Dr Michael Silverman of BioStrategics Consulting Ltd discusses some of the more common mistakes made in Phase 2 clinical study design and implementation.
Every so often, I run into someone who sees Phase 2 clinical studies as an unnecessary stage of drug development. Let me explain.
A few weeks ago, I addressed an issue regarding Phase 2 clinical trials, and mentioned that Phase 2 was a fertile topic for further exploration. Many times, inventors, corporate executives and investors – who, understandably, are anxious to see a new technology advanced to market as quickly as possible – do not see Phase 2 as stage that adds value to the overall process of commercialization. Instead, it is sometimes perceived as a step that unnecessarily adds time and distance to the drug development process – not to mention, of course, costs. While in some unusual and uncommon situations – for example, very rare diseases or dramatically effective therapies for previously untreatable diseases – it is possible that a program can jump directly form Phase 1 to a registration trial. However, in the vast majority of cases Phase 2 clinical studies are an absolutely critical step to establishing the benefit and utility of a new drug.
I have definitely seen Phase 2 trials that subtract, rather than add, value to a product. Such trials often make unsupportable assumptions, cut corners, and fail to gather the right information to position a product for the registration trials. In the worst of cases, Phase 2 trials are sufficiently misguided to require a “do-over”. So I’ve imagined a list of protocol design adventures that can make Phase 2 as harrowing as a blindfolded drive down the Interstate:
Misstep 1: Use a limited dose range. You already “know” the “right” dose, so why waste time with dose finding? Keep it simple, and don’t bother with those wimpy low doses that are sure to fail, or those scary high doses that will just cause side effects you don’t want anyone to know about. The inevitable result is an inconclusive dose response and a need to repeat the study with more dose levels and/or a different dose range.
Misstep 2: Don’t bother with that control group. You already “know” your drug works; all you need is a few cooperative patients to convince everyone else. Placebos are for sissies! The absence of a control group inevitably means the data has no true basis for comparison or calibration; to gain this information, a new trial must be conducted, this time with a control arm.
Misstep 3: Biostatisticians are for sissies too! Don’t worry about sample size and power – just go with your gut. Why bother with 120 patients if, under the more optimistic scenario, it will only take 60 to show efficacy? This is probably the most frequent mistake made and the most costly. The incremental cost of using sufficient patients to demonstrate a significant effect is the difference between success and failure of a clinical program.
Misstep 4: Define your own patient population in a data-free zone. Don’t fret that your animal and human data all point to a specific and limited target population – test the drug in a much larger and heterogeneous group, because that’s the largest market opportunity! The problem here is that the trial has too many drug failures, and the resulting “noise” drowns out the efficacy that may exist in a subpopulation of patients who truly benefit.
I’m sure you have your own favorite examples, which I would enjoy hearing about. Together, we can probably make a very long, and instructive, list of what NOT to do in Phase 2. Send me your favorite Phase 2 Missteps to add to the list.
Copyright © 2011 BioStrategics Consulting Ltd.