P16.04.B Palliative CSF diversion in symptomatic leptomeningeal metastasis: a feasibility analysis of risks and benefits

Abstract

Abstract Background Leptomeningeal metastasis (LM) represents a terminal condition in a subset of patients with primary extra-cranial malignancies. With improved survival rates under novel systemic therapies for extra-cranial primary tumors, the role of cerebrospinal fluid (CSF) diversion via ventriculo-peritoneal shunts (VP-shunt) for symptom control of hydrocephalic LM is becoming increasingly important. This study hence aimed to describe this patient cohort and weigh out benefits against adverse events of palliative VP-shunt placement. Material and Methods A single-center retrospective analysis of all consecutive adult patients with VP-shunt placement over a period of six years was performed and clinical data (primary disease, clinical presentation, CSF diagnostics, previous therapies) in addition to surgical data (shunt devices, surgeries, and complications) and survival data were collected. Results were compared to a ‘non-oncological’ shunt cohort and statistical analyses was performed. Results 38 patients with a median age of 53 (18-78) years (13 males, 25 females) underwent VP-Shunt placement. The most common underlying oncological conditions were breast cancer (n=21, 55%) and non-small cell lung cancer (NSCLC, n=11, 29%). The median time between primary tumor and LM diagnosis was 23 months (0 to 180 months). 14 patients (37%) presented with end-stage disease at primary tumor diagnosis, 25 (66%) patients had cranial metastases. Most patients presented with classical symptoms of intracranial hypertension and 11 patients (38%) had already received intrathecal treatment or radiotherapy before shunting. After shunting, symptom relief was achieved in 30 patients (79%) and 24 patients (63%) were eventually discharged home after surgery. Subsequently, 63% of the shunted patients underwent systemic or radiotherapy and 55% received intrathecal therapy. Revision surgery was performed for valve malfunctions and infections (n=3, 8% for either) or distal catheter malposition (n=2, 5%) but comparison with a ‘non-oncological’ shunt cohort showed no differences in complication rates. Median survival from shunting was 2.1 months (95% CI 0.5 to 3 months). Conclusion VP-shunt placement could relieve symptoms of intracranial hypertension secondary to LM of primary solid tumors without a higher risk of shunt complications compared to non-oncological patients, with many LM patients persuing further oncologic therapy. However, decision-making regarding VP-shunt placement in LM patients still retains a palliative nature.