Eradication of Helicobacter pylori Infection in Primary Low-Grade Gastric Lymphoma of Mucosa-Associated Lymphoid Tissue

Methods

Twenty-six patients (14 women, 12 men; median age, 60 years [range, 21 to 76 years]) with histologic diagnoses of localized (stage IE), primary, B-cell, low-grade gastric MALT lymphoma and H. pylori infection (diagnosed using endoscopic biopsy specimens) were entered into our study. Almost all patients had a prolonged history (as long as 4 years) of gastric symptoms before diagnosis of H. pylori infection, and most of them had already been given antacid medication, motility stimulant agents, or both. Before patients received treatment for H. pylori infection, the following staging procedures were done: physical examination, routine laboratory tests, chest radiographs, abdominal ultrasound or computed tomographic scans, endoscopic gastric biopsy specimens, and bone marrow biopsy specimens.

Patients were treated for H. pylori infection for 2 weeks with amoxicillin (500 mg three times daily) plus metronidazole (400 mg three times daily) and colloidal bismuth (120 mg four times daily; 13 patients) or omeprazole (20 mg twice daily; 13 patients). Twenty patients were treated immediately after diagnosis of H. pylori infection, and 6 were treated after an observation period of several months (median, 5 months; range, 3 to 36 months) without endoscopic and histologic changes of the MALT lymphoma.

One patient had a brief response to first-line combination chemotherapy with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP regimen), but the patient later had a local relapse. This patient still had histologic characteristics of low-grade gastric MALT lymphoma when treatment for H. pylori infection was given.

To reevaluate the condition of the patients after treatment for H. pylori infection, biopsy specimens were obtained from any abnormal area in the stomach and randomly obtained from the rest of the stomach. These multiple (range, 8 to 20 specimens), random biopsy specimens of the gastric mucosa were obtained within 6 months of treatment (3 months for 19 patients, 4 months for 4 patients, and 6 months for the remaining 3 patients); patients were reevaluated every 3 to 6 months thereafter. The patients with persistent H. pylori infection had additional treatment for 2 weeks with a different regimen: omeprazole (20 mg twice daily), metronidazole (400 mg three times daily), and azithromycin (500 mg daily for 3 consecutive days every week).

Restaging with chest radiographs and abdominal ultrasound was done again after 1 year. In 9 patients, a bone marrow biopsy specimen was obtained again after 6 months. Histologic responses were graded according to the histologic scoring system recently proposed for the diagnosis of gastric MALT lymphoma [2]. We considered a post-treatment score of 2 or less to be evidence of lymphoma regression. Binomial exact 95% CIs were calculated for outcome percentages.

Results

Complete eradication of H. pylori was initially achieved in 21 patients (81%; 95% CI, 61% to 93%); in the remaining 4 patients, second-line antibiotic treatment was needed to eradicate the microorganism. The overall eradication rate was therefore 96% (25 of 26 patients; 95% CI, 80% to 99%). After treatment for H. pylori infection, the median follow-up was 12 months (range, 3 to 36 months). One patient, treated after an observation period of 36 months without clinical or histologic change, showed persistent gastric infection and unchanged lymphoma features on biopsy specimens obtained 3 months after treatment. The patient received additional antibiotic treatment but has not been reevaluated with endoscopy.

All but one of the patients were symptomatic at presentation; the main symptoms were pain (60% of patients), dyspepsia (30% of patients), and nausea and vomiting (7% of patients). One patient presented with gastrointestinal hemorrhage. Symptoms disappeared in 77% of patients or markedly diminished after antibiotic treatment. At the endoscopic evaluation done before treatment, 11 patients had gastric ulcers, 9 had endoscopic evidence of gastritis, and the remaining 6 had abnormal, congestive (hyperemic) gastric mucosa. In all patients, these endoscopic features improved after eradication of H. pylori infection and all of the ulcers healed, but only one patient had a complete recovery of the gastric mucosa.

Histologic regression of the MALT lymphoma (Table 1) was observed in 15 patients (60%; CI, 39% to 79%) after H. pylori eradication, including the 1 patient who relapsed after chemotherapy. Five patients had complete disappearance of any histologic evidence of lymphoma (histologic score, 0 to 1), and 10 patients had residual lymphoid follicles but no longer had lymphoepithelial lesions (histologic score, 2). Three patients showed persistent suspicious, probably reactive, lymphoid infiltrate in the lamina propria (histologic score, 3) and were considered to have not responded to treatment. The other patients showed no change or only partial improvement in the lymphoma histologic pattern in repeat gastric biopsy specimens.

Our study suggests that regression of the lymphoma may require a prolonged period. Only 8 patients showed histologic regression on the first biopsy specimen after antibiotic treatment, but when endoscopic specimens were obtained again 3 or more months later, regression was clearly evident in 7 additional patients. At a median follow-up of 12 months, all patients were alive and 14 of 15 responders, including the 1 patient who relapsed after chemotherapy, were free of lymphoma. A bone marrow specimen from 1 patient showed evidence of relapse shortly thereafter, with a response duration of only 4 months; however, no evidence of MALT lymphoma was found in repeat gastric biopsy specimens (histologic score, 2).

Discussion

A strong causal relation between H. pylori infection and gastritis, duodenal ulcer, or both has been shown; moreover, increasing epidemiologic and histopathologic evidence indicates a link between H. pylori infection and gastric tumors, both carcinomas [7] and lymphomas [3, 8]. Further, the ability of H. pylori to stimulate cellular proliferation in low-grade gastric MALT lymphoma has been reported [9], and eradication of H. pylori may inhibit the growth of the lymphoma [2]. The rarity of extra-abdominal spread low-grade gastric MALT lymphoma may also be partially explained by the role of H. pylori in the pathogenetic process [9]. Our data confirm, in a larger series of patients, the recent observations that eradication of H. pylori can lead to a regression of the lymphoma [2, 10] and further support the concept of a causal correlation between H. pylori infection and gastric lymphoma.

The case of low-grade gastric MALT lymphomas is not the only one in which a bacterial infection has been implicated in the pathogenesis of a lymphoma. Antibacterial therapy has already been reported to be useful in the treatment of α-chain disease (a small-intestinal lymphoma) [11], providing indirect evidence that bacteria may have a pathogenetic role in lymphomas.

Only careful, prolonged follow-up will ascertain whether treatment for H. pylori can definitely cure gastric MALT lymphoma. Currently, this represents one of the many questions in the treatment of gastric lymphoma (others include those about the necessity for surgery and role of chemotherapy) [12]. Our data show that for low-grade MALT gastric lymphoma, an antibiotic treatment designed to eradicate H. pylori appears to be mandatory before further therapeutic options are considered. An international randomized trial will soon investigate whether it is beneficial to add cytotoxic chemotherapy to such an antibiotic treatment.