The Foundation for Taxpayer and Consumer Rights in Santa Monica, Calif., and the Public Patent Foundation in New York are alleging that when University of Wisconsin-Madison researcher James Thomson isolated human embryonic stem cells in 1998, he wasn't the first to do it.
IPBiz notes that a lot hinges on the meaning of the word "it." If "it" is "isolating human embryonic stem cells," then Thomson was first. The papers filed with the re-exam request allege both anticipation (exactly this was done before) and obviousness grounds for invalidity. Since human stem cells weren't done before (and the reference cited by PubPat does NOT explain how to isolate human stem cells), the anticipation argument won't go far. IPBiz has discussed the obviousness arguments (for example, here)
Gallagher mentions the old numbers for re-examinations at the USPTO: One or more of a patent's claims are changed 59% of the time when a third party has requested a re-exam, said Brigid Quinn, a patent office spokeswoman. All of the claims are confirmed 29% of the time and the patent is canceled 12% of the time, she said. IPBiz notes that a lot of re-exams have been conducted since these numbers were generated.