Survival of Stomach Cancer Cases in Khon Kaen, Thailand 2000-2012

Background: Stomach cancer is an aggressive malignancy that is difficult to detect at an early stage and therefore is characterized by poor survival rates. Over the last two decades, there has been no report of gastric cancer survival in Khon Kaen province, Thailand. The aim of this retrospective cohort study was to provide up-to-date information about the survival of gastric cancer patients in this province. Materials and Methods: Data from Khon Kaen population-based cancer registry, Faculty of Medicine, Khon Kaen University were newly obtained on 650 patients who were diagnosed with stomach cancer during the period 1 January, 2000 to 31 December, 2012. These were then followed up until death or the end of the study (31 December 2014). We calculated the observed survival with the actuarial life table method, and relative survival, defined as the ratio of observed survival in the group of the stomach cancer patients to the expected survival in the entire Thai population from the estimated generation life tables for Thailand of five-year birth cohorts from 1900 - 2000. Results: The 5 year observed and 5 year relative survival rates were 17.2 % (95% CI: 13.54-21.14) and 18.2 % (95% CI: 14.3-22.4), respectively. The highest 5 year relative survival rates were demonstrated among patients aged 45-65, with stage I or II lesions, with adenocarcinomas, with a body of stomach location, well differentiated and receiving surgery and/or chemotherapy. Conclusions: The observed and relative survival rates were close to each other. Our findings provide basic information beneficial to development of an effective treatment system and appropriately improved population-based cancer registration.


Introduction
International Agency on Research for Cancer (IARC) determined that stomach cancer ranks the third leading cause of cancer-specific mortality, which was 8.8%. They estimated age-standardized incidence rate (ASR) were 17.4 per 100,000 for males and 7.5 per 100,000 for females. Moreover, the number of deaths will be 1,174,013 in 2030 by the prediction of IARC (Wang et al., 2012;Ferlay et al., 2015). In Thailand, the overall estimated age-standardized incidence rates (ASR) were 3.6 per 100,000 for males and 2.5 per 100,000 for females (Imsamran et al., 2015).
Stomach cancer is one of the most common forms of malignancies, include Thailand (Suwanrungruang et al., 2006). It is an aggressive malignancy that is difficult to detect at early stage and over poor outcome (Wang et al., 2007;Zhang et al., 2015). The burden and severity of a cancer were reflected by patterns of survival (Che et al., survival, which use to estimated survival rate. As the cancer survival was a key measure of the effectiveness of health -care system (De Angelis et al., 2014). Therefore, the study of survival of cancer patients is essential for monitoring the successful cancer control (Dickman et al., 1999). Especially in relative survival, we use this approach as a key indicator for monitoring progress against cancer in population.
Population -based cancer registries collected the information on all cancer cases in defined areas. These data should confine analysis of survival to those cases who are residents of the registry area (Jensen, 1991). For most cancer registries, cause of death information obtained from death certificate is either unavailable or unreliable due to misclassification errors. Therefore, instead of calculating the probability of surviving cancer in the usual way, considering deaths from other causes as censoring events, the concept of relative survival was developed by comparing the observed survival probability of a group of cancer patients with the survival of a similar cancer -free group (Cho et al., 2011).
In Thailand, the population-based cancer registry at the Faculty of Medicine, Srinagarind hospital, Khon Kaen University, started in January 1988 which had the reports of the five years relative survival of the stomach cancer in Khon Kaen province which were 23.4% in both sexes and 14.9%, 23.3% in male and female respectively (Sriamporn et al., 1995;Vatanasapt et al., 1998). Over the last two decades, there has been no report of the gastric cancer survival in Khon Kaen province, Thailand. The aim of this retrospective cohort study was to provide up-to-date information about the survival of gastric cancer patients in this province.

Cases definitions
This study, we assessed all new cases of stomach cancer recorded in the population -based cancer registry of Khon Kaen province according to the International Classification of Diseases for Oncology (ICD -O 3 rd edition, code C16.0 -16.9) from January 1st, 2000 to December 31 th , 2012 (n=650). Cases reported only in death certificated (n=10) and patients who diagnosed with multiple primaries were not included in our study.

Follow -up
Through record linkage with the continuously updated Death Registry of the Nation Health Security Office (NHSO), Thailand and the medical records, we obtained for each patient the number of years of observation until the date of death or the end of the study (December 31 th , 2014).

Statistical methods
Percentages were used to describe categorical data and mean with standard deviations or medians with ranges were used to describe continuous data. We calculated the observed survival, with the actuarial life table method, and the relative survival, defined as the ratio of observed survival in the group of the stomach cancer patients to the expected survival in the entire Thai population from the estimated generation life tables for Thailand of five -year birth cohort: 1900 -2000 (Prasartkul and Rakchanyaban, 2002), corresponding to the patient group with respect to gender, 5 -year age group, and 5 -year calendar period of observation. Consequently, an adjustment was made for deaths from causes other than stomach cancer. All analyses were conducted using Stata version 10.0 (Stata Corp LP, 2007).

The ethics consideration
The study was approved by the Khon Kaen University Ethics Committee for Human Research. The reference number is HE 581401.

Demographic characteristics of stomach cancer
The distribution of demographic characteristics was summarized in Table 1. The amount of stomach cancer patients with 650, 365 (56.15 %) were male. The median (min : max) age was 61 (22: 92) years. Table 2 shows the frequencies and the distribution of pathological characteristics of cases. The most commonly specified anatomical sites of stomach cancer that identified, histological type, histological grading and stage of diseases were antrum (8.15%), Adenocarcinoma (56.92%), poorly differentiated (28.77%) and stage IV (38.77%) respectively.

Survival rate of stomach cancer
The total follow-up person time was 562.86 personyears. The overall mortality rate and the median survival time were 80.48 per 100 person-years (95% CI: 73.40-88.25) and 5.16 months (95% CI: 4.44-6.00) respectively. The overall of the observed survival rate and the relative survival rate were demonstrated by the Figure 1. Figure 2 shows the observed survival rate and the relative survival rate with 95% confident intervals. Table 3 presented 1, 2, 3, 4, 5 year observed survival rate and relative survival rate, which had 5 year observed survival rate (17.16 (95% CI: 13.54-21.14)) and 5 year relative survival rate   Table  4 revealed the 5 year observed survival rate and 5 year relative survival rate of stomach cancer patients with dependence on gender, age groups, stage of diseases, histological type, site of diseases, histological grading, surgery and treatment groups.   Year from diagnosis Year from diagnosis relative survival Lower 95% CI /Upper 95% CI Overall Relative survival

Discussion
The objective of this study was to provide up-to-date of survival rate of stomach cancer in Khon Kaen province.
Furthermore, this result less than many previous studied which in U.S.A. by Hudahl et al. (5 year RS;1985-199628%), Thailand by Sriamporn et al. (5 year RS;1985-199223.4%), Estonia by Innos et al. (5 year RS;199520%, 2000-200422%, 2005; 22%), European 22 countries by Sant et al. (5 year RS;1985-199623%), Korea by Jung et al. (5 year RS;1993-199743.8%, 199850.3%), Korea by Bae et al. (5 year RS;1985;43.9%), Japan by Tsukuma et al. (5 year RS;1993-199662.1%), Finland by Brenner and Hakulinen (5 year RS;1990-199425.9%, 199528.8%) and China by Xiang et al. (5 year RS;1992-199530%) (Sriamporn et al., 1995;Hundahl et al., 2000;Bae et al., 2002;Sant et al., 2003;Brenner and Hakulinen, 2005;Tsukuma et al., 2006;Jung et al., 2007;Xiang et al., 2010;Innos et al., 2014). In Japan where it had the highly 5 year relative survival rate due to Japan has implemented a stomach cancer screening program especially among high risk group since 1971. Then, they have more efficient to detect gastric cancer patients in early stages, as a result, the 5 year relative survival rate and the prognosis of gastric cancer patients becomes better. Over the last two decades, Sriamporn et al. have been reported the 5 years relative survival rate of stomach cancer higher than this study result; The rates of Sriamporn et al. was higher than this study can be possible because the limitation of followed up at that time, which increased censoring. According to an increasing of censored, the relative survival rates were overestimated.
In conclusion, the observed and relative survival rates were close to each other. The survival among Khon Kaen province population was indicated by the 5 year relative survival rate of gastric cancer. Our findings can comprise as a basic information that beneficial to develop the basic structure of treatment system and helpful in appropriate improved population-based cancer registry, as well.