Clinical Neurology and Neurosurgery 139 (2015) 119–124
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Correlation of diffusion tensor and dynamic susceptibility contrast MRI with DNA ploidy and cell cycle analysis of gliomas Anastasia Zikou a , George A. Alexiou b,∗ , George Vartholomatos c , Anna Goussia d , Vasileios G. Xydis a , Paraskevi Kosta a , Spyridon Voulgaris b , Athanasios P. Kyritsis e , Maria I. Argyropoulou a a
Department of Radiology, University Hospital of Ioannina, Ioannina, Greece Department of Neurosurgery, University Hospital of Ioannina, Ioannina, Greece c Haematology Laboratory-Unit of Molecular Biology, University Hospital of Ioannina, Ioannina, Greece d Department of Pathology, University Hospital of Ioannina, Ioannina, Greece e Department of Neurology, University Hospital of Ioannina, Ioannina, Greece b
a r t i c l e
i n f o
Article history: Received 11 August 2015 Received in revised form 25 August 2015 Accepted 22 September 2015 Available online 26 September 2015 Keywords: Glioma MRI Diffusion tensor imaging Perfusion imaging Flow cytometry Cell cycle Ploidy
a b s t r a c t Objectives: Flow cytometry provides a powerful tool to assess cells in G0 /G1 , S and G2 /M phase and ploidy. The purpose of the present study was to investigate the correlation between diffusion tensor (DTI) and dynamic susceptibility contrast (DSC) MRI metrics with cell cycle analysis findings in gliomas. Patients and methods: We studied thirty patients who were operated on for glioma. DTI and DSC MRI were performed within a week prior to surgical excision. Lesion/normal ratios were calculated for the ADC, FA and rCBV. In an excised tumour sample flow cytometric analysis was performed. Results: There were 24 glioblastomas, 2 anaplastic astrocytomas, 1 oligoastrocytoma and 3 diffuse astrocytomas. There were significant differences between low and high-grade gliomas for rCBV and ADC values. Low grade tumours had higher G0 /G1 phase fraction and lower S-phase, G2 /M, S + G2 /M and S + G2 /M/G0 /G1 fractions There was a significant negative correlation between rCBV and G0 /G1 phase fraction and a positive correlation with G2 /M, S + G2 /M and the S + G2 /M/G0 /G1 fraction. Significant correlation was also observed between FA ratio and S + G2 /M/G0 /G1 . There was a negative significant correlation between ADC and S + G2 /M and the S + G2 /M/G0 /G1 fraction. There were 21 (70%) diploid and 9 (30%) aneuploid tumours. No significant difference was found between diploid and aneuploid tumours with respect to rCBV, ADC and FA values. Conclusion: Dynamic susceptibility contrast MRI and diffusion tensor imaging metrics are correlated to tumour aggressiveness as assessed by cell cycle analysis. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Gliomas are the most frequent primary malignant brain tumours. Glioblastoma is the most common and most devastating glioma in adults and despite complete surgical resection and chemo/radiation therapy it has a 5-year relative survival of approximately 5% [1,2]. Thus, pathological grading is of outmost importance for establishing prognosis, next to patient’s age and karnofsky performance scale. Additionally, tumor’s proliferation potential is also an important predictor of tumour behaviour [3]. Several times, for establishing the diagnosis, tissue sample is
∗ Corresponding author at: Aetideon 52, Holargos, 11561 Attiki, Greece. E-mail address:
[email protected] (G.A. Alexiou). http://dx.doi.org/10.1016/j.clineuro.2015.09.013 0303-8467/© 2015 Elsevier B.V. All rights reserved.
obtained by stereotactic surgery which has the disadvantage of providing a small sample and there is always the risk of sampling error. Especially in glioblastomas, the necrotic part may comprise more than 80% of the tumour [4]. Furthermore, given the enhanced chemotherapy methods, and radiotherapy modes of administration that are currently available, it is of outmost importance the identification of tumour response to therapy at short follow-up times. Thus, a non-invasive imaging modality that would provide clues for the characterization of gliomas grade of malignancy, proliferation potentials, response to treatment and overall prognosis is of paramount significance. DNA ploidy and cell-cycle distribution of tumor’s cells, as determined by flow cytometry, provide a powerful tool for establishing tumor’s grade and prognosis [5]. In gliomas, G0 /G1 and S-fraction of cell cycle could distinguish low from high-grade gliomas with high
120
A. Zikou et al. / Clinical Neurology and Neurosurgery 139 (2015) 119–124
sensitivity and specificity. Furthermore, glioma patients with G0 /G1 value greater than 69% and S-phase lower than 6% had better survival [5]. It has been shown previously that diffusion and perfusion MRI provides excellent visualization of high-grade and low-grade gliomas, however is unknown whether diffusion and perfusion MRI metrics correlate with tumour proliferative activity as assessed by flow cytometry [6]. Thus, the aim of this study was to investigate the correlation between diffusion tensor (DTI) and dynamic susceptibility contrast (DSC) MRI metrics with cell cycle and DNA ploidy findings in gliomas.
with the mean values of the ROI in the normal region. The ROIs were evaluated for eligibility independently by two experienced neuroradiologists; any possible disagreements were solved by consensus. 4. DNA analysis protocol (Ioannina protocol)
Patients hospitalized in the Neurosurgical Department of our institution over a 3-year period, who underwent surgery for an intracranial lesion suspicious of glioma and were evaluated by brain MRI within a week prior to the surgical excision were included in the study. A tumour sample was taken during surgery for immediate flow cytometry analysis trying to avoid areas of necrosis. The preoperative CT/MRI findings and the intraoperative findings were not disclosed to the investigator who performed cell cycle analysis. The histopathological examination of the tissue samples was performed from the same neuropathologist and diagnosed tumours were graded according to the World Health Organization 2007 classification scheme. Regarding flow cytometry analysis, the tumours were classified as low-grade (WHO grade I and II) and high-grade (WHO grade III and IV). The study had the approval of our Institutional Review Board and was in accordance with the principles of the Declaration of Helsinki.
Immediately after tumour excision, tumour samples (0.5–2 mm) are minced (Medimachine System, BD Bioscience) for 1 min in PBS buffer (Ca2+ and Mg2+ free, with 0.5 mg/ml RNase) and a cell suspension is obtained. The suspension is then filtered (Consult No 10, Medicons, BD Bioscience) and cells are counted using an automated haematology analyzer to a final concentration of 1.0 × 106 cells/ml. Cells are then processed immediately for staining by adding propidium iodine (PI) (125 g/ml) and after 3 min flow cytometric analysis is performed. All the stained samples are analyzed using a FACSCalibur flow cytometer, equipped with 2 lasers (488 nm, 635 nm) and 6 parameters (FSC, SSC, FL1–FL4) and using CellQuest software. Chicken red blood cells and normal cells obtained from the peripheral blood mononuclear cells (PBMCs) (Ficoll-Paque separation med Pharmacia) are used as the standard to define the position of the diploid G0 /G1 peak in the DNA histograms. These cells can then be mixed with the sample in a second tube before staining and used as a reference to determine the degree of DNA content aberration. The DNA index was calculated as the ratio between the modal channel of the G0 /G1 peak of the aneuploid cells (tumour cells) and that of the diploid cells. We examined the role of G0 /G1 , S-phase, G2 /M, S + G2 /M and S + G2 /M/G0 /G1 phase fraction as biological markers of tumor’s aggressiveness.
3. Imaging protocol
5. Statistical analysis
All MR examinations were performed on the same 1.5-tesla MR unit (Gyroscan Intera; Philips Medical Systems, Best, The Netherlands) using a head coil. The imaging protocol consisted of: (a) a T1-weighted high resolution (1 mm × 1 mm × 1 mm) 3dimentional spoiled gradient echo sequence (repetition time [TR]:25 ms, echo time [TE]:4.6 ms, acquisition matrix: 256 × 228, field of view [FOV]:220 mm), which was used for structural imaging before and after intravenous injection of gadolinium – DTPA, (b) a single shot, multi-slice, spin-echo echo-planar imaging sequence (TR:9807 ms, TE:131 ms, FOV:230 mm, acquisition matrix:128 × 128, slice thickness:3 mm, max b-value:700 s/mm2 , 16 non-collinear diffusion directions), which was used for Apparent Diffusion Coefficient (ADC) and Fractional Anisotropy (FA) measurements. (c) T2* gradient-echo, multishot EPI sequence (TR: 702 ms, TE 30 ms), flip angle: 40◦ , FOV: 250 mm, slice thickness: 7 mm, gap: 0, EPI factor: 17, acquisition matrix: 128 × 51,dynamic scans: 50, imaging time per dynamic scan: 2.1 s, 0.1 mmol/Kg gadolinium is typically injected via an 18 gauge i.v. catheter at 5 cc/s, which was used for rCBV measurements (d) T2-weighted turbo spin echo, axial plane, [TR]:3000 ms, [TE]:90 ms, acquisition matrix:250 × 250, field of view [FOV]:230 mm, slice thickness 6 mm, gap 0.6 mm (e) T2-weighted inversion recovery based sequence for CSF suppression, sagittal plane, [TR]:6300 ms, [TE]:120 ms, inversion recovery[IR]:2150 ms, field of view [FOV]:250 mm, slice thickness 6 mm, gap 0.6, acquisition matrix:250 × 250. A semiquantitative method of image analysis was applied, by calculating the lesion-to-normal (L/N) ratio: a region-of-interest (ROI) was manually defined in the enhancing region of the lesion on a transverse slice showing maximal tumour size in T1-weighted imaging with contrast medium and a second region was drawn on the contralateral normal brain side. Areas of necrosis were excluded. The L/N ratio was calculated by dividing the mean of values derived for every pixel in the given ROI in the tumour region
The rCBV, FA ratio, ADC ratio, G0 /G1 , S-phase, G2 /M, S + G2 /M, S + G2 /M/G0 /G1 , between low and high-grade groups were compared using the two-sided, nonparametric Mann–Whitney U test. Correlation among ADC ratio, FA ratio, rCBV, and G0 /G1 , S-phase, G2 /M, S + G2 /M, S + G2 /M/G0 /G1 and tumour ploidy were analyzed statistically using Spearman Rho test. A 2-sided p-value
