P&S Medical Review: Nov 1993, Vol.1, No.1
Treatment of Gastric Lymphoma

JASON CHRISTIE, M.D.
Columbia University College of Physicians and Surgeons,
New York, NY.

Abstract

Primary gastric lymphomas are uncommon tumors, constituting less than 2% of all primary stomach malignancies, although the incidence appears to be increasing. The stomach, however, is the most frequent site of extra-nodal non-Hodgkin's lymphoma. Diagnosis is difficult and may be complicated by several factors: vague symptoms on clinical presentation, problems in pathological diagnosis, and difficulties in staging with radioimaging and endoscopic biopsy. There is considerable controversy concerning the treatment of gastric lymphoma. Traditionally, surgery has been the first line of therapy; however, radiotherapy and chemotherapy have gained support, particularly for earlier stage lesions. This review describes the pathology and behavior of gastric lymphoma with particular reference to staging and therapeutic interventions.

Epidemiology and Pathology

Gastric lymphoma is an uncommon malignancy, accounting for less than 2% of primary gastric cancers. [1] However, the gastrointestinal tract is the most commonly involved site of extranodal non-Hodgkin's lymphoma, with the stomach most frequently involved. [2] In one series, almost 75% of primary gastrointestinal lymphomas were of gastric origin. [3] Gastric lymphomas are more prevalent in patients over the age of 50, and men are affected two to three times more frequently than women. [4],[5]

Lymphomas of the stomach are thought to arise in the mucosa and submucosa from so-called mucosa-associated lymphoid tissue (MALT). [6] Most gastric lymphomas are of the B-cell type and frequently display a diffuse growth pattern involving the mucosa, submucosa and muscularis. [2] Gastric lymphoma is virtually always of the non-Hodgkin's type. The Rappaport, Lukes-Collins, Working Formulation, and Kiel systems have all been used for classification. Gastric lymphomas are most commonly of the diffuse histiocytic (Rappaport) or diffuse large cell type (Working Formulation). [7],[8]

Microscopically, low grade gastric lymphomas may be difficult to distinguish from pseudolymphoma, a term used to describe the lymphocytic infiltration of the gastric mucosa which may occur with chronic gastritis and peptic ulceration. Solitary lymphocytic lesions (focal lymphoid hyperplasia) of the gastric mucosa may also be termed pseudolymphoma. Microscopic features suggestive of malignant lymphoma include: prominent lymphoepithelial lesions (lymphoid infiltration of glands without destruction), Dutcher bodies, and moderate cytologic atypia. [9]

For definitive diagnosis of any lymphoid lesion of the stomach, immunohistochemical marker studies should be employed. Demonstration of a monoclonal population of lymphocytes confirms the diagnosis of malignant lymphoma. However, it should be noted that gastric pseudolymphomas can become malignant, and the two conditions can coexist. [10]

Clinical Presentation and Diagnosis

Gastric lymphoma and adenocarcinoma may be difficult to distinguish, both clinically and pathologically. Both of these malignancies may diffusely infiltrate the gastric wall or cause mucosal ulceration. The clinical presentation and the radiographic and endoscopic appearance of gastric lymphoma may be similar to that of peptic ulcer disease, gastritis, or other gastric neoplasms. The most common presenting symptoms are abdominal pain, nausea and vomiting, anorexia, weight loss, and bleeding. More advanced lesions may present with weakness, hemorrhage, pyloric stenosis, or signs of perforation. [1],[5],[7] Early symptoms are vague, and many patients go long periods of time before the diagnosis is established and the proper treatment is initiated. In 1982 Dworkin et al [5] reported a median duration of symptoms of 10.3 months prior to treatment.

Physical examination is often unremarkable. Abdominal tenderness occurs in 35% of patients, an abdominal mass is palpated in 20-30% of patients, and hepatomegaly exists in 14% of patients. [2],[11] Diagnosis can frequently be suspected based on the results of contrast enhanced radiographic studies. However, the sensitivity of such studies is low, and the findings for gastric lymphoma are often non-specific. The primary method for diagnosis of gastric lymphoma is esophagogastroduodenoscopy (EGD). EGD not only provides a means for pathologic diagnosis through biopsy, but it is also important in the workup of other disorders with overlapping symptoms. It should be noted, however, that obtaining a diagnostic biopsy may be difficult because gastric lymphoma spreads submucosally and has vague surface qualities; therefore, many specimens may be required for accurate diagnosis. [1]

Staging

Prior to initiating treatment of newly diagnosed gastric lymphoma, it is important to rule out the presence of systemic lymphoma. Bone marrow biopsy, chest radiography, and abdominal CT are necessary for evaluation and future management. [1],[11] In addition, all patients thought to have primary gastric lymphoma should undergo indirect laryngoscopy to rule out involvement of Waldeyer's ring. [11]

Staging systems vary considerably by institution. Most centers use the Ann Arbor staging system, in which stage IE is tumor confined to the GI tract, IIE has regional lymph node involvement, IIIE has spread to other organs within the abdomen, and IV is spread beyond the abdomen. [11] Many clinicians also employ the Musshoff modification for Stage IIE lesions. This system divides IIE lesions into those that involve regional lymph nodes only (IIE1) and those that involve extraregional nodes (IIE2). Several authors have suggested that such staging modification may have prognostic significance. [7]

The evaluation of nodal extension is important for staging. However, the accuracy of CT scan for such evaluation has recently been questioned. Regional gastric lymph nodes are difficult to image by CT, and false-negative scans can lead to erroneous pathologic down-staging.[4] Endoscopic ultrasonography is a new technique that some have employed as a way of more accurately estimating both the depth of invasion and involvement of regional nodes.[12] More studies are required to delineate the precise role of endoscopic ultrasonography in the evaluation of gastric malignancies. However, it is a promising tool which may aid in the diagnosis, staging, and management of gastric lymphomas in the future.

Prognosis

The authors of numerous clinicopathologic studies on primary gastric lymphoma have generally agreed that the initial stage at presentation is a significant prognostic variable. [1],[2],[5],[7] Poorer prognosis has been associated with size greater than 5 cm, extension of tumor through the serosa, lymph node involvement, and older age. [5],[7] Good prognostic factors include surgical resectability, female sex, and early stage (IE and IIE). [5],[7],[8]

Many authors have suggested that histological grading is significantly associated with prognosis. However, because of the variety of classification systems, comparison among studies is difficult. With the concept of extranodal non-Hodgkin's lymphomas arising in MALT, a new grading system has been established. Recently, Cogliatti et al [13] reported the prognostic relevance of including grading based on the concept of MALT. Histologically, lymphomas arising from MALT are low-grade B-cell type and can be distinguished from high-grade malignant B-cell lymphomas with or without a low-grade component. Cogliatti's group found a significant association between histologic grade and 5-year survival rates, with MALT lymphomas carrying a much better prognosis. The investigators concluded that gastric MALT lymphomas tend to remain localized within the gastrointestinal tract, and morphologically and phenotypically resemble splenic marginal zone B-cells. [7],[9],[13]

Azab and coworkers [7] performed a multivariate analysis of 106 cases and found only three variables to be significantly associated with prognosis of gastric lymphomas: clinical stage, surgical resection, and histologic grade. They also found that when achievement of initial clinical remission was considered, the only significant prognostic factor was histological grade. Thus, the authors conclude a second set of prognostic factors should be considered after primary treatment, which include histological grade and whether clinical remission is achieved.

Treatment

There is much controversy surrounding the treatment of gastric lymphoma, and therapeutic regimens vary from institution to institution. Some centers advocate the surgery alone while others prefer non-operative treatment with radiation, chemotherapy or both.

Surgery

Traditionally, aggressive surgical resection has been the mainstay of treatment. [1],[5],[8] Aside from attempted cure, surgery provides the most accurate means of both grading and staging disease. [8] In the past, radical gastrectomy with en bloc resection of spleen and regional nodes was performed almost exclusively to assure accurate margins. However, with better understanding of tumor biology and increased use of adjuvant therapy, lesser procedures are now more accepted. [1],[8],[14] The goal of surgery is resection of all gross disease and involved lymph nodes, but in cases in which gastric lymphoma extends into the esophagus or duodenum, total resection may not be mandatory. Several authors have found no decrease in survival for patients with involved margins if adjuvant radiation therapy (RT) or chemotherapy were given.[7],[8],[14] Currently, splenectomy is only indicated in cases of direct tumor extension.[1]

Resectability rates range from 60% to 88%. Five year survival following potentially curative resection ranges from 50% to 87%.[1],[5] Five year survival in tumors judged non-resectable ranges from 6% to 40%.[1] It is important to note that the issue of resectability takes into account many factors at the time of operation including size and stage of the tumor, as well as the age and general state of health of the patient. Thus, whether or not a tumor is resectable reflects many variables which affect the survival of the patient.

Operative mortality rates range from 2.3% to 25% and are generally higher for palliative procedures,[1],[8] which have traditionally been performed not only for symptomatic relief but to remove tumor mass and avoid hemorrhage or perforation.[11] However, the necessity for debulking to avoid complications has more recently been questioned.[15] In a 1991 study by Maor and co-workers of 34 stage IE and IIE gastric lymphoma patients treated with only RT and chemotherapy, no patients developed bleeding or perforation.[15]

Radiotherapy

The role of RT in the management of gastric lymphoma is controversial. The use of RT as monotherapy for stage IE disease has been advocated by Burgers et al.[4] Their group found survival rates in stage IE patients treated with radiation alone comparable to those treated with resection and radiation combined (both near 85%). In contrast, several authors have found little benefit to RT alone[1],[7] and there is need for further study.

Radiation therapy has also been advocated as adjuvant therapy for both potentially curative and palliative resections. As with the use of radiotherapy alone, the role of RT in the treatment of patients who have undergone partial or complete surgical resection is not firmly established. Several studies have found no significant improvement of overall survival rates with administration of radiation after attempted curative resection, mostly in early stage disease.[1],[5],[7],[11],[13] Interestingly, some of these studies did show an increase in 5-year survival in patients with residual tumor who received RT. Shimm, et al[14] found an increased survival with adjuvant radiation in patients with poor prognostic factors: penetration of the bowel wall, positive regional nodes or positive surgical margins.

Although the role of adjuvant RT remains controversial, the majority of reports in the literature show no benefit when combined with curative resection. Furthermore, gastric lymphoma seems to be a more systemic disease with the majority of recurrences occurring at extraabdominal sites.[7],[8] In the absence of obvious persistent local disease, the need for additional local therapy with RT is put in question by these reports. The need for systemic postoperative treatment, as is provided by chemotherapy, is apparent.

Chemotherapy

As stated earlier, most primary gastric lymphomas are of the diffuse histiocytic or the diffuse large cell type. These subtypes seem to be quite responsive to current chemotherapeutic regimens.[7],[15] In addition, successful adjuvant chemotherapy allows for adoption of less aggressive surgical procedures, thus decreasing morbidity and mortality.[8]

The results of most post-operative chemotherapy trials have been encouraging. For stages IE and IIE, several authors have found excellent 5-year disease free survival for patients treated with chemotherapy after surgery.[1],[7],[8] Therapeutic regimens usually include CHOP (cyclophosphamide, Adriamycin, vincristine, and prednisone), CHOP-bleo (added bleomycin), COPP-bleo (cyclophosphamide, vincristine, procarbazine, prednisone, and bleomycin), or CVP (cyclophosphamide, vincristine, and prednisone).[1] Perhaps the most marked improvement in survival has been in the management of advanced stage disease. Solidoro, et al [16] reported on a series of 36 patients with stage IV disease, 18 of whom were treated with chemotherapy (CHOP or COPP-bleo alone or combined with surgery), while 16 received no chemotherapy. At the conclusion of the study period, 78% of patients treated with chemotherapy survived (mean survival 3.8 years). The 16 patients who received no chemotherapy had all died during the study period (median survival 5 months).

The issue of whether chemotherapy, alone or in combination with RT, should be used in the absence of surgery remains largely controversial. Several authors have reported good results with only RT and chemotherapy, and cite the advantages of stomach conservation and avoidance of surgical morbidity. [11],[15] Maor et al [15] reported improvement in 5-year survival with the combination of chemotherapy and RT compared with surgical therapy alone. In a study of 34 patients with stage IE and IIE disease, 68% of patients treated with combined RT and chemotherapy were free of disease at five to fifteen years after diagnosis. There was no significant difference in survival between IE and IIE. The authors conclude that surgery has a role only as salvage if non-surgical therapy fails. It is important to note that the Ann Arbor staging system was used in this study without the addition of Musshof's criteria. Only one of their patients had involvement in extraregional nodes (IIE2). Thus, the vast majority of their patients had only IIE1 disease which may account for the good results in their IIE patients. In addition, 5 of the 6 patients who died of lymphoma had recurrent local disease. In surgically treated patients most recurrences are extraabdominal, and local disease is well controlled.

Conclusion

It is difficult to ascertain which treatment is best from a review of the literature. Results for many therapeutic regimens are comparable. However, most studies are small, retrospective, and are plagued by selection bias. There is a clear need for more thorough prospective study. Until there is a more definitive answer, treatment options should be discussed with patients and therapy tailored to each individual's needs and preferences.

Patients with a diagnosis of early stage gastric lymphoma made by endoscopic biopsy should initially undergo surgical resection. Surgical therapy provides local control which may not be covered by chemotherapy and allows for correction of possible errors in preoperative staging. Adjuvant chemotherapy should be considered even in early stage disease, given that most failures of surgical therapy are extraabdominal.

In individuals with evidence of invasion and a definitive biopsy diagnosis, chemotherapy should be considered the mainstay of treatment with either surgery or radiation providing local control. In those patients with non-diagnostic biopsies, surgical exploration and resection are needed.

REFERENCES

1. Frazee RC, Roberts J. Gastric lymphoma treatment: medical versus surgical. Surg Clin North Am 1992;72:423-31.

2. Coulson WF. The Stomach. In: Coulson WF, ed. Surgical Pathology, Philadelphia:J.B. Lippincott Company, 1988:123-5.

3. Weingrad DN, Sherlock P, Straus D, et al. Primary gastrointestinal lymphoma: A 30 year review. Cancer 1982;49:1258-65.

4. Burgers JMV, Taal BG, Van Heerde P, et al. Treatment results of primary stage I and II non-Hodgkin's lymphoma of the stomach. Radiother Oncol 1988;11:319-26.

5. Dworkin B, Lightdale CJ, Weingrad DN, et al. Primary gastric lymphoma: a review of 50 cases. Dig Dis Sci 1982;27:986-92.

6. Isaacson PG, Wright DH. Extranodal malignant lymphoma arising from mucosa-associated lymphoid tissue. Cancer 1984;53:2515-24.

7. Azab MB, Henry-Amar M, Rougier P, et al. Prognostic factors in primary gastrointestinal non-Hodgkin's lymphoma. Cancer 1989;64:1208-17.

8. Shiu MH, Nisce LZ, Pinna A, et al. Recent results of multimodal therapy of gastric lymphoma. Cancer 1986;58:1389-99.

9. Zukerberg LR, Ferry JA, Southern JF, et al. Lymphoid infiltrates of the stomach. Am J Surg Pathol 1990;14:1087-99.

10. Brooks JJ, Enterline HT. Gastric pseudolymphoma: Its three subtypes and relation to lymphoma. Cancer 1983;51:476-86.

11. Haber DA, Mayer RJ. Primary gastrointestinal lymphoma. Semin Oncol 1988;15:154-69.

12. Caletti GC, Lorena Z, Bolondi L, et al. Impact of endoscopic ultrasonography on the diagnosis and treatment of primary gastric lymphoma. Surgey 1988;103:315-20.

13. Cogliatti SB, Schmid U, Schumacher U, et al. Primary B-cell gastric lymphoma: A clinicopathological study of 145 patients. Gastroenterology 1991;101:1159-70.

14. Shimm DS, Dosoretz DE, Anderson T, et al. Primary gastric lymphoma: An analysis with emphasis on prognostic factors and radiation therapy. Cancer 1983;52:2044-8.

15. Maor MH, Velasquez WS, Fuller LM, et al. Stomach conservation in stages IE and IIE gastric non-Hodgkin's lymphoma. J Clin Oncol 1990;8:266-71.

16. Solidoro A, Payet C, Sanchez-Lithon J, et al. Gastric Lymphomas: Chemotherapy as a primary treatment. Semin Surg Oncol 1990;6:218-25.