LYON, France -- The severity of precancerous lesions tracks to specific bacterial strains, investigators here reported, strengthening the association between Helicobacter pylori infection and gastric cancer.
LYON, France, Aug. 28 -- The severity of precancerous lesions tracks to specific bacterial strains, investigators here reported, strengthening the association between Helicobacter pylori infection and gastric cancer.
Only H. pylori strains containing cytotoxin-associated (cagA) gene DNA predicted gastric dysplasia severity, Martyn Plummer, of the International Association for Research on Cancer, and colleagues reported in the Aug. 28 issue of the Journal of the National Cancer Institute.
CagA-positive lesions also were more likely to progress, they said.
The results suggest that the proportion of gastric cancers attributable to H. pylori worldwide may exceed a recent estimate of 63%, they added.
The worldwide incidence of gastric carcinoma exceeds 900,000 cases annually. However, the absolute risk of the cancer is small in comparison to the prevalence of H. pylori infection (estimated at 80-90% in developing countries).
Some evidence suggests that cancer evolves from specific genetic lineages of H. pylori, the authors noted. Potential suspects include cagA, which has been shown to increase the risk of atrophic gastritis and gastric cancer.
Data for the current study came from patients who had biopsies as part of a trial of gastric cancer control in Venezuela. Frozen biopsy specimens were available for 2,145 of the 2,200 study participants. H. pylori genotyping showed that 947 participants were cagA negative and 852 were cagA positive.
CagA-positive individuals had double the risk of chronic gastritis compared with cagA-negative study participants. The odds ratio was 7.35 for intestinal metaplasia and 16.7 for dysplasia for the cagA-positive group versus the cagA-negative participants.
"Thus, the odds ratio increased with severity of disease," the authors noted.
Using individuals with normal mucosa or superficial gastritis as control subjects, the odds ratio for dysplasia was 15.5 (95% confidence interval [CI] = 6.42 to 37.2) in cagA-positive individuals compared with uninfected individuals and 0.90 (95% CI = 0.37 to 2.17) for individuals infected with cagA-negative H. pylori compared with uninfected individuals.
Follow-up gastroscopy results were available for 1,474 participants. Evaluation of the rate of lesion progression showed that uninfected participants had a progression rate of 76.5 per 1,000 person-years, compared with 76.0 for cagA-negative participants, and 80.3 for cagA-positive individuals. Regression rates were 137.8, 149.2, and 126.9 per 1,000 person-years, respectively.
Patients with cagA-positive infection had an odds ratio of 1.21 for progression and 0.81 for regression compared with uninfected individuals, but neither difference was statistically significant.
"In this study, the risk of precancerous lesions conferred by H. pylori was specific to cagA positive strains," the authors stated. "Failure to distinguish between cagA-positive and negative strains would have led to a substantial underestimation of the relative risk."
As an example, a calculation based on H. pylori-positive versus negative patients (without genotyping) resulted in an odds ratio of 4.2 for dysplasia compared with 15.5 when cagA-positive patients were compared with uninfected patients.
"Our study adds to the emerging body of evidence that the strength of the association between H. pylori and gastric carcinoma has been underestimated," the authors concluded.