Feature Request: Gut Mucosa Metabolism Scalar

[prvmalik]

Gut metabolism by CYP3A4 and CYP2C19 is poorly estimated by PKSim (see midazolam model file as an example).

This can be addressed by adding a scalar to Vmax or Specific clearance of the relevant CYP process in the intracellular mucosa compartment of the small intestine segments.

This may reflect: A) a higher fraction unbound of the drug in the gut than calculated, making it more available for metabolism B) a higher expression of CYP in the gut than we have parameterized C) longer residence time in the gut mucosal cells than we have parameterized

Simplest way is to offer a scalar.

Delaying the basolateral permeability out of the mucosa to increase residence time and therefore metabolism does not scale well to special populations, limiting its utility. It is also difficult to implement for non-experienced users.

Further we could change the expression of CYP3A4 in the gut mucosa. But this causes issues when you want to model drugs in a DDI scenario that are both substrates of CYP3A4 - if one requires a higher expression level to capture gut metabolism and the other does not.

Paul

[StephanSchaller]

Thanks for taking this up. For which model is gut CYP2C19 metabolism an issue?

[prvmalik]

Fair to ask for proof; there is no existing model I can offer you for failure of PKSim on 2C gut metabolism. But the relevant enzymes where this may be applicable are CYP3A4, CYP2C9 and 2C19, SULT1A and UGT1A based on expression profiles in the gut.

Gut metabolism is well documented in literature as relevant for ~10-20 existing CYP3A4 substrates, with increasing importance based on the properties of modern compounds. See doi: 0.1124/dmd.114.059147 or doi: 10.1517/17425255.4.7.909 for some lists of older compounds where PKSim would likely not capture gut metabolism.

Compounds also tend to be more vulnerable to gut metabolism when they are substrates for efflux transporters, which may effectively increase residence time in gut. It is terribly difficult to model or identify efflux transporters in the intestines with current methods; in these cases a gut metabolism scalar may also be quite helpful.

Perhaps while we are at it we can update regional expression profiles in the mucosa for relevant gut enzymes in the gene expression database from sources such as doi: 10.1111/bcpt.13137

[prvmalik]

See also this paper where CYP2C19 expression had to be decreased to properly capture gut metabolism of R-omeprazole: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7764797/

[prvmalik]

Update: CYP1A1 and UGT isoforms are also relevant to this

[RobinM92]

Dear @prvmalik

Thanks for this, do you know whether this will be implemented into PKSim? And would adapting the relative expression in the intestine be a workaround for now?

Thanks! Robin

[prvmalik]

Hi Robin

No plans that I have heard.

Adjusting relative expression is ok. Few other approaches like estimating basolateral intestinal permeability to increase residence time in the mucosa. Have to ensure that if you adjust relative expression that you use a unique enzyme for each drug when simulating DDIs (which can get laborious and complicated).

Paul