Abstract 301


While whole brain radiotherapy (WBRT) has long been considered the standard consolidative therapy in primary CNS lymphoma (PCNSL), concerns regarding irreversible neurocognitive effects of brain irradiation have prompted development of dose-intensive induction and consolidative chemotherapeutic approaches, with the aim to eliminate brain irradiation. We present the updated results of the first Phase II multicenter trial, conducted by a major cooperative group within the United States, to evaluate an intensive chemotherapy-alone strategy in newly diagnosed patients with PCNSL.


Induction chemotherapy consisted of methotrexate (MTX), temozolomide, rituximab (MT-R) with HD-MTX (8 gm/m2) administered every 2 weeks × 8, weekly rituximab × 6 and temozolomide (150 mg/m2) starting day +7 and continued monthly × 5. Patients who achieved a CR received intensive consolidation cytarabine 2 gm/m2 BID on days 1–4 with etoposide 40 mg/kg over 96 h (EA). Assessment of BCL6 and MYC expression by lymphoma cells was performed by immunohistochemical analysis of diagnostic specimens and scored (10% increments) by a pathologist who was blinded to clinical outcome.


44 newly diagnosed, immunocompetent patients with PCNSL were treated at 12 CALGB centers between 2005 and 2009. Patient characteristics were as follows: median age was 61 yr (range 12–76), median ECOG PS was 1, 58% were IELSG risk group 2–3. 98% of tumors were large B-cell lymphoma. 66% of patients exhibited CR to induction MT-R. With median overall follow-up of 5.3 years, 21 out of 46 patients exhibited disease progression and 17 have died. There was one treatment-related death (sepsis) during consolidation. There has been no evidence for significant treatment-related neurotoxicity. The median progression-free survival (PFS) is 4.0 years and estimated PFS rates with 95% confidence limits at 1, 2, 3 and 4 years are 0.66 (0.50, 0.76), 0.59 (0.43, 0.72), 0.52 (0.36, 0.64) and 0.47 (0.32, 0.61). The probability of 2-year PFS for patients who completed the entire regimen is 0.69 (0.47, 0.94). The estimated overall survival (OS) rate with 95% confidence limits at 4 years is 0.65 (0.49,0.77). Remarkably, event-free survival (EFS) was similar in patients older and younger than 60 (p< 0.47). While ECOG PS>1 and high IELSG score showed a trend toward inferior EFS, the most significant clinical prognostic variable was treatment delay: those patients who started remission induction therapy more than 30 days after diagnosis exhibited shorter EFS than patients who received MT-R within a month (2-sided p-value = 0.05). There was no relationship between treatment delay and IELSG prognostic score. While high MYC expression (>50% lymphoma nuclei) was detected in 54% of cases, MYC was not prognostic. By contrast, high BCL6 expression (≥30% of lymphoma nuclei) was detected in 59% of cases and correlated as a continuous variable with inferior progression-free survival, event-free survival and overall survival. The 2-sided p-values for these models were p=0.045, p=0.019 and p=0.045 (log-rank test).


CALGB 50202 (Alliance) demonstrates that induction MT-R followed by EA consolidation is feasible in the multicenter setting and yields rates of PFS and OS in newly diagnosed PCNSL patients that are at least comparable to combined modality treatment involving reduced dose whole brain irradiation. The MT-R-EA regimen is well-tolerated in patients age >60 and has similar efficacy in this population as in younger patients. Based upon these encouraging results, a successor, intergroup, randomized phase II trial, CALGB 51101 (Alliance) has been activated that compares dose-intensive consolidation (EA) with myeloablative chemotherapy and autologous stem cell transplant in this first randomized trial in PCNSL in which neither arm involves WBRT. Our observations regarding the association of treatment delay with adverse prognosis suggest that prompt initiation of therapy for PCNSL patients may translate into improved outcomes. In this first prospective analysis of molecular biomarkers in PCNSL in the setting of a clinical trial, high BCL6 expression was found to correlate with inferior outcome. This observation raises the possibility that in future studies BCL6 could be used as a biomarker in risk-adapted therapy and supports the investigation of BCL6 antagonists in the treatment of this disease. Supported by LLS and by NCI CA13908301.


No relevant conflicts of interest to declare.

This icon denotes a clinically relevant abstract

Author notes


Asterisk with author names denotes non-ASH members.