Center News

Lung cancer's non-smoking gun

Study reveals increased lung-cancer risks among Asian-American women; exposure to unrefined cooking oils may be a factor

Nov. 17, 2005
Meira Epplein and Dr. Stephen Schwartz

Meira Epplein, Cancer Prevention Program, and Dr. Stephen Schwartz, Epidemiology, found a fourfold increase in the incidence of lung cancer for Chinese-American women and a twofold increase for Filipino-American women after factoring in smoking prevalence. Studies have found similar results among Chinese women living in Asia.

Photo by Todd McNaught

Public Health Sciences Division researchers report finding a higher risk of lung cancer among women of Chinese and Filipino descent living on the West Coast of the United States than would be expected based on their tobacco use.

The study, in the November issue of Cancer Causes and Control, examined data from Asian-Americans living in three metropolitan areas: the San Francisco Bay area, Los Angeles County, and the 13 counties of northwest Washington. The researchers found a fourfold increase in the incidence of lung cancer for Chinese-American women and a twofold increase for Filipino-American women versus non-Hispanic whites of the same age living in the same areas, after taking into account smoking prevalence for each population.

The lead author of the study is Meira Epplein of the Cancer Prevention Program in PHS. The findings represent her thesis work for a Master of Science from the Department of Epidemiology at the University of Washington. Dr. John Potter, director of PHS, is a co-author of the study.

Previous studies have found that Chinese women living in Asia have a higher incidence of lung cancer than would be expected from their smoking habits. The recent study by Epplein and colleagues extends this finding to Chinese women living in the United States. The question remains why?

The possible cooking connection

"One possible reason is you might think that Chinese women have a greater susceptibility to the carcinogenic effects of cigarette smoke than Chinese men, but we haven't seen evidence supporting that hypothesis in literature or in the current research," Epplein said. Several environmental factors have been posited to explain the effect in Chinese women living in Asia, including diet, hormones and pollution from coal. But the explanation favored by Epplein is prolonged exposure to unrefined cooking oils. "In these studies in Asia where they've looked at why there's excess lung cancer in Chinese women, the most consistent findings have been an association with cooking using unrefined oils such as rapeseed oil," Epplein said.

The study also found that certain types of lung cancer are more prevalent in Chinese and Filipino women living in the United States. The risk of adenocarcinoma was elevated in both groups, and was responsible for the majority of the excess risk in these populations, similar to the findings of excess lung cancer by histological type in Asia. "Adenocarcinoma is on the rise in the United States, whereas squamous cell carcinoma, previously the predominant type of lung cancer in the United States, has been on the decline over the past few decades."

"This important study shows that not only is there an excess of adenocarcinoma in non-smoking women in China, but there is also an excess of this same kind of lung cancer among Chinese women in the United States," Potter said about the findings. "Given what we know so far, it is likely that there are carcinogens other than cigarette smoke that are responsible for this pattern of disease risk. Further studies are needed to find these causative agents," he said.

The study used retrospective methods to gather data about smoking prevalence (how many people currently smoke), cancer incidence (the number of new cancer cases), and population from three separate sources and then integrated the data to calculate the risk of lung cancer in each of the three Asian-American subpopulations, relative to the risk for non-Hispanic whites. The researchers obtained data on cancer incidence among non-Hispanic whites, Chinese, Japanese and Filipinos living in the three regions of the American west from the National Cancer Institute's Surveillance Epidemiology and End Results (SEER) database. The Cancer Surveillance System (CSS) housed in the PHS Division provides cancer data for SEER from the 13 counties of northwest Washington. Dr. Stephen Schwartz of the Epidemiology Program in PHS and a co-author of the study is co-principal investigator of CSS. The data on smoking prevalence were obtained from a statewide California study performed in the early 1990s and the population data were from the 2000 United States Census.

To obtain relative cancer risks for each Asian subgroup, the researchers took the cancer incidence rates from SEER for non-Hispanic whites (based on age, sex and geographic area) and multiplied each by the ratio obtained by dividing the percentage of smokers in each Asian subgroup by the percentage of whites who smoke. This gives an estimate of the incidence rate of cancer among non-Hispanic whites if they smoked at the Asian levels. Then they multiplied these estimates by the population of each Asian subgroup living in each of the three regions (obtained from the 2000 census) to get the expected number of lung-cancer cases in each group based on smoking habits. This expected number was compared to the observed number of cases by race from the SEER data to obtain the relative risk. Statistical tests then estimated the reliability of the results. The data show that Chinese and Filipino women living on the West Coast have a fourfold and twofold higher risk of lung cancer, respectively than would be expected based on their smoking rates. Neither Japanese-American women, nor any of the three racial subgroups of Asian-American men had elevated lung-cancer rates.

Factoring migration

Epplein cautioned that there are several limitations to the study. "One slightly unusual aspect of the paper is that I had to use data from a number of different sources, and that's obviously not ideal," Epplein said. For example, the data on smoking prevalence came from a study of California residents published in 1992 by the California Department of Health Services. "They had the most detailed data on smoking prevalence for Asian sub-races that I could find, and their methods, using random-digit dialing for adults throughout the state, were good," Epplein said of the California study. But in Epplein's opinion, the biggest limitation to the study is the issue of migration. "By our saying that there is an increased risk in Chinese women in America, we may not be saying anything more than the studies in China, Singapore and Taiwan have said; this increased risk we've found could be among those same women, now living in the United States. If we had data on place of birth and date of migration — if foreign born — we could determine whether this increase in risk applies to American-born Chinese women specifically," Epplein said.

So, should all Chinese and Filipino women living in the United States make an appointment with their doctor for lung-cancer screening? Not yet, Epplein said. "While the relative risk is high, the absolute number of Chinese and Filipino women developing lung cancer is still rather small."

Avenues for exploration

Nonetheless, the study does have some important implications. "Whether their exposure was in China or here, I think it is still an important finding for American public-health policy people to know that there is this increased lung-cancer risk among Chinese women residing in the United States," Epplein said. The study also suggests several possible avenues for further research into lung-cancer epidemiology in migrant populations of Asian Americans. From Epplein's perspective, the possibility of a connection between cooking oils and cancer in this population needs further examination. "In addition to creating the impetus for future cohort and case-control studies in Asian-American women, I think that this study points to the need for more laboratory studies on unrefined cooking oils to see if exposure to this substance is possibly contributing to the increased lung-cancer incidence found here," she said.

Epplein is currently the international research project manager working under Potter to develop a program that will establish a worldwide cohort for examining gene-environment interactions in cancer risk.

In addition to Epplein, Potter and Schwartz, the senior author of the current study was Dr. Noel Weiss of the Epidemiology Program in PHS. Weiss was chair of Epplein's master's thesis committee and is currently chair of Epplein's doctoral dissertation committee in the Department of Epidemiology at the UW.

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.