Should Phase II Trials Target Only One Population?

In this post, we share why it can sometimes be a good idea to expand Phase II trials. While doing so may seem counterintuitive to most medical affairs professionals, consider our reasoning on how this can mitigate Phase III failure. In other words, we suggest why expanding Phase II could save your drug candidate.

Phase III failures can be mitigated if the patient population is properly categorized prior to the initiation of a full-scale clinical demonstration of efficacy. But assuming there is robust evidence that the desired pharmacologic activity is evident in man, wouldn't an alternative approach for clinical development be to consider the conduct of more Phase II trials in several variants of the population(s) considered for treatment?

The Problem with the Phase II and Phase III Relationship

Phase II results are evaluated to determine effect size and estimate the Phase 3 sample size/power for the 'pivotal' demonstration of efficacy. If the primary endpoint is not positive, then secondary assessments are evaluated. Unfortunately, these evaluations are often not pre-specified and no statistical adjustments for multiple comparisons are made in the analyses. Expecting a large treatment effect with a small number of patients often forgets the old adage 'beware of small N's'.

Free ebook download: Learn the 4-step process that simplifies the site  budgeting process.

When Phase II results in having a small effect size, a large N is required for the pivotal Phase III trial. Depending on the indication, these trials are now extending beyond one year. The expense associated with a trial with these requirements is only possible by an organization with very deep pockets.

The Solution for Phase III May Actually Lie in Phase II

For this reason, wouldn't it be an enhancement for clinical development to spend more time in Phase II and with several candidate populations? 

Doing this would also provide the opportunity to explore alternative endpoints than ones previously used for prior approvals for this indication. For example, a drug candidate with potent anti-inflammatory actions will have numerous disease targets from gastrointestinal to dermatologic.  Thus, enlarging the phase II testing may lead to several different opportunities, some more promising than others.

End of Phase II discussions with appropriate regulatory agencies would allow the 'new' endpoint to be evaluated and perhaps utilized in a Phase III program.

Industry Factors to Consider

It is recognized that bodies such as EMA require specified endpoints in their guidelines and are reluctant to change. However, if robust data from the alternative and 'standard' endpoints were available, negotiation of a change in current guideline specified endpoints could be undertaken. 

Discussion should be possible by the very nature of the word 'guideline'. For example, the FDA now accepts a single endpoint for a demonstration of efficacy as an alternative to a co-primary endpoint in Alzheimer's trials.

All endpoint assessments of a treatment effect and safety data need to subject to state of the art 'predictive analytics'.  These analytics can properly characterize the probability of success of further development or as an alternative reveal an estimation of the futility of achieving a positive result.