Precedent often trumps science when it comes to drug development. A common refrain is, “This is what we did at xxx”, “This is how they did it” and so on. So much so that we are doing things because that was how it was done. While there are many reasons to not reinvent the wheel and to go with precedent, there are several occasions when the wheels of development can be made more efficient. I am certain you can relate to one or more of these.
The use of Data Monitoring Committees (DMCs) is an example. About 20 to 25 years ago, DMCs were unusual. The Sponsor took responsibility for trial strategy and execution. But lately, DMCs have become de rigueur. The Contract Research Organization expects it, the Institutional Review Board sometimes insists on it. Current FDA regulations, however, do not make it obligatory to use DMCs in trials, save under 21 CFR 50.24(a)(7)(iv) for research studies in emergency settings wherein informed consent is excepted. As anyone who has used a DMC will testify, putting together and providing unblinded data to a DMC on a recurring basis is an onerous task. Therefore, whether one needs a DMC should be carefully considered. Recently, we were involved in a trial for a topical where the Sponsor was planning on a DMC. A risk assessment revealed little, if any, risk; for example, the topical had systemic exposures that were 1000-fold lower than that at the maximum tolerated oral dose. Our recommendation to not have a DMC was met with resistance, but we prevailed. The study completed uneventfully.
We also challenge the notion of the double-blind study design. For non-pivotal trials in particular, we believe the traditional double-blind design, where the Sponsor, Principal Investigator and patient are all blinded, is not sacrosanct. As an example, for the longest time, ‘First in Human’ studies have been double blinded as this is what has always been done. However, in such a setting, this time consuming and laborious study design is dispensable. There is no reason for the Sponsor to be blinded. Instead, an unblinded Sponsor can be invaluable, reviewing accruing PK and safety data and intervening promptly, planning the next study, etc. In a recent instance, when a Sponsor unblinded design was proposed for an FIH study, three of the four CROs under consideration suggested we replace it with a DMC.
Another example is in pediatric drug development. Here, it is common practice to divide the pediatric patient population into smaller age groups because that is how it is done; quite often there is no scientific basis for this. While it may be prudent to test a drug in adults (> 18 years), there is no real need to then test it in adolescents (12 to 18) before testing in children (say < 12) – though this is often the norm. Generally speaking, if the anticipated exposure in the younger age groups does not exceed the maximum exposure in adults, then it may be safe to proceed with evaluating in children.
Rare disease is often made to seem like it is an indication by itself. Nothing can be farther from the truth. Each rare disease brings its own development plan, just as for a disease which is not rare. Additionally, rare diseases encompass such a wide prevalence that it is almost critical to think about indication and prevalence and not just rare disease.
In most instances, clinical development seems to follow a guideline, but as one regulator remarked: “Rigorously following Regulatory Guidelines is the last refuge of those who don’t know how to develop medicines!!” (https://www.ema.europa.eu/en/documents/presentation/presentation-key-aspects-non-clinical-pharmacology-pharmacokinetics-evaluation-safety-david-r-jones_en.pdf)
Okay, we may not endorse this in its entirety, but……could not help quoting.
For more blogs, please visit: https://www.rxmd.com/insights