Analysis of Surveillance Imaging and Outcomes for Relapsed CNS Lymphoma Using Novel Agents

Introduction: Primary central nervous system lymphoma (PCNSL) is a rare entity with approximately 1,500 cases diagnosed per year in the United States. Despite effective frontline and consolidation therapy, up to 50% of patients with PCNSL will suffer disease relapse, which carries a poor prognosis. Relapse of systemic lymphoma in the CNS (secondary CNS lymphoma, SCNSL) is associated with a poor prognosis overall. Surveillance imaging is generally recommended following completion of primary therapy, although there is limited data to guide decisions regarding interval and duration of surveillance imaging following consolidation. Prior studies have suggested that asymptomatic relapses identified with surveillance imaging may confer a similar or better prognosis (Fossard, 2017). In addition, recent incorporation of novel therapies including Bruton's tyrosine kinase inhibitors (BTKi) like ibrutinib or immunomodulatory agents including lenalidomide or pomalidomide have improved outcomes, although data are limited by small sample sizes.

We hypothesized that the utility of surveillance imaging to detect relapses will differ according to post-consolidation timeline, and that the use of novel agents may improve outcomes compared to cytotoxic chemotherapy.

Methods: We performed a single-center retrospective analysis of patients with diagnosed CNS lymphoma treated at the University of Nebraska Medical Center (UNMC) between 2004 and 2022. Details regarding demographics, disease characteristics, treatment, and surveillance imaging were collected.

Results: There were 108 patients found to have either PCNSL or isolated SCNSL who were treated at UNMC in this period, 70 with PCNSL and 38 with SCNSL. The majority of included cases (83%) were treated between 2014-2022.

A total of 37 patients were found to have relapsed/refractory CNS disease, 34 of which underwent therapy. Out of 37 relapses, 19 occurred less than 1 year after the initiation of first-line therapy, 5 occurred between 1 and 2 years after the initiation of therapy, and 13 occurred greater than 2 years after the initiation of therapy. Of the 37 relapses, 11 (30%) were asymptomatic with relapse found on routine imaging. Six of these relapses were found within the first year, two were found between years one and two, and three were found between years two and three. On multivariate analysis, PCNSL was associated with a reduced risk of relapse compared to SCNSL (odds ratio = 0.4 P = 0.029). There was no statistical difference noted for performance status (ECOG 0-1 vs 2-4), methotrexate dose, age, or type of consolidation (transplant versus chemotherapy).

The efficacy of chemotherapy regimens versus novel agents (ibrutinib & lenalidomide) for relapsed PCNSL and SCNSL was also evaluated. The progression-free restricted mean survival time (RMST) was similar at 17.6 and 19.8 months for chemotherapy and novel agents (P = 0.65), respectively. The overall RMST was 19.0 months and 27.0 months, chemotherapy versus novel agents (P = 0.38), and not statistically significant. Fourteen patients who underwent second-line therapy had PCNSL, of which 5 were treated with chemotherapy and 9 were treated with novel agents (7 ibrutinib, 2 lenalidomide). 20 patients who underwent second-line therapy had SCNSL, of which 9 were treated with chemotherapy and 11 with novel agents (10 ibrutinib, 1 lenalidomide).

Conclusion: In this single center retrospective analysis, the majority of patients with PCNSL are alive and relapse-free following frontline therapy and consolidation therapy. The majority of relapses occurred within the first 2 years of surveillance, though many (13/37, 35%) occurred after 2 years. Asymptomatic relapses were detected by surveillance imaging in 11/37 (30%), though none were identified after 3 years. Further analysis with a larger cohort is necessary to definitively understand the role of surveillance imaging. Outcomes for relapsed PCNSL and SCNSL appear favorable compared to prior retrospective studies, and there was no difference observed between the use of traditional chemotherapy versus novel agents on survival outcomes. The limited sample size within this cohort precludes a definitive analysis, and a larger multicenter study is recommended to definitively evaluate the impact of ibrutinib, lenalidomide, and other novel agents on outcomes for relapsed PCNSL and SCNSL.

Lunning:Bristol Myers Squibb: Consultancy, Research Funding; Abbvie: Consultancy; Genmab: Consultancy. Vose:Pfizer: Research Funding; Novartis: Honoraria; GenMab: Honoraria, Research Funding; Abbvie: Honoraria, Research Funding.