Search

BACKGROUND In WHO grade II diffuse gliomas (low-grade gliomas, hereafter called LGGs), chemotherapy and radiotherapy contribute to prolonged survival but could induce somatic mutations. The optimal timing of treatment in LGGs remain poorly understood. To delineate this, we designed a mathematical model for tumor growth and investigate the association among the treatment, malignant transformation (MT), and the accumulation of somatic mutations revealed by whole exome sequencing (WES) in LGGs. METHODS Totally, 290 patients with LGGs between 1990 and 2018 were analyzed. We assessed the statuses of IDH mutation and 1p19q co-deletion in all tumors. Among all, 114 patients (39%) underwent MT during follow-up periods (mean: 82.6 months). Tumor volume was evaluated with FLAIR and/or T2-weighted MRI. MT was evaluated with contrast-enhanced MRI and/or pathological diagnosis. To investigate the number of somatic mutations in a cohort of LGGs and their patient matched recurrence, WES was performed on 88 serial samples collected at least two time-points from 39 patients. RESULTS Oligodendroglioma, IDH-mutant and 1p/19q-codeleted (OD) showed longer transformation-free survival compared to other subtypes. An exponential model was chosen to estimate growth rate in LGGs, since the exponential model provided a better fit to our data as compared to a linear model. The growth rate significantly decreased in the middle of chemotherapy and after radiotherapy. By contrast, these treatments increased the number of somatic mutations identified by WES and the rate of MT in each subtype. The increasing number of mutations in recurrent tumors showed strong correlation with the rise in MT rate. Based on the growth rate and the risk of MT, optimal timing of treatments could be calculated for each genetic subtype. CONCLUSIONS The mathematical model and WES analysis delineates the optimal timing of treatments in each subtype, which will help to decide the treatment for LGGs.

RET finger protein (RFP) plays a pivotal role in the acquisition of chemoresistance via formation of a complex with nuclear transcription factor Y and histone deacetylase 1, producing specific enhancer activity. We hypothesized that chemoresistance mediated by RFP may result from aberrant deacetylation of H3K27 and dysregulation of novel cis-regulatory (active) enhancers. Therefore, we investigated the effects of RFP on active enhancers and gene expression and evaluated the effects of RFP depletion on the growth of glioma cells treated with temozolomide both in vitro and in vivo. We found that the combination of RFP depletion and TMZ treatment markedly suppressed the growth of glioma cells and extended the survival time of intracranial tumor-bearing mice compared to that of TMZ alone. Chromatin immunoprecipitation and whole-transcriptome sequencing revealed that RFP depletion weakened a significant number of enhancers, and diminished RNA with functions related to mitosis, DNA-dependent DNA replication and cell cycle. Further, the transcriptomes of FOXO1 and TBP2 were significantly increased, while that of PARP-binding protein (PARPBP) was decreased, resulting in induction of reactive oxygen species and cell death. This study suggests that RFP contributes to chemoresistance via aberrant deacetylation of histone H3K27 and dysregulation of RFP-associated active-enhancers in glioma and that the combination of targeting RFP and TMZ has potential as an effective novel treatment strategy for lethal glioma.