Socio-demographic risk factors of esophageal carcinoma: A case control study in a tertiary care hospital, Kabul Afghanistan

Unmatched case control study of socio-demographic risk factors and was conducted We enrolled and analyzed 132 EC cases and 132 normal controls to nd out the associated risk factors for EC Result lower to low socioeconomic status, and having positive family history in rst degree our


Introduction
Non-communicable diseases are expanding all around quickly, and are now responsible for the majority of global deaths. The vast majority of non-communicable diseases belong to cancer (1). Esophageal cancer (EC) is the 7th most common cancer in term of incidence and 6th common cancer in term of cancer death worldwide. In 2018 worldwide, 572,000 new cases were diagnosed and as overall 509,000 Interestingly, in developed countries in studies with wide range of educational histories and income range, SES is still associated with the risk of EC (8). As the cancer higher incidence belt is from northern Africa, to middle east and central Asia including Afghanistan to parts of China (9). Afghanistan which is a limited resources country, still no study has been published to evaluate the associated risk factors with EC. Current study is the rst study conducted to see the associated risk factors with EC.

Methodology:
Design This was unmatched case control study design to explore the sociodemographic risk factors association with EC.

Study population and setting
The data was collected from EC patients and control groups during time period of January 2019 up to January 2020. A total of 264 (132 cases and 132 controls) participants were included in the study. All EC cases were diagnosed and con rmed by histopathologic examination at the department of Pathology and Clinical Laboratory at French Medical Institute for Mother and Children (FMIC). The control groups were selected from adult participants who were visiting the hospital for routine health check-up with no cancer, cancer history and signs of cancer. All the questions were asked by direct interview from participants in local languages (Pashto & Farsi) following informed consent and using a structured questionnaire.

Risk factors
Variable included; Age by year, gender is selected as an independent/exposure variable both male and female are included in this case-control study, provinces (country zone) of living. The living situation is also classi ed as rural and urban area. The urban living de ned as living in the capital or center cities of the provinces while living in the districts and villages of the country de ned as rural area.
Weight and height counted by Kg and cm, respectively. Ethnicity is classi ed according to available ethnicity groups in the country. The common ethnicity group in Afghanistan are Pashtun, Tajik, Hazara, Turk (Uzbek and Turkmen) and the remaining population which make less percentage are labeled as other ethnicity group.
Education level was categorized as illiterate/primary school (who never gone school or completed only elementary school), Secondary/high school or higher (beyond secondary school, graduated from high school or more).
The SES was categorized as low-income, middle/high-income as follows; Participants who don't have own house, permanent job, and other property are known as low-income, while participants who have own house, car, and other property are known as middle/high-income.
Family history of the cancer for each participant was asked about the history of cancer in rst degree family as yes or no. The cancer history was classi ed for site of the cancer history as esophageal and non-esophageal and also relation of study participant with family member who had cancer.

Histopathology
The tissue biopsies were submitted in formalin. Grossly, all the tissues were small gray-white endoscopic biopsied specimens. Microscopic slides of the tissue block were made and stained with hematoxylin and Eosin (H&E) stain. The stained slides were seen under microscope for diagnosis and histologic subtype of the tumor.

Statistical analysis
Statistical Package for the Social Sciences (SPSS, version 25) was used for analyzing the data. Mean and standard deviation were calculated for continuous variables (age, weight and height of the participants), frequency and proportion were calculated for categorical variables. Binary logistic regression test of univariate analysis was used to estimate the risk of hypothesized risk factors for their unadjusted associations with EC. Signi cant variables (p-value < 0.05) in univariate logistic regression were further analyzed in multivariant logistic regression model to see independent associations with EC.

Results
Among total number of 132 EC cases, ESCC was the predominant type (74%) and 26% were EAC cases.
Majority of our study population was male (54.5%). Table 1 shows the detail descriptive information of risk factors. Age group of EC cases ranged from 30-85 years. 55-64 year was the common age group for cases with mean age of 59.48 ± 9.9 years compared to control group mostly in the age group of 45-54 years with mean age of 48.05 ± 11.02 years. Statistically it showed the risk of the getting EC is higher in older age group (OR: 1.070, 95% CI, p-value < 0.001). In Tables 2 and 3 the detail information of univariate and multivariate analysis for cases and controls are described, respectively.

Discussion
The great majority of the EC cases are ESCC and EAC which are making 95% of all EC cases. Among these two most common types ESCC is more prevalent in under-developed countries and EAC is prevalent in developed countries (10,11). However, the ESCC remains the most common histologic subtype of EC worldwide (12). Since the esophagus is lined by squamous epithelium, it could be a reason ESCC is more predominant. Moreover, other risk factors are also having important role. In our study the predominant subtype was ESCC (68%).
EC occurs mostly in older age groups. The risk rises with age, with an average age of 67 years at diagnosis (13). In our study the mean age of case groups was 59.4 years old compared to mean age of control (48 years). Study in India (14),Tanzania (15) and China (16) reported the majority of EC cases occurred in age group of > 60 years, > 65 years, and 60-69 years, respectively. Although EC occurs in the older age groups the lowest age for EC in the present study is 30 years. Study in more endemic area for EC, such as China, Japan and Iran also reported that EC starting from 30 years old age onward (17). The likely reason why EC occurs in older age maybe due to increase in the exposure to environmental risk factors and certain speci c genes that are more likely to be altered and mutated by increasing age.
The EC most commonly affect males than female population. The incidence of EAC is 6-10 times higher in male than female and the incidence of ESCC is 2-3 times higher (18). In our study most of the EC cases were in males and showed 3.3-time associated risk of EC in male groups. Similar distribution were reported by study was conducted in Tanzania (15), an ecological study, which conducted based on GLOBOCAN project of World Health Organization (WHO) for Asian counters estimated that 70.33% incidence of EC cases in male and 29.87% in female (19). In addition, American Society of Cancer (20) and cancer research from United Kingdom (UK) countries reported the higher incidence of EC in male over female (21). Equally or higher distribution of EC among females is a rare epidemiologic feature of EC which is reported in Linxian, China. This may show a single, very powerful risk factor shared by both genders (22). By the reviewing of literatures, it was not clear why the EC cases occur more in male than female, but probably due to the exposure to different environmental factors. As, the EC most common in farmers and workers in agricultures as men commonly do these works in rural areas. Others risk factors such as smoking, alcohol consumption and snuff dipping are also common among males.
The current study showed belonging to low SES comparing to middle/high SES, has the higher associated risk with EC (OR = 14.08). In addition, according to the education level, participants who were illiterate/having primary school education comparing to those who studied up to high school or more are at higher risk of EC (OR = 11.21). Many studies and different countries reported low SES and low education are related to the increase incidence of EC. Study was obtained in India indicated that 30.91% of EC patients were illiterates, 73.91% patients belonged to lower SES (23). Case control studies are conducted in Chinese and Iranian population sample revealed strong association between low SES and an increased risk of EC (24,25). Low SES and increases the risk of EC incidences, are also reported in developed countries than might be expected. A case-control analysis in the United States by Gammon et al. estimated that the risk of EC among those with low income and low education was higher (26). The correlation of low SES and EC was documented by a case control study in Sweden, the unquali ed employees were at 3.7 times the risk of AC and 2.1 times the risk of ESCC (27). Low education has also had negative impact on the prognosis and survival rate of the EC. In this regard a cohort study in Sweden reveals clear association between lower education and increased mortality after esophagectomy of EC patients (28).
Very large number of the EC patients in current study were living in rural areas (92.4%) and suggest an increased associated risk of EC among people living in rural areas (OR = 25.16). This ndings are in line with study has been conducted in Turkey and reported the association of EC and rural population (p < 0.001) (29) and study in India also showed high prevalence of EC in rural areas (30).
Despite, the occupation did not show signi cant association with risk of EC in multivariate model in our study, but the majority of the EC patients were unemployed (93.9%). Of these unemployed EC patients 75% were farmer and all female EC patients were housewives. Studies in Brazil and Iran reported the that EC was more prevalent in farmers in high prevalent areas (25,31). The work environment in agriculture is complex, with many potential hazardous exposures, such as pesticides, herbicides, fertilizers, dusts, zoonotic microbes, and sunlight (32). The reason why it is not signi cant in our study, will be belongs to small sample size and this need larger and wide study on this class population.
Weight and height did not show statistically signi cance association with EC in our study. A metaanalysis study showed high risk of obesity with EAC but inverse association to ESCC (33). Another study in Netherland also showed signi cant association between obesity and overweight with EC (34). No signi cance of EC and body mass index (BMI) in our study is because all cancer patients had the weight loss during the time of the diagnosis of the cancer.
In addition to environmental risk factors, hereditary susceptibility in esophageal carcinogenesis is another noticeable risk factor. Meanwhile, the co-occurrence of EC in family members is not always related to hereditary susceptibility but it also be consider as environmental factors (35). A study by Chen et al (36), reported that EC patients with positive family history increased 2-fold risk of developing of EC, while for whose both parents had history of cancer, 8-fold risk had been observed. Another study in high endemic area of the Iran showed more than 2-times risk for people who had positive rst-degree family history of the cancer (37).
In the present study, the same association between family history of cancer in rst degree family and EC was identi ed (OR = 4.71). The speci c genetic expression related to EC in Afghanistan, however, have not been researched yet, which is important in prevention and treatment of EC. However, a multi-center casecontrol study in USA revealed no statistically signi cant risk of positive family history with EC (38). The inconsistency in the result from different articles might be due to different genetic susceptibility and different type environmental risk factors exposures.
To the best of our knowledge, this is the rst study in Afghanistan in case control design which estimate the associated risk factors with EC. The study has been conducted in one of the main pathology centers of the country which receive biopsy samples from all around the country.
The limitations of this study includes small sample size due to the limited number of oncology center in the country. Secondly due to a smaller number of the EC cases in particular EAC cases, we were not able to run statistical analyses for both subtypes of EC separately. Thirdly, the risk factors are not studied in details because this is the rst research assessing the risk factors in Afghanistan.

Conclusion
There is a growing occurrence of EC in Afghanistan, but very limited data are available. Our study concluded that EC is common cancer in older ages groups (> 65y) and in male over female. Our study also provided the evidence that living in rural areas, being un-educated or having lower education, belonging to low SES, and having family history in rst degree relatives may contribute to the etiology of EC. These positive associations can help to suggest EC preventive measures and screening programs in early detection of EC in people who affected by above positive risk factors. Availability of data and materials: All data generated or analyzed during this study are included in this published article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests Funding: The authors received no speci c funds for this study.
Authors' contributions: The hypotheses were made by RS. RS collected the data. The data analyzed by RS, NA and TA. RS wrote the manuscript and reviewed the literatures. NA, TA, JAG and RS were the major contributors for critically revising the manuscript for important intellectual content. JAG has given expert opinion and nal approval of the version to be published. All authors read and approved the nal manuscript.