Test stratifies patients based on likely response to standard of care
Though rare, glioblastoma multiforme (GBM) is the most common and most aggressive primary form of brain cancer. GBM tumors arise from astrocytes—star-shaped cells that comprise the supportive tissue of the brain. These tumors are typically malignant since the cells reproduce quickly and are nourished by a large network of blood vessels.
GBM can be extremely difficult to treat because the tumors contain many different types of cells. Some cells may respond well to certain therapies, while others may not. For most patients, the goal is to first relieve pressure on the brain and then remove as much tumor as possible through surgery—a complicated procedure due to the finger-like tentacles of a glioblastoma. Radiation and chemotherapy are also used to slow the growth of tumors that cannot be removed.
DecisionDx™-GBM is a robust multivariate diagnostic test that determines the molecular signature of a GBM tumor. The assay probes tumor tissue for expression of nine specific genes that have been shown in studies to be highly associated with outcome in GBM patients treated with standard of care (radiation with concurrent and adjuvant temozolomide). Knowing whether these genes are over- or under-expressed, and in what combination, can give physicians a glimpse into how a tumor might behave in response to standard therapy alone.
DecisionDx-GBM was discovered at The University of Texas M.D. Anderson Cancer Center and licensed to Castle Biosciences for clinical and research purposes. Castle has performed and completed CLIA development and validation through its licensed laboratory.
It is known that some GBM tumors exhibit a good, sustained response to first line treatment while others are refractory. However, traditional histopathologic methods used to diagnose GBM are unable to differentiate between these two tumor types.
The DecisionDx-GBM assay was developed to fulfill the need for prospective risk stratification, which may allow more individualized treatment options to be considered early on in diagnosis. With an accurate picture of the tumor’s likely response to standard of care, patients and their physicians can make more informed treatment decisions.
Castle’s most recent clinical update found an overall median survival of 71 weeks for the entire GBM population. However, significant differentiation in progression-free survival and overall survival occur between the long-term responder group and the short-term or refractory responder group (median overall survival: 377 weeks vs 56 weeks; p<0.0001).
Through multivariate analysis, DecisionDx-GBM has been shown to be independent of age, performance score (p=0.0003), and MGMT methylation (p=0.0055). In the same analysis, MGMT methylation was not found to be independent of DecisionDx-GBM (p=0.0602).