The saga over the validity of Genentech's widely licensed recombinant antibody patent, Cabilly 2, inched forward in late February when the US Patent and Trademark Office (USPTO) rejected it officially for the fourth time (though two were partially withdrawn). The patent covers a fundamental method of using host cell culture to make therapeutic antibodies; in 2007, it generated $133 million for Genentech, of San Francisco. Genentech can respond to the rejection, then appeal the decision to the patent office's board of appeals and later to the courts. Cabilly 2 remains valid until the end of the proceedings, which could take up to a decade. The legal costs are trivial compared with the licensing revenue, so the company has every reason to continue fighting. The rejection may be the USPTO's most thorough response to Cabilly 2 since the start of its re-examination in 2005, meaning that “Genentech has some work to do here to overcome this,” says Ivor Elrifi, co-chair of intellectual property at Mintz Levin in Boston. The USPTO argues that Cabilly 2 is a variation of the art described in a number of earlier patents, including Genentech's Cabilly 1. While there are some in the industry who perceive Genentech as having an unfair monopoly with its Cabilly patents, others suggest the ruling is a sign that the patent office is increasingly rejecting patents based on prior art. (Nat. Biotechnol, 25, 699–700, 2007.) For example, Alice Martin, a partner at Barnes & Thornburg in Chicago, says the decision could “backfire on biotech companies.” EW