Gastric MALT Lymphoma: Cure the Infection, Cure the Tumor?


The revised European-American lymphoma classification of the International Lymphoma Working Group formally recognizes the existence of mucosa-associated lymphoid tissue (MALT) lymphomas. One cannot take the name too seriously; MALT lymphomas may occur in the thyroid gland, which does not seem to have any mucosa for the lymphoid tissue to associate with. In any case, most MALT lymphomas arise in the gastrointestinal tract, and more than 90% of all gastric MALT tumors are associated with chronic Helicobacter pylori infection. Clinical studies have shown remissions of the gastric MALT lymphomas after bacteriological cure of H. pylori infection, but long-term follow-up of these patients is lacking. Neubauer and colleagues report on clinical follow-up of 50 patients and add information from PCR determination of B-cell monoclonality to the traditional criteria for follow-up.

Neubauer et al identified 50 patients by recruitment of candidates from a central referral pathologist. When tissue containing a gastric MALT tumor was received, the referring physician was contacted to enroll patients in the study. Patients considered to be stage I-E after clinical examination, CT scan of the chest and abdomen, endoscopic ultrasound, and bone marrow biopsy were enrolled.

After workup, patients were treated with two-week courses of amoxicillin 750 mg tid and omeprazole 40 mg tid. Endoscopy was repeated in four weeks and then monthly until a complete remission was achieved. After complete remission, endoscopy was done twice a year and the other staging procedures were done yearly. Non-responding patients were given alternative treatments at week 22.

Conventional criteria for the diagnosis of MALT lymphoma were employed. Complete histological remission was defined as absence of lymphoma cells and replacement by an empty lamina propria with small basal clusters of lymphocytes and scattered plasma cells. Partial regression was declared when the histologic examination noted only partial depletion of atypical lymphoid cells from the lamina propria or focal lymphoepithelial destruction. Monoclonal B cells were detected by PCR amplification of rearranged immunoglobulin heavy-chain variable region genes.

All 50 patients were cured of H. pylori by the antibiotic regimen. Two of the patients required a second antimicrobial treatment with metronidazole (800 mg/day), clarithromycin (500 mg/day), and omeprazole (40 mg/day) for seven days. One patient had a relapse of a MALT tumor associated with reinfection, which cleared with the triple regimen resulting in a second remission. Forty patients (80%) went into a complete remission (CR), four had a partial remission, and six did not respond.

The 50 patients were followed for a median of 24 months. Four of the 40 patients who achieved a complete remission relapsed at six months, seven months, 14 months, and 16 months after entering remission. A fifth developed a high-grade lymphoma in the nasal cavity four months after treatment. DNA sequencing of the CDR3 region of the immunoglobulin heavy chains showed that the high-grade lymphoma was not related to the original low-grade gastric MALT lymphoma. Three of the relapsing patients were H. pylori negative. No relapses occurred after 500 days of remission. Two patients died of cardiovascular disease.

Three of the four patients with a partial remission underwent a gastrectomy resulting in relapse-free survivals of 10+ months, 16+ months and 38+ months. The fourth patient died of a stroke in partial remission. Four of the six patients who did not respond to antibiotic therapy underwent gastric surgery. These four showed high-grade lymphoma in deeper mucosal areas. Three of the lymphomas were high-grade, B-cell lymphomas. One was high-grade T-cell lymphoma.

Surgical exploration showed regional node involvement in three patients. Two of the operated patients died at 3.0 and 5.5 months after surgery, and the other is in a continuous remission. One of the unoperated patients entered a durable complete remission after five cycles of CHOP chemotherapy. The sixth patient is stable with disease at 31 months.

Molecular analysis showed monoclonal bands in 38 of 50 patients at diagnosis. Thirty-two of these patients entered a complete remission. Thirty-one of the 32 patients with monoclonal bands were available for PCR, and 22 (71%) continuously tested positive despite being histologically normal. Some patients have been in remission, though PCR positive, as long as 23 months.

Cure of H. pylori is associated with an 80% CR rate in gastric MALT lymphomas in several series. The current series reports follow-up on these patients. So far, four patients have had gastric relapses, and one developed a second lymphoma in the nasal cavity that was histologically and molecularly distinct from the primary gastric lymphoma. Three of the four local relapses occurred in the absence of bacteriological relapse. The quality of the response to the eradication of H. pylori may indicate the clonal progression of the B-cell tumor. Among the four patients who were referred for surgery after failure of the lymphoma to respond to H. pylori eradication, all four had high-grade lymphoma in deeper mucosal areas.

Regression of the low-grade MALT lymphoma took much more time than it takes to see regression with chemotherapy or radiation. Several patients needed more than six months to achieve a CR, and the median time to CR was 5.5 months. Patients showed a continuous, gradual response to therapy. The clinical meaning of persistent PCR positivity is not clear. However, the presence of such cells does not appear to predict for early relapse of the lymphoma. (Neubauer A, et al. J Natl Cancer Inst 1997;89:1350-1355.)


MALT lymphomas are rare. They account for about 2-4% of lymphomas worldwide and gastric MALT lymphomas are a common subtype. But, such lymphomas may involve the lung, breast, lacrimal glands, thyroid, and other tissues, as well. The lesson we have learned from gastric MALT lymphoma is that chronic antigenic stimulation may well play an important pathogenetic role in the development of other forms of MALT lymphoma. Unfortunately, the primary antigenic stimulus may be harder to define and eradicate in the other tissues that may be affected by MALT lymphoma.

Early estimates were that about 50% of patients with gastric MALT lymphoma responded to eradication of gastric H. pylori infection. It now appears that 80% complete response rates are achievable. There are probably at least two factors contributing to this difference. First, patient selection may play a role. Although the majority of patients have disease localized to the stomach, some patients have disease that has spread to gastric nodes or even more widespread sites. It is difficult to imagine that tumor cells that have developed the capacity to spread have also retained dependence on antigenic stimulation for their proliferation. Thus, restricting study to patients with stage I disease is likely to lead to higher response rates.

A second factor that may have contributed to the lower response rates associated with H. pylori eradication in the early reports was the timing after treatment at which response was assessed. If one assesses response one month after the completion of the antibiotic course, response rates will be underestimated. It takes time for the lymphoma to regress when the antigenic stimulus is removed. We have had calls from a number of patients who wondered what to do next when a course of antibiotics recently completed did not seem to produce the desired effects. The answer is to wait. Responses may take six months to be manifested. There are two key things to be done to monitor this treatment approach. First, document that the infection has been eradicated. If the organism persists, treat it again. Second, perform gastric ultrasound endoscopy monthly in an effort to determine whether the tumor is growing. If the gastric wall thickness increases in an objectively documented fashion, move to another treatment approach. However, if the tumor is stable or subtly decreasing, keep waiting.

If the tumor grows or fails to regress in the six months following H. pylori eradication, a number of treatment options are available. The tumor is responsive to systemic combination chemotherapy regimens. However, we favor approaching a local problem with locally active methods, if possible. An approach that does not yet have prospective data to support it, but that seems rational is to locally resect the tumor with a segmental gastric resection followed by involved-field radiation therapy. The major concern about surgical intervention is that the local spread of the disease through the submucosal may make it difficult to keep the surgical procedure limited. Surgeons are justifiably anxious about leaving tumor behind and when they cut across tumor-involved stomach wall, the impulse is to take the entire stomach out. Total gastrectomy is clearly a treatment that is worse than the disease in this case. The advantage of a surgical procedure in this instance is that it enables the physician to detect the histologic progression to diffuse large cell lymphoma that may occur in a small subset of patients and permits accurate surgical staging. However, the surgeon should be implored not to remove more than one-half of the stomach, regardless of what he/she finds.

If the surgical staging reveals histologic progression or nodal spread, systemic combination chemotherapy may be indicated. If the disease has been entirely removed by the surgical procedure, radiation therapy to the stomach may be effective at preventing disease recurrence. However, it is essential to add that this recommended treatment approach has not been validated and has no data to support it. It represents our clinical judgment, and several elements in this recommendation are arguable.

Several important lessons have been learned from gastric MALT lymphoma. It validates the notion that chronic antigenic stimulation may lead to lymphoma. This further supports models of carcinogenesis that identify stages of transformation that may be reversible up to a point. It adds impetus to the efforts to develop effective methods of treating viruses such as human papilloma virus and human herpes virus-8 in the hope that their eradication might be a successful way of treating some early cervical dysplasia or cervical cancer and Kaposi’s sarcoma, respectively. Another lesson is that the persistence of abnormal cells in an indolent lymphoma is not tantamount to recurrence. The abnormal B cells that are detectable in patients in remission may well be incapable of expanding without the antigenic stimulus of the organism. Finally, the high frequency of cytogenetic abnormalities found in gastric MALT lymphoma (often involving chromosomes 3 and 18) suggest that even cells with genetic abnormalities may be dependent on exogenous growth signals to survive. Efforts to identify growth factor dependence have gained support from the gastric MALT lymphoma story.

H. pylori is prevalent in the population and has clearly been shown to be involved in the pathogenesis of peptic ulcer disease, gastric carcinoma, gastric MALT lymphoma, gastric large cell lymphoma, and gastric carcinoid tumor. Indeed, one wonders whether there is any significant pathologic condition involving the stomach that is not related to this organism. Should there be a concerted effort to eradicate H. pylori from the human population? This seems like a no-brainer; the gain might not be as great as would be seen with the elimination of cigarette smoking, but eradication of H. pylori may improve health in many millions of people.