BioCentury's websites will be down for upgrades starting at 9 p.m. PDT on Monday, August 26. We expect the downtime to last no more than 6 hours, and we apologize for any inconvenience.

12:00 AM
 | 
Feb 16, 2015
 |  BioCentury  |  Regulation

Scrutinizing dose finding

How companies, EMA hope to improve dose-finding in Phase II

EMA officials have hinted they would be willing to approve drugs with less clinical data if companies employ more robust dose-finding methodologies, but companies want more than a hint before they abandon traditional dose-finding approaches.

At a workshop held in December, regulators from Europe and the U.S. as well as company representatives agreed the traditional dose-finding approach should be abandoned in favor of new modeling and simulation methodologies that better characterize the dose-response relationship. Both groups acknowledged that a better understanding of dose response can shorten development timelines and reduce overall costs for industry while giving regulators better data to make benefit-risk decisions.

The regulators clearly expected better data and greater efficiency to be enough to drive industry adoption. But sponsors said additional guidance and regulatory incentives - such as only requiring a single Phase III trial if data were supplemented with a large, well-characterized Phase II dose-finding trial - will be necessary to get R&D heads on board.

The opportunity

Attendees at the joint workshop hosted by EMA and the European Federation of Pharmaceutical Industries and Associations (EFPIA) largely shared the belief that a majority of clinical programs do not adequately characterize the dose-response relationship in Phase II testing.

Most companies continue to use the traditional method of pairwise comparison, in which multiple doses are compared to placebo and each dose arm is powered to demonstrate statistical significance. One weakness of the approach is that the actual shape of the dose-response curve is usually unknown.

In addition, because pairwise comparison requires statistical significance which in turn requires large cohorts for each arm, only two or three dose arms are usually included. The combination of uncertainty about the shape of the dose-response curve and the costs of enrolling big enough cohorts to do pairwise comparison often results in companies just making a best guess about the one or two doses to take into Phase III testing.

These uncertainties have predictable knock-on effects in Phase III and in registration. Phase III programs can fail because the dose was not high enough to produce the necessary effect size, or because the dose was too high and resulted in unacceptable side effects.

...

Read the full 1816 word article

User Sign in

Trial Subscription

Get a 4-week free trial subscription to BioCentury

Article Purchase

$150 USD
More Info >