Kari Hemminki; Asta FOrsti; Jan Sundquist; Seyed Mohsen Mousavi
K. Hemminki (&); A. Forsti; S. M. Mousavi Division of Molecular Genetic Epidemiology, German Cancer Research Centre (DKFZ), 69120 Heidelberg, Germany
K. Hemminki; A. Forsti; J. Sundquist Center for Primary Care Research, Lund University, Malmo, Sweden.
K. Hemminki; Center for Family and Community Medicine, Karolinska Institute, 14183 Huddinge, Sweden.
J. Sundquist; Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, CA, USA.
S. M. Mousavi; Cancer Research Center of Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran.
E-mail: [email protected].
Abstract
Breast cancer incidence has markedly increased in Western countries for reasons that are not entirely understood. We characterized periodic and age-specific incidence trends of breast cancer in immigrants who migrated from low incidence areas to Sweden. The incidence in immigrants was compared to that in native Swedes and standardized incidence ratios (SIRs) were calculated, based on the Swedish Family-Cancer Database. Age-specific incidence data for low and high incidence populations were obtained from Cancer Incidence in Five Continents IX and NORDCAN. For immigrants from the seven lowest countries/regions 535 breast cancers were identified; the SIRs ranging from 0.45 for Turkish to 0.70 for Greek women. The SIR increased somewhat with the length of stay in Sweden, from 0.55 for stay between 0 and 10 years to 0.59 for a stay of 20+ years. The age-specific incidence curves for these immigrants weresuperimposable upon the earliest Swedish (year 1960) or Danish (1943) rates. These rates differed from the current Swedish rates by a much lower postmenopausal compo-nent. Large incidence differences were also observed between white Californians and immigrants from China and Korea. Our results show that the main difference between high and low incidence areas is in postmenopausal cancer which has increased preferentially during the past century. Immigrants from low risk areas to Sweden show age-specific incidence patterns of Swedes half a century ago. These differences offer opportunities for the identifi-cation of factors underlying breast cancer etiology and tools for prevention.
Keywords: Ethnic differences; Age-incidence; Risk factors; Environmental effect; Incidence change.
Introduction
The incidence of female breast cancer is increasing in developing Asian and African countries but the rates are no more than 25% of those in North America and Western Europe. There is also a major difference in age-specific incidence rates between low risk and high risk regions, which can be observed in ad hoc studies and in Cancer Incidence in Five Continents data [1-5]. In developing countries maximal incidence rates occur in the age band 45 to 54 years, while in developed countries maximal inci-dence rates are reached at older ages [1, 5-7]. Data from Cancer Incidence in Five Continents IX are shown in Fig.1 for China, India, Algeria, Sweden and USA, all countries covered by well-functioning regional or national cancer registries (note the logarithmic y-axis). The incidence differences are small until age 30 years, when the low incidence rates diverge and reach a maximum at age 50 years. The postmenopausal rates are over 350/100,000 for USA and Sweden while those for developing countries remain within the interval 50-100/100,000. Mammography has shifted the age maxima towards younger ages even in high incidence countries, including Sweden, with a nation-wide mammographic screening program, which was implemented between years 1986 and 1997 [8-10].
Figure 1. Age-specific incidence rate for female breast cancer in selected countries or regions based on Cancer Incidence in Five Continents IX in 1998 to 2002 (logarithmic scale) [1]
A careful analysis of the time-dependent and ethnic variation in age-specific incidence patterns and, particu-larly, after immigration to a new environment, may advance understanding on breast cancer etiology and pre-vention. We analyzed age-specific incidence relationships for breast cancer in female immigrants to Sweden and compared the results to the historic data from Sweden and Denmark. The study is based on the nation-wide Family- Cancer Database with 11.8 million individuals, among whom some 15% are foreign born [11].
Subjects and methods
The Swedish Family-Cancer Database was first assembled from the national databases in 1996 and since then it has been periodically updated [12]. The Database contains those born in Sweden since 1932 with their biological parents and additionally data on immigrants are included. This Database is the largest in the world on familial cancer and its updated version (2008, VIII) which has been sup-plied with longitudinal demographic and socio-economic data from each national census of 1960, 1970, 1980 and 1990 has been used for this study [11]. The parental information was classified according to the country of birth. The incidence in immigrants was compared to that in native Swedes and standardized incidence ratios (SIRs) were calculated, as previously [13]. The expected numbers were calculated for 5-year age groups, sex, time periods (10-year bands from 1958 to 2006) age at first child birth (<20, 20-24, 25-29, 30+) and parity (0, 1, 2?) in the native Swedish reference population. Based on SIRs, immigrants with seven lowest significant risks were considered for age-specific analyses (Turkey SIR 0.45, Chile 0.54, Southeast Asia 0.57, Indian subcontinent 0.61, Africa, excluding North Africa, 0.64, East Asia 0.71 and Greece 0.70). Age-specific incidence data were calculated for cancer diagnosed from 1990 through 2006. The Cancer Incidence in Five Continents IX data, covering year 1998-2002, were available at the IARC web site CancerMondial [1]. The NORDCAN data for Sweden and Denmark were also accessed through the IARC web site.
Results
The age-specific incidence data among immigrants from seven countries/regions with the lowest breast cancer risk are shown in Fig. 2, divided in three groups based on the length of stay in Sweden (0-10, 11-19 and 20+). The curves were reasonably similar up to age 50-59 years, but at a higher age, the incidence increased steeply for women who had stayed 20+ years in Sweden. The SIR was 0.58 (N = 535, 95% CI 0.53-0.63) for all selected immigrants in years 1990-2006 compared to native Swedes (SIR = 1.00). By the length of stay, the SIRs were 0.55 for 0-10 years, 0.58 for 11-19 years and 0.59 for 20+ years.
Figure 2. Age-specific incidence rates of breast cancer among Swedes and immigrants with the lowest rates according to the length of stay in Sweden from 1990 to 2006. Case numbers are shown for each age group. The lowest breast cancer incidence countries/regions: Turkey, Greece, Southeastern Asia, East Asia, Chile, India, North Africa and other African
Cancer registration started relatively early in the Nordic countries and we accessed the data for the earliest age- specific data for breast cancer through NORDCAN. The data for Denmark from year 1943 and Sweden from year 1960 are shown together with the data from the seven Swedish immigrant groups with the lowest breast cancer incidence (Fig. 3). The curves are practically superimposable.
Figure 3. Age-specific incidence rates of breast cancer among Swedes in year 1960, Danes in year 1943 and immigrants with the lowest rates from 1990 to 2006. The lowest breast cancer incidence countries/regions: Turkey, Greece, Southeastern Asia, East Asia, Chile, India, North Africa and other African
The main Asian immigration to USA started after 1965 when immigration laws were changed. Over 2.4 million Chinese and 1 million Koreans have immigrated mainly to California [14]. The incidence rates for breast cancer are lowest for these ethnic groups among the main ethnic groups in California [15]. The Los Angeles and San Francisco cancer registries report data by ethnicity as shown in Fig. 4 for years 1998-2002, based on Cancer Incidence in Five Continents [1]. The age-specific incidence curves for non-Hispanic white populations are superimposable (and only Los Angeles curve is shown), leveling off at a rate of 500/100,000, compared to a rate at 200/100,000 or less for the Asian immigrants (logarithmic scale). The age-standardized incidence for Los Angeles whites was 110.0/100,000 for 0-85+ years, 34.4/100,000 for 0-49 years and 412.4/100,000 for 50-85+ years [1]; incidence difference between these age groups was thus 12.0-fold, typical of the high incidence countries (Sweden 10.9-fold). For Los Angeles Chinese, the incidence rates were 52.1, 26.6 and 153.9/100,000, respectively, giving a 5.8-fold difference between age groups below 50 years and older. Among Los Angeles Koreans, the difference was 6.2-fold. These are typical of the incidence differences for women in developing countries (Shanghai 6.2-fold, Chen-nai 6.7-fold and Algeria 4.9-fold) [1].
Figure 4. Age-specific incidence rates of breast cancer among white Californians and Asian immigrants in 1998-2002 based on Cancer Incidence in Five Continents IX (logarithmic scale) [1]
Discussion
Many authors have previously presented evidence that premenopausal and postmenopausal breast cancers are different diseases [16,17]. The novelty of this work is thatwe can show, based on the current and historic Nordic data, that the main increase in breast cancer burden accumulated over the past half century has been in postmenopausal women. As a consequence, the incidence difference between postmenopausal (age 50+ years) and premenopausal breast cancer increased from about 6 in developing countries to 10-12 in high incidence countries. Remarkably, the age-specific incidence curves of the Swedish immigrants from low risk areas resemble the Swedish and Danish rates half a century ago. Such differences can also be observed for Asian immigrants to California compared to the white population. In Swedish immigrants, the rates for postmenopausal breast cancer increased somewhat by the length of stay in Sweden (SIR 0.55 0-10 years and SIR 0.59 20+ years). Larger changes have been observed for Chinese immigrants in Californian whose rates more than doubled after 22 years of stay compared to the newly arrived immigrants and then reached 2/3 of the rates for those born in California [18]. That study also showed that Chinese who immigrated to California at age over 40 years had essentially the risk of the newly arrived, implying that immigration does not change breast cancer risk after age 40 years. We have reviewed age-specific incidence data on all other populations for which data are available by ethnicity in Cancer Incidence in Five Continents IX, including Hawaii, Singapore and Israel. For Hawaii, Chinese and Filipino the rates leveled off at age 50 yeas similar to Chinese in California. The same patterns were observed for Non-Jews in Israel. The ethnic groups in Singapore had more complex patterns.
The causes of the vast increase in breast cancer inci-dence in the developed counties are not known precisely but decreased physical inactivity, weight gain, delayed childbirth and earlier menarche have been implicated [19]. The Swedish immigrant data were adjusted for parity and age at first childbirth but factors such as lactation and attendance for mammographic screening could not be controlled for. Limited historical data from developing countries suggest that lactation has often been continued for 2 years [20]. Long lactation times have also been recorded among immigrants to Sweden: 6-month breast-feeding rates were not different from Swedes but 12-month rates were lower for Swedes (18%) than for e.g. Africans (44%) and Middle Easterners (36%) [21]. The participation of non-Nordic immigrants to the national mammographic screening program has been somewhat lower (about 80%) compared to Swedes (90%) [22]. However, the difference is not large and, at least transiently, mammography increases cancer incidence [23-25].
The causes of the vast increase in breast cancer inci-dence in the developed counties are not known precisely but decreased physical inactivity, weight gain, delayed childbirth and earlier menarche have been implicated [19]. The Swedish immigrant data were adjusted for parity and age at first childbirth but factors such as lactation and attendance for mammographic screening could not be controlled for. Limited historical data from developing countries suggest that lactation has often been continued for 2 years [20]. Long lactation times have also been recorded among immigrants to Sweden: 6-month breast-feeding rates were not different from Swedes but 12-month rates were lower for Swedes (18%) than for e.g. Africans (44%) and Middle Easterners (36%) [21]. The participation of non-Nordic immigrants to the national mammographic screening program has been somewhat lower (about 80%) compared to Swedes (90%) [22]. However, the difference is not large and, at least transiently, mammography increases cancer incidence [23-25].
Acknowledgements
This work was supported by the Swedish Council for Working Life and Social Research, the Swedish Cancer Society and Deutsche Krebshilfe. The used database was created by linking registers maintained at Statistics Sweden and the Swedish Cancer Registry.
References
Curado MP, Edwards B, Shin HR, Storm H, Ferlay J, Heanue M, Boyle P (eds) (2007) Cancer Incidence in Five Continents, vol IX. IARC Scientific Publications No. 160. IARC, Lyon Kwong A, Cheung P, Chan S, Lau S (2008) Breast cancer in Chinese women younger than age 40: are they different from their older counterparts? World J Surg 32:2554-2561
Mousavi SM, Montazeri A, Mohagheghi MA, Jarrahi AM, Ha- rirchi I, Najafi M, Ebrahimi M (2007) Breast cancer in Iran: an epidemiological review. Breast J 13:383-391
Parkin DM, Sitas F, Chirenje M, Stein L, Abratt R, Wabinga H (2008) Part I: cancer in Indigenous Africans—burden, distribution, and trends. Lancet Oncol 9:683-692
Tavassoli F, Devilee P (eds) (2003) Tumours of the breast and female genital organs. IARC Press, Lyon Mousavi SM, Zheng T, Dastgiri S, Miller AB (2009) Age distribution of breast cancer in the Middle East, implications for screening. Breast J 15:677-679
Smigal C, Jemal A, Ward E, Cokkinides V, Smith R, Howe HL, Thun M (2006) Trends in breast cancer by race and ethnicity: update 2006. CA Cancer J Clin 56:168-183
Hemminki K, Lorenzo Bermejo J (2005) Effects of screening for breast cancer on its age-incidence relationships and familial risk. Int J Cancer 117:145-149
Hemminki K, Rawal R, Lorenzo Bermejo J (2004) Mammo- graphic screening is dramatically changing age-incidence data for breast cancer. J Clin Oncol 22:4652-4653
Olsson S, Andersson I, Karlberg I, Bjurstam N, Frodis E, Ha- kansson S (2000) Implementation of service screening with mammography in Sweden: from pilot study to nationwide pro-gramme. J Med Screen 7:14-18
Hemminki K, Ji J, Brandt A, Mousavi SM, Sundquist J (2010) The Swedish Family-Cancer Database 2009: prospects for histology-specific and immigrant studies. Int J Cancer 126:22592267
Hemminki K, Li X, Plna K, Granstrom C, Vaittinen P (2001) The nation-wide Swedish Family-Cancer Database: updated structure and familial rates. Acta Oncol 40:772-777
Hemminki K, Li X, Czene K (2002) Cancer risks in first generation immigrants to Sweden. Int J Cancer 99:218-228
McCracken M, Olsen M, Chen MS Jr, Jemal A, Thun M, Cok- kinides V, Deapen D, Ward E (2007) Cancer incidence, mortality, and associated risk factors among Asian Americans of Chinese, Filipino, Vietnamese, Korean, and Japanese ethnicities. CA Cancer J Clin 57:190-205
Deapen D, Liu L, Perkins C, Bernstein L, Ross RK (2002) Rapidly rising breast cancer incidence rates among Asian- American women. Int J Cancer 99:747-750
de Waard F (1979) Premenopausal and postmenopausal breast cancer: one disease or two? J Natl Cancer Inst 63:549-552
Jatoi I, Anderson WF, Rosenberg PS (2008) Qualitative age- interactions in breast cancer: a tale of two diseases? Am J Clin Oncol 31:504-506
Ziegler RG, Hoover RN, Pike MC, Hildesheim A, Nomura AM, West DW, Wu-Williams AH, Kolonel LN, Horn-Ross PL, Rosenthal JF, Hyer MB (1993) Migration patterns and breast cancer risk in Asian-American women. J Natl Cancer Inst 85:1819-1827
Boyle P, Levin B (eds) (2008) World cancer report 2008. IARC, Lyon Grummer-Strawn L (1996) The effect of changes population characteristics on breastfeeding trends in fifteen developing countries. Int J Epidemiol 25:94-102
Wallby T, Hjern A (2009) Region of birth, income and breastfeeding in a Swedish county. Acta Paediatr 98:1799-1804
Lagerlund M, Maxwell AE, Bastani R, Thurfjell E, Ekbom A, Lambe M (2002) Sociodemographic predictors of non-attendance at invitational mammography screening—a population-based register study (Sweden). Cancer Causes Control 13:73-82
IARC (2002) Breast cancer screening. IARC Press, Lyon Miller A (2010) Conundrums in screening for cancer. Int J Cancer 126:1039-1046
Zahl PH, Maehlen J, Welch HG (2008) The natural history of invasive breast cancers detected by screening mammography. Arch Intern Med 168:2311-2316
Bernstein L (2008) Identifying population-based approaches to lower breast cancer risk. Oncogene 27(Suppl 2):S3-S8
IARC (2002) Weight control and physical activity. IARC, Lyon Parkin DM, Bray FI, Devesa SS (2001) Cancer burden in the year 2000. The global picture. Eur J Cancer 37(Suppl 8):S4-S66
Hemminki K, Li X (2002) Cancer risks in second-generation immigrants to Sweden. Int J Cancer 99:229-237
Friedenreich CM, Woolcott CG, McTiernan A, Ballard-Barbash R, Brant RF, Stanczyk FZ, Terry T, Boyd NF, Yaffe MJ, Irwin ML, Jones CA, Yasui Y, Campbell KL, McNeely ML, Karvinen KH, Wang Q, Courneya KS (2010) Alberta physical activity and breast cancer prevention trial: sex hormone changes in a yearlong exercise intervention among postmenopausal women. J Clin Oncol 28:1458-1466
Jatoi I, Miller A (2003) Why is breast-cancer mortality declining? Lancet Oncol 4:251-254