There is no way to do this. The current pdf conversion engine destroys any data which would allow this to happen. That engine is on the way out though, a new conversion engine is in the works. However I don't believe seamlessly converting the existing TOC is at the top of the to-do list, but it could get added in at some point.
You can also try enabling the Preprocess option under structure detection - this enables some more aggressive searching to identify different sections of the pdf - it may do what you want in some cases.
|