Characterization of Microwave Generator Energy and Ablation Volumes Following Transarterial Embolization in an In Vivo Porcine Liver Model.

PURPOSE: This study aims to characterize the relationship between ablation zone volume (AZV) and microwave ablation (MWA) energy in an in vivo porcine liver model following arterial embolization. MATERIALS AND METHODS: With animal IRB approval, eleven female swine underwent either right (n= 5) or left (n= 6) hepatic artery embolization under fluoroscopic guidance. Subsequently, ultrasound guided MWA was performed in each liver segment (left lateral, left medial, right medial, right lateral) at either 30 Watts (W) (n=4 lobes), 60W (n=4), 65W (n=20), 90W (n=8), 120W (n=4), or 140W (n=4) continuously for 5 minutes. Post-procedural volumetric segmentation was performed on standardized multiphase T1 MRI sequences. RESULTS: AZVs in embolized lobes (15.8 ± 10.6 mL) were significantly larger than non-embolized lobes (11.2 ± 6.5 mL, P <0.01). MWA energy demonstrated significant positive linear correlation with both embolized (R2=0.66, P <0.01) and non-embolized lobes (R2=0.64, P < 0.01). The slope of the linear models corresponded to a 0.95 ± 0.16 and 0.54 ± 0.09 mL/kJ increase in ablation volume per applied kJ of energy (E) in embolized and non-embolized lobes, respectively. In the multivariate model, embolization status significantly modified the relationship between E and AZV as described by the interaction term: 0.42*E*(Embolization Status), (P = 0.031). CONCLUSION: Linear models demonstrate a near 1.8 fold increase in ratio of AZV per unit E, R(AZV:E), when applied to embolized lobes relative to non-embolized lobes. Absolute AZV differences between embolized and non-embolized lobes were greater at higher power MWA.

Cerebellar Mutism Syndrome and Dentato-Thalamo-Cortical Tract Disruption in Diffusion Tractography Following Surgery for Medulloblastoma.

Background Cerebellar mutism syndrome (CMS), a complication following medulloblastoma surgery, has been linked to dentato-thalamo-cortical tract (DTCT) injury; the association of the degree of DTCT injury with severity of CMS-related symptoms has not been investigated. Purpose To investigate the association between severity of CMS-related symptoms and degree and patterns of DTCT injury with use of diffusion tensor imaging (DTI), and if laterality of injury influences neurologic symptoms. Materials and Methods This retrospective case-control study used prospectively collected clinical and DTI data on patients with medulloblastoma enrolled in a clinical trial (between July 2016 and February 2020) and healthy controls (between April and November 2017), matched with the age range of the participants with medulloblastoma. CMS was divided into types 1 (CMS1) and 2 (CMS2). Multivariable logistic regression was used to investigate the relationship between CMS likelihood and DTCT injury. Results Overall, 82 participants with medulloblastoma (mean age, 11.0 years ± 5.2 [SD]; 53 male) and 35 healthy controls (mean age, 18.0 years ± 3.06; 18 female) were included. In participants with medulloblastoma, DTCT was absent bilaterally (AB), absent on the right side (AR), absent on the left side (AL), or present bilaterally (PB), while it was PB in all healthy controls. Odds of having CMS were associated with higher degree of DTCT damage (AB, odds ratio = 272.7 [95% CI: 269.68, 275.75; P < .001]; AR, odds ratio = 14.40 [95% CI: 2.84, 101.48; P < .001]; and AL, odds ratio = 8.55 [95% CI: 1.15, 74.14; P < .001). Left (coefficient = -0.07, χ2 = 12.4, P < .001) and right (coefficient = -0.15, χ2 = 33.82, P < .001) DTCT volumes were negatively associated with the odds of CMS. More participants with medulloblastoma with AB showed CMS1; unilateral DTCT absence prevailed in CMS2. Lower DTCT volumes correlated with more severe ataxia. Unilateral DTCT injury caused ipsilateral dysmetria; AB caused symmetric dysmetria. PB indicated better neurologic outcome. Conclusion The severity of CMS-associated mutism, ataxia, and dysmetria was associated with DTCT damage severity. DTCT damage patterns differed between CMS1 and CMS2. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Dorigatti Soldatelli and Ertl-Wagner in this issue.

Endoportal Radiofrequency Ablation and Stent Placement in Patients with Portal Vein Tumor Thrombosis from Hepatocellular Carcinoma: A Study on Feasibility and Safety.

Background: Hepatocellular carcinoma (HCC) is the most common type of liver cancer, with 10-40% of cases involving portal vein tumor thrombosis (PVTT), leading to poor outcomes and a short survival. The effectiveness of PVTT treatment in patients with HCC is still controversial. Methods: This prospective dual-center study cohort comprised 60 patients with HCC and PVTT who underwent PVR-EPRFA-ST using a novel intravascular radiofrequency system followed by vascular stent placement across the PVTT stenosed segment under fluoroscopy guidance. Results: PVR-EPRFA-ST was technically and clinically successful in 54/60 (90%) and 37/54 (68.5%) patients, respectively. The mean tumor size, PVTT length, post-ablation luminal diameter, and median duration of the recanalized PV patency were 8.6 ± 3.4 cm, 4.1 ± 2.1 cm, 10.3 ± 1.8 mm, and 13.4 months. Higher technical and clinical success rates were associated with a longer survival (177 ± 17.3 days, HR: 0.3, 95%CI 0.12-0.71, p = 0.04; and 233 ± 18.3 days, HR: 0.14, 0.07-0.27, p < 0.001). A shorter survival was associated with Child-Pugh C (HR: 2.7, p = 0.04), multiple tumors (HR: 1.81, p = 0.03), and PVTT length (HR: 1.16, p = 0.04). Conclusions: PVR-EPRFA-ST was feasible and effective for the treatment of selected patients with PVTT, especially in patients with Child-Pugh A/B, single tumors, or a shorter PVTT length.

Current and Future States of Natural Killer Cell-Based Immunotherapy in Hepatocellular Carcinoma.

Natural killer (NK) cells are innate lymphoid cells that exhibit high levels of cytotoxicity against NK-specific targets. NK cells also produce various cytokines, and interact with T cells, B cells, and dendritic cells to effectively serve as frontliners of the innate immune system. Produce various cytokines, and interact with T cells, B cells, and dendritic cells to effectively serve as frontliners of the innate immune system. Moreover, NK cells constitute the second most common immune cell in the liver. These properties have drawn significant attention towards leveraging NK cells in treating liver cancer, especially hepatocellular carcinoma (HCC), which accounts for 75% of all primary liver cancer and is the fourth leading cause of cancer-related death worldwide. Notable anti-cancer functions of NK cells against HCC include activating antibody-dependent cell cytotoxicity (ADCC), facilitating Gasdermin E-mediated pyroptosis of HCC cells, and initiating an antitumor response via the cGAS-STING signaling pathway. In this review, we describe how these mechanisms work in the context of HCC. We will then discuss the existing preclinical and clinical studies that leverage NK cell activity to create single and combined immunotherapies.

[11C]-methionine positron emission tomography in the evaluation of pediatric low-grade gliomas.

Background: [11C]-Methionine positron emission tomography (PET; [11C]-MET-PET) is principally used for the evaluation of brain tumors in adults. Although amino acid PET tracers are more commonly used in the evaluation of pediatric brain tumors, data on [11C]-MET-PET imaging of pediatric low-grade gliomas (pLGG) is scarce. This study aimed to investigate the roles of [11C]-MET-PET in the evaluation of pLGGs. Methods: Eighteen patients with newly diagnosed pLGG and 26 previously treated pLGG patients underwent [11C]-MET-PET met the inclusion and exclusion criteria. Tumor-to-brain uptake ratio (TBR) and metabolic tumor volumes were assessed for diagnostic performances (newly diagnosed, 15; previously treated 26), change with therapy (newly diagnosed, 9; previously treated 7), and variability among different histology (n = 12) and molecular markers (n = 7) of pLGGs. Results: The sensitivity of [11C]-MET-PET for diagnosing pLGG, newly diagnosed, and previously treated combined was 93% for both TBRmax and TBRpeak, 76% for TBRmean, and 95% for qualitative evaluation. TBRmax showed a statistically significant reduction after treatment, while other PET parameters showed a tendency to decrease. Median TBRmax, TBRpeak, and TBRmean values were slightly higher in the BRAFV600E mutated tumors compared to the BRAF fused tumors. Median TBRmax, and TBRpeak in diffuse astrocytomas were higher compared to pilocytic astrocytomas, but median TBRmean, was slightly higher in pilocytic astrocytomas. However, formal statistical analysis was not done due to the small sample size. Conclusions: Our study shows that [11C]-MET-PET reliably characterizes new and previously treated pLGGs. Our study also shows that quantitative parameters tend to decrease with treatment, and differences may exist between various pLGG types.

Comprehensive analysis of MYB/MYBL1-altered pediatric-type diffuse low-grade glioma.

BACKGROUND: Pediatric-type diffuse low-grade gliomas (pLGG) harboring recurrent genetic alterations involving MYB or MYBL1 are closely related tumors. Detailed treatment and outcome data of large cohorts are still limited. This study aimed to comprehensively evaluate pLGG with these alterations to define optimal therapeutic strategies. METHODS: We retrospectively reviewed details of pLGG with MYB or MYBL1 alterations from patients treated or referred for pathologic review at St. Jude Children's Research Hospital. Tumor specimens were centrally reviewed, and clinical data were collated. RESULTS: Thirty-three patients (18 male; median age, 5 y) were identified. Two tumors had MYBL1 alterations; 31 had MYB alterations, MYB::QKI fusion being the most common (n=10, 30%). Most tumors were in the cerebral hemispheres (n=22, 67%). Two patients (6%) had metastasis at diagnosis. The median follow-up was 6.1 years. The 5-year event-free survival (EFS) rate was 81.3±8.3%; the 5-year overall survival (OS) rate was 96.4±4.1%. Patients receiving a near-total or gross-total resection had a 5-year EFS of 100%; those receiving a biopsy or subtotal resection had a 5-year EFS rate of 56.6±15.2% (p<0.01). No difference in EFS was observed based on location, histology, or molecular alterations. However, the tumors that progressed or metastasized may have distinct methylation profiles with evidence of activation of the MAPK and PI3K/AKT/mTOR pathways. CONCLUSIONS: pLGG with MYB/MYBL1 alterations have good outcomes. Our findings suggest that surgical resectability is a crucial determinant of EFS. Further characterization is required to identify optimal treatment strategies for progressive tumors.

Pigs as Clinically Relevant Models for Synergizing Interventional Oncology and Immunotherapy.

Traditionally, rodent cancer models have driven preclinical oncology research. However, they do not fully recapitulate characteristics of human cancers, and their size poses challenges when evaluating tools in the interventional oncologists' armamentarium. Pig models, however, have been the gold standard for validating surgical procedures. Their size enables the study of image-guided interventions using human ultrasound (US), computed tomography (CT), and magnetic resonance (MR) imaging platforms. Furthermore, pigs have immunologic features that are similar to those of humans, which can potentially be leveraged for studying immunotherapy. Novel pig models of cancer are being developed, but additional research is required to better understand both the pig immune system and malignancy to enhance the potential for pig models in interventional oncology research. This review aims to address the main advantages and disadvantages of using a pig model for interventional oncology and outline the specific characteristics of pig models that make them more suitable for investigation of locoregional therapies.

High-grade glioma in infants and young children is histologically, molecularly, and clinically diverse: Results from the SJYC07 trial and institutional experience.

BACKGROUND: High-grade gliomas (HGG) in young children pose a challenge due to favorable but unpredictable outcomes. While retrospective studies broadened our understanding of tumor biology, prospective data is lacking. METHODS: A cohort of children with histologically diagnosed HGG from the SJYC07 trial was augmented with nonprotocol patients with HGG treated at St. Jude Children's Research Hospital from November 2007 to December 2020. DNA methylome profiling and whole genome, whole exome, and RNA sequencing were performed. These data were integrated with histopathology to yield an integrated diagnosis. Clinical characteristics and preoperative imaging were analyzed. RESULTS: Fifty-six children (0.0-4.4 years) were identified. Integrated analysis split the cohort into four categories: infant-type hemispheric glioma (IHG), HGG, low-grade glioma (LGG), and other-central nervous system (CNS) tumors. IHG was the most prevalent (n = 22), occurred in the youngest patients (median age = 0.4 years), and commonly harbored receptor tyrosine kinase gene fusions (7 ALK, 2 ROS1, 3 NTRK1/2/3, 4 MET). The 5-year event-free (EFS) and overall survival (OS) for IHG was 53.13% (95%CI: 35.52-79.47) and 90.91% (95%CI: 79.66-100.00) vs. 0.0% and 16.67% (95%CI: 2.78-99.74%) for HGG (p = 0.0043, p = 0.00013). EFS and OS were not different between IHG and LGG (p = 0.95, p = 0.43). Imaging review showed IHGs are associated with circumscribed margins (p = 0.0047), hemispheric location (p = 0.0010), and intratumoral hemorrhage (p = 0.0149). CONCLUSIONS: HGG in young children is heterogeneous and best defined by integrating histopathological and molecular features. Patients with IHG have relatively good outcomes, yet they endure significant deficits, making them good candidates for therapy de-escalation and trials of molecular targeted therapy.

Constitutional balanced translocations involving SMARCB1: A rare cause of rhabdoid tumor predisposition syndrome.

Rhabdoid Tumor Predisposition Syndrome 1 (RTPS1) confers an increased risk of developing rhabdoid tumors and is caused by germline mutations in SMARCB1. RTPS1 should be evaluated in all individuals with rhabdoid tumor and is more likely in those with a young age at presentation (occasionally congenital presentation), multiple primary tumors, or a family history of rhabdoid tumor or RTPS1. Proband genetic testing is the standard method for diagnosing RTPS1. Most known RTPS1-related SMARCB1 gene mutations are copy number variants (CNVs) or single nucleotide variants/indels, but structural variant analysis (SVA) is not usually included in the molecular evaluation. Here, we report two children with RTPS1 presenting with atypical teratoid/rhabdoid tumor (ATRT) who had constitutional testing showing balanced chromosome translocations involving SMARCB1. Patient 1 is a 23-year-old female diagnosed with pineal region ATRT at 7 months who was found to have a de novo, constitutional t(16;22)(p13.3;q11.2). Patient 2 is a 24-month-old male diagnosed with a posterior fossa ATRT at 14 months, with subsequent testing showing a constitutional t(5;22)(q14.1;q11.23). These structural rearrangements have not been previously reported in RTPS1. While rare, these cases suggest that structural variants should be considered in the evaluation of children with rhabdoid tumors to provide more accurate genetic counseling on the risks of developing tumors, the need for surveillance, and the risks of passing the disorder on to future children. Further research is needed to understand the prevalence, clinical features, and tumor risks associated with RTPS1-related constitutional balanced translocations.

Effects of a comprehensive brain computed tomography deep learning model on radiologist detection accuracy.

OBJECTIVES: Non-contrast computed tomography of the brain (NCCTB) is commonly used to detect intracranial pathology but is subject to interpretation errors. Machine learning can augment clinical decision-making and improve NCCTB scan interpretation. This retrospective detection accuracy study assessed the performance of radiologists assisted by a deep learning model and compared the standalone performance of the model with that of unassisted radiologists. METHODS: A deep learning model was trained on 212,484 NCCTB scans drawn from a private radiology group in Australia. Scans from inpatient, outpatient, and emergency settings were included. Scan inclusion criteria were age ≥ 18 years and series slice thickness ≤ 1.5 mm. Thirty-two radiologists reviewed 2848 scans with and without the assistance of the deep learning system and rated their confidence in the presence of each finding using a 7-point scale. Differences in AUC and Matthews correlation coefficient (MCC) were calculated using a ground-truth gold standard. RESULTS: The model demonstrated an average area under the receiver operating characteristic curve (AUC) of 0.93 across 144 NCCTB findings and significantly improved radiologist interpretation performance. Assisted and unassisted radiologists demonstrated an average AUC of 0.79 and 0.73 across 22 grouped parent findings and 0.72 and 0.68 across 189 child findings, respectively. When assisted by the model, radiologist AUC was significantly improved for 91 findings (158 findings were non-inferior), and reading time was significantly reduced. CONCLUSIONS: The assistance of a comprehensive deep learning model significantly improved radiologist detection accuracy across a wide range of clinical findings and demonstrated the potential to improve NCCTB interpretation. CLINICAL RELEVANCE STATEMENT: This study evaluated a comprehensive CT brain deep learning model, which performed strongly, improved the performance of radiologists, and reduced interpretation time. The model may reduce errors, improve efficiency, facilitate triage, and better enable the delivery of timely patient care. KEY POINTS: • This study demonstrated that the use of a comprehensive deep learning system assisted radiologists in the detection of a wide range of abnormalities on non-contrast brain computed tomography scans. • The deep learning model demonstrated an average area under the receiver operating characteristic curve of 0.93 across 144 findings and significantly improved radiologist interpretation performance. • The assistance of the comprehensive deep learning model significantly reduced the time required for radiologists to interpret computed tomography scans of the brain.

CSF cytology of common primary CNS neoplasms categorized by CNS WHO 2021.

OBJECTIVE: The detection of neoplastic cells in cerebral spinal fluid (CSF) is pivotal for the management of patients with central nervous system (CNS) tumours. This article delves into the CSF cytological characteristics of common CNS neoplasms, aligning with the 2021 World Health Organization (WHO) classification of CNS tumours. METHODS: A retrospective review of CSF specimens positive for primary CNS neoplasms was performed at three tertiary medical centres. Only cases that had histopathologic confirmation and/or molecular workup were included. RESULTS: Common primary CNS neoplasms seen in CSF cytology specimens include medulloblastoma, (non-WNT/non-SHH as well as SHH-activated and TP53 mutant), pineoblastoma, atypical teratoid/rhabdoid tumour (AT/RT), IDH-wildtype glioblastoma, and primary diffuse large B-cell lymphoma of the CNS. Ependymomas and germinomas can also have CSF involvement but are less common. Although the typical histologic architecture of these tumours may not be preserved in the CSF, unique cytomorphologic features such as nuclear moulding, nuclear pleomorphism, rhabdoid cells, prominent nucleoli and rosette formation can still be appreciated. CONCLUSION: Adopting the updated terminology and correlating cytologic observations with molecular findings will streamline the diagnostic process, reducing the complexities and ambiguities pathologists often encounter when analysing CSF specimens for potential primary CNS neoplasms.

GRP78-CAR T cell effector function against solid and brain tumors is controlled by GRP78 expression on T cells.

Lack of targetable antigens is a key limitation for developing successful T cell-based immunotherapies. Members of the unfolded protein response (UPR) represent ideal immunotherapy targets because the UPR regulates the ability of cancer cells to resist cell death, sustain proliferation, and metastasize. Glucose-regulated protein 78 (GRP78) is a key UPR regulator that is overexpressed and translocated to the cell surface of a wide variety of cancers in response to elevated endoplasmic reticulum (ER) stress. We show that GRP78 is highly expressed on the cell surface of multiple solid and brain tumors, making cell surface GRP78 a promising chimeric antigen receptor (CAR) T cell target. We demonstrate that GRP78-CAR T cells can recognize and kill GRP78+ brain and solid tumors in vitro and in vivo. Additionally, our findings demonstrate that GRP78 is upregulated on CAR T cells upon T cell activation; however, this expression is tumor-cell-line specific and results in heterogeneous GRP78-CAR T cell therapeutic response.

Characterizing hepatocellular carcinoma stem markers and their corresponding susceptibility to NK-cell based immunotherapy.

Introduction: Hepatocellular carcinoma (HCC) is the most common primary liver cancer and is the fourth-leading cause of all cancer-related deaths around the world. Liver transplantation, surgery, and local ablation are curative therapies for early-stage HCC. However, post-treatment outcomes can vary based on histopathologic stage. Poorly-differentiated HCC are associated with higher rates of tumor progression and lower overall survival compared to well-differentiated HCC after therapy. In this study, we aimed to characterize the cancer stem cell (CSC) profile of histopathologically-proven well and poorly-differentiated HCCs in an in-vitro environment. We characterized the stem-like profile of each type of HCC based on their surface markers and susceptibility to NK cell-mediated cytotoxicity. Methods: Flow cytometry was used to quantify differential expression of MHC-class I, CD54, and CD44 between well- and poorly-differentiated HCCs. Primary untreated NK cells, IL-2 stimulated primary NK cells, and supercharged (sNK) cell-mediated cytotoxicity was assessed against well- and poorly-differentiated HCCs. IFN-γ supernatant from each respective NK cell experimental arm was also used to induce differentiation of HCCs. Finally, we characterized the temporal NK effector cell cytotoxicity using real-time quantitative analysis of imaging and impedance (eSight study). Results: Poorly-differentiated HCCs demonstrated low surface expression of MHC-class I and CD54, and high expression of CD44. Treatment of NK cells secreted IFN-γ or IFN-γ cytokine induced differentiation in HCCs. Poorly-differentiated HCCs in comparison to well-differentiated HCC were more susceptible to NK cell-mediated cytotoxicity in primary NK cells, IL-2 stimulated primary NK cells, and sNK cells. sNK cells induced significantly higher cytotoxicity against well-differentiated HCCs in comparison to untreated or IL-2-stimulated primary NK cells. These findings were recapitulated with real-time quantitative imaging analysis. Conclusions: Poorly-differentiated HCCs were found to have surface marker patterns of CSCs, making them highly susceptible to NK cell-based immunotherapy. NK-cell based therapy can potentially be leveraged as a neoadjuvant or adjuvant therapy in poorly-differentiated HCCs. Supercharged NK cells, which can be rapidly expanded to therapeutic levels, are uniquely capable of lysing both poorly- and well-differentiated HCCs. This finding suggests that sNK cells not only exhibit enhanced features against NK cells' targets but also are capable of activating T cells to induce cytotoxicity against well-differentiated HCCs with high expression of MHC class I.

Combination transarterial chemoembolization and microwave ablation vs. microwave ablation monotherapy for hepatocellular carcinomas greater than 3 cm: a comparative study.

PURPOSE: To evaluate the efficacy of combination therapy using transarterial chemoembolization with microwave ablation (MWA) therapy vs. MWA monotherapy for hepatocellular carcinomas (HCCs) >3 cm in size. METHODS: This two-arm retrospective observational study included patients with HCCs >3 cm who underwent either combination therapy (29 patients) or MWA monotherapy (35 patients) between 2014 and 2020. The treatment outcomes related to primary treatment efficacy, local tumor progression (LTP), tumor control rate, and overall survival were compared between each cohort. RESULTS: The technical success and primary efficacy were 96.56% and 100.00% in the combination therapy cohort, and 91.42% and 100.00% in the MWA cohort, respectively, over a mean follow-up period of 27.6 months. The 1- and 3-year rates of LTP-free survival were 78.57% and 69.56% in the combination therapy cohort, vs. 72.45% and 35.44% in the MWA cohort, respectively (P = 0.001). The overall progression-free survival was longer in the combination therapy cohort compared with the MWA cohort (median: 56.0 vs. 13.0 months; P = 0.017). With the incorporation of additional locoregional therapy, the overall survival rates were not significantly different, with 1- and 3-year overall survival rates of 100.00% and 88.71% in the combination therapy cohort and rates of 90.15% and 82.76% in the MWA cohort, respectively (P = 0.235). CONCLUSION: The combination therapy provided significantly longer upfront LTP-free survival in HCCs >3 cm when compared with the MWA treatment alone, albeit with similar local tumor control and overall survival rates when accounting for additional locoregional therapies.

Pediatric diencephalic tumors: a constellation of entities and management modalities.

The diencephalon is a complex midline structure consisting of the hypothalamus, neurohypophysis, subthalamus, thalamus, epithalamus, and pineal body. Tumors arising from each of these diencephalic components differ significantly in terms of biology and prognosis. The aim of this comprehensive review is to describe the epidemiology, clinical symptoms, imaging, histology, and molecular markers in the context of the 2021 WHO classification of central nervous system neoplasms. We will also discuss the current management of each of these tumors.

Central Nervous System Tumor With BCL6 Corepressor Internal Tandem Duplication: Treatment Course of a Long-term Survivor.

Central nervous system (CNS) tumor with BCL6 corepressor (BCOR) internal tandem duplication (ITD) is a newly described CNS tumor, characterized by in-frame ITDs of the BCOR gene. There is no standard practice regarding the management of this tumor. We report the clinical course of a 6-year-old boy who presented to the hospital with worsening headaches. Computed tomography scan showed a large right-sided parietal supratentorial mass and brain magnetic resonance imaging confirmed a 6×8×6.7 cm lobulated, solid but heterogeneous mass in the right parieto-occipital region. While initial pathology suggested a WHO grade 3 anaplastic meningioma, additional investigation with molecular analysis confirmed the diagnosis of high-grade neuroepithelial tumor with BCOR exon 15 ITD. This diagnosis was renamed CNS tumor with BCOR ITD in the 2021 WHO CNS tumor classification. The patient received 54 Gy of focal radiation and has no evidence of disease recurrence after 48 months from the end of treatment. As this is a newly discovered entity with only a few previous reports in the scientific literature, this report presents a unique treatment for this CNS tumor compared with those previously described.

Machine-learning versus traditional approaches for atherosclerotic cardiovascular risk prognostication in primary prevention cohorts: a systematic review and meta-analysis.

BACKGROUND: Cardiovascular disease (CVD) risk prediction is important for guiding the intensity of therapy in CVD prevention. Whilst current risk prediction algorithms use traditional statistical approaches, machine learning (ML) presents an alternative method that may improve risk prediction accuracy. This systematic review and meta-analysis aimed to investigate whether ML algorithms demonstrate greater performance compared with traditional risk scores in CVD risk prognostication. METHODS AND RESULTS: MEDLINE, EMBASE, CENTRAL, and SCOPUS Web of Science Core collections were searched for studies comparing ML models to traditional risk scores for CVD risk prediction between the years 2000 and 2021. We included studies that assessed both ML and traditional risk scores in adult (≥18 year old) primary prevention populations. We assessed the risk of bias using the Prediction Model Risk of Bias Assessment Tool (PROBAST) tool. Only studies that provided a measure of discrimination [i.e. C-statistics with 95% confidence intervals (CIs)] were included in the meta-analysis. A total of 16 studies were included in the review and meta-analysis (3302 515 individuals). All study designs were retrospective cohort studies. Out of 16 studies, 3 externally validated their models, and 11 reported calibration metrics. A total of 11 studies demonstrated a high risk of bias. The summary C-statistics (95% CI) of the top-performing ML models and traditional risk scores were 0.773 (95% CI: 0.740-0.806) and 0.759 (95% CI: 0.726-0.792), respectively. The difference in C-statistic was 0.0139 (95% CI: 0.0139-0.140), P < 0.0001. CONCLUSION: ML models outperformed traditional risk scores in the discrimination of CVD risk prognostication. Integration of ML algorithms into electronic healthcare systems in primary care could improve identification of patients at high risk of subsequent CVD events and hence increase opportunities for CVD prevention. It is uncertain whether they can be implemented in clinical settings. Future implementation research is needed to examine how ML models may be utilized for primary prevention.This review was registered with PROSPERO (CRD42020220811).

Machine Learning Augmented Interpretation of Chest X-rays: A Systematic Review.

Limitations of the chest X-ray (CXR) have resulted in attempts to create machine learning systems to assist clinicians and improve interpretation accuracy. An understanding of the capabilities and limitations of modern machine learning systems is necessary for clinicians as these tools begin to permeate practice. This systematic review aimed to provide an overview of machine learning applications designed to facilitate CXR interpretation. A systematic search strategy was executed to identify research into machine learning algorithms capable of detecting >2 radiographic findings on CXRs published between January 2020 and September 2022. Model details and study characteristics, including risk of bias and quality, were summarized. Initially, 2248 articles were retrieved, with 46 included in the final review. Published models demonstrated strong standalone performance and were typically as accurate, or more accurate, than radiologists or non-radiologist clinicians. Multiple studies demonstrated an improvement in the clinical finding classification performance of clinicians when models acted as a diagnostic assistance device. Device performance was compared with that of clinicians in 30% of studies, while effects on clinical perception and diagnosis were evaluated in 19%. Only one study was prospectively run. On average, 128,662 images were used to train and validate models. Most classified less than eight clinical findings, while the three most comprehensive models classified 54, 72, and 124 findings. This review suggests that machine learning devices designed to facilitate CXR interpretation perform strongly, improve the detection performance of clinicians, and improve the efficiency of radiology workflow. Several limitations were identified, and clinician involvement and expertise will be key to driving the safe implementation of quality CXR machine learning systems.