Self-assembly of a layered two-dimensional molecularly woven fabric.

Fabrics-materials consisting of layers of woven fibres-are some of the most important materials in everyday life1. Previous nanoscale weaves2-16 include isotropic crystalline covalent organic frameworks12-14 that feature rigid helical strands interlaced in all three dimensions, rather than the two-dimensional17,18 layers of flexible woven strands that give conventional textiles their characteristic flexibility, thinness, anisotropic strength and porosity. A supramolecular two-dimensional kagome weave15 and a single-layer, surface-supported, interwoven two-dimensional polymer16 have also been reported. The direct, bottom-up assembly of molecular building blocks into linear organic polymer chains woven in two dimensions has been proposed on a number of occasions19-23, but has not previously been achieved. Here we demonstrate that by using an anion and metal ion template, woven molecular 'tiles' can be tessellated into a material consisting of alternating aliphatic and aromatic segmented polymer strands, interwoven within discrete layers. Connections between slowly precipitating pre-woven grids, followed by the removal of the ion template, result in a wholly organic molecular material that forms as stacks and clusters of thin sheets-each sheet up to hundreds of micrometres long and wide but only about four nanometres thick-in which warp and weft single-chain polymer strands remain associated through periodic mechanical entanglements within each sheet. Atomic force microscopy and scanning electron microscopy show clusters and, occasionally, isolated individual sheets that, following demetallation, have slid apart from others with which they were stacked during the tessellation and polymerization process. The layered two-dimensional molecularly woven material has long-range order, is birefringent, is twice as stiff as the constituent linear polymer, and delaminates and tears along well-defined lines in the manner of a macroscopic textile. When incorporated into a polymer-supported membrane, it acts as a net, slowing the passage of large ions while letting smaller ions through.

Efficacy of MEK inhibition in patients with histiocytic neoplasms.

Histiocytic neoplasms are a heterogeneous group of clonal haematopoietic disorders that are marked by diverse mutations in the mitogen-activated protein kinase (MAPK) pathway1,2. For the 50% of patients with histiocytosis who have BRAFV600 mutations3-5, RAF inhibition is highly efficacious and has markedly altered the natural history of the disease6,7. However, no standard therapy exists for the remaining 50% of patients who lack BRAFV600 mutations. Although ERK dependence has been hypothesized to be a consistent feature across histiocytic neoplasms, this remains clinically unproven and many of the kinase mutations that are found in patients who lack BRAFV600 mutations have not previously been biologically characterized. Here we show ERK dependency in histiocytoses through a proof-of-concept clinical trial of cobimetinib, an oral inhibitor of MEK1 and MEK2, in patients with histiocytoses. Patients were enrolled regardless of their tumour genotype. In parallel, MAPK alterations that were identified in treated patients were characterized for their ability to activate ERK. In the 18 patients that we treated, the overall response rate was 89% (90% confidence interval of 73-100). Responses were durable, with no acquired resistance to date. At one year, 100% of responses were ongoing and 94% of patients remained progression-free. Cobimetinib treatment was efficacious regardless of genotype, and responses were observed in patients with ARAF, BRAF, RAF1, NRAS, KRAS, MEK1 (also known as MAP2K1) and MEK2 (also known as MAP2K2) mutations. Consistent with the observed responses, the characterization of the mutations that we identified in these patients confirmed that the MAPK-pathway mutations were activating. Collectively, these data demonstrate that histiocytic neoplasms are characterized by a notable dependence on MAPK signalling-and that they are consequently responsive to MEK inhibition. These results extend the benefits of molecularly targeted therapy to the entire spectrum of patients with histiocytosis.

Tracking tumour evolution in glioma through liquid biopsies of cerebrospinal fluid.

Diffuse gliomas are the most common malignant brain tumours in adults and include glioblastomas and World Health Organization (WHO) grade II and grade III tumours (sometimes referred to as lower-grade gliomas). Genetic tumour profiling is used to classify disease and guide therapy1,2, but involves brain surgery for tissue collection; repeated tumour biopsies may be necessary for accurate genotyping over the course of the disease3-10. While the detection of circulating tumour DNA (ctDNA) in the blood of patients with primary brain tumours remains challenging11,12, sequencing of ctDNA from the cerebrospinal fluid (CSF) may provide an alternative way to genotype gliomas with lower morbidity and cost13,14. We therefore evaluated the representation of the glioma genome in CSF from 85 patients with gliomas who underwent a lumbar puncture because they showed neurological signs or symptoms. Here we show that tumour-derived DNA was detected in CSF from 42 out of 85 patients (49.4%) and was associated with disease burden and adverse outcome. The genomic landscape of glioma in the CSF included a broad spectrum of genetic alterations and closely resembled the genomes of tumour biopsies. Alterations that occur early during tumorigenesis, such as co-deletion of chromosome arms 1p and 19q (1p/19q codeletion) and mutations in the metabolic genes isocitrate dehydrogenase 1 (IDH1) or IDH21,2, were shared in all matched ctDNA-positive CSF-tumour pairs, whereas growth factor receptor signalling pathways showed considerable evolution. The ability to monitor the evolution of the glioma genome through a minimally invasive technique could advance the clinical development and use of genotype-directed therapies for glioma, one of the most aggressive human cancers.

Can the introduction of a 12-lead ECG help reduce mortality in those presenting with foot ulceration to multidisciplinary diabetic foot clinics? An observational evaluation of a real-world implementation pilot in England.

AIMS/HYPOTHESIS: The risk of dying within 2 years of presentation with diabetic foot ulceration is over six times the risk of amputation, with CVD the major contributor. Using an observational evaluation of a real-world implementation pilot, we aimed to assess whether for those presenting with diabetic foot ulceration in England, introducing a 12-lead ECG into routine care followed by appropriate clinical action was associated with reduced mortality. METHODS: Between July 2014 and December 2017, ten multidisciplinary diabetic foot services in England participated in a pilot project introducing 12-lead ECGs for new attendees with foot ulceration. Inception coincided with launch of the National Diabetes Footcare Audit (NDFA), whereby all diabetic footcare services in England were invited to enter data on new attendees with foot ulceration. Poisson regression models assessed the mortality RR at 2 and 5 years following first assessment of those receiving care in a participating pilot unit vs those receiving care in any other unit in England, adjusting for age, sex, ethnicity, deprivation, type and duration of diabetes, ulcer severity, and morbidity in the year prior to first assessment. RESULTS: Of the 3110 people recorded in the NDFA at a participating unit during the pilot, 33% (1015) were recorded as having received an ECG. A further 25,195 people recorded in the NDFA had attended another English footcare service. Unadjusted mortality in the pilot units was 16.3% (165) at 2 years and 37.4% (380) at 5 years for those who received an ECG, and 20.5% (430) and 45.2% (950), respectively, for those who did not receive an ECG. For people included in the NDFA at other units, unadjusted mortality was 20.1% (5075) and 42.6% (10,745), respectively. In the fully adjusted model, mortality was not significantly lower for those attending participating units at 2 (RR 0.93 [95% CI 0.85, 1.01]) or 5 years (RR 0.95 [95% CI 0.90, 1.01]). At participating units, mortality in those who received an ECG vs those who did not was lower at 5 years (RR 0.86 [95% CI 0.76, 0.97]), but not at 2 years (RR 0.87 [95% CI 0.72, 1.04]). Comparing just those that received an ECG with attendees at all other centres in England, mortality was lower at 5 years (RR 0.87 [95% CI 0.78, 0.96]), but not at 2 years (RR 0.86 [95% CI 0.74, 1.01]). CONCLUSIONS/INTERPRETATION: The evaluation confirms the high mortality seen in those presenting with diabetic foot ulceration. Overall mortality at the participating units was not significantly reduced at 2 or 5 years, with confidence intervals just crossing parity. Implementation of the 12-lead ECG into the routine care pathway proved challenging for clinical teams-overall a third of attendees had one, although some units delivered the intervention to over 60% of attendees-and the evaluation was therefore underpowered. Nonetheless, the signals of potential mortality benefit among those who had an ECG suggest that units in a position to operationalise implementation may wish to consider this. DATA AVAILABILITY: Data from the National Diabetes Audit can be requested through the National Health Service Digital Data Access Request Service process at: https://digital.nhs.uk/services/data-access-request-service-dars/dars-products-and-services/data-set-catalogue/national-diabetes-audit-nda.

Directed evolution of biomass intensive CHO cells by adaptation to sub-physiological temperature.

We report a simple and effective means to increase the biosynthetic capacity of host CHO cells. Lonza proprietary CHOK1SV® cells were evolved by serial sub-culture for over 150 generations at 32 °C. During this period the specific proliferation rate of hypothermic cells gradually recovered to become comparable to that of cells routinely maintained at 37 °C. Cold-adapted cell populations exhibited (1) a significantly increased volume and biomass content (exemplified by total RNA and protein), (2) increased mitochondrial function, (3) an increased antioxidant capacity, (4) altered central metabolism, (5) increased transient and stable productivity of a model IgG4 monoclonal antibody and Fc-fusion protein, and (6) unaffected recombinant protein N-glycan processing. This phenotypic transformation was associated with significant genome-scale changes in both karyotype and the relative abundance of thousands of cellular mRNAs across numerous functional groups. Taken together, these observations provide evidence of coordinated cellular adaptations to sub-physiological temperature. These data reveal the extreme genomic/functional plasticity of CHO cells, and that directed evolution is a viable genome-scale cell engineering strategy that can be exploited to create host cells with an increased cellular capacity for recombinant protein production.

CEST MR fingerprinting (CEST-MRF) for brain tumor quantification using EPI readout and deep learning reconstruction.

PURPOSE: To develop a clinical CEST MR fingerprinting (CEST-MRF) method for brain tumor quantification using EPI acquisition and deep learning reconstruction. METHODS: A CEST-MRF pulse sequence originally designed for animal imaging was modified to conform to hardware limits on clinical scanners while keeping scan time under 2 min. Quantitative MRF reconstruction was performed using a deep reconstruction network (DRONE) to yield the water relaxation and chemical exchange parameters. The feasibility of the six parameter DRONE reconstruction was tested in simulations using a digital brain phantom. A healthy subject was scanned with the CEST-MRF sequence, conventional MRF and CEST sequences for comparison. Reproducibility was assessed via test-retest experiments and the concordance correlation coefficient calculated for white matter and gray matter. The clinical utility of CEST-MRF was demonstrated on four patients with brain metastases in comparison to standard clinical imaging sequences. Tumors were segmented into edema, solid core, and necrotic core regions and the CEST-MRF values compared to the contra-lateral side. RESULTS: DRONE reconstruction of the digital phantom yielded a normalized RMS error of ≤7% for all parameters. The CEST-MRF parameters were in good agreement with those from conventional MRF and CEST sequences and previous studies. The mean concordance correlation coefficient for all six parameters was 0.98 ± 0.01 in white matter and 0.98 ± 0.02 in gray matter. The CEST-MRF values in nearly all tumor regions were significantly different (P = 0.05) from each other and the contra-lateral side. CONCLUSION: Combination of EPI readout and deep learning reconstruction enabled fast, accurate and reproducible CEST-MRF in brain tumors.

Fast, light, and scalable: harnessing data-mined line annotations for automated tumor segmentation on brain MRI.

OBJECTIVES: While fully supervised learning can yield high-performing segmentation models, the effort required to manually segment large training sets limits practical utility. We investigate whether data mined line annotations can facilitate brain MRI tumor segmentation model development without requiring manually segmented training data. METHODS: In this retrospective study, a tumor detection model trained using clinical line annotations mined from PACS was leveraged with unsupervised segmentation to generate pseudo-masks of enhancing tumors on T1-weighted post-contrast images (9911 image slices; 3449 adult patients). Baseline segmentation models were trained and employed within a semi-supervised learning (SSL) framework to refine the pseudo-masks. Following each self-refinement cycle, a new model was trained and tested on a held-out set of 319 manually segmented image slices (93 adult patients), with the SSL cycles continuing until Dice score coefficient (DSC) peaked. DSCs were compared using bootstrap resampling. Utilizing the best-performing models, two inference methods were compared: (1) conventional full-image segmentation, and (2) a hybrid method augmenting full-image segmentation with detection plus image patch segmentation. RESULTS: Baseline segmentation models achieved DSC of 0.768 (U-Net), 0.831 (Mask R-CNN), and 0.838 (HRNet), improving with self-refinement to 0.798, 0.871, and 0.873 (each p < 0.001), respectively. Hybrid inference outperformed full image segmentation alone: DSC 0.884 (Mask R-CNN) vs. 0.873 (HRNet), p < 0.001. CONCLUSIONS: Line annotations mined from PACS can be harnessed within an automated pipeline to produce accurate brain MRI tumor segmentation models without manually segmented training data, providing a mechanism to rapidly establish tumor segmentation capabilities across radiology modalities. KEY POINTS: • A brain MRI tumor detection model trained using clinical line measurement annotations mined from PACS was leveraged to automatically generate tumor segmentation pseudo-masks. • An iterative self-refinement process automatically improved pseudo-mask quality, with the best-performing segmentation pipeline achieving a Dice score of 0.884 on a held-out test set. • Tumor line measurement annotations generated in routine clinical radiology practice can be harnessed to develop high-performing segmentation models without manually segmented training data, providing a mechanism to rapidly establish tumor segmentation capabilities across radiology modalities.

Dabrafenib as a diagnostic and therapeutic strategy for the non-surgical management of papillary craniopharyngioma.

PURPOSE: Papillary craniopharyngiomas can cause considerable morbidity due to mass effect and potential surgical complications. These tumors are known to harbor BRAF V600 mutations, which make them exquisitely sensitive to BRAF inhibitors. METHODS: The patient is a 59 year old man with a progressive suprasellar lesion that was radiographically consistent with a papillary craniopharyngioma. He was consented to an Institution Review Board-approved protocol, which permits sequencing of cell free DNA in plasma and the collection and reporting of clinical data. RESULTS: The patient declined surgical resection and was empirically treated with dabrafenib at 150 mg twice daily. Treatment response was demonstrated after 19 days, confirming the diagnosis. After achieving a near complete response after 6.5 months on drug, a decision was made to deescalate treatment to dabrafenib 75 mg twice daily with subsequent tumor stability for 2.5 months. CONCLUSION: Patients with a suspected papillary craniopharyngioma can be challenged with dabrafenib as a potentially effective diagnostic and therapeutic strategy, given that rapid regression with dabrafenib is only observed in tumors harboring a BRAF V600 mutation. Further work is needed to explore the optimal regimen and dose of the targeted therapy.

The treatment of aggressive prolactinomas with everolimus.

INTRODUCTION: Aggressive prolactinomas are life-limiting tumors without a standard of care treatment option after the oral alkylator, temozolomide, fails to provide tumor control. METHODS: We reviewed an institutional database of pituitary tumors for patients with aggressive prolactinomas who progressed following treatment with a dopamine receptor agonist, radiotherapy and temozolomide. Within this cohort, we identified four patients who were treated with everolimus and we report their response to this therapy. Treatment response was determined by a neuroradiologist, who manually performed volumetric assessment and determined treatment response by Response Assessments in Neuro-Oncology (RANO) criteria. RESULTS: Three of four patients who were treated with everolimus had a biochemical response to therapy and all patients derived a clinically meaningful benefit based upon suppression of tumor growth. While the best overall response as assessed by RANO criteria was stable disease for the four patients, a minor regression in tumor size was appreciated in two of the four patients. CONCLUSION: Everolimus is an active agent in the treatment of prolactinomas that warrants further investigation.

Direct Evaluation of Treatment Response in Brain Metastatic Disease with Deep Neuroevolution.

Cancer centers have an urgent and unmet clinical and research need for AI that can guide patient management. A core component of advancing cancer treatment research is assessing response to therapy. Doing so by hand, for example, as per RECIST or RANO criteria, is tedious and time-consuming, and can miss important tumor response information. Most notably, the prevalent response criteria often exclude lesions, the non-target lesions, altogether. We wish to assess change in a holistic fashion that includes all lesions, obtaining simple, informative, and automated assessments of tumor progression or regression. Because genetic sub-types of cancer can be fairly specific and patient enrollment in therapy trials is often limited in number and accrual rate, we wish to make response assessments with small training sets. Deep neuroevolution (DNE) is a novel radiology artificial intelligence (AI) optimization approach that performs well on small training sets. Here, we use a DNE parameter search to optimize a convolutional neural network (CNN) that predicts progression versus regression of metastatic brain disease. We analyzed 50 pairs of MRI contrast-enhanced images as our training set. Half of these pairs, separated in time, qualified as disease progression, while the other 25 image pairs constituted regression. We trained the parameters of a CNN via "mutations" that consisted of random CNN weight adjustments and evaluated mutation "fitness" as summed training set accuracy. We then incorporated the best mutations into the next generation's CNN, repeating this process for approximately 50,000 generations. We applied the CNNs to our training set, as well as a separate testing set with the same class balance of 25 progression and 25 regression cases. DNE achieved monotonic convergence to 100% training set accuracy. DNE also converged monotonically to 100% testing set accuracy. We have thus shown that DNE can accurately classify brain metastatic disease progression versus regression. Future work will extend the input from 2D image slices to full 3D volumes, and include the category of "no change." We believe that an approach such as ours can ultimately provide a useful and informative complement to RANO/RECIST assessment and volumetric AI analysis.

Semisupervised Training of a Brain MRI Tumor Detection Model Using Mined Annotations.

Background Artificial intelligence (AI) applications for cancer imaging conceptually begin with automated tumor detection, which can provide the foundation for downstream AI tasks. However, supervised training requires many image annotations, and performing dedicated post hoc image labeling is burdensome and costly. Purpose To investigate whether clinically generated image annotations can be data mined from the picture archiving and communication system (PACS), automatically curated, and used for semisupervised training of a brain MRI tumor detection model. Materials and Methods In this retrospective study, the cancer center PACS was mined for brain MRI scans acquired between January 2012 and December 2017 and included all annotated axial T1 postcontrast images. Line annotations were converted to boxes, excluding boxes shorter than 1 cm or longer than 7 cm. The resulting boxes were used for supervised training of object detection models using RetinaNet and Mask region-based convolutional neural network (R-CNN) architectures. The best-performing model trained from the mined data set was used to detect unannotated tumors on training images themselves (self-labeling), automatically correcting many of the missing labels. After self-labeling, new models were trained using this expanded data set. Models were scored for precision, recall, and F1 using a held-out test data set comprising 754 manually labeled images from 100 patients (403 intra-axial and 56 extra-axial enhancing tumors). Model F1 scores were compared using bootstrap resampling. Results The PACS query extracted 31 150 line annotations, yielding 11 880 boxes that met inclusion criteria. This mined data set was used to train models, yielding F1 scores of 0.886 for RetinaNet and 0.908 for Mask R-CNN. Self-labeling added 18 562 training boxes, improving model F1 scores to 0.935 (P < .001) and 0.954 (P < .001), respectively. Conclusion The application of semisupervised learning to mined image annotations significantly improved tumor detection performance, achieving an excellent F1 score of 0.954. This development pipeline can be extended for other imaging modalities, repurposing unused data silos to potentially enable automated tumor detection across radiologic modalities. © RSNA, 2022 Online supplemental material is available for this article.

Quantitative assessment of circulating tumor cells in cerebrospinal fluid as a clinical tool to predict survival in leptomeningeal metastases.

PURPOSE: Circulating tumor cells in cerebrospinal fluid are a quantitative diagnostic tool for leptomeningeal metastases from solid tumors, but their prognostic significance is unclear. Our objective was to evaluate CSF-CTC quantification in predicting outcomes in LM. METHODS: This is a single institution retrospective study of patients with solid tumors who underwent CSF-CTC quantification using the CellSearch® platform between 04/2016 and 06/2019. Information on neuroaxis imaging, CSF results, and survival was collected. LM was diagnosed by MRI and/or CSF cytology. Survival analyses were performed using multivariable Cox proportional hazards modeling, and CSF-CTC splits associated with survival were identified through recursive partitioning analysis. RESULTS: Out of 290 patients with CNS metastases, we identified a cohort of 101 patients with newly diagnosed LM. In this group, CSF-CTC count (median 200 CTCs/3 ml) predicted survival continuously (HR = 1.005, 95% CI: 1.002-1.009, p = 0.0027), and the risk of mortality doubled (HR = 2.84, 95% CI: 1.45-5.56, p = 0.0023) at the optimal cutoff of ≥ 61 CSF-CTCs/3 ml. Neuroimaging findings of LM (assessed by 3 independent neuroradiologists) were associated with a higher CSF-CTC count (median CSF-CTCs range 1.5-4 for patients without radiographic LM vs 200 for patients with radiographic LM, p < 0.001), but did not predict survival. CONCLUSION: Our data shows that CSF-CTCs quantification predicts survival in newly diagnosed LM, and outperforms neuroimaging. CSF-CTC analysis can be used as a prognostic tool in patients with LM and provides quantitative assessment of disease burden in the CNS compartment.

Salvage resection plus cesium-131 brachytherapy durably controls post-SRS recurrent brain metastases.

BACKGROUND: Salvage of recurrent previously irradiated brain metastases (rBrM) is a significant challenge. Resection without adjuvant re-irradiation is associated with a high local failure rate, while reirradiation only partially reduces failure but is associated with greater radiation necrosis risk. Salvage resection plus Cs131 brachytherapy may offer dosimetric and biologic advantages including improved local control versus observation, with reduced normal brain dose versus re-irradiation, however data are limited. METHODS: A prospective registry of consecutive patients with post-stereotactic radiosurgery (SRS) rBrM undergoing resection plus implantation of collagen-matrix embedded Cs131 seeds (GammaTile, GT Medical Technologies) prescribed to 60 Gy at 5 mm from the cavity was analyzed. RESULTS: Twenty patients underwent 24 operations with Cs131 implantation in 25 tumor cavities. Median maximum preoperative diameter was 3.0 cm (range 1.1-6.3). Gross- or near-total resection was achieved in 80% of lesions. A median of 16 Cs131 seeds (range 6-30), with a median air-kerma strength of 3.5 U/seed were implanted. There was one postoperative wound dehiscence. With median follow-up of 1.6 years for survivors, two tumors recurred (one in-field, one marginal) resulting in 8.4% 1-year progression incidence (95%CI = 0.0-19.9). Radiographic seed settling was identified in 7/25 cavities (28%) 1.9-11.7 months post-implantation, with 1 case of distant migration (4%), without clinical sequelae. There were 8 cases of radiation necrosis, of which 4 were symptomatic. CONCLUSIONS: With > 1.5 years of follow-up, intraoperative brachytherapy with commercially available Cs131 implants was associated with favorable local control and toxicity profiles. Weak correlation between preoperative tumor geometry and implanted tiles highlights a need to optimize planning criteria.

Temperature and exudativory as drivers of the marmoset (Callithrix spp.) daily activity period.

Primates are affected by fluctuations in ambient temperatures, mostly through thermoregulatory costs and changes in the availability of food. In the present study, we investigate whether the ambient temperature and proxies of food availability affect the activity period of marmosets (Callithrix spp.). We predicted that: (i) at colder sites, marmosets would spend more time at sleeping sites; (ii) midday resting bouts would be longer at hotter sites; (iii) the onset/cessation of activity and resting behavior at midday would be more closely related to temperature than food availability, and (iv) highly exudativorous groups would have higher total levels of resting. We compiled data on the onset and cessation of activity and the time spent resting at midday from seven marmoset studies from sites with a wide range of temperatures. We used generalized linear mixed models to verify the relationship between the dependent variables (lag between dawn and the onset of activities, lag between cessation of activities and dusk, and proportion of resting during midday) and the minimum and maximum temperatures at the respective study sites, together with proxies of food availability (exudativory rates, the amount of habitat available per individual, and net primary productivity) using each sample month as a sampling unit and the identity of the study as a categorical random factor. At colder sites and during colder months, the marmosets left sleeping trees later in the morning and ceased their activities earlier, while at hotter sites and during hotter months, they spent more time resting during midday. More exudativorous groups become active later in the morning, but also ceased their activities later. The abundance of food did not affect the timing of activities. We provide evidence that both low and high temperatures affect marmosets' activities, and that their activity period appears to be more influenced by the thermal environment than food availability.

Fundamental Insights into Graphene Strain Sensing.

Graphene has been studied extensively for use in flexible electronics as ultrasensitive and wide-area strain sensors. Many sensors demonstrated so far rely on graphene networks, such that the spatial resolution is compromised, and they are unable to measure strain variations on a fine scale such as those resulting from substrate/interface failure. In this study, mono-/few-layer graphene are demonstrated to be good candidates for strain sensing with high spatial resolution to evaluate features <100 nm. The fundamentals of strain sensing-interaction with the target-have been discussed to shed light on the sensitivity and durability for future sensor fabrication. The proof-of-concept strain sensors have been shown to be able to monitor different states, e.g., the initiation and evolution, of crazes. The analysis also leads to the evaluation of interfacial energy and realization of high local strain in graphene that is applicable for other 2D materials for ultrasensitive strain sensing and bandgap opening applications.

Deep Learning Achieves Neuroradiologist-Level Performance in Detecting Hydrocephalus Requiring Treatment.

In large clinical centers a small subset of patients present with hydrocephalus that requires surgical treatment. We aimed to develop a screening tool to detect such cases from the head MRI with performance comparable to neuroradiologists. We leveraged 496 clinical MRI exams collected retrospectively at a single clinical site from patients referred for any reason. This diagnostic dataset was enriched to have 259 hydrocephalus cases. A 3D convolutional neural network was trained on 16 manually segmented exams (ten hydrocephalus) and subsequently used to automatically segment the remaining 480 exams and extract volumetric anatomical features. A linear classifier of these features was trained on 240 exams to detect cases of hydrocephalus that required treatment with surgical intervention. Performance was compared to four neuroradiologists on the remaining 240 exams. Performance was also evaluated on a separate screening dataset of 451 exams collected from a routine clinical population to predict the consensus reading from four neuroradiologists using images alone. The pipeline was also tested on an external dataset of 31 exams from a 2nd clinical site. The most discriminant features were the Magnetic Resonance Hydrocephalic Index (MRHI), ventricle volume, and the ratio between ventricle and brain volume. At matching sensitivity, the specificity of the machine and the neuroradiologists did not show significant differences for detection of hydrocephalus on either dataset (proportions test, p > 0.05). ROC performance compared favorably with the state-of-the-art (AUC 0.90-0.96), and replicated in the external validation. Hydrocephalus cases requiring treatment can be detected automatically from MRI in a heterogeneous patient population based on quantitative characterization of brain anatomy with performance comparable to that of neuroradiologists.

High-resolution three-dimensional chromatin profiling of the Chinese hamster ovary cell genome.

Chinese hamster ovary (CHO) cell lines are the pillars of a multibillion-dollar biopharmaceutical industry producing recombinant therapeutic proteins. The effects of local chromatin organization and epigenetic repression within these cell lines result in unpredictable and unstable transgene expression following random integration. Limited knowledge of the CHO genome and its higher order chromatin organization has thus far impeded functional genomics approaches required to tackle these issues. Here, we present an integrative three-dimensional (3D) map of genome organization within the CHOK1SV® 10E9 cell line in conjunction with an improved, less fragmented CHOK1SV 10E9 genome assembly. Using our high-resolution chromatin conformation datasets, we have assigned ≈90% of sequence to a chromosome-scale genome assembly. Our genome-wide 3D map identifies higher order chromatin structures such as topologically associated domains, incorporates our chromatin accessibility data to enhance the identification of active cis-regulatory elements, and importantly links these cis-regulatory elements to target promoters in a 3D promoter interactome. We demonstrate the power of our improved functional annotation by evaluating the 3D landscape of a transgene integration site and two phenotypically different cell lines. Our work opens up further novel genome engineering targets, has the potential to inform vital improvements for industrial biotherapeutic production, and represents a significant advancement for CHO cell line development.

The effect of surgery on radiation necrosis in irradiated brain metastases: extent of resection and long-term clinical and radiographic outcomes.

OBJECTIVE: Radiation therapy is a cornerstone of brain metastasis (BrM) management but carries the risk of radiation necrosis (RN), which can require resection for palliation or diagnosis. We sought to determine the relationship between extent of resection (EOR) of pathologically-confirmed RN and postoperative radiographic and symptomatic outcomes. METHODS: A single-center retrospective review was performed at an NCI-designated Comprehensive Cancer Center to identify all surgically-resected, previously-irradiated necrotic BrM without admixed recurrent malignancy from 2003 to 2018. Clinical, pathologic and radiographic parameters were collected. Volumetric analysis determined EOR and longitudinally evaluated perilesional T2-FLAIR signal preoperatively, postoperatively, and at 3-, 6-, 12-, and 24-months postoperatively when available. Rates of time to 50% T2-FLAIR reduction was calculated using cumulative incidence in the competing risks setting with last follow-up and death as competing events. The Spearman method was used to calculate correlation coefficients, and continuous variables for T2-FLAIR signal change, including EOR, were compared across groups. RESULTS: Forty-six patients were included. Most underwent prior stereotactic radiosurgery with or without whole-brain irradiation (N = 42, 91%). Twenty-seven operations resulted in gross-total resection (59%; GTR). For the full cohort, T2-FLAIR edema decreased by a mean of 78% by 6 months postoperatively that was durable to last follow-up (p < 0.05). EOR correlated with edema reduction at last follow-up, with significantly greater T2-FLAIR reduction with GTR versus subtotal resection (p < 0.05). Among surviving patients, a significant proportion were able to decrease their steroid use: steroid-dependency decreased from 54% preoperatively to 15% at 12 months postoperatively (p = 0.001). CONCLUSIONS: RN resection conferred both durable T2-FLAIR reduction, which correlated with EOR; and reduced steroid dependency.

Salvage resection of recurrent previously irradiated brain metastases: tumor control and radiation necrosis dependency on adjuvant re-irradiation.

PURPOSE: The efficacy of salvage resection (SR) of recurrent brain metastases (rBrM) following stereotactic radiosurgery (SRS) is undefined. We sought to describe local recurrence (LR) and radiation necrosis (RN) rates in patients undergoing SR, with or without adjuvant post-salvage radiation therapy (PSRT). METHODS: A retrospective cohort study evaluated patients undergoing SR of post-SRS rBrM between 3/2003-2/2020 at an NCI-designated cancer center. Cases with histologically-viable malignancy were stratified by receipt of adjuvant PSRT within 60 days of SR. Clinical outcomes were described using cumulative incidences in the clustered competing-risks setting, competing risks regression, and Kaplan-Meier methodology. RESULTS: One-hundred fifty-five rBrM in 135 patients were evaluated. The overall rate of LR was 40.2% (95% CI 34.3-47.2%) at 12 months. Thirty-nine (25.2%) rBrM treated with SR + PSRT trended towards lower 12-month LR versus SR alone [28.8% (95% CI 17.0-48.8%) versus 43.9% (95% CI 36.2-53.4%), p = .07 by multivariate analysis]. SR as re-operation (p = .03) and subtotal resection (p = .01) were independently associated with higher rates of LR. On univariate analysis, tumor size (p = .48), primary malignancy (p = .35), and PSRT technique (p = .43) bore no influence on LR. SR + PSRT was associated with an increased risk of radiographic RN at 12 months versus SR alone [13.4% (95% CI 5.5-32.7%) versus 3.5% (95% CI 1.5-8.0%), p = .02], though the percentage with symptomatic RN remained low (5.1% versus 0.9%, respectively). Median overall survival from SR was 13.4 months (95% CI 10.5-17.7). CONCLUSION: In this largest-known series evaluating SR outcomes in histopathologically-confirmed rBrM, we identify a significant LR risk that may be reduced with adjuvant PSRT and with minimal symptomatic RN. Prospective analysis is warranted.

Revisiting the Language of Glycoscience: Readers, Writers and Erasers in Carbohydrate Biochemistry.

The roles of carbohydrates in nature are many and varied. However, the lack of template encoding in glycoscience distances carbohydrate structure, and hence function, from gene sequence. This challenging situation is compounded by descriptors of carbohydrate structure and function that have tended to emphasise their complexity. Herein, we suggest that revising the language of glycoscience could make interdisciplinary discourse more accessible to all interested parties.