Distribution and failure patterns of primary central nervous system lymphoma related to the hippocampus: Implications for hippocampal sparing irradiation.

Jooho Lee,Hye Jo Ryu

DOI: https://doi.org/10.1200/go.2023.9.supplement_1.20

2023-08-01

JCO Global Oncology

Abstract:20 Background: Hippocampus (HC) injury by conventional whole brain radiotherapy (C-WBRT) contributes to the neurocognitive decline in primary central nervous system lymphoma (PCNSL). Hippocampal avoidance (HA–WBRT) could minimize neurocognitive impairment by reducing the radiation dose to HC. However, its feasibility in PCNSL has not been examined regarding the incidence of HC involvement and failures. In this retrospective study, we assessed the risk of hippocampal area involvement at diagnosis and after treatments in PCNSL patients. Methods: We identified 278 immunocompetent PCNSL patients diagnosed between 2000 and 2020. After high dose methotrexate-based induction chemotherapy followed by observation or consolidation including RT, cytarabine alone, or autologous stem cell transplantation (ASCT). HC was contoured on T1 MRI image and expanded with a 5mm margin, generating hippocampal avoidance region (HAR). The extent of initial and recurrent lesions was evaluated using pre-induction and post-consolidation T1 contrast-enhanced MRI images. HC failure was defined as recurrence or progression at HAR and those who progressed after induction were excluded. The median follow up was 38.7 months (3.1-239.4). Results: Of 278 patients diagnosed with PCNSL, 39.9% of them had initial lesions at HAR. After induction therapy, 212 evaluable patients received consolidation as follows: RT (n=145, 68.4%) consisting of C-WBRT (n=114), HA-WBRT (n=23), and focal RT (n=8), observation (n=38, 17.9%), cytarabine only (24, 11.3%), and ASCT (n=5, 2.4%). Intracranial failures occurred in 47.6% (n=101) of patients, with 33.7% (n=34) of them in HAR. The multivariate analysis identified multifocal disease (HR 3.86, 95% CI 1.15-9.73, p=0.004) as the only factor associated with the risk of HC failure. Those with unifocal lesion outside HAR showed the lowest HC failure rate, 7.0%, while the highest HC failure rate, 25.4% was observed in the subgroup with multifocal disease within HAR at diagnosis. In the lowest risk group (unifocal lesion outside HAR, n=66), C-WBRT was not significantly associated with HC failure (HR 0.57, CI 0.09-3.33, P=.572) or intracranial failure (HR 0.88, CI 0.40-1.91, P=.748). Conclusions: Our data suggest the HA-WBRT could be explored in patients whose lesion is unifocal and located outside HAR. WBRT, including the hippocampus, did not significantly change HC failure for patients without initial HAR involvement. Further prospective study will be warranted to assess the feasibility of HA-WBRT in the subgroup with low risk of HC failure.