Warning: Declaration of My_Walker::start_el(&$output, $item, $depth, $args) should be compatible with Walker_Nav_Menu::start_el(&$output, $data_object, $depth = 0, $args = NULL, $current_object_id = 0) in /home2/ajnrblog/public_html/ajnrdigest/wp-content/themes/ajnr/functions.php on line 258
Glioma: Application of Histogram Analysis of Pharmacokinetic Parameters from T1-Weighted Dynamic Contrast-Enhanced MR Imaging to Tumor Grading - AJNR News Digest
September 2014
Brain

Glioma: Application of Histogram Analysis of Pharmacokinetic Parameters from T1-Weighted Dynamic Contrast-Enhanced MR Imaging to Tumor Grading

Seung Chai Jung

Seung Chai Jung

Seung Hong Choi

Seung Hong Choi

It is not popular in clinical practice for radiologists to perform analysis based on whole-lesion segmentation of brain lesions for interpretation and diagnosis on MRI. The same method would be better applied to glioma grading by using MR perfusion imaging. However, many investigators performed quantitative analyses only from parts of the entire tumor in glioma grading. This discrepancy could make observers or interpreters have a distorted perspective, which could result in incorrect interpretation. We’ve paid attention to the discrepancy in the analyses by using MR perfusion imaging. We thought that if we wanted to analyze the entire tumor, histogram analysis would be one of the most effective methods. We have studied histogram analyses from entire-tumor data of gliomas.1,2 We also believed that even though small errors occurred in the analyses, the result could not be easily distorted due to the strength of huge data. This concern led to the present study.

T1-weighted dynamic contrast-enhanced (T1W DCE) MR imaging may represent the vascular permeability and angiogenic activity in gliomas. Ktrans, Ve, and Vp have been introduced as the representative parameters in glioma grading. Ktrans and Ve are known to reflect the vascular permeability; Vp may reflect angiogenic acitivity. Therefore, we wondered which parameter is better in determining the grade of gliomas and whether the histogram analysis from entire-tumor data is useful or not.

In the present study, the histogram analysis in perfusion parameters (Ktrans, Ve, Vp) from entire-tumor data based on T1W DCE MR imaging was useful and feasible for glioma grading even though it was time-consuming work. Ktrans, Ve, and Vp could contribute to the differentiation of grade of gliomas, but Ktrans of the 98th percentile was the most significant parameter. The cutoff values between high- and low-grade gliomas were Ktrans of the 98th percentile (0.277 minute-1; AUC, 0.912), Ve of the 90th percentile (19.7%; AUC, 0.939), and Vp of the 84th percentile (11.7%; AUC, 769). The cutoff values belong to relatively low values in each parametric value, which may mean that low-grade gliomas could also have relatively high

perfusion parametric values. Hence, the glioma grading in T1W DCE MR perfusion imaging should be performed on the basis of entire-tumor data. In clinical practice, we also are concerned with the distribution or rate of higher or lower perfusion parametric values in the entire tumor, which may contribute to predicting where the tumor develops or transforms into high grade.

We presented some limitations in the present study. These became other starting points for future studies. We used a fixed baseline T1 (1000 ms). The method gave us consistent and reproducible information, but it is not physiologic. Therefore, we need a physiologic and consistent baseline T1 value. The manual drawing of ROIs is necessary to convert to a semiautomatic or automatic method in terms of building a basic and reproducible measurement method. We actually performed a comparison study between manual and semiautomatic segmentation methods in T2-weighted DSC MR around the same time as the present study. The conclusion was that semiautomatic segmentation was a more reproducible and consistent method than the manual method.3 The semiautomatic method could be applied to T1W DCE MR imaging. Histogram analyses based on entire-tumor data also applied to an animal model.4 The clinical application of the quantitative analyses by using entire lesions is important above all things. Hence, semiautomatic or automatic method can be a research focus, from defining lesions to presenting the histogram analyses. A time-consuming method is finally avoided for users. In addition, we are wondering if T1W DCE MR perfusion parameters work in treatment responses. Therefore, the present study raised more interesting issues, and we are trying to answer those questions.

References

  1. Chu HH, Choi SH, Ryoo I, et al. Differentiation of true progression from pseudoprogression in glioblastoma treated with radiation therapy and concomitant temozolomide: comparison study of standard and high-b-value diffusion-weighted imaging. Radiology 2013;269:831–40, 10.1148/radiol.13122024
  2. Kang Y, Choi SH, Kim YJ, et al. Gliomas: Histogram analysis of apparent diffusion coefficient maps with standard- or high-b-value diffusion-weighted MR imaging—correlation with tumor grade. Radiology 2011;261:882–90, 10.1148/radiol.11110686
  3. Jung SC, Choi SH, Yeom JA, et al. Cerebral blood volume analysis in glioblastomas using dynamic susceptibility contrast-enhanced perfusion MRI: a comparison of manual and semiautomatic segmentation methods. PLoS One 2013;8:e69323, 10.1371/journal.pone.0069323
  4. Song YS, Park CM, Lee SM, et al. Reproducibility of histogram and texture parameters derived from intravoxel incoherent motion diffusion-weighted MRI of FN13762 rat breast carcinomas. Anticancer Res 2014;34:2135–44

 

Read this article at AJNR.org . . .