A distinct core regulatory module enforces oncogene expression in KMT2A-rearranged leukemia.

Acute myeloid leukemia with KMT2A (MLL) rearrangements is characterized by specific patterns of gene expression and enhancer architecture, implying unique core transcriptional regulatory circuitry. Here, we identified the transcription factors MEF2D and IRF8 as selective transcriptional dependencies of KMT2A-rearranged AML, where MEF2D displays partially redundant functions with its paralog, MEF2C. Rapid transcription factor degradation followed by measurements of genome-wide transcription rates and superresolution microscopy revealed that MEF2D and IRF8 form a distinct core regulatory module with a narrow direct transcriptional program that includes activation of the key oncogenes MYC, HOXA9, and BCL2. Our study illustrates a mechanism of context-specific transcriptional addiction whereby a specific AML subclass depends on a highly specialized core regulatory module to directly enforce expression of common leukemia oncogenes.

Genetic and transcriptional evolution alters cancer cell line drug response.

Human cancer cell lines are the workhorse of cancer research. Although cell lines are known to evolve in culture, the extent of the resultant genetic and transcriptional heterogeneity and its functional consequences remain understudied. Here we use genomic analyses of 106 human cell lines grown in two laboratories to show extensive clonal diversity. Further comprehensive genomic characterization of 27 strains of the common breast cancer cell line MCF7 uncovered rapid genetic diversification. Similar results were obtained with multiple strains of 13 additional cell lines. Notably, genetic changes were associated with differential activation of gene expression programs and marked differences in cell morphology and proliferation. Barcoding experiments showed that cell line evolution occurs as a result of positive clonal selection that is highly sensitive to culture conditions. Analyses of single-cell-derived clones demonstrated that continuous instability quickly translates into heterogeneity of the cell line. When the 27 MCF7 strains were tested against 321 anti-cancer compounds, we uncovered considerably different drug responses: at least 75% of compounds that strongly inhibited some strains were completely inactive in others. This study documents the extent, origins and consequences of genetic variation within cell lines, and provides a framework for researchers to measure such variation in efforts to support maximally reproducible cancer research.

Code-free machine learning for object detection in surgical video: a benchmarking, feasibility, and cost study.

OBJECTIVE: While the utilization of machine learning (ML) for data analysis typically requires significant technical expertise, novel platforms can deploy ML methods without requiring the user to have any coding experience (termed AutoML). The potential for these methods to be applied to neurosurgical video and surgical data science is unknown. METHODS: AutoML, a code-free ML (CFML) system, was used to identify surgical instruments contained within each frame of endoscopic, endonasal intraoperative video obtained from a previously validated internal carotid injury training exercise performed on a high-fidelity cadaver model. Instrument-detection performances using CFML were compared with two state-of-the-art ML models built using the Python coding language on the same intraoperative video data set. RESULTS: The CFML system successfully ingested surgical video without the use of any code. A total of 31,443 images were used to develop this model; 27,223 images were uploaded for training, 2292 images for validation, and 1928 images for testing. The mean average precision on the test set across all instruments was 0.708. The CFML model outperformed two standard object detection networks, RetinaNet and YOLOv3, which had mean average precisions of 0.669 and 0.527, respectively, in analyzing the same data set. Significant advantages to the CFML system included ease of use, relatively low cost, displays of true/false positives and negatives in a user-friendly interface, and the ability to deploy models for further analysis with ease. Significant drawbacks of the CFML model included an inability to view the structure of the trained model, an inability to update the ML model once trained with new examples, and the inability for robust downstream analysis of model performance and error modes. CONCLUSIONS: This first report describes the baseline performance of CFML in an object detection task using a publicly available surgical video data set as a test bed. Compared with standard, code-based object detection networks, CFML exceeded performance standards. This finding is encouraging for surgeon-scientists seeking to perform object detection tasks to answer clinical questions, perform quality improvement, and develop novel research ideas. The limited interpretability and customization of CFML models remain ongoing challenges. With the further development of code-free platforms, CFML will become increasingly important across biomedical research. Using CFML, surgeons without significant coding experience can perform exploratory ML analyses rapidly and efficiently.

Robotic and robot-assisted skull base neurosurgery: systematic review of current applications and future directions.

OBJECTIVE: The utility of robotic instrumentation is expanding in neurosurgery. Despite this, successful examples of robotic implementation for endoscopic endonasal or skull base neurosurgery remain limited. Therefore, the authors performed a systematic review of the literature to identify all articles that used robotic systems to access the sella or anterior, middle, or posterior cranial fossae. METHODS: A systematic review of MEDLINE and PubMed in accordance with PRISMA guidelines performed for articles published between January 1, 1990, and August 1, 2021, was conducted to identify all robotic systems (autonomous, semiautonomous, or surgeon-controlled) used for skull base neurosurgical procedures. Cadaveric and human clinical studies were included. Studies with exclusively otorhinolaryngological applications or using robotic microscopes were excluded. RESULTS: A total of 561 studies were identified from the initial search, of which 22 were included following full-text review. Transoral robotic surgery (TORS) using the da Vinci Surgical System was the most widely reported system (4 studies) utilized for skull base and pituitary fossa procedures; additionally, it has been reported for resection of sellar masses in 4 patients. Seven cadaveric studies used the da Vinci Surgical System to access the skull base using alternative, non-TORS approaches (e.g., transnasal, transmaxillary, and supraorbital). Five cadaveric studies investigated alternative systems to access the skull base. Six studies investigated the use of robotic endoscope holders. Advantages to robotic applications in skull base neurosurgery included improved lighting and 3D visualization, replication of more traditional gesture-based movements, and the ability for dexterous movements ordinarily constrained by small operative corridors. Limitations included the size and angulation capacity of the robot, lack of drilling components preventing fully robotic procedures, and cost. Robotic endoscope holders may have been particularly advantageous when the use of a surgical assistant or second surgeon was limited. CONCLUSIONS: Robotic skull base neurosurgery has been growing in popularity and feasibility, but significant limitations remain. While robotic systems seem to have allowed for greater maneuverability and 3D visualization, their size and lack of neurosurgery-specific tools have continued to prevent widespread adoption into current practice. The next generation of robotic technologies should prioritize overcoming these limitations.

Author Correction: Mitochondrial metabolism promotes adaptation to proteotoxic stress.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Mitochondrial metabolism promotes adaptation to proteotoxic stress.

The mechanisms by which cells adapt to proteotoxic stress are largely unknown, but are key to understanding how tumor cells, particularly in vivo, are largely resistant to proteasome inhibitors. Analysis of cancer cell lines, mouse xenografts and patient-derived tumor samples all showed an association between mitochondrial metabolism and proteasome inhibitor sensitivity. When cells were forced to use oxidative phosphorylation rather than glycolysis, they became proteasome-inhibitor resistant. This mitochondrial state, however, creates a unique vulnerability: sensitivity to the small molecule compound elesclomol. Genome-wide CRISPR-Cas9 screening showed that a single gene, encoding the mitochondrial reductase FDX1, could rescue elesclomol-induced cell death. Enzymatic function and nuclear-magnetic-resonance-based analyses further showed that FDX1 is the direct target of elesclomol, which promotes a unique form of copper-dependent cell death. These studies explain a fundamental mechanism by which cells adapt to proteotoxic stress and suggest strategies to mitigate proteasome inhibitor resistance.

Neural network modeling for prediction of recurrence, progression, and hormonal non-remission in patients following resection of functional pituitary adenomas.

PURPOSE: Functional pituitary adenomas (FPAs) cause severe neuro-endocrinopathies including Cushing's disease (CD) and acromegaly. While many are effectively cured following FPA resection, some encounter disease recurrence/progression or hormonal non-remission requiring adjuvant treatment. Identification of risk factors for suboptimal postoperative outcomes may guide initiation of adjuvant multimodal therapies. METHODS: Patients undergoing endonasal transsphenoidal resection for CD, acromegaly, and mammosomatotroph adenomas between 1992 and 2019 were identified. Good outcomes were defined as hormonal remission without imaging/biochemical evidence of disease recurrence/progression, while suboptimal outcomes were defined as hormonal non-remission or MRI evidence of recurrence/progression despite adjuvant treatment. Multivariate regression modeling and multilayered neural networks (NN) were implemented. The training sets randomly sampled 60% of all FPA patients, and validation/testing sets were 20% samples each. RESULTS: 348 patients with mean age of 41.7 years were identified. Eighty-one patients (23.3%) reported suboptimal outcomes. Variables predictive of suboptimal outcomes included: Requirement for additional surgery in patients who previously had surgery and continue to have functionally active tumor (p = 0.0069; OR = 1.51, 95%CI 1.12-2.04), Preoperative visual deficit not improved after surgery (p = 0.0033; OR = 1.12, 95%CI 1.04-1.20), Transient diabetes insipidus (p = 0.013; OR = 1.27, 95%CI 1.05-1.52), Higher MIB-1/Ki-67 labeling index (p = 0.038; OR = 1.08, 95%CI 1.01-1.15), and preoperative low cortisol axis (p = 0.040; OR = 2.72, 95%CI 1.06-7.01). The NN had overall accuracy of 87.1%, sensitivity of 89.5%, specificity of 76.9%, positive predictive value of 94.4%, and negative predictive value of 62.5%. NNs for all FPAs were more robust than for CD or acromegaly/mammosomatotroph alone. CONCLUSION: We demonstrate capability of predicting suboptimal postoperative outcomes with high accuracy. NNs may aid in stratifying patients for risk of suboptimal outcomes, thereby guiding implementation of adjuvant treatment in high-risk patients.

A systematic review of virtual reality for the assessment of technical skills in neurosurgery.

OBJECTIVE: Virtual reality (VR) and augmented reality (AR) systems are increasingly available to neurosurgeons. These systems may provide opportunities for technical rehearsal and assessments of surgeon performance. The assessment of neurosurgeon skill in VR and AR environments and the validity of VR and AR feedback has not been systematically reviewed. METHODS: A systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines was conducted through MEDLINE and PubMed. Studies published in English between January 1990 and February 2021 describing the use of VR or AR to quantify surgical technical performance of neurosurgeons without the use of human raters were included. The types and categories of automated performance metrics (APMs) from each of these studies were recorded. RESULTS: Thirty-three VR studies were included in the review; no AR studies met inclusion criteria. VR APMs were categorized as either distance to target, force, kinematics, time, blood loss, or volume of resection. Distance and time were the most well-studied APM domains, although all domains were effective at differentiating surgeon experience levels. Distance was successfully used to track improvements with practice. Examining volume of resection demonstrated that attending surgeons removed less simulated tumor but preserved more normal tissue than trainees. More recently, APMs have been used in machine learning algorithms to predict level of training with a high degree of accuracy. Key limitations to enhanced-reality systems include limited AR usage for automated surgical assessment and lack of external and longitudinal validation of VR systems. CONCLUSIONS: VR has been used to assess surgeon performance across a wide spectrum of domains. The VR environment can be used to quantify surgeon performance, assess surgeon proficiency, and track training progression. AR systems have not yet been used to provide metrics for surgeon performance assessment despite potential for intraoperative integration. VR-based APMs may be especially useful for metrics that are difficult to assess intraoperatively, including blood loss and extent of resection.

The SAFE-T assessment tool: derivation and validation of a web-based application for point-of-care evaluation of gastroenterology fellow performance in colonoscopy.

BACKGROUND AND AIMS: Attending assessment is a critical part of endoscopic education for gastroenterology fellows. The aim of this study was to develop and validate a concise assessment tool to evaluate real-time fellow performance in colonoscopy administered via a web-based application. METHODS: The Skill Assessment in Fellow Endoscopy Training (SAFE-T) tool was derived as a novel 5-question evaluation tool that captures both summative and formative feedback adapted into a web-based application. A prospective study of 15 gastroenterology fellows (5 fellows each from years 1 to 3 of training) was performed using the SAFE-T tool. An independent reviewer evaluated a subset of these procedures and completed the SAFE-T tool and Mayo Colonoscopy Skills Assessment Tool (MCSAT) for reliability testing. RESULTS: Twenty-six faculty completed 350 SAFE-T evaluations of the 15 fellows in the study. The mean SAFE-T overall score (year 1, 2.00; year 2, 3.84; year 3, 4.28) differentiated each sequential fellow year of training (P < .0001). The mean SAFE-T overall score decreased with increasing case complexity score, with straightforward cases compared with average cases (4.07 vs 3.50, P < .0001), and average cases compared with challenging cases (3.50 vs 3.08, P = .0134). In dual-observed procedures, the SAFE-T tool showed excellent inter-rater reliability with a kappa agreement statistic of 0.898 (P < .0001). Correlation of the SAFE-T overall score with the MCSAT overall hands-on and individual motor scores was excellent (each r > 0.90, P < .0001). CONCLUSIONS: We developed and validated the SAFE-T assessment tool, a concise and web-based means of assessing real-time gastroenterology fellow performance in colonoscopy.

Simulated outcomes for durotomy repair in minimally invasive spine surgery.

Minimally invasive spine surgery (MISS) is increasingly performed using endoscopic and microscopic visualization, and the captured video can be used for surgical education and development of predictive artificial intelligence (AI) models. Video datasets depicting adverse event management are also valuable, as predictive models not exposed to adverse events may exhibit poor performance when these occur. Given that no dedicated spine surgery video datasets for AI model development are publicly available, we introduce Simulated Outcomes for Durotomy Repair in Minimally Invasive Spine Surgery (SOSpine). A validated MISS cadaveric dural repair simulator was used to educate neurosurgery residents, and surgical microscope video recordings were paired with outcome data. Objects including durotomy, needle, grasper, needle driver, and nerve hook were then annotated. Altogether, SOSpine contains 15,698 frames with 53,238 annotations and associated durotomy repair outcomes. For validation, an AI model was fine-tuned on SOSpine video and detected surgical instruments with a mean average precision of 0.77. In summary, SOSpine depicts spine surgeons managing a common complication, providing opportunities to develop surgical AI models.

Pilot Analysis of Surgeon Instrument Utilization Signatures Based on Shannon Entropy and Deep Learning for Surgeon Performance Assessment in a Cadaveric Carotid Artery Injury Control Simulation.

BACKGROUND AND OBJECTIVES: Assessment and feedback are critical to surgical education, but direct observational feedback by experts is rarely provided because of time constraints and is typically only qualitative. Automated, video-based, quantitative feedback on surgical performance could address this gap, improving surgical training. The authors aim to demonstrate the ability of Shannon entropy (ShEn), an information theory metric that quantifies series diversity, to predict surgical performance using instrument detections generated through deep learning. METHODS: Annotated images from a publicly available video data set of surgeons managing endoscopic endonasal carotid artery lacerations in a perfused cadaveric simulator were collected. A deep learning model was implemented to detect surgical instruments across video frames. ShEn score for the instrument sequence was calculated from each surgical trial. Logistic regression using ShEn was used to predict hemorrhage control success. RESULTS: ShEn scores and instrument usage patterns differed between successful and unsuccessful trials (ShEn: 0.452 vs 0.370, P < .001). Unsuccessful hemorrhage control trials displayed lower entropy and less varied instrument use patterns. By contrast, successful trials demonstrated higher entropy with more diverse instrument usage and consistent progression in instrument utilization. A logistic regression model using ShEn scores (78% accuracy and 97% average precision) was at least as accurate as surgeons' attending/resident status and years of experience for predicting trial success and had similar accuracy as expert human observers. CONCLUSION: ShEn score offers a summative signal about surgeon performance and predicted success at controlling carotid hemorrhage in a simulated cadaveric setting. Future efforts to generalize ShEn to additional surgical scenarios can further validate this metric.

Expert surgeons and deep learning models can predict the outcome of surgical hemorrhage from 1 min of video.

Major vascular injury resulting in uncontrolled bleeding is a catastrophic and often fatal complication of minimally invasive surgery. At the outset of these events, surgeons do not know how much blood will be lost or whether they will successfully control the hemorrhage (achieve hemostasis). We evaluate the ability of a deep learning neural network (DNN) to predict hemostasis control ability using the first minute of surgical video and compare model performance with human experts viewing the same video. The publicly available SOCAL dataset contains 147 videos of attending and resident surgeons managing hemorrhage in a validated, high-fidelity cadaveric simulator. Videos are labeled with outcome and blood loss (mL). The first minute of 20 videos was shown to four, blinded, fellowship trained skull-base neurosurgery instructors, and to SOCALNet (a DNN trained on SOCAL videos). SOCALNet architecture included a convolutional network (ResNet) identifying spatial features and a recurrent network identifying temporal features (LSTM). Experts independently assessed surgeon skill, predicted outcome and blood loss (mL). Outcome and blood loss predictions were compared with SOCALNet. Expert inter-rater reliability was 0.95. Experts correctly predicted 14/20 trials (Sensitivity: 82%, Specificity: 55%, Positive Predictive Value (PPV): 69%, Negative Predictive Value (NPV): 71%). SOCALNet correctly predicted 17/20 trials (Sensitivity 100%, Specificity 66%, PPV 79%, NPV 100%) and correctly identified all successful attempts. Expert predictions of the highest and lowest skill surgeons and expert predictions reported with maximum confidence were more accurate. Experts systematically underestimated blood loss (mean error - 131 mL, RMSE 350 mL, R2 0.70) and fewer than half of expert predictions identified blood loss > 500 mL (47.5%, 19/40). SOCALNet had superior performance (mean error - 57 mL, RMSE 295 mL, R2 0.74) and detected most episodes of blood loss > 500 mL (80%, 8/10). In validation experiments, SOCALNet evaluation of a critical on-screen surgical maneuver and high/low-skill composite videos were concordant with expert evaluation. Using only the first minute of video, experts and SOCALNet can predict outcome and blood loss during surgical hemorrhage. Experts systematically underestimated blood loss, and SOCALNet had no false negatives. DNNs can provide accurate, meaningful assessments of surgical video. We call for the creation of datasets of surgical adverse events for quality improvement research.

Deep Neural Networks Can Accurately Detect Blood Loss and Hemorrhage Control Task Success From Video.

BACKGROUND: Deep neural networks (DNNs) have not been proven to detect blood loss (BL) or predict surgeon performance from video. OBJECTIVE: To train a DNN using video from cadaveric training exercises of surgeons controlling simulated internal carotid hemorrhage to predict clinically relevant outcomes. METHODS: Video was input as a series of images; deep learning networks were developed, which predicted BL and task success from images alone (automated model) and images plus human-labeled instrument annotations (semiautomated model). These models were compared against 2 reference models, which used average BL across all trials as its prediction (control 1) and a linear regression with time to hemostasis (a metric with known association with BL) as input (control 2). The root-mean-square error (RMSE) and correlation coefficients were used to compare the models; lower RMSE indicates superior performance. RESULTS: One hundred forty-three trials were used (123 for training and 20 for testing). Deep learning models outperformed controls (control 1: RMSE 489 mL, control 2: RMSE 431 mL, R2 = 0.35) at BL prediction. The automated model predicted BL with an RMSE of 358 mL (R2 = 0.4) and correctly classified outcome in 85% of trials. The RMSE and classification performance of the semiautomated model improved to 260 mL and 90%, respectively. CONCLUSION: BL and task outcome classification are important components of an automated assessment of surgical performance. DNNs can predict BL and outcome of hemorrhage control from video alone; their performance is improved with surgical instrument presence data. The generalizability of DNNs trained on hemorrhage control tasks should be investigated.

Validation of Machine Learning-Based Automated Surgical Instrument Annotation Using Publicly Available Intraoperative Video.

BACKGROUND: Intraoperative tool movement data have been demonstrated to be clinically useful in quantifying surgical performance. However, collecting this information from intraoperative video requires laborious hand annotation. The ability to automatically annotate tools in surgical video would advance surgical data science by eliminating a time-intensive step in research. OBJECTIVE: To identify whether machine learning (ML) can automatically identify surgical instruments contained within neurosurgical video. METHODS: A ML model which automatically identifies surgical instruments in frame was developed and trained on multiple publicly available surgical video data sets with instrument location annotations. A total of 39 693 frames from 4 data sets were used (endoscopic endonasal surgery [EEA] [30 015 frames], cataract surgery [4670], laparoscopic cholecystectomy [2532], and microscope-assisted brain/spine tumor removal [2476]). A second model trained only on EEA video was also developed. Intraoperative EEA videos from YouTube were used for test data (3 videos, 1239 frames). RESULTS: The YouTube data set contained 2169 total instruments. Mean average precision (mAP) for instrument detection on the YouTube data set was 0.74. The mAP for each individual video was 0.65, 0.74, and 0.89. The second model trained only on EEA video also had an overall mAP of 0.74 (0.62, 0.84, and 0.88 for individual videos). Development costs were $130 for manual video annotation and under $100 for computation. CONCLUSION: Surgical instruments contained within endoscopic endonasal intraoperative video can be detected using a fully automated ML model. The addition of disparate surgical data sets did not improve model performance, although these data sets may improve generalizability of the model in other use cases.

Transcriptional Plasticity Drives Leukemia Immune Escape.

Relapse of acute myeloid leukemia (AML) after allogeneic bone marrow transplantation has been linked to immune evasion due to reduced expression of major histocompatibility complex class II (MHCII) genes through unknown mechanisms. In this work, we developed CORENODE, a computational algorithm for genome-wide transcription network decomposition that identified a transcription factor (TF) tetrad consisting of IRF8, MYB, MEF2C, and MEIS1, regulating MHCII expression in AML cells. We show that reduced MHCII expression at relapse is transcriptionally driven by combinatorial changes in the expression of these TFs, where MYB and IRF8 play major opposing roles, acting independently of the IFNγ/CIITA pathway. Beyond the MHCII genes, MYB and IRF8 antagonistically regulate a broad genetic program responsible for cytokine signaling and T-cell stimulation that displays reduced expression at relapse. A small number of cells with altered TF abundance and silenced MHCII expression are present at the time of initial leukemia diagnosis, likely contributing to eventual relapse. SIGNIFICANCE: Our findings point to an adaptive transcriptional mechanism of AML evolution after allogeneic transplantation whereby combinatorial fluctuations of TF expression under immune pressure result in the selection of cells with a silenced T-cell stimulation program. This article is highlighted in the In This Issue feature, p. 369.

Utility of the Simulated Outcomes Following Carotid Artery Laceration Video Data Set for Machine Learning Applications.

Importance: Surgical data scientists lack video data sets that depict adverse events, which may affect model generalizability and introduce bias. Hemorrhage may be particularly challenging for computer vision-based models because blood obscures the scene. Objective: To assess the utility of the Simulated Outcomes Following Carotid Artery Laceration (SOCAL)-a publicly available surgical video data set of hemorrhage complication management with instrument annotations and task outcomes-to provide benchmarks for surgical data science techniques, including computer vision instrument detection, instrument use metrics and outcome associations, and validation of a SOCAL-trained neural network using real operative video. Design, Setting, and Participants: For this quailty improvement study, a total of 75 surgeons with 1 to 30 years' experience (mean, 7 years) were filmed from January 1, 2017, to December 31, 2020, managing catastrophic surgical hemorrhage in a high-fidelity cadaveric training exercise at nationwide training courses. Videos were annotated from January 1 to June 30, 2021. Interventions: Surgeons received expert coaching between 2 trials. Main Outcomes and Measures: Hemostasis within 5 minutes (task success, dichotomous), time to hemostasis (in seconds), and blood loss (in milliliters) were recorded. Deep neural networks (DNNs) were trained to detect surgical instruments in view. Model performance was measured using mean average precision (mAP), sensitivity, and positive predictive value. Results: SOCAL contains 31 443 frames with 65 071 surgical instrument annotations from 147 trials with associated surgeon demographic characteristics, time to hemostasis, and recorded blood loss for each trial. Computer vision-based instrument detection methods using DNNs trained on SOCAL achieved a mAP of 0.67 overall and 0.91 for the most common surgical instrument (suction). Hemorrhage control challenges standard object detectors: detection of some surgical instruments remained poor (mAP, 0.25). On real intraoperative video, the model achieved a sensitivity of 0.77 and a positive predictive value of 0.96. Instrument use metrics derived from the SOCAL video were significantly associated with performance (blood loss). Conclusions and Relevance: Hemorrhage control is a high-stakes adverse event that poses unique challenges for video analysis, but no data sets of hemorrhage control exist. The use of SOCAL, the first data set to depict hemorrhage control, allows the benchmarking of data science applications, including object detection, performance metric development, and identification of metrics associated with outcomes. In the future, SOCAL may be used to build and validate surgical data science models.

An In Vivo CRISPR Screening Platform for Prioritizing Therapeutic Targets in AML.

CRISPR-Cas9-based genetic screens have successfully identified cell type-dependent liabilities in cancer, including acute myeloid leukemia (AML), a devastating hematologic malignancy with poor overall survival. Because most of these screens have been performed in vitro using established cell lines, evaluating the physiologic relevance of these targets is critical. We have established a CRISPR screening approach using orthotopic xenograft models to validate and prioritize AML-enriched dependencies in vivo, including in CRISPR-competent AML patient-derived xenograft (PDX) models tractable for genome editing. Our integrated pipeline has revealed several targets with translational value, including SLC5A3 as a metabolic vulnerability for AML addicted to exogenous myo-inositol and MARCH5 as a critical guardian to prevent apoptosis in AML. MARCH5 repression enhanced the efficacy of BCL2 inhibitors such as venetoclax, further highlighting the clinical potential of targeting MARCH5 in AML. Our study provides a valuable strategy for discovery and prioritization of new candidate AML therapeutic targets. SIGNIFICANCE: There is an unmet need to improve the clinical outcome of AML. We developed an integrated in vivo screening approach to prioritize and validate AML dependencies with high translational potential. We identified SLC5A3 as a metabolic vulnerability and MARCH5 as a critical apoptosis regulator in AML, both of which represent novel therapeutic opportunities.This article is highlighted in the In This Issue feature, p. 275.

BAF Complex Maintains Glioma Stem Cells in Pediatric H3K27M Glioma.

Diffuse midline gliomas are uniformly fatal pediatric central nervous system cancers that are refractory to standard-of-care therapeutic modalities. The primary genetic drivers are a set of recurrent amino acid substitutions in genes encoding histone H3 (H3K27M), which are currently undruggable. These H3K27M oncohistones perturb normal chromatin architecture, resulting in an aberrant epigenetic landscape. To interrogate for epigenetic dependencies, we performed a CRISPR screen and show that patient-derived H3K27M-glioma neurospheres are dependent on core components of the mammalian BAF (SWI/SNF) chromatin remodeling complex. The BAF complex maintains glioma stem cells in a cycling, oligodendrocyte precursor cell-like state, in which genetic perturbation of the BAF catalytic subunit SMARCA4 (BRG1), as well as pharmacologic suppression, opposes proliferation, promotes progression of differentiation along the astrocytic lineage, and improves overall survival of patient-derived xenograft models. In summary, we demonstrate that therapeutic inhibition of the BAF complex has translational potential for children with H3K27M gliomas. SIGNIFICANCE: Epigenetic dysregulation is at the core of H3K27M-glioma tumorigenesis. Here, we identify the BRG1-BAF complex as a critical regulator of enhancer and transcription factor landscapes, which maintain H3K27M glioma in their progenitor state, precluding glial differentiation, and establish pharmacologic targeting of the BAF complex as a novel treatment strategy for pediatric H3K27M glioma. See related commentary by Beytagh and Weiss, p. 2730. See related article by Mo et al., p. 2906.

A Guide to Annotation of Neurosurgical Intraoperative Video for Machine Learning Analysis and Computer Vision.

OBJECTIVE: Computer vision (CV) is a subset of artificial intelligence that performs computations on image or video data, permitting the quantitative analysis of visual information. Common CV tasks that may be relevant to surgeons include image classification, object detection and tracking, and extraction of higher order features. Despite the potential applications of CV to intraoperative video, however, few surgeons describe the use of CV. A primary roadblock in implementing CV is the lack of a clear workflow to create an intraoperative video dataset to which CV can be applied. We report general principles for creating usable surgical video datasets and the result of their applications. METHODS: Video annotations from cadaveric endoscopic endonasal skull base simulations (n = 20 trials of 1-5 minutes, size = 8 GB) were reviewed by 2 researcher-annotators. An internal, retrospective analysis of workflow for development of the intraoperative video annotations was performed to identify guiding practices. RESULTS: Approximately 34,000 frames of surgical video were annotated. Key considerations in developing annotation workflows include 1) overcoming software and personnel constraints; 2) ensuring adequate storage and access infrastructure; 3) optimization and standardization of annotation protocol; and 4) operationalizing annotated data. Potential tools for use include CVAT (Computer Vision Annotation Tool) and Vott: open-sourced annotation software allowing for local video storage, easy setup, and the use of interpolation. CONCLUSIONS: CV techniques can be applied to surgical video, but challenges for novice users may limit adoption. We outline principles in annotation workflow that can mitigate initial challenges groups may have when converting raw video into useable, annotated datasets.

Use of surgical video-based automated performance metrics to predict blood loss and success of simulated vascular injury control in neurosurgery: a pilot study.

OBJECTIVE: Experts can assess surgeon skill using surgical video, but a limited number of expert surgeons are available. Automated performance metrics (APMs) are a promising alternative but have not been created from operative videos in neurosurgery to date. The authors aimed to evaluate whether video-based APMs can predict task success and blood loss during endonasal endoscopic surgery in a validated cadaveric simulator of vascular injury of the internal carotid artery. METHODS: Videos of cadaveric simulation trials by 73 neurosurgeons and otorhinolaryngologists were analyzed and manually annotated with bounding boxes to identify the surgical instruments in the frame. APMs in five domains were defined-instrument usage, time-to-phase, instrument disappearance, instrument movement, and instrument interactions-on the basis of expert analysis and task-specific surgical progressions. Bounding-box data of instrument position were then used to generate APMs for each trial. Multivariate linear regression was used to test for the associations between APMs and blood loss and task success (hemorrhage control in less than 5 minutes). The APMs of 93 successful trials were compared with the APMs of 49 unsuccessful trials. RESULTS: In total, 29,151 frames of surgical video were annotated. Successful simulation trials had superior APMs in each domain, including proportionately more time spent with the key instruments in view (p < 0.001) and less time without hemorrhage control (p = 0.002). APMs in all domains improved in subsequent trials after the participants received personalized expert instruction. Attending surgeons had superior instrument usage, time-to-phase, and instrument disappearance metrics compared with resident surgeons (p < 0.01). APMs predicted surgeon performance better than surgeon training level or prior experience. A regression model that included APMs predicted blood loss with an R2 value of 0.87 (p < 0.001). CONCLUSIONS: Video-based APMs were superior predictors of simulation trial success and blood loss than surgeon characteristics such as case volume and attending status. Surgeon educators can use APMs to assess competency, quantify performance, and provide actionable, structured feedback in order to improve patient outcomes. Validation of APMs provides a benchmark for further development of fully automated video assessment pipelines that utilize machine learning and computer vision.