Correlation of microvascular fractal dimension with positron emission tomography [(11)C]-methionine uptake in glioblastoma multiforme: preliminary findings

Abstract

Neuroradiological and metabolic imaging is a fundamental diagnostic procedure in the assessment of patients with primary and metastatic brain tumors. The correlation between objective parameters capable of quantifying the neoplastic angioarchitecture and imaging data may improve our understanding of the underlying physiopathology and make it possible to evaluate treatment efficacy in brain tumors. Only a few studies have so far correlated the quantitative parameters measuring the neovascularity of brain tumors with the metabolic profiles measured by means of amino acid uptake in positron emission tomography (PET) scans. Fractal geometry offers new mathematical tools for the description and quantification of complex anatomical systems, including microvascularity. In this study, we evaluated the microvascular network complexity of six cases of human glioblastoma multiforme quantifying the surface fractal dimension on CD34 immunostained specimens. The microvascular fractal dimension was estimated by applying the box-counting algorithm. As the fractal dimension depends on the density, size and shape of the vessels, and their distribution pattern, we defined it as an index of the whole complexity of microvascular architecture and compared it with the uptake of (11)C-methionine (MET) assessed by PET. The different fractal dimension values observed showed that the same histological category of brain tumor had different microvascular network architectures. Fractal dimension ranged between 1.19 and 1.77 (mean: 1.415+/-0.225), and the uptake of (11)C-methionine ranged between 1.30 and 5.30. A statistically significant direct correlation between the microvascular fractal dimension and the uptake of (11)C-methionine (p=0.02) was found. Our preliminary findings indicate that that vascularity (estimated on the histologic specimens by means of the fractal dimension) and (11)C-methionine uptake (assessed by PET) closely correlate in glioblastoma multiforme and that microvascular fractal dimension can be a useful parameter to objectively describe and quantify the geometrical complexity of the microangioarchitecture in glioblastoma multiforme.