At a recent debate on clinical trials for Pulmonary Arterial Hypertension (sponsored by the Drug Information Association and the Pulmonary Vascular Research Institute) Bob Temple (CDER Deputy Director for Clinical Science) commented that the distribution of positive results for individual patients (as well as a drug's mean benefit across the population of treated patients in a clinical trial) can be useful in assessing efficacy for pulmonary arterial hypertension.
The issue is – the few versus the many.
"It really isn't the average that matters to people most. For various reasons - analytical, historical and others - we use mean results. But that's not what happens to individual people. ...Of greater interest, this is true for a lot of situations, are the effects on individuals or the distribution of those effects, which we are not so used to measuring."
Even if a drug only has a 10 percent effect, Temple said, "it might have a rather large effect in a fraction of the population," so a look at distribution is useful. "We do think about that, mostly though," he conceded, "when you have context that has a safety problem or the effect is particularly important."
There are increasing opportunities in early studies to identify the responders and incorporate that information into the design of later-phase trials, Temple noted. If sponsors find a likely response predictor, "genomic or some other kind of finding, there's nothing that stops the later trials from stratifying that predictor and analyzing the groups separately or even making the result in the responder subgroup the primary analysis."
Temple's remarks came during a debate on whether clinical trials for PAH drugs should be required to show that patients in the treatment arm improve their distance by a minimum amount during the six-minute walk.
The question, he noted, is what is a big enough improvement? The answer can come from the patients themselves, in how they feel about walking or their life activities, he suggested, "We're very interested in looking more at well-validated patient reported outcomes."
Let's remember and embrace that last comment, "well-validated patient reported outcomes."
Bob’s comments, coming on the heels of the FDA’s recent documents regarding clinical trial designs (Guidance for Industry: Adaptive Design Clinical Trials for Drugs and Biologics), seem to reinforce the agency’s Critical Path philosophy of broadening decision-making criteria about a drug based strictly on generalized results from large scale RCTs. Trial designs that focus on patient selection for ideal response is a key tenet of the personalized medicine approach that industry and FDA are hoping to create and for which the Critical Path’s Reagan Udall Foundation was created to foster.
That being the case, perhaps we should take the Wisdom of Temple to our nation’s Temple of Wisdom and suggest to Congress that a certain percentage of PDUFA fees be used to fund the FDA’s Critical Path program.
The issue is – the few versus the many.
"It really isn't the average that matters to people most. For various reasons - analytical, historical and others - we use mean results. But that's not what happens to individual people. ...Of greater interest, this is true for a lot of situations, are the effects on individuals or the distribution of those effects, which we are not so used to measuring."
Even if a drug only has a 10 percent effect, Temple said, "it might have a rather large effect in a fraction of the population," so a look at distribution is useful. "We do think about that, mostly though," he conceded, "when you have context that has a safety problem or the effect is particularly important."
There are increasing opportunities in early studies to identify the responders and incorporate that information into the design of later-phase trials, Temple noted. If sponsors find a likely response predictor, "genomic or some other kind of finding, there's nothing that stops the later trials from stratifying that predictor and analyzing the groups separately or even making the result in the responder subgroup the primary analysis."
Temple's remarks came during a debate on whether clinical trials for PAH drugs should be required to show that patients in the treatment arm improve their distance by a minimum amount during the six-minute walk.
The question, he noted, is what is a big enough improvement? The answer can come from the patients themselves, in how they feel about walking or their life activities, he suggested, "We're very interested in looking more at well-validated patient reported outcomes."
Let's remember and embrace that last comment, "well-validated patient reported outcomes."
Bob’s comments, coming on the heels of the FDA’s recent documents regarding clinical trial designs (Guidance for Industry: Adaptive Design Clinical Trials for Drugs and Biologics), seem to reinforce the agency’s Critical Path philosophy of broadening decision-making criteria about a drug based strictly on generalized results from large scale RCTs. Trial designs that focus on patient selection for ideal response is a key tenet of the personalized medicine approach that industry and FDA are hoping to create and for which the Critical Path’s Reagan Udall Foundation was created to foster.
That being the case, perhaps we should take the Wisdom of Temple to our nation’s Temple of Wisdom and suggest to Congress that a certain percentage of PDUFA fees be used to fund the FDA’s Critical Path program.
