This case control study demonstrates that hereditary factors, especially familial history of cancer and possession of blood group O, are associated with the development of gastric cancer under the age of fifty. To our knowledge, this may be the first study showing a correlation between blood group O and the development of gastric cancer in a specific category of patients.
Risk factors for gastric cancers have been explored in a number of previous studies, including genetic factors such as blood group. Haenszel et al. suggested an association between gastric cancer and blood type A, supporting the view that genetic factors have a role in the development of gastric cancer . Our study emphasizes the role of genetic factors in one subcategory of patients, those who develop gastric cancer under the age of fifty. Blood group A is more strongly associated with the diffuse histopathological type of gastric cancer than the intestinal type [21, 22]. In our study, most of the patients had a diffuse rather than intestinal type, so we could not test this association. A larger sample size would be needed. On the other hand, there might be a higher prevalence of Helicobacter pylori in our community, causing a higher incidence of the diffuse type of gastric cancer. However, there were no significant differences in histological type of gastric cancer between the case and control groups. In a study by Su et al. in 2001, a total of 6685 patients with esophageal carcinoma and 2955 patients with cardiac cancer in the Chaoshan district were retrospectively assessed for their association with ABO blood groups. Su et al. showed that the distribution of ABO blood groups in patients with esophageal carcinoma or cardiac cancer was similar to that in the normal local population, but there was an association between blood group B and the development of cancer of cardia in males .
In our study, approximately 54% of the case group had a familial history of cancer compared to 11% of the control group. This seems compatible with the findings of a population-based case-control study of stomach cancer in Warsaw, Poland. Here, the investigators interviewed 464 cases and 480 controls to evaluate the role of family history and other risk factors. A greater than threefold increase in risk was associated with a history of gastric cancer in a first degree relative (OR = 3.5), but no excess risk was seen for other forms of cancer. The risk associated with familial occurrence was not significantly modified by gender, age or ABO blood type, and did not vary with Lauren histological classification24. Despite the relatively large sample size in the Polish study, younger patients were not evaluated as a separate category. This may explain the difference between their results and ours. Furthermore, they defined "positive family history" as having a first-degree relative with gastric cancer. In contrast, we considered all types of malignancy in first and second-degree relatives; though the familial incidence of other (non-GI) malignancies in our study may reflect their higher frequency in the community rather than any genetic risk factor. The Polish study did not confirm previous results on the correlation between blood group A and gastric cancer. Moreover, another study by Parsonnet et al on 90 cases and 89 controls showed no association between ABO blood group and malignancy . In a multicentric study in Italy, 1016 patients with gastric cancer and 1623 population controls were interviewed to determine family histories of gastric, esophageal and colorectal cancer. A significant association was found with history of gastric cancer in a sibling or parent (odds ratios 2.6 and 1.7, respectively). Among the adult siblings of controls and cases, the prevalences of gastric cancer reported at interview were 1 and 2.7%, respectively. A further increase was noted in families with at least one affected parent (1.4 and 5.7%). The risk of gastric cancer associated with a positive family history was greater (increased about 2-fold) among residents of low-risk areas. Among the cases, there was no relationship between family history of gastric cancer and blood group A or histological type according to the Lauren classification . In our study, there was no significant relationship between the histological type of the cancer and positive family history or blood group. However, this does not prove that the two variables do not correlate; an association might become apparent with a larger study. Mecklin et al studied the clinical and histopathological characteristics of gastric carcinoma in young patients (under 40 years old) in Finland in 1988. In 94% of the young patients, the carcinoma was of the diffuse type. They showed a poor prognosis, an equal sex ratio, and a strong association with blood group A in their study group. They also found a highly significant over-representation of gastric cancer in the parents of the index cases (p < 0.001) . The difference between Mecklin's study and ours in the blood groups identified as risk factors may reflect ethnic differences; both studies confirm a significant correlation between a specific blood group and the development of gastric cancer. Future studies may use linkage analysis to detect genetic abnormalities in chromosomal regions that are located near the genes encoding the ABO antigens.
Matching the geographical origins of the cases and controls could have improved the power of our study by excluding ethnic factors from the study population. However, this is very difficult to achieve in such studies because there is a high rate of combinations between races in the country. In addition, the sample size was too small for such effects to be excluded. However, there were no significance differences between the two groups in respect of the origins of the subjects.
In conclusion, our results show that familial history of cancer, and hereditary factors including blood group, have a role in the development of gastric cancer in young patients. The role of environmental factors may be more important in older patients and can be considered in future studies.