Sure - Belimumab for Lupus probably qualifies as one of the most complicated ever.
Also note that I am not sure I would call logical ANDs of different severities of the same type of event a real composite since they don't necessitate the strange equating of different kinds of events. In contrast, try to explain in English the proved benefit of Belimumab and why being 1% worse on x was considered equal and offsetting to a 2% benefit on y. (Note that this isn't meant to be a condemnation of Belimumab - only that true composites, especially within individual patients, are rare for a reason.)