Background The current treatment management for primary central nervous system lymphoma (PCNSL) relies on high-dose methotrexate (HD-MTX)-based regimens, followed by consolidation therapy (1-3), including whole-brain radiotherapy (WBRT) (4). However, concerns persist regarding WBRT-related neurotoxicity and cognitive impairment, particularly in older patients. Whether dose-reduced WBRT in patients achieving complete remission (CR) can mitigate these risks, or whether WBRT alone is sufficient for patients ineligible for systemic therapy, remains uncertain.

Patients and methods Consecutive patients with biopsy-proven PCNSL who received cranial irradiation between January 2000 and December 2022 were identified. Dates of symptom onset, diagnosis, first hospital visit, first chemotherapy cycle, radiotherapy initiation, relapse or progression, death, and last follow-up were recorded to compute overall response rate (ORR), progression-free survival (PFS) and OS. Patients completed Minimum Mental State Examination (MMSE) evaluation. Survival analyses were performed using Kaplan-Meier method.

Results Of 244 biopsy-proven PCNSL patients, 62 who received brain irradiation were screened. The median age was 64 years (range 28-81), with 32.26% (20/62) were ≥60 years old. Males comprised 56.45% (35/62). Disease was unifocal in 41.94% (26/62) and multifocal in 58.06% (36/62). Deep brain involvement was observed in 54.84% (34/62) of patients, with the most common tumor location of frontal lobe (28/62, 25.69%).

Resection was performed in 61.29% (38/62), and stereotactic biopsy in 38.71% (24/62). First-line chemotherapy was administered to 82.26% (51/62) patients. 29 patients (46.77%) received lumbar puncture and MTX or Ara-c plus dexamethasone were used as intrathecal chemotherapy agents.

WBRT was used for consolidative intention in 53.23% (33/62) of patients, as palliative treatment after surgery/biopsy in 32.26% (20/62), and palliative treatment after progression of chemotherapy in 14.52% (9/62). Median overall treatment time was 17 days (range 13-22). Pre-WBRT disease status was CR in 32 patients (51.61%), PR in 6 patients (9.68%), and sable or progressive disease (SD/PD) in 24 patients (38.71%). After WBRT, 45 patients (72.58%) achieved CR and 17 patients (27.42%) still had residual tumor. WBRT doses were 23.4Gy (n=21, 33.87%), 30Gy (n=20, 32.26%), 36Gy (n=2, 3.22%), 40Gy (n=7, 11.29%), 50Gy (n=7, 11.29%), <20Gy (n=5, 8.06%), respectively. A boost to the tumour bed was delived in 18 patients (29.03%). 7 patients (11.29%) received oral lenalidomide (10mg) as maintenance after WBRT. A total of 42 patients (≤30Gy) completed MMSE test at 6-month intervals after completion of WBRT. The mean MMSE scores stratified by educational attainment were 23 (illiterate), 24 (primary school), and 26.5 (secondary education or higher), indicating preserved cognition across literacy levels.

The median follow-up was 77.8 months, with a mean time between initial symptoms occur and the date of first visit to hospital of 27.4 days (range 0-183). The relapse or disease progression occurred in 41.94% (26/62) of patients, while 48.39% (30/62) patients died. The median OS was 77.03 months (95% CI, 52.01-102.05), the 1-year OS rate was 85.2%, the 2-year OS rate was 71.6%, and the 5-year OS was 58.6%.

In univariate analysis, the variables associated with increased OS were receiving systemic chemotherapy (p=0.048), consolidiative WBRT after chemotherapy (p=0.037), no residual disease before WBRT (p=0.001), CR after WBRT (p=0.001) and intrathecal chemotherapy (p=0.008). Gender (p=0.375), age>50 years (p = 0.086), the ECOG (p=0.794), elevated LDH level (p=0.653), elevated protein level of CSF (p=0.508), deep brain involvement (p=0.101), multifocal disease (p=0.364), non-GCB phenotype (p=0.186), the radiotherapy dose of the whole brain less than 30Gy (p=0.149), a boost to the tumour bed (p=0.589), IELSG score (p=0.537), MSKCC score (p=0.193), and receiving salvage chemotherapy (p=0.777) did not show statistically significant association with OS.

Conclusion OS was superior when WBRT was given as consolidation after systemic therapy and when pre-WBRT CR was achieved. Low-dose WBRT (≤30Gy) could not negatively affect OS, with no evidence of significant cognitive decline, especially in older patients. Prospective trials evaluating reduced-dose WBRT with neurocognitive endpoints are warranted.

This content is only available as a PDF.
2025
Sign in via your Institution