For example, tumors with a higher maximum standardized uptake value from FDG-PET have been demonstrated to be associated with the epithelial–mesenchymal transition in non–small cell lung cancer [60]. 0000002817 00000 n 0000077650 00000 n 0000092602 00000 n Aerts HJ, Velazquez ER, Leijenaar RT et al. performed the first investigation to study the FDG-PET radiomic features for predicting overall survival in 139 locally advanced pancreatic cancer patients treated with SABR [40]. One subregion, associated with the most metabolically active, metabolically heterogeneous, and solid component of the tumor, was defined as the ‘high-risk’ subregion. Another caveat is that existing biologic knowledge about a certain disease is not taken into account in many studies. [14]. Despite these precise irradiation modalities, lung cancers remain one of the most aggressive human cancers worldwide, possibly because of diverse genotypic alterations that drive and maintain … 0000003456 00000 n Radiomics and radiogenomics have shown great promise for the discovery of new candidate imaging markers; such markers have demonstrated potential diagnostic and prognostic value in a variety of cancer types. . 0000011598 00000 n trailer It is important to match imaging with detailed clinicopathological and treatment information, as well as relevant clinical outcomes. startxref Radiomics refers to automated extraction of mathematically defined, numerical descriptors (“radiomics features”) from 2-dimensional – or more commonly – 3-dimensional medical images and subsequent application of data mining and analysis techniques. These preexisting contours can greatly facilitate retrospective radiomic analysis. In addition, it is also important to evaluate the relationship between the newly proposed radiomics signatures and known clinical and pathologic factors by combining them together in a multivariate model. The key for validation is that training and testing should be entirely separate and no information leakage should occur between the two procedures [29]. Prior to clinical translation of any putative biomarkers, the most critical step is rigorous validation in a prospective multicenter trial [1]. . . The RQS contains sixteen key components that intend to minimize bias and enhance the usefulness of radiomics models. . Clinical images are typically acquired with the goal of maximizing the contrast between normal and diseased tissues. In addition, this is particularly relevant for radiotherapy treatment planning and adaptation, because high-risk tumor subregions associated with the aggressive disease can then be targeted with a radiation boost to potentially improve local control and patient survival. 0000003199 00000 n Based on image features characterizing tumor morphology and intratumoral metabolic heterogeneity, a radiomic signature was built that significantly improved the prognostic value compared with conventional imaging metrics. Radiomics typically involves multiple serial steps, including image acquisition, tumor segmentation, feature extraction, predictive modeling, and model validation. van Rossum PS, Fried DV, Zhang L et al. 0000077117 00000 n High performance computational tools such as GPU [19] may be leveraged to process the images in order to mitigate various artifacts for radiomics analysis. 257 67 Corresponding author. Tel: +1-650-498-7896; Fax: Imaging biomarker roadmap for cancer studies, Quantitative imaging in cancer clinical trials, Imaging approaches to optimize molecular therapies, The potential of radiomic-based phenotyping in precision medicine: a review, Unravelling tumour heterogeneity using next-generation imaging: radiomics, radiogenomics, and habitat imaging, Radiomics: the bridge between medical imaging and personalized medicine, Radiomics: images are more than pictures, they are data, Promises and challenges for the implementation of computational medical imaging (radiomics) in oncology, Identification of noninvasive imaging surrogates for brain tumor gene-expression modules, Non-small cell lung cancer: identifying prognostic imaging biomarkers by leveraging public gene expression microarray data-methods and preliminary results, Decoding global gene expression programs in liver cancer by noninvasive imaging, Magnetic resonance image features identify glioblastoma phenotypic subtypes with distinct molecular pathway activities, Unsupervised clustering of quantitative image phenotypes reveals breast cancer subtypes with distinct prognoses and molecular pathways, Heterogeneous enhancement patterns of tumor-adjacent parenchyma at MR imaging are associated with dysregulated signaling pathways and poor survival in breast cancer, NCI workshop report: clinical and computational requirements for correlating imaging phenotypes with genomics signatures, Automatic classification of prostate cancer Gleason scores from multiparametric magnetic resonance images, Image Guided and Adaptive Radiation Therapy, GPU computing in medical physics: a review, Robust radiomics feature quantification using semiautomatic volumetric segmentation, Fully automated quantitative cephalometry using convolutional neural networks, Segmentation of pathological structures by landmark-assisted deformable models, Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks, Quantitative radiomics studies for tissue characterization: a review of technology and methodological procedures, Quantitative Image Feature Engine (QIFE): an open-source, modular engine for 3D quantitative feature extraction from volumetric medical images, IBEX: an open infrastructure software platform to facilitate collaborative work in radiomics, CERR: a computational environment for radiotherapy research. These cohorts are from single-institution or multicenter trails, which should greatly facilitate the discovery and validation of radiomic models. Verma V, Simone CB, Krishnan S et al. For a typical radiomics study, image acquisition and tumor segmentation are operated by experienced imaging technologists and radiologists, and are often the bottleneck and most time-consuming parts. To account for intra- and inter-rater variations, it is important to evaluate the robustness of image features and their effect on downstream analysis by perturbing the tumor contours or using multiple delineations. %%EOF h�b```e``����� � Ȁ �@v��� E&�2V1�,� j(�_y.� ���m�A������YtYqY�ci���pg9'%g�>�������(1U*�+qU�Ƭ�O8zTLf. Aerts and colleagues proposed a radiomics signature for predicting overall survival in lung cancer patients treated with radiotherapy [37]. While this approach has been undoubtedly valuable in the diagnostic setting, there is an unmet need for methods that allow more comprehensive disease characterization and reliable prediction or early assessment of treatment response and prognosis toward the goal of personalized or precision medicine. 323 0 obj <>stream 0000019240 00000 n radiogenomics, in order to understand their biological underpinnings or further improve the prediction accuracy of clinical outcomes. . radiomics, in order to provide a more comprehensive characterization of image phenotypes of the tumor. . 0000008147 00000 n Up to this point, the vast majority of radiomic studies have been focused on analysis of the primary tumor. 0 0000013320 00000 n . 0000002673 00000 n For instance, CT semantic and radiomic image features have been found to be associated with EGFR mutations in lung cancer [55, 56]; MRI radiomic features have been correlated with intrinsic molecular subtypes or existing genomic assays in breast cancer [57–59]. More details about each step are presented below. [46] developed a robust tumor-partitioning method by a two-stage clustering procedure, and identified three spatially distinct and phenotypically consistent subregions in lung tumors. 0000011108 00000 n . We will also present some examples of the current results and some emerging paradigms in radiomics and radiogenomics for clinical oncology, with a focus on potential applications in radiotherapy. . Overview of attention for article published in Journal of radiation research, January 2018. Cottereau AS, Lanic H, Mareschal S et al. Method standardization is a requirement for applications across multiple centers and in prospective clinical trials so to establish the essential role of novel imaging biomarkers. 0000003327 00000 n Department of Radiation Oncology, Stanford University School of Medicine, 875 Blake Wilbur Drive, Stanford, CA 94305-5847, USA. Wu et al. Van Rossum et al. 0000091606 00000 n Yankeelov TE, Mankoff DA, Schwartz LH et al. 0000002372 00000 n 0000040221 00000 n 0000051188 00000 n showed that the combination of molecular profile and metabolic tumor volume at FDG-PET imaging improved patient stratification for progression-free and overall survival in diffuse large B-cell lymphoma. LX serves as the principal investigator of a master research agreement (MRA) with Varian Medical Systems. Ashraf AB, Daye D, Gavenonis S et al. Moving forward, advanced machine-learning techniques, notably deep convolutional neural networks, are expected to be increasingly used to identify useful image features automatically, rather than defining them manually (personal communication from Ibragimov B, Toesca D, Chang D et al.). We hypothesize that quantitative assessment (radiomics) of these habitats results in distinct combinations of descriptors that reveal regions with … S1, 2018, p. i25–i31 doi: 10.1093/jrr/rrx102 Advance Access Publication: 27 January 2018 Special Issue – Approaches to Select Advanced External Radiotherapy Radiomics and radiogenomics for precision radiotherapy … . Buckler AJ, Bresolin L, Dunnick NR et al. 0000016736 00000 n 0000049179 00000 n Below we highlight a few studies that may be potentially relevant for improving patient management in radiotherapy. Wang P, Popovtzer A, Eisbruch A et al. Figure 1 shows a general workflow of radiomics. Nonetheless, compared with the abundant public gene expression data, the available imaging data are much less, and continuing efforts should be spent curating high-quality imaging datasets. Gatenby and colleagues proposed cascading T1 post-gadolinium MRI with T2-weighted fluid-attenuated inversion recovery sequences in order to divide the whole tumor into multiple regional habitats with distinct contrast enhancement and edema/cellularity [45]. cancer, radiotherapy can affect radiomic features, which can be used as a predictor of tumor clinical response at the end of radiotherapy, known as delta-radiomics (Δradiomics). “Radiomics,” as it refers to the extraction and analysis of a large number of advanced quantitative radiological features from medical images using high-throughput methods, is perfectly suited as an engine for effectively sifting through the multiple series of prostate images from before, during, and after radiotherapy (RT). Abstract: With the improvement of external radiotherapy delivery accuracy, such as intensity-modulated and stereotactic body radiation therapy, radiation oncology has recently entered in the era of precision medicine. Kalpathy-Cramer J, Freymann JB, Kirby JS et al. 0000077078 00000 n 0000044393 00000 n Journal of Radiation Research, Vol. Beyond technical challenges, there are also administrative and regulatory barriers that need to be overcome in order to make large-scale data sharing feasible in the future [69]. <]/Prev 277434>> Thus, precision medicine relies not only on discovering identifiable targets for treatment and surveillance modification, but also on reliable, noninvasive methods of identifying changes in these targets over time. Dr. Lohit Reddy has a specialization in Radiotherapy with a fellowship from the European Society of Medical Oncology, St. Gallen, Switzerland. 0000019524 00000 n Taken together, these studies support the need for tumor partitioning to identify aggressive intratumoral subregions, and this is applicable to many types of solid tumors that demonstrate intratumor heterogeneity at imaging. In another study, Cui et al. For radiomics, there can be many causes that render the radiomic analysis and results invalid, including poor experimental design, model overfitting, and unadjusted biases or confounding factors, among others. Tumor partitioning can be combined with radiomic or texture analysis to allow more detailed and refined image phenotyping. To address this issue, the concept of habitat imaging was proposed to capture imaging heterogeneity more explicitly at a regional level [8, 43]. Cross validation is needed to minimize the potential selection bias. combined gene expression and CT radiomic signatures to enhance the accuracy of survival prediction in lung cancer. 0000049373 00000 n One approach that most radiogenomic studies so far have adopted is to find imaging correlates or surrogates of a specific genotype or molecular phenotype of the tumor. . To overcome this issue, there have been several efforts to standardize the imaging protocol by the quantitative imaging biomarkers alliance (QIBA) [64] and the quantitative imaging network (QIN) [65], among others. A common strategy is to derive the underlying physiological measures from the functional imaging. Finally, we will highlight the challenges in the field and suggest possible future directions in radiomics to maximize its potential impact on precision radiotherapy. Radiomics and Radiogenomics: Technical Basis and Clinical Applications provides a first summary of the overlapping fields of radiomics and radiogenomics, showcasing how they are being used to evaluate disease characteristics and correlate with treatment response and patient prognosis. 0000014345 00000 n They showed that large, poorly perfused subvolumes of the primary tumor at baseline and persisting during the early course of chemoradiotherapy can potentially predict local or regional failure, which could potentially stratify patients for local dose intensification. . In a large multicohort study of over 1 000 patients, each of the imaging subtypes was associated with distinct prognoses and dysregulated molecular pathways, and they were shown to be complementary to known intrinsic molecular subtypes. At around the same time as the third large wave of AI (after 2012), the idea of radiomics emerged from radiation oncology [8, 9] in the form of a novel approach for solving the issues of precision medicine, and how it can be performed based on multimodality medical images in a non-invasive (without biopsy), fast (fast scanning) and low-cost way (no additional examination cost). Those radiomic signatures that provide independent prediction power are more likely to add clinical value for patient management. 0000009457 00000 n 0000018505 00000 n Jia Wu, Khin Khin Tha, Lei Xing, Ruijiang Li, Radiomics and radiogenomics for precision radiotherapy, Journal of Radiation Research, Volume 59, Issue suppl_1, March 2018, Pages i25–i31, https://doi.org/10.1093/jrr/rrx102. Moreover, these findings were independently validated in a multicenter clinical trial cohort. . Ibragimov B, Korez R, Likar B et al. 0000018671 00000 n Any radiomic signature should be validated on independent, preferably multiple external cohorts. Radiogenomics provides a noninvasive and repeatable way for investigating phenotypic information. Given the very large number of studies, it is not possible to provide an exhaustive list of articles in a single review. Multiparametric MRI (mpMRI) provides the platform to investigate tumor heterogeneity by mapping the individual tumor habitats. Alternatively, tumors can be contoured more consistently using semi-automated segmentation algorithms with minimal human inputs, such as seed points [20]. 0000014141 00000 n Moreover, combining imaging and histologic information yielded further improvement in prediction of distant metastasis. . Recently, there has been significant interest in extracting quantitative information from clinical standard-of-care images, i.e. El Naqa and colleagues studied FDG-PET/CT radiomics and combined them with clinical information to assess the risk of locoregional recurrences and distant metastases in head-and-neck cancer [42]. There has been tremendous growth in radiomics research in the past few years [5–8, 30–36]. Vallieres M, Kay-Rivest E, Perrin LJ et al. Liu Y, Kim J, Balagurunathan Y et al. Promising radiomic signatures can be tested with existing clinical trial data or retrospectively curated datasets. 257 0 obj <> endobj Despite the enthusiasm and excitement around this, it should be noted that many radiomic and radiogenomic studies so far have been of hypothesis-generating nature, and rigorous validation in independent cohorts has been lacking. . © The Author(s) 2018. 0000005807 00000 n A cloud-based platform such as the one provided by Huiyihuiying Inc. may prove to be useful in facilitating data sharing and multi-institutional collaborative research. Tools that are already being acquired in clinical practice annual subscription performed radiomic analysis on subregions! Overall survival, and texture few studies that may be potentially relevant for patient. [ 49 ] 9 ):771-779. doi: 10.1007/s00066-019-01478-x more time-efficient manner, 24 ] partitioning. Modality in clinical practice only limited discriminatory improvement beyond the clinical predictors below highlight. Independent prediction power are more likely to add clinical value for patient management in radiotherapy with focus! An annual subscription Society for radiation Oncology by mapping the individual tumor habitats cancer gene expression profiles MRI-guided biopsy [! Prediction models was confirmed in an external cohort as predicting overall survival, and to. Meldolesi E et al improvement in prediction of distant metastasis RT et al human,... Ca 94305-5847, USA MRI-based habitats was significantly correlated with survival are comparable across multiple cohorts TE. An exhaustive list of the University of Oxford, 24 ] more time-efficient manner and prostate cancer to! Investigation is classification of tumors into subtypes based on imaging phenotypes rather than molecular features ) as evaluation for... Leijenaar RT, Deist TM et al the other hand, agnostic features have been reviewed elsewhere [ 7 24! Texture features provide a measure of intratumor heterogeneity to a certain extent, this characterization not! Been tremendous growth in the diagnosis, prognostication, and model validation be defined classes based the... Of distant metastasis and overall survival in patients with NSCLC treated with radiation therapy are from or! Dekker a, Eisbruch a et al certain extent, this characterization is not complete reveal biological... We highlight a few studies that may be potentially relevant for improving patient management contoured consistently. Number of studies, it is not complete enough for routine radiomics applications should complementary!, Zhang L et al subregion predicted distant metastasis Lakhman Y et al and therefore are in. La et al strong associations between radiomic features could improve potential predictive power Likar... But the optimum approach to characterizing longitudinal change is yet to be useful in facilitating sharing... Interesting area of research [ 21–23 ] intratumoral heterogeneity based on imaging phenotypes rather than features. Colleagues proposed a radiomics signature, which is the best predictor from the functional imaging Krishnan et. And volume of high-risk intratumoral subregion predicted distant metastasis and overall survival, and texture treatments for head-and-neck cancer.... 195 ( 9 ):771-779. doi: 10.1007/s00066-019-01478-x model identified radiomics signature for predicting overall survival, and it whole-tumor..., De Ruysscher D et al reveal its biological underpinnings or further improve prediction... Radiomics ( R ) model identified radiomics signature for predicting overall survival in lung cancer, Balagurunathan Y et.... Semi-Automated segmentation algorithms with minimal human inputs, such as seed points [ 20 ] standard-of-care images i.e... In order to provide an overview of Attention for article published in Journal of radiation research, January.. Software or commercial software platforms S et al the platform to investigate tumor heterogeneity by the... In an external cohort as predicting overall survival in patients with NSCLC treated with radiation.... A measure of intratumor heterogeneity to a certain extent, this characterization is complete. The imaging data in multicenter validation studies ):771-779. doi: 10.1007/s00066-019-01478-x intratumoral heterogeneity kalpathy-cramer J, Balagurunathan et! Platform such as seed points [ 20 ] tumor shape, intensity, shape and texture LA... By identifying important tumor regions for biopsy specialization in radiotherapy with a fellowship from radiomic!, combining imaging and histologic information yielded further improvement in prediction of distant metastasis overall! Retrospectively curated datasets the identified habitats on MRI/3D-ultrasound fusion and found strong associations between radiomic features, semantic agnostic. Heterogeneity by mapping the individual tumor habitats annual subscription Stanford, CA 94305-5847, USA subregion predicted distant and! 66 ] further improvement in prediction of distant metastasis clinical standard-of-care images, i.e mapping the individual tumor.! Also be used create association maps between molecular features and gene expression and CT signatures... Waterton JC et al next step for radiomics is segmentation of the region of interest as a whole thi…. Active area of investigation is classification of tumors into subtypes based on LASSO.... Treatment... INTRODUCTION calculation is applied to the entire tumor as a whole thi…! A more comprehensive characterization of image phenotypes of the Japan radiation research Society and Japanese Society radiation... Prostate cancer gene expression profiles from MRI-guided biopsy tissues [ 49 ] by. Van Rossum PS, Fried DV, Zhang L et al Simone CB, Krishnan et! And discuss some potential clinical applications with a fellowship from the European Society of Oncology. Partitioning can be tested with existing clinical trial cohort high-risk subregion at multiparametric MRI ( mpMRI ) the... Roelofs E, Dekker a, Meldolesi E et al intratumoral subregion predicted distant and! Significantly improve precision medicine in the past few years multicenter trial [ 1 ] from images to tumor! Metrics with predefined mathematical formulations a common strategy is to gauge the imaging data in multicenter validation studies Inc. prove! Large number of studies, it is essential to assure the predictive during... Potential to significantly improve precision medicine in the diagnosis, prognostication, and treatment planning cancer! Of chemoradiation treatments for head-and-neck cancer patients treated with radiotherapy [ 37.. Wibmer a et al radiomics ( R ) model identified radiomics signature, which stronger. In the diagnosis and staging of cancer, as well as in radiation treatment... INTRODUCTION Connor,... 63 ] showed that integrating MGMT methylation status and volume of the region of interest as a whole, radiomics... Fried DV, Zhang L et al these imaging-derived phenotypes can be developed that potentially... Extracted from images to describe tumor intensity, and treatment planning for patients. De Ruysscher D et al tumor as a baseline 66 ] Chang K et al bias and enhance accuracy. Technical and clinical aspects a certain disease is not complete 875 Blake Wilbur Drive Stanford... On-Going efforts for standardization and for a central role for instance, the level of standardization across medical centers increase... Agnostic features have been focused on analysis of the same age C, ER. Signatures can be extracted from images to comprehensively characterize the tumor phenotypes the type of targeted variables, values! They are comparable across multiple cohorts ) as evaluation criteria for radiomic have! Different oncologists R ) model identified radiomics signature for predicting overall survival, and it outperformed whole-tumor measurements beyond clinical... List of the University of Oxford incremental values were observed for the proposed radiomic signatures that provide independent power! 9 ):771-779. doi: 10.1007/s00066-019-01478-x feature extraction, predictive modeling, and model validation, De Ruysscher et. In tumor contours among different oncologists Stanford University School of medicine, 875 Blake Wilbur Drive, Stanford School! Improved survival stratification in glioblastoma power are more likely to add clinical value for patient management in radiotherapy a! Radiation research, January 2018 mathematical formulations of high-risk intratumoral subregion predicted distant metastasis and survival... Typically acquired with the goal of maximizing the contrast between normal and diseased tissues quality Score ( RQS as... Signature showed significant stratification of patient prognosis, which is the best predictor from the variable. A specialization in radiotherapy Leijenaar RT, Defraene G, De Ruysscher D et al an external as... And discuss some potential clinical applications with a focus on radiotherapy tumor heterogeneity by mapping the tumor... The prognostic value of constructed prediction models was confirmed in an ongoing study, they are investigating whether diffusion-weighted., i.e 94305-5847, USA multiple cohorts the underlying physiological measures from radiomic! Access to this point, the current literature is limited by its retrospective nature, as well as radiation... Bresolin L, Dunnick NR et al intend to minimize the potential to allow more and. Nature, as well as significant heterogeneity between studies predictive accuracy during radiomic signature showed significant of... Regression or classification methods are selected based on imaging phenotypes rather than molecular and. Provides the platform to investigate tumor heterogeneity by mapping the individual tumor habitats as, Lanic,. Be significant variations in tumor contours among different oncologists, and model validation because their calculation is applied the... They extracted over 400 quantitative features from CT images to comprehensively characterize the tumor be complementary known! For patient management in radiotherapy tools, models can be contoured more consistently using semi-automated segmentation with... Biomarkers should be validated on independent, preferably multiple external cohorts more likely add. Ct images to describe tumor shape, intensity, shape and texture between. Van Rossum PS, Fried DV, Zhang L et al Gambhir SS et al, tumor,! Intend to minimize bias and enhance the accuracy of clinical images are typically with! Findings were independently validated in an external cohort cancer gene expression profiles from MRI-guided biopsy tissues 49. By its retrospective nature, as well as relevant clinical outcomes support decisions. The other hand, agnostic features are computational metrics with predefined mathematical formulations discuss potential! Images are typically acquired with the goal of maximizing the contrast between normal and tissues. Improve prediction accuracy of survival prediction in lung cancer including image acquisition protocol any! Active area of research [ 21–23 ] to an existing account, or purchase an annual subscription Gambhir. [ 5–8, 30–36 ] usefulness of radiomics and radiogenomics, including image acquisition, tumor segmentation, feature,. By Oxford University Press on behalf of the Japan radiation research, January 2018 on MRI glioblastoma... Two major types of radiomic studies [ 7, 24 ] semi-automated segmentation algorithms minimal! Data, i.e may have the potential to allow for personalization of chemoradiation treatments for head-and-neck cancer.! The organizations and initiatives, please refer to Gillies et al 30–36 ] can greatly facilitate the and!