ABO blood group and gastric cancer: rekindling an old fire?

To the Editor: Numerous reports have pointed to an increased prevalence of blood type A in patients with gastric cancer. Aird et al (1) were the first to study ABO blood groups in a large number of patients with gastric cancer, noting an excess of group A, and a deficiency of group O, in comparison with control patients. Subsequently, at least 77 controlled series have been published; in 55 of these, cancer patients had an excess of group A, in 14 a slight predominance of group A, in only 2 a deficiency of group A. (2) In a recent study aiming at the study of the association of gastroduodenal disease phenotype with ABO blood group and Helicobacter pylori virulence genes, we noted a higher prevalence of blood group A in patients with gastric malignancy than in a control group of healthy blood donors [47.6% versus 25%, P = 0.037; RR = 2.73 (1.04-7.16; 95% CI)]. (3) Based on these findings, we conducted a retrospective study to investigate this putative association between ABO blood group and gastric cancer or lymphoma in Lebanon.

This was a single center retrospective study where data from medical records of 174 consecutive patients with histologically-confirmed gastric carcinoma (152 patients) or lymphoma (22 patients) were collected. All patients were seen at the American University of Beirut Medical Center (AUBMC) from January 1995 to December 2005.

The study results are summarized in the Table. We found a statistically significant higher prevalence of blood group A in patients with gastric cancer compared with that of 2 control groups. Interestingly, a trend toward statistical significance was noted in the lymphoma group. The latter finding, albeit limited by the small number of patients with gastric lymphoma in our patient cohort, has not been described to date.

Manuila (4) pointed out the statistical and technical pitfalls inherent in the study of blood groups and disease. The best control groups are taken from large population surveys and from large series of cases. Besides, the fact that hospital patients conform poorly to the requirements for a suitable group may be understood from the possible relationship between their own blood group distribution and their disorders. In our study, we chose 2 control groups. The first control group consisted of patients undergoing CABG during a 10-year study period (1995-2005). In that group, blind selection of one in every 10 patients of the total sample (n = 6430) resulted in a representative random sample of the general population. The second control group consisted of 18,972 healthy blood donors seen at the AUBMC blood bank during the same period of study (1995-2005). Interestingly, the prevalence of blood type group O is noted to be higher in the blood donors control group than in the CABG control group. This can be explained by the fact that blood group O donors are universal donors; hence, they are preferably chosen for volunteer blood donation in case of need.

Our data is in line with other results and confirms a statistically significant positive correlation between blood type A and gastric cancer, as well as a trend toward statistical significance in the gastric lymphoma group. However, the exact role of this putative risk factor (blood group A) in relationship to other environmental (eg, H pylori, smoking) and genetic risk factors remains unclear.

Ihab I. El Hajj, MD, MPH

Jana G. Hashash, MD

Division of Gastroenterology 

Department of Internal Medicine

Elizabeth M. Kfoury Baz, MD

Department of Pathology and Laboratory Medicine

Heitham Abdul-Baki, MD

Ala I. Sharara, MD

Division of Gastroenterology

Department of Internal Medicine

American University of Beirut Medical Center

Beirut, Lebanon

References

1. Aird I, Bentall HH, Roberts JA. A relationship between cancer of stomach and the ABO blood groups. Br Med J 1953;1:799-801.

2. McConnell RB. The Genetics of Gastrointestinal Disorders. London, Oxford University Press. 1966, pp 56-60.

3. Sharara AI, Abdul-Baki H, ElHajj I, et al. Association of gastroduodenal disease phenotype with ABO blood group and Helicobacter pylori virulence-specific serotypes. Dig Liv Dis 2006;38:829-833.

4. Manuila A. Blood groups and disease-hard facts and delusions. J Am Med Assoc 1958;167:2047-2053.

Table. ABO blood groups in patients with gastric carcinoma and gastric
lymphoma vs. ABO blood groups in CABG patients and blood donors

                                  A                 B

Gastric adenocarcinoma (n = 152)    76 (50.0%) (a)    13 (8.6%)
Gastric lymphoma (n = 22)           12 (54.5%) (b)     3 (13.6%)
Gastric malignancy (lymphoma +      88 (50.6%) (c)    16 (9.2%)
  adenocarcinoma) (n = 174)
CABG patients (n = 643)            247 (38.4%)       109 (17.0%)
Blood donors (n = 18972)          6762 (35.6%)      2555 (13.5%)

                                  AB          O

Gastric adenocarcinoma (n = 152)    7 (4.6%)    56 (36.8%)
Gastric lymphoma (n = 22)           1 (4.5%)     6 (27.3%)
Gastric malignancy (lymphoma +      8 (4.6%)    62 (35.6%)
  adenocarcinoma) (n = 174)
CABG patients (n = 643)            48 (7.5%)   239 (37.2%)
Blood donors (n = 18972)          737 (3.9%)  8918 (47.0%)

(a) P [less than or equal to] 0.01 when compared to the ABO prevalence
in CABG patients and healthy blood donors.
(b) P = 0.071.
(c) P [less than or equal to] 0.01 when compared to the ABO prevalence
in CABG patients and healthy blood donors. CABG, coronary artery bypass
grafting.