Pancreatic cancer is one of the deadliest forms of cancer, causing an estimated 432,000 deaths worldwide each year [
Due to the lethality of pancreatic cancer, mortality rates may act as surrogates for incidence rates. Mortality varies by province in Egypt, with northern districts having an average rate that is 2.85 times the rate of southern districts [
While mortality rates of pancreatic cancer across Egypt have been previously reported, this is the first study to describe the geographic distribution of incidence in Egypt. The purpose of this study is to characterize the distribution of pancreatic cancer incidence in the province of Gharbiah using population-based cancer registry data and to lay the groundwork for future etiologic studies to investigate the possible association between regional distribution of pancreatic cancer incidence and potential risk factors.
The study population included all men and women 18 years of age and older who resided in the Gharbiah province and were diagnosed with pancreatic cancer between 1999 and 2010, as recorded in the Gharbiah Population-based Cancer Registry (GPCR) [
The province of Gharbiah is located in the central Nile Delta region and consists of eight districts with a total population of approximately 4 million people, according to the latest Egyptian census of 2006 [
The GPCR, located in the capital city of Tanta, was established in 1998 as a part of the Middle East Cancer Consortium (MECC) and received funding from the U.S. National Cancer Institute in Bethesda and the Egyptian Ministry of Health and Population until being discontinued due to lack of funding in 2011. The GPCR is known to be of high quality in terms of completeness and reliability. The International Association of Cancer Registries (IARC), the Surveillance Epidemiology and End Results (SEER) group of the National Cancer Institute, and the Department of Epidemiology of the University of California at Irvine have conducted validation and control quality checks with the MECC showing high data quality of this registry [
The GPCR used active case finding to collect information on all cancer cases within the entire Gharbiah province. Pancreatic cancer cases included in the registry were classified based on the World Health Organization’s ICD-O-2 coding for 1999-2000, followed by the ICD-O-3 coding from 2001 onward [
Population data for each district of Gharbiah were obtained from the 1996 and 2006 censuses conducted by the Central Agency for Public Mobilization and Statistics (CAPMAS) [
Data were obtained electronically from the GPCR, and descriptive statistics were generated using SAS Studio 3.8. Crude age- and sex-specific incidence rates (IR) were then calculated by dividing the number of pancreatic cancer cases per group (E) by the corresponding age- and sex-specific population estimates calculated with CAPMAS census data. Age-standardized rates were then calculated by direct standardization utilizing the Segi-Doll world standard [
Shapefiles for Egyptian provinces and districts were downloaded from the Humanitarian Data Exchange and imported into ArcMap 10.6.1. A map layer created by the team was utilized to discriminate between urban and rural areas. Incidence data were imported into ArcMap in the form of a geodatabase, and choropleth maps were created.
A total of 1,089 pancreatic cancer cases (36.3% female and 63.7% male) were reported to the GPCR from 1999 to 2010 and included in this study (Table
Clinical and demographic characteristics of 1,089 pancreatic cancer cases in Gharbiah during the period of 1999-2010 by urban and rural places of residence.
Variable | Description | Urban no. (%) | Rural no. (%) | Total cases |
---|---|---|---|---|
Year of diagnosis | 1999 | 36 (8.20) | 31 (4.77) | 67 (6.15) |
2000 | 30 (6.83) | 38 (5.85) | 68 (6.24) | |
2001 | 29 (6.61) | 35 (5.38) | 64 (5.88) | |
2002 | 41 (9.34) | 42 (6.46) | 83 (7.62) | |
2003 | 36 (8.20) | 45 (6.92) | 81 (7.44) | |
2004 | 44 (10.02) | 44 (6.77) | 88 (8.08) | |
2005 | 29 (6.61) | 62 (9.54) | 91 (8.36) | |
2006 | 42 (9.57) | 58 (8.92) | 100 (9.18) | |
2007 | 39 (8.88) | 65 (10.00) | 104 (9.56) | |
2008 | 31 (7.06) | 77 (11.85) | 108 (9.92) | |
2009 | 44 (10.02) | 83 (12.77) | 127 (11.66) | |
2010 | 38 (8.66) | 70 (10.77) | 108 (9.92) | |
All years | 439 | 650 | 1089 | |
Age | 0-24 | 1 (0.23) | 3 (0.46) | 4 (0.37) |
25-29 | 0 (0.00) | 3 (0.46) | 3 (0.28) | |
30-34 | 7 (1.59) | 8 (1.23) | 15 (1.38) | |
35-39 | 13 (2.96) | 23 (3.54) | 36 (3.31) | |
40-44 | 17 (3.87) | 38 (5.85) | 55 (5.05) | |
45-49 | 37 (8.43) | 65 (10.00) | 102 (9.37) | |
50-54 | 73 (16.63) | 90 (13.85) | 163 (14.97) | |
55-59 | 71 (16.17) | 105 (16.15) | 176 (16.16) | |
60-64 | 76 (17.31) | 103 (15.85) | 179 (16.44) | |
65-69 | 52 (11.85) | 84 (12.92) | 136 (12.49) | |
70-74 | 46 (10.48) | 66 (10.15) | 112 (10.28) | |
75+ | 46 (10.48) | 62 (9.54) | 108 (9.92) | |
Sex | Female | 273 (62.19) | 421 (64.77) | 694 (63.73) |
Male | 166 (37.81) | 229 (35.23) | 395 (36.27) | |
Basis of diagnosis1 | Death certificate only | 54 (12.41) | 19 (2.92) | 73 (6.73) |
CT scan/US/MRI | 60 (13.79) | 128 (19.69) | 188 (17.33) | |
Exploratory surgery/ERCP | 103 (23.68) | 158 (24.31) | 261 (24.06) | |
Radiology with tumor marker | 60 (13.79) | 105 (16.15) | 165 (15.21) | |
Cytology/hematology | 9 (2.07) | 12 (1.85) | 21 (1.94) | |
Histology of metastases | 32 (7.36) | 87 (13.38) | 119 (10.96) | |
Histology of primary | 116 (26.7) | 141 (21.69) | 257 (23.69) | |
Autopsy with histology | 1 (0.23) | 0 (0.00) | 1 (0.09) | |
District | Tanta | 148 (33.71) | 137 (21.08) | 285 (26.17) |
Mehalla | 153 (34.85) | 139 (21.38) | 292 (26.81) | |
Kafr El-Zayat | 46 (10.48) | 72 (11.08) | 118 (10.84) | |
Zefta | 24 (5.47) | 55 (8.46) | 79 (7.25) | |
Samanoud | 21 (4.78) | 51 (7.85) | 72 (6.61) | |
El Santa | 19 (4.33) | 81 (12.46) | 100 (9.18) | |
Kotour | 13 (2.96) | 65 (10.00) | 78 (7.16) | |
Basyoun | 15 (3.42) | 50 (7.69) | 65 (5.97) |
1Four cases (0.37%) had missing data on the basis of diagnosis.
Figure
Map of pancreatic cancer incidence rates in urban and rural Gharbiah for males and females (1999-2010).
Age-adjusted pancreatic cancer incidence rates per 100,000 with 95% confidence intervals for 1,089 cases in Gharbiah by district, sex, and urban/rural place of residence (1999-2010).
Region | Urban | Rural | Total | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
M | F | Overall | No. cases | M | F | Overall | No. cases | M | F | Overall | Number of cases | |
Tanta | 4.45 | 3.59 | 4.04 | 148 | 4.77 | 2.76 | 3.70 | 137 | 4.60 | 3.16 | 3.88 | 285 |
(3.52-5.39) | (2.69-4.49) | (3.39-4.70) | (3.74-5.79) | (2.03-3.50) | (3.08-4.32) | (3.91-5.29) | (2.85-3.73) | (3.43-4.33) | ||||
Mehalla | 4.63 | 3.34 | 4.05 | 153 | 5.65 | 2.12 | 3.84 | 139 | 5.08 | 2.71 | 3.93 | 292 |
(3.68-5.57) | (2.50-4.17) | (3.41-4.69) | (4.54-6.76) | (1.46-2.79) | (3.20-4.48) | (4.37-5.80) | (2.18-3.25) | (3.48-4.38) | ||||
Kafr El-Zayat | 9.58 | 5.93 | 7.89 | 46 | 4.02 | 2.48 | 3.21 | 72 | 5.31 | 3.08 | 4.18 | 118 |
(6.21-12.95) | (2.93-8.93) | (5.61-10.17) | (2.82-5.22) | (1.57-3.38) | (2.47-3.95) | (4.10-6.53) | (2.17-3.99) | (3.42-4.93) | ||||
Zefta | 5.58 | 1.42 | 3.58 | 24 | 2.50 | 1.87 | 2.18 | 55 | 3.19 | 1.76 | 2.46 | 79 |
(3.13-8.02) | (0.03-2.81) | (2.15-5.01) | (1.62-3.37) | (1.12-2.62) | (1.60-2.75) | (2.32-4.07) | (1.11-2.41) | (1.92-3.01) | ||||
Samanoud | 5.77 | 4.59 | 5.15 | 21 | 3.61 | 2.39 | 2.96 | 51 | 4.08 | 2.71 | 3.38 | 72 |
(2.63-8.91) | (1.41-7.76) | (2.95-7.36) | (2.32-4.90) | (1.37-3.41) | (2.15-3.77) | (2.86-5.30) | (1.72-3.70) | (2.60-4.16) | ||||
El Santa | 11.02 | 5.44 | 8.32 | 19 | 4.69 | 2.22 | 3.36 | 81 | 5.24 | 2.45 | 3.76 | 100 |
(5.03-17.01) | (1.09-9.80) | (4.58-12.06) | (3.41-5.96) | (1.41-3.02) | (2.63-4.09) | (3.97-6.52) | (1.64-3.26) | (3.02-4.50) | ||||
Kotour | 14.28 | 11.27 | 12.94 | 13 | 10.42 | 3.06 | 6.45 | 65 | 10.70 | 3.62 | 6.91 | 78 |
(3.70-24.85) | (2.25-20.29) | (5.91-19.98) | (7.44-13.40) | (1.65-4.48) | (4.88-8.01) | (7.84-13.55) | (2.17-5.07) | (5.38-8.44) | ||||
Basyoun | 4.40 | 2.38 | 3.43 | 15 | 5.29 | 2.06 | 3.57 | 50 | 5.09 | 2.15 | 3.56 | 65 |
(1.67-7.13) | (0.29-4.46) | (1.69-5.16) | (3.54-7.05) | (1.02-3.10) | (2.58-4.56) | (3.60-6.58) | (1.21-3.09) | (2.69-4.42) | ||||
Total | 5.21 | 3.63 | 4.45 | 439 | 4.68 | 2.31 | 3.43 | 650 | 4.87 | 2.72 | 3.78 | 1089 |
(4.59-5.82) | (3.08-4.19) | (4.04-4.87) | (4.23-5.12) | (2.01-2.61) | (3.17-3.70) | (4.51-5.23) | (2.46-2.99) | (3.56-4.01) |
Age-specific pancreatic cancer incidence rates per 1,089 cases in the Gharbiah district of Egypt (1999-2010).
In all districts, incidence rates were higher in urban areas (4.45/100,000) compared to rural areas (3.43/100,000). The highest incidence rates were observed in urban centers of Kotour, El Santa, and Kafr El-Zayat districts (12.94, 8.32, and 7.89 per 100,000, respectively). The districts with the highest urban-rural incidence rate ratios (IRRs) include El Santa (
This was the first study to investigate the geographic distribution of pancreatic cancer incidence in Egypt. The study revealed two interesting trends. First, the results showed that incidence rates were significantly higher in urban compared to rural districts. Second, the incidence rates of pancreatic cancer varied greatly by district.
Regarding the difference between urban and rural areas, we found that urban areas had incidence rates that were 1.3 times the rates of rural areas (95% CI: 1.29-1.31). The highest urban incidence was found in Kotour (14.28 per 100,000 in males and 11.27 per 100,000 in females) compared with the lowest rural incidence rate in Zefta (2.50 per 100,000 in males and 1.87 per 100,000 in females). These findings are in concordance with two studies conducted in China which demonstrated that pancreatic cancer incidence and mortality rates are higher in urban compared to rural areas [
We believe it is unlikely that the observed differences between urban and rural rates were due to disparities in access to medical care. Across Egypt, 95% of the population lives within 5 km of a primary health care facility [
Differences in urban and rural lifestyles and occupational exposures are likely to contribute to the observed differences in urban and rural incidence rates. Obesity and diabetes have been shown to vary significantly between urban and rural parts of Egypt; according to a 1995 study, 49% of urban populations with higher incomes in Egypt were obese and 20% had diabetes, compared to rural populations where only 16% were obese and 4.9% had diabetes [
In regard to occupational exposures, a hospital-based study conducted in the nearby province of Dakahleia found that farming, which is highly correlated with rural residence, was significantly associated with pancreatic cancer risk [
Regarding the observed regional variation, our results are consistent with other epidemiological studies demonstrating regional variation of pancreatic cancer incidence and mortality rates within the U.S., China, Japan, and France [
As the districts in Gharbiah are relatively homogenous in terms of lifestyle, age distributions, and sex ratios, regional variation in incidence rates may be attributed to differences in exposure to environmental risk factors. It is well documented that the Nile Delta region has some of the highest levels of environmental contamination in Egypt by heavy metals, hydrocarbons, and pesticides; pollution levels are particularly high in urban and industrial areas [
This study has multiple strengths. Mainly, it is the first study to characterize the geographic distribution of pancreatic cancer incidence in Egypt. The GPCR is a high-quality registry that received regular quality checks by the IARC and is routinely cited by Cancer Incidence in Five Continents [
With these strengths, the study had limitations. There is the possibility that in Egypt and in other regions of the world, people may die of pancreatic cancer before seeking medical care or without obtaining a medical diagnosis first. However, we have no reason to believe that this would lead to differential misclassification. Because population-based cancer registries are not designed for specific research investigations, the variables in the registry were limited to clinical and descriptive epidemiologic information. Well-designed case-control studies that can assess individual risk factors such as smoking frequency, occupational exposures, and diet are needed. Additionally, although we had a reasonable sample size of 1,089 patients for this rare cancer, we were unable to conduct analyses on a finer spatial resolution. Utilizing more nuanced urbanization classifications, such as distinguishing semiurban areas, would also be useful to examine whether there is a linear correlation between pancreatic cancer incidence and the rate of urbanization. The population estimates used to calculate incidence differed slightly from the years that cases were reported but were utilized based on the availability of census data from 1996 and 2006. Lastly, while we are able to provide rationale for the differential rates by urbanization, data were not available to definitively establish causality.
This study demonstrates that pancreatic cancer incidence is higher in urban compared to rural areas in Gharbiah, Egypt. The study also reveals geographic variation in pancreatic cancer incidence by district of residence. It highlights the importance of cancer registration in Egypt and a need for future registration to provide relevant public health insights. This study provides a preliminary examination into pancreatic cancer distribution in Egypt; however, we recommend the completion of more advanced geospatial analyses with larger datasets to confirm these initial findings. In a country undergoing great urban growth, understanding the epidemiology of pancreatic cancer and investigating the risk factors which may vary between urban and rural settings and between districts will be essential for gaining a better understanding of the disease etiology and for guiding cancer prevention and control strategies.
All data used in the study were provided by the Gharbiah Cancer Society. Data are made available to researchers following a formal proposal for a specific study and request for data.
The authors declare that there is no conflict of interest regarding the publication of this article.
We are grateful for Dr. Mohamed Hablas, Khaled Daboos, and other personnel at the Gharbiah Cancer Society for their assistance and support throughout this study. This work was supported by the Cancer Epidemiology Education in Special Populations (CEESP) program through a National Institute of Health/National Cancer Institute R25 grant [grant number CA112383].