TORONTO — People with colon cancer who smoked and drank alcohol appeared to develop the disease up to a decade earlier in life than those without the habits, a study has shown, suggesting that some people should be screened for the disease much earlier than currently recommended.

Under North American guidelines, people with no family history of colorectal cancer are advised to start testing for pre-cancerous intestinal polyps at age 50. Those with a close relative affected by the disease would begin regular testing earlier.

But a U.S. study of more than 160,000 people with colon cancer suggests that age 50 may be too late for those who smoke tobacco and/or use alcohol to begin screening, which often includes a colonoscopy to view inside the large intestine.

"What we found was not only was the age of onset (of cancer) earlier, but it was actually significantly earlier, in some close to a decade,'' said principal researcher Dr. Hemant Roy, a gastroenterologist at Northwestern University.

Men tended to develop colorectal cancer at an earlier age than women, whether or not they smoked or drank alcohol, Roy said Monday from Evanston, Ill. "But the effect of tobacco and alcohol was equally great in women and men.''

The study, which analyzed data from a registry of colorectal cancer patients between 1993 and 2003, found that alcohol and tobacco users were diagnosed with the disease 7.8 years earlier on average (age 63 in women and 62 in men) than those who never drank or smoked.

Non-smokers who drank and non-drinkers who smoked were on average 5.2 years younger at cancer diagnosis than those who neither smoked nor drank, the study showed.

"When you put those two together (smoking and drinking), you got kind of an added effect,'' said Roy.

Smoking appeared to affect women in particular: women who smoke but never drank developed cancer 6.3 years earlier than those who never drank or smoked, compared with 3.7 years for men.

Roy said the study has limitations because the data did not include how much patients drank or smoked or for how long. Genetic predisposition because of a family member with the disease was also not captured in the data. More detailed research is needed to pin down the risks with greater certainty, he added.

Still, the research suggests screening smokers and drinkers at a younger age could mean detecting pre-cancerous lesions earlier to prevent the disease, the second most deadly cancer for both men and women after lung cancer. In 2005, almost 20,000 Canadians were diagnosed with colorectal cancer and an estimated 8,400 died.

"Everybody needs to be screened for colon cancer, but our data would suggest that looking at environmental factors needs to be factored into the equation as to when to start screening,'' Roy said.

"The idea would be to pick a time not to diagnose colon cancer early, but rather to pick up the polyps that would lead to colon cancer and remove them so you prevent these people from getting colon cancer.''

Dr. Sharlene Gill, a medical oncologist at the B.C. Cancer Agency, called the study interesting "because it goes beyond the way we currently define risk for screening guidelines for colorectal cancer.''

But Gill said the limitations of the research mean its findings are not confirmatory -- and would not affect current screening guidelines in Canada.

"It's interesting and we probably still should be individualizing our recommendations for patients, based on what their situation is,'' she said from Vancouver. "This might add to that, but it's not enough to change population guidelines.''

However, the research does show that lifestyle has an impact on the risk of developing colorectal cancer, Gill said. "The risk is real and ... if we modify our lifestyle, we can modify our risk.''

Roy said that although the evidence from his study is not definitive, it still makes sense for people to maintain a healthy lifestyle.

"Don't smoke. Use alcohol in moderation,'' he said. "Exercise, because we know a sedentary lifestyle predisposes to colorectal cancer.''

The study is published in Tuesday's issue of the Archives of Internal Medicine.