How To Design a Clinical Trial
Designing and executing a clinical trial that meets scientific and marketing requirements can be a tall order. Lots of variable exist, and often a meaningful clinical study result is a moving target. So study design requires significant expertise in the therapeutic area, and an understanding of market dynamics.
Here are four main questions to ask.
1.) What are the rationale and central questions to the study? There are a number of questions that need answers, but in general, clinical studies should start with one or two central questions, and a reason for why the material should be studied. This results in the development of primary and secondary endpoints. Are you trying to see whether a nutritional product can improve joint pain in baby boomers, or muscle pain in athletes? Because the study design may be completely different for what may appear to be very similar studies.
How the product will be perceived by the market post-study is also important. What study endpoints will allow for solid marketing claims? If your product has a significant effect in the study, will it help to differentiate your product against the leaders in the category? In the drug industry, studies on new products are compared against the “standard of care,” and the approach for supplement clinicals can take the same approach, particularly if the product is not very well differentiated in other ways. Are there new mechanisms of action or emerging markers that can be added as secondary endpoints, which would help to differentiate your product?
Accumulation of data to support safety and global regulatory acceptance such as GRAS determinations should always be an objective, so any efficacy study is also a great opportunity to inexpensively accumulate safety data.
2.) What is the dose? Often, this is the most challenging and critical question across all drug and nutrition clinical studies. For many products that are complex mixtures of active compounds, pharmacokinetics or bioavailability is unknown or untenable, making dosing a wild guess. In cases where there are only a couple active compounds, bioavailability should be assessed before moving on to clinical efficacy trials.
In cases where bioavailability cannot be easily determined, a dose-response study (using multiple doses) should be performed. Ideally, a dose-response study observes a small effect at a lower dose, and a greater effect at a higher dose. In other cases, a linear dose-response relationship should not be assumed; a higher dose may not work as well (or reveal safety issues) compared to a lower dose.
Market considerations, such as cost per day and number of capsules should also be included in this evaluation. While a randomized, placebo-controlled clinical trial is wonderful to have, if the product never reaches the shelf (or the dose is too high for the consumer to stomach) then the best-designed study is like a tree falling in the woods.
3.) How many subjects are needed for the study to be adequately powered? A minimum requirement today for nutritional products is that the changes in the group taking the active dose must be significantly different than the changes in the placebo or control group. It makes no sense to design and invest in a study that will show no difference between your product and a sugar pill. For some subjective measures such as pain, the placebo effect and inter-individual variation can be very high, due to the subjective and ever-changing nature of pain perception. In this case, the number of subjects required to get reliable statistical separation between the active versus control groups is relatively high. For other endpoints, such as blood concentrations of actives in pharmacokinetic studies, placebo effects are almost nil, and therefore a lower ‘n’ is likely to result in significant changes versus controls.
4.) What is the budget and timeline? Research is an investment, one that can be expensive and time-consuming. For example, if the therapeutic area and endpoints include testing of blood markers, then the drawing, processing and testing of blood samples is a major cost center in the research budget. Common blood markers such as blood lipids are relatively easy using standard kits, while other less standard markers can require method development and increase costs, and may provide unreliable data that needs to be repeated.
A university-based study offers the independence and clout of world-class clinical studies, but the prestige can be balanced with increased costs and more uncertainty in the timeline, particularly when your study is relatively small and relies on shared resources. While a contract research organization is often faster than a university, this option can also come with greater costs. A research services contract with a detailed protocol and time-based milestones is critical to have in place.
Ethical approval (typically through an Institutional Review Board, or IRB) is also required for all human studies. Some research centers can get IRB approval within a month, while others are mired in bureaucracy and generally take six months or more.
Recruiting also contributes to the study timeline. If you are excluding a lot of lifestyle factors, then your available population is low, and getting the required number of subjects can be costly if not impossible. Many clinical studies never get off the ground when recruiting is not taken into account.
Lastly, it is critical to do the homework up front and ask a lot of questions. Make sure you have someone in your corner, who speaks the language and is looking out for your best interests. Only then can you ensure the returns on your research investment are maximized.
By: Blake Ebersole
This article was previously published in Natural Products Insider, June 2015.