For APIs that are small molecules, as long as your reactor is closed and what is outside the reactor does not really matter, what you really care about is what the API sees (if you pretend you are the API). With biological processes, it tends to be a little tighter specification, is that what you’re getting at?

Well conceivably: for example, in the example that you described, does “closed” mean the lid is on the entrance to the tank or does it mean the lid is in place and sealed so it’s gas-tight? No, we are talking about contained systems, totally contained where the API, the intermediate or even the raw materials or API starting materials are not exposed to the environment. It is the same for biotech, however most contained biotech procedures that I have seen are typically done in a Class 100,000 environment and as you know there are a lot of things that are not controlled in this Class of environment. For example the operators will typically wear some sort of a uniform but there is no attempt to cover all body surfaces. So I would agree that there are probably some similarities between biotech and small chemical entities. An example is where we needed to add a tank to do a refolding step and the tank was going to be located in a room adjacent to the facility that wasn’t in a classified area. Since the material would be pumped into the tank, where the refolding took place, and then pumped back out into the facility, would you really need to have the strict environmental classification for that room? The question is what happens if you have to open up the system? A typical problem in purification is column chromatography where people argue that the column is certainly not a closed system, but that is operated as a contained system. That is fine as long as you’re talking about routine operation of the column, however you have to think about how you add and take the load off the column and also how you pack the column because typically the column has to be physically open when you’re packing it. But it is not that these situations are impossible to deal with it is just that you may need other kinds of controls whenever a part of the system has to be opened. I would like to clarify that under no circumstances would a guidance document that is not legally binding supersede a regulation and especially an established regulation that is legally binding. If that were the case with Q7A, FDA would have never signed off on this document. We were very conscious that Q7A, particularly the section on APIs produced by biotech fermentation, should not be in conflict with the biological regulations.