Something Fishy About Tuna and Melanoma Risk

Heavier consumption of fish, including tuna and other non-fried fish, was associated with an increased risk of melanoma, findings from a large prospective cohort study indicated.

Examining data on nearly half a million participants from the NIH-AARP Diet and Health Study revealed that individuals in the highest quintile of total fish intake had a 22% higher risk for malignant melanoma compared to those in the bottom quintile, after multivariable adjustment that included melanoma-specific risk factors (HR 1.22, 95% CI 1.11-1.34, P=0.001 for trend).

A similarly higher risk was also seen for melanoma in situ (HR 1.28, 95% CI 1.13-1.44, P=0.002 for trend), according to the study from Eunyoung Cho, ScD, of Brown University School of Public Health in Providence, Rhode Island, and colleagues.

"Our findings may be explained by contaminants in fish, such as polychlorinated biphenyls, dioxins, arsenic and mercury," the group wrote in Cancer Causes & Control. "Higher fish intake is associated with higher level of body burden of each of these contaminants, which are associated with higher risk of skin cancer."

The associations were consistent across several demographic and lifestyle factors and for both tuna and non-fried fish:

• Tuna, malignant melanoma: HR 1.20 (95% CI 1.09-1.31)
• Tuna, melanoma in situ: HR 1.17 (95 CI 1.05-1.31)
• Non-fried fish, malignant melanoma: HR 1.18 (95% CI 1.07-1.30)
• Non-fried fish, melanoma in situ: HR 1.25 (95% CI 1.11-1.42)

Fried fish, meanwhile -- which has never been accused of being healthy -- was associated with a small though non-significant lower risk for malignant melanoma (HR 0.90, P=0.06 for trend).

"The authors did their best to control for confounding factors, but technically, it is just rather difficult to adequately control for the most important one -- lifetime sun exposure," Teresa Fung, ScD, RD, of Harvard T. H. Chan School of Public Health in Boston, told MedPage Today.

Fung noted that the researchers used geographical ultraviolet radiation (UVR) quantity to approximate sun exposure, and also controlled for physical activity, as people tend to exercise outdoors. "Those were helpful and important, but sun exposure remained inadequately controlled for," she said.

"I am not alarmed by the results of this study, and there is no need to reduce fish intake," said Fung. "However, for general good nutrition practice, fried fish in the form of fish sandwich (white bread) is generally not a healthy food."

For their study, Cho's group examined data from 491,367 participants (ages 50-71) from the NIH-AARP Diet and Health Study, which in 1995-1996 mailed questionnaires to AARP members in six states (California, Florida, Louisiana, New Jersey, North Carolina, Pennsylvania) and two metropolitan areas (Detroit, Atlanta).

People with a history of cancer, who had died, moved out of the study area, or who had an extreme caloric intake (less than 426 kcal or above 6,760 kcal) were excluded from the study. About 90% of the study population were non-Hispanic white.

At baseline, people who consumed higher quantities of fish were more likely to be men, be younger in age, as well as higher education levels, physical activity levels, body mass index, daily caloric intake, and alcohol intake.

Individuals were followed for a median 15.5 years, for a total of more than 6.6 million person-years of follow-up. By the end of the study period, there were 5,034 total cases of malignant melanoma and 3,234 cases of melanoma in situ. Median age at diagnosis was 70.8 years.

Other covariates for the multivariable analysis included age, sex, education, family history of cancer, alcohol intake, smoking history, and daily caloric intake. Fish types included canned tuna, fried fish or fish sticks, or non-fried fish (cod, clams, shrimp, flounder, crabs, etc.).

A limitation of the study included the assumption that dietary and lifestyle habits did not change from participants' baseline questionnaire, the authors noted. Furthermore, UVR exposure was not estimated based on individuals' behavior. Researchers also lacked data on melanoma risk factors such as mole count, hair color, and history of sunburn.

Comment