Andra Krauze

Bio: Andra Krauze is an academic researcher from National Institutes of Health. The author has contributed to research in topics: Medicine & Prostate cancer. The author has an hindex of 11, co-authored 33 publications receiving 530 citations. Previous affiliations of Andra Krauze include University of Alberta & BC Cancer Agency.

A Phase 2 Study of Concurrent Radiation Therapy, Temozolomide, and the Histone Deacetylase Inhibitor Valproic Acid For Patients With Glioblastoma

Abstract: Purpose Valproic acid (VPA) is an antiepileptic agent with histone deacetylase inhibitor (HDACi) activity shown to sensitize glioblastoma (GBM) cells to radiation in preclinical models. We evaluated the addition of VPA to standard radiation therapy (RT) plus temozolomide (TMZ) in patients with newly diagnosed GBM. Methods and Materials Thirty-seven patients with newly diagnosed GBM were enrolled between July 2006 and April 2013. Patients received VPA, 25 mg/kg orally, divided into 2 daily doses concurrent with RT and TMZ. The first dose of VPA was given 1 week before the first day of RT at 10 to 15 mg/kg/day and subsequently increased up to 25 mg/kg/day over the week prior to radiation. VPA- and TMZ-related acute toxicities were evaluated using Common Toxicity Criteria version 3.0 (National Cancer Institute Cancer Therapy Evaluation Program) and Cancer Radiation Morbidity Scoring Scheme for toxicity and adverse event reporting (Radiation Therapy Oncology Group/European Organization for Research and Treatment). Results A total of 81% of patients took VPA according to protocol. Median overall survival (OS) was 29.6 months (range: 21-63.8 months), and median progression-free survival (PFS) was 10.5 months (range: 6.8-51.2 months). OS at 6, 12, and 24 months was 97%, 86%, and 56%, respectively. PFS at 6, 12, and 24 months was 70%, 43%, and 38% respectively. The most common grade 3/4 toxicities of VPA in conjunction with RT/TMZ therapy were blood and bone marrow toxicity (32%), neurological toxicity (11%), and metabolic and laboratory toxicity (8%). Younger age and class V recursive partitioning analysis (RPA) results were significant for both OS and PFS. VPA levels were not correlated with grade 3 or 4 toxicity levels. Conclusions Addition of VPA to concurrent RT/TMZ in patients with newly diagnosed GBM was well tolerated. Additionally, VPA may result in improved outcomes compared to historical data and merits further study.

Bringing the heavy: carbon ion therapy in the radiobiological and clinical context

Abstract: Radiotherapy for the treatment of cancer is undergoing an evolution, shifting to the use of heavier ion species. For a plethora of malignancies, current radiotherapy using photons or protons yields marginal benefits in local control and survival. One hypothesis is that these malignancies have acquired, or are inherently radioresistant to low LET radiation. In the last decade, carbon ion radiotherapy facilities have slowly been constructed in Europe and Asia, demonstrating favorable results for many of the malignancies that do poorly with conventional radiotherapy. However, from a radiobiological perspective, much of how this modality works in overcoming radioresistance, and extending local control and survival are not yet fully understood. In this review, we will explain from a radiobiological perspective how carbon ion radiotherapy can overcome the classical and recently postulated contributors of radioresistance (α/β ratio, hypoxia, cell proliferation, the tumor microenvironment and metabolism, and cancer stem cells). Furthermore, we will make recommendations on the important factors to consider, such as anatomical location, in the future design and implementation of clinical trials. With the existing data available we believe that the expansion of carbon ion facilities into the United States is warranted.

Automated Glioma Grading on Conventional MRI images Using Deep Convolutional Neural Networks

Abstract: Purpose Gliomas are the most common primary tumor of the brain and are classified into grades I-IV of the World Health Organization (WHO), based on their invasively histological appearance. Gliomas grading plays an important role to determine the treatment plan and prognosis prediction. In this study we propose two novel methods for automatic, non-invasively distinguishing low-grade (Grades II and III) glioma (LGG) and high-grade (grade IV) glioma (HGG) on conventional MRI images by using deep convolutional neural networks (CNNs). Methods All MRI images have been preprocessed first by rigid image registration and intensity inhomogeneity correction. Both proposed methods consist of two steps: (a) three-dimensional (3D) brain tumor segmentation based on a modification of the popular U-Net model; (b) tumor classification on segmented brain tumor. In the first method, the slice with largest area of tumor is determined and the state-of-the-art mask R-CNN model is employed for tumor grading. To improve the performance of the grading model, a two-dimensional (2D) data augmentation has been implemented to increase both the amount and the diversity of the training images. In the second method, denoted as 3DConvNet, a 3D volumetric CNNs is applied directly on bounding image regions of segmented tumor for classification, which can fully leverage the 3D spatial contextual information of volumetric image data. Results The proposed schemes were evaluated on The Cancer Imaging Archive (TCIA) low grade glioma (LGG) data, and the Multimodal Brain Tumor Image Segmentation (BraTS) Benchmark 2018 training datasets with fivefold cross validation. All data are divided into training, validation, and test sets. Based on biopsy-proven ground truth, the performance metrics of sensitivity, specificity, and accuracy are measured on the test sets. The results are 0.935 (sensitivity), 0.972 (specificity), and 0.963 (accuracy) for the 2D Mask R-CNN based method, and 0.947 (sensitivity), 0.968 (specificity), and 0.971 (accuracy) for the 3DConvNet method, respectively. In regard to efficiency, for 3D brain tumor segmentation, the program takes around ten and a half hours for training with 300 epochs on BraTS 2018 dataset and takes only around 50 s for testing of a typical image with a size of 160 × 216 × 176. For 2D Mask R-CNN based tumor grading, the program takes around 4 h for training with around 60 000 iterations, and around 1 s for testing of a 2D slice image with size of 128 × 128. For 3DConvNet based tumor grading, the program takes around 2 h for training with 10 000 iterations, and 0.25 s for testing of a 3D cropped image with size of 64 × 64 × 64, using a DELL PRECISION Tower T7910, with two NVIDIA Titan Xp GPUs. Conclusions Two effective glioma grading methods on conventional MRI images using deep convolutional neural networks have been developed. Our methods are fully automated without manual specification of region-of-interests and selection of slices for model training, which are common in traditional machine learning based brain tumor grading methods. This methodology may play a crucial role in selecting effective treatment options and survival predictions without the need for surgical biopsy.

A Prospective Study on 18F-DCFPyL PSMA PET/CT Imaging in Biochemical Recurrence of Prostate Cancer

Abstract: 18F-DCFPyL (2-(3-{1-carboxy-5-[(6-18F-fluoro-pyridine-3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid), a prostate-specific membrane antigen–targeting radiotracer, has shown promise as a prostate cancer imaging radiotracer. We evaluated the safety, sensitivity, and impact on patient management of 18F-DCFPyL in the setting of biochemical recurrence of prostate cancer. Methods: Subjects with prostate cancer and biochemical recurrence after radical prostatectomy or curative-intent radiotherapy were included in this prospective study. The subjects underwent 18F-DCFPyL PET/CT imaging. The localization and number of lesions were recorded. The uptake characteristics of the 5 most active lesions were measured. A pre- and posttest questionnaire was sent to treating physicians to assess the impact on management. Results: One hundred thirty subjects were evaluated. 18F-DCFPyL PET/CT localized recurrent prostate cancer in 60% of cases with a prostate-specific antigen (PSA) level of ≥0.4 to

Brain tumor segmentation using holistically nested neural networks in MRI images

Abstract: Purpose Gliomas are rapidly progressive, neurologically devastating, largely fatal brain tumors. Magnetic Resonance Imaging (MRI) is a widely used technique employed in the diagnosis and management of gliomas in clinical practice. MRI is also the standard imaging modality used to delineate the brain tumor target as part of treatment planning for the administration of radiation therapy. Despite more than 20 years of research and development, computational brain tumor segmentation in MRI images remains a challenging task. We are presenting a novel method of automatic image segmentation based on holistically-nested neural networks that could be employed for brain tumor segmentation of MRI images. Methods Two preprocessing techniques were applied to MRI images. The N4ITK method was employed for correction of bias field distortion. A novel landmark-based intensity normalization method was developed so that tissue types have a similar intensity scale in images of different subjects for the same MRI protocol. The holistically-nested neural networks (HNN), which extend from the convolutional neural networks (CNN) with a deep supervision through an additional weighted-fusion output layer, was trained to learn the multi-scale and multi-level hierarchical appearance representation of the brain tumor in MRI images and was subsequently applied to produce a prediction map of the brain tumor on test images. Finally, the brain tumor was obtained through an optimum thresholding on the prediction map. Results The proposed method was evaluated on both the Multimodal Brain Tumor Image Segmentation (BRATS) Benchmark 2013 training data sets, and clinical data from our institute. A dice similarity coefficient (DSC) and sensitivity of 0.78 and 0.81 were achieved on 20 BRATS 2013 training data sets with High Grade Gliomas (HGG), based on a two-fold cross-validation. The HNN model built on the BRATS 2013 training data was applied to 10 clinical data sets with HGG from a locally developed database. DSC and sensitivity of 0.83 and 0.85 were achieved. A quantitative comparison indicated that the proposed method outperforms the popular fully convolutional network (FCN) method. In terms of efficiency, the proposed method took around 10 hours for training with 50,000 iterations, and approximately 30 seconds for testing of a typical MRI image in the BRATS 2013 data set with a size of 160×216×176, using a DELL PRECISION workstation T7400, with an NVIDIA Tesla K20c GPU. Conclusions An effective brain tumor segmentation method for MRI images based on a HNN has been developed. The high level of accuracy and efficiency make this method practical in brain tumor segmentation. It may play a crucial role in both brain tumor diagnostic analysis and in the treatment planning of radiation therapy. This article is protected by copyright. All rights reserved.

Radiobiology for the Radiologist

Abstract: The text consists of two sections, one for those studying or practicing diagnostic radiology, nuclear medicine and radiation oncology; the other for those engaged in the study or clinical practice of radiation oncology--a new chapter, on radiologic terrorism, is specifically for those in the radiation sciences who would manage exposed individuals in the event of a terrorist event. The 17 chapters in Section I represent a general introduction to radiation biology and a complete, self-contained course especially for residents in diagnostic radiology and nuclear medicine that follows the Syllabus in Radiation Biology of the RSNA. The 11 chapters in Section II address more in-depth topics in radiation oncology, such as cancer biology, retreatment after radiotherapy, chemotherapeutic agents and hyperthermia.

Deep learning in medical imaging and radiation therapy.

Abstract: The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and strategies that researchers have taken to address these challenges; and (c) identify some of the promising avenues for the future both in terms of applications as well as technical innovations. We introduce the general principles of DL and convolutional neural networks, survey five major areas of application of DL in medical imaging and radiation therapy, identify common themes, discuss methods for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.