By Jim Shrine

Data from a Phase II trial of Xoma Corp. and Genentech Inc.'s anti-CD11a antibody showed highly significant statistical improvement in patients with moderate to severe psoriasis.

The companies knew the results were good in April but specifics were disclosed for the first time at the Canadian Dermatology Association meeting in Vancouver. The plan now is to move the hu1124 antibody into a Phase III program, likely in the fourth quarter, said Ellen Martin, director of corporate communications at Xoma.

"That's clearly the next step," Martin said. "We're considering additional indications, as well. Psoriasis has become a very interesting model for testing T cell-mediated drugs. I think we have the advantage of a good mechanism and an early start. Of course, Genentech as a partner is hard to beat."

The randomized, double-blind, placebo-controlled Phase II trial in Canada tested the antibody in 145 patients. The primary endpoint was fair or better improvements in a physician's global assessment (PGA) score, with fair meaning 25 percent. The company also measured "good" (50 percent improvement) and "excellent" (75 percent). The score represents improvement of clinical signs and symptoms, as compared to baseline in a six-point scale.

The data showed 48 percent of patients treated with the 0.3 milligrams per kilogram dose had a PGA score at 56 days of at least 50 percent improvement compared to 15 percent of placebo patients (p=0.0002). Twenty-five percent had a PGA score of 75 percent or greater improvement, compared to two percent among placebo patients (p=0.0004).

Patients were given weekly intravenous injections for eight weeks. They were evaluated seven days after the final dose, or 56 days after beginning treatment, and also were evaluated at 91 and 140 days.

Martin said the focus was on scores of "good" or better, an endpoint that is more typical in U.S. trials. The Phase III trial is expected to be designed similarly, she said, but specifics will not be known until later.

"It's not clear at this time whether there will be one or more trials required," Martin said. "We 're also looking at alternative forms of administration."

Genentech, of South San Francisco, and Xoma, of Berkeley, Calif., began collaborating on hu1124 in April 1996. It was Genentech's product, and it enlisted Xoma's help because of the latter company's ability to help accelerate development through its clinical and manufacturing capabilities, Martin said.

Genentech has paid Xoma about $34.5 million, with $23.5 million of it coming in the form of long-term convertible loans that help fund Xoma's development costs. They can be paid back in cash, stock or through royalties.

In April, the companies agreed to continue their collaboration in a deal that assigned Genentech 75 percent of development costs and resulting profits in the U.S., and Xoma 25 percent of each. Outside the U.S., Xoma would receive royalties on sales.

That collaborative structure is similar for any other indications they decide to pursue. Among the T cell-mediated chronic inflammatory disorders expected to be considered for future testing are rheumatoid arthritis, multiple sclerosis and inflammatory bowel disease. The antibody has been tested only in psoriasis so far.

Xoma's stock (NASDAQ:XOMA) gained 34.38 cents Wednesday, to close at $6.468. Genentech's stock stopped trading following the decision last month by Basel, Switzerland-based F. Hoffmann-La Roche Ltd. to exercise its option to purchase all of Genentech.