Making sense of artificial intelligence and large language models-including ChatGPT-in pediatric hematology/oncology.

ChatGPT and other artificial intelligence (AI) systems have captivated the attention of healthcare providers and researchers for their potential to improve care processes and outcomes. While these technologies hold promise to automate processes, increase efficiency, and reduce cognitive burden, their use also carries risks. In this commentary, we review basic concepts of AI, outline some of the capabilities and limitations of currently available tools, discuss current and future applications in pediatric hematology/oncology, and provide an evaluation and implementation framework that can be used by pediatric hematologist/oncologists considering the use of AI in clinical practice.

Short-Course Empiric Antibiotics in Children Undergoing Allogeneic Hematopoietic Cell Transplantation.

Fever is common in children undergoing hematopoietic cell transplantation (HCT). Empiric antibiotic (EA) therapy is initiated and often continued until neutrophil engraftment. Prolonged antibiotic exposure reduces microbiome diversity and causes overgrowth of pathogenic organisms, leading to such complications as infections from antibiotic-resistant organisms and Clostridium difficile colitis. Shorter courses of EA therapy have been studied in adults undergoing HCT without significant safety concerns, but data in children are lacking. We instituted a single-center preintervention/ postintervention quality improvement (QI) project to assess the feasibility of short-course EA therapy for first fever in patients undergoing HCT. We aimed to reduce the median duration of broad-spectrum antibiotic use in eligible patients from 20 days in 2020 to 10 days in 2021. Patients were eligible for the intervention, limiting EAs to 7 days for first fever, if they were admitted for their first allogeneic HCT, were afebrile for >24 hours, had no infection requiring systemic treatment, and were hemodynamically stable. Outcome measures included days of EA therapy for first fever and total broad-spectrum antibiotic use during the period of hospitalization, defined as the time from the start of conditioning to 30 days after HCT or hospital discharge, whichever occurred first. Balancing measures included bloodstream infection (BSI), fever, and intensive care (ICU) admission within 3 days of stopping EA therapy. Project criteria were applied retrospectively to patients who underwent HCT in 2020 to construct a preintervention short-course-eligible cohort. During the intervention period, 41 patients underwent allogeneic HCT, of whom 17 (41%) were eligible for short-course EA therapy. Among eligible patients, the median age was 5.3 years, 47% had an underlying malignancy, and 88% received myeloablative conditioning. There were no differences in demographic or HCT characteristics between patients eligible for short-course EA during the intervention and preintervention period (n = 24). The short-course EA schedule was adhered to by 14 of the 17 eligible patients (82%). The duration of EA for first fever and total broad-spectrum antibiotic use was significantly decreased in the short-course EA-eligible patients compared to the preintervention cohort, from a median of 17 days to 8 days and from 20 days to 10 days, respectively (P < .01). Of the 14 patients adhering to short-course EA, 2 experienced a balancing measure of recurrent fever requiring resumption of EA, but no infection was identified. There were no BSIs, ICU admissions, or deaths during the hospitalization period in patients who received short-course EA. In this single-center QI project, short-course EA for initial fever was successfully applied to children undergoing allogeneic HCT using strict criteria and led to a significant decrease in broad-spectrum antibiotic use during hospitalization. These results should be validated in a prospective clinical trial to include the impact of short-course EA on antibiotic-resistant organisms, the intestinal microbiome, and HCT outcomes.

Cell Therapy Informatics: Updates on the Integration of HCT/IEC Functionalities into an Electronic Medical Record System in the US to Promote Efficiency, Patient Safety, Research, and Data Interoperability.

The use of electronic health/medical record (EMR) systems has streamlined medical practice and improved efficiency of clinical care in recent years. However, EMR systems are not generally well designed to support research and tracking of longitudinal outcomes across populations, which are particularly important in hematopoietic stem cell transplantation (HCT) and immune effector cell therapy (IEC), where data reporting to registries and regulatory agencies are often required. Since its formation in 2014, the HCT EMR user group has worked with a large EMR vendor (Epic) to develop many functionalities within the EMR to improve the care of HCT/IEC patients and facilitate the capture of HCT/IEC data in an easily interoperable format. Awareness and the widespread adoption of these new tools among transplant centers remains a challenge, however. In this report, we aim to increase awareness and adoption of these new features in the Epic EMR across the transplantation community, advocate for the use of data standards, and promote future collaboration with other commercial EMRs to develop standardized HCT/IEC content to improve patient care and facilitate interoperable data exchange.

Tailoring Therapy for Children With Neuroblastoma on the Basis of Risk Group Classification: Past, Present, and Future.

For children with neuroblastoma, the likelihood of cure varies widely according to age at diagnosis, disease stage, and tumor biology. Treatments are tailored for children with this clinically heterogeneous malignancy on the basis of a combination of markers that are predictive of risk of relapse and death. Sequential risk-based, cooperative-group clinical trials conducted during the past 4 decades have led to improved outcome for children with neuroblastoma. Increasingly accurate risk classification and refinements in treatment stratification strategies have been achieved with the more recent discovery of robust genomic and molecular biomarkers. In this review, we discuss the history of neuroblastoma risk classification in North America and Europe and highlight efforts by the International Neuroblastoma Risk Group (INRG) Task Force to develop a consensus approach for pretreatment stratification using seven risk criteria including an image-based staging system-the INRG Staging System. We also update readers on the current Children's Oncology Group risk classifier and outline plans for the development of a revised 2021 Children's Oncology Group classifier that will incorporate INRG Staging System criteria to facilitate harmonization of risk-based frontline treatment strategies conducted around the globe. In addition, we discuss new approaches to establish increasingly robust, future risk classification algorithms that will further refine treatment stratification using machine learning tools and expanded data from electronic health records and the INRG Data Commons.

Wheezing is common in children with sickle cell disease when compared with controls.

In children with sickle cell disease (SCD), wheezing may occur in the absence of asthma. However, the prevalence of wheezing in children with SCD when compared with children without SCD (controls) in the same setting is unknown. Using a case-control study design, we tested the hypothesis that children with SCD would have a higher rate of wheezing than those without SCD. We enrolled 163 children with SCD (cases) and 96 children without SCD (controls) from a community hospital in Nigeria. Parent reports of respiratory symptoms were identified based on responses to questions taken from the American Thoracic Society Division of Lung Diseases' Questionnaire. The median age was 8.5 years for children with SCD and 7.7 years for controls. Cases were more likely than controls to report wheezing both with colds (17.3% vs. 2.1%, P<0.01) and without colds (4.9% vs. 0%, P=0.03). Cases had 9.8 times greater odds of wheezing (95% confidence interval, 2.3-42.2). In the multivariable model, the only variable associated with wheezing was SCD status (odds ratio=18.7, 95% confidence interval, 2.5-142; P=0.005). Children with SCD experience a significantly higher rate of wheezing when compared with children of similar age without SCD.