Citations

Introduction: Left and right ventricular systolic dysfunction (LVSD and RVSD respectively) contribute to considerable pediatric morbidity. Both LVSD and RVSD can occur in children before symptom onset, emphasizing the need for early detection. Existing deep-learning algorithms can identify LVSD in adults using 12-lead ECG; however, their efficacy in children is uncertain. Aims: We aimed to create a deep-learning model for LVSD and RVSD detection in pediatric patients using an AI-enabled ECG and compare novel pediatric models to a previously validated adult LVSD model. Methods: We identified 10,214 pediatric patients at the Mayo Clinic with a transthoracic echocardiogram (TTE) and a 10-second, 12-lead surface ECG within 14 days of TTE, performed from 2002 -2022. Our cohort comprised the first TTE-ECG pair from each patient, totaling 138 with EF ≤ 35%, 312 with EF ≤ 50%, 482 with RVSD, and 9,567 without any LVSD or RVSD. The cohort was split into training, validation, and testing datasets to develop neural networks for systolic dysfunction detection. Results: The cohort was 52.3% male, with an average age of 10.3 years. Novel models generated in the pediatric cohort demonstrated excellent performance in systolic dysfunction detection (Figure 1). Pediatric models for EF ≤ 35% and EF ≤ 50% achieved test AUCs of 0.92 (95% CI 0.89, 0.95) and 0.89 (95% CI 0.85, 0.94) respectively. The RVSD model reached a test AUC of 0.91 (95% CI 0.88, 0.94). A combined model for EF ≤ 35% or RVSD exhibited a test AUC of 0.90 (95% CI 0.87, 0.93), sensitivity 0.88, specificity 0.82, PPV 0.20, and NPV 0.95. A combined model for EF ≤ 50% or RVSD exhibited a test AUC of 0.88 (95% CI 0.85, 0.91). The previously validated adult data-derived model achieved an AUC of 0.86 (95% CI 0.83, 0.89) for EF ≤ 50% in children. Conclusions: AI-enabled ECGs demonstrate efficient detection of both LVSD and RVSD in pediatric patients. Future multi-site validation is needed to further establish the reliability of these models.

Introduction: Biologic sex and hormonal concentrations shape ECG parameters due to sex hormone effects on cardiac function. AI-enabled ECGs have previously been shown to accurately predict patient sex in adults and correlate with sex hormone levels. Aims: We aimed to test the ability of AI-enabled ECGs to predict pediatric patient sex and explore the influence of pubertal development. Methods: Two AI-enabled ECG models were created using a convolutional neural network trained on pediatric 10-second, 12-lead surface ECGs. The first model was trained de novo using pediatric data. The second model used transfer learning from a previously validated adult data-derived algorithm. We analyzed the first ECG from 89,063 pediatric patients (aged <19 years) recorded from 1988-2022, and divided the cohort into training, validation, and testing datasets. Subgroup analysis was performed on prepubertal (0-7 years), pubertal (8-14 years), and post-pubertal (15-18 years) patients. Results: The cohort consisted of 46.6% males, with 21,434 prepubertal, 26,300 pubertal, and 41,329 post-pubertal children. No significant performance difference was observed between the two models. Both models demonstrated 81% accuracy and an overall AUC of 0.91 in the testing cohorts, with high discriminatory ability in post-pubertal teenagers (AUC = 0.98). (Figure 1) Conclusions: AI-ECG predicts pediatric patient sex with varying performance across pubertal stages. The highest discriminatory ability is observed post-puberty, decreasing in pubertal and prepubertal children. The results suggest hormonal and physiological changes during puberty may influence cardiac electrophysiology measurably. More research is required to interpret CNN-predicted sex discordance with actual sex and its implications for future cardiovascular risk.

AbstractObjectivesCorpus callosotomy (CC) is used to reduce seizures, primarily in patients with generalized drug‐resistant epilepsy (DRE). The invasive nature of the procedure contributes to underutilization despite its potential superiority to other palliative procedures. The goal of this study was to use a multi‐institutional epilepsy surgery database to characterize the use of CC across participating centers.MethodsData were acquired from the Pediatric Epilepsy Research Consortium (PERC) Surgery Database, a prospective observational study collecting data on children 0–18 years referred for surgical evaluation of DRE across 22 U.S. pediatric epilepsy centers. Patient, epilepsy, and surgical characteristics were collected across multiple CC modalities. Outcomes and complications were recorded and analyzed statistically.ResultsEighty‐three patients undergoing 85 CC procedures at 14 participating epilepsy centers met inclusion criteria. Mean age at seizure onset was 2.3 years (0–9.4); mean age for Phase I evaluation and surgical intervention were 9.45 (.1–20) and 10.46 (.2–20.6) years, respectively. Generalized seizure types were the most common (59%). Complete CC was performed in 88%. The majority of CC procedures (57%) were via open craniotomy, followed by laser interstitial thermal therapy (LiTT) (20%) and mini‐craniotomy/endoscopic (mc/e) (22%). Mean operative times were significantly longer for LiTT, whereas mean estimated blood loss was greater in open cases. Complications occurred in 11 cases (13%) and differed significantly between surgical techniques (p < .001). There was no statistically significant difference in length of postoperative stay across approaches. Mean follow‐up was 12.8 months (range 1–39). Favorable Engel outcomes were experienced by 37 (78.7%) of the patients who underwent craniotomy, 10 (58.8%) with LiTT, and 12 (63.2%) with mc/e; these differences were not statistically significant.SignificanceCC is an effective surgical modality for children with DRE. Regardless of surgical modality, complication rates are acceptable and seizure outcomes generally favorable. Newer, less‐invasive, surgical approaches may lead to increased adoption of this efficacious therapeutic option for pediatric DRE.

Abstract Objective Dravet syndrome (DS) is a developmental and epileptic encephalopathy characterized by high seizure burden, treatment‐resistant epilepsy, and developmental stagnation. Family members rate communication deficits among the most impactful disease manifestations. We evaluated seizure burden and language/communication development in children with DS. Methods ENVISION was a prospective, observational study evaluating children with DS associated with SCN1A pathogenic variants ( SCN1A + DS) enrolled at age ≤5 years. Seizure burden and antiseizure medications were assessed every 3 months and communication and language every 6 months with the Bayley Scales of Infant and Toddler Development 3rd edition and the parent‐reported Vineland Adaptive Behavior Scales 3rd edition. We report data from the first year of observation, including analyses stratified by age at Baseline: 0:6–2:0 years:months (Y:M; youngest), 2:1–3:6 Y:M (middle), and 3:7–5:0 Y:M (oldest). Results Between December 2020 and March 2023, 58 children with DS enrolled at 16 sites internationally. Median follow‐up was 17.5 months (range = .0–24.0), with 54 of 58 (93.1%) followed for at least 6 months and 51 of 58 (87.9%) for 12 months. Monthly countable seizure frequency (MCSF) increased with age (median [minimum–maximum] = 1.0 in the youngest [1.0–70.0] and middle [1.0–242.0] age groups and 4.5 [.0–2647.0] in the oldest age group), and remained high, despite use of currently approved antiseizure medications. Language/communication delays were observed early, and developmental stagnation occurred after age 2 years with both instruments. In predictive modeling, chronologic age was the only significant covariate of seizure frequency (effect size = .52, p = .024). MCSF, number of antiseizure medications, age at first seizure, and convulsive status epilepticus were not predictors of language/communication raw scores. Significance In infants and young children with SCN1A + DS, language/communication delay and stagnation were independent of seizure burden. Our findings emphasize that the optimal therapeutic window to prevent language/communication delay is before 3 years of age.

Progression of myeloproliferative neoplasms to blast phase (BPMPN) is associated with lack of response to conventional therapies and dire clinical outcomes. Consequently, there is a major unmet need to develop new therapies for BPMPN. Chromothripsis, the process of catastrophic shattering and haphazard repair of chromosomes, is a key contributor to somatic variation in cancer, but this phenomenon has not yet been described in BPMPN. More broadly, whether chromothripsis might result in actionable molecular events that are amenable to targeting remains an open question. To characterise the contribution of structural variants to BPMPN, we first performed integrated copy number and mutation profiling in 64 BPMPN patients by SNP karyotyping and targeted next generation sequencing. We observed a recurrent pattern of chromothripsis that involved chromosome 21, which together with other structural variants led to amplification of a common region of chromosome 21 (‘chr21amp‘) in ~25% of patients (GISTIC q-val&lt;0.01, Fig 1A). Chr21amp was associated with TP53 mutations and a higher number of copy number alterations. Patients with chr21amp had a particularly aggressive and treatment-resistant phenotype, with 0% surviving 12 months compared to 46% in the non-chr21amp pts ( p=0.0007), retaining significance in multivariate analysis including after correction for TP53 mutation status. Whole genome sequencing confirmed that the chromosomal rearrangements resulting in chr21amp occurred by different mechanisms, ranging from simple amplification to highly complex chromothriptic events involving multiple chromosomes. There were no recurrent translocation partners or mutations. The minimally amplified region (MAR) spanned 2.7Mb and contained 24 genes, with a median copy number of 3.5 (range 2.7-8.3) Single-cell transcriptomics combined with allelic resolution genotyping revealed that chr21amp was present in the dominant subclone and occurred subsequent to JAK2V617F and mut TP53 acquisition. Chr21amp was detectable in phenotypic HSCs and throughout early stages of hematopoiesis, but not in mature erythroid cells, consistent with a differentiation block. Of the 24 genes in the minimally amplified region, only one gene, DYRK1A, a serine threonine kinase linked to cell proliferation and survival, was both differentially expressed (single-cell and bulk RNAseq) and differentially accessible (ATACseq). To explore the functional role of DYRK1A in BPMPN we performed shRNA and CRISPR-mediated DYRK1A-knockdown and knockout (KO) in BPMPN cell lines (HEL and SET-2), which led to impaired cell proliferation. The DYRK1A inhibitors EHT1610 and GNF2133 also led to dose-dependent growth inhibition. DYRK1A-KO BPMPN HEL or SET2 cell clones showed a reduced ability to propagate leukemia in vivo with a significant survival advantage vs. wild type control mice. BPMPN chr21amp+ primary patient CD34+ cells were highly sensitive toDYRK1A inhibitors, while healthy control CD34+ cells were unaffected Prior studies have shown that DYRK1A activates the DREAM complex, a transcriptional repressor of DNA-repair pathways. In chr21amp patient cells, the DREAM DNA repair gene signature was significantly downregulated (NES -1.74, q-val &lt;0.001), while conversely in CRISPR DYRK1A KO SET2 cells the transcriptional DNA repair signature was upregulated (NES 1.76, q-val &lt;0.001). In functional assays, CRISPR DYRK1A KO was protective against DNA damage, with a reduction in γ-H2AX foci after etoposide treatment or irradiation ( p&lt;0.01 for both). A second mechanism of leukemogenesis emerged from geneset enrichment analyses, which suggested enhanced JAK-STAT signaling in chr21amp BPMPN cells and downregulation in the CRISPR DYRK1A KO context. We validated this by showing that DYRK1A overexpression activates and potentiates STAT5B transcriptional activity in a luciferase reporter assay. Finally, we noted that the STAT target BCL2 was selectively upregulated in chr21amp cells. BCL2 inhibition showed strong synergy with DYRK1A inhibitors for induction of BPMPN cell apoptosis (Bliss synergy score 15). Collectively, these findings define the chr21amp event as a novel prognostic biomarker in BPMPN. We pinpoint DYRK1A amplification as a central driver of genomic instability and exacerbated JAK-STAT signalling, for the first time linking chromothripsis to a specific druggable target ( Fig 1B).

Background Treating patients with multiple myeloma (MM) is a complex endeavor that requires effective clinical decision-making from a team of specialists and supporting clinicians. Additionally, treatment with chimeric antigen receptor (CAR) T-cell therapy presents unique challenges, requiring communication and collaboration between the oncology clinic and CAR T-cell treatment centers. To improve collaborative MM care, we conducted a quality improvement (QI) study, bringing together community and academic oncology centers to assess current practice patterns and develop strategies to overcome barriers in the treatment of patients with MM across the continuum of care. Methods This QI initiative is comprised of baseline provider surveys (Community n = 21; Academic n = 18) and pre- and post-surveys of providers (n = 31) participating in small-group, team-based audit-feedback (AF) sessions (Table 1). Survey questions were designed to assess knowledge, confidence, and experiences in keeping up with the latest clinical evidence and guideline recommendations and collaboration and communication practices between centers. Care teams from each clinic, along with a nationally renowned expert MM physician, participated in audit and feedback sessions to (a) assess system-specific practice gaps identified via the provider surveys, (b) prioritize areas for improvement, and (c) develop action plans for addressing root causes. Results Provider Surveys: When asked their top challenge encountered in managing patients with MM, community-based providers reported keeping up with clinical evidence for new and emerging therapies (48%), whereas academic-based providers reported knowing how to optimally sequence therapies (50%). Further differences between community and academic clinicians were evident in their self-reported confidence in their ability to stay up-to-date with new data on CAR T-cell therapies and other MM treatment (high/very high confidence: 84% academic; 24% community; p &lt; .001), ability to align clinical practices with evidence-based guidelines (high/very high confidence: 72% academic; 34% community; p = .018), and ability to sequence lines of therapy and differentiate between novel therapies for MM (high/very high confidence: 72% academic; 29% community; p = .007). Despite these differences, care teams from both settings reported low percentages of eligible patients have received CAR T-cell therapy for MM, citing patient resistance, cost and difficulty gaining insurance approval as top barriers (Figure 1). When asked to rate the level of communication and collaboration between academic and community centers, 44% of academic providers and 35% of community providers rated very good/excellent. The top barrier among providers not indicating very good/excellent communication from both settings was lack of time for our clinic's staff to communicate. AF Sessions: Providers participating in the small group AF sessions reported gaining access to newer therapies (55%) as their biggest challenge regarding MM management. Clinicians indicated gains in moderate/high/very high confidence in their ability to align treatment decisions with guidelines and clinical evidence for MM (78% to 92%) and ability to sequence treatment options across multiple lines of therapy for patients with MM (80% to 100%). Care teams set goals to enhance knowledge of clinical efficacy and safety of new and emerging therapies for MM. Action plans to achieve these goals include developing a hematology disease specific tumor board to collaborate on patient care plans, further train designated staff members focused on CAR T-cell/MM, discuss relevant new data to improve the team's confidence in therapy sequencing and early referrals to CAR T-cell centers. Conclusions Through this QI initiative, clinical teams identified barriers to optimal MM care, creating and implementing specific action plans to increase knowledge of current and emerging therapies for MM to enhance individualized treatment plans. While meaningful confidence gains were demonstrated, these data underscore clinical practice gaps to address in future initiatives to support evidence-based, collaborative care plans for patients with MM. Study Sponsor Statement The study reported in this abstract was funded by educational grants from Bristol Myers Squibb. The grantors had no role in the study design, execution, analysis, or reporting.