Diffuse Large B-Cell Lymphoma in the Very Elderly, a Study of Outcome in Patients Aged 80 and Older.

The trial showing that the addition of rituximab to the CHOP regimen increases complete response (CR) rate and overall survival (OS) in elderly patients (pts) with diffuse large B-cell lymphoma (DLBCL) did not include pts over the age of 80 (Coiffier, 2002, NEJM). There is very limited data on octogenarians with DLBCL. These pts are often treated with reduced-dose therapeutic regimens, which may lead to a poorer OS (Meyer, 1995, J Clin Oncol). We were interested in examining the proportion of pts with DLBCL aged 80 and over in our institution; to characterize the determinants of their OS; to assess how many received standard therapy; and, to define which characteristics lead physicians to withhold therapy. Patients and method: A retrospective chart analysis was conducted of clinical and pathological data on all new cases of DLBCL from 2004 to 2008. Clinical data collected included International Prognostic Index (IPI), LDH, therapy received, and survival. Pathologic data included presence or absence of bcl-6 and/or bcl-2 expression, and proliferative fraction as determined by the level of Ki-67. This study received IRB approval. A Kaplan-Meier survival analysis was performed, Fisher's exact test was used to estimate p value for proportions, and confidence intervals were estimated for means. Results: We identified 54 new cases of DLBCL with a median age of 69.5 years. Fifteen pts (27.8%) were age 80 or older. Male to female ratio was 1:1.5. The average IPI was greater in pts above 80 than in pts below 80 years of age, 3.0 [95% confidence interval (CI), 2.17-3.83] vs. 1.9 [95% CI, 1.51-2.29 ], respectively. The LDH was high in 57.9% (22/38) of pts in the younger group vs. 69.2% (9/13) of pts in the older group (p=0.52). The presence of bcl-6 in the younger group versus the older group was 88.6% (31/35) and 84.6% (11/13), respectively (p = 0.65). The expression of bcl-2 in the younger group versus the older group was 55.9% (19/34) and 66.7% (8/12), respectively (p=0.73). The proportion of younger pts vs. older pts with high Ki67 (>80%) was 45.7% vs. 53.8%, respectively (p= 0.75). Thirty six pts under age 80 received RCHOP compared to 7 pts 80 years and over (97.3% vs. 50%) (p=0.0006). There did not seem to be a correlation between IPI or co-morbidity and whether pts received treatment. However, therapy tended to be offered to outpatients but withheld from patients hospitalized at time of diagnosis. Chemotherapy had to be reduced or terminated due to poor tolerance in 1 patient in the younger cohort vs. 2 pts in the older cohort. Amongst the older group who did not receive RCHOP, 6 pts (85.7%) did not receive any chemotherapy, while 1 patient received 5 cycles of RCEPP. Kaplan Meier analysis for overall survival of pts under age 80 compared to over age 80 is shown below (HR = 7.0, 95% CI [2.0 – 24.2]), (p = 0.002). All pts above age 80 who did not receive any chemotherapy died vs. 25.0% (1/4) who received full dose RCHOP, vs. 66.7% (2/3) who received reduced RCHOP. The CR rate was 88.6% (31/35) for those aged under 80 vs. 75% (6/8) for those above age 80 (p=0.31), while the relapse rate was 11.4% (4/35) vs. 25% (2/8), respectively (p = 0.31). Conclusions: Though limited, these data reveal interesting findings about pts aged over 80 with DLBCL. With our aging population, more physicians will be faced with the question of how to treat the very elderly with this diagnosis. These pts had a poorer overall survival which, in large part, can be attributed to the fact that therapy was never administered. Age above 80 and hospitalization seemed to be the only clear reasons for withholding therapy. Elderly pts tended to have higher IPIs, but did not seem to have histologically more aggressive tumours. Interestingly, the pts over the age of 80 who died did tend to have more frequent expression of bcl-2. In contrast, all elderly pts who survived were bcl-6 positive. When standard therapy is given to pts over the age of 80, their response rates do not differ from those in younger pts. Further research into pts above 80 years of age with DLBCL is needed to confirm our findings. Nonetheless, our results raise questions about how we approach treatment of DLBCL in the very elderly and suggest that age alone should not prevent standard doses of therapy from being administered.

1. Kaplan-Meier survival curves for the two age groups:

No relevant conflicts of interest to declare.