Abstract 3923

Poster Board III-859

Introduction

Wide use of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) has improved the clinical outcome for elderly patients with DLBCL; however, a higher prevalence of coexisting disorders remains a problem. Correlation between their comorbidities and prognosis has not yet been well investigated.

Patients and methods

We retrospectively analyzed all patients over 65 years old who had been newly diagnosed with DLBCL at our institution from 2001 to 2008. To assess their comorbid medical status, we calculated the Charlson Comorbidity Index (CCI) for patient excluding primary disease. Prognostic factors were identified by Cox proportional hazards regression model. We classified patients into a low CCI group (CCI 0-1) and a high CCI group (CCI 2 or higher). Kaplan-Meyer curves for each group were evaluated by logrank test.

Results

A total of 80 patients were enrolled in this analysis. The median age was 73 (range 66-90) and the median observation period was 28 months (range 4-90 months). 62 patients (77.5%) were treated with R-CHOP, 15 (18.6%) underwent some other regimen, and 3 (3.8%) were given best supportive care only. According to revised International Prognostic Index (r-IPI), 43 patients were in the good risk group and the others were in the poor risk group. The estimated 3 year over all survival (OS) rate for these groups were 90% and 45% (p<0.0001). As for CCI, 14 patients (17.5%) were assigned to the high CCI group. Multivariate analysis revealed high CCI was associated with worse OS, while independent of r-IPI [Hazard Ratio (HR) 3.20, 95% Confidence interval (CI) 1.28-7.41, p=0.0145]. Among r-IPI poor risk patients, the high CCI group was inferior to the low CCI group for the 3 year OS rate (14% vs 56% p=0.0358), whereas this was not significant among r-IPI good risk patients (69% vs 94% p=0.0617).

Conclusions

Among elderly patients with DLBCL, high CCI is independently associated with poor survival. Patients having both poor r-IPI and high CCI may need discrete strategies.

Disclosures:

No relevant conflicts of interest to declare.

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution