Original paperDiffusion kurtosis imaging in head and neck cancer: A correlation study with dynamic contrast enhanced MRI
Introduction
The head and neck (HN) region is anatomically complex, because it is characterized by heterogeneous tissues with several histological types of tumors. Magnetic resonance imaging (MRI) is largely used for the differential diagnosis, staging, and follow-up of primary tumors and cervical nodes, thanks to its high soft-tissue contrast [1]. If combined with conventional morphological sequences, functional MRI techniques, such as dynamic-contrast enhanced MRI (DCE-MRI) and diffusion-weighted imaging (DWI), allow a better tumor characterization due to availability of additional information on microvascular and cellular architecture of tissues [2], [3], [4], [5]. In several studies, DCE-MRI and DWI have shown the potential for differentiating benign from malignant HN pathologies, elucidating early treatment-induced tissue changes, predicting the treatment response to therapy, and assessing scar/recurrent cancers [6], [7], [8], [9].
Specifically, DCE-MRI requires the injection of an intravenous bolus of gadolinium contrast agent, followed by a rapid serial acquisition of T1-weighted images, which can be post-processed to derive several perfusion metrics using a pharmacokinetic model or simpler semiquantitative methods [10]. In the former case, information on the vessel permeability can be obtained using the transfer constants from the plasma space to the extracellular extravascular space (EES) and vice versa, Ktrans and Kep [11]; in the latter case, semiquantitative model-free parameters such as the initial area under the gadolinium curve (IAUGC) can be derived [5].
Conversely, DWI is a quantitative technique, which is largely applied—at first to the brain and then to various body regions—to investigate the tissue microstructure and the presence of pathological alterations by estimating the water molecule diffusion [12]. Performing DWI of HN cancers is technically challenging, because of possible motion and/or susceptibility artifacts; however, thanks to the considerable advances in image acquisition and processing, it is now widely used [3]. These technical improvements have recently allowed the implementation of more specialized DWI techniques known as Diffusion Kurtosis Imaging (DKI), which has been proposed to take into account the non-Gaussian diffusion effects observed at high b-values (>2000 s/mm2). These effects can be attributed to the presence of tissue microstructures and barriers, which cause a diffusion hindrance. The DKI analysis is based on an exponential function that yields two additional parameters [13]: the apparent diffusion coefficient (Dapp), which is corrected for the non-Gaussian behavior of water molecules, and the apparent diffusional kurtosis parameter (Kapp) [14].
Whilst the biophysical meaning of DCE-MRI parameters is well understood, a clear biophysical interpretation of DWI/DKI-derived metrics remains elusive [14]; only a few studies have focused on a better comprehension of Dapp and Kapp with regard to HN cancer [9], [15].
We have hypothesized that the DKI parameters may have a strict connection to DCE-MRI parameters and intratumoral heterogeneity, as they may be influenced by the complexity and integrity of both the vascular and extracellular compartments.
Although DCE-MRI was proposed in several HN studies [15], [16], [17], it was rarely acquired with an adequate scan time (≥10 min) to allow an accurate estimate of ve (the volume of EES per unit volume of tissue), as indicated by the literature [18].
The aim of this study was to investigate the biophysical meaning of DKI parameters via correlations with more consolidated perfusion parameters obtained from a long DCE-MRI scan.
Section snippets
Patient population
The present retrospective study was authorized by the hospital ethics committee. Patients were eligible to be included in the study if they had (a) a newly diagnosed HN tumor, (b) a biopsy- or fine needle aspiration-based histological confirmation, (c) an age of > 18 years and (d) underwent pre-treatment MRI, including DKI and DCE-MRI, according to the protocol described below. The exclusion criteria for the patients were: (a) previous treatment for tumors at the primary site and/or the neck,
Results
From September 2016 to May 2018, a total of 22 patients with HN tumors were included in this retrospective study with authorization from the hospital ethics committee. All patients underwent pre-treatment MRI according to the imaging protocol described in the previous section. The study population included 16 men and 6 women, with an average age of 56.8 years (range, 21–79 years). Multiple lesions were identified in some patients; therefore a total of 26 lesions were analyzed. For one patient,
Discussion
DKI is an emerging technique in oncological applications and it is a research tool for the evaluation of HN lesions [6], [15], [26], [27], [28]. To contribute to a better understanding of DKI [14], we investigated the associations between the DKI parameters and the more consolidated perfusion parameters obtained from DCE-MRI. In fact, DCE-MRI allows to investigate the tissue vascularity by composite parameters as Ktrans, which is influenced by blood flow and vessel permeability [10], but it
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Acknowledgement
The authors would like to thank the lead GE Healthcare MR education specialist Carmelo Parisi for his contribution to the MR sequence optimization. The authors are also grateful to the anonymous reviewers for their constructive criticisms and comments.
References (41)
- et al.
State of the art MRI in head and neck cancer
Clin Radiol
(2018) - et al.
Diffusion-Weighted Imaging in Head and Neck Cancer: Technique, Limitations, and Applications
Magn Reson Imaging Clin N Am
(2018) - et al.
Dynamic Contrast-Enhanced MR Imaging in Head and Neck Cancer
Magn Reson Imaging Clin N Am
(2018) - et al.
Evaluation of Head and Neck Tumors with Functional MR Imaging
Magn Reson Imaging Clin N Am
(2016) - et al.
Histogram analysis parameters identify multiple associations between DWI and DCE MRI in head and neck squamous cell carcinoma
Magn Reson Imaging
(2018) - et al.
Advanced diffusion models in head and neck squamous cell carcinoma patients: Goodness of fit, relationships among diffusion parameters and comparison with dynamic contrast-enhanced perfusion
Magn Reson Imaging
(2017) - et al.
3D Slicer as an image computing platform for the Quantitative Imaging Network
Magn Reson Imaging
(2012) - et al.
Correlations Between DCE MRI and Histopathological Parameters in Head and Neck Squamous Cell Carcinoma
Transl Oncol
(2017) - et al.
Oral carcinoma: Clinical evaluation using diffusion kurtosis imaging and its correlation with histopathologic findings
Magn Reson Imaging
(2018) - et al.
How clinical imaging can assess cancer biology
Insights Imaging
(2019)
Functional magnetic resonance imaging of head and neck cancer: performance and potential
Neuroradiol J
Integrating dynamic contrast-enhanced magnetic resonance imaging and diffusion kurtosis imaging for neoadjuvant chemotherapy assessment of nasopharyngeal carcinoma
J Magn Reson Imaging.
Diffusion-weighted imaging outside the brain: Consensus statement from an ISMRM-sponsored workshop
J Magn Reson Imaging
Diffusional kurtosis imaging: the quantification of non-gaussian water diffusion by means of magnetic resonance imaging
Magn Reson Med
Differentiation of olfactory neuroblastomas from nasal squamous cell carcinomas using MR diffusion kurtosis imaging and dynamic contrast-enhanced MRI
J Magn Reson Imaging
Cited by (8)
MRI and Clinical Biomarkers Overlap between Glaucoma and Alzheimer’s Disease
2023, International Journal of Molecular SciencesOptimizing b-values schemes for diffusion MRI of the brain with segmented Intravoxel Incoherent Motion (IVIM) model
2023, Journal of Applied Clinical Medical Physics
- 1
The first two authors contributed equally to this work.