Revised immunity; drug slows diabetes in young patients. (This Week).

A drug fashioned from a mouse antibody has halted the progression of diabetes in children and young adults who are newly diagnosed with the disease. By blunting the immune system's attack on insulin-making cells in these patients, the treatment may offer a way to forestall the disease.

The work represents the second time in the past 6 months that scientists have reported success in thwarting type I, or juvenile-onset, diabetes in people. In the earlier experiment, patients received a fragment of a protein that protects cells against extreme heat or other stress (SN: 12/1/01, p. 341). The newer study tested an antibody drug called hOKT3[gamma]l (Ala-Ala).

In both studies, scientists prevented immune cells from unleashing an attack on insulin-producing islet cells in the pancreas. Insulin is a hormone essential for sugar metabolism. If the islet cells are destroyed, a person needs regular injections of insulin to process sugars and starches.

The researchers identified two dozen people, ages 7 to 30, who seemed destined for just such a life-long routine. All had started to show diabetes symptoms within the previous 6 weeks, and all had tested positive for at least one of three antibodies commonly detected in people with type I diabetes. These antibodies target one's own tissues and so are called autoantibodies.

The scientists randomly assigned 12 participants to receive infusions of hOKT3[gamma]l (Ala-Ala) for 14 days. The other 12 patients didn't receive the antibody. All the patients got insulin injections as needed.

Blood tests done a year later showed that 9 of the 12 antibody-treated patients could make insulin after a meal as well as or better than they did at the start of the study. In contrast, only 2 of 12 patients not treated with the antibody maintained their earlier insulin-making capabilities.

By another measure, 7 of the 12 antibody-treated patients showed undiminished or improved production of insulin after a year. Only 1 of the 12 without the drug kept up their insulin production, the scientists report in the May 30 New England Journal of Medicine.

The drug is a monoclonal antibody that's a hybrid of mouse and human antibodies. The human portion of hOKT3[gamma]l (Ala-Ala) keeps the body from recognizing it as a mouse protein and launching an inflammatory response.

The hybrid antibody binds to a molecule called CD3 that's found on immune cells called T cells. "This is the business end of a T cell," says study coauthor Kevan C. Herold, an immunologist at the Naomi Berrie Diabetes Center and Columbia University in New York. Past research had revealed that T cells trigger the production of autoantibodies and orchestrate the release of inflammatory agents and immune cells that attack islet cells.

By binding to CD3, the drug appears to prevent T cells from instigating the onslaught. "How that inactivation occurs ... we're not sure," Herold says.

Despite that uncertainty and the small number of people in the current study, the findings suggest that exploiting the immune system will eventually work against type I diabetes, says Luc Teyton, an immunologist at the Scripps Research Institute in La Jolla, Calif.

Many people test positive for autoantibodies against islet cells but never develop type I diabetes, indicating that the immune system can sometimes right itself, says immunologist and pediatrician Michael Dosch of the University of Toronto and the Hospital for Sick Children, also in Toronto. This study suggests that in newly diagnosed diabetes patients, immune intervention using hOKT3[gamma]l (Ala-Ala) might "tip the balance" away from islet-cell destruction because the treatment lasted only 2 weeks yet had effects that persisted for a year in many patients, he says.

Herold says that he and his colleagues are currently planning studies that will enroll more patients, including some who have had diabetes symptoms for up to 6 months.