Paradise Lost and Found in Answers to Challenges in E&L: Extractable Study Variables (Part1)
Welcome to another blog entry from Maven E&L. Check out others here. One of the things about the topic of extractables and leachables (E&L), which is both a blessing and a curse is the lack of a clear instruction set.
It's a blessing because, for a scientist like me it gives me interest and stimulation to find new and interesting ways to tackle the challenges which E&L poses.
It's a curse perhaps because this leads to difference of view and opinion on the right way to approach a given task.
I have written before about some of these task such as design of extractable or leachable studies but on this occasion I wanted to highlight one particular variable which is part of an extractable study.
This is the selection of the ratio of sample to extraction solvent.
On the face of it's a simple enough thing, and certainly something you will need to address in every extractable study. So, where to start. Do you just weigh in (excuse the pun) and begin your extraction?
Analytical chemists working in this area will want to achieve a number of things, among these are:
Have confident that extraction will remove substances from the material or component in question
Achieve a detection limit, whereby substances removed can be precisely and accurately determined
There are many factors which can impact this but clearly the amount of sample compared to extraction solvent you used in your experiment is a major one.
Too little sample and you cannot expect to achieve the required detection limit
Too much sample and you run the risk of incomplete extraction due to solubility and other concerns
So you might think there would be some very clear guidelines or standards on this one. Well the answer here is like all things E&L not as clear as you might like or expect.
Therefore, at this point I like to ask a favour. If you know of a peer reviewed source which might be quoted can be please leave it as a comment here or email me at Jason.Creasey@MavenEandL.com.
So what is out there? I like to start with a reference to Biophorum. This organisation has in recent year published widely on E&L. In April they updated their best practice guide for extractables testing of SUS component which many people have found very useful. Within this guide they recommend a sample to surface area to the extracting solvent ratio volume (cm2/mL) ratio of 6:1 or higher and they include a reference.
That reference is to USP chapter <88>, This chapter is a chapter about in-vivo biocompatibility of plastics. A strange choice you might think but one of the very few they could have chosen and probably the most relevant, and as you will see below commonly used.
So what does USP <88> have to say on this subject?
Well, it lays out six different sample to volume ratios based around sample form in units of square centimeters per 20mL of extracting medium. These range from 120 cm2 per 20mL to 25cm2 per 20mL and samples as films or sheet, tubing, slabs (moulded items) and elastomers which one should use to compare a sample prepare a sample extract for an in-vivo toxicological study. The gifted mathematicians among you will have worked out that that give ratios from 6:1 to 1.25:1.
I do not know where these values come from. Do they seem reasonable? Yes.
Are they referenced to a source? No
Indeed this chapter has been widely used in other USP relevant chapters, USP <665> the yet to be finalised chapter on plastics used in manufacture of pharmaceuticals and biopharmaceuticals also uses 6:1 in general but does suggest filters should adopt a different ratio. Its accompanying guidance <1665> suggests a target of 6:1 and contains the phrase "The target SA/V of 6cm2/mL was adopted from relevant regulatory guidelines and standards"
My point here is not to suggest this ratio or approach is incorrect but to highlight it seems to have been created and is now being adopted on the basis of very little science. A test for biocompatibility seems a very different requirement to a modern analytical test for substances at the trace level. I suspect both Biophorum and USP found it as difficult as I have to find any good sources to reference on this subject.
So I am requesting, if you can, to supply a rationale. If the rationale is it seems about right, well that is OK. However, wouldn't it be great if we had a better basis than that?
USP <1663> in its guide to assessment of extractables associated with pharmaceutical packaging systems, does not set a ratio. It introduces a term "extraction stoichiometry" to discuss the aforementioned ratio of sample to solvent but does not offer a value, approach or reference to move this forward. Equally the PQRI documents have nothing to add - "Controlled Extraction Studies should employ vigorous extraction with multiple solvents of varying polarity."
So let's look elsewhere, ISO10993, part 18 has again recently published a standard on chemical charactisteration. Within that standard, Annex D is a whole chapter on "principles of sample extraction". Which recognises extraction is complex, in its 11 pages on the subject it suggests, "..a surface area/volume ratio that exceeds clinical use exposure..".
Nothing to disagree with there but goes on to describe in quite some details a whole number of other factors (time, temperature, pH) but nothing which provides a guide to the ratio.
In conclusion, extractable experiments are being conducted right now and choices are being made on that sample surface area to volume ratio during the extraction step. Please help the E&L community by thinking about why you are selecting the a particular ratio and when you have thought about it, publish that thought process for us all to reference, so we will not have to use a USP <88> chapter dedicated to toxicological in-vivo testing.
If you want to discuss this and other E&L topics some more, please join my newly established E&L forum here.