Prognostic Impact of Age at the Time of Diagnosis in Korean Patients with Diffuse Large B-cell Lymphoma in the Rituximab Era: A Single Institution Study.

PURPOSE: In contrast to the Western diffuse large B-cell lymphoma (DLBCL), prognostic impact of age in a Korean population with DLBCL has not been fully evaluated. MATERIALS AND METHODS: Six hundred and eight DLBCL patients treated with rituximab-containing chemotherapeutic regimens from January 2002 to March 2012 in Asan Medical Center were enrolled. Survival models using the restricted cubic spine-transformed age variable were constructed to evaluate non-linear relationships between age and survival outcome. Finally, age was categorized according to the conventional international prognostic index (IPI), National Comprehensive Cancer Network (NCCN)-IPI, and Grupo Español de Linfomas/Trasplante Autólogo de Médula Ósea (GELTAMO)-IPI schemes and the prognostic implications were evaluated. RESULTS: The relative hazard did not change significantly during the first to fifth decades, but began to increase exponentially in patients aged over 62 years. This pattern or relationship was also retained in a multivariate model fitted to the age-adjusted IPI and relative dose intensity. Multivariate survival analysis revealed that age > 75 years, but not age > 60 years, was associated independently with poor overall and progression-free survival when the relative dose intensity and age-adjusted IPI were taken into account. CONCLUSION: The outcome of DLBCL in Korean populations may deteriorate rapidly as age exceeds 62 years. Therefore, a consensus cutoff value for age in Korean DLBCL patients should be determined to better predict prognosis.

A new extranodal scoring system based on the prognostically relevant extranodal sites in diffuse large B-cell lymphoma, not otherwise specified treated with chemoimmunotherapy.

Extranodal involvement is a well-known prognostic factor in patients with diffuse large B-cell lymphomas (DLBCL). Nevertheless, the prognostic impact of the extranodal scoring system included in the conventional international prognostic index (IPI) has been questioned in an era where rituximab treatment has become widespread. We investigated the prognostic impacts of individual sites of extranodal involvement in 761 patients with DLBCL who received rituximab-based chemoimmunotherapy. Subsequently, we established a new extranodal scoring system based on extranodal sites, showing significant prognostic correlation, and compared this system with conventional scoring systems, such as the IPI and the National Comprehensive Cancer Network-IPI (NCCN-IPI). An internal validation procedure, using bootstrapped samples, was also performed for both univariate and multivariate models. Using multivariate analysis with a backward variable selection, we found nine extranodal sites (the liver, lung, spleen, central nervous system, bone marrow, kidney, skin, adrenal glands, and peritoneum) that remained significant for use in the final model. Our newly established extranodal scoring system, based on these sites, was better correlated with patient survival than standard scoring systems, such as the IPI and the NCCN-IPI. Internal validation by bootstrapping demonstrated an improvement in model performance of our modified extranodal scoring system. Our new extranodal scoring system, based on the prognostically relevant sites, may improve the performance of conventional prognostic models of DLBCL in the rituximab era and warrants further external validation using large study populations.

A randomized comparison of peripheral blood hematopoietic progenitor cell level of 5/mm3 versus 50/mm3 as a surrogate marker to initiate efficient autologous blood stem cell collection.

We previously showed that at least 5/mm(3) hematopoietic progenitor cells (HPCs) could be used as a marker for initiating autologous blood stem cell collection (ABSCC). However, the timing of efficient ABSCC following mobilization is still to be determined. We conducted a prospective, randomized comparison of 5/mm(3) versus 50/mm(3) peripheral blood (PB) HPCs as a surrogate marker to initiate efficient ABSCC. Forty-five consecutive patients, 26 with multiple myeloma (MM) and 19 with non-Hodgkin's lymphoma (NHL), were enrolled between October 2004 and October 2006. Chemotherapy was cyclophosphamide 4 g/m(2) for MM and ESHAP (etoposide, methylprednisolone, high-dose cytarabine, and cisplatin), with or without Rituximab, for NHL. Circulating HPCs were monitored daily with the Sysmex SE9000 automated hematology analyzer, and harvested CD34+ cells were counted by flow cytometry. ABSCC was initiated when HPC levels reached at least 5/mm(3) (HPC5 group) or 50/mm(3) (HPC50 group). The median number of harvested CD34+ cells was 15.0 x 10(6)/kg and 21.0 x 10(6)/kg in the HPC5 and HPC50 groups, respectively (P = 0.23). Optimal collection (>5 x 10(6) CD34+ cells/kg) in a single session (day 1) was attained in 15 HPC5 patients (63%) and in 14 HPC50 patients (67%), and targeted collection of 5 x 10(6) CD34+ cells/kg was achieved in 100 and 95% of HPC5 and HPC50 patients, respectively (P = 0.47), with a median number of 1 apheresis in both groups (P = 0.58). There were no between group differences in optimal collection rate on day 1, median number of aphereses to achieve optimal collection, and overall optimal collection rate. HPC > or = 5/mm(3) and > or =50/mm(3) are both reliable indices for the timing of ABSCC.