Bahareh Yazdizadeh et al Asian Pacific Journal of Cancer Prevention, Vol 6, 2005 130 Asian Pacific J Cancer Prev, 6, 130-134 RESEARCH COMMUNICATION Introduction Cancers of upper gastrointestinal tract, mainly esophagus and stomach are the most frequent cancers in Iran (Mosavi- Jarrahi et al., 2001). The existence of geographic variation in the frequency of GI cancer in the country has been noted since the first report on high rate of esophageal cancer in north-eastern part of Iran was published (Kemet and Mahboubi, 1972). The variations seen in the incidence of different cancers in the country have been subject of interest among epidemiologist, both locally and internationally. There is consensus among Iranian epidemiologists to divide the country into two regions for cancer incidence, the Caspian littoral for high rates of upper GI cancer mainly esophagus and gastric cancers, and other parts of the country with low rates of esophageal cancers and high rate of skin Time Trends in the Occurrence of Major GI Cancers in Iran Bahareh Yazdizadeh 1 , Alireza Mosavi-Jarrahi 1,2 , Hossein Mortazavi 3 , Mohammad Ali Mohagheghi 1 , Soroush Tahmasebi 1 , Azin Nahvijo 1 1 The Cancer Institute Research Center, the Imam Khomeini Medical Center. 2 Dept. of Social Medicine, Medical School, Shaheed Beheshti University of Medical Sciences. 3 Dept. Radiation Oncology, Jorjani Hospital, Shaheed Beheshti University of Medical Sciences. Corresponding Authors Address: Alireza Mosavi-Jarrahi, P.O. Box 18575-4194, Tehran, I. R. of Iran Tel: +98-21-23872567 Fax: +98- 21-6428655 rmosavi@yahoo.com Abstract Objective: The aim of this study was to study the changes in occurrence of esophageal, stomach and colon cancers (cancers of interest) over the last 30 years in Iran. Material and Methods: Cancer cases referred to two main cancer centers in the country (the Shiraz cancer center and the Tehran cancer center) during last 30 years and published by the two centers were utilized. Morbidity odds ratios (MOR) were used to study trend in the occurrence of each cancer site in each center. For this purpose the cancers of interest were considered as cases; childhood cancers as controls; and calendar year as exposure. A regression line was fitted to morbidity odds ratios over years and the slope of the regression line was considered to indicate the overall trend. MORs and 95% CIs comparing the last five and first five years were computed to measure the magnitude of the change over time. Result: The overall trend for esophageal cancer was decrease (slopes = -0.02 for Shiraz and -0.03 for Tehran); for stomach was increase (slopes = 0.04 for Shiraz and 0.08 for Tehran), and for colon cancer was sharp increase (slopes = 0.02 for Shiraz and 0.10 for Tehran). The magnitude of changes showed stomach cancer to increase by 35% in Shiraz (MOR = 1.35 with 95% CI 1.1, 1.65) and 13% in Tehran (MOR = 1.13 with 95% CI 0.96, 1.38), esophageal cancer to decrease by 20% in Shiraz (MOR = 0.82 with 95% CI 0.62, 1.11) and 50% in Tehran (MOR = 0.52 with 95% CI 0.45, 0.60), and colon cancer to increase by 65% in Shiraz (MOR = 1.65 with 95% CI 1.26, 2.16) and 82% in Tehran (MOR = 1.82 with 95% CI 1.52, 2.25). Conclusion: During the last thirty years the occurrence of major GI cancers has changed in Iran with sharp increase in colon cancer, slight to moderate increase in stomach cancer and sharp decrease in esophageal cancer. Key Words: Cancer - time trends - esophagus - stomach - colon - Iran cancers and malignancy of lymphoproliferative and hemopoietic systems (Mosavi-Jarrahi et al., 2001). However, certain changes in patterns of cancer occurrence are anticipated, especially since the country has transformed from a static-agricultural society to a dynamic-urban society during the last 50 years. Epidemiologic investigations of major GI cancers in the countries where reliable cancer registry is available have generated data demonstrated certain trends in incidences, mainly lowering cancers of esophagus and stomach and increase in the incidence of colon and lower GI cancers (Muir and Nectoux, 1996). Although, there are recent population data regarding the incidence of GI cancers in Iran, due to lack of population based cancer registries, monitoring the changes in cancer morbidity and mortality over years is not as straight-forward as comparing rates at different times and