Terminology for meta content inside text documents

My problem is that I want to systematically handle document internal meta-content in NLP processing, but I don't know how to find relevant resources.

By meta-content I'm referring to content that exists within the documents in the corpus, but that describe/refer to other content in the document. For example, in the standard memo headers:

From: Bob
To: Alice
Date: 3 March 2022
Re: Quarterly Sales Figures

body

the strings From, To, Date, Re are meta content that specify the role/function/meaning of the content that follows them. What term/terms are used in the NLP processing community for referring to this kind of meta-content?

All I can think of is meta-data, but in the NLP context, that applies to attaching metadata to elements of the test context. Instead I'm looking for approaches for how to handle these types of elements in a processing workflow; and to do that I need to know the correct terminology.