Abstract

The aim of this study was to evaluate the influence of duration of first remission (FR) on overall survival (OS) of patients (pts) with DLBCL in a population-based setting, and further to assess the possible differences in clinical features or IPI between pts with different duration of FR. We analyzed all DLBCL pts registered on the database of the Scotland and Newcastle Lymphoma Group. Further inclusion criteria were age >18 yrs and defined clinical stage (CS) at presentation, treatment with anthracycline based chemotherapy, and no immunotherapy or stem cell transplantation in first line treatment. According to duration of FR four groups were defined; “refractory”-pts who did not respond to first line treatment or relapsed within 9 months from diagnosis, “early relapse”-pts who relapsed within 9–18 months from diagnosis, “late relapse”-pts who relapsed after 18 months from diagnosis and “no relapse”-pts who did not relapse. From 1990 to 2003, 2025 DLBCL pts were registered within the database and 1391 pts fulfill inclusion criteria. The median age at diagnosis was 64 yrs, 40% of pts were aged ≤60yrs, 51% of pts were male, 79% of pts had ECOG <1. 18% of pts were diagnosed with CS I, 28% with CS II, 25% with CS III and 29% with CS IV. B-symptoms were present in 42% of pts, extranodal involvement in 34%, and marrow involvement in 16% and bulky disease in 53%. IPI was calculated for 803 pts–38% of them had low IPI, 26% low intermediate IPI, 22% high intermediate IPI and 14% high IPI. 5yr progression free survival and OS were respectively 37% and 46%. 566 pts (41%) were in “refractory–group”, 92 (7%) pts in “early relapse–group”, 120 (9%) pts in “late relapse–group” and 597 pts (43%) in “no relapse–group”, 16 pts was excluded due to short follow-up time. Among pts ≤60yrs, the proportion were 33%, 7%, 8% and 52% respectively and among pts >60 yrs 46%, 7%, 9% and 38%, respectively. 5 yrs OS was 8% in “refractory–group”, 16% in “early relapse–group”, 50% in “late relapse–group” and 90% in “no relapse–group”, differences between the groups were statistically significant, p<0.001. These results were confirmed for pts ≤60 yrs–5 yrs OS was 13%, 31%, 66% and 94% respectively, p<0.001 and for pts >60yrs 4%, 6%, 40% and 85%, respectively, p<0.001. Importantly, 45% of pts in “refractory group” had low and low-intermediate IPI, 57% in “early relapse group”, 79% in “late relapse group” and 80% in “no relapse group” in evaluation of all pts, p<0.001. In evaluation of pts ≤60yrs, the proportions for age-adjusted IPI were 31%, 53%, 77% and 77%, respectively, p<0.001 and of pts >60yrs 32%, 50%, 76% and 68%, respectively, p<0.001. There were statistical significant differences in IPI and age-adjusted IPI between “refractory” and “no relapse” groups tested for all pts, pts ≤60yrs and pts >60yrs, however there were no statistically significant differences between “early relapsed” and “late relapse” groups. To assess the further potential differences in clinical features at presentation between “primary refractory” vs “no relapse” pts and “early relapse” vs “late relapse” pts, the groups were compared for the following factors: age, sex, CS, B-symptoms, ECOG, extranodal and bone marrow involvement, bulky disease, haemoglobin, leukocytes, LDH, albumin, urea, and alkaline phosphatase serum level. In evaluation of all pts and pts ≤60 years the “primary refractory–group” differed statistically significantly from the “no relapse group” in all evaluated factors except of sex and in evaluation of pts >60yrs in all factors. The “early relapse group” differed statistically significantly from the “late relapse group” only in ECOG performance status in evaluation of all pts, CS and bulky disease in evaluation of pts ≤60 years and age and CS in evaluation of pts >60yrs. In a population-based study of pts with DLBCL treated with anthracyclines, duration of FR defined groups of pts with statistically different OS. It was impossible to distinguish entirely by clinical features or IPI at diagnosis. We conclude, that the currently used prognostic indices based on clinical factors need to be enhanced using biological features of the tumor.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author