Search

Background: The standard of care for glioblastoma is surgical resection followed by combination temozolomide and radiation. Magnetic resonance imaging (MRI) is used preoperatively for tumor resection planning. In some instances, MRI is also obtained postoperatively to assess for any complications and to determine extent of resection. There is some question whether early routine postoperative imaging of patients after tumor resection is beneficial to long-term outcomes, especially with the increased scrutiny of increasing health care costs., Methods: In this study we retrospectively analyze patients with glioblastoma treated at our institution, comparing the difference in overall survival and treatment regimens between patients who had early postoperative MRI versus patients who did not., Results: We determine that in our cohort of 125 patients, those with early postoperative MRI had no statistically significant overall survival difference compared with patients with no early postoperative MRI (P = 0.996). The median survival for the group with postoperative MRI was 378 days (95% confidence interval [CI], 242-443 days), and the median survival for the group without postoperative MRI was 308 days (95% CI, 203-445 days). Early postoperative MRI also did not significantly alter therapeutic regimens., Conclusions: Although early postoperative MRI may not significantly affect patient overall survival from a statistical standpoint or therapeutic regimens, this type of imaging may be important to hone resident and attending skill. We encourage other institutions to perform similar analyses to determine the overall survival benefit of early postoperative imaging after glioma resection for patients with glioblastoma., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Glioblastoma is a devastating malignancy with a dismal survival rate. Currently, there are limited prognostic markers of glioblastoma including IDH1, ATRX, MGMT, PTEN, EGFRvIII, and others. Although these biomarkers for tumor prognosis are available, a surgical biopsy must be performed for these analyses, which has morbidity involved. A non-invasive and readily available biomarker is sought after which provides clinicians prognostic information. Sodium is an electrolyte that is easily and quickly obtained through analysis of a patient's serum. Hyponatremia has been shown to have a predictive and negative prognostic indication in multiple cancer types, but the role of glioblastoma patients' serum sodium at the time of diagnosis in predicting glioblastoma patient survival has not been determined. We assessed whether hyponatremia at the time of glioblastoma diagnosis correlates to patient survival and show that in our cohort of 200 glioblastoma patients, sodium, at any level, did not significantly correlate to glioblastoma survival, unlike what is seen in multiple other cancer types. We further demonstrate that inducing hyponatremia in an orthotopic murine model of glioblastoma has no effects on tumor progression and survival.

Human hemochromatosis protein (HFE) is involved in iron metabolism. Two major HFE polymorphisms, H63D and C282Y, have been associated with an increased risk of cancers. Previously, we reported decreased gender effects in overall survival based on H63D or C282Y HFE polymorphisms patients with glioblastoma multiforme (GBM). However, the effect of other single nucleotide variation (SNV) in the HFE gene on the cancer development and progression has not been systematically studied. To expand our finding in a larger sample, and to identify other HFE SNV, we analyzed the frequency of somatic SNV in HFE gene and its relationship to survival in GBM patients using The Cancer Genome Atlas (TCGA) GBM (Caucasian only) database. We found 9 SNVs with increased frequency in blood normal of TCGA GBM patients compared to the 1000Genome. Among 9 SNVs, 7 SNVs were located in the intron and 2 SNVs (i.e., H63D, C282Y) in the exon of HFE gene. The statistical analysis demonstrated that blood normal samples of TCGA GBM have more H63D (p = 0.0002, 95% Confidence interval (CI): 0.2119-0.3223) or C282Y (p = 0.0129, 95% CI: 0.0474-0.1159) HFE polymorphisms than 1000Genome. The Kaplan-Meier survival curve for the 264 GBM samples revealed no difference between wild type (WT) HFE and H63D, and WT HFE and C282Y GBM patients. In addition, there was no difference in the survival of male/female GBM patients based on HFE genotype. There was no correlation between HFE expression and survival. In conclusion, the current results suggest that somatic HFE polymorphisms do not impact GBM patients' survival in the TCGA data set of GBM.