Dead ends and discouragement can force a whole field into dormancy for years, until something new restores its promise. That's what happened with the use of antibodies to treat cancer, first seen as a bright magic bullet, then scorned as an over-hyped disappointment. Mark Kaminski's recent success using the method to treat a form of lymphoma is a sort of prince's kiss, reviving the field.

The idea of treating cancer with antibodies dates back to German bacteriologist Paul Ehrlich, whose pioneering studies of the immune system won him a Nobel Prize in 1908. An antibody's normal job is to seek out and disable bacteria, viruses, toxins or other foreign substances that enter the body. By attaching tumor-killing substances to antibodies, it might be possible to make the antibodies home in on tumors, delivering their deadly cargo without harming normal cells, Ehrlich thought. Other scientists tried to perfect the approach but had trouble producing pure lines of antibodies that would target only tumor cells and give consistent results.

A major advance came about 25 years ago from immunologists Georges Kohler and Cesar Milstein, also Nobel laureates. They found a way to make pure lines of antibodies, called monoclonal antibodies, in large amounts. These antibodies were so selective, they might be able to kill tumor cells themselves, without additional tumor-killing substances, researchers thought.

The next challenge was finding a target on tumor cells that the antibodies would recognize and home in on. That's where science hit a snag. No one could find a good target that appeared only on tumor cells, not on normal cells. The next best thing would be a target that always appeared on tumor cells, but was found on normal cells only during certain developmental stages. By aiming antibodies at such a target, relatively few normal cells would be affected. A protein called CD20 was just that sort of target.

At first, however, no one seemed to realize how useful CD20 could be. It was the mid-1980s, and scientists who had been seeking a magic bullet were disappointed and "very, very depressed," says Kaminski, a U-M associate professor of internal medicine.

Denied funding to continue their work on monoclonal antibodies, many scientists drifted into other areas of research. But Kaminski, U-M colleague Richard Wahl and a handful of scientists at other institutions "continued to work stubbornly" on the problem. They began to realize that perhaps Ehrlich had been right all along. Antibodies alone might not do the job, but armed with a powerful tumor killer, such as radiation, they could be effective.

Persistence and scientific savvy paid off. Kaminski's group developed a treatment for B-cell lymphoma, the most common form of non-Hodgkin's lymphoma. The treatment uses anti-CD20 antibodies, tagged with radioactive iodine that can deliver radiation to tumor cells. Because CD20 is on virtually all lymphoma cells, but on only a small population of normal cells, the treatment spares normal tissue. The technique also stimulates the immune system to attack any remaining cancer. In initial trials on 59 patients who hadn't responded to conventional chemotherapy, 20 went into complete remission after treatment, with only minimal side effects. Ten of those 20 continue to be cancer-free anywhere from one year to more than 4.5 years later, with no further treatment.--Nancy Ross-Flanigan