Intensive chemotherapy and immunotherapy in patients with newly diagnosed primary CNS lymphoma: CALGB 50202 (Alliance 50202)

Abstract

Purpose: Concerns regarding neurocognitive toxicity of whole-brain radiotherapy (WBRT) have motivated development of alternative, dose-intensive chemotherapeutic strategies as consolidation in primary CNS lymphoma (PCNSL). We performed a multicenter study of high-dose consolidation, without WBRT, in PCNSL. Objectives were to determine: one, rate of complete response (CR) after remission induction therapy with methotrexate, temozolomide, and rituximab (MT-R); two, feasibility of a two-step approach using high-dose consolidation with etoposide plus cytarabine (EA); three, progression-free survival (PFS); and four, correlation between clinical and molecular prognostic factors and outcome.

Patients and methods: Forty-four patients with newly diagnosed PCNSL were treated with induction MT-R, and patients who achieved CR received EA consolidation. We performed a prospective analysis of molecular prognostic biomarkers in PCNSL in the setting of a clinical trial.

Results: The rate of CR to MT-R was 66%. The overall 2-year PFS was 0.57, with median follow-up of 4.9 years. The 2-year time to progression was 0.59, and for patients who completed consolidation, it was 0.77. Patients age > 60 years did as well as younger patients, and the most significant clinical prognostic variable was treatment delay. High BCL6 expression correlated with shorter survival.

Conclusion: CALGB 50202 demonstrates for the first time to our knowledge that dose-intensive consolidation for PCNSL is feasible in the multicenter setting and yields rates of PFS and OS at least comparable to those of regimens involving WBRT. On the basis of these encouraging results, an intergroup study has been activated comparing EA consolidation with myeloablative chemotherapy in this randomized trial in PCNSL, in which neither arm involves WBRT.

Conflict of interest statement

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

Figures

Fig 1.

Protocol schema. Patients were restaged after 4 months of high-dose methotrexate-based therapy (seven doses of high-dose methotrexate, every 2 weeks; six doses of weekly rituximab; and 4 months of temozolomide over 5 days [MT-R]). Patients who achieved a complete response (CR) or CR/unconfirmed (CRu) received an additional course of high-dose methotrexate plus one of temozolomide. Three to 5 weeks later, patients received intensive consolidation with etoposide plus cytarabine (EA). High-dose EA chemotherapy doses were based on corrected body weight (kg), defined as ideal weight plus 0.25 (actual weight − ideal weight), as described previously. IV, intravenous; PD, progressive disease; PO, orally; PR, partial response; SD, stable disease.

Fig 2.

Outcome for all 50202 study patients; y-axis refers to cumulative probability of event. (A) Time to progression (TTP) for all patients; median TTP was 4.0 years (22 patients experienced disease progression). Estimated TTPs at 1, 2, 3, and 4 years were 0.66 (95% CI, 0.50 to 0.78), 0.59 (95% CI, 0.43 to 0.72), 0.52 (95% CI, 0.36 to 0.65), and 0.48 (95% CI, 0.33 to 0.63), respectively. (B) TTP for those patients (n = 27) who completed entire treatment protocol (induction plus consolidation). One- and 2-year probabilities of TTP from start of etoposide plus cytarabine consolidation were 0.85 (95% CI, 0.64 to 0.94) and 0.69 (95% CI, 0.47 to 0.83), respectively. (C) Overall survival (OS) for all patients; median OS has not been reached. Estimated OS at 1, 2, 3, and 4 years were 0.75 (95% CI, 0.59 to 0.85), 0.70 (95% CI, 0.52 to 0.80), 0.70 (95% CI, 0.52 to 0.80), and 0.65 (95% CI, 0.49 to 0.77), respectively. TTP is defined as time from date of study entry until progression or date of last follow-up while in remission, with censoring of deaths not resulting from progressive lymphoma. OS is defined as time from date of study entry until death resulting from any cause or date of last follow-up while in remission.

Fig 3.

Clinical prognostic variables and their relationship to progression-free survival (PFS); median PFS survival was 2.4 years (22 patients who experienced disease progression plus two patients achieving complete response who succumbed to sepsis and lung cancer, respectively). Estimated PFS at 1, 2, 3, and 4 years were 0.64 (95% CI, 0.48 to 0.76), 0.57 (95% CI, 0.41 to 0.70), 0.50 (95% CI, 0.34 to 0.64), and 0.47 (95% CI, 0.32 to 0.61; not shown). (A) PFS was similar for patients age > 60 years (n = 23) and for younger patients (n = 21; P = .48). (B) PFS was shorter for patients with Eastern Cooperative Oncology Group performance status of 2 (n = 8; P < .06). (C) There was a trend between shorter PFS and highest International Extranodal Lymphoma Study Group risk score of 4 to 5 (P = .16). (D) Treatment delay was associated with shorter PFS. Patients with delayed initiation of remission induction therapy, beyond 30 days after diagnosis, experienced significantly shorter PFS compared with patients whose therapy began within 1 month of diagnosis (P = .050). Three-year PFS was 0.59 (95% CI, 0.40 to 0.73) for those without treatment delay and 0.2 (95% CI, 0.03 to 0.47) for those with treatment delay. PFS is defined as time from date of study entry until progression, death resulting from any cause, or date of last follow-up while in remission. There was no association between malignant CSF cytology at pretreatment staging and response rate or outcome.

Fig 4.

BCL6 expression is associated with short time to progression (TTP) and overall survival (OS) in patients with primary CNS lymphoma (PCNSL) treated in the 50202 study. (A) Example of strong nuclear BCL6 expression in a PCNSL case from patient treated in study (40× magnification). (B) High BCL6 expression (ie, > 60% of lymphoma nuclei) is associated with short TTP (two-sided P < .016). (C) High BCL6 is also associated with shorter OS (two-sided P < .009). For BCL6, monoclonal antibody Pg-B6p (Dako, Carpenteria, CA) was used. For MYC (not shown), monoclonal antibody Y69 (Epitomics, Burlingame, CA) was used. An automated immunostainer with iView diaminobenzidine detection (Ventana Medical Systems, Oro Valley, AZ) was used with CC1 heat-induced epitope retrieval for both assays.