C-met, a new prognostic biomarker in glioblastoma multiforme.

Rikke Hedegaard Dahlrot, Stine Asferg Petterson, Simon Kjaer Hermansen, Bjarne Winther Kristensen, Steinbjorn Hansen

Publikation: Bidrag til tidsskriftTidsskriftsartikelForskningpeer review

Abstract

2088Background: C-met is a tyrosine kinase receptor involved in growth, invasiveness and malignant progression in glioblastoma multiforme (GBM). Activation of the C-met pathway increases resistance towards DNA damage in glioma cell lines and it has been shown that an orally administrated C-met kinase inhibitor inhibits intracranial glioma growth in immunodeficient mice, suggesting that C-met is a potential target in glioma treatment. The purpose of the present study was to investigate the prognostic value of C-met in GBMs and subsequently correlate the prognostic value to known clinical variables, identified in a population-based cohort. Methods: Tissue samples from 186 GBM patients were analyzed using immunohistochemical staining and advanced quantitative image analysis. This provided continuous measurements based on staining intensity. Results: Median intensity was 70.8 (range 15.5-200.1). When divided at the median C-met was not prognostic. Further exploration showed that dichotomizing at an intensity of 75 (60% vs. 40%), high levels of C-met were associated with poor survival. However; C-met is a time-dependent factor and no prognostic effect was observed within the first 8.5 months after diagnosis (HR 0.97, 95% CI 0.62-1.51, p = 0.892). After 8.5 months, patients with high levels of C-met had a significantly poorer survival as compared to patients with low levels (HR 2.06, 95% CI 1.33-3.18, p = 0.001). This was significant in multivariate analysis adjusted for clinical variables; C-met (HR 1.89, 95% CI 1.22-2.93, p = 0.004), age (HR 1.02, 95% CI 1.00-1.03, p = 0.037), performance status (HR1.55, 95% CI 1.38-1.74, p = 0.004), and tumor crossing midline (HR 1.72, 955 CI 0.75-3.92, p = 0.201). Interestingly; C-met was only prognostic in patients who received post-surgical treatment, whereas no effect was observed in the 31 patients who underwent surgery only (HR 0.54, 95% CI 0.26-1.15, p = 0.113). Conclusions: Using advanced quantitative image analysis, we found that C-met was an independent prognostic factor of poor survival in GBMs. The effect did not appear within the first 8.5 months after the diagnosis and it was only seen in patients who received post-surgical treatment.
OriginalsprogDansk
TidsskriftJournal of Clinical Oncology
Vol/bind31
Udgave nummer15_suppl
Sider (fra-til)2088-2088
Antal sider1
ISSN0732-183X
DOI
StatusUdgivet - 20 maj 2013

Citationsformater