin V, Contribution of different design components to the efficiency of response-adaptive dose-ranging studies, JSM 2013 Program. Visit: www.amstat.org/meetings/jsm/2013/onlineprogram/ abstractdetails.cfm?abstractid=307341
9. Bornkamp B et al, Innovative approaches for designing and analyzing adaptive dose-ranging trials, Journal of Biopharmaceutical Statistics 17: pp965-995, 2007
10. Dragalin V et al, A simulation study to compare new adaptive dose-ranging designs, Statistics in Biopharmaceutical Research 2: pp487-512, 2010
11. Mielke T, Efficient designs for dose-response studies under model uncertainty, JSM 2013 Program. Visit: www.amstat.org/meetings/jsm/2013/onlineprogram/abstractdetails.cfm?abstractid=309849
12. Dragalin V, Optimal design of experiments for delayed responses in clinical trials, in mODa 10 – Advances in model-oriented design and analysis, contributions to statistics, pp55-61, 2013
13. FACTS. Visit: www.smarterclinicaltrials.com/what-we-offer/facts
14. ADDPLAN 6. Visit: www.addplan.com


Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:
0
     

There are no comments in regards to this article.

spacer
Vladimir Dragalin is Senior Vice President, Software Development and Consulting at the Aptiv Solutions Innovation Centre. Previously, he led the adaptive trial design centre at Quintiles, as well as holding positions at Wyeth Research, GSK, the University of Rochester and at research institutions in Europe. Vladimir earned his PhD in Probability Theory and Mathematical Statistics from the Steklov Mathematical Institute, Moscow. He is an elected member of PhRMA Biostatistics and Data Management Technical Group and a Fellow of the American Statistical Association.
spacer
Vladimir Dragalin
spacer
spacer
Print this page
Send to a friend
Privacy statement
News and Press Releases

Signant Health Appoints Lawrence Miller as Chief Technology Officer

LONDON & PHILADELPHIA – June 13, 2019: Signant Health, formerly CRF Bracket, has appointed Lawrence Miller as Chief Technology Officer. Signant unites electronic Clinical Outcomes Assessment (eCOA), eConsent, Patient Engagement, Interactive Response Technology (IRT), Clinical Supply Management and Endpoint Quality into the industry’s most comprehensive patient-centric suite of technology solutions and services. An accomplished innovator, Miller brings unique vision and expertise to a Signant team already committed to advancing clinical research with its next-generation technology portfolio.
More info >>

White Papers

The Power of Execution in Clinical Research

Medpace

Medpace, long known for its excellence in conducting Phase I-IV studies on a global platform, is particularly noted for its therapeutic excellence. When that expertise is combined with operational precision, the results are nothing short of stellar. The Medpace full-service outsourcing approach – combined with therapeutic expertise, regulatory know how, and global presence – delivers clinical studies with an industry recognized track record.
More info >>

 
Industry Events

SCOPE Europe – Summit for Clinical Ops Executives Europe

17-18 September 2019, Barcelona, Spain

Building on the success of its long-running US-based counterpart, Summit for Clinical Trials Operations Executives Europe (SCOPE Europe) is a two-day event serving clinical innovation leaders, clin ops execs, functional heads, category managers, and those responsible for the planning and management of clinical trials in Europe and the Rest of the World. The event will feature opening and closing plenary keynote sessions, five conference tracks, interactive breakout discussion groups, and networking. It is an opportunity for leaders from large, mid-sized, and small pharma, specialty pharma, biotech, vendor companies, and academic research centers to come together to share best practices and discuss the new era of analytics-driven and patient-centric trials.
More info >>

 

 

©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement