Gastric cancer in East Azerbaijan, Iran: Five-year survival analysis of population-based cancer registry results

Use your smartphone to scan this QR code and download this article ABSTRACT Introduction: Gastric cancer is the most common cancer with significant increasing trends during the last decade in Iran. The aim of this study was to evaluate the epidemiologic profile of gastric cancer along with gastric cancer-specific survival analysis. Methods: This was an analytical cross-sectional study in which all gastric cancer data were analyzed using the database of the East Azerbaijan Population-Based Cancer Registry (EA-PBCR). The incidents of definitive gastric cancer diagnosis were between the period ofMarch 20th , 2015 toMarch 19th , 2017 ( = 3 Iranian solar years). The survival analysis was performed using the Kaplan-Meier method and life tables for 1to 5-year survival data. The Log-rank test and Cox regression were computed to test the equality of survival function and mortality hazard. Results: Overall, 2,631 newly diagnosed gastric cancer cases were registered for 3 years. Gastric cancer was 2.35 times more common in men than women. The most common age group was the 7th decadewith 531 (31.2%) gastric cancer cases. Most of the gastric cancer cases were non-cardia (n = 2,244, 85.29%) cancer, and the proportion of non-cardia to cardia gastric cancer was 5.8:1. Overall survival was 60.1%, and 1to 5-year survival proportions were 91.61%, 64.21%, 58.53%, 30.14% and 24.77%, respectively. Cardia cancers had a worse survival rate than non-cardia cancers, and the hazard of mortality was 1.33 times higher in cardia than noncardia cancers (hazard ratio or HR = 1.33; 95% CI: 1.05 1.68; P = 0.017). Conclusion: Non-cardia gastric cancer is still the most dominant subsite in East Azerbaijan, Iran. There was a higher 1to 5year survival proportion in East Azerbaijan, with lower overall mortality rates, compared to other regions of Iran.


INTRODUCTION
The incidence of gastric cancer is particularly high in Iran, where it remains a leading cause of cancerrelated deaths. Gastric cancer is the most common cancer among Iranian men, with an age-standardized incidence rate (ASR) of 21.6 per 100,000 males, and is third among females (ASR = 9.44/100,000 women) 1,2 . The incidence and mortality of gastric cancer, by contrast, rank fifth and third, respectively, in both genders worldwide 1 . Studies have shown that there has been an overall decrease of ASRs (between 2007 and 2017) by 6% worldwide and for all socio demographic index (SDI) quintiles. Gastric cancer was the most common cancer in men compared to women, especially in highmiddle SDI countries, with 1 in 33 men and 1 in 78 women worldwide developing stomach cancer over a lifetime 3 . Although the global incidence of gastric cancer has decreased in recent decades, in Iran as well 4 , there is still a high incidence of this cancer in the north and northwest regions of Iran compared to other regions of the country. Clustering the incidence of gastric cancer during 2004-2008 for the provincial-level, and in 2010 for the country-level, showed that the north and northwest of Iran are high-risk clusters [4][5][6] . The Ardebil Province in Northwest Iran has been introduced as a high-risk area-with the highest ASRs and age-standardized mortality rates (ASMRs) of gastric cancer [7][8][9][10] . Moreover, there has been a six-fold difference in gastric cancer mortality rates between the northern and southern parts of Iran 11 . Recently, in the East Azerbaijan Province (located in the northwest of Iran), gastric cancer has become the most common cancer with significant increasing trends-by the overall annual percentage change (APC) statistical test-which showed 1.8% APC in both genders during the last decade, which was higher in men (APC, 3.2%) than in women (APC, 1.3%) 12,13 . Since the incidence and mortality rates of gastric cancer have varied significantly in recent years in the different regions and provinces of Iran 4,14,15 , it is necessary to have an epidemiological study of the incidence, gender and age distribution, and morphological type of gastric cancer. The aim of this study was to evaluate the epidemiologic profile of gastric cancer, along with 1-to 5-year gastric cancer specific survival analysis, in cases from East Azerbaijan, Iran.

Data Sources
All newly diagnosed cancer data were collected in EA-PBCR database, during the period of the 20 th of March 2015 to the 19 th of March 2017 (= 3 Iranian solar years), and from different data sources. The major sources of data collection related to the cancer cases were reports from pathology laboratories, hospital records, radiotherapy centers, and death registry databases. The main demographic data, including the first and last name, national identification (NID) number, date of birth, date of diagnosis, gender, place of birth, and contact information were all recorded in the database. The morphology (including histology, behavior and grade of the cancer) and topography (primary site of the tumor) were coded and collected according to the International Classification of Disease for Oncology (ICD-O-3).

Population
This was an analytical cross-sectional study in which all gastric cancer data were analyzed using the database of the East Azerbaijan Population Based Cancer Registry (EA-PBCR), which included the definitive diagnoses of gastric cancer cases in the entire East Azerbaijan Province. The East Azerbaijan Province is located in the northwest of Iran, and has a population of approximately 4 million people from the 20 counties, 62 cities, and 44 districts which comprise the largest Azeri Ethnic population in Iran.

Quality Control
Statistical data analysis, as well as quality and consistency checks, were performed to ensure clean and non-duplicated data. Most of the cases provided their address and contact information; follow-up and outcome data were obtained by contacting the patients' relatives and/or from the Hospital Information System (HIS).

Statistical Analysis
Gastric cancer-specific 1-to 5-year survival analysis and mortality rates were calculated using STATA MP 14.2 (Stata Corp LP, College Station, TX, USA). The survival status was considered from the date of diagnosis (as new cases were recorded in the EA-PBCR) to the date of gastric cancer-related death, and for censored cases until the last follow-up time. Patients were followed until the end of September 2019. The survival analysis was performed using the Kaplan-Meier method and life tables for 1-to 5-year survival data. The log-rank test and Cox regression analysis were used to test the equality of survival function and mortality hazard (HR, with 95% CI) by age, gender, morphology, subsite, stage and grade.

Study Sample
We registered 2,631 new gastric cancer cases during 3 years: 918 (34.89%) cases during 2015, 782 (29.72%) during 2016, and 931 (35.39%) during 2017. Of the cases, 1,849 (70.19%) were males and 784 (29.81%) were females; the male to female ratio was 2.35:1. The mean age of the patients was 67.52 (SD 12.77) years, with a range of 21-99 years. The most common age group was the 7 th decade-with 531 (31.2%) cases having gastric cancer. The most common morphological type was intestinal type adenocarcinoma (n = 1322, 50.25%), followed by diffuse type AC (n = 677, 25.73%). Of the cases, 387 (14.71%) had cardia gastric cancer and 2,244 (85.29%) had non-cardia gastric cancer; the proportion of noncardia to cardia was 5.8:1. The stage of tumor was defined for 257 cases, whereby 53.70% of the cases (n = 138) were in stage II, and 27.63% of the cases were in stage III. The grade of cancer was available in 2,445 cases; of these, 384 (15.71%) were grade I, 150 cases (6.13%) were grade II, 253 cases (10.35%) were grade III, and most of the cases (n = 1,658, 67.81%) were grade IV gastric cancer. The descriptive data are shown in Table 1.

Survival Analysis
Of the 2,631 cases, survival analysis was performed for 1,288 cases (we did not have access to contact information for the other 1,343 cases). From these, 514 (39.90%) cases died and overall survival was 60.1% for the 5-year survival analysis. The 1-year to 5-year survival proportions were 91.61%, 64.21%, 58.53%, 30.14% and 24.77%, respectively. The overall mortality incidence rate was 17.19 per 100,000 people, and mean follow-up time was 23.22 (range of 6 -36) months. Log-rank test for equality of survival functions did not show any significant difference among cases ≤ 65 years and > 65 years (P log-rank = 0.881). There was not any significant difference of survival functions between males and females (P log-rank = 0.959), but  survival functions showed significant differences with respect to grade of tumor (P log-rank = 0.000) and stage of cancer (P Log-rank = 0.000), as higher stages and grades had worse survival functions. The morphological type of cancer did not have a significant effect on survival function (P log-rank = 0.316); however, cases with cardia tumors had worse survival than non-cardia tumors (P log-rank = 0.012) (Figure 1 ). Cox regression analysis at univariate level showed that the hazard of mortality was 1.33 times higher in cardia than non-cardia tumors (HR =  Ref.
---Ref.  to mortality hazards. The results were presented in detail in Table 3.

DISCUSSION
According to the last published results of the EA-PBCR, gastric cancer was the most common cancer among men (ASR 29.7) and the third common in women (ASR 13.3). Therefore, we aimed to evaluate the epidemiologic profile of gastric cancer-along with 1-to 5-year gastric cancer specific survival analysis-in East Azerbaijan, the northwest region of Iran. During the 3 years of study, 2,631 newly diagnosed gastric cancer cases were registered in East Azerbaijan. The estimated gastric cancer specific Age-Standardized Mortality Rates (ASMRs) in 29 different provinces of Iran revealed extremely high mortality rates. There is a 6-fold difference between the northern (ASMR = 29.1 per 100,000) and southern (ASMR = 5.0 per 100,000) regions of Iran, with an average ASMR of 15 for males and 8.1 for females 11 . Iran is a large and geographically diverse country with several ethnic groups. Therefore, the variation in ASMRs and the high gastric cancer mortality rate in Iran may be linked to gene-environmental factors. However, different environmental risk factor exposures, including H. pylori infections and patterns, may also contribute to these variations. The overall 5-year mortality rate in our study was 17.19 per 100,000; this rate has been on a tremendous decline during the last decade. According to the results of a study by Zendehdel et al., the ASMR in Iran in 2012 was 27.6 per 100,000 in men (ranking as the 2 nd highest ASMR), and 13.6 per 100,000 in women (ranking as the 4 th highest ASMR) 11 . Our results showed better 1-to 5-year survival proportions compared with previous reports from Iran [19][20][21][22] . According to other European reports, Iran still have lower survival rates than most developed countries 23 . In our study, since more than half of the cases were in stage II of disease, and non-cardia gastric cancer was more common than cardia (85% vs. 15%), these results may have led to better prognosis. Cancers of the gastric cardia have epidemiological characteristics which are more similar to those of esophageal adenocarcinoma, and with poor prognosis; incidentally, the incidence of cardia cancers has been increasing, particularly in high-income countries 24,25 . Evidence from previous studies have shown that the prognosis is better for non-cardia than cardia cancers [26][27][28] . There was a significant difference in survival function between cardia and non-cardia gastric adenocarcinoma in this study as well, with cardia cases having worse survival than non-cardia cases. Indeed, gastric cardia cancers had a 1.33-fold higher mortality hazard ratio than non-cardia cancers. Similar findings to ours have been reported in other studies conducted in Iran [19][20][21][22] . Regardless of the overall incidence of gastric cancer, the proportion of non-cardia to cardia involvement in different population types seems to have different patterns 7 . Malekzadeh et al. emphasized this theory of higher rates of gastric cardia vs. non-cardia in the north and northwest regions of Iran, and vice versa for the south region and low risk areas of Iran, such as Khuzestan 7 . However, the results of our study showed an obvious higher incidence of non-cardia cancers in East Azerbaijan; the 85% prevalence of H. pylori infection in the East Azerbaijan population probably explains the higher incidence of non-cardia cancer in the present study 29,30 . Gastric cancer is more common in men than women in most countries. The male to female ratio has ranged from 2.2:1 in developed countries, to 1.83:1 in developing countries 16 . In the present study, about 70% of gastric cases were males, and the male to female ratio was 2.35:1. Although the gastric cancer mortality rate is 2-3 times higher in men than women 1,31 , there was no significant difference in survival function among genders in this study. While most other studies from Iran have shown lower overall survival in males 19,20 , gender did not have any significant impact on mortality hazard of gastric cancer, according to the results of our study.

STRENGTHS AND LIMITATIONS OF THE STUDY
Population-based cancer registries are the best resources for epidemiological and surveillance systems and studies. In this study, we utilized one of the best and most reliable population-based cancer registry data in recent years. Indeed, the East Azerbaijan population-based cancer registry (EA-PBCR) has been established with high quality and timeliness in the region. EA-PBCR is in its first step toward improving the quality indicators, including comparability, validity, timeliness, and completeness of cancer data. However, we performed the survival analysis only on approximately half of the registered gastric cancer cases in this study because we could not access the contact information of the others. Moreover, the stage of the disease is not a mandatory variable in the recent Iranian National Cancer Registry program; therefore, the stage was not available for most of gastric cancer cases. We hope to achieve better and more complete data concerning the stage and grade of cancers since a nationwide program has been recently instated in Iran for cancer staging, and is an ongoing plan at present.

CONCLUSION
In conclusion, non-cardia cancer is still the dominant subsite of gastric cancer in East Azerbaijan. However, we had higher 1-to 5-year survival proportions in East Azerbaijan, with improvement in mortality rates in the region. Additional studies of gastric cancer, particularly with respect to cancer subsite and cancer morphology, would be most informative.

ACKNOWLEDGMENTS
We would like to thank you from The East Azerbaijan Population Based Cancer Registry (EAPBCR) for providing the data of this study.

AUTHOR'S CONTRIBUTIONS
PJ, RD: substantial contributed to conception and design, acquisition of data, and analysis and interpretation of data; PJ, RD: drafting the article, and revised manuscript critically for important intellectual content; MHS, HMA: revised manuscript critically for important intellectual content. All authors read and approved the final manuscript.

AVAILABILITY OF DATA AND MATERIALS
Data and materials used and/or analysed during the current study are available from the corresponding author on reasionable request.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE
The East Azerbaijan Population Based Cancer Registry (EAPBCR) provided the data of this study. The ethics committee of Tabriz University of Medical Sciences has been approved this project, and all patients information and records are confidential (Grant Number: IR.TBZMED.REC.1395.1333). We followed ethics rules of EAPBCR for manuscript publication.

CONSENT FOR PUBLICATION
Not applicable.