An intention-to-treat analysis is a way to include patients who, for whatever reason, failed to comply with a clinical trial's protocol following randomization. Imagine a clinical trial designed to measure experimental treatment A against standard treatment B. After patients are randomized into their groups, some patients in the A group might not show up for a scheduled dosing or drop out of the trial before investigators can measure the trial's outcome goal. An intention-to-treat analysis would include these patients as members of group A regardless.

Getty Images Scientists At Work

Image source: Getty Images.

It's easy to see how an intention-to-treat analysis can dilute an experimental new drug's efficacy results. If you're investing in smaller biotech companies with futures that hang on clinical trial data, you should know the U.S. Food and Drug Administration has a long-standing policy that insists on maintenance of existing data for all patients, regardless of compliance issues. Across the Atlantic, regulators aren't always as strict.

Look at these, not those

As an example, take a look at how PTC Therapeutics learned this the hard way. In one analysis of a phase 3 trial designed to support approval of its Duchenne muscular dystrophy drug, Translarna, patients who began the study capable of walking 300 meters in six minutes showed a highly significant improvement over similar patients receiving a placebo.

However, the intent-to-treat population included patients capable of walking only 150 meters, and the complete group treated with Translarna didn't show a significant improvement.

PTC Therapeutics tried to submit an application based on its analysis of the group of patients who outperformed a placebo, but the FDA refused to accept it. The European Medicines Agency, however, allowed the drug a conditional approval that could be revoked if safety issues outweighed its ability to slow progression of the muscle-wasting disease.

Beware of modified analyses

Keep your eyes open for modified intent-to-treat analyses massaged to make a drug look better than it might otherwise. For example, removing patients who never started treatment after they were randomized might be sensible. On the other hand, patients without outcome data should be included if they discontinued treatment because of side effects, as they represent real-world possibilities.

In a nutshell, the FDA demands strictly objective endpoints in the clinical trials used to support a new drug's application. Modified analyses might impress some investors, but the agency isn't nearly as flexible.

This article is part of The Motley Fool's Knowledge Center, which was created based on the collected wisdom of a fantastic community of investors. We'd love to hear your questions, thoughts, and opinions on the Knowledge Center in general or this page in particular. Your input will help us help the world invest, better! Email us at knowledgecenter@fool.com. Thanks -- and Fool on!

The Motley Fool has no position in any of the stocks mentioned. Try any of our Foolish newsletter services free for 30 days. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.