I have two types of data, one is 200 drugs' plasma concentrations, another is the binding affinity(Ki, Kd or IC50) between drug and its targets downloaded in Chembl database.
Now I want to filter the drug-target interactions using these two data. In my opinions, when the binding affinity is small enough, in other words, the IC50 is too large, normal dose of drug cannot let the target work, so these targets should be abandoned. To this end, I compare the drug plasma concentration with the IC50, for example, If I have compound X binding to proteins A and B with IC50 at 0.1 uM and 10 uM, If the drug X's plasma concentration is 1 uM, then only X-A would be retained.
But a big problem is IC50 is usually measured in vitro, and targets are distributed in specific tissues, so the plasma concentration may be not equal to the concentration in vitro, or the concentration in tissues. So is my original idea feasible? How can I convert the plasma concentration to the concentration in tissues? Thank you very much.
Dear Jixin Cao,
I see different aspects you may need to look into.
1-Regarding the relevant concentration in vivo ("converting" plasma to tissue concentration). A first point is that usually only the free drug is available to interact with the target. So you should look for plasma protein binding/unbound fraction (fu) information, and calculate free concentrations (Cu=Conc x fu). Because fu can actually be small for some drugs, it can have a strong impact. Then you may assume that average Cu in tissue is the same as in plasma. This will be a reasonable assumption for small molecules with high permeability across membranes. It may be acceptable for a small molecule with a target in extracellular fluid, even with limited permeability, in most tissues (but not all, e.g. brain penetration may be hindered). It will generally not be true for biologics or within cells for low permeability compounds.
2-Free fraction in in vitro assay may also need to be considered. Usually drug with low fu in plasma may also experience binding in vitro. This may need to be factored in (and probably difficult to find in literature).
3-Are all the values from the same in vitro assay, what do they represent? For example, there may be significant differences in potency between different assay set-ups (maybe due to tracer ligand used, its concentration and affinity for target). Assays more representative of the physiological conditions should be favoured.
4-Is IC50 sufficient? I would agree that if drug concentration is much lower than IC50, no relevant interaction may be expected. Based on concentration and Ki/Kd/IC50, you could estimate target occupancy. However, if the aim is to estimate efficacious doses (or which drugs may exert significant pharmacological effects on which targets), then IC50 may not be sufficient (although the drug interacts with the target it may not lead to much observable effects). It is not uncommon that > IC75 or IC90 is needed to achieve efficacy.
5-Plasma time profiles. Do you know when the plasma concentrations were measured with respect to dose? Was it at Cmax, Cmin, other times? If Cmax, then your estimate will be relevant at that time but target occupancy may be much lower overall (i.e. good target engagement at Cmax may not lead to efficacy). Some targets will require to be inhibited by 90% (Cu>=IC90) for a substantial part of the dosing interval to lead to pharmacological effects in vivo.
In summary, drug with plasma concentrations much below IC50 are unlikely to have relevant pharmacology on that target. Quite often plasma concentrations need to be much higher than in vitro assays to lead to significant target engagement, let alone efficacy. A number of considerations / corrections need to be done for a better comparison.
The well accepted “free drug hypothesis” for small molecule drugs assumes that only the free (unbound) drug concentration at the therapeutic target is able to elicit pharmacological effect. Unbound (free) drug concentrations in plasma are readily measurable and are often used as surrogates for the drug concentrations at the site of pharmacological action in PKPD analysis and clinical dose projection in drug discovery. Furthermore, for permeable compounds at pharmacokinetic steady-state, the tissue free drug concentration is likely a close approximation of that in plasma. However, several factors can create and maintain dis-equilibrium between the free drug concentration in plasma and tissue leading to free drug concentration asymmetry. These factors include drug uptake and extrusion mechanisms involving uptake and efflux drug transporters, intracellular biotransformation of prodrugs, membrane receptor-mediated uptake of antibody-drug conjugates, pH gradients, unique distribution properties (covalent binders, nanoparticles), and local drug delivery (e.g. inhalation). The impact of these factors on the free drug concentrations in tissues can be represented by Kp,uu, the ratio of free drug concentration between tissue and plasma at steady state. This review will focus on situations in which tissue free drug concentrations may be different from those in plasma (e.g. Kp,uu > or
The concentration in tissues is not often measured directly, but you might have other information you could use to get an idea of the partitioning between plasma and tissues (conc tissues/conc plasma) such as the Log P (measured in vitro) or volume of distribution (Vd or Vss, measured in vivo). You could use this to weight the plasma concentration data e.g. molecules with higher Log P, Vd or Vss ought to higher tissue concentration.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs