Laure F. Pittet

Background: The development of food sensitisation (FS) and food allergy (FA) reflects complex interactions between genetic, environmental and immunological factors, which conventional univariate approaches may fail to capture. This exploratory study applied penalised regression to determine whether early‐life immune responses to common pathogens are associated with infant FS and FA. Methods: At 1 year of age, FS was determined by skin prick test and FA by allergy/tolerance history and oral food challenge. Elastic net regression (ENR) was applied to cytokine response data (13 cytokines) from whole blood stimulation (11 microbial stimuli) collected at 6 months of age ( n = 147), together with longitudinal clinical data from MIS BAIR trial infants. Stably selected variables (0.75) were analysed by multivariable logistic regression to explore the direction of association of the stably selected variables. Results: For FS, 13 cytokine‐stimulant pairs were stably selected in cytokine‐only ENR models. After including early‐life clinical factors, four pairs were retained. Two were inversely associated with FS ( Haemophilus influenzae IL‐1Ra and Pam3CSK4 MIP‐1α). Two were positively associated ( Streptococcus pyogenes IL‐1Ra and peptidoglycan IFN‐γ). For FA, four cytokine‐stimulant pairs were selected in cytokine‐only ENR models. Two were positively associated with FA ( Escherichia coli ‐stimulated MCP‐1 and S. pyogenes IP‐10), and two inversely associated (Pam3CSK4 IL‐1Ra and H. influenzae IL‐1Ra). No cytokine responses were selected after inclusion of clinical factors in the FA model. The most consistently selected clinical factors were infant eczema for FS and topical corticosteroid treatment by 6 months for FA likely due to disease severity. Conclusion: Early‐life cytokine responses to pathogens associated with FS and FA at 1 year were identified through stability selection. While hypothesis‐generating, findings were consistent with a role for early immune perturbation in the development of food sensitisation. image

Background: Congenital cytomegalovirus (cCMV) is the most common congenital infection, with approximately 17%–20% of infected neonates developing permanent sequelae. Early risk stratification remains challenging yet crucial for therapeutic decision-making. We aimed to characterize national epidemiology, identify early predictors of sequelae, and assess the impact of early antiviral treatment on outcomes in infants with cCMV. Methods: We conducted a nationwide prospective observational study between 2017 and 2024, analyzing birth and 1-year follow-up data of infants with confirmed cCMV in Switzerland. Results: Two hundred and nine cCMV-infected children were included. The most common findings at birth were microcephaly (35%), intrauterine growth restriction (34%) and neuroimaging abnormalities (36%). Antiviral treatment was administered to 36% (75/209) of patients, among whom 75% (56/75) had treatment initiation within the first month of life. Multivariate analysis identified hepatosplenomegaly (odds ratio 13.2, 95% confidence interval, 2.3–75.2) and hypotonia at birth (odds ratio 6.6, 95% confidence interval, 1.1–41.1) as independent predictors of late-onset sensorineural hearing loss (SNHL). At the 1-year follow-up assessment, early treatment initiation was associated with significantly lower rates of neurodevelopmental disorders compared with delayed treatment initiation (23% vs. 50%, P = 0.004). Conclusions: This Swiss nationwide study confirms the previously reported association between hepatosplenomegaly at birth and late-onset SNHL and identifies hypotonia as a new independent predictor of late-onset SNHL. The data also confirm the crucial role of early antiviral treatment, particularly for neurodevelopmental disorders, emphasizing the importance of prompt cCMV diagnosis and treatment when indicated.

Background: Children with rheumatologic diseases often require immunosuppressive therapies, which can compromise their immune response and increase susceptibility to infections. Despite the heightened risk, vaccination coverage in this population is frequently suboptimal, largely due to concerns over vaccine safety, especially live attenuated vaccines, and the absence of clear, universally adopted guidelines. Objective: This retrospective study aims to evaluate the vaccination status and serological immunity of pediatric patients with rheumatologic diseases at the time of diagnosis, prior to initiation of immunosuppressive therapy. Methods: We conducted a retrospective review of medical records from children under 16 years of age diagnosed with a rheumatologic disease at Geneva University Hospitals (HUG) between 2005 and 2023. Demographic data, clinical diagnosis, vaccination history, and available vaccine-specific serologies were collected and analysed. Results: Seventy-four patients were included, with a median age at diagnosis of 6 years. At the time of diagnosis, vaccination coverage was highest for Haemophilus influenzae type b (Hib; 93%), followed by diphtheria and tetanus (91%), measles-mumps-rubella (MMR; 80%), pertussis and poliomyelitis (77–78%), pneumococcus (77%), varicella-zoster virus (VZV; 72%), and hepatitis B virus (HBV; 64%). Baseline serological testing was available in only half of the patients (38/74, 51%). Among those tested, seroprotection was highest for Hib (100%), followed by tetanus (91%), measles (89%), diphtheria (88%), hepatitis B (58%), varicella (47%), and pneumococcus (42%). Longitudinal follow-up revealed limited and inconsistent serological monitoring, with only a minority of patients undergoing repeat testing, and very few receiving booster vaccinations, despite the generally stable protection levels observed over time. Conclusions: Vaccination coverage in children with rheumatologic diseases remains suboptimal, particularly for varicella, hepatitis B, and pneumococcus. These findings underscore the critical importance of optimizing vaccination coverage at diagnosis and establishing systematic serologic surveillance protocols in this high-risk population, rather than relying solely on vaccine history.

Immunization is one of the most effective interventions to prevent infectious diseases. Identifying individuals at risk of non-adherence to immunization schedules could enable early interventions to improve coverage. With the increasing availability of large datasets with sociodemographic, medical, and geographic data, artificial intelligence (AI) offers new possibilities for identifying predictive patterns. We performed a systematic literature review of original studies that used AI to predict immunization non-adherence in children. Searches were conducted in PubMed, Embase, and Web of Science and no restrictions were applied regarding publication year, language, or study design. Study characteristics, dataset size, AI methods, performance metrics, and validation approaches were extracted. Risk of bias was assessed using PROBAST + AI. A total of 843 records were identified; after screening, 12 studies were included. Nine of 12 studies were conducted in low- and middle-income countries, used cross-sectional survey data, and analyzed vaccines from routine immunization schedules. Model performance was generally high; 9/12 studies reported an accuracy or an area under the receiver operating characteristic curve exceeding 80%, but no study conducted external validation. Predictors clustered into seven categories: maternal, paternal, family, child, geographic, wealth, and others. Among maternal and child-related factors, prenatal care and previous immunization history had a high predictive weight. Conclusion: AI demonstrates a promising potential for identifying individuals at risk of vaccine non-adherence. However, the lack of external validation raises concerns about model overfitting and limits real-world applicability. Demographic health surveys were routinely repeated across countries and time periods and could therefore represent an ideal resource for external validation. What is known: • Geographic, sociodemographic, and medical factors contribute to non-adherence to the immunization schedule.  • Identifying these individuals can guide early interventions to improve immunization coverage. What is new: • Artificial intelligence can reveal unknown influential factors; prenatal care and delivery setting emerged as key factors in this systematic review.

Objectives: Bacillus Calmette-Guérin (BCG), similar to some other live vaccines, may offer partial protection against unrelated infections. During the COVID-19 pandemic, several trials tested whether BCG provides protection against COVID-19. Design: Meta-analysis of randomised placebo-controlled trials which provided mortality data. Setting: High-income settings as well as low- and middle-income settings (LMICs). Participants: Trial populations were healthcare workers (HCW) or individuals of older age, followed for 6-12 months. Interventions: BCG vaccine versus placebo. Main outcome measure: Overall mortality and COVID-19 mortality. Results: Nine of the 15 RCTs reported at least one death. Trial populations were healthcare workers (4 RCTs) or individuals of older age (5 RCTs), in Europe (5 RCTs) or LMICs (4 RCTs) with a total of 8,169 participants randomly allocated to BCG (18 deaths) and 8,176 to placebo (37 deaths). In a meta-analysis, receiving BCG vs placebo was associated with a 51% (95% CI: 15 to72%) reduction in all-cause mortality. The all-cause mortality RRR was 78% (−3 to 95%) in HCWs (Europe and LMICs) and 43% (95% CI: −6 to 69%) in the elderly. It was 37% (95 CI: −15 to 66%) in Europe and 90% (95% CI: 22 to 99%) in LMICs. The relative reduction of deaths (6 BCG; 12 placebo) from COVID-19 was 50% (95 CI: −33 to 81). Conclusions: BCG may have reduced all-cause mortality in the context of the COVID-19 pandemic.