R-Mini-CHOP Versus R-Bendamustine As Primary Treatment for Diffuse Large B-Cell Lymphoma (DLBCL) in Elderly and Comorbid Patients

Introduction: Incidence of diffuse large B-cell lymphoma (DLBCL) increases in part due to a growing elderly population. Age, comorbidities and different drug metabolism confer to increased risk for treatment-dependent toxicity. Thus, standard R-CHOP frequently is no treatment option. Therapeutic strategies vary from adapted CHOP-protocols to Bendamustine, both in combination with Rituximab (R). R-mini-CHOP is an attenuated CHOP-regimen with reduced cyclophosphamide and doxorubicin doses that recently has been shown to be a good compromise between efficacy and safety in patients older than 80 years. R-Bendamustine is well established in indolent lymphoma and shows efficacy in small DLBCL series of elderly and co-morbid patients. No comparison between R-mini-CHOP and R-Bendamustine is available.

Patients/methods: We retrospectively analyzed 24 consecutive patients with newly diagnosed DLBCL unfit for R-CHOP. All finished firstline therapy. Median observation time was 15 months. Ten patients received R-mini-CHOP and 14 R-Bendamustine (R-B) as 1^st line therapy. Baseline characteristics were age, ECOG, comorbidities quantified with the relevant somatic comorbidity index (RSC) based on cumulative illness rating scale for geriatrics (CIRS-G), and risk stratification according to the international prognostic index (IPI). Primary endpoint was progression free survival (PFS), secondary endpoints overall response rate and survival (ORR/OS), rate of complete remission (CR) and graded toxicity according to NCI-CTCAE 4.0.

Results: Median age of the 14 R-B treated patients was 83 years (62-92), ECOG score was ≥ 2 in 36%. 50% had an RSC ≥ 3, Ann-Arbor-Stage was ≥ 3 in 57% and IPI was ≥ 3 in 57%.

Median age of the 10 R-mini-CHOP patients was 81 years (68-87), ECOG was ≥ 2 in 30%. 40% had an RSC ≥ 3, Ann-Arbor-Stage was ≥ 3 in 70% and IPI was ≥ 3 in 60%.

Median 5 (2-9) cycles of R-B and 4 (2-6) cycles of R-mini-CHOP were administered, and 36% vs. 40% reached the planned six cycles. The ORR for R-B was 64% with med. PFS of 6 (2-41) months and med. OS of 15 (4-56) months. The ORR for R-mini-CHOP was 60% with a med. PFS of 6 (2-34) months and med. OS of 8 (3-36) months. 21% (R-B) vs. 20% (R-mini-CHOP) achieved CR.

Non hematologic toxicity was mainly grade I and II. Grade III/IV hematologic toxicity occurred in 7% vs. 22% of the administered cycles of R-B/R-mini-CHOP, hence 21% vs. 50% patients were affected.

Conclusions: This exploratory retrospective analysis confirms R-B and R-mini-CHOP as reasonable treatment options for 1^st line treatment of elderly and comorbid DLBCL-patients. Toxicity was well manageable in both treatment arms. Superior hematologic tolerability of R-B compared to R-mini-CHOP warrants a treatment recommendation in particular for patients with impaired hematologic reserve. However, our data cannot recommend any of both regimens as “winner”. For evidence based decision-making prospective and comparative investigation is required.