Dose-finding with Sequential Parallel Comparison Designs

Posted by Esha Senchaudhuri

Apr 23, 2015 3:53:00 PM

Last week the Cytel Blog discussed the benefits of using the Adaptive Maximizing Design [AM Design] for dose-finding trials involving clinical utility limiting therapies. However, there are other ways that a dose-finding trial can make use of frequent-adaptation maximizing designs. Here we look at what happens to early phase clinical development when an AM Design combines with another adaptive design that is slowly gaining popularity, namely the Sequential Parallel Comparison Design. 


 

Visit Cytel Consulting

Last week the Cytel Blog discussed the benefits of using the Adaptive Maximizing Design [AM Design] for dose-finding trials involving clinical utility limiting therapies [1]. Clinical utility limiting therapies are those in which high levels of adverse side effects place limits on dose selection. For example, when testing a new pain medication, randomization into a high dose arm is likely to cause nausea and vomiting. Randomization into a low dose-arm, by contrast, will not alleviate a sufficient amount of pain.

Given that we do not know the correct dose to administer to a patient prior to a dose-finding trial, it is worthwhile to invest in a trial design which predictably randomizes as many patients as possible into dose-arms near or at the correct dose. A new paper published in Therapeutic Innovation & Regulatory Science argues that the AM Design achieves this, particularly when compared to a conventional trial [2].

A natural question arises as to whether or not such a design will make the trial more expensive. One of the benefits of the AM Design is that it has been known to increase the likelihood of finding the right dose, when compared to a conventional trial. As a result, it is a good investment when preparing for Phase 3. 

AM Designs can also combine with other adaptive clinical trial designs to determine optimal clinical utility even faster. 

One of our clients recently came to us for a new drug that would benefit from a Sequential Parallel Comparison Design [SPCD]. SPCD’s are two-stage designs meant for therapeutic areas known to have high levels of placebo response (e.g. psychiatry.) The second stage of an SPCD trial offers the chance to compare efficacy only amongst placebo non-responders (i.e. the first stage establishes which patients do and do not respond to placebo; the second stage compares responses of the non-responders.)

SPCD are sometimes associated with Phase 3 trials. However, statisticians are developing new uses of the SPCD for the purposes of dose-finding [3].

Working off of a familiar SPCD Model, Cytel designed the two stages of the SPCD to reflect proof-of-concept and dose-finding. The second, dose-finding stage of the SPCD was then designed as an AM Design. This leveraged the benefits of the AM Design improve dose-selection, while also utilizing the hallmarks of the SPCD (i.e. smaller sample size). The overall result was an estimated 3-6 month expedited development time.  


Related Items of Interest

[1] Phase 2 Designs for Clinical Utility Limiting Therapies

[2] Bolognese, James A., Ruth Ann Subach, and Franck Skobieranda. "Evaluation of an Adaptive Maximizing Design Study Based on Clinical Utility Versus Morphine for TRV130 Proof-of-Concept and Dose-Regimen Finding in Patients With Postoperative Pain After Bunionectomy." Therapeutic Innovation & Regulatory Science (2015): 2168479015577866.

[3] Wang, Jessie J., and Anastasia Ivanova. "Dose finding with the sequential parallel comparison design." Journal of biopharmaceutical statistics 24.5 (2014): 1091-1101.


Topics: Cytel Consulting, Early Phase Trials, Clinical Development Strategy, Adaptive Clinical Trials, psychiatry and neuroscience

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