Are you interested in only those cases where a secondary endpoint later became a primary endpoint, or are you interested in the more general cases of drugs that failed in their first indication but succeeded in a subsequent indication? The latter set includes many examples of which Viagra is perhaps the most famous.
A couple that are going on now: *Allos, new pivotal based on prior ph III trial subset with breast cancer. *Intermune: new pivotal based on prior pivotal data showing less severe ipf cases had mortality benefits
For a good example of one that did work and the only one I can think of, NTMD, subset in african americans. Rest of study failed miserably.
Market Data 
Markets 
AI
