Radiomics in adaptive Radiotherapy

Application for Adaptive Head&Neck, Brain and Liver Radiotherapy

The aim of biologically adapted radiotherapy (RT) is to shape or paint the prescribed radiation dose according to biological properties of the tumor in order to increase local control rates in the future. Human tumors are known to present with an extremely heterogeneous tissue architecture leading to highly variable local cell densities and chaotic vascular structures leading to tumor hypoxia and regions of increased radiation resistance. The goal of biologically adapted RT or dose painting is to individually adapt the radiation dose to biological features of the tumor as noninvasively assessed with functional MRI and PET imaging in order to overcome increased radiation resistance.

For this purpose, we are developing the platform and methodology to engage functional imaging (MRI and PET) into the radiation treatment planning process and then using texture analysis and radiomics to identify imaging biomarkers for adaptive radiotherapy. We are looking at external beam (conventional dose), stereotactic radiosurgery (SRS) (high dose) and Y90 transarterial radioembolization treatments for head and neck, brain and liver tumors.

Quantitative FLAIR MR imaging as a metric for MR guided Radiation Treatment Planning

The standard-of-care for glioblastoma is resection followed by concurrent temozolomide chemoradiotherapy (RT+TMZ) with adjuvant TMZ chemotherapy. Conventional Gadolinium (Gd) contrast-enhanced T1-weighted (T1W) and T2W or fluid attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) used for radiation treatment planning may inadequately represent the extent of this heterogeneous disease. Differentiating hypercellularity components of glioblastoma from high-vascular components, edema, and normal tissue is a challenge using FLAIR and conventional (b≤1000 s/mm2) diffusion-weighted imaging (DWI).[2] The hypercellularity components may appear nonenhanced on Gd-enhanced T1W-MRI and indistinguishable from edema and normal tissue on FLAIR-MRI thus may receive inadequate treatment. We proposed to use new type of texture analysis technique on T2-FLAIR images to differentiate tumor vs edema on MR images.

Methods: Eighteen GBM and ten meningioma or cerebral metastasis patients were selected. Neuroradiologist using semiautomatic algorithm to contour edema and tumor ROIs in all patients Figure 1. Texture Analysis (First-Order Statistics, Second-Order Statistics and Higher-Order Statistics) was applied to all ROIs, both with and without normalization. More than 280 different texture parameters were calculated for each ROI. The (least absolute shrinkage and selection operator) LASSO was applied on set of texture parameters to select one with highest association for distinguishing tumor cell infiltrative vs vasogenic edema. These variables used for ROC analysis and constructed all relevant plots.

Results: Variables selected for each scenario from the LASSO procedure are shown in Table 1. First-Order Statistics using 1%-Percentile feature was the only parameter chosen in all four scenarios with the best discriminant ability for meningioma, both with and without normalization. Second-Order Statistics using Correlation feature was also selected across scenarios, although the angle and the magnitude varied. For GBM, Higher-Order Statistics using Gray Level Non-uniformity and Short Run Emphasis features provided the best discrimination for images without and with normalization, respectively. Figure 2 displays ROC results showcasing both the single best discriminator and the discriminant ability of the model using all variables selected by LASSO. All univariate models had good discriminant ability (AUC>0.83), and all multivariate models had excellent discriminant ability (AUC>0.93). Figure 3 shows the sorted values for the best discriminator stratified by tissue.

Conclusions: Texture parameters, and only a small subset of such, show excellent ability to discriminate edema from tumor tissue through its most discriminating features. Future studies will evaluate Diffusion-weighted Imaging (DWI) images and ADC maps with some transformations techniques applied to input images (Gabor Transform, s-Transform, Hartley Transform, and so on). In addition, our future efforts will be focused on discriminating vasogenic edema from tumor cell infiltrative which is very important for Radiotherapy processes. For this purpose, we will incorporate additional ROIs in our study.

Publications

1- Sangjune Laurence Lee, Warren Foltz, Brandon Driscoll, Fatemi-Ardekani A, Cynthia Menard, Catherine Coolens, and Caroline Chung, “Comparison of Arterial Input Functions by Magnitude and Phase Signal Measurement in Dynamic Contrast Enhancement MRI using a Dynamic Flow Phantom” (Submitted to the journal of Magnetic Resonance in Medicine).

2- Edward Florez, Ali Asri, Majid Khan, David Joyner, Ellen Parker, Bruce Schlakman, Fatemi-Ardekani A, “Quantitative FLAIR MR imaging as a metric for MR guided RTP (MRgRTP)”, (Manuscript prepared for submission to the journal of Medical Physics)

Quantitive Magnetic Resonance Imagine for response assessment of liver metastases to Y-90 brachytherapy

Y-90 brachytherapy, which is a transarterial radioembolization (TARE) with yttrium 90 microspheres, has become a preferred treatment option for liver metastases, but the response criteria are in need of confirmation. Post response assessment is primarily done using radiologic imaging, but it has been made difficult due to the wide array of post-treatment findings and a somewhat limited ability to distinguish malignant lesions from benign. This study will help determine the clinical significance of post respond quantitative magnetic resonance imaging (MRI) in individuals treated with Y90 brachytherapy

We are collecting 150 patients retrospective post TARE with Y90 micro-sphere treatments. we are using diffusion weighted imaging (DWI), Dynamic contrast enhanced MRI (DCE-MRI), SPECT and PET FDG imaged to model the quantitative assessment of tumor necrosis, tumor growth or reduction and normal tissue (healthy liver) complications such as peritumoral edema, rim enhancement, ill-defined geographic areas of hypoattenuation, capsular retraction, and perihepatic fluid and pleural effusions.

And finally we will correlated our results with the modified Response Evaluation Criteria in Solid Tumors (mRECIST) assessment and look for improvemenets and recommendations.

Publications

Florez E, Fatemi A, Claudio PP and Howard CM. Emergence of Radiomics: “ Novel Methodology Identifying Imaging Biomarkers of Disease in Diagnosis, Response, and Progression”. SM J Clin. Med. Imaging. 2018; 4(1): 1019.
Ali Fatemi, Bart Moris, Chunli Claus Yang, Edward Florez, Sirinvasan Vijayakumar, Mahhava Madhava R. Kanakamedala, Somayeh Taghizadeh,“Multiparametric quantitative MRI as a metric for radiation treatment planning”, Theranostics of Can Res. 1(1). TCR.000503. 2017
E Florez*, T Nicholas, S Lirette, Ali Fatemi, WE-F-205-08, “Quantitative FLAIR MR Imaging as a Metric for MR Guided Radiation Treatment Planning (MRgRTP)”, Med Phys, 2016 Jun;43(6):523
Edward Florez, Todd Nicholas, Seth T.Lirette, Ali Fatemi, “Developing texture analysis technique using Fluid-attenuated inversion recovery (FLAIR) to differentiate tumor from edema in common primary intracranial tumors” (Submitted to the journal of Technology in Cancer Research and Treatment, Oct 2017)
Edward Florez, Todd Nicholas, Seth T.Lirette, Ali Fatemi, “Multiparametric MR imaging in the assessment of primary brain tumors through radiomic features as a metric for guided radiation treatment planning” (Submitted to the journal of Medical Imaging, Dec 2017)
Charles S. Adcock, Kevin Zand, Akash Patel,Ali Fatemi, “Review of treatment response assessment in Yttrium-90 transarterial radioembolization of liver malignancies”, (submitted to the journal Magnetic Resonance Imaging.
Ali Fatemi , Edward Florez, Madhava R. Kanakamedala, Sirinvasan Vijayakumar, “Head and Neck cancer adaptive radiotherapy using generalized multiparametric respond mapping”, (Manuscript in preparation for submission to the journal of clinical oncology)
Charles S. Adcock, Akash Patel, Candace M. Howard-Claudio , Edward Florez, Ali Fatemi, “Evaluation of liver tumor respond to Y-90 brachytherapy using generalized parametric reposes mapping”, Manuscript in preparation for submission to the journal of Magnetic Resonance Imaging (JMRI)

Google Sites

Report abuse