New drug could heal hard-to-mend fractures. (Bone Builder).

A synthetic compound can heal broken bones that are so damaged they don't knit on their own, a study in rats and dogs shows. Encouraged by the findings, scientists are already testing the compound in people. If the experimental drug--so far, called only CP-533,536--passes safety and effectiveness trials, it could become an important treatment for the very worst of fractures.

The compound works by binding to a receptor molecule on the surface of bone-building cells. A natural compound called prostaglandin E2 normally attaches to this receptor, which sets off a flurry of bone-repair signals in the cell. When tested as a drug, however, prostaglandin E2 had dire side effects.

CP-533,536 appears to mimic the prostaglandin's bone-building benefits without its downside, says study coauthor David D. Thompson of Pfizer in Groton, Conn. He and his colleagues report their findings in an upcoming issue of the Proceedings of the National Academy of Sciences.

Prostaglandin E2 actually binds to several molecules on cell surfaces, some of which might account for its side effects, Thompson says. Previous studies suggested that one receptor is particularly instrumental to bone growth, so his team screened hundreds of compounds to find ones that would only latch onto that cell-surface molecule. They then modified one of the selected compounds to improve its binding. They named the result CP-533,536.

Next, the researchers injected CP-533,536 near fractures in some rats and gave other rats inert shots at their injury sites. Only the drug-treated rats showed significant gains in bone density.

The researchers then tested CP-533,536 on dogs, some with severe breaks that would be unlikely to heal on their own and others with more modest fractures. Most of the dogs getting the drug healed fully within 24 weeks of surgery, regardless of the severity of the break. Dogs not getting CP-533,536 failed to heal at the worst breaks and mended the modest breaks more slowly than did treated dogs.

None of the dogs given CP-533,536 exhibited the harsh diarrhea, lethargy, or kidney and heart damage seen in past animal tests of prostaglandin E2.

Because of these side effects, "nobody has ever developed clinical utility for prostaglandin E2 itself," says Lawrence G. Raisz of the University of Connecticut in Farmington. "This compound [CP-533,536] could be a much better way of doing this."

Of the 6.2 million fractures diagnosed in the United States each year, up to 10 percent don't heal properly, says Thomas A. Einhorn of Boston University School of Medicine. Many of these are the result of violent accidents, but some simply occur in parts of the body, such as the shinbone, that have poor circulation. If CP-533,536 tests well in people, it might facilitate healing in such cases and eliminate the need for some bone grafts, he says.

As such, it would join another line of pharmaceuticals approved for fractures in the past few years (SN: 6/3/00, p. 357). Those so-called bone morphogenic proteins also heal severely damaged bones, but they don't work on all patients.

It may also be possible, Einhorn says, to develop an oral treatment using CP-533,536.