Cancer rates can differ even within nations. In the U.S., lung cancer mortality rates are significantly higher in the southeast, where tobacco is cultivated and cigarette smoking rates are high ⁴. In China, lung cancer mortality rates are high in the north, in areas with widespread industrial and household pollution from fossil fuels ⁴. There are also large regional differences in nasopharyngeal cancer mortality rates in China, with higher rates in the southeast among the Cantonese; explanations may be genetic and dietary factors ⁵. Higher nasopharyngeal cancer mortality rates in California are also attributable in part to its large Cantonese immigrant population.

Data Availability and Quality. Data on cancer incidence and mortality among Asians and Asians-Americans and Pacific Islanders (AAPIs) are available from cancer registries in the U.S. and Asian nations. The Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute is a population-based cancer registry in five states and six metropolitan areas, covering approximately 14% of the U.S. population. SEER collects data on cancer incidence and mortality, as well as demographic characteristics of the cases ⁶. Among AAPIs, recent SEER cancer incidence data (1988-1992) are available for Chinese-, Filipino-, Hawaiian-, Japanese-, Korean-, and Vietnarnese-Americans. Corresponding mortality data are available for Americans of Chinese, Filipino, Hawaiian, and Japanese ancestry ⁷. The numbers of cancers among individual groups of AAPIs are very small, and rates are unstable. Corresponding sources of comparative cancer incidence and mortality data in Asia include: the Osaka registry for Japanese populations; the Shanghai, Hong Kong, and Singapore registries for Chinese populations; the Manila registry for Filipinos; and Singapore and Bombay registries for Indian populations. For Pacific Islanders, registries include the Maori in New Zealand and Native Hawaiians. An ethnic identifier for American~Samoans was not recorded in the U.S. census until 1980 ⁸.

Limitations of the data include differences in accuracy of census figures, cancer ascertainment, diagnostic accuracy, and coding practices among nations. When migration from a country is continuous and includes illegal immigrants, the number of immigrants may be under-estimated, thereby inflating reported cancer rates ⁹. In addition, information on race/ethnicity in census, death certificate, or cancer registry data may be inaccurate ². It is difficult to identify persons of second-, third- and subsequent generations, particularly when inter-marriages have occurred.

Migrants may not be representative of the population of their country of origin. Migrants may be healthier than the general population of their country of origin ("healthy migrant effect"), and differ with respect to religion, ethnic group, or socioeconomic status ³. For example, Asian women who migrated to the U.S. from rural areas in Asia reportedly have lower breast cancer rates than migrants from urban areas ¹⁰ .

Aims and Methods of Migrant Studies of AAPIs. Migrant studies can reveal clues to cancer etiology. Changes in cancer rates in migrants can identify carcinogenic effects of environment and lifestyle factors, which can be altered to reduce risk. Comparisons of cancer rates among migrants and their descendents might also reveal changes in cultural practices that affect cancer risk. Acculturation, the extent of lifestyle changes among migrants, depends on many factors. Within AAPIs, acculturation tends to increase with number of generations and years since migration, improved socioeconomic status, and residence in communities with few immigrants. Because acculturation is difficult to measure, the number of generations since immigration to the U.S. is often used as a surrogate. When comparing different ethnic sub-groups of AAPIs, the proportion of first-generation immigrants merits consideration. For example, most Vietnamese-Americans were born outside of the U.S., while most Japanese-Americans were born in the U.S.⁷. Exposures in the country of origin may be an important factor in cancer among first-generation immigrants, who tend to be less acculturated.

Changes in cancer rates among the first-generation immigrants suggest the role of etiologic factors with short latency to disease occurrence. Changes in later generations suggest the role of exposures during early life, carcinogens with latent periods of decades, or the effects of acculturation in the subsequent generations² For example, the incidence rates of both colon cancer and female breast cancer are low in most Asian countries and high in the U.S. Migrant studies show that breast cancer rates tend to remain low in first-generation Asian-American women, but increase in their descendants to approach those of the general U.S. population ² In contrast, colon cancer incidence rates often start to rise in first-generation Asian immigrants to the U.S. II, suggesting that latency periods are shorter for colon cancer than for breast cancer. One hypothesis for the increased risk for colon cancer within the first generation of AAPIs is due to rapid increase in red meat consumption.

Several types of research designs have been employed in migrant studies ² Conceptually, the simplest are studies of incidence and mortality rates among migrants, which are compared with corresponding rates in the country of adoption and the country of origin. When data are available, migrants can be grouped by generation and by time since migration for subset analyses. In addition, data on cancer rates can be used in correlational studies to generate etiologic hypotheses and identify cancer risk factors.

Migrant studies help identify special health needs, which has implications for public health programs. For example, the age-adjusted cervical cancer incidence rate for Vietnamese-American women is more than twice that for other American women ^{7, 12, 13} Low cervical screening rates have been reported among Vietnamese-American and other AAPI women ^{12, 14} Cervical cancer is detectable by screening, and programs are needed to overcome cultural, language, or economic barriers and to increase Pap smear screening among these high-risk women.

Patterns of Selected Cancers among Asian-Americans and Pacific Islanders. Cancer incidence rates vary among sub-groups of AAPIs. By aggregating AAPIs, important differences among ethnic groups can be overlooked. For example, Japanese-, Chinese-, Korean-, and Filipino~Amefl'can males have lower rates of lung cancer than U.S. white males, but Native Hawaiian males have higher rates ⁷.

AAPJs and the Lung Cancer Pandemic. Tobacco is the major cause of lung cancer, and international variations in rates of lung cancer incidence are mainly due to patterns of cigarette smoking. Tobacco use has decreased in some developed countries, but has increased in developing countries, especially in Asia ¹⁵. Since China lifted import licensing requirements for tobacco products in 1994, the international cigarette companies have aggressively advertised in China ^{I 6}. The estimated per capita consumption of smoking among Chinese increased by 260% in the last two decades ¹⁵. The increase is due to rising disposable income with economic improvements, the relatively low cost of cigarettes in Asia, and lack of knowledge about smoking-related diseases ¹⁶. In the U.S., cigarette smoking rates tend to be lower among AAPIs in aggregate (15%) than among whites (26%). The exceptions are men from Southeast Asia and Native Hawaiians, who have a higher prevalence of smoking than whites (approximately 34-43% among Southeast Asian men and 34% among Native Hawaiians) ^{17, 18}. Consequently, lung cancer incidence rates have historically been higher in the U.S. than in China and Japan, but the difference is narrowing due to increased tobacco use in Asia ¹⁹. Recent data show that lung cancer incidence rates among men in Shanghai and Manila are similar to U.S. whites, but higher than rates in Korea and Vietnam ¹. Asian women tend to be non-smokers 20 and have low rates of lung cancer ¹. First-generation Japanese and Chinese immigrants have intermediate lung cancer rates when compared to the high U.S. rates and low rates in Asia ²

Tobacco is a major cause of illness and death and a global public health problem. In addition to lung cancer, tobacco causes cancers of the oral cavity, larynx, esophagus, and bladder ²¹, as well as heart disease, stroke, and chronic obstructive pulmonary disease. Smoking is also dangerous for pregnant women and their children²² . The World Health Organization (WHO) estimates that tobacco consumption kills 3.5 million persons per year worldwide (one third of whom are in developing countries). The number will likely increase to ten million per year (approximately 70% from developing countries) in 40 years, if current trends continue ²² . U.S. tobacco companies advertise and use aggressive marketing strategies overseas, especially in developing nations, and exported approximately 11 billion packs of cigarettes in 1997 alone ²³ Health education, tobacco cessation programs, advertising restrictions, tobacco control legislation, and taxing policies are urgently needed to prevent young people worldwide from starting to smoke and to aid smokers in quitting..

Rising Colon, Prostate, and Breast Cancer Rates in AAPIs. Colon, prostate, and breast cancers occur more frequently in the U.S. and other western countries than in Asia ^{1, 2} . Rates of the three cancers tend to rise in Asian migrants to the U.S., due perhaps to changes in diet and other lifestyle factors ^{1, 2, 24, 25}. Compared with Japanese men in Hiroshima, Japanese men in Hawaii consume more animal protein, cholesterol, and total fat, and consequently tended to be taller and more obese ²⁶ . In Asia, the westernization of lifestyle in Japan has also been associated with increased colon cancer rates ¹.Age-adjusted colon cancer incidence rates among Japanese-Americans now approximate those for U.S. whites, whereas Filipino-, Hawaiian-, Korean-, and Vietnamese-Americans still have lower rates ^{7, 27}.

Incidence rates of prostate and breast cancers among AAPIs are also intermediate between rates in Asia and U.S whites ^{1, 24, 25}. There is diversity in the age-adjusted prostate cancer incidence rates among AAPI men, with higher rates among Japanese-Americans than Korean-, Chinese-, Filipino-Americans, and Native Hawaiians. However, prostate cancer rates among all AAPI sub-groups are lower than rates for African-American and U.S. white men ⁷. Breast cancer incidence rates for Chinese, Japanese, Korean, Filipino, and Vietnamese women are also higher in the U.S. than in Asia (Figure l)^{l, 2}. Among AAPIs, the many lifestyle changes brought about by migration to the U.S. are postulated to influence reproductive patterns and endocrine function, and affect breast cancer risk. In 1988-1992, the age-adjusted incidence rates for invasive breast cancer were highest in white American women, followed by Hawaiian, African-American, and Japanese-American women. Low incidence rates were reported in Korean-, Vietnamese-, and Native-American women ⁷.

A western pattern of diet, with higher consumption of red meat and animal fat, is associated with increased risk for colon and prostate cancer ^{28, 29}. Studies investigating diet and breast cancer are not as conclusive, but evidence suggests that diet during childhood affects breast cancer risk ³⁰. Japanese-American women who migrated to the U.S. at early ages are at greater risk for breast cancer than those who migrated at later ages ²⁴. Reproductive factors associated with a western lifestyle, such as a later age at first pregnancy and having fewer children, are also associated with an increased risk for breast cancer in women ³⁰.

Decline in Stomach and Liver Cancers among AAP!s. In contrast to colon, prostate, and breast cancers, some cancers occur more frequently in Asia than in the U.S., including stomach and liver cancers. Rates of both cancers decline among Asian migrants to the U.S. (Figure 2) ^{1, 2, 25, 31}, indicating the importance of environmental factors. The rates of stomach cancer in migrants decline rapidly with succeeding generations, with some residual excess in second-generation Japanese- and Chinese-Americans ², ¹¹. Smoked, salted, and pickled foods, low consumption of fresh fruits and vegetables, and infection with H pylon bacteria (which is reported to be higher in most Asian countries) are postulated risk factors for stomach cancer ³² . Among AAPIs, stomach cancer incidence rates are higher among Japanese-, Korean-, Vietnamese-Americans, and Native Hawaiians, when compared with Filipinos and non-Hispanic whites in the U.S^{. 7}.

Likewise, liver cancer incidence and mortality rates are much lower in the U.S. than in Asian nations, with intermediate rates among Asian migrants to the U.S^{. l, 2, 25, 31}. Within the U.S., liver cancer incidence rates are higher among Vietnamese-, Korean-, Chinese-, Filipino-, Japanese-, and African-Americans, with the excess primarily among men ⁷. Liver cancer incidence rates are lower among later generations of Asian-Americans, compared to first generation Asian-Americans, but still exceed rates among U.S. whites ³¹. Hepatitis B virus infections and aflatoxins are etiologic agents ³³, and the excess of liver cancer among AAPIs emphasizes the importance of Hepatitis B vaccination in this population.

Conclusion: Migrant studies of AAPIs provide exceptional opportunities to gain knowledge about cancer etiology. The data are useful in distinguishing between effects of genetic and environmental factors, and in identifying modifiable environmental and lifestyle factors that elevate cancer risk.

FIGURE LEGENDS

Figure 1. Higher age-adjusted female breast cancer incidence rates, 1988-1992, among Asian-Americans and Pacific Islanders, compared with corresponding populations in Asia. All rates are age-adjusted to the standard world population, except for Vietnamese-Americans. Incidence rates in Asia are for Shanghai (Chinese), Miyagi (Japanese), Kangwha (Koreans), and Hanoi (Vietnamese). ^{1, 7}.

Figure 2. Lower age-adjusted stomach cancer incidence rates by ethnic group, both sexes, 1988-1992, for Asian Americans and Pacific Islanders, compared with corresponding populations in Asia. All rates are age-adjusted to the standard world population, except for VietnameseAmericans. Incidence rates in Asia are for Shanghai (Chinese), Miyagi (Japanese), Kangwha (Koreans), and Hanoi (Vietnamese). ^{1, 7}

REFERENCES

1 Parkin D, Whelan S, Ferlay J, Raymond L, Young J. Cancer Incidence in Five

Continents. IACR Publication No.143. Lyon: International Agnecy for Research on Cancer; 1997.

20. World Health Organization. Tobacco epidemic in the Western Pacific. WHO web page.1997; http ://www.who.int/inf-fs/en/fact 175 .html:Fact Sheet No.175.

23. Exporting tobacco addiction from the USA. Lancet. I 998;35 1:1597.