Landis' Doctor: Test is 'False Positive'

July 29, 2006 — -- Tour de France winner Floyd Landis' personal physician told "Good Morning America Weekend Edition" he believes a test that showed Landis had elevated testosterone during the race was a false positive.

"I've talked to countless doctors in the past couple days [and] everybody has had the same opinion," Dr. Brent Kay said. "There's no reason for a professional cyclist to be taking a body-building, mass-building steroid during the Tour de France. Floyd has had some 20 odd tests this year. They've all been normal."

Landis denied Friday taking illegal drugs during cycling's most celebrated race, claiming his own physiology, not performance-enhancing drugs, led to the indication of unusual levels of testosterone.

"The levels I have had during the Tour and all my career are absolutely natural," he said.

However, some doctors say if that's the case, Landis' testosterone levels should always appear abnormal, which he says has never happened before.

The questionable sample was taken after Landis' remarkable win in the 17th stage of the Tour de France. The night before the test, Landis said he drank whiskey and beer.

"I don't know the night before would produce that result," Kay said. "But there's a long list of potential contributors that would cause a false positive test [or] cause the test to look abnormal when, in fact, it's not abnormal."

On CNN's "Larry King Live," Landis said he requested a backup sample be tested immediately.

Doping scandals are not new to cycling. Seven-time Tour de France winner Lance Armstrong faced numerous doping allegations. In an exclusive interview with ABC News, Armstrong said he knows how his former teammate feels.

"I've lived that life," he said. "I know what it's like to be accused of things. You have to fight back and answer the questions and be very specific -- and be aggressive."

Landis said his team has asked him not to race while the investigation is ongoing. He worries that the cloud of suspicion will follow, no matter the results of the second test.