When the U.S. Food and Drug Administration approved implanting microchips in humans, the manufacturer said it would save lives, letting doctors scan the tiny transponders to access patients’ medical records almost instantly. The FDA found "reasonable assurance" the device was safe, and a subagency even called it one of 2005’s top "innovative technologies."

But neither the company nor the regulators publicly mentioned this: A series of veterinary and toxicology studies, dating to the mid-1990s, stated that chip implants had "induced" malignant tumors in some lab mice and rats.

"The transponders were the cause of the tumors," said Keith Johnson, a retired toxicologic pathologist, explaining the findings of a 1996 study he led.

Leading cancer specialists reviewed the research for The Associated Press and, while cautioning that animal test results do not necessarily apply to humans, said the findings troubled them. Some said they would not allow family members to receive implants, and all urged further research before the glass-encased transponders are widely implanted in people.

2,000 implanted in humans

To date, about 2,000 radio frequency identification, or RFID, chips have been implanted in humans worldwide, according to VeriChip Corp. The company, which sees a target market of 45 million Americans for its medical monitoring chips, insists the devices are safe.

"We stand by our implantable products which have been approved by the FDA and/or other U.S. regulatory authorities," said Scott Silverman, chairman and chief executive officer of the Delray Beach, Fla., company.

Management was "not aware of any studies that have resulted in malignant tumors" in lab animals, but he added that millions of pets have been implanted with microchips, without reports of significant problems.

The FDA also stands by its approval of the technology but declined repeated requests to specify what studies it reviewed before approving the implants.

The agency is overseen by the Department of Health and Human Services, which, at the time of VeriChip’s approval, was headed by Tommy Thompson. Two weeks after the device’s approval was formally announced on Jan. 10, 2005, Thompson left his Cabinet post, and by July was a board member of VeriChip Corp. and its parent company, Applied Digital Solutions. He was compensated in cash and stock options.

Also making no mention of the findings on animal tumors was a June report by the ethics committee of the American Medical Association, which touted the benefits of implantable RFID devices.

Had committee members reviewed, or even been aware of, the literature on cancer in chipped animals?

No, said Dr. Steven Stack, an AMA board member.

Published in veterinary and toxicology journals between 1996 and 2006, the studies found that lab mice and rats injected with microchips sometimes developed subcutaneous "sarcomas," or malignant tumors, most of them encasing the implants.

*A 1998 study in Ridgefield, Conn., of 177 mice reported cancer incidence to be slightly higher than 10 percent, a result the researchers described as "surprising."

*A 2006 study in France detected tumors in 4.1 percent of 1,260 microchipped mice. This was one of six studies in which the scientists did not set out to find microchip-induced cancer but noticed the growths incidentally.

*In 1997, a study in Germany found cancers in 1 percent of 4,279 chipped mice. The tumors "are clearly due to the implanted microchips," the authors wrote.

Caveats accompanied the findings. "Blind leaps from the detection of tumors to the prediction of human health risk should be avoided," one study cautioned. Also, none of the studies had a control group that did not get chips, so the normal rate of tumors cannot be determined and compared to the rate with chips implanted.

Specialists see red flags

Still, specialists at some pre-eminent cancer institutions said the findings raised red flags.

Dr. Cheryl London, a veterinarian oncologist at Ohio State University, noted it’s easier to cause cancer in mice than people. "So it may be that what you’re seeing in mice represents an exaggerated phenomenon of what may occur in people."

Tens of thousands of dogs have been chipped, she said, and veterinary pathologists haven’t reported outbreaks of related sarcomas. (Published reports detailing malignant tumors in two chipped dogs turned up in AP’s four-month examination of research on chips and health. In one dog, the researchers said cancer appeared linked to the presence of the embedded chip; in the other, the cancer’s cause was uncertain.)

Nonetheless, London saw a need for a 20-year study of chipped canines. Dr. Chand Khanna, a veterinary oncologist at the National Cancer Institute, also backed such a study, saying current evidence "does suggest some reason to be concerned about tumor formations."

Meanwhile, the animal study findings should be disclosed to anyone considering a chip implant, the cancer specialists agreed