P14.23 Risk of venous thromboembolism (VTE) in grade II-IV gliomas as a function of molecular subtype

Abstract

Abstract BACKGROUND VTE affects up to 30% of patients with glioblastoma (GBM, WHO grade IV), but little is known about its incidence in lower-grade gliomas (LGG, WHO grade II-III). It has been suggested that isocitrate dehydrogenase (IDH) mutation status dramatically decreases the incidence of VTE in glioma patients, through a combination of F3 promoter hypermethylation leading to lower expression of the procoagulant protein tissue factor, and an increased production of D-2-hydroxyglutarate, which has anticoagulant properties (Unruh et al, 2016). Our objective was to determine the incidence of VTE in LGG and stratify VTE risk by molecular subtype in gliomas grade II-IV. MATERIAL AND METHODS We performed a retrospective analysis of 590 glioma patients with molecular testing seen at our institution (UVa) from January 2005 to August 2017. We divided LGG patients into 3 groups: IDH-wildtype (IDHwt); IDH-mutant (IDHmt), 1p/19q-codeleted; and IDHmt, 1p/19q-intact. GBM patients were divided according to MGMT methylation status. Estimates of cumulative incidence of VTE were calculated with death as competing risk, and significance testing was determined using the Fine and Gray model. RESULTS Of 256 LGG patients (147 grade II and 109 grade III), 81 were IDHwt, 113 IDHmt and 1p/19q-codeleted, and 62 IDHmt and 1p/19q-intact. There were 334 GBM patients, with MGMT methylation status available in 263 (98 (37%) methylated and 165 (63%) unmethylated). With a median follow-up of 545 days, the overall incidence of VTE was 8.2% for grade II, 9.2% for grade III and 30.5% for grade IV. The 6-, 12- and 24-month VTE incidence was 4.1%, 4.8% and 5.4% respectively for grade II, 4.6%, 7.3% and 9.2% for grade III and 23.1%, 26.6% and 29% for grade IV. In LGG patients, VTE incidence was slightly higher in IDHwt tumors (11.1%) vs IDHmt, 1p/19q-codeleted (8.8%) and IDHmt, 1p/19q-intact tumors (4.8%). However, this difference was not statistically significant (IDHwt vs IDHmt, 1p/19q-codeleted, sub-distribution hazard ratio (SHR)=1.67, 95% CI=0.59–4.72; IDHwt vs IDHmt, 1p/19q-intact, SHR=1.87, 95% CI=0.54–6.53). In GBM patients, there was no difference in the VTE incidence according to MGMT methylation status (SHR=0.99, 95% CI=0.64–1.54). CONCLUSION In our cohort, the risk of VTE in GBM patients was consistent with historical data; patients with LGG also had a higher VTE risk compared to the general population. In contrast to other retrospective studies in which the incidence of VTE for grade II-IV IDHmt gliomas was 0% (Unruh et al, 2016; Nazari et al, 2018), our data suggest that VTEs do occur in IDHmt LGG patients, although at a lower rate than in IDHwt. MGMT methylation does not seem to influence the incidence of VTE. VTE risk stratification in GBM patients based on IDH mutation is forthcoming.