Introduction: Central nervous system (CNS) relapse in large B cell lymphoma (LBCL) is a rare, but catastrophic event associated with poor prognosis without an effective standard prophylactic strategy. Intravenous and/or intrathecal chemotherapy have been proposed for patients at high risk for CNS relapse (i.e., CNS-IPI of 4 or greater) with mixed results due to the lack of prospective trials comparing different approaches. Large retrospective cohort studies have challenged the value of CNS prophylaxis. We performed a systematic review and meta-analysis (MA) assessing the role of chemoprophylaxis in preventing CNS relapse for LBCL patients treated with chemoimmunotherapy.

Methods: A systematic literature review was conducted using the major online databases, including English language studies published from 2011 until 2022. Data were pooled from the studies with a comparator group using the random effects model to report odds ratio (OR) and 95% confidence intervals (95% CI). Data from single-arm studies were pooled using the random effects model to report the pooled proportion with 95% CIs.

Results: There were 16 studies included in the MA (12 retrospective and 4 prospective) assessing the role of CNS prophylaxis in a total of 11,530 patients. The incidence of CNS relapse appeared to be higher among patients who received CNS chemoprophylaxis (n=2108) compared with patients without chemoprophylaxis(n=9422) (OR 1.70, 95% CI 1.10-2.61) (figure 1). Among the single-arm studies, the pooled proportion of the incidence of CNS relapse, isolated CNS relapse and secondary CNS relapse was 5% (95% CI 4%-7%), 5% (95% CI 4%-7%), and 1% (95% CI 0%-2%), respectively. Results are confounded by the retrospective design of most studies, different types of chemoprophylaxis and heterogeneous patient population.

Discussion: This systematic review and MA demonstrated consistent rates of CNS relapse as seen in prior large cohorts, and did not show evidence for the use of chemoprophylaxis in patients with LBCL to prevent CNS relapse. Given the scarcity of high quality data due to the absence of prospective trials and the lack of motivation by multiple stakeholders to develop these type of studies, the totality of evidence grouped in this MA constitutes one of the most robust datasets arguing against chemoprophylaxis for prevention of CNS relapse in patients with LBCL.

Chavez:GenMab: Consultancy; Adicet: Consultancy; Janssen: Research Funding; ADC Therapeutics: Research Funding; Kite Pharma: Consultancy; Abbvie: Consultancy; Beigene: Honoraria; Astrazeneca: Research Funding, Speakers Bureau; Merck: Research Funding; TG Therapeutics: Honoraria; MorphoSys/Incyte: Speakers Bureau; Epizyme: Honoraria, Speakers Bureau.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution