Summaries of Must-Read Clinical Literature, Guidelines, and FDA Actions
Model Stratifies Outcomes in Glioblastoma Patients
JAMA Oncol; ePub 2017 Jan 12; Bell, Pugh, et al
A model that uses recursive partitioning analysis (RPA) in patients with glioblastoma treated with radiation and temozolomide improves outcome stratification over existing models, according to a study involving 452 individuals.
Specimens from the RPA model were analyzed for protein biomarkers representing key pathways in glioblastoma. Investigators looked at overall survival, and checked results in an independent data set (n=176).
In an analysis of all specimens, MGMT, survivin, c-Met, pmTOR, and Ki-67 protein levels were significant on single-marker multivariate analysis of overall survival.
Significant protein biomarkers and certain clinical variables were incorporated into a new model and tested a subset of 166 patients. Among the results:
- Higher MGMT protein level was significantly linked with decreased MGMT promoter methylation and vice versa.
- MGMT protein expression was able to predict overall survival better than MGMT promoter methylation.
- Prognostic significance was confirmed in the independent data set.
Bell E, Pugh S, McElroy J, et al. Molecular-based recursive partitioning analysis model for glioblastoma in the temozolomide era: A correlative analysis based on NRG oncology RTOG 0525. [Published online ahead of print January 12, 2017]. JAMA Oncol. doi:10.1001/jamaoncol.2016.6020.