A bone-building drug now approved only for cancer patients dramatically reduced hip, vertebra and other fractures in women suffering from osteoporosis - with just a 15-minute intravenous dose once a year, a new study shows.

But the international, three-year study of about 8,000 women also raises safety questions. A rare, serious heartbeat irregularity occurred more often in women on the drug, called zoledronic acid, than those given a placebo.

And the study was too small to rule out an increased risk of jawbone deterioration, a frightening side effect estimated to occur in 3 percent to 10 percent of cancer patients on more frequent doses of the drug.

Novartis Pharmaceuticals Corp. got approval six years ago to market zoledronic acid as Zometa for cancer-related bone weakening. The North Jersey-based company hopes that, with the latest results, it will get the Food and Drug Administration's blessing to begin marketing the drug for osteoporosis later this year under a new name, Reclast.

The drug reduced the risk of hip fracture 41 percent, general fractures 33 percent, and vertebral fractures 77 percent. Cracking and compression of the vertebrae cause the hunched-back appearance that is a classic, painful sign of osteoporosis.

"This data is extremely exciting. It's revolutionary, in my opinion," said Audrey Kriegman, Novartis' senior medical director for Reclast. "Any female with osteoporosis would be eligible."

The study appears in today's New England Journal of Medicine.

Only a dozen years ago, weakened, thinning bones were considered an untreatable part of aging. In 1995, Merck & Co. Inc.'s blockbuster Fosamax ushered in the current era. About nine osteoporosis drugs, worth $6.2 billion last year, were taken by tens of millions of women - and more than a few men.

Zometa and Fosamax are part of the most popular and effective class of such drugs, called bisphosphonates, which work by inhibiting the breakdown of bone. There are other classes of osteoporosis drugs that are not as effective.

But a surprising problem has become apparent as treatment options have multiplied: Most patients quit the drugs after a short period.

Manufacturers have come up with increasingly potent pills and shots so that patients only need a weekly, monthly, or quarterly dose. Still, doctors say, patients don't stick with it.

"Seventy percent of patients quit bisphosphonates within a year," said Dennis Black, a University of California, San Francisco, biostatistician who led the new study.

Novartis, which sponsored the research, believes Reclast's once-a-year infusion could change that.

"It was very easy to do. I was hooked up to an IV, and it took 15 minutes," said Diana Mahon, 69, a Pittsburgh-area osteoporosis patient who participated in the study. "With this, it would be similar to getting a mammogram once a year."

Why don't patients stay on the bone medicines?

Often, the problem is side effects. Mahon, for example, developed a backache on Evista, a hormonal drug. On Fosamax, her esophagus became scarred and narrowed, a complication related to the drug's effects on the gastrointestinal system.

"After a few months, I was having trouble swallowing. I thought I wasn't chewing my food enough. I ended up in the emergency room," recalled Mahon, whose daughter takes Fosamax with no problems.

There are other obstacles.

"With the elderly, they forget to order it," said Cheryl Malloy of the Visiting Nurse Association of Greater Philadelphia. "And cost is a concern."

Indeed, Oretha Riddick, 83, of Mount Airy, used to go off Fosamax whenever she ran out of samples from her doctor - until Malloy got her a prescription paid by Medicare.

Kriegman said she expected Reclast to cost about $1,000, the same as an infusion of Zometa. That price would be competitive with the annual cost of other bisphosphonates, doctors say.

The biggest question - one that cannot be fully answered yet - is the risk-benefit trade-off for Reclast. Drugs that have been pulled off the market, such as the pain-reliever Vioxx, have shown that some side effects are difficult or impossible to detect until the medication is being used by millions of people.

In the osteoporosis study, about 7 percent of patients taking the drug had irregular heartbeats, compared with 5.3 percent on placebo. More worrisome, 50 patients on zoledronic acid had to be hospitalized for a heart arrhythmia called atrial fibrillation, compared with 20 in the placebo group. The women's average age was 73.

A recent review of Merck's 1997 study of Fosamax also found "the possibility" of an increased risk of atrial fibrillation; it occurred in 47 patients on the drug, vs. 31 on a placebo, according to a letter published in today's New England Journal.

Atrial fibrillation is a risk factor for stroke, but stroke was not more common in the patients taking zoledronic acid.

"We're not 100 percent sure it's the effect of the drug," Black, in California, said of the new study. "It's hard to come up with a biological mechanism. It might be something about how the drug affects intracellular calcium."

Another side effect, jaw osteonecrosis - jawbone death - has recently been recognized as a rare problem of certain bisphosphonates. Most of the reported cases have involved cancer patients taking intravenous Zometa or Aredia every few months, but cases have also been linked to oral Fosamax.

Last year, Merck estimated the risk of osteonecrosis to be less than one out of 100,000 patients per year. Felicia Cosman, an osteoporosis expert who was an author of the new study, speculated that Reclast's risk would be just as remote.

John R. Kalmar, an Ohio State University oral pathologist who has written a patient-information sheet for the American Dental Association, challenged that assumption. A new Australian study has found that jawbone damage occurs in one to four out of every 10,000 patients on oral bisphosphonates, Kalmar said.

"These drugs are not metabolized," Kalmar said. "They bind to the bone tightly, so it takes a long time to be filtered out through the blood and urine. If you give your body a year to recover from a dose, you might actually reduce adverse effects.

"The problem is, we really don't know. For them to say, 'We know it's going to have the level of risk of other oral forms' - they don't know. It's pure conjecture."

Contact staff writer Marie McCullough at 215-854-2720 or mmccullough@phillynews.com.