Original article
Substitution of 11C-methionine PET by perfusion MRI during the follow-up of treated high-grade gliomas: Preliminary results in clinical practiceSubstitution du PET-méthionine par la perfusion IRM lors du suivi thérapeutique des gliomes de haut grade : résultats préliminaires en pratique clinique

https://doi.org/10.1016/j.neurad.2009.04.005Get rights and content

Summary

Purpose

Our aim was to compare perfusion magnetic resonance imaging (MRI) and positron emission tomography (PET) using carbon-11 labelled methionine (MET) in gliomas and their value in differentiating tumour recurrence from necrosis.

Materials and methods

We retrospectively reviewed 28 patients with a high-grade glioma. A total of 33 MR perfusions and MET-PET were ultimately analysable for comparison between the relative cerebral blood volume (rCBV) and MET-PET examinations. Intra- and interobserver reproducibility was assessed and diagnostic value of rCBV compared to MET-PET and histology was assessed by the area under the receiver operating characteristic (ROC) curve.

Results

ROC curve analysis showed that rCBV had at least equal performances in differentiating tumour recurrence and necrosis than MET-PET. Cut-off value of rCBV for differentiating tumour from necrosis was 182% with a sensitivity of 81.5% and a specificity of 100%.

Conclusion

In clinical practice, perfusion MRI could replace MET-PET for differentiating necrosis from tumour recurrence.

Introduction

Magnetic resonance imaging (MRI) is nowadays the technique of choice for accurate brain tumour work-up. It is the key modality for characterizing lesion topography and morphology, evaluation of the tumoral extension and plays an important role in the follow-up during treatment. For additional physiological information, morphological MRI can be assisted by complementary MR techniques such as spectroscopy, diffusion weighted imaging and perfusion weighted imaging (PWI); those so called “functional imaging” techniques are a growing necessary step in tumoral work-up [1].

PWI can be used to study the tumour vasculature [2], [3] and pinpoint subsection of a given brain tumour where vascularity and endothelial permeability are more altered [4], [5], being thus a useful tool for tumour characterization and treatment planning [6]; moreover, there is an increasing interest in the use of PWI as an indicator of the tumoral response to therapy [7], [8].

Lately, new treatment protocols, combining fractionated radiotherapy and temozolomide, have increased life expectancy to a survival rate of 26.5% at 2 years [9] against 10% at 6 months when conventional radiotherapy is used alone [10]. As a result, the prevalence and extent of posttreatment necrosis has increased [9], [11], [12], [13] making it difficult to determine if an enhancing posttreatment lesion should be considered as a local tumoral recurrence or necrosis: indeed they share the same MRI signs such as enhancement, mass effect and vasogenic oedema [2], [4], [14], [15]. This MRI lack of specificity has prompted the development of other techniques such as carbon-11 labelled methionine (MET) positron emission tomography (PET), which is an indirect method for highlighting tumour proliferation and micro vessel density [16], [17]. It enables a better tissue characterization and is suited for differentiating residual tumour from necrosis with a specificity and sensibility of 61 and 100%, respectively [9], [17], [18].

Nowadays, in such situation and in clinical practice, the MET-PET, with its low specificity, is nevertheless the reference examination when one must plan the future therapeutic scheme of a given patient where the differential diagnosis between recurrence and necrosis must be made [19]. The present study was thus undertaken to examine whether PWI could replace MET-PET in such clinical situation.

First pass PWI, has the big advantage of being performed during the “anatomical” conventional MRI, reducing therefore the number of imaging sessions and the global cost, a non-negligible advantage for the patient and nowadays health policy [11], [18].

Section snippets

Patients

Twenty-eight patients (16 males and 12 females), with a mean age of 51-years-old (ranging from 25 to 74-years-old) and with a histological proven intraaxial primitive brain tumour were included in this retrospective study.

The study was carried out in compliance with our institutional ethic committee (CE Accred. No. 2008/29/246).

All tumours were high-grade gliomas and precise diagnostic was obtained by surgery or stereotactic biopsy: according to the World Health Organization (WHO) histologic

Number of perfusion included into analysis

Out of the 28 patients, a total a 47 PWI were obtained and analysable. Of those 47 PWI, 14 examinations had to be excluded because not performed with a MET-PET. Finally, 33 combined MR and PET examinations were included for further analysis.

MR and PET combined examination analysis

Of the 33 combined MR and PET studies, 31 matched perfectly: 25 studies were positive in both modalities and six were negative. No false positive (type I error) was observed. Nevertheless, two negative MR perfusion corresponded to two positive MET-PET,

Discussion

In clinical neuro-oncologic imaging, conventional MRI is the cornerstone technique for the follow-up of treated high-grade glioma. The main goal of repeated MR acquisition is to detect, precociously, new abnormal modification in the vicinity of the treated tumour site. Most of high-grade tumoral lesions are characterized by a proliferation of immature blood vessels with breakages in the BBB (Fig. 3) [4], [5], [14], [20] whereas radiotherapy induces an acute transient vasodilatation and

Conflicts of interests

None.

References (27)

  • J. Provenzale et al.

    Correlation of relative permeability and relative cerebral blood volume in high-grade neoplasms

    AJR

    (2006)
  • J. Dennie et al.

    NMR imaging of changes in vascular morphology due to tumor angiogenesis

    Magn Reson Med

    (1998)
  • L. Ludemann et al.

    Comparison of dynamic contrast-enhanced MRI with WHO tumor grading for gliomas

    Eur Radiol

    (2001)
  • Cited by (44)

    • Pseudoprogression versus true progression in glioblastoma patients: A multiapproach literature review. Part 2 – Radiological features and metric markers

      2021, Critical Reviews in Oncology/Hematology
      Citation Excerpt :

      Dandois et al. reported that perfusion MRI and MET PET both assisted in diagnosing TP in 28 HGG patients (14 GBM). Of 33 combined MRI and PET studies, 31 matched perfectly, and rCBV had equal performance with MET-PET (Dandois et al., 2010). Another study reported that perfusion MRI with rCBV was superior to both MET-PET and FDG-PET for the differentiation between treatment-related changes and TP (Kim et al., 2010).

    • Radiotherapy target volume definition in newly diagnosed high grade glioma using <sup>18</sup>F-FET PET imaging and multiparametric perfusion MRI: A prospective study (IMAGG)

      2020, Radiotherapy and Oncology
      Citation Excerpt :

      Therefore, 18F-FET PET may result in change in radiotherapy target volume, whereas rCBV could be useful to guide a dose escalation boost inside the initial target volume. Previous studies have shown that tumor volumes determined by amino acid L-[methyl-11C] methionine (MET) PET and blood volume imaging are closely correlated [29,30]. Sadeghi et al. found a high correlation between rCBV values and MET uptake values in a group of 18 patients diagnosed with glial tumors [30], indicating that the degree of vascularity is a major factor in the transfer of MET within gliomas.

    • Imaging Glioblastoma Posttreatment: Progression, Pseudoprogression, Pseudoresponse, Radiation Necrosis

      2019, Radiologic Clinics of North America
      Citation Excerpt :

      In high-grade gliomas, tumor growth leads to neovascularity, increased microvascular density, and slower flow in collateral vasculature, which translates into elevated CBV.58,59 Multiple studies examining DSC-MR in the setting of posttreatment glioma demonstrate elevated rCBV in the setting of tumor progression.33,50,54,60–66 Although a recent survey indicated that approximately 50% of surveyed institutions do not process quantitative parameter maps, published studies show that inspection of color maps in the absence of parameter computation does not reliably differentiate progression from psuedoprogression.67

    View all citing articles on Scopus
    View full text