Radiotherapy (WBRT) or Autologous Stem-Cell Transplantation (ASCT) for Primary CNS Lymphoma in Patients 60 Years of Age and Younger: Long-Term Results of the Intergroup Anocef-Goelams Randomized Phase II Precis Study

Introduction: We previously reported the results of the PRECIS trial with a median follow-up of 33 months. Both whole brain radiotherapy (WBRT) and autologous stem cell transplantation (ACST) were effective according to the predetermined threshold. However, more relapses occurred in the WBRT arm. The 2-year event-free survival (EFS) from consolidation (relapse or death defined as event) were 69% (95% CI, 57% to 83%) and 87% (95% CI, 77% to 98%) after WBRT and ASCT, respectively (p = 0.03). Overall survival (OS) was similar in both arms. Cognitive impairment was observed after WBRT, whereas cognitive functions were preserved or improved after ASCT. A longer follow-up is required to better assess the impact of the treatment on relapse, survival and late complications. We report here the results of the PRECIS trial with a median follow-up of 98.3 months [min= 4,1 - max= 131.1], focusing on the per protocol population from time of consolidation.

Methods: Immunocompetent patients (18 to 60 years of age) with untreated primary CNS lymphoma (PCNSL) were randomly assigned upfront to receive WBRT (Arm A) or ASCT (Arm B) as consolidation treatment after an induction chemotherapy consisting of two cycles of R-MBVP (rituximab, methotrexate, VP16, BCNU, prednisone) followed by two cycles of R-AraC (rituximab, cytarabine). Intensive chemotherapy consisted of thiotepa (250 mg/m2/d D9; D8; D7), busulfan (8 mg/kg D6 through D4), and cyclophosphamide (60 mg/kg/d D3; D2). WBRT delivered 40 Gy (2 Gy/ fraction). Cognitive functions were prospectively assessed until disease progression and focused on global cognitive function, episodic verbal memory, attention and mental flexibility, and psychoaffective status.

Results: 140 patients were randomized (Arm A: N = 70; Arm B: N = 70). Fifty-three and 44 patients completed WBRT and ASCT respectively (per protocol population), including 3 and 5 patients who were in progressive disease (PD) at time of WBRT and ASCT, respectively.

8-y EFS from time of consolidation in the per protocol population was 75.9% [63.3-91.0] and 39.9% [26.8-59.3] after ASCT and WBRT, respectively (p = 0.007) (fig 1a). The risk of relapse was significantly decreased after ASCT compared to WBRT (8-y relapse-free interval 94.1% [86.4-100] vs 47.6% [34.2-66.3], (p <0.001) (fig 1b). The 8-year overall survival from time of consolidation was similar in both arms, 63.4% [49.8 - 80.6] and 69.3% [56.7-84.8] in the WBRT and ASCT arms, respectively (fig1c). Among the 24 patients who relapsed after WBRT, 13 patients received subsequent salvage chemotherapy and consolidative ASCT, and seven of these patients were disease-free at last follow-up.

Causes of deaths after WBRT (n = 17) were PD (n = 12), neurotoxicity (n = 3), second-line ASCT (n = 2). After ASCT, causes of deaths (n = 14) were treatment-related death (n = 5, including 2 occurring > 100 days post-ASCT, and 2 in patients in PD before ASCT), PD (n = 4), neurotoxicity following salvage WBRT (n = 1), second solid cancer (n = 3) and undetermined in one patient.

In multivariate analysis, ECOG, disease status at the end of induction, and protein level in the CSF at diagnosis were independent prognostic factors for OS. Disease status at the end of induction and intraocular involvement at diagnosis were independent prognostic factors for EFS.

Cognitive decline that could be fatal was only observed in patients who received WBRT. Imaging analysis of post consolidation leukoencephalopathy is ongoing.

Conclusions: Consolidation with ASCT after HD-MTX based induction chemotherapy resulted in an excellent disease control but with a higher treatment-related mortality than WBRT. Severe Cognitive decline and late treatment-induced neurotoxic deaths were observed after WBRT. Intensity of the thiotepa-busulfan-cyclophosphamide regimen used before ASCT should be slightly reduced to improve the benefit/risk ratio of ASCT in first-line treatment of young patients with PCNSL.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal