Prognostic Impact of Past History of Other Malignancies in Diffuse Large B Cell Lymphoma

Continual improvements of diagnosis and treatment methods for malignant tumors have improved the survival rates of cancer patients, and treatment strategy for the cancer survivors should be discussed more carefully. The risk and prognosis of secondary primary malignancy (SPM) in the non-Hodgkin's lymphomas (NHLs) have been investigated intensively, while little is known about the clinical and prognostic impacts of the past history of cancer: past cancer (PC) and the co-existence of synchronous cancer (SC) in the treatment of NHLs. In this study, we retrospectively analyzed 809 patients with diffuse large B-cell lymphoma (DLBCL) who were histologically diagnosed and treated between January 2006 and February 2016 at seven independent institutes involved in the Kyoto Clinical Hematology Study Group (KOTOSG), and investigated the frequency and clinical impacts of PC and SC in DLBCL. Based on the Surveillance Epidemiology and End Results Program (SEER) definition for the chronicity of malignant tumors, DLBCL with PC was defined as patients with precedent cancer(s) that occurred more than two months before the diagnosis of DLBCL, DLBCL with SC was defined as patients diagnosed with lymphoma and other malignancies within two months, and DLBCL with MPM was defined as patients including both DLBCL with PC and DLBCL with SC. As the result, 123 (15.2%) of 809 DLBCL patients were defined as patients with MPM, including patients with PC (n=94) and patients with SC (n=29). The median age of DLBCL with MPM was statistically higher than that of patients without MPM (75 vs. 70 years old; p<0.001), while there was no significant difference in the the Ann Arbor stages or international prognostic index (IPI)-defined disease risk between patients with MPM and patients without MPM. Frequent past/synchronous other cancers (>5%) included gastric cancer (29.2%), colorectal (20.3%), prostate (12.2%), breast (12.2%), lung cancer (8.1%) and bladder cancer (5.7%). Stomach cancer was the most frequent in both PC and SC, and approximately two thirds (65.5%) of concomitant cancers were those detected by upper gastrointestinal survey, i.e., stomach cancer and upper aerodigestive tract cancer, in SC. We next investigated the prognostic impact of MPM. Importantly, both OS and PFS of DLBCL with MPM were significantly shorter than those without MPM (the 3-year OS: 56.2% vs. 74.6%, p<0.001, the 3-year PFS: 49.3% vs. 64.2%, p<0.01). Multivariate analysis incorporating age, gender, clinical stage, IPI disease risk and treatment approach as other variables revealed that the presence of MPM was found to be one of independent risk factors for both the shorter OS and PFS (hazard ratio = 1.68; 95% CI = 1.22-2.31, p<0.001 for OS and hazard ratio = 1.58; 95% CI = 1.19-2.09, p=0.002 for PFS). There was no significant difference in OS and PFS between PC and SC, and actually, the survival curves for OS and PFS of PC and SC largely superimposed in our cohort. When we assessed the cause of death, the rates of death by DLBCL were not significantly different between DLBCL patients with MPM (57.7%) and without MPM (64.5%), and malignancy (lymphoma and/or other cancer)-unrelated death rates was also not significantly different between DLBCL with MPM (32.7%) and DLBCL without MPM (32.3%) during the observation period. We next analyzed the prognostic impact of MPM according to the IPI-defined risks. As the results, MPM significantly associated with the shorter OS in IPI-low risk (the 3-year OS: 91.2% vs. 74.5%, p=0.018, the 3-year PFS: 85.5% vs. 71.6%, p=0.016) and IPI-high risk DLBCLs (the 3-year OS: 51.4% vs. 19.1%, p<0.001, the 3-year PFS: 38.7% vs. 7.3%, p=0.003), but not in IPI-low-intermediate risk or IPI-high-intermediate risk. Multivariate analyses revealed that MPM was one of the independent poor prognostic factors only in the IPI-high risk DLBCL (hazard ratio = 1.97; 95% CI = 1.22-3.16, p<0.01), but not in the IPI-low risk patients (hazard ratio = 1.91; 95% CI = 0.84-4.34, p=0.12). Approximately 50% (16 of 30) IPI-high risk patients with MPM died by the progression of lymphoma. Our data showed that MPM affect the survival of DLBCL patients. Of particular, MPM is an independent poor prognostic factor in IPI-high risk DLBCL who might require the additional clinical approach for improving the survival. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki, and was approved by the Institutional Review Board of our institutes.