Gliomas are the most frequent primary tumors of the brain. They can be divided into grade II-IV astrocytomas and grade II-III oligodendrogliomas, based on their histomolecular profile. The prognosis and treatment is highly dependent on grade and well-identified prognostic and/or predictive molecular markers. Multi-parametric MRI, including diffusion weighted imaging, perfusion, and MR spectroscopy, showed increasing value in the non-invasive characterization of specific molecular subsets of gliomas. Radiolabeled amino-acid analogues, such as 18F-FET, have also been proven valuable in glioma imaging. These tracers not only contribute in the diagnostic process by detecting areas of dedifferentiation in diffuse gliomas, but this technique is also valuable in the follow-up of gliomas, as it can differentiate pseudo-progression from real tumor progression. Since multi-parametric MRI and 18F-FET PET are complementary imaging techniques, there may be a synergistic role for PET-MRI imaging in the neuro-oncological imaging of primary brain tumors. This could be of value for both primary staging, as well as during treatment and follow-up.

BACKGROUND Venous thromboembolism (VTE) is a common complication in patients with glioblastoma. Despite high incidence of up to 30% per year, concerns about bleeding complications have limited the use of primary anticoagulant prophylaxis. Finding a suitable biomarker to assess the risk of occurrence is therefore of utmost clinical interest. We performed an exploratory study of preoperative routinely used haematological markers as predictor for the development of VTE in glioblastoma patients. MATERIAL AND METHODS Data was retrospectively collected from an existing database of 307 patients diagnosed with glioblastoma by the Oncology Network South-East Netherlands (OnzoZON) between 2006 and 2020. Collected preoperative haematological markers included: haemoglobin, platelets, lactate dehydrogenase, neutrophils, lymphocytes, albumin and derived ratios. In addition, type and date of VTE were retrieved from medical records. Receiver operating curve was used to identify the optimal cut-off values of the preoperative haematological markers. Univariate and multivariate logistic regression analyses were performed to predict VTE for each haematologic marker independently. Variables included in the multivariate analyses were age, gender, type of surgery, Karnofsky performance score, MGMT status, weight, height and BMI, already available from the primary database. RESULTS In the total dataset, 45 patients (15%) suffered from a VTE, most common pulmonary embolism (51%) followed by deep vein thrombosis (31%). Mean time from diagnosis until VTE was 4.3 months (SD = 5.5). Preoperative haemoglobin value was available for analyses in 265 patients, platelets value in 226, lactate dehydrogenase in 98, neutrophils in 133, lymphocytes in 133 and albumin in 56 patients. A preoperative lactate dehydrogenase value > 243 U/L was found to increase the risk of VTE in both univariate and multivariate analysis (P <0.05). Seventeen out of 98 patients of whom lactate dehydrogenase level was available suffered from a VTE, most common pulmonary embolism (59%), followed by deep vein thrombosis (29%) and cerebral venous sinus thrombosis (12%). An elevated lactate dehydrogenase in serum increased the odds for getting a VTE by 3.2 (1.1–9.4). None of the other investigated haematological markers or ratios were found to be significantly correlated with the occurrence of VTE in our study. CONCLUSION Glioblastoma initiates locally haemostatic abnormalities, that propagate systemically though circulating mediators. Our exploratory analysis shows for the first time that preoperative lactate dehydrogenase levels might aid clinicians in identifying patients at risk for a venous thromboembolism. Ultimately this could lead to preventive measures and patient education, but larger and prospective validation of these findings is warranted.