CME Radiology Grand Rounds

Focused ultrasound (FUS) lesioning of the brain has been of interest since its inception in the 1950s, but it has never been accepted because of the limitations with sonicating through the intact human cranium. Recent advances in transducer technology have now enabled for precise transcranial focusing deep within the brain. High intensity focused ultrasound combined with MRI allows for a theoretically powerful technology where noninvasive lesioning can be performed deep within the brain with continuous clinical and radiographic monitoring. The application of this technology for the treatment of neurologic disorders will be presented including the results of a phase 1 clinical trial of FUS thalamotomy for the treatment of essential tremor.

Emerging from the early recognition that computers might play a powerful role in the future of health care and medical research, biomedical informatics has become a mature but interdisciplinary discipline that draws on many component sciences. These include computer science, decision science, information science, statistics, organizational theory, and management science—as well as the clinical and basic biomedical sciences. In light of the increasing recognition of its key role in modern care delivery as well as clinical and translational research, I offer a definition of the field and a perspective on its role in clinical medicine, public health, and biomedical research. Its academic role as well as its accomplishments and challenges in practice settings will be discussed. Imaging is an important element in the broad spectrum of the informatics discipline. I will close with some thoughts about future directions and synergies with other biomedical disciplines, since informatics does not exist in a vacuum and all informatics research is inherently collaborative.

Endoscopic microscopy is a new field where microscopic images are obtained from living patients. This capability opens up possibilities for obtaining histopathologic diagnoses from tissues that are difficult or unsafe to sample, screening entire organs for occult microscopic disease, and understanding disease mechanisms in vivo. In this talk, I will describe some endoscopic microscopy techniques developed, including optical coherence tomography (OCT), optical coherence microscopy (OCM), and confocal microscopy (CM) and will discuss how these methods can potentially impact patient care.

Breast MRI is the most sensitive clinical test for breast cancer detection and has huge future clinical potential. Yet, the use of Breast MRI in the preoperative setting has been questioned despite finding additional cancers. Additionally, since 2007 breast MRI screening has been recommended for high risk patients yet many are not referred for screening. Assessment of treatment response is currently best performed by breast MRI but are there any new promising competing modalities? A review of the current indications and applications of breast MRI and where we may be headed.

MRgFUS is an emerging treatment modality for the noninvasive treatment of a variety of disorders. It is FDA approved for the treatment of uterine fibroids and is in clinical trials for the treatment of neuropathic pain, essential tremor, brain tumors, painful bone metastases, prostate cancer, and liver cancer. For these applications, MRI provides image guidance for targeting, temperature imaging, and treatment assessment. Focused ultrasound is also being investigated for neuromodulation and for opening the blood brain barrier for drug delivery. This talk will review the clinical applications, ongoing basic science research, as well as discuss current work to address ongoing engineering challenges.

The National Lung Screening Trial (NLST) was a randomized trial in 53,454 participants comparing low dose helical CT (LDCT) to chest radiography (CXR) in individuals at high risk of lung cancer. Published in 2011, the results showed a 20% relative reduction in lung cancer mortality using LDCT relative to CXR; a 6.7% decrease in all-cause mortality was also observed. Data on screening results, types of diagnostic procedures, screen-related complications, lung cancers and stage distribution are presented. Preliminary calculations of cost effectiveness will be described, along with screening-related harms and how these factors may influence the implementation of CT screening in routine practice.

High resolution and high field Magnetic Resonance Neurography is shown to have excellent anatomic capability. There have been considerable advances in the technology in recent years leading to various feasibility studies using different structural and functional imaging approaches in both clinical and research settings. This talk will discuss high resolution 3 Tesla MR Neurography techniques, outline the normal and abnormal imaging appearances of the peripheral nerves, discuss the accuracy of the technique in the smaller and larger peripheral nerves and its clinical impact with relevant surgical correlations. Finally, future directions including functional imaging and whole body MR Neurography will be discussed.

Errors in clinical medicine have received a great deal of attention since the publication by The Institute of Medicine of the report entitled to err is human. An abundant literature exists on the types of errors encountered in adult radiology. However, disease entities encountered in pediatric radiology practice vary considerably from adult practice and the frequency, types and causes of diagnostic error in imaging children have not been studied widely. The goals of this presentation are to describe the patterns of diagnostic error in our subspecialty, to identify situations in which risk of diagnostic error is high, and to suggest strategies for systematic improvement.

Introduction of functional imaging (fMRI) and high fields for brain imaging coincides in time. Since their introduction approximately two decades ago, there has been a revolution in the ability to image brain function going from early experiments demonstrating relatively course images of activity in the visual cortex to mapping cortical columns and to “brain reading” that constructs mental experiences of an individual. These functional studies have been complemented with imaging of morphology, providing increasingly exquisite depictions of cerebral blood vessels, axonal fibers, myelin distribution etc. These developments have been marked by incessant improvements in instrumentation, image acquisition and reconstruction methods, and, in case of fMRI, a significant expansion in our knowledge of neurovascular coupling. Collectively, this body of work has brought us recently to the point of depicting functional activity and anatomy in three dimensions in the entire human brain with submillimeter resolution. The pace of development and discovery continues unabated after two decades, with increasing use of ultrahigh magnetic fields (7 Tesla and higher) that exploit new engineering solutions to ultrahigh field challenges, accelerations in image acquisition speeds that enable high resolution whole brain imaging in the subsecond domain, incorporation of motion correction approaches that allow us to exploit the high resolution acquisition capability, and the use of novel image analysis methods that in turn impact how we acquire images. Aspects of these developments will be reviewed and their implications for future developments will be discussed.

Molecular imaging is not only an increasingly used laboratory tool but is beginning to be implemented clinically – both for identifying pathological tissue non-invasively and for guiding therapy. Examples derived primarily from oncology will be discussed, with an emphasis on how the reagents and techniques developed can be used in current clinical practice. Some of the challenges inherent to clinical translation of molecular imaging and strategies to overcome them will also be discussed.

Currently, tumor staging and re-staging involves acquiring and repeating a series of diagnostic tests, which are time-consuming, cause direct and indirect accumulated expenses, accumulated radiation exposures, and in children, may require several or prolonged sedations. The ultimate goal of our research is to develop one single, noninvasive staging test, which could provide a sensitive detection of primary tumors and their metastases, a specific characterization of the cellular/molecular composition and aggressiveness of malignant lesions and quantitative assessment of response or failure to treatment. Towards this end, we are investigating novel, second-generation ultrasmall superparamagnetic iron oxide nanoparticles (USPIO; hydrodynamic diameter <50 nm), which have recently entered clinical applications (e.g. ferumoxytol/Feraheme) or which are expected to enter the clinic very soon (P904, GEH121333). This presentation will discuss distinguishing physico-chemical, physiological and safety characteristics of these novel USPIO compared to first generation USPIO and SPIO (> 50 nm). Examples from our pre-clinical investigations and our ongoing clinical trial will demonstrate how these novel USPIO can be utilized to improve tumor staging with whole body MR Imaging, to detect tumor-associated inflammation, and to monitor therapy response. It is important to understand capabilities and limitations of these new nanoparticle-based imaging approaches, as they are being integrated into hybrid imaging techniques and expected to develop rapidly within the clinical realm. The close interaction of nanoparticle development and clinical advances in novel, cuttingedge imaging technologies will provide a unique platform to realize novel, non-invasive, more efficient, and more accurate approaches for “one-stop-shop” staging of cancer patients.

Cancer Immunotherapeutics are progressing rapidly into the human clinical arena. At this time the immune response is generally studied by sampling the peripheral blood, assuming that the cells of interest are trafficking in this compartment rather than having made their way to the target site. In mice, animals can be sacrificed and tumors, spleen and lymph nodes can be sampled to gain a better understanding of the immune response. Given the impracticalities of performing similar studies in human, it is paramount to develop imaging studies of the immune response in humans. A variety of potential imaging technologies will be presented; optical, nuclear, and MRI based.

Worldwide, lung cancer is the most common cause of major cancer mortality in men and the second most common in women. Recent therapeutic advances have led to a revolution in the lung cancer field in discovering therapeutically tractable oncogene dependency, that have major implications for patient evaluation and approach to diagnosis. The 2011 IASLC/ATS/ERS Classification of Lung Adenocarcinoma addresses these issues. This classification was developed based on an evidence-based approach by an international multidisciplinary panel including pathologists, oncologists/respiratory physicians, radiologists, molecular biologists, and thoracic surgeons. Multiple paradigm shifts are outlined that have a major impact on clinical practice for pathologists as well as the entire multidisciplinary team caring for lung cancer patients.

Major changes are also recommended for adenocarcinomas diagnosed in resection specimens:1 1) the term bronchioloalveolar carcinoma (BAC) should not be used anymore, because tumors previously classified under as BAC are represented by five different tumors in this classification; 2), for solitary tumors measuring ≤3cm, new concepts of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) have been introduced for lesions that have no invasion or ≤5mm invasion, respectively; these patients should have 100% or near 100% disease free survival (DFS); 3) for invasive adenocarcinomas, comprehensive histologic subtyping is recommended for evaluation with classification according to the predominant subtype; 4) micropapillary adenocarcinoma is proposed as a new subtype with a poor prognosis; 5) the term lepidic replaces BAC for tumors with a predominant component formerly called non-mucinous BAC, and the term lepidic predominant adenocarcinoma is recommended along with discontinuing the term “mixed subtype”; 6) invasive mucinous adenocarcinoma (IMA) is the term used to replace those formerly classified as mucinous BAC. IMA are strongly correlated with KRAS mutation Recently strong survival correlations were demonstrated using this classification in Stage I adenocarcinomas with the following 5-year DFS: AIS and MIA (100%), lepidic (90%), acinar (84%), papillary (83%), and there was a poor prognostic group: micropapillary (67%), solid (70%), colloid (71%) predominant tumors as well as invasive mucinous adenocarcinoma (75%).15 Comprehensive histologic subtyping is performed by making semiquantitative estimation of each of the patterns in 5% increments. A deliberate choice needs to be made to give one pattern the largest percentage. It is useful to record in diagnostic reports each adenocarcinoma subtype that is present with the percentages. This approach may also provide a basis for architectural grading of lung adenocarcinomas. Since this classification was initially published there are a growing number of studies of resected lung adenocarcinomas that have demonstrated its utility in identifying significant prognostic subsets and molecular correlations according to the predominant patterns. Several articles also have shown good radiologic pathologic correlation with this proposed classification.

In summary this classification outlines many paradigm shifts in lung cancer diagnosis that crystalize the importance of histology and genetics in personalized medicine for lung cancer patients. Evidence-based recommendations are made that will transform the clinical practice of all physicians involved with lung cancer diagnosis including radiologists.

Computed Tomography (CT) is a powerful diagnostic imaging tool, with more than 70 million CT scans obtained in 2007 in the US alone. CT technology has dramatically evolved over the last decades, and substantial innovations are introduced into clinical practice in relatively short intervals. While novel powerful CT technology has allowed a wide range of new applications to emerge, this poses several challenges: (1) First, the essence of the innovations published in the technical literature (if published at all) is not necessarily easy to understand. The more digestible marketing material provided by the vendors is typically insufficient to assess and compare scanner capabilities. (2) Once purchased, the implementation of new CT technology into an existing practice is another challenge: New protocols have to be developed, and new applications have to be specified and coordinated with existing equipment. This is not straightforward, because clinical strengths and limitations of a new technology have not yet been evaluated. Early ‘clinical’ publications from ‘luminary sites’ tend to overstate the strengths of a new system. (3) Third, new scanner capabilities have substantially increased the complexity of CT scanner operation, not only for special applications. (4) Finally, there is growing public concern regarding radiation exposure from CT with recent regulatory reverberations in California. While the overall benefits of a clinically indicated CT are usually not in doubt, the rapid evolution of CT makes it difficult to design and interpret clinical effectiveness trials which require long study periods to complete.

This presentation will provide a brief overview of current CT scanner technology. Innovations in hardware, such as x-ray tubes, numbers of x-ray sources, detector bank design, as well as new scanning modes, dual energy / spectral CT, and new image reconstruction algorithms will be briefly explained. The main portion of this talk will focus on the most recent CT scanner upgrades and purchases at Stanford Hospital and Clinics, as an illustration how new technology is embedded in a highly subspecialized academic practice. This will include a discussion of the scanner specifications, strengths and limitations of each machine, and how these factors determine expected applications, and scanner locations. This talk will also explain how we plan to implement new protocols, train technologists and house-staff, and maintain enterprise wide dose hygiene and regulatory standards.

Technical emphasis will be on latest dual-source CT technology (Siemens Flash; Stanford Hospital), and on latest iterative image reconstruction technology (‘VEO’, GE Discovery 750HD; SMIC). The secondgeneration dual-source technology provides unprecedented temporal resolution and acquisition speeds, for all acute, pediatric, and cardiovascular applications. Iterative image reconstructions have great dose savings potential, but the relationship between radiation dose, image noise, and image quality is fundamentally different when compared to current CT technology. This is important to understand in order to take advantage of the full capabilities of such new technology.

The world of scientific publishing is changing. The pace of change continues to increase in medical publishing, the role of the journal is evolving but the need for high quality, relevant content remains greater than ever. Information volume is increasing, online interconnectivity and means of access to information, for education or research will demand technology developments that increase journal functionality, discoverability, and deliverability. This lecture will introduce these themes from the market at large and the perspective of the Journal of Vascular and Interventional Radiology.

The increasing ability to establish genotypic characteristics of individual patients and the genotypes of diseases such as cancer and the ability to establish better phenotypic cohorts is resulting in more precise diagnosis. In turn, this will allow more precise selection of therapy. Imaging is poised to play major roles in the age of precision medicine. Among many other applications, imaging biomarkers serve as proxies for phenotype manifestation of disease and imaging is often the most immediate and best way of assessing response to therapy with the unique ability to survey the entire body.