Innovative Study Design in Phase I: Adaptive Clinical Trials

Smarter Science, Smarter Study Design

We all know that some drugs work better for certain age groups, genders, and other demographic categories. But what if a drug works well for just a tiny subset of patients within an already targeted group?  How do we identify these patients in a big sea of participants?

Nowhere is this question more important than in oncology research, in which the variety of symptoms and characteristics among patients with similar cancers adds an additional layer of complexity to an already perplexing condition. Often, cancer treatments are highly effective for about 5% of patients who share certain biomarkers, or measurable biological indicators (e.g., tumors with specific molecular features). However, researchers’ ability to develop these highly targeted treatments has been thwarted by traditional clinical trial design, in which simple comparisons between large, roughly homogenous patient groups determine the results. As a result, the benefit of a drug for a very small patient subgroup is often masked.

Recognizing the weaknesses of conventional clinical trials, cancer researchers have adopted adaptive study design. Adaptive trials “learn” as they accumulate data, using early results to adjust aspects of the study protocol, such as recruitment, dosing, and sampling occasions. Most commonly, researchers eliminate groups of patients with nonresponsive biomarkers early on and increase recruitment of patients with a good chance of positive results. This type of patient selection may seem blasphemous, but for the costly treatments and fast response needed in cancer trials, it’s a welcome innovation.

Phase I Redefined

Phase I takes on new meaning in adaptive trials. Most cancer drug trials already include patients in Phase I in order to evaluate tolerability, but in adaptive Phase I trials, effectiveness is also considered. In fact, researchers use early results gathered in adaptive Phase I trials to define which research questions are worth pursuing in Phase II and III.  In traditional design, of course, the later phases are planned well in advance.

By evaluating biomarkers early on, researchers get a head start on identifying the patients that are most likely to benefit from a treatment and can eliminate the rest sooner. This allows for smaller, faster trials that lead to lower overall costs. Likewise, shorter study timelines mean that patients in need get help quicker.

Although traditional Phase I trials that test solely for safety and PK/PD in healthy volunteers are still the norm outside of oncology, this is likely to change in coming years. As the industry moves toward a more personalized approach to medicine (as many industry experts predict), adaptive clinical trials will become commonplace across therapeutic areas. In turn, early phase research will have an increasingly important role in identifying which drugs work and for whom.