Obesity May Mask Prostate Cancer Risk by Lowering PSA Levels

October 9, 2006

NASHVILLE, Tenn. -- Weight, height, and race independently influence prostate-specific antigen (PSA) levels, and each may distort the risk of prostate cancer, researchers here reported.

NASHVILLE, Tenn., Oct. 9 -- Weight, height, and race independently influence prostate-specific antigen (PSA) levels, and each may distort the risk of prostate cancer, found researchers here.

Regardless of race, obese men had lower levels of PSA than normal-weight men, while taller men had higher levels of percent free PSA (%fPSA), reported Jay Fowke, Ph.D., of Vanderbilt, and colleagues, in Cancer, in a study released in advance of the Nov.15 issue.

As a result, an obese man with a slightly elevated PSA might be at higher risk for prostate cancer than a man with a similar PSA and a normal body mass index (BMI), they suggested.

Prior studies suggested that obese men had lower PSAs than leaner men and that Caucasian men may also have lower PSA levels than African American men, but the relevance of body size to racial disparities in PSA levels has been unclear, Dr. Fowke said.

Epidemiological evidence shows that obese patients and African Americans are diagnosed with more advanced disease. African Americans are less likely to have a PSA test, while PSA levels may lag in obese patients. Body fat, indirectly measured by BMI, the researchers said, decreases the amount of circulating PSA. In addition, tall height has been shown to affect percent free PSA (% fPSA).

To analyze the relationships between BMI, height, and race and the prostate-antigen triad, including PSA, free PSA (fPSA), and %fPSA, the researchers studied 149 Caucasian men and 150 African-American men. The men, ranging in age from 40 to 79, from the Southern Community Cohort Study, completed an extensive in-person interview and donated blood.

PSA levels decreased with increasing BMI (PSA = 0.72, 0.69, 0.67, 0.59 ng/mL for BMIs ranging from 18.5 to < 25, 25 to < 20, 30 to < 25, and ?35, respectively; P trend = 0.18). In sum, for a BMI ?35, the PSA was only 0.59 ng/mL, whereas for a BMI of < 25, the PSA was 0.72ng/mL.

Overall, mean PSA was approximately 22% higher among men with a BMI less that 25 compared with men with a BMI greater than 35, the researchers said.

The inverse relationship was strongest among men younger than age 60. PSA levels for each increasing BMI category decreased to 0.81, 0.76, 0.66, 0.59 ng/mL, respectively; P trend = 0.02). Also, free PSA (fPSA) decreased significantly with BMI among men younger than 60 years (P trend = 0.04).

In contrast, % fPSA was not associated with BMI. However, it increased 27% across height categories (P trend = 0.02), and 28% with the highest height quartile. Mechanisms linking height to %fPSA levels are speculative, but may be related to IGF levels associated with growth hormone, the investigators said.

PSA levels were significantly lower among Caucasian men (African Americans = 0.87 ng/mL; Caucasian men =0.63 ng/mL; P

Also, associations between body size and PSA or %PSA were not significantly affected by race. In addition, adjustment for BMI and height had no effect on racial disparity in PSA levels (African Americans, PSA=0.87 ng/mL, Caucasians, PSA =0.63 ng/mL; P

Furthermore, they said, the stronger association between BMI and PSA among younger men suggests the BMI effect may be obscured by PSA determinants associated with aging. Thus, the PSA-obesity relationship seems unlikely to diminish the high PSA levels typically observed among older patients or those with advanced prostate cancer.

In conclusion, Dr. Fowke wrote that the data suggest that race, BMI, and height are independently associated with PSA and %fPSA levels. As for the clinical relevance of the study, he said, "Whether equal PSA levels in an obese versus a thin man convey the same biologic relevance is unclear, but our findings suggest that clinical suspicion might be heightened with a marginally elevated PSA level in an obese person."