Published: May 15, 2013

By Michael Smith , North American Correspondent, MedPage Today

Action Points

• For middle-age men, good physical fitness reduces the risk of lung and colorectal cancer.
• Point out that if men who are fit in their 50s do develop those cancers, as well as prostate cancer, the risk of dying appears to be lower.

CHICAGO — For middle-age men, good physical fitness reduces the risk of lung and colorectal cancer, a researcher said.

And if men who are fit in their 50s do develop those cancers, as well as prostate cancer, the risk of dying appears to be lower, according to Susan Lakoski, MD, of the University of Vermont in Burlington, and colleagues.

Fitter men also had — as might be expected — a lower risk of dying from cardiovascular causes, Lakoski told reporters in a telephone press briefing in advance of the 2013 meeting in Chicago of the American Society of Clinical Oncology.

But the effects of fitness on cancer risk and mortality are new and have important clinical implications, Lakoski told MedPage Today during an interview earlier.

When primary care doctors think about preventing cancer and cardiovascular disease, she said, “they need to think about fitness … because we know it is prognostic of major outcomes.”

Lakoski said doctors have to go beyond writing a prescription for exercise to think about actually measuring fitness. “Physical activity and fitness are very different,” she said in the media conference.

The findings are both novel and exciting, commented society president Sandra Swain, MD, of Washington Hospital Center in the District of Columbia.

Importantly, she noted, the researchers found that weight is not the issue. “Even if men are not obese, they still have an increased risk of cancer if they aren’t fit,” she told reporters.

Lakoski and colleagues followed 17,049 men in the Cooper Center Longitudinal Study, a prospective observational cohort of participants undergoing a preventive health at the Cooper Clinic in Dallas.

As part of the exam, the men — with an average age of 50 at the time — took a treadmill test to assess their maximal oxygen uptake (VO2max) during exercise.

After a median follow-up of more than 20 years, the researchers used Medicare claims data to identify participants who had died or developed lung, colorectal, or prostate cancer — the three most common types of cancer among U.S. men.

All told, 2,332 men were diagnosed with prostate cancer, 276 were diagnosed with colorectal cancer, and 277 were diagnosed with lung cancer. There were 769 deaths, 347 of them due to cancer and 151 because of cardiovascular disease.

Compared with the men who were in the lowest quintile of fitness, those in the highest fifth were less likely to get lung or colorectal cancer, although there was no significant difference in prostate cancer risk.

Specifically, compared with the least fit men, the fittest men had a hazard ratio:

• For lung cancer of 0.32 with a 95% confidence interval from 0.20-0.51, which was significant at P<0.001.
• For colorectal cancer of 0.62 with a 95% confidence interval from 0.40-0.97, which was significant at P=0.05.
• The analysis was adjusted to take into account such factors as smoking, body mass index, and age.

If participants developed one of the three cancer types, mortality was significantly lower among the fittest men, with a hazard ratio of 0.36, which was significant at P<0.001.

In addition, Lakoski reported, the risks of all-cause mortality and cardiovascular death were lower for the fittest men, compared with the least fit. The hazard ratios were 0.42 and 0.39, respectively, and both were significant at P<0.001.

Lakoski told MedPage Today that an important aspect of the study is that it used a treadmill test, an objective assessment of fitness, rather than relying on self-reported exercise. But she cautioned that the researchers don’t know what happens when fitness changes over time, since they only measured it once.

She added it’s not possible to establish causality because of the study’s observational design, or to tease out any biological mechanism that might account for the observed associations.