WASHINGTON (Reuters) - Older men should not take testosterone as an anti-aging agent until there has been a large-scale study of the hormone therapy's true risks and benefits, a U.S. advisory committee said on Wednesday.

While there is some evidence that replacement of the male hormone could combat declining strength, failing memory, and decreased sex drive, the panel appointed by the Institute of Medicine said there was not enough data to show it was safe.

"There is still much we don't know about normal levels of testosterone at different ages," said committee chairman Dan Blazer, professor of psychiatry and behavioral sciences at Duke University Medical Center.

It was unclear how decreased testosterone levels affect men's health and whether testosterone therapy might increase the risk of prostate cancer, Dr. Blazer said.

The panel said men should heed the experience of the Women's Health Initiative study. That large trial showed that, contrary to expectations, estrogen replacement offered few, if any benefits, to postmenopausal women, and even increased their risk of breast cancer and lung clots.

"It is precisely the concern and hope that we can avoid a similar situation with respect to testosterone," said Richard Hodes, director of the National Institute on Aging (NIA), which commissioned the Institute of Medicine study.

The NIA plans to fund studies of testosterone replacement as recommended by the Institute of Medicine panel.

Testosterone replacement is approved by the Food and Drug Administration to treat young men with severe testosterone deficiencies. But it is increasingly being used "off-label" by older men who have been diagnosed with low testosterone levels, patients with AIDS-related wasting, and men who use it in the belief that it may keep them young and vital.

A key concern is that testosterone replacement could accelerate benign enlargement of the prostate or, more dangerously, fuel prostate tumor growth.

There have been only 31 trials of testosterone replacement in men over age 65, and most of those were too short or had too few patients to prove efficacy or safety, panel member William Hazzard told Reuters.

Copyright © 2003 Reuters Limited. All rights reserved. Republication or redistribution of Reuters Limited content, including by framing or similar means, is expressly prohibited without the prior written consent of Reuters Limited. Reuters Limited shall not be liable for any errors or delays in the content, or for any actions taken in reliance thereon.