Abstract 2693


Treatment of Diffuse Large B-Cell Lymphoma (B-DLCL) is not well coded in the elderly patients. They may receive full dose immunochemotherapy, low dose chemotherapy or palliative treatment regarding co morbidities, Performans Status (PS), psychological, social or mental state. The lack of age-adapted prognosis factors including geriatric scales induce a subjective choice for the treatment. The purpose of the study is to evaluate the outcome of all the patients treated in a single institute for a B-DLCL, with comparison of age of diagnosis and treatment received.


All patients with B-DLCL, age≥70 years, treated in the Paoli-Calmettes institute between 1995 and 2008 were included, excepted patients with intra-ocular and cerebral localizations or with a “Burkitt-like” histology. Were also excluded patients with incomplete data. Treatments were simplified for statistic analysis in three types: CHOP Like (CH-L): three chemotherapies with anthracyclin (or etoposide in place if cardiac impossibility) with conventional doses, mini-CHOP-Like (mCH-L): with reduce doses of anthracyclin and cyclophosphamide, or COP Like (C-L): two agents without anthracyclin. Factors studied in the different items are systematically Age (70–79 vs olders), PS (0–1 vs 2–4), LDH, Ann Arbor stage (AA:1–2 vs 3–4) and type of chemotherapy.


From 1995 to 2008, 212 patients with B-DLCL were admitted in the Paoli-Calmettes institute for a B-DLCL. The median age was 76 years [range 70–90], 70% of the patients had a PS=0–1 and 30% a PS=2–4, LDH was increased in 55% of patients, AA was 3–4 in 58% of cases. The repartition of chemotherapy was 56% for CH-L, 33% for mCH-L and 11% for C-L. In the 70–79 age subgroup, CH-L is predominant (67% vs 25% for the older patients, p<0.0001). Four patients died before therapy initiation. Survival curves for mCH-L and C-L are identical, with no difference of population characteristics and patients are grouped for the final analysis (mCH-CL group). Rituximab was added to the chemotherapy in 63% of cases. Overall Survival at 12 and 60 month was respectively 73% and 47% with a median [IC95] of 48.8 month. Age at diagnosis is statistically significant with a 5-year survival of 53% and 29% for respectively 70–79 years and older (p=0.0045). Patients characteristics of the age subgroups are different only for the type of chemotherapy infused (p<0.0001) and not for the others factors analyzed. Choice of chemotherapy was also important (P=0.0011) with an OS of 55% and 37% respectively for CH-L and mCH-CL protocols. In this case, patients characteristics are different in term of date of diagnosis (p<0.001), age of diagnosis (p<0.001), PS (0.04) and AA (0.011). Surprisingly, there is no difference in OS when rituximab was given (p=0.7), and despite the difference of treatment, there is no difference of incidence of relapse in the two age groups (p=0.97).


Survival of our elderly population of patients with B-DLCL is comparable to the literature. With non-selected patients, repartition of factors from the IPI score is not different in the two age subgroups, but the more intensive chemotherapy is given in the less older patients. Moreover, OS is increased in this CH-L protocol, in contrast with the same incidence of relapse. Furthermore, the use of rituximab, a major treatment of B-DLCL in the elderly, do not influence OS in this non-selected population of patients. These data confirm the requirement of a more discriminant prognosis model than the IPI score for the daily practice, including relevant geriatric factors.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution