It's a good question that you (and related to the one Bio asked in a subsequent post) pose here and worth a greater discussion at some point. Leaving aside the particulars of this case (i.e. I haven't had the opportunity as yet to read all that might have been disclosed about these emails, meetings, etc.) and in conjunction with my belief that Dr. Woodcock is, herself, a thoughtful and forward thinking fixture at the Agency, I think generally it's bad form to co-author with a single party in the midst of a drug review process involving that same single party and others.
Historically, the Agency has used the "taskforce" approach in pulling together experts to deal with a variety of issues (some time pressing like this Heparin example, others somewhat less, such as the Clinical Trials Working Group and the Critical Path Initiatives) but in those cases, when documents are authored, they are typically authored by the taskforce or working group. It would be interesting to know in how many cases FDA supported/worked with research that ended up in publication but where FDA authorship was not claimed.
I do not at all subscribe to the idea that attending a conference that happens to also be attended by a company rep itself is a conflict, nor, in a vacuum to co-authoring research pieces to advance public health and safety. The group invited to participate and/or lead such research, however, needs to be considered when FDA is going to lend itself as an author, because without question, it's a positive for any company to have co-published with an FDA official with a prominent role in an ongoing review of their drug.
You can go back to the Bexxar/Zevalin (obviously there are more recent cases as well) to see what happens when you're NOT on the 'right' side with FDA, no matter your data/clinical benefits.
But again, your question is a good one, and worth consideration and discussion.
ST