5 Scenarios When ‘Keep it Simple’ May Be Bad Advice for Clinical Trial Designers

Posted by Esha Senchaudhuri

Sep 18, 2014 10:44:00 AM

When designing clinical trials, many trial designers are advised to keep the trial simple. Prima facie, the keep it simple principle seems like sound advice. There are various logistical uncertainties that arise when implementing a clinical trial, and the more simple a trial – so conventional wisdom says – the easier it is to respond to these uncertainties.

According to Zoran Antonijevic, a Senior Director at Cytel Consulting, there is reason to doubt such conventional wisdom. After all, flexibility is hardly a virtue of a traditional trial design. Simple designs may seem to make it easier to monitor data and report results. However, a flexible design can better address remaining uncertainties in product development. These uncertainties are related to treatment effect, dose selection, or a sub-population that would experience the best benefit/risk from the treatment.














Here are five benefits a trial may have to forego when designers stay true to the 'Keep It Simple' Principle.

(1)    De-risking drug development:

Perhaps the most compelling reason to adapt for trial sponsors is the capacity of an adaptive trial to mitigate the financial risks of drug development. Powerful computer simulations can take information gathered from one or more interim looks to give potential investors a reasonable idea of how well a trial might perform. One example of this is the design of the VALOR trial, which allowed potential investors to invest depending on whether or not interim looks were favorable (or unfavorable), or promising.

(2) Improving probability of success through better dose/subpopulation selection:

Through improved dose selection and opportunities to drop trial arms after interim looks, adaptive trials are sometimes able to offer more accurate calculations of a drug’s probability of success, at least when compared with more conventional trials. For example, the ADVENT trial for the drug Crofelmer, was a Phase 3 trial with four treatment arms. The trial continued with two arms and after an interim look that dropped the two sub-optimal doses. Given the number of Phase 2 trials that move to Phase 3 with questionable dose-selection, an adaptive design can be useful in addressing residual uncertainties regarding the dose.  

(3) Discontinuation of inefficient treatments or doses:

Adaptive designs allow trial sponsors to discover ineffective treatments earlier during a trial. Not only does this reduce trial time, but from an ethical perspective it ensures that patients are not receiving subpar doses. While in a conventional trial it could take years to determine that the new drug is not as efficacious as standard of care, adaptive designs allow such discoveries to happen in the interim. Furthermore, when it is unclear which dose is optimal, dropping sub-optimal arms ensures that every patient receives quality care.

(4) Reductions in costs and development time:

The flexibility of adaptive designs provide numerous ways in which to reduce cost and development time. Upfront program and portfolio strategy becomes critical for circumventing several potential pitfalls for Phase 1, 2 and 3. See the following blog posts on how adaptive designs can make trials more efficient.

  1. Predictive Enrichment Trials
  2. Cardiovascular Outcome Trials
  3. Phase 3 Relapsed/Refractory AML Trial
  4. Oncology Trials

(5)    A Win-Win for Patient Enrollment:

This creates a related benefit to trial sponsors. One of the most difficult and expensive aspects of implementing a clinical trial is patient enrollment. 80% of clinical trials registered at clinicaltrials.gov fail to meet enrollment deadlines, and 50% enroll one or fewer patients.  Since the average set-up cost per trial site is approximately $70,000 USD, the inability to secure patients can be very costly. An adaptive design can help overcome difficulties with enrollment, simply be improving patient confidence in the investigated therapeutic, which in turn would increase likelihood of participation. 

Related Items of Interest

Oncology & Precision Mediine

Cyrus Mehta, et al., 'Biomarker Driven Adaptive Population Enrichment for Oncology Trials with Time to Event Endpoints,' (forthcoming: Statistics in Medicine) PaperSlides

Adaptive Designs for Precision Medicine: A Look at Pfizer's Xalkori Trial

Overcoming Prowell's Pitfalls: Cytel Weighs in on Strategies for Oncology Drug Development

Powering Oncology Trials for Success

Cytel Joins DIA Discussion on Predictive Enrichment


Adaptive Designs for Financial Program & Portfolio Optimization

Why Using Adaptive Designs Can Attract Investors to Your Trial 

To Adapt or Not to Adapt? 10 Simple Steps to Deciding Whether Your Next Trial Should be Adaptive

Impact of Study Design and Development Strategy on Pharmaceutical Programs & Portfolios

Beyond Borders: Biotechnology Industry Report 2014 (Ernst & Young) 


Other Adaptive Designs

Cardiovascular Outcome Trials

Adaptive Designs for Risk Based Monitoring

Seamless Adaptive Designs: The ADVENT Trial

Empirical Study Confirms Positive Impact of Adaptive Designs

Why Adaptive Sample Size Re-estimation Preserves Type 1 Error


Topics: Oncology, Promising Zone, sample size re-estimation, Enrichment, Cytel Consulting, Efficacy, Interim Analyses, forecasting, optimization, Program and Portfolio Optimization, R&D, Adaptive Clinical Trials

The Cytel blog keeps you up to speed with the latest developments in biostatistics and clinical biometrics.  Sign up for updates direct to your inbox. You can unsubscribe at any time.


Posts by Topic

see all

Recent Posts