In June, 2018, FDA distributed a guidance document in the form of a Q&A describing under what cases material consistent with the PI but not in the PI could be disseminated.
This guidance clearly defines that information inconsistent with the approved PI cannot be disseminated in promotional material. These include:
1. Unapproved indications – including use as a monotherapy (if approved in combination), use to treat a different stage or therapy of disease, use to treat patients not included in patient population studied for drug approval)
2. Expanded patient populations if PI has limitations to the patient population
3. Conditions of use/handling/storage that are outside of what is in the PI
4. Dosing (e.g., amount, route of administration, strength) different from what is in the PI
FDA also indicated additionally, if the information communicated increases the chance that someone will be harmed by use of the product or if the instructions for how to use the product in the PI were insufficient to use the product in the way described by the communication, then, in both cases, the communication could not be used.
At the heart of this guidance is the idea that promotion has to be truthful and not misleading, some of the parameters of which are described within this guidance. The importance of this guidance is that it allows the pharmaceutical industry to speak to things that they have not allowed previously, most or all of which have been the subject of OPDP action letters in the past.
What the guidance seems to say is that a number of areas that previously were either “gray”, undefined or clearly forbidden are now possible IF all the considerations above for the communication are met. These areas that now appeared to be allowable are:
• Comparative studies – Routinely OPDP has said previously that companies should not talk about these kinds of studies unless FDA has seen the data first. In other words, you had to have that information in the clinical section of your PI.
• Providing context around (and softening of) statements regarding adverse events. In the past if you had nausea as your adverse event and in subsequent studies not in the NDA you found that prophalaxis with nausea-reducing drugs eliminated most of this nausea, you really couldn’t say that as, once again, FDA hadn’t reviewed the data in those studies. This was already mentioned in a 2014 guidance.
• Onset of action – Previously, if your PI was silent to onset of action you couldn’t mention it in your promotional material unless it was a fact but was just not mentioned in your PI
• Long-term safety and efficacy – Previously, you couldn’t talk about long-term efficacy if that data wasn’t in your label and FDA had not seen that data in their review. Now, if data exists, you may be able to market to it without that data being included in a supplement to your PI.
• Sub-group analysis – As part of what was referred to, in a negative way, as ‘data mining’ it used to be prohibited to promote to sub-groups if those sub-groups were not specifically tested for in the endpoints of clinical studies. Now it appears that if the data is present you can promote to these sub-groups.
• Composite endpoints – Previously you couldn’t say anything about the individual endpoints that make up a composite endpoint. Now you can, with proper qualification, give some information about the results of the individual endpoints within a composite endpoint.
• Product convenience and mechanism of action – If new information not in the PI is available then it can be promoted to. Interestingly, for product convenience this extends to comparative studies with competitive products, as well.
• Tolerability with concommittant drugs – Again, this normally had to be spelled out in the PI as did all information about drug-drug interactions. Now it appears that information generated after the drug was approved can be marketed to, even if it is never added to the PI.
Of course, in all these cases the statements have to be truthful and not misleading and, as such would have to be scientifically sound. Depending on the type of claim, the guidance also says that the evidence necessary might be substantial evidence and might only have to be adequate evidence. Thus, you probably couldn’t do a sub-group analysis on blacks if only two blacks were enrolled in your piviotal study. However, if you did an entirely separate study on blacks and it represented substantial evidence, the way I read this guidance is that you could promote to that study.
As is always the case, any information used in promotional material has to be truthful and not misleading. This is always a high bar to hurdle and companies should make sure they are on strong statistical and scientific ground before they make claims that are based on data not addressed in their approved PI.
