Meningeal carcinomatosis in patients with breast cancer: clinical features and prognostic factors in 91 patients from a single institution.

Abstract

Abstract Abstract #1090 Background
 Breast cancer is the leading solid tumor responsible for meningeal carcinomatosis (MC) and the incidence is rising. To date, there is no standard of care and few prospective trials. Intrathecal treatment is provided but the modalities remain controversial.
 Methods
 From our data base we recorded 91 patients with breast MC admitted to Institut Curie between 2000-2007. Lumbar punction was performed in all patients for cerebrospinal fluid (CSF) cytology, protein level and dosage of CYFRA 21. Intra-thecal treatment: methotrexate 15 mg day1 to 5 and hydrocortisone acetate day 1 with oral folinic acid rescue day 1 to 5 every two weeks.
 Results
 Median age was 53 years (range 30-78). Tumor type was ductal carcinoma (63%), lobular (28%), ER positive 74%, HER2 overexpressed 10% and triple negative 31%. All patients presented other metastatic site: bone (67%), liver (42%) lung (29%), brain (38%). The median interval between primary tumor and diagnosis of MC was 60 months.
 Amongst the 80 patients treated (87%) with intra-thecal MTX, clinical response was observed in 88% of cases. There was a strong correlation between clinical response and survival (p<0.001). Normalization of cerebrospinal fluid cytology (CSFC) was observed in 13 % of patients and was associated with better survival: 13 months versus 2 months (p=0.003). Improvement of other biologic parameters as CSF protein and CYFRA 21 was observed in 17% and 80% of patients respectively. The median survival was 4.5 months (0-53). In multivariate analysis, prognostic factors were: performance status 3-4 (p<0.0001), more than 3 chemotherapy regimens before MC diagnosis (p=0.04), lack of ER (p=0.04) and high CYFRA 21 at diagnosis (p=0.01). With these factors a prognostic score may select patients with a true benefit to IT MTX (20% survival > 1 year) or patients for whom treatment will have no impact on survival (< 3 months).
 Conclusion: This score has to be confirmed in prospective study and new modalities of treatments need to be evaluated according to the biology of disease. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 1090.