http://scopeblog.stanford.edu/archives/2011/04/is-type-2-diabe.html So does this smack of pharma trying to sell more rituximab (dangerous potential side effects http://www.rxlist.com/rituxan-drug.htm )
It's been known for a while that 10-15% of people diagnosed type 2 have antibodies associated with type 1, even with an onset and disease course characteristic of type 2 diabetes. I think there's evidence that dysglycemia furthers diabetogenic autoimmunity- in particular, the portion of people with genes for monogenic diabetes who then develop type 1.