I'm assuming you're looking to create this for a PDF transform. I recommend getting a copy of either DITA for Practitioners or DITA for Print. What you need to do is create a plugin to override the base XSL and extract using XPATH to grab both mainbooktitle and topicmeta/title (the exact xpath is more complex than what I've written here).
You haven't specified precisely what you want to do with the titles, so it's difficult to come up with a meaningful example.
HTH.