Tweetorial: Oral Absorption vs. Oral Bioavailability
13 August 2021
This is the first truly epic tweetorial thread I’ve written, and it got great traction: over 8,000 views. This is a really fundamental topic. And at the core of it something that’s bothered me for many years: the relentless focus in the literature on optimization of bioavailability rather than absorption. There are deep flaws, perhaps fatal flaws, in all of those kinds of studies. Only recently have medicinal chemists really started paying proper attention to oral absorption and doing the small amount of math needed to do this analysis right.
So here we set the record straight, and put everyone on notice: now you know. Change course and fly right!
One small regret: on post 7, I wish I’d edited the Monopoly card to say “Go Directly to the Liver”. Alas.
Time for another PK tweetorial, this time about the oft-misunderstood distinction between oral absorption and oral bioavailability. These are not the same thing, with different meanings and different implications for medicinal chemistry. 1/
Warning for the math-phobic: this topic does require a smidge of math to understand, but nothing beyond basic high school algebra. Not too scary, promise. 😱2/
Oral bioavailability (F) first. This is an experimentally determined parameter defined as the ratio of the area under the drug concentration-time curve (AUC) from oral (PO) administration to the AUC from intravenous (IV) administration, normalized for the respective doses. 3/
Equation form: F = (AUC_po / AUC_iv) * (dose iv / dose po) ; sometimes this is multiplied by 100 to express as a percentage (%F) rather than a fraction (F). 4/
F reflects the fraction of the dose that makes it to systemic circulation following oral administration. It's compared to a theoretical maximum in systemic circulation from IV dosing, which removes all the barriers between the gut and systemic circulation. 5/
But now let's consider what those barriers are. There are 3 main ones: 1) absorption through the gut wall, usually through passive diffusion; 2) metabolism in the gut, which can come from both microbiome sources and intestinal CYP enzymes; 3) first-pass hepatic metabolism. 6/
The last one is important & follows basic physiology. After absorption, drugs enter the portal vein & go straight to the liver. All oral drugs must pass through the liver once (thus "first-pass effect") before reaching systemic circulation. Do not pass Go, do not collect $200. 7/
Thus, drugs are subject to clearance in the liver before reaching systemic circulation. Clearance is obtained from an IV PK experiment, where it's basically the only thing operating on the drug, because the other barriers (absorption and gut metabolism) are not present. 8/
So another, theoretical formulation of oral bioavailability is F = fa * fg * fh, where fa is the fraction absorbed, fg is the fraction gut available (gut metabolism), and fh is the fraction hepatically available (a term for the first pass effect). 9/
Now it starts to become clearer that absorption (fa) is a component of bioavailability (F), but there are other, very relevant factors at play, principally the first-pass effect as measured by fh. 10/
Fortunately, with some assumptions, we can get a handle on fa. fh can also be determined experimentally. fh = 1-(Cl/Qh), where Cl is the experimental IV clearance and Qh is hepatic blood flow, a basic physiologic parameter that's well-established for many species. 11/
Math time! F = fa * fg * fh. Substituting for fh = 1-(Cl/Qh) gives F = fa * fg * (1- (Cl/Qh)). So then rearranging a bit, fa * fg = F / (1-(Cl/Qh)). 12/
We can now estimate the fraction absorbed (technically the product of fa * fg, as these are difficult to independently dissect out) entirely from experimental parameters. But it requires both an IV and a PO PK experiment to do so. 13/
An important thing to grasp here is that bioavailability is dependent on, and limited by, hepatic (liver) clearance. Consider a theoretical case where absorption is complete (fa = 1 & fg = 1). Where is the upper limit? 14/
(Forgot to put this in the original thread, ugh!)
The absolute maximum bioavailability obtainable is F_max = fa * fg * fh = 1 * 1 * (1-(Cl/Qh)) = 1-(Cl/Qh). This should make intuitive sense. If Cl/Qh = 0.25 (or 25%), 25% of the drug is removed by the liver on first pass. In this case, bioavailability CANNOT exceed 75%. 15/
We can thus approximate the fraction absorbed (technically, fa * fg) as the ratio of the experimental bioavailability divided by this theoretical maximum bioavailability if absorption was complete. This has big consequences for medicinal chemists during optimization. 16/
Here's a test case of what I like to call the "Tale of Two Compounds". Assume they’re otherwise identical, and both have a measured bioavailability of 5%. This seems to be "poor", but... is it? Compound A has a clearance of 90% Qh, and compound B has a clearance of 20% Qh. 17/
Compound A's theoretical maximum bioavailability is 10%. Compound B's is 80%. Now when we calculate the apparent fraction absorbed, A's is (5%/10%) = 0.50, whereas B's is (5%/80%) = 0.063. A is relatively well-absorbed, whereas B is poorly absorbed. 18/
A is limited by its awful clearance, and the chemist should focus on fixing that first. B is limited by its awful absorption, which is a different set of challenges. But this is not obvious at first if you only focus on the bioavailability, which is the same in both cases. 19/
(Aside for the nitpickers in the audience: technically the focus for A should be on reducing *unbound* clearance, as usually total clearance is what’s measured, and has no correction for protein binding. Free drug hypothesis we’ll save for another day.) /20
There are many analyses out there in the literature on ways to improve bioavailability. In my mind, every one of these studies that tracks trends in F and not fa, which is corrected for clearance, is flawed. Many medicinal chemists, judged by publications, do the same. /21
Bottom line: both IV & PO PK legs are required to develop an integrated understanding of an orally dosed compound's pharmacokinetics. Focusing just on raw F and not calculating the apparent fraction absorbed can send you careening off trying to solve the wrong problem. /end
