It is known that although only the (S)-enantiomer of the infamous sedative thalidomide possesses teratogenic properties, it is not very useful to administer the pure (R)-enantiomer since it is racemized within the body. Are there other known examples of drugs that are stereoconverted in vivo? How does the body perform the conversion?
Although I can't think of any drug examples other than thalidomide, here's information on thalidomide's mechanism:
The chiral carbon of thalidomide can tautomerize in basic conditions into an enol, which is achiral. A reversal back to the ketone results in a mix of (R) and (S) enantiomers.
In the body, this tautomerization is generally catalyzed by basic amino acids. Specifically, albumin is the main catalyst in humans.
While this is beyond the scope of your question, the reason that only (S)-thalidomide causes birth defects is that it can insert itself into DNA and suppresses certain genes necessary for embryonic development.
Drugs racemised in-vivo
The following table shows some of the most common (not an exhaustive list) of drugs racemised in-vivo.
Since the biological targets are necessarily chiral, the molecules formed upon epimerisation (inversion at the centre being racemised) are usually inactive against the desired target (annoying, but ultimately fine), issues only arise when the epimerised molecule has its own biological activity, which is often undesirable (for instance in the case of toxicity of thalidomide).
Pathways of racemisation
The mechanisms by which drugs are racemised is not always well known/studied (i.e. it isn't always possible to pin down the exact thing responsible, however two generalised 'mechanistic pathways' have been proposed: 1
1) formation of a covalent intermediate followed by a stepwise enzyme-catalyzed transformation
2) interaction of two opposing metabolic processes, such as oxidation and reduction
One thing that is generally agreed upon, is that the racemisation is the result of more than simple deprotonation/protonation (such as the way chiral centres adjacent to aldehydes tend to epimerise), since the pH in the body is, in most places, not basic enough to do so.
The interconversion of (R) and (S) ibuprofen is one example where a mechanism has been proposed, and involves multiple intermediates:
1: Curr. Drug Metab. 2004, 5, 517–533.
Apologies for the lack of reference to the table, it was taken from a set of lecture notes, though I suspect the lecturer had previously copied it from the primary literature somewhere
Ibuprofen is another example (have a look at e.g. here).
Ibuprofen is an optically active compound with both S and R-isomers, of which the (S)-isomer is the more biologically active. Although it is possible to isolate the enantiopure compound, Ibuprofen is produced industrially as a racemate. An isomerase (alpha-methylacyl-CoA racemase) converts (R)-ibuprofen to the active (S)-enantiomer.