Treating mice that develop an early stage of prostate cancer with antioxidants increases the proliferation of cells in the pre-cancerous lesions. The findings by Vanderbilt University Medical Center researchers, reported in the journal PLoS ONE, provide insight into the recent controversy regarding antioxidants and prostate cancer prevention.

"The dogma in the field has been that antioxidants are good for you - that they inhibit tumorigenesis - and we saw the opposite," said Sarki Abdulkadir, Ph.D., associate professor of Pathology, Microbiology and Immunology and of Cancer Biology.

The results in mice also mirror recent findings from a large clinical study that tested antioxidants for prostate cancer prevention. SELECT (the Selenium and Vitamin E Cancer Prevention Trial) demonstrated a 17 percent increased risk of prostate cancer in men who took vitamin E supplements compared to placebo.

The SELECT findings surprised the prostate cancer field, but they validated what Abdulkadir and graduate student Erin Martinez were observing at the time in a mouse model of early prostate cancer. The mice are missing a gene (Nkx3.1) that is commonly mutated in human prostate cancer, and they develop pre-cancerous lesions.

Abdulkadir and Martinez had previously demonstrated that the Nkx3.1-negative mice have high levels of reactive oxygen species (oxidative stress) in the prostate. It fit the hypothesis, Martinez said, that this elevated oxidative stress was promoting tumorigenesis, and that blocking it with antioxidants would prevent cancer development.

But when the researchers treated the mice with the antioxidant N-acetylcysteine, they found more cell proliferation in the pre-cancerous lesions. Because the results didn't fit the prevailing dogma, Abdulkadir and Martinez were suspicious of their findings. The human trial data erased their doubts.

"When we saw the SELECT results, we could tell that our study was going to have a real impact on the field," Martinez said.

The Nkx3.1-negative mice will be useful for exploring how antioxidants, including vitamin E, promote prostate cancer development - a particularly important clinical issue given the widespread consumption of nutritional supplements that contain vitamin E.

In the current studies, the researchers demonstrated that antioxidant treatment promotes the expression of a network of pro-proliferative genes. Their results suggest that reactive oxygen species may be inhibiting proliferation in this early-stage model, and that antioxidant treatment removes this "brake" on tumor progression, Martinez said. However, this may not be true in later stages of prostate cancer.

"Reactive oxygen species aren't doing the same thing in every tissue and at every point during tumorigenesis, so antioxidant treatment may have different effects in early versus advanced disease," Martinez said.

The researchers plan to test the effects of antioxidant treatment at different stages of tumorigenesis in the Nkx3.1-negative mouse model. They can delete additional genes to cause the pre-cancerous lesions to progress to advanced prostate cancer.

Abdulkadir and Martinez also are exploring whether their mouse model findings extend to humans. They know from the work of other investigators that men with inherited genetic variations that impair Nkx3.1 are more susceptible to developing prostate cancer. They will now assess, using samples from the SELECT trial, whether antioxidant treatment modifies the risk of prostate cancer in individuals with Nkx3.1 genetic variations.

"We hypothesize that the SELECT trial participants who have the Nkx SNPs (genetic variations) - and have decreased levels of Nkx3.1 like the mice - will have a heightened risk of developing prostate cancer when they're treated with antioxidant, in this case vitamin E," Martinez explained. "They may have been the group driving the observed increase in risk, and it's important to know that antioxidant supplements may be bad for this specific population."

Philip Anderson and Monica Logan also contributed to the PLoS ONE studies. The research was supported by grants from the National Cancer Institute (CA094858, CA068485) of the National Institutes of Health.