Evidence-based review of intrathecal chemotherapy in outcome of patients with leptomeningeal metastasis

Abstract

1529 Background: Intrathecal chemotherapy (ITC) is commonly employed to treat leptomeningeal metastasis (LM). An evidence-based literature review supporting this practice is lacking. We therefore searched the literature for evidence of efficacy of ITC on LM patient outcome in order to provide evidence-based treatment recommendations. Methods: Computer-assisted searches of the Medline and Cochrane Library databases were performed. Only prospective, randomized, controlled trials in which ITC chemotherapy was used to treat LM were selected. Study design, patient characteristics, interventions, and outcome measures were independently compiled by each of the authors using customized data extraction forms. Each article was assigned a class of evidence by each of the authors. Disagreements were resolved by consensus. Results: Five articles met the criteria. Four studies compared different ITC regimens. One compared ITC to controls who did not receive ITC. Study populations included solid tumors (3 [2 of these included lymphoma]), lymphoma only (1), breast cancer only (1). Positive CSF cytology was a requirement for study entry in three trials. In each trial patients were allowed to receive CNS radiation as needed and non-investigational chemotherapy to treat the underlying cancer. Outcome measures included neurological status (5), survival (5), toxicity (5), CSF cytology (4), cause of death (3), time to neurologic progression (3), and quality of life (2). All studies lacked masked outcome assessment, and some suffered from additional design flaws including insufficient size, inequalities in baseline characteristics, inadequate characterization of treatment arms, and poor choice or definition of primary endpoints. Consequently, all studies provided only Class III evidence. Conclusions: There is no Class I or II evidence that ITC improves outcome in patients with LM. Improved quality of life, survival, or time to neurologic progression in treatment responders vs. non-responders in some of these trials provides a weak suggestion that effective therapy improves clinical outcome. Well-designed, adequately powered studies are urgently needed to determine efficacy of ITC in LM. No significant financial relationships to disclose.