Developmental Trajectories of Autism
Most autistic individuals show improved communication and social functioning as they age, but not all do. Trajectory group membership is correlated with socioeconomic status. Future research should investigate what drives these correlations.
KEY FINDINGS: Most autistic individuals show improved communication and social functioning as they age, but not all do. Trajectory group membership is correlated with socioeconomic status. Future research should investigate what drives these correlations.
BACKGROUND: The goal of this study was to describe the typical, longitudinal, developmental trajectories of communication and social functioning in individuals with autism spectrum disorder from childhood through adulthood and to determine the correlates of these trajectories.
DETAILS: Children with autism spectrum disorder who were born in California from 1992 through 2016 and enrolled with the California Department of Developmental Services were identified. Subjects with <4 evaluations in the database were excluded, resulting in a sample of 71,285 individuals. Score sequences were constructed based on evaluative items for communication and social functioning. Typical trajectories were identified using group-based latent trajectory modeling, and logistic regression was used to determine the odds of classification into a social adolescent decline trajectory by individual-, family-, and zip code-level factors. Six typical patterns of communication functioning and 7 typical patterns of social functioning were identified. Whereas the majority of autistic individuals exhibit improved communication functioning as they age, the majority of individuals exhibit steady social functioning. A small group of individuals (5.0%) exhibits high social functioning in childhood that declines in adolescence. Membership in this adolescent decline group is associated with maternal non-Hispanic white race and ethnicity, female sex, moderate levels of maternal education, lower zip code-level median home values and population density, and higher zip code-level inequality.
Our Most Popular Resources
Copyright © American Academy of Pediatrics. All rights reserved.
Source: Fountain, C., Winter, A. S., Cheslack-Postava, K., et al. (2023). Developmental Trajectories of Autism . XXXXXXXX. 2023; 152(3): e2022058674. Published: September, 2023. DOI: 10.1542/peds.2022-058674.
Next-Generation Phenotyping May Strengthen Variant Interpretation in Mowat-Wilson Syndrome
In a proof-of-concept study, GestaltMatcher ranked Mowat-Wilson syndrome as the leading diagnosis in a 4-year-old child with a de novo ZEB2 variant. Three of 4 facial images met the PP4 moderate threshold and 1 met the supporting threshold. Integration of facial, HPO, and molecular data through PEDIA ranked ZEB2 as the most likely disease-causing gene. The findings suggest that AI-assisted phenotyping may provide quantitative support for variant interpretation while complementing expert clinical genetics assessment.
KEY FINDINGS: Computational facial phenotyping using GestaltMatcher identified MWS as the leading diagnosis in a child with a suspected de novo ZEB2 variant and provided quantitative evidence that supported variant interpretation. Three of 4 facial images reached the PP4 moderate threshold, while 1 reached the supporting threshold. Integration of facial phenotype, HPO information, and molecular data through PEDIA ranked ZEB2 first and contributed to classification of the variant as likely pathogenic. The findings provide a proof of concept for incorporating NGP into rare-disease diagnostic and variant-interpretation workflows, although larger, multicenter studies are needed to establish reproducibility and generalizability.
BACKGROUND: Mowat-Wilson syndrome (MWS) is a rare neurodevelopmental disorder caused by pathogenic variants in the ZEB2 gene and characterized by developmental impairment, epilepsy, congenital anomalies, and distinctive craniofacial features. Although next-generation phenotyping (NGP) tools such as GestaltMatcher can assist in recognizing rare genetic disorders, their use as quantitative evidence for variant interpretation within the American College of Medical Genetics and Genomics (ACMG) framework remains less established. This study evaluated whether computational facial phenotyping and multimodal clinical data could support variant prioritization and the ACMG PP4 phenotype-specificity criterion in MWS.
DETAILS: The investigators applied GestaltMatcher to a 4-year-old child with an undiagnosed neurodevelopmental disorder, suspected MWS, and a de novo ZEB2 variant. Facial photographs were analyzed alongside Human Phenotype Ontology (HPO) terms and simulated exome data using the PEDIA framework. Bayesian likelihood modeling was used to establish Gestalt score thresholds corresponding to supporting, moderate, strong, and very strong PP4 evidence. Brain MRI was also analyzed for structural abnormalities associated with MWS. The GestaltMatcher model had been trained using 9,671 images from 7,294 patients representing 275 disorders. For the MWS analysis, the dataset included 11 individuals with MWS and 14 age-matched controls.
Our Most Popular Resources
GestaltMatcher ranked MWS as the leading diagnosis, while integration through PEDIA also ranked ZEB2 as the most likely disease-causing gene. Across the 4 facial images from the proband, 3 achieved the PP4 moderate threshold and 1 achieved the PP4 supporting threshold. The established Gestalt score thresholds were 0.5111 for supporting, 0.5672 for moderate, 0.6609 for strong, and 0.7632 for very strong evidence.
Using the NGP-derived PP4 moderate evidence together with the original ACMG evidence, the ZEB2 variant was classified as likely pathogenic. Brain MRI demonstrated subtle thinning of the corpus callosum, a finding consistent with previously reported MWS-associated neuroimaging abnormalities. MRI-derived features also differentiated individuals with MWS from age-matched controls in exploratory analysis.
An additional exploratory case involving an infant with molecularly confirmed MWS showed that GestaltMatcher could prioritize MWS based solely on infant facial characteristics. The authors emphasize that NGP is intended to complement, rather than replace, expert clinical genetic assessment.
Copyright © Skyscape. All rights reserved.
Source: Hsieh, T. C., Todd, D., Warner, T., et al. Leveraging Next-Generation Phenotyping in Dysmorphology to Support Variant Interpretation in Mowat-Wilson Syndrome. Neurology Genetics. 2026; 12(4): e200384. Published: July 1, 2026. DOI: 10.1212/NXG.0000000000200384
Long-Term Outcomes of Azacitidine, Venetoclax, and Gilteritinib in Newly Diagnosed FLT3-Mutated Acute Myeloid Leukemia
In a phase II study of 30 patients with newly diagnosed FLT3-mutated AML, azacitidine, venetoclax, and gilteritinib produced a 96% CR/CRi rate. At a median follow-up of 41.5 months, median RFS and OS were 23.4 and 29.7 months, with 3-year rates of 43% and 46%, respectively. Baseline RAS pathway mutations were associated with poorer outcomes, while 67% of evaluable relapses lacked detectable FLT3 mutations. The triplet produced durable remissions but frequently required treatment dose or duration reductions because of myelosuppression.
KEY FINDINGS: In this phase II study, azacitidine, venetoclax, and gilteritinib produced high remission rates and encouraging long-term survival in patients with newly diagnosed FLT3-mutated AML who were unfit for intensive chemotherapy. The 3-year RFS and OS rates were 43% and 46%, respectively, although outcomes were less favorable in patients with FLT3-ITD mutations and baseline RAS pathway mutations. Most evaluable relapses were FLT3-negative, suggesting that resistance may involve mechanisms beyond persistent FLT3-mutated disease. Dose or duration reductions were frequently required during consolidation, primarily reflecting the challenge of managing myelosuppression. Randomized studies are needed to establish the role of this triplet regimen relative to current standard approaches.
BACKGROUND: Relapse after frontline azacitidine and venetoclax remains a major challenge in patients with FLT3-mutated acute myeloid leukemia (AML), with relapse often associated with expansion of FLT3-mutated clones. This phase II study evaluated the long-term efficacy and safety of adding the FLT3 inhibitor gilteritinib to azacitidine and venetoclax in adults with newly diagnosed FLT3-mutated AML who were considered unfit for intensive chemotherapy.
DETAILS: Thirty patients with newly diagnosed FLT3-mutated AML were treated with the azacitidine, venetoclax, and gilteritinib triplet regimen. The median age was 71 years, and 22 (73%) patients had FLT3-ITD mutations. Fourteen patients (47%) proceeded to allogeneic hematopoietic stem cell transplantation in first remission. The median follow-up was 41.5 months.
Our Most Popular Resources
The complete remission or complete remission with incomplete hematologic recovery rate was 96%. Eleven patients (37%) subsequently relapsed, and among evaluable relapses, the FLT3 mutation was no longer detectable in 67%, indicating that relapse mechanisms were not necessarily driven by persistent FLT3-mutated disease.
The median relapse-free survival (RFS) was 23.4 months and median overall survival (OS) was 29.7 months. At 3 years, RFS was 43% and OS was 46%. Among patients with FLT3-ITD-mutated AML, median RFS and OS were 17.0 months and 21.8 months, respectively, with 3-year RFS and OS rates of 32% and 36%. Baseline RAS pathway mutations were associated with poorer outcomes.
Survival outcomes were similar regardless of whether patients underwent allogeneic hematopoietic stem cell transplantation in first remission. Among patients receiving at least one consolidation cycle, 68% required a reduction in the dose or duration of at least one study drug, highlighting the need to manage treatment-related myelosuppression during prolonged therapy.
Copyright © Skyscape. All rights reserved.
Source: Short, N. J., Kantarjian, H. M., Daver, N., et al. Long-Term Outcomes of Azacitidine, Venetoclax and Gilteritinib in Newly Diagnosed FLT3-Mutated AML. Blood Advances.. 2026; 10(17): 5743–5750. Published: May 27, 2026. DOI: 10.1182/bloodadvances.2026019841.
[Posted 2/Sep/2026]
AUDIENCE: Hematology, Oncology
KEY FINDINGS: In this phase II study, azacitidine, venetoclax, and gilteritinib produced high remission rates and encouraging long-term survival in patients with newly diagnosed FLT3-mutated AML who were unfit for intensive chemotherapy. The 3-year RFS and OS rates were 43% and 46%, respectively, although outcomes were less favorable in patients with FLT3-ITD mutations and baseline RAS pathway mutations. Most evaluable relapses were FLT3-negative, suggesting that resistance may involve mechanisms beyond persistent FLT3-mutated disease. Dose or duration reductions were frequently required during consolidation, primarily reflecting the challenge of managing myelosuppression. Randomized studies are needed to establish the role of this triplet regimen relative to current standard approaches.
BACKGROUND: Relapse after frontline azacitidine and venetoclax remains a major challenge in patients with FLT3-mutated acute myeloid leukemia (AML), with relapse often associated with expansion of FLT3-mutated clones. This phase II study evaluated the long-term efficacy and safety of adding the FLT3 inhibitor gilteritinib to azacitidine and venetoclax in adults with newly diagnosed FLT3-mutated AML who were considered unfit for intensive chemotherapy.
DETAILS: Thirty patients with newly diagnosed FLT3-mutated AML were treated with the azacitidine, venetoclax, and gilteritinib triplet regimen. The median age was 71 years, and 22 (73%) patients had FLT3-ITD mutations. Fourteen patients (47%) proceeded to allogeneic hematopoietic stem cell transplantation in first remission. The median follow-up was 41.5 months.
The complete remission or complete remission with incomplete hematologic recovery rate was 96%. Eleven patients (37%) subsequently relapsed, and among evaluable relapses, the FLT3 mutation was no longer detectable in 67%, indicating that relapse mechanisms were not necessarily driven by persistent FLT3-mutated disease.
The median relapse-free survival (RFS) was 23.4 months and median overall survival (OS) was 29.7 months. At 3 years, RFS was 43% and OS was 46%. Among patients with FLT3-ITD-mutated AML, median RFS and OS were 17.0 months and 21.8 months, respectively, with 3-year RFS and OS rates of 32% and 36%. Baseline RAS pathway mutations were associated with poorer outcomes.
Survival outcomes were similar regardless of whether patients underwent allogeneic hematopoietic stem cell transplantation in first remission. Among patients receiving at least one consolidation cycle, 68% required a reduction in the dose or duration of at least one study drug, highlighting the need to manage treatment-related myelosuppression during prolonged therapy.
Copyright © Skyscape. All rights reserved.
Source: Short, N. J., Kantarjian, H. M., Daver, N., et al. Long-Term Outcomes of Azacitidine, Venetoclax and Gilteritinib in Newly Diagnosed FLT3-Mutated AML. Blood Advances.. 2026; 10(17): 5743–5750. Published: May 27, 2026. DOI: 10.1182/bloodadvances.2026019841.
Parental Responsibility May Drive Hesitation Toward AI Educational Services
This research is the first to uncover the impact of educational approaches on individuals' guilt and downstream behaviours in the AI-in-Education field, shedding light on attribution as its underlying mechanism and offering actionable strategies to enhance individuals' WOM. The findings offer novel insights to AI-human interaction psychological research and hold practical implications for AI-in-Education industry practitioners.
KEY FINDINGS: Across five experimental studies, AI educational services were associated with greater guilt, lower perceived value, and, in several conditions, less willingness to recommend the approach compared with direct parental engagement. The findings indicate that reluctance to use AI for children's education may be linked less to perceptions of AI capability and more to the belief that educating one's children is a parental responsibility. The study also suggests that framing AI use as necessary because of an individual's limitations, or demonstrating that other parents use AI services, may improve positive WOM toward these services.
BACKGROUND: Artificial intelligence (AI) educational services are increasingly positioned as tools that can assist children with learning and tutoring. However, the decision to use these services may be influenced by more than their perceived educational capability. This research examined how choosing AI educational services rather than direct parental involvement affects guilt, perceived value, and willingness to recommend the approach to others. Across five experimental studies, the investigators also examined whether perceived parental responsibility, intrinsic reasons for using AI, and conformity influence these responses.
DETAILS: The research used experimental designs comparing parental engagement with AI educational services across homework and writing-tutoring scenarios. Study 1a included 191 participants after exclusion of 9 cases; Study 1b included 200 participants; Study 2 included 200 participants; Study 3 included 390 participants; and Study 4 included 400 participants. Participants were recruited through the Credamo online platform.
Our Most Popular Resources
In Study 1a, participants who considered using AI educational services reported greater guilt and assigned a lower monetary valuation than those who personally tutored their children. Mean guilt scores were 2.74 (SD 1.81) with AI educational services versus 2.09 (SD 1.58) with parental engagement (t=2.67, p=.008). Mean valuation was 1219.31 (SD 1154.32) versus 2097.99 (SD 2217.96), respectively (t=3.41, p=.001).
Study 1b reproduced these findings without using pictures and after accounting for parental status. Guilt was higher with AI educational services than with parental engagement (5.23 [SD 2.94] vs 3.79 [SD 2.47]; F(1, 198)=14.07, p<.001, ηp2=.07). Valuation was lower with AI educational services (1219.62 [SD 1581.27] vs 2494.11 [SD 2530.90]; F(1, 198)=18.24, p=.001, ηp2=.08).
Study 2 extended the analysis to willingness to recommend the educational approach. Participants using AI educational services reported greater guilt, lower valuation, and lower word-of-mouth (WOM) intentions than participants engaging in education themselves. Among participants with children, guilt means were 3.09 (SD 2.35) for AI educational services and 2.32 (SD 2.31) for parental engagement (F=4.18, p=.043, ηp2=.03). Valuation was 477.83 (SD 547.20) versus 968.14 (SD 780.09), respectively (F=21.14, p<.001, ηp2=.12), while WOM was 7.83 (SD 1.65) versus 8.23 (SD 1.41) (F=4.77, p=.030, ηp2=.03).
Perceived responsibility for children's education emerged as an important explanatory mechanism. Attribution scores among participants with children were 6.27 (SD 2.50) in the AI condition and 8.79 (SD 0.86) in the parental-engagement condition (F=87.92, p<.001, ηp2=.36). Mediation analysis showed that attribution significantly mediated the relationship between educational approach and guilt, valuation, and WOM, with effects of 0.6116, -183.9280, and -0.8910, respectively.
Study 3 demonstrated that the reason for choosing AI could alter the pattern of WOM responses. When participants lacked the ability to tutor their children, those using AI educational services reported greater positive WOM than those personally tutoring their children: 7.60 (SD 1.47) versus 6.76 (SD 2.29), p=.006. In the control condition, the pattern was reversed, with WOM scores of 7.36 (SD 1.90) for AI educational services and 7.86 (SD 1.23) for parental engagement (p=.04).
Study 4 found that social conformity also influenced WOM. Overall, WOM was lower with AI educational services than with parental engagement: 7.59 (SD 2.19) versus 8.30 (SD 1.10) (F(1, 396)=17.23, p.001, ηp2=.04). When participants were given information that other parents were using AI educational services, the difference was no longer statistically significant (7.88 [SD 1.94] vs 8.24 [SD 1.23], p=.133). Without such conformity information, WOM was significantly lower for AI educational services (7.29 [SD 2.39] vs 8.36 [SD 0.94], p<.001).
Copyright © Skyscape. All rights reserved.
Source: Shao, A., Lu, Z., Liu, S. Q., et al. Demystifying the mist: Why do individuals hesitate to accept AI educational services?. British Journal of Psychology. 2026; 117(3): 932-956. Published: August, 2026. DOI: 10.1111/bjop.70040.
US Measles Cases Rise in 2026 as Outbreaks Continue Amid Declining MMR Coverage
US measles activity has increased substantially in 2026, with 2,465 confirmed cases and 38 new outbreaks reported as of August 6. Declining MMR coverage may increase community vulnerability to transmission. Maintaining high vaccination coverage remains central to preventing measles outbreaks and sustaining elimination.
KEY FINDINGS: As of August 6, 2026, the US had reported 2,465 confirmed measles cases and 38 new outbreaks, with 94% of cases outbreak-associated. The increase occurs alongside a decline in kindergarten MMR coverage to 92.5% in 2024-2025 from 95.2% in 2019-2020. CDC emphasizes that measles can spread rapidly in communities with lower vaccination coverage, while 2 doses of MMR vaccine provide 97% protection against measles.
BACKGROUND: Measles was officially eliminated in the United States in 2000 following widespread use of the measles, mumps, and rubella (MMR) vaccine. However, declining vaccination coverage and increasing global measles activity have increased opportunities for measles transmission following importation into the United States.
DETAILS: As of August 6, 2026, the Centers for Disease Control and Prevention (CDC) reported 2,465 confirmed measles cases in the United States in 2026. Of these, 2,449 cases were reported by 47 jurisdictions, while 16 cases occurred among international visitors to the United States. Thirty-eight new outbreaks had been reported during 2026.
Our Most Popular Resources
Overall, 94% of confirmed cases in 2026 (2,309 of 2,465) were associated with outbreaks, including 936 cases from outbreaks beginning in 2026 and 1,373 from outbreaks that began in 2025. For comparison, 2,289 confirmed cases and 48 outbreaks were reported during the full year of 2025; 90% of cases (2,066 of 2,289) were outbreak-associated.
CDC reports confirmed measles cases notified by jurisdictions as of noon on Thursdays. An outbreak is defined as 3 or more related cases. State and CDC counts may differ because jurisdictions update and publicly report their data on different schedules.
MMR vaccination coverage among US kindergartners declined from 95.2% during the 2019-2020 school year to 92.5% during the 2024-2025 school year, leaving approximately 286,000 kindergartners at risk during the 2024-2025 school year. CDC notes that communities with vaccination coverage below the 95% level are more vulnerable to measles outbreaks. The 2026 measles case count reported by CDC as of August 6, 2026, had already exceeded the total number of confirmed cases reported during all of 2025 (2,465 vs 2,289). The high proportion of outbreak-associated cases indicates sustained transmission within affected communities.
The burden of measles remains closely associated with vaccination status. CDC reports that 2 doses of MMR vaccine are 97% effective at preventing measles, while 1 dose is 93% effective. Breakthrough infections can occur, particularly during outbreaks with high levels of circulating measles virus, and account for approximately 10% of all measles infections.
Copyright © Skyscape. All rights reserved.
Source: CDC: Measles Cases and Outbreaks. Centers for Disease Control and Prevention (CDC). 2026; Published: August 7, 2026.
Global Forecasts Reveal Rising Antimicrobial Resistance Threats in Children
Global pediatric antimicrobial resistance increased from 2004 to 2022 across all regions. Resistance was highest among Gram-negative pathogens, children in intensive care, those with sepsis, and resource-limited settings. By 2035, carbapenem resistance is projected to reach 35% in Klebsiella species and 82% in Acinetobacter baumannii.
KEY FINDINGS: This global pediatric surveillance study demonstrates a sustained increase in antimicrobial resistance among children, with the greatest concern involving Gram-negative pathogens, intensive care settings, sepsis cases, and resource-limited regions. Increasing resistance to Watch and Reserve antibiotics may compromise treatment options for severe childhood infections. Forecasted growth in carbapenem resistance among Klebsiella species and Acinetobacter baumannii highlights the need for strengthened antimicrobial stewardship, surveillance systems, and development of effective pediatric treatment strategies.
BACKGROUND: Antimicrobial resistance (AMR) threatens the effectiveness of antibiotic therapy for severe childhood infections, yet comprehensive pediatric AMR surveillance data across multiple regions remain limited. This study evaluated global and temporal patterns of antimicrobial resistance among children using the World Health Organization (WHO) Access, Watch, and Reserve (AWaRe) antibiotic classification framework and projected future resistance trends
DETAILS: This cross-sectional surveillance study analyzed pediatric bacterial isolates from the Antimicrobial Testing Leadership and Surveillance (ATLAS) database collected between January 2004 and December 2022. The study included 106 581 isolates from 106 581 children aged 0 to 18 years across 82 countries. Data were analyzed from February 2024 to April 2026. Resistance trends were assessed by geographic region, age group, clinical setting, infection syndrome, and pathogen type, with spatiotemporal models used to forecast resistance patterns through 2035.
Our Most Popular Resources
The analysis categorized antibiotics according to the WHO AWaRe framework: Access antibiotics used for first-line treatment, Watch antibiotics with higher resistance potential, and Reserve antibiotics intended for difficult-to-treat infections. The study evaluated WHO priority pathogens and examined resistance patterns among different pediatric populations, including children with sepsis and respiratory infections.
From 2004 to 2022, pediatric AMR increased across all regions, with higher resistance levels and faster increases observed in resource-limited settings. Resistance to Access-group antibiotics was highest overall, with a mean resistance of 36% (range, 2%-66%), compared with Watch-group antibiotics at a mean of 22% (range, 1%-47%) and Reserve-group antibiotics at a mean of 13% (range, 0%-30%).
Resistance to higher-tier antibiotics increased substantially in vulnerable clinical groups. In intensive care units, Watch-group resistance increased from 15% (517/3564) to 33% (2910/8748) (P < .001), particularly among children aged 0 to 2 years, where resistance increased from 12% (325/2649) to 32% (1257/3959) (P < .001). Among children with sepsis, Watch-group resistance increased from 15% (298/2030) to 30% (1409/4705) (P < .001), while Reserve-group resistance increased from 3% (16/474) to 26% (746/2824) (P < .001).
Among critical pathogens, Acinetobacter baumannii demonstrated the highest overall resistance, exceeding 55% in every AWaRe antibiotic category in 2022. Klebsiella species showed the fastest increases in resistance, particularly to third- or fourth-generation cephalosporins and carbapenems. Forecasts estimated that by 2035, carbapenem resistance would reach 35% (95% uncertainty interval [UI], 29%-40%) in Klebsiella species and 82% (95% UI, 77%-85%) in A baumannii.
Copyright © Skyscape. All rights reserved.
Source: Hu, Y. J., Qiu, H., Harwell, J. I., et al. Childhood Antimicrobial Resistance With Global Forecasts. JAMA Pediatrics. 2026; Published: July 20, 2026. DOI: 10.1001/jamapediatrics.2026.2808.
Specialty: