A Significant Better Two Year Survival in the Very Elderly Patients with a Diffuse Large B-Cell Lymphoma (DLBCL) Who Did Complete Standard Therapy: A Population-Based Cohort Study On Treatment, Toxicity and Outcome.

With the aging of the population clinicians are increasingly confronted with elderly, often frail, patients with a DLBCL. While the very elderly patients, defined by the age of ≥75 year, are often excluded from clinical trials, only few prospective data in the rituximab-era are available on the outcome in this patient cohort. To asses efficacy, tolerability and safety of standard intensive rituximab-containing therapy in those patients, a descriptive population-based, cohort study was performed. Detailed information on treatment, toxicity and outcome in all patients aged ≥75 year, diagnosed with DLBCL in the Dutch province Friesland was prospectively gathered and analyzed.

Since 2005 all patients diagnosed with a haematological malignancy in Friesland, a province with 600.000 inhabitants, are prospectively registered and followed in a population-based registry, the Hemobase^R. For this analysis data of all patients aged ≥ 75 years with a newly diagnosed DLBCL over a period of six years were retrieved from the Hemobase^R. Data of clinical characteristics, treatment and its adaptations, treatment-related toxicity and outcome were obtained. Treatment modality was divided in four groups: R-CHOP chemotherapy, other rituximab-containing therapy, palliative radiotherapy and only supportive care. Follow-up was completed until 31 December 2011. Cox Proportional hazards model was used for identifying significant prognostic risk factors. Kaplan-Meier curves for each group were evaluated by a logrank test.

From 2005 until the end of 2011 103 patients aged ≥ 75 years were diagnosed with a DLBCL. The median age was 81 years (range 75 – 96) with a slight female predominance (57%). The median observation period was 13 months (range 1 – 78) and 31 months for those still alive at time of evaluation.19 patients (19%) had stage 1 disease; in 74 patients (71%) advanced disease (stage 2–4) was reported and in 10 patients (10%) staging was incomplete. In 84 patients (81%) an age adjusted International Prognostic Index was calculated, 43% of them had an aaIPI of 2 or 3. Of the patients with stage 1 disease, 13 (68%) received three courses of R-CHOP and radiotherapy with a curative intent. Eight (62%) of them did complete this therapy. Of the other 84 patients 57 (68%) received R-CHOP chemotherapy with a curative intent. 31 (54%) of them completed at least six cycles. Ultimately 39 patients (56%), who did start with a standard therapy regimen, completed their treatment. A complete remission was achieved in 30 patients (77%). In the remaining group 10 patients (10%) received alternative, suboptimal, rituximab-based chemotherapy, 9 (9%) of patients was treated with radiotherapy. In 12 patients (11%) only supportive care was given. Severe toxicity (grade 3–4), occurred in 66% of all patients treated with rituximab-containing chemotherapy. Toxicity grade 3–4 was the main reason for receiving less then six cycles of chemotherapy (44%). Ten toxicity-related deaths (13%) were observed in patients treated with R-CHOP, with eight deaths due to infectious complications. The two-year survival was 67% for the elderly who completed chemotherapy, 25% for those treated with incomplete or inferior chemotherapy regimens and 20% for those receiving palliative radiotherapy or supportive care. In a multivariate analysis completing therapy and not age was significantly associated with a better survival (p<0.001).

Over a period of six years the majority of patients in Friesland with a DLBCL and aged ≥ 75 years was treated with a R-CHOP and 56% could complete this treatment. The ability to complete such an intensive treatment was significantly associated with a better 2-years survival (67%) compared to suboptimal chemotherapy (25%) or either palliative radiotherapy or supportive care (20%). Severe treatment-related toxicity was also observed in 66% patients, with 13% toxicity-related deaths. Age only is a poor indicator to differentiate the fit elderly patient, who will benefit from intensive therapy, from the frail. Other strategies for patient selection, such as the implication of a comprehensive geriatric assessment, should be further evaluated in very elderly patients with DLBCL.

No relevant conflicts of interest to declare.