With PDFNet SDK you can achieve this in couple of ways:
· Create a temp (memory) document B and then copy a set of pages from the source document A (using B.InsertPages(A, pages)). Then convert B to XOD.
· Remove pages you don’t need from A then convert to XOD.
The latter option is more efficient.
If you are using DocPub you can do the same with -a option. For example ...
docpub -a 1-10,12,18 in.pdf -f xod