From January 2000 to September 2012, 50 patients with the diagnosis of gastric MALT lymphoma were treated with RT. Two patients lost to follow-up after RT were excluded from the evaluation. The remaining 48 patients were enrolled in this study.

All patients underwent diagnostic endoscopy and had pathologic confirmation of their diagnosis at local clinics or our institution. Although there are some differences between the cases, the main histologic features are a diffuse infiltration of lymphoid cells with lymphoepithelial lesion. The lymphoid cells showed centrocyte-like morphology, pale cytoplasm with small to medium sized, slightly irregularly shaped nuclei containing dispersed chromatin. Scattered large cells resembling centroblasts are present. Occasionally, plasma cells are also present. The lymphoid cells are immunoreactive for CD20+, CD5-, CD10-, and cyclin D1-, histologically consistent with the low-grade MALT lymphoma. Disease stage was determined based on the Ann Arbor Classification of Extranodal Lymphoma (modified by Musshoff) [6]. Work-up studies comprised patient medical history including duration and presence of local or systemic symptoms, physical examination, laboratory tests such as complete blood counts, peripheral blood smear, lactate dehydrogenase, β2-microglobulin levels, and evaluation of renal and liver function. Other staging procedures included bone marrow biopsy, chest radiographs, abdominal and pelvic computed tomography scan, and/or gastric endoscopic ultrasonography (EUS). H. pylori infection was determined by histological examination, rapid urease test, urea breath test, and/or serum anti-H. pylori immunoglobulin G antibody.

Anti-H.pylori antibiotic therapy using lansoprazole, clarithromycin, and amoxicillin was administered to 30 of 31 patients with H. pylori infection and to three of 13 patients without H. pylori infection for 7 or 14 days. Four patients without information regarding H. pylori infection status did not receive H. pylori eradication. Thirty-four of the 48 patients received RT alone. Ten patients received chemotherapy before RT with one of three regimens, including cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP), rituximab-CHOP (R-CHOP), or fludarabine, mitoxantrone, and dexamethasone (FND). Three patients underwent surgery followed by chemotherapy and RT. One patient received surgery followed by RT. All patients received RT of median total dose of 30.6 Gy (range, 30.6 to 45.0 Gy), with a fraction size of 1.8 Gy for 3.5 to 5 weeks. Patients were treated primarily with opposed anterior and posterior two fields, or occasionally three or four fields targeting the stomach and perigastric lymph nodes with additional margins using two-dimensional (Fig. 1A) or three-dimensional conformal RT (Fig. 1B), with or without respiratory gating. All patients were treated in the supine position with an empty stomach by a 6- or 10-MV X-ray of linear accelerator on an outpatient basis.

Response assessment was performed every three months by gastroduodenoscopy, including gathering biopsy samples, until one year after completion of RT, and every six months thereafter. Pathologic complete remission (pCR) was defined as the total disappearance of clinical evidence for lymphoma and an absence of histologic evidence for lymphoma on biopsy specimens. In cases with complete remission, endoscopic examinations and biopsies were performed at regular intervals. Blood work, including a complete blood cell count and screening profile, was performed biannually for the first two years, and thereafter annually. Chest X-ray and computed tomography (CT) scans of the abdomen and pelvis were also performed annually. The overall survival (OS), cause-specific survival (CSS), and event-free survival (EFS) rates were analyzed by Kaplan-Meier methods. OS was measured from the date of start of RT to death from any cause, and CSS was measured from the date of start of RT to the date of death due to gastric MALT lymphoma. EFS was measured from the date of start of RT to the date of death from any cause, diagnosis of relapse of MALT lymphoma or transformation to a DLBCL. Toxicity was evaluated according to Common Terminology Criteria for Adverse Events (CTCAE v.4.0).

Forty-one patients were diagnosed with low-grade MALT lymphoma, and the remaining seven patients had low-grade MALT lymphoma mixed with high-grade transformation components (Table 1). Forty of these patients (83.3%) had an Eastern Cooperative Oncology Group (ECOG) performance score of 0. The median age of the patients was 53 years with a range of 45 to 73. Thirty-two patients (66.7%) had lesions located mainly in the mid to the lower stomach. Two patients had additional involvement of the duodenum (stage IIE₂). Twenty-five patients (61.0%) had ulcerative or infiltrative endoscopic type tumors. Thirteen patients (27.1%) also exhibited intestinal metaplasia. H. pylori infection was noted in 31 patients (64.6%).

Among the 41 patients with low-grade MALT lymphoma (Fig. 2A), 31 patients underwent RT alone. Twenty-three of these patients had H. pylori infection and underwent H. pylori eradication therapy. Of the 23 patients receiving eradication, 14 patients showed persistent disease after eradication, and the remaining nine patients underwent RT immediately after eradication without response evaluation. Another seven out of the 41 patients with low-grade MALT lymphoma underwent chemotherapy prior to RT. Of these seven, six patients showed persistent disease after H. pylori eradication, while the other patient did not undergo eradication. Of the seven patients who received chemotherapy prior to RT, five showed complete remission after chemotherapy, and two showed partial remission. Another two out of the 41 patients, who did not initially undergo H. pylori eradication, underwent a wedge resection plus chemotherapy prior to RT. One remaining patient did not undergo H. pylori eradication, but had a total gastrectomy prior to RT. Median dose of RT in these 41 low grade patients was 30.6 Gy (range, 30.6 to 43.2 Gy).

Among the seven mixed grade patients (Fig. 2B), three patients had RT alone, two immediately following H. pylori eradication without response evaluation and one without H. pylori eradication. Three other mixed grade patients had chemotherapy plus RT. Two of these patients showed persistent disease after H. pylori eradication, while the other patient did not undergo eradication. Of the three patients who received chemotherapy prior to RT, two showed complete remission after chemotherapy. The other patient, who had also shown persistent tumor after initial eradication, showed partial remission. The remaining one patient with mixed grade MALT lymphoma, who also did not undergo H. pylori eradication, underwent a total gastrectomy plus chemotherapy followed by RT. The median dose of RT in these seven mixed grade patients was 37.8 Gy (range, 30.6 to 45.0 Gy)

The follow-up period for all patients ranged from 6 to 158 months, with a median of 48 months. The 5-year OS rate was 90.3%, CSS was 100%, and EFS was 85.2%. All thirty-four patients, 31 of low-grade and 3 of mixed-grade, treated with RT alone achieved pCR. In 10 patients who received chemotherapy prior to RT, five of seven patients with low grade and two of three patients with mixed grade achieved complete remission to chemotherapy and received consolidative involved field RT. The remaining three of 10 patients who exhibited a partial response received salvage RT, two of whom achieved a complete response. One showed a persistent tumor combined with newly developed a gastric DLBCL. The other four patients who received surgery initially, with or without chemotherapy, received consolidative RT.

Among all patients, three patients died without evidence of disease, one from renal insufficiency at 56 months, one from intracranial hemorrhage at 12 months, and one from an unknown cause at 16 months post-RT. During the follow-up period, three patients developed second malignancies (Table 2). One patient with initially low-grade lymphoma showed persistent tumor combined with DLBCL three months after chemotherapy and RT. This patient initially underwent H. pylori eradication, but had persistent disease. Following eradication, she received three cycles of FND chemotherapy, but responded only partially. The patient then received salvage RT, but showed persistent tumor mixed with DLBCL. Thereafter, the patient was salvaged completely with four cycles of R-CHOP augmented with four cycles of Rituximab and is alive without disease at 31 months after RT. The second patient had a DLBCL 38 months after completion of RT for initial low-grade MALT lymphoma. She received eight cycles of R-CHOP for newly developed DLBCL and has no evidence of disease at 64 months after RT. The third patient was diagnosed with gastric adenocarcinoma and bronchus-associated lymphoid tissue (BALT) lymphoma at 44 and 51 months after completion of RT for initial low-grade MALT lymphoma. This patient had endoscopic mucosal resection for gastric adenocarcinoma (pT1 lesion), but no further treatment for BALT and expired due to renal failure at 57 months after RT. Follow-up endoscopy study for all patients after RT revealed that 13 patients had H. pylori reinfection, of whom 11 received primary or secondary eradication. Five out of these 13 patients did not have a history of H. pylori infection originally.

Acute complications were grade 2 (n=31) or grade 1 (n=1) with complaints of nausea, vomiting, dyspepsia, anorexia, diarrhea, or abdominal pain. No grade 3 or higher complications developed either as acute or late complications. Late grade 2 complications developed in three patients with gastric erosion, erythematous gastritis, and gastric ulcer. One patient, who initially had gastroptosis, developed left kidney atrophy. Because of gastroptosis, an extremely large RT field size was used to cover the entire stomach such that the entire left kidney was included within the RT field (Fig. 1C). This patient had been diagnosed with early gastric adenocarcinoma treated with endoscopic mucosal resection and BALT lymphoma (no treatment) after RT completion as described above. However, she died due to renal insufficiency with Waldenström's macroglobulinemia and metabolic acidosis at 56 months post-RT. Her kidney disease was not due to prior RT because ultrasonography of both kidneys at 52 months after RT showed mild atrophic change of the left kidney, but normal echogenicity with an intact right kidney (Fig. 1D).