M Balm1, J Hammack. 1. Department of Neurology, Mayo Clinic, Rochester, Minn, USA.
Abstract
OBJECTIVE: To determine factors that are predictive of survival among patients with leptomeningeal carcinomatosis. BACKGROUND: Studies of potential prognostic factors in leptomeningeal carcinomatosis have produced conflicting results. Reasons for the discrepancies may be methodological differences in case ascertainment, treatment protocols, and limitations due to the size of the study group. DESIGN AND METHODS: We reviewed the medical records of 126 patients with cytologically confirmed leptomeningeal carcinomatosis seen at the Mayo Clinic in Rochester, Minn, from 1983 to 1994. Clinical, radiographic, and cerebrospinal fluid (CSF) parameters at the time of presentation are summarized. Treatment response, complications, and cause of death are also discussed. Using the forward stepwise Cox model, independent predictors of survival were identified. RESULTS: Independent negative predictors of survival include elevated CSF protein (P < .001) and clinical involvement of the cerebral leptomeninges (P = .05). Independent positive predictors of survival were longer duration of neurological symptoms at the time of presentation (P < .005), treatment with intrathecal or intraventricular (intra-CSF) chemotherapy (P = .01), and female sex (P = .02). Other variables, including age, primary tumor type, and extent of systemic disease, were not predictive. CONCLUSIONS: We conclude that female sex, longer duration of neurological symptoms, absence of cerebral leptomeningeal clinical involvement, and absence of elevated CSF protein independently predict better survival. Patients treated with intra-CSF chemotherapy also survived longer.
OBJECTIVE: To determine factors that are predictive of survival among patients with leptomeningeal carcinomatosis. BACKGROUND: Studies of potential prognostic factors in leptomeningeal carcinomatosis have produced conflicting results. Reasons for the discrepancies may be methodological differences in case ascertainment, treatment protocols, and limitations due to the size of the study group. DESIGN AND METHODS: We reviewed the medical records of 126 patients with cytologically confirmed leptomeningeal carcinomatosis seen at the Mayo Clinic in Rochester, Minn, from 1983 to 1994. Clinical, radiographic, and cerebrospinal fluid (CSF) parameters at the time of presentation are summarized. Treatment response, complications, and cause of death are also discussed. Using the forward stepwise Cox model, independent predictors of survival were identified. RESULTS: Independent negative predictors of survival include elevated CSF protein (P < .001) and clinical involvement of the cerebral leptomeninges (P = .05). Independent positive predictors of survival were longer duration of neurological symptoms at the time of presentation (P < .005), treatment with intrathecal or intraventricular (intra-CSF) chemotherapy (P = .01), and female sex (P = .02). Other variables, including age, primary tumor type, and extent of systemic disease, were not predictive. CONCLUSIONS: We conclude that female sex, longer duration of neurological symptoms, absence of cerebral leptomeningeal clinical involvement, and absence of elevated CSF protein independently predict better survival. Patients treated with intra-CSF chemotherapy also survived longer.