This topic came up yesterday on the PKP mailing list, and James suggested I might want to comment from a librarian perspective.
On a practical note, I suspect that what the original editor may have been asking is some way of indicating or certifying that a particular article is the immutable, "final" version. The short answer is that, in OJS, once an article has been published (made available, either openly or via subscription), that should constitute the "version of record".
Whether or not a DOI (or, say, just the URL) is used to uniquely identify the article doesn't matter so much as long as the content doesn't change after it has been published -- in OJS parlance, that new galleys aren't uploaded to replace the originally-published ones. This can be a fuzzy line, for example, what constitutes typographic changes or corrections versus what requires publishing an erratum or formal correction (or even retraction).
"Version of record" becomes more important if a journal is publishing content ahead of typesetting and proofing, and if it allows author self-archiving in advance of journal publication. In these cases, again, the journal has to decide what their policy is, and enforce it via their own processes (eg. not adding a DOI or other metadata to an article until it is considered "final").
So, I think James' recommendation stands: while it doesn't really make sense to have an explicit "version of record" flag, depending on the journal's policies and practices, you could probably derive it from the metadata that's available.