Hi,

Probably because these are (usually also) not measured in clinical trials. The main interest and focus will be on the active metabolite(s), generally.

Also, I assume metabolites will often not significantly differ that much with their parent drug in terms of Vd, since they own similar (base)structures/backbones.

In terms of pharmacokinetic modeling, the reason this assumption is made is to avoid issues of model identifiability. If you have access to Peter Bonate's textbook on PKPD modeling and simulation, browse the first chapter to understand this, The underlying concept here is that - Volume is a proportionality constant that relates amount in a compartment to the concentration (Amount/Concentration). E.g., after an IV bolus dose, when we are dealing with the central compartment of the parent, we know the "true" amount of parent drug in that compartment, but the amount of parent being converted into metabolite is at best an estimate given the data we have. As it is an estimate, we never know the "true dose" that is going into the metabolite compartment without which the volume term becomes unidentifiable, and hence it is for convenience assumed to be equal to that of the parent. 

Volume is a proportionality constant that relates amount in a compartment to the concentration (Amount/Concentration). E.g., after an IV bolus dose, when we are dealing with the central compartment of the parent, we know the "true" amount of parent drug in that compartment, but the fraction (fm) of parent being converted into metabolite is at best an estimate given the data we have. As it is an estimate, we never know the "true dose" that is going into the metabolite compartment without which the volume term becomes unidentifiable, and hence it is for convenience assumed to be equal to that of the parent. Note that even if you don't make this assumption during your modeling work, the software will spit out a volume estimate, but you have to acknowledge that  the estimate is Vm/fm, where fm is the fraction of parent drug metabolized to metabolite, 

Two cases where you can estimate the volume of the metabolite - 1) if you know the true clearance of the metabolite, for which one has to conduct a study where the metabolite is given directly as a dose and you measure the "true" clearance of the metabolite. You can then use this "true" metabolite clearance in the parent-metabolite model to estimate the volume. 2) if your parent drug is 100% metabolized, then the fraction fm will be 1 and the volume can be identifiable. 

Note: This discussion above assumes that the parent drug is given via IV route of administration where bioavailability, F is 100%. If F

Thanks Vijay. The parent drug is administered orally and is not 100% metabolized. I have "true" clearance values from published studies where the metabolite was orally administered. Is it ok to use these clearance values to estimate volume of distribution of the metabolite in parent-metabolite modelling?

Is the CL of the metabolite, CL/F, i.e., is the bioavailability of the oral metabolite drug 100%? If not, you will be using CL/F as the clearance of metabolite and not CL