Primary Digestive Tract Lymphoma: A Prospective Multicentric Study of 91 Patients zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA AGNES RUSKONi-FOURMESTRAUX,* PHILIPPE AEGERTER,* ALAIN DELMER,* NICOLE BROUSSE,§ ANNIE GALIAN,’ JEAN-CLAUDE RAMBAUD,” and the GROUPE D’ETUDE DES LYMPHOMES DIGESTIFS’ *Departments of Gastroenterology and Hematology, Hdtel-Dieu, Paris; *Department of Biostatistics, HBpital Ambroise Pare, Boulogne; “Departments of Pathology, Hdpital Necker and HBpital LariboisiBre. Paris; I’Department of Gastroenterology, Hapital Saint-Lazare, Paris: and “Fo&ation Francaise de Canc&ologie Digestive, Dijon, France Bacfwound: The optimal management of primary gas- trointestinal lymphoma including the use of surgery remains unsettled. This prospective study aimed to determine the prognostic factors of primary gastroin- testinal lymphoma and to evaluate a therapeutic strat- egy based on surgical tumor reduction, followed by che- motherapy adapted to the histological type of tumor and tumor removal or nonremoval. Methods: Ninety- one patients were included (mean age, 50.4 years) with mainly gastric (61%) or several digestive tract or- gan (18%) involved sites in clinical stages I, (43%), II, (27%), and IV (30%). Three therapeutic groups were defined: group I, patients with low-grade lymphoma (n = 28); group II, patients with high-grade lymphoma who had complete tumor removal (n = 24); and group Ill, patients with high-grade lymphoma who had only partial or no tumor removal (n = 39). Each group un- derwent adapted chemotherapy. Results: The respec- tive overall 5-year survival rates for groups I, II, and Ill were 81% 4 6%, lOO%, and 56% + 8%, respectively (P < 0.0001). By Cox multivariate analysis, the prog- nostic factors of survival were age under 65 (P < 0.05), gastric localization (P < 0.05), stage I, (P < O.OOl), and radical or incomplete surgery zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA (P < 0.01). Conclusions: Combined radical surgery and chemother- apy according to histological grading is associated with prolonged remission in patients with primary di- gestive tract lymphoma. Moreover, compared with che- motherapy alone, incomplete resection of tumor is as- sociated with increased survival of high-grade lymphomas. T zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA he prevalence of primary digestive tract lympho- mas (PDTL) seems to be increasing, as recently suggested by several studies.lm5 There is no current agreement on the best therapeutic approach to PDTL, because prospective studies with accurate histological typing are scarce6’ and adequate clinical staging is even rarer. Most reports concern retrospective studies of poorly characterized patients given various treat- ments. Little is therefore known about the respective efficacy of the three main therapeutic approaches, sur- gery, chemotherapy, and radiotherapy, used either alone or in combination. This prospective and multi- centric study aimed to determine the prognostic fac- tors of PDTL and to evaluate a therapeutic strategy based on surgical tumor reduction followed by chemo- therapy adapted to histological subtype and tumor re- sectability. Patients and Methods Patients From January 1984 through January 1990,108 adult patients with previously untreated PDTL were enrolled in a prospective multicenter therapeutic study. The primary character of the lymphoma was defined according to Lewin et al. i.e., either predominant gastrointestinal lesions with- out previous peripheral lymphadenopathy or any lym- phoma whose clinical presentation was related to digestive tract involvement. ‘O,” Exclusion criteria were age more than 75 years, Burkitt’s and lymphoblastic subtypes of non- Hodgkin’s lymphoma, and Mediterranean or human immu- nodeficiency virus-related lymphoma. Informed consent was obtained from the patients, and the protocol was ap- proved by the Saint-Lazare Ethics Committee. Histopathology The histology of each tumor was reviewed by at least two pathologists based on endoscopic biopsy specimens and/ or surgical material. For therapeutic purposes, two grades of malignancy were considered: low-grade lympho- mas consisting of small cell proliferations and high-grade lymphomas including tumors containing at least one compo- nent of large cells. Thus, the high-grade group of PDTL included the intermediate and high-grade lymphomas of the U.S. National Cancer Institute’s working formulation, and the low-grade group comprised the diffuse small cleaved cell Abbreviations used in this paper: MALT, mucosa-associated lym- phoid tissue; PDTL, primary digestive tract lymphoma. 0 1993 by the American Gastroenterological Association 0016-5085/93/$3.00