Introduction: A hallmark of cancerous cells is an ability to maintain the ends of their chromosomes using a telomere maintenance mechanism, telomerase activity or the alternative lengthening of telomeres (ALT). In glioblastoma those with ALT positive tumors have an increased survival compared to those with non-ALT tumors. However, this is not the case for all with some ALT positive tumors associated with poorer survival. To better predict survival we investigated gene expression differences in ALT positive glioblastoma associated with poor patient survival to better understand the underlying mechanism
Methods: RNAseq data from ALT glioblastomas associated with poor patient outcome were analysed to identify gene signatures. Markers associated with gene expression signatures of interest were confirmed in a cohort of 129 glioblastomas collected in New Zealand. Glioblastomas were selected to include all ALT positive tumors (n=36) and a subset of non-ALT tumors (n=93) for comparison. Markers were confirmed using immunohistochemistry on paraffin embedded tumor sections and realtime PCR using frozen tumor samples. Gene signatures were correlated with patient survival.
Results: ALT positive tumors associated with poor patient survival had an activated microglia gene signature by RNAseq. In a larger cohort, 38% percent of ALT positive tumors had an activated microglia gene signature by real time PCR and an increased percentage of activated microglia by immunohistochemistry (CD163 and P2RY12 double positive cells). An activated microglia gene signature was absent in non-ALT tumors. In the ALT positive tumor group patients with activated microglia had a reduced median survival compared to patients without activated microglia (7.7 versus 18 months, P < 0.0001). Only one patient (8%, 1/13) with an activated microglia tumor survived post 20 months compared to eleven without activated microglia (48%, 11/23).
Conclusions: Activated microgila in ALT positive tumours are associated with poorer patient outcome.
Patient Care: Identifying patients with poor prognosis would help understanding tumour behaviour and targeting treatment accordingly.
Learning Objectives: Identifying GBMs with poor prognosis.