Primary central nervous system lymphoma (PCNSL) - retrospective review of 62 PMCI patients

Purpose: To assess outcome of patients receiving treatment for PCNSL in terms of response rate, patterns of failure and overall survival. Methods: The records of all patients presenting to PMCI between 1982 and 1994 with the diagnosis of cerebral lymphoma were obtained. All patients had a histologically confirmed diagnosis of PCNSL, no extra-CNS disease, and negative serology and/or no risk factors for HIV. Response was assessed by the Cotswolds criteria. All patients received dexamethasone and were planned to receive cranial irradiation. The median dose to the target volume was 50.4 Gy. Seven patients also received spinal axis radiation. Twenty-two patients received additional chemotherapy; 7 intrathecal, 9 systemic and 6 both. In 80% of cases the systemic chemotherapy was cyclophosphamide, doxorubicin, vincristine, prednisolone (CHOP), usually post radiotherapy. Only one patient was lost to follow-up. All other patients were followed until death, with a median follow-up for living patients of 5.4 years. Results: Sixty-two patients were identified. The median age was 60 years; males 53%; females 47%; 85% had WHO performance status ≥ 2 at presentation; all were B-cell lymphomas - 52% large cell histology; cerebrospinal fluid (CSF) status was known in 55% of which 29% had positive cytology. Sixty-one patients received cerebral irradiation. The objective response rate was 68% (CR 29%, CRu 15%, PR 24%). The estimated median survival was 20.6 months (95% confidence interval 12.4 - 33.4 months). Estimated two and five year survivals were 44% and 27% respectively. Progression/relapse included the CNS/CSF in all patients except 3 who had ocular relapses only. Known sites of CNS/CSF relapse/progression were: initial brain site only 14%, new brain site only 14%, both 10%, CSF alone 27%, CSF plus another site 14%, spinal cord alone 9%. Two patients had both CNS and extra-CNS relapse: 1 in spleen/mesenteric nodes, 1 in liver. On univariate analysis male gender, age less than 60 years, performance status < 2, treatment with ≥ 45 Gy to target volume and treatment with additional chemotherapy, were associated with a significantly better overall survival (p < 0.05). However on multivariate analysis only age and performance status remained significant prognostic variables. Estimated five year survival was 38% and 18% for age < 60 yrs and ≥60 yrs respectively, and 78% and 21% for performance status < 2 and ≥ 2 respectively. Conclusions: These results are comparable with other published series. Patient outcome is strongly influenced by age and performance status. No survival benefit with chemotherapy nor radiation dose to the target volume ≥ 45 Gy could be demonstrated in this series when adjusted for age and performance status. It is possible that studies suggesting better survival for patients treated with chemoradiation compared to radiation alone may reflect patient selection rather than treatment variables. Alternatively, the drugs used in this series may not be the most active for this disease, in contrast to extra- CNS non-Hodgkins lymphoma, or that post radiation chemotherapy may not represent optimal timing. All relapses, apart from the ocular, involved the central nervous system; thus current management is unsatisfactory for long term 'locoregional' control. The high incidence of CSF involvement both at presentation and relapse suggests that greater attention needs to be paid to this region at both diagnostic work-up and treatment