Clinico-pathological and biological characterization of a cohort of 140 patients affected by relapsed/refractory Diffuse Large B Cell Lymphoma: looking for prognostic factors.

Diffuse Large B Cell Lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma. Although most patients respond efficaciously to the first line therapy, about 30-40% are primary refractory or present relapse after an initial response, constituting a group with poor prognosis. Many efforts have been made to find prognostic factors, able to identify high-risk patients; however, the currently available scores are often inadequate to this purpose. In this study we compared the clinical and biological features of a cohort of relapsed/refractory patients (R/R, n = 140), with those a cohort of patients not affected by relapse after at least 5 years of follow-up (controls, n = 45). We divided the R/R patients according to the time of relapse in three subgroups – refractory (characterized by persistence of disease or recurrence within 9 months from diagnosis, n = 72), early relapsed (with recurrence of disease within 10 and 24 months, n = 35) and late relapsed (with recurrence of disease beyond 24 months from diagnosis, n = 33). We also performed gene expression profiling (GEP) analysis on a subgroup of patients, aiming at recognizing differentially expressed genes; through this analysis, we identified the B1 subunit of NADH:Ubiquinone Oxidoreductase (NDUFB1) as a gene with enhanced expression in R/R patients. We further verified NDUFB1 expression and protein levels in DLBCL cell lines and performed ad hoc immunohistochemistry on patients’ samples. Our results show that the R/R subgroups differentiate in terms of clinical and biological features, but also in terms of outcomes, with an inferior post-relapse overall survival (OS) for refractory patients. In the whole R/R cohort, we confirmed the prognostic value of the International Prognostic Index (IPI) and the Revised-IPI (R-IPI), even when calculated at relapse, and of well-known adverse factors (B-symptoms, advanced stage, lactate dehydrogenase (LDH) increase, bulky disease, extra-nodal involvement), but we also found a novel correlation between male sex and inferior progression-free survival (PFS) and OS, and between inflammatory indexes as neutrophils/lymphocytes (N/L) ratio, Systemic Immune-Inflammation Index (SII) and C-reactive protein value/albumin value (CAR) and outcome. As for immunohistochemistry data, high Ki67 values correlated with reduced OS, while NDUFB1 overexpression caused a PFS disadvantage. We detected a trend of more frequent altered expression of P53 in the R/R cohort, in which all patients with enhanced expression were refractory, while all cases with a “null” phenotype belonged to the late relapsed group. In the second line setting, 23% of patients underwent autologous stem cell transplant (ASCT); transplanted patients had a post-relapse OS significantly superior to patients who did not receive transplant. Globally, we assessed the importance of the time of relapse for prognosis prediction, and identified some subgroup-specific features and variables impacting on outcome. These results, if validated in larger cohorts could contribute to the formulation of new prognostic scores, improving risk-stratification in DLBCL.