Abstract

Abstract BACKGROUND Radiation therapy(IMRT) protocols for glioblastoma utilize isotropic margin expansion around neurosurgical cavities/residual contrast enhancing(T1gd) disease. However pathways of infiltration may be associated with white matter tract(WMT) involvement varying between neuroanatomical subsites. Posterior frontoparietal tumours involving posterior cingulate region (PostCing) may have extensive infiltration centrally. This study aims to quantify the PostCing infiltration pathway/relapse sites. METHODS Consecutive patients managed for GBMwt with IMRT between 1/2016-6/2022 were classified into 15 neuroanatomical subsites. Tumour segmentation on T1-gadolinium-enhanced(T1gd) and T2-weighted(T2) sequences was performed at diagnosis/relapse. Subsites of Frontal Lobe (Posterior) and Parietal Lobe (Superior/Posterior/Lateral) were identified. On both sequences involvement of adjacent/distant regions (cingulate gyrus, cingulum, splenium, ventricular trigone, parahippocampal-hippocampal gyrus) were identified; as were major WMT: cingulum-parahippocampal formation(CING-PHG), and arcuate fasciculus–inferior longitudinal fasciculus(AF-ILF). Survival was analysed in relation to pattern of infiltration and site of relapse. RESULTS Of 348 patients with glioblastoma, fifty-five(16%) involved PostCing region. Pre-IMRT features included near-total resection in 25%; impaired ECOG 2-3 in 47%; and 58% MGMT-methylated. Parietal lobe tumours(67%) were predominantly posterior(42%) with only 7% lateral parietal. WMT involvement was evident in 94% with T1gd(62%), and T2 alone(33%) abnormality. Infiltration to distal sites of cingulum/splenium/PHG was frequent and predictable; correlating to cingulum pathway involvement in 70% of patients. Median RFS and OS was 13.2 months(95%CI: 11.8-14.8) and 16.6 months(95%CI:14.8-18.5) respectively. Distant relapse was evident in 53% of relapses. In the forty-four Stupp60Gy patients, distant relapse occurred later and was associated with improved overall survival(p = 0.02). No difference in RFS(p = 0.82) or OS(p = 0.31) was noted if tract involvement was either T1gd or T2 alone abnormality. CONCLUSION For glioblastoma in posterior cingulate region, the infiltration and relapse pathways followed white matter tracts in a predictable distribution with high rate of distant disease at diagnosis and relapse. IMRT potentially may be individualised for neuroanatomical subsite and non-isotropic target volume expansion.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call