Simple but powerful prognostic scoring model for MALT lymphoma: a retrospective study

Abstract

Dear Editor, Little is known about the prognostic factors of posttreatment extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT lymphoma) [1]. To address this issue, we carried out a retrospective cohort study with MALT lymphoma patients treated at our institute. We reviewed the patients with MALT lymphoma treated at The University of Tokyo Hospital between February 1997 and April 2010. Patients who had undergone extensive survey for precise staging at initial presentation and with clinical follow-up data for at least 24 months after initial therapy were included in this study. Impact of clinical variables on survival was assessed using the log-rank test (in univariate analyses), and the factors showing relevant association (P<0.10) were subjected to the multivariate analysis with Cox proportional hazard model. Parameter selection was optimized with Akaike’s information criteria. The log-rank test was used to compare different groups of patients. We identified 53 patients eligible to the inclusion criteria. The characteristics of these patients at diagnosis are summarized in Fig. 1a. Thirty-nine patients achieved complete remission, while relapse was seen in 11 patients. Median overall survival (OS; time from histological diagnosis to death from any causes), progression free survival (PFS; time from initial treatment to the date of first progression), and disease-free survival (DFS; time from first confirmation of CR after therapy to the day of losing CR) were 60 months (range 5– 148 months), 42 months (range 1–131 months), and 56 months (range 6–128 months), respectively. Univariate analyses identified low hemoglobin level (<120 g/L) (P00.0024), abnormal LDH level (P00.007), higher ECOG performance status scoring (P00.021), and peripheral lymph node involvement (P00.046) as possible contributing factors for OS; low hemoglobin level (P00.067), abnormal sIL-2R level (P00.018), peripheral lymph nodes involvement (P00.034), and higher clinical stage according to the Ann Arbor classification (P00.007) for PFS. Only higher clinical stage (P00.008) affected DFS. We conducted the multivariate analysis for PFS because PFS is one of the most important markers for treatment efficacy for MALT lymphoma. Because the Ann Arbor staging and hemoglobin level at the initiation of treatment were revealed as independent prognostic factors on poor PFS (P00.007 and P00.041; hazard ratio, 3.58 and 2.97), we created a simple scoring model for PFS by allocating one point each for these two factors and stratifying patients to three groups according to the total score. This model successfully classified patients into three groups with different PFS (P00.004) (Fig. 1b). This model was also valid for DFS (P<0.001) (Fig. 1c) but not for OS (P00.063). K. Morita :Y. Nannya : T. Yoshizato :M. Kurokawa Department of Hematology & Oncology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan