1. Zanier, E. R., Bertani, I., Sammali, E., Pischiutta, F., Chiaravalloti, M. A., Vegliante, G., … Chiesa, R. (2018). Induction of a transmissible tau pathology by traumatic brain injury. Brain, 2685–2699. https://doi.org/10.1093/brain/awy193
Traumatic brain injury is a risk factor for subsequent neurodegenerative disease, including chronic traumatic encephalopathy, a tauopathy mostly associated with repetitive concussion and blast, but not well recognized as a consequence of severe traumatic brain injury. The authors show that a single severe brain trauma is associated with the emergence of widespread hyperphosphorylated tau pathology in a proportion of humans surviving late after injury. In parallel experimental studies, in a model of severe traumatic brain injury in wild-type mice, they found progressive and widespread tau pathology, replicating the findings in humans. Scary part: Brain homogenates from these mice, when inoculated into the hippocampus and overlying cerebral cortex of naïve mice, induced widespread tau pathology, synaptic loss, and persistent memory deficits. The observation that TBI-induced tau pathology can be transmitted between wild-type mice and can cause persistent memory deficits satisfies the main requirement for a toxic, bona fide prion.
These results indicate that tau prions are generated in TBI, providing a mechanistic explanation for how a biomechanical insult might trigger self-sustained neurodegeneration. They also conclude that the demonstration of tau prions in human TBI will urgently require studies to determine whether neurosurgical intervention in such patients should conform to well established precautions for transmission of prion disease, as suggested by the Centers for Disease Control for CJD.
8 Figures
2. Trapp, B. D., Vignos, M., Dudman, J., Chang, A., Fisher, E., Staugaitis, S. M., … Rudick, R. A. (2018). Cortical neuronal densities and cerebral white matter demyelination in multiple sclerosis: a retrospective study. The Lancet Neurology, 17(10), 870–884. https://doi.org/10.1016/S1474-4422(18)30245-X
Brains and spinal cords were removed at autopsy from patients, who had died with multiple sclerosis, at the Cleveland Clinic. Visual examination of centimeter-thick slices of cerebral hemispheres was done to identify brains without areas of cerebral white-matter discoloration that were indicative of demyelinated lesions (referred to as myelocortical multiple sclerosis) and brains that had cerebral white-matter discolorations or demyelinated lesions (referred to as typical multiple sclerosis). These individuals with myelocortical multiple sclerosis were matched by age, sex, MRI protocol, multiple sclerosis disease subtype, disease duration, and Expanded Disability Status Scale, with individuals with typical multiple sclerosis.
Brains and spinal cords were collected from 100 deceased patients between May, 1998, and November, 2012, and this retrospective study was done between Sept 6, 2011, and Feb 2, 2018. 12 individuals were identified as having myelocortical multiple sclerosis and were compared with 12 individuals identified as having typical multiple sclerosis. Demyelinated lesions were detected in spinal cord and cerebral cortex, but not in cerebral white matter, of people with myelocortical multiple sclerosis. Cortical demyelinated lesion area was similar between myelocortical and typical multiple sclerosis. Spinal cord demyelinated area was significantly greater in typical than in myelocortical multiple sclerosis.
The authors compared pathological and MRI characteristics of people with myelocortical multiple sclerosis with those of people with typical multiple sclerosis (ie, individuals with cerebral white-matter demyelination). Both groups exhibited similar cerebral white-matter MRI abnormalities. Furthermore, MRI-defined lesions in cerebral white matter in myelocortical multiple sclerosis were associated with, and presumably due to, swelling of myelinated axons. Cortical neuronal density in myelocortical multiple sclerosis brains was significantly decreased compared with that in aged-matched control brains and was similar to neuronal density in typical multiple sclerosis brains. Their findings provide pathological evidence that cerebral white-matter demyelination and neurodegeneration can be independent events in multiple sclerosis.
The findings also highlight the non-specific nature of the MRI abnormalities that are traditionally assessed in patients with multiple sclerosis and their lack of specificity for demyelination. Inclusion of patients with myelocortical multiple sclerosis in immunomodulatory clinical trials could explain the variable responses of individuals to treatment as monitored with MRI.
2 Tables, 4 Figures with single MR figure.
3. Kriegeskorte, N., & Douglas, P. K. (2018). Cognitive computational neuroscience. Nature Neuroscience, 21 (September). https://doi.org/10.1038/s41593-018-0210-5
The authors argue that task-performing computational models that explain how cognition arises from neurobiologically plausible dynamic components will be central to a new cognitive computational neuroscience. They briefly trace the steps of the cognitive and brain sciences and then review several exciting recent developments that suggest that it might be possible to meet the combined ambitions of cognitive science (to explain how humans learn and think) and computational neuroscience (to explain how brains adapt and compute) using neurobiologically plausible artificial intelligence models.
Despite methodological challenges many of the findings of cognitive neuroscience provide a solid basis on which to build. As one example, the findings of face-selective regions in the human ventral stream have been thoroughly replicated and generalized. Nonhuman primates probed with fMRI exhibit similar face-selective regions, which had evaded explorations with invasive electrodes because the latter do not provide continuous images over large fields of view. Localized with fMRI and probed with invasive electrode recordings, the primate face patches revealed high densities of face selective neurons, with invariances emerging at higher stages of hierarchical processing. The example of face perception illustrates, on one hand, the solid progress in mapping the anatomical substrate and characterizing neuronal responses and, on the other, the lack of definitive computational models. The literature does provide clues to the computational mechanism. A brain-computational model of face recognition will have to explain the spatial clusters of face selective units and the selectivity and invariances observed with fMRI and invasive recordings.
3 Figures and 5 side boxes which explain various computation models.
4. Stewart, D. R., Korf, B. R., Nathanson, K. L., Stevenson, D. A., & Yohay, K. (2018). Care of adults with neurofibromatosis type 1: a clinical practice resource of the American College of Medical Genetics and Genomics (ACMG). Genetics in Medicine, 20(7), 671–682. https://doi.org/10.1038/gim.2018.28
A work group of experts sought to determine the prevalence, morbidity and mortality, and available treatments of common and emerging NF1-related clinical problems in adults. Work-group members identified peer-reviewed publications from PubMed. Publications derived from populations and multi-institution cohorts were prioritized. Recommendations for management arose by consensus from this literature and the collective expertise of the authors.
In NF1, malignant peripheral nerve sheath tumor (MPNST), a type of soft-tissue sarcoma, frequently arises from a preexisting plexiform neurofibroma, a benign, congenital lesion affecting approximately 50% of NF1 patients. MPNST is clinically aggressive and tends to metastasize early. The risk of NF1-associated MPNST to ages 30, 50, and 85 years was 8.5%, 12.3%, and 15.8% respectively. These data also showed that high-grade MPNSTs were usually fatal, contributing significantly to NF1 mortality. There are no pathognomonic molecular or immunohistochemical studies for MPNST and the histologic features are nonspecific. Low-grade MPNST accounts for ~ 5% of NF1-associated MPNST and is associated with a 100% ten-year survival, in contrast with high-grade tumors (~20% 5-year survival). Surgery remains the cornerstone of treatment for high-grade MPNST, with the aim of achieving clear margins.
Clinical suspicion (pain, rapid growth, neurologic symptoms, deep, truncal location of PN), and awareness of known risk factors (germline microdeletion of the NF1 locus, previous radiation) remain paramount for early detection of MPNST, which is facilitated by targeted MRI imaging. The optimal use, timing, and utility of 18F-FDG PET, PET/CT, and MRI to screen for MPNST are not known. In the largest study of its kind to date (116 lesions, 59 with histologic confirmation) from 105 NF1 patients, used 18F-FDG-PET and PET/ CT to diagnose NF1-associated MPNST with a sensitivity of 0.89 and a specificity of 0.95. Review of 12 studies of 353 NF1 patients revealed that the maximum standardized uptake value (SUVmax) of 3.5 is a commonly accepted threshold for biopsy of a lesion, but this finding needs validation.
Regarding pheochromocytomas: They should be considered in hypertensive NF1 patients who are over 30 years of age, pregnant, and/ or have paroxysmal hypertension, hypertension-associated headache, palpitations, or sweating. Biochemical or imaging screening in asymptomatic patients with NF1 for pheochromocytoma is not recommended. These appear to be exclusively in the adrenal glands.
NF1 causes an increased risk for several other tumor types. In the Finnish epidemiological study of 1,404 persons with NF1, other non-NF1-specific cancers for which a statistically significant excess has been found include those of the brain and central nervous system, “other endocrine glands, gastrointestinal stromal tumors (GIST), malignant fibrous histiocytoma, and rhabdomyosarcoma. The excess of brain tumors included higher grade, non-optic pathway gliomas. An epidemiologic analysis of adult glioma (n = 489) and adolescent and adult NF1 (n = 2,108) found a 20-100-fold increased risk (vs. US rates) of developing malignant glioma, including glioblastoma multiforme.
5. D’Hondt, S., Van Damme, T., & Malfait, F. (2018). Vascular phenotypes in nonvascular subtypes of the Ehlers-Danlos syndrome: A systematic review. Genetics in Medicine, 20(6), 562–573. https://doi.org/10.1038/gim.2017.138
The Ehlers-Danlos syndrome (EDS) is an umbrella term for a group of clinically and genetically heterogeneous connective tissue disorders. Over the past two decades the Villefranche Nosology (1997) has been the standard for classifying EDS. It recognized six subtypes, most of which were caused by defects in the primary structure of collagen or collagen-modifying enzymes. Recent discoveries have, however, expanded the pathogenic spectrum to include EDS variants that are caused by defects in both noncollagenous extracellular matrix proteins and intracellular processes. This has led to an EDS reclassification: a task that was recently accomplished by an international EDS consortium (2017).
Historically, arterial aneurysm and dissection have been synonymous with the vascular type of EDS (vEDS). This type of EDS is characterized by the presence of a thin, translucent skin, which bruises very easily, and joint hypermobility, which is often confined to the small joints. The clinical picture is, however, dominated by a remarkable vascular fragility that leads to spontaneous rupture of blood vessel walls, often without preceding vascular dilatation or aneurysm formation. Other life-threatening complications include rupture of the gastrointestinal (GI) tract, gravid uterus, or other internal organs, such as liver or spleen. The calculated median survival for vEDS patients is 48 years, with most deaths resulting from arterial rupture. vEDS is caused by heterozygous mutations in the type III procollagen encoding gene COL3A1.
Vascular complications have been described in other, “nonvascular” subtypes of EDS. Most of these reports are anecdotal, and the occurrence of such complications in the different EDS subtypes is not well documented. This review will help familiarize clinicians with the spectrum of vascular complications in nonvascular EDS subtypes as well as guide follow-up and management.
Nonvascular types include Classical (type V collagen), Hypermobility (unknown defect), Kyphoscoliosis (Lysyl hydroxylase deficiency), Arthrochalasia (type I collagen), Dermatosporaxis (type I collagen processing), Periodontitis type, Fibronectin-deficient EDS, Brittle cornea syndrome, Spondylodysplastic, and Familial hypermobility syndrome, among others!
Vascular complications in this “nonvascular” group (n = 100) was categorized as either (i) hematoma (53/100, 53%), (ii) intracranial hemorrhage (18/100, 18%), (iii) spontaneous arterial dissection (16/100, 16%), (iv) arterial aneurysm (5/100, 5%), (v) GI bleeding (1/100, 1%), (vi) perioperative hemorrhage (5/100 5%), or (vii) sporadic vascular complication (2/100, 2%).
Vascular complications, for example, are most frequently reported in musculocontractural EDS (CHST14/DSE) and in Classical-like (clEDS) (autosomal recessive, TNXB gene and Tenascin X protein), being present in about two thirds and half of the patients, respectively.
3 tables, 2 figures, no imaging.
6. Neelapu, S. S., Tummala, S., Kebriaei, P., Wierda, W., Gutierrez, C., Locke, F. L., … Shpall, E. J. (2018). Chimeric antigen receptor T-cell therapy-assessment and management of toxicities. Nature Reviews Clinical Oncology, 15(1), 47–62. https://doi.org/10.1038/nrclinonc.2017.148
Immunotherapy using T cells genetically engineered to express a chimeric antigen receptor (CAR) is rapidly emerging as a promising new treatment for hematological and non‑hematological malignancies. CAR‑T‑cell therapy can induce rapid and durable clinical responses, but is associated with unique acute toxicities, which can be severe or even fatal. Cytokine-release syndrome (CRS), the most commonly observed toxicity, can range in severity from low-grade constitutional symptoms to a high-grade syndrome associated with life‑threatening multiorgan dysfunction; rarely, severe CRS can evolve into fulminant hemophagocytic lymphohistiocytosis (HLH). Neurotoxicity, termed CAR‑T‑cell-related encephalopathy syndrome (CRES), is the second most-common adverse event, and can occur concurrently with or after CRS. Intensive monitoring and prompt management of toxicities is essential to minimize the morbidity and mortality associated with this potentially curative therapeutic approach.
CRES typically manifests as a toxic encephalopathy, with the earliest signs being diminished attention, language disturbance, and impaired handwriting; other symptoms and signs include confusion, disorientation, agitation, aphasia, somnolence, and tremors. In severe
cases of CRES (grade >2), seizures, motor weakness, incontinence, mental obtundation, increased intracranial pressure, papilledema, and cerebral edema can also occur.
MRI and CT scans of the brain are usually negative for any anatomical pathology that would account for the neurotoxicity symptoms observed in patients treated with CAR‑T‑cell therapy, although rare cases of reversible T2/fluid attenuated inversion recovery (FLAIR) MRI hyperintensity involving the thalami, dorsal pons, and medulla, and cerebral edema have been reported.
Life‑threatening cerebral edema, although very rare in patients treated with cellular immunotherapy, tends to have a very rapid course with ensuing brain death within 24 h. Notably, in March 2017, five deaths attributed to cerebral edema were reported following treatment of the patients with one anti‑CD19 CAR‑T‑cell product (JCAR015) as part of a multicenter clinical trial. The sponsor has now halted development of this agent.
3 figures, 3 tables and 5 boxes with various treatment regimens, and toxicity grading schemes.
7. Hosny, A., Parmar, C., Quackenbush, J., Schwartz, L. H., & Aerts, H. J. W. L. (2018). Artificial intelligence in radiology. Nature Reviews Cancer, 18(8), 500–510. https://doi.org/10.1038/s41568-018-0016-5
As imaging data are collected during routine clinical practice, large data sets are — in principle — readily available, thus offering an incredibly rich resource for scientific and medical discovery. Radiographic images, coupled with data on clinical outcomes, have led to the emergence and rapid expansion of radiomics as a field of medical research. Early radiomics studies were largely focused on mining images for a large set of predefined engineered features that describe radiographic aspects of shape, intensity and texture. More recently, radiomics studies have incorporated deep learning techniques to learn feature representations automatically from example images, hinting at the substantial clinical relevance of many of these radiographic features. Within oncology, multiple efforts have successfully explored radiomics tools for assisting clinical decision making related to the diagnosis and risk stratification of different cancers. For example, studies in non- small-cell lung cancer (NSCLC) used radiomics to predict distant metastasis in lung adenocarcinoma and tumor histological subtypes as well as disease recurrence, somatic mutations, gene expression profiles and overall survival. Such findings have motivated an exploration of the clinical utility of AI- generated biomarkers based on standard- of-care radiographic images— with the ultimate hope of better supporting radiologists in disease diagnosis, imaging quality optimization, data visualization, response assessment and report generation. In this Opinion article, the authors start by establishing a general understanding of AI methods particularly pertaining to image- based tasks. They then explore how up- and-coming AI methods will impact multiple radiograph- based practices within oncology. Finally, they discuss the challenges and hurdles facing the clinical implementation of these methods.
As just one example of the challenges faced by this technology, the recent paradigm shift from programs based on well- defined rules to others that learn directly from data has brought certain unforeseen concerns to the spotlight. A strong theoretical understanding of deep learning is yet to be established despite the reported successes across many fields — explaining why deep learning layers that lie between inputs and outputs are labelled as ‘hidden layers’. Identifying specific features of an image that contribute to a predicted outcome is highly hypothetical, causing a lack of understanding of how certain conclusions are drawn by deep learning. This lack of transparency makes it difficult to predict failures, isolate the logic for a specific conclusion or troubleshoot inabilities to generalize to different imaging hardware, scanning protocols and patient populations. Not surprisingly, many uninterpretable AI systems with applications in radiology have been dubbed ‘black- box medicine’.
3 Figures, 2 boxes
8. Broekman, M. L., Maas, S. L. N., Abels, E. R., Mempel, T. R., Krichevsky, A. M., & Breakefield, X. O. (2018). Multidimensional communication in the microenvirons of glioblastoma. Nature Reviews Neurology, 14(August), 1–14. https://doi.org/10.1038/s41582-018-0025-8
Glioblastomas are heterogeneous and invariably lethal tumors. They are characterized by genetic and epigenetic variations among tumor cells, which makes the development of therapies that eradicate all tumor cells challenging and currently impossible. An important component of glioblastoma growth is communication with and manipulation of other cells in the brain environs, which supports tumor progression and resistance to therapy. Glioblastoma cells recruit innate immune cells and change their phenotype to support tumor growth. Tumor cells also suppress adaptive immune responses, and our increasing understanding of how T cells access the brain and how the tumor thwarts the immune response offers new strategies for mobilizing an antitumor response. Tumors also subvert normal brain cells — including endothelial cells, neurons and astrocytes — to create a microenviron that favors tumor success. Overall, after glioblastoma- induced phenotypic modifications, normal cells cooperate with tumor cells to promote tumor proliferation, invasion of the brain, immune suppression and angiogenesis. This glioblastoma takeover of the brain involves multiple modes of communication, including soluble factors such as chemokines and cytokines, direct cell–cell contact, extracellular vesicles (including exosomes and microvesicles) and connecting nanotubes and microtubes. Understanding these multidimensional communications between the tumor and the cells in its environs could open new avenues for therapy.
2 boxes, 4 figures.
