Incidence and Risk Factors for Central Nervous System Relapse in Very Elderly Patients over 80 with Diffuse Large B-Cell Lymphoma: A Retrospective Analysis of Two Lysa Studies

Background. The prevalence of diffuse large B-cell lymphoma (DLBCL) in patients aged over 80 years is reported to have tripled compared to patients in their sixties. Treating very elderly patients is particularly challenging given the likelihood of comorbidities and concerns over risks of toxicity. One of the most devastating and rapidly fatal complications in DLBCL is central nervous system (CNS) relapse. Most studies reporting incidence and risk factors of CNS relapse concern DLBCL patients under the age of 80 years, and little is known about CNS recurrence in the very elderly, aged over 80 years. CNS prophylaxis is rarely implemented in this population due to the burden of comorbidities, frequent antiplatelet or anticoagulant treatment along with renal failure and hypoalbuminemia, as well as potential prophylaxis toxicity of IV high dose methotrexate and/or the invasiveness of IT therapy.

Aim. We retrospectively evaluated the incidence of CNS relapse, risk factors and specific survival in very elderly DLBCL patients aged 80 years

Patients and Methods. Data were collected retrospectively from two multicentre, open-label, single arm phase II LYSA trials (LNH 03-7B (NCT01087424) between 2004 and 2007, LNH09-7B (NCT01195714) between 2009 and 2013) evaluating the addition of rituximab or ofatumumab to miniCHOP as front-line therapy.

Results. A total of 270 elderly patients were included in the two trials, 150 treated with R-miniCHOP and 120 with O-miniCHOP. Median age was 83 years (range 79-95) and none received CNS prophylaxis. At inclusion, most patients (76%) presented with disseminated disease (Ann Arbor stage III or IV), and 37% had at least two extra-nodal sites. Overall 18% of patients had bone marrow involvement, while renal, adrenal, testis or cavum involvement was rare (4%, 2%, 3%, and 2%, respectively). After a median follow-up of 28.7 months (range 0.1-72.1), 8 (3%) cases of CNS relapse were reported. Median time between inclusion and CNS relapse was 19.2 months (range 3.2 - 32.6). Cumulative 1 and 2-year incidence of CNS relapse was 1.4% (95% CI 3.2-25.4) and 2.5% (95% CI 6.9-9.2), respectively. At inclusion, clinical characteristics of the CNS relapse patients were not significantly different from patients without relapse. At CNS relapse, all patients except one present a performance status ECOG > 2. neurological symptoms were either mild with loss of autonomy, asthenia, hearing impairment, urinary incontinence, or more prominent with delirium, aphasia, intracranial hypertension, or consciousness disorder. Treatment of CNS relapse was a supportive care based- treatment with corticosteroid only in 5 patients, radiation therapy in one patient, and chemotherapy for 2 patients including rituximab-temodal (5 cycles) in one patient and rituximab -aracytine-vepeside (2 cycles) for the other one. Patients survived for a median of 1.5 months after CNS relapse diagnosis (range, 0.4 to 4). The CNS international prognostic index (CNS-IPI) classified 33 (12%) patients in the low-risk group, 164 (62%) in the intermediate-risk group, and 71 (26%) in the high-risk group. The low-risk and intermediate-risk CNS-IPI groups showed 2-year rate of CNS disease of 3%, and the high-risk of 2.8% (p=0.9483). Conclusion. Incidence of CNS relapse in very elderly previously untreated patients is approximately 3% and is associated with a very poor survival. The absence of prophylaxis in this population did not appear to have a strong impact on CNS relapse incidence. Higher number of patients is warranted to identify risk factors for CNS relapse in this population