Trends in the Incidence of Stomach Cancer in Golestan Province, a High-risk Area in Northern Iran, 2004–2016

Fatemeh Ghasemi-Kebria, MSc1; Taghi Amiriani, MD1; Abdolreza Fazel, MD2,3; Mohammad Naimi-Tabiei, MD2; Alireza Norouzi, MD1; Masoud Khoshnia, MD1; Mohammadreza Seyyedmajidi, MD1; Abdollah Pooshani, MD1; SeyedeFatemeh Mousaviemadi, MD1; Houshang Poorkhani, MD4; SeyedMehdi Sedaghat, MD5; Faezeh Salamat, MSc1; Susan Hasanpour-Heidari, MSc1; Nastaran Jafari-Delouie, MSc1; Masoomeh Gholami, MSc6; Shahryar Semnani, MD3,1; Gholamreza Roshandel, PhD1*; Elisabete Weiderpass, PhD7†; Reza Malekzadeh, MD8*


Introduction
Stomach cancer was the most common cancer in the world less than a century ago. 1 More recent reports have suggested decreasing trends in incidence and mortality rates of this cancer worldwide. 2 Stomach cancer is now the fifth most common malignancy worldwide, after cancers of the lung, breast, colorectum and prostate, with 1 033 701 new cases (5.7% of the total) estimated in 2018. 3 It is the third most common cause of cancer-related death with 782 685 deaths (8.2% of the total) in 2018. 3 There are large differences in the incidence of stomach cancer by country. The highest estimated age standardized incidence rates (ASRs) as well as the highest mortality rates are reported from East Asia. 3 H. pylori infection may play an important role in this geographical variation, while other environmental factors should also be mentioned in this context. 4 The ASRs of stomach cancer in the West Asian population are 19.6 and 9.2 in men and women, respectively. 5 The results of cancer registry studies in Iran have shown that stomach cancer is one of the most incident cancer types, in particular in northwestern and northern regions of the country. 6 Epidemiological studies have indicated increasing trends in the incidence rates of stomach cancer in Iran. [5][6][7][8] The Golestan province, located in northeastern Iran, has been known as a high-risk area for upper GI cancers since the 1970s. 9 Previous reports from the Golestan province suggested that stomach cancer was the most common cancer in men and the third malignancy in women from 2004 to 2008. 10 The aim of this study is to describe the incidence rates as well as the temporal and geographical variations of stomach cancer in the Golestan province during the 13-year period (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016).

Methods
The Golestan province is located in the northeast of Iran, encompassing14 counties, 33 cities and 1051 villages. According to the 2016 National Population and Housing Census by the Center of Statistics of Iran, the Golestan province has a population of about 1 900 000, of whom 50% are men and about 50% live in urban areas. 11 The Golestan Population-based Cancer Registry (GPCR), a voting member of the International Association of Cancer Registries (IACR) since 2007, is a high-quality cancer registry collecting data on cancer patients throughout the province. The protocol and standards of data collection in the GPCR were described previously. 12 Briefly, the GPCR collects data on new incident primary cancers. The sources of GPCR data include health care centers and all public and private diagnostic and therapeutic centers throughout the province.
Data on cancer mortality were also obtained from the Golestan Death Registry unit in the deputy of health of the Golestan University of Medical Sciences. Cancer mortality data were linked to the incidence data to define cases with death certificate only.
The GPCR observes the third edition of the International Classification of Diseases for Oncology (ICD-O-3) for coding tumor characteristics, including topography, morphology, behavior and grade. 13 For the present study, data on newly diagnosed primary stomach cancers in the Golestan province during 2004-2016 were obtained from the GPCR. CanReg-5, a free software created and published by the International Agency for Research on Cancer (IARC) was used for data entry and analysis. 12,14 ASRs were calculated using the world standard population. We used the 18-group Segi's World standard population to calculate the ASR. Data on the Golestan population (2004-2016) was obtained from the statistics office of the deputy of health of the Golestan University of Medical Sciences ( Figure S1). To assess the geographical distribution of stomach cancer in the study population, we calculated the ASR of stomach cancer by major subdivisions (cities) of the Golestan province and presented the results on the map of Golestan. The joinpoint regression analysis was used to assess the temporal trends in incidence rate of stomach cancer. We used the Joinpoint software version 4.6.0.0 for this analysis. Average annual percent change (AAPC) and corresponding 95% confidence intervals (CI) were calculated. We also calculated the contribution of population aging, population size and risk to the changes in the incidence of stomach cancer using a previously described method. 15 P values less than 0.05 were considered statistically significant.

Results
Overall, 2,964 new cases of stomach cancer were registered in the GPCR during the study period. Among these, 2,041 (68.9%) were men and 923 (31.1%) were women with a mean (standard deviation) age of 65.3 (13.5) and 62.6 (13.5) years in men and women, respectively. Also, 1418 (47.8%) of patients lived in urban areas (cities) and 1546 (52.2%) lived in rural areas (villages). Table 1 shows the numbers, crude rate, and ASR of stomach cancer by gender, residence area, and calendar years. The age specific rate (per 100 000 person-years) of stomach cancer indicated increasing rates after 40 years of age in both genders and residence areas ( Figure 1).
As shown in Figure 2, there was a significant decreasing trend between 2004 and 2010 (AAPC = -4.36; P < 0.01), and a plateau during 2010-2016 (AAPC=0.83; P value =0.48) in the incidence of stomach cancer in the population of men in Golestan.
Our results suggested a significant decreasing trend in the ASR of stomach cancer in the rural population (AAPC= -2.14; P < 0.01), while there were no significant time trends in stomach cancer rates in urban areas (AAPC= 0.36; P = 0.46) (Figure 3).
The results of partitioning analysis are shown in Table  2 Table 2 shows that changes in population structure, including population aging and population size, resulted in a 43.26% increase in the number of new cases of stomach cancer during 2004-2016. The results also suggested that changes in risk factors could result in a 20.93% decrease in the number of stomach cancer cases during the study period.  As shown in Figure 4, there were geographical diversities in the ASR of stomach cancer in the subdivisions of the Golestan province. We found higher rates for stomach cancer in eastern parts of the province, especially in Kalaleh city (ASR = 38.9 in men and 15.8 in women) (Figure 4).

Discussion
We presented the 13-year trends in incidence of stomach cancer in high-risk area in the Golestan province, northern Iran. The ASRs of stomach cancer were 26.9 and 12.2 in men and women, respectively. According to the Globocan 2018 estimates, the worldwide ASRs of stomach cancer were 15.7 and 7.0 for men and women, respectively 3 . The Globocan 2018 report also suggested lower ASRs of stomach cancer in West Asian countries (11.3 and 6.0 in men and women, respectively). 3 The results of the IARC Cancer in Five Continents project suggested relatively low rates of stomach cancer in some neighboring countries of Iran including Saudi Arabia (3.5 and 2.1 in men and women, respectively), Kuwait (2.6 and 2.2 in men and women, respectively), and Qatar (6.6 and 4.9 in men and women, respectively) as well as some high rates of this cancer in parts of Turkey including Erzurum and Trabzon. 16 According to the most recent report form the Iranian National Cancer Registry, the ASRs of stomach cancer in Iran were reported at 21.2 and 9.4 in men and women, respectively. 17 There were diversities in incidence rates of stomach in different parts of Iran with lower rates in central and eastern parts Iran and higher rates in northern and northwestern areas including the Golestan province. 17,18 Previous reports from the Golestan province also suggested high rates of stomach cancer in this area. 9,10,19 Therefore, these findings confirm that the Golestan province is a high-risk area for stomach cancer.
The high prevalence of Helicobacter pylori infection in the Golestan province, 20,21 a class I risk factor for gastric cancer, 22 may be considered as a potential explanation for the high rates of stomach cancer in this area. In populations where the rates of H. pylori infection remain high, stomach cancer remains a major health problem even with implementation of other interventions. 2,23 Low consumption of fruit and vegetable, 24,25 tobacco smoking, 4,26 high consumption of salt [27][28][29] and opium consumption [30][31][32] were proposed as other risk factors for stomach cancer. The results from a large-scale cohort study 33 suggested the high prevalence of these risk factors and their relationship with esophageal cancer in Golestan, Iran. 34,35 Therefore, these risk factors may partly explain the high incidence of stomach cancer in this high-risk population. Further studies are warranted to clarify the issue in this area.
In the present study, stomach cancer was significantly higher in men and there was a peak in age-specific rates after the age of 40 years, which may be partly explained by the higher prevalence of known risk factors in men in the Golestan province, including alcohol consumption and cigarette smoking. 35 These finding are in line with previous reports, suggesting gender and age as risk factors for stomach cancer. 36,37 These factors should be considered in designing cancer control programs for stomach cancer.
Our findings also suggested higher rates of stomach cancer in rural areas (mostly villages). Similar disparities were reported in incidence rates of esophageal cancer in this population. 38 A number of known risk factors of esophageal cancer in Golestan, including low socioeconomic, opium and tobacco consumption, poor oral health, drinking un-piped water and exposure to polycyclic aromatic hydrocarbons [39][40][41][42][43][44] are likely more prevalent in the rural (village) population, although studies are scarce. 45 Most of these risk factors are common risk factors for both esophageal and gastric cancers. Further studies are needed to clarify this point and it should be mentioned by health policy makers. Table 2 shows that changes in risk factors contributed to a 20.93% decrease in the number of new cases of stomach cancer in 2016 compared to 2004. The results of joinpoint regression analysis also suggested significant decreasing trends in the incidence rates of stomach cancer in men and the rural population. These changes may be partly explained by the reduction in the prevalence of some risk factors (e.g. low socioeconomic status). This risk lowering due to improved socioeconomic status may reflect reduced H. pylori infection rates, 46 especially in areas with high rates of H. pylori infection such as the Golestan province. 20,21 This point should be further investigated in future studies.
Our findings also showed a plateau of decreasing trend in incidence of stomach cancer in men of the province after the year 2010. Similar findings have been previously   reported. 47,48 Our results also suggested a slight increasing trend in the rates in the urban population. Life style changes toward Westernized habits (e.g. fast food consumption, low physical activities) may be proposed to partly explain these trends.
Other factors, such as interventions, or changes in the data quality from the GPCR, may have also affected these trends. There were, however, no marked changes in GPCR protocols (sources, methods of data collection, definitions, and rules) since its start in 2004, 49 while no populationlevel cancer screening programs have been implemented in our population. Change in access to diagnostic and therapeutic services may, however, be in part responsible for the trends in the incidence rates of stomach cancer. Although there was no considerable change in the number of main sources used by the GPCR (pathology centers and hospitals) during the study period, the availability of other diagnostic services (e.g. imaging centers) as well as physicians (specialists and subspecialists), especially in the private sector, increased from 2004 to 2016.
Despite the decreasing trends in the ASR, there was a 22.33% increase in the number of new cases of stomach cancer during the study period, mainly due to changes in the population structure (population aging and population size). These findings may suggest an increasing trend in the burden of stomach cancer in the Golestan population. This point should be taken into account in future studies and should be considered as top priority and appropriate interventions should be designed and implemented for this cancer in this high-risk area.
We found geographical diversities in the incidence rates of stomach cancer with the highest rates in eastern parts of the Golestan province, especially Kalaleh city. These geographical patterns almost match those of esophageal cancer, suggesting (at least some) common risk factors for these two cancers in the Golestan province. Further investigations are needed to clarify this matter.
As in other population-based cancer registries, lack of data on risk factors was the major limitation of this study. Therefore, we could not provide appropriate interpretations for differences and trends in the incidence rates of stomach cancer in our population. This point should be addressed in future studies (e.g. case-control or cohort studies) to identify the most important risk factors of stomach cancer in Golestan, Iran.
In conclusion, our results emphasized the high incidence rates of stomach cancer in the Golestan province. We found higher rates of stomach cancer in men and the rural population. Our findings also suggested temporal and geographical diversities in the incidence rates of stomach cancer in this high-risk population. There were significant decreasing trends in the ASRs of stomach cancer in men (during 2004 to 2010) and the rural population (during 2004 to 2016). The incidence rates of stomach cancer were considerably higher in eastern parts of Golestan, especially Kalaleh city, which had been well known as high-risk regions for esophageal cancer. Stomach cancer should be addressed by local health policy makers as top priority in our population. Further studies are warranted to determine risk factors related to these trends and diversities.