Hormone-blocking drugs can quicken death in some prostate cancer patients, it seems.
Men given the drugs alone were slightly more likely to die of the disease in the next six years than men who'd gotten medical monitoring but no or delayed treatment.
In any case the drugs did not improve survival chances of older men whose disease hadn't spread, US researchers said.
The study involved nearly 20,000 medicare patients in the U.S. with prostate cancer that hadn't spread. A surprising 41 per cent got only drug treatment, in shots or implants, showing that the therapy has become a popular alternative to surgery and radiation, the study authors said.
The drugs are typically given in a doctor's office and can have serious side effects including increasing risks for diabetes, heart diseases, impotence and bone loss.
Patients often believe that any treatment is better than nothing, said lead author Dr. Grace Lu-Yao of the Robert Wood Johnson Medical School in New Jersey.
"What we are saying is doing something may not always be the best choice, because given the overall picture, this doesn't really give you any proven benefit," she said.
She said the researchers hoped the study would prompt doctors to avoid hormone-blocking drugs alone in older men whose disease hasn't spread.
The results appear in Wednesday's Journal of the American Medical Association.
The researchers examined medical records for 19,271 medicare prostate cancer patients over 66 and followed them for an average of about six years. That observational approach is less rigorous than studies in which researchers randomly choose patients for certain treatments and compare results. But Lu-Yao said "it's the best evidence so far."
There were 11,045 deaths from all causes, including 1,560 from prostate cancer. Overall, survival was the same for those treated and those not given drugs; however, almost 20 per cent of drug-treated men died from prostate cancer, versus about 17 per cent in the other group.
Other experts said the study gives doctors important information about how to treat older men with slow-growing disease that hasn't spread beyond the prostate. However, the study didn't look at whether hormone-blocking drugs alone benefit younger men or compare that treatment with radiation or surgery.
Randomized studies have shown that the drugs can benefit men with more aggressive disease when used along with surgery or radiation. But research is sparse on using hormone-blockers alone or in patients with localized cancer, like those in their study, the authors said.
The drugs block production of testosterone, which feeds cancer cells. They are sometimes given in addition to surgery or radiation; using them alone is a less traditional but increasingly used approach, particularly among older men whose prostate cancer hasn't spread, the study authors said.
The study does suggest that men with local but aggressive disease treated with the drugs are less likely to die from prostate cancer than the men not given drugs. It's possible that men with aggressive disease had undetected stray cancer cells beyond the prostate that hormone treatment helped attack, but Lu-Yao said those results were not definitive.
She also noted that the average age of patients studied was 77, so the results couldn't be applied to younger men, who were more likely to be treated with surgery than older men, news agency AP reported..
While other studies have suggested that many prostate cancer patients do fine with no treatment, only "a minority of doctors feel comfortable doing that," said Dr. Otis Brawley, the American Cancer Society's chief medical officer.
"I would hope that a study like this would make us be much more cautious" about giving hormone-blocking treatment, he said.