Machine Learning Prediction of Treatment Response to Inhaled Corticosteroids in Asthma.

BACKGROUND: Although inhaled corticosteroids (ICS) are the first-line therapy for patients with persistent asthma, many patients continue to have exacerbations. We developed machine learning models to predict the ICS response in patients with asthma. METHODS: The subjects included asthma patients of European ancestry (n = 1371; 448 children; 916 adults). A genome-wide association study was performed to identify the SNPs associated with ICS response. Using the SNPs identified, two machine learning models were developed to predict ICS response: (1) least absolute shrinkage and selection operator (LASSO) regression and (2) random forest. RESULTS: The LASSO regression model achieved an AUC of 0.71 (95% CI 0.67-0.76; sensitivity: 0.57; specificity: 0.75) in an independent test cohort, and the random forest model achieved an AUC of 0.74 (95% CI 0.70-0.78; sensitivity: 0.70; specificity: 0.68). The genes contributing to the prediction of ICS response included those associated with ICS responses in asthma (TPSAB1, FBXL16), asthma symptoms and severity (ABCA7, CNN2, PTRN3, and BSG/CD147), airway remodeling (ELANE, FSTL3), mucin production (GAL3ST), leukotriene synthesis (GPX4), allergic asthma (ZFPM1, SBNO2), and others. CONCLUSIONS: An accurate risk prediction of ICS response can be obtained using machine learning methods, with the potential to inform personalized treatment decisions. Further studies are needed to examine if the integration of richer phenotype data could improve risk prediction.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine response in adults with predominantly antibody deficiency.

Background: Patients with predominantly antibody deficiency (PAD) have lower anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike antibody levels after initial 2-dose SARS-CoV-2 vaccination than healthy controls do; however, the anti-spike antibody responses and neutralization function in patients with PAD following subsequent immunizations remain understudied. Objective: We sought to characterize anti-spike antibody responses in adults with PAD over the course of 5 SARS-CoV-2 vaccine doses and identify diagnostic and immunophenotypic risk factors for low antibody response. Methods: We evaluated anti-spike antibody levels in 117 adult patients with PAD and 192 adult healthy controls following a maximum of 5 SARS-CoV-2 immunizations. We assessed neutralization of the SARS-CoV-2 wild-type strain and the Omicron BA.5 variant and analyzed infection outcomes. Results: The patients with PAD had significantly lower mean anti-spike antibody levels after 3 SARS-CoV-2 vaccine doses than the healthy controls did (1,439.1 vs 21,890.4 U/mL [P < .0001]). Adults with secondary PAD, severe primary PAD, and high-risk immunophenotypes had lower mean anti-spike antibody levels following vaccine doses 2, 3, and/or 4 but not following vaccine dose 5. Compared with patients with mild and moderate PAD, patients with severe PAD had a higher rate of increase in anti-spike antibody levels over 5 immunizations. A strong positive correlation was observed between anti-spike antibody levels and neutralization of both the SARS-CoV-2 wild-type strain and the Omicron BA.5 variant. Most infections were managed on an outpatient basis. Conclusions: In all of the patients with PAD, anti-spike antibody levels increased with successive SARS-CoV-2 immunizations and were correlated with neutralization of both the SARS-CoV-2 wild-type strain and the Omicron BA.5 variant. Secondary PAD, severe primary PAD, and high-risk immunophenotypes were correlated with lower mean anti-spike antibody levels following vaccine doses 2 through 4. Patients with severe PAD had the highest rate of increase in anti-spike antibody levels over 5 immunizations. These data suggest a clinical benefit to sequential SARS-CoV-2 immunizations, particularly among high-risk patients with PAD.

Racial and ethnic disparities in early mortality among patients with inborn errors of immunity.

BACKGROUND: Racial and ethnic disparities in life expectancy in the United States have been widely documented. To date, there remains a paucity of similar data in patients with inborn errors of immunity (IEIs). OBJECTIVE: Our aim was to examine racial and ethnic differences in mortality due to an IEI in the United States. METHODS: We analyzed National Center for Health Statistics national mortality data from 2003 to 2018. We quantified age-adjusted death rate and age-specific death rate as a result of an IEI for each major racial and ethnic group in the United States and examined the association of race and ethnicity with death at a younger age. RESULTS: From 2003 to 2018, IEIs were reported as the underlying or contributing cause of death in 14,970 individuals nationwide. The age-adjusted death rate was highest among Black patients (4.25 per 1,000,000 person years), compared with 2.01, 1.71, 1.50, and 0.92 per 1,000,000 person years for White, American Indian/Alaska Native, Hispanic, and Asian/Pacific Islander patients, respectively. The odds of death before age 65 years were greatest among Black patients (odds ratio [OR] = 5.15 [95% CI = 4.61-5.76]), followed by American Indian/Alaska Native patients (OR = 3.58 [95% CI = 2.30-5.82]), compared with White patients. The odds of death before age 24 years were greater among Hispanic patients than among non-Hispanic patients (OR = 3.60 [95% CI = 3.08-4.18]). CONCLUSION: Our study highlights racial and ethnic disparities in mortality due to an IEI and the urgent need to further identify and systematically remove barriers in care for historically marginalized patients with IEIs.

Enrollment of Pediatric Patients in COVID-19 Interventional Trials.

This cross-sectional study uses information gathered from ClinicalTrials.gov to assess the inclusion of children in COVID-19 interventional trials conducted in the US.

Multiomics analysis identifies BIRC3 as a novel glucocorticoid response-associated gene.

BACKGROUND: Inhaled corticosteroid (ICS) response among patients with asthma is influenced by genetics, but biologically actionable insights based on associations have not been found. Various glucocorticoid response omics data sets are available to interrogate their biological effects. OBJECTIVE: We sought to identify functionally relevant ICS-response genetic associations by integrating complementary multiomics data sets. METHODS: Variants with P values less than 10-4 from a previous ICS-response genome-wide association study were reranked on the basis of integrative scores determined from (1) glucocorticoid receptor- and (2) RNA polymerase II-binding regions inferred from ChIP-Seq data for 3 airway cell types, (3) glucocorticoid response element motifs, (4) differentially expressed genes in response to glucocorticoid exposure according to 20 transcriptomic data sets, and (5) expression quantitative trait loci from GTEx. Candidate variants were tested for association with ICS response and asthma in 6 independent studies. RESULTS: Four variants had significant (q value < 0.05) multiomics integrative scores. These variants were in a locus consisting of 52 variants in high linkage disequilibrium (r2 ≥ 0.8) near glucocorticoid receptor-binding sites by the gene BIRC3. Variants were also BIRC3 expression quantitative trait loci in lung, and 2 were within/near putative glucocorticoid response element motifs. BIRC3 had increased RNA polymerase II occupancy and gene expression, with glucocorticoid exposure in 2 ChIP-Seq and 13 transcriptomic data sets. Some BIRC3 variants in the 52-variant locus were associated (P < .05) with ICS response in 3 independent studies and others with asthma in 1 study. CONCLUSIONS: BIRC3 should be prioritized for further functional studies of ICS response.

New-onset hypogammaglobulinaemia and infectious complications associated with rituximab use in childhood-onset rheumatic diseases.

OBJECTIVE: To investigate the incidence and risk factors for hypogammaglobulinaemia and infectious complications associated with rituximab treatment in childhood-onset rheumatic diseases. METHODS: We performed a single-centre retrospective study of patients (n = 85) treated at Boston Children's Hospital (BCH) from 2009 to 2019. Study subjects included patients (ages 6-24 years) who received rituximab for the treatment of a childhood-onset rheumatic disease. RESULTS: New-onset hypogammaglobulinaemia developed in 23 (27.1%) patients within 18 months of rituximab induction treatment. Twenty-two patients (25.9%) developed at least one infectious complication in the 18 months following the first rituximab infusion; of these, 11 (50%) had serious infections requiring inpatient treatment. After adjusting for potential confounders, exposure to pulse corticosteroid therapy in the month prior to rituximab use was a significant predictor of both new-onset hypogammaglobulinaemia (odds ratio [OR] 3.94; 95% CI: 1.07, 16.0; P = 0.044) and infectious complications (OR 15.3; 95% CI: 3.04, 126.8; P = 0.003). Post-rituximab hypogammaglobulinaemia was the strongest predictor of serious infectious complications (OR 7.89; 95% CI: 1.41, 65.6; P = 0.028). Younger age at rituximab use was also a significant predictor of new-onset hypogammaglobulinaemia (OR 0.83; 95% CI: 0.70, 0.97; P = 0.021). Compared with other rheumatic diseases, patients with vasculitis had a higher likelihood of developing infectious complications, including serious infections. CONCLUSION: Although rituximab was well tolerated in terms of infectious complications in the majority of patients with childhood-onset rheumatic diseases, a substantial proportion developed new-onset hypogammaglobulinaemia and infectious complications following treatment. Our study highlights a role for heightened vigilance of rituximab-associated hypogammaglobulinaemia and infections in paediatric patients with rheumatic conditions.

Improved Disease Course Associated With Early Initiation of Biologics in Polyarticular Juvenile Idiopathic Arthritis: Trajectory Analysis of a Childhood Arthritis and Rheumatology Research Alliance Consensus Treatment Plans Study.

OBJECTIVE: To investigate the effects of early introduction of biologic disease-modifying antirheumatic drugs (bDMARDs) on the disease course in untreated polyarticular juvenile idiopathic arthritis (JIA). METHODS: We analyzed data on patients with polyarticular JIA participating in the Start Time Optimization of Biologics in Polyarticular JIA (STOP-JIA) study (n = 400) and a comparator cohort (n = 248) from the Childhood Arthritis and Rheumatology Research Alliance Registry. Latent class trajectory modeling (LCTM) was applied to identify subgroups of patients with distinct disease courses based on disease activity (clinical Juvenile Arthritis Disease Activity Score in 10 joints) over 12 months from baseline. RESULTS: In the STOP-JIA study, 198 subjects (49.5%) received bDMARDs within 3 months of baseline assessment. LCTM analyses generated 3 latent classes representing 3 distinct disease trajectories, characterized by slow, moderate, or rapid disease activity improvement over time. Subjects in the rapid improvement trajectory attained inactive disease within 6 months from baseline. Odds of being in the rapid improvement trajectory versus the slow improvement trajectory were 3.6 times as high (95% confidence interval 1.32-10.0; P = 0.013) for those treated with bDMARDs ≤3 months from baseline compared with subjects who started bDMARDs >3 months after baseline, after adjusting for demographic characteristics, clinical attributes, and baseline disease activity. Shorter disease duration at first rheumatology visit approached statistical significance as a predictor of favorable trajectory without bDMARD treatment. CONCLUSION: Starting bDMARDs within 3 months of baseline assessment is associated with more rapid achievement of inactive disease in subjects with untreated polyarticular JIA. These results demonstrate the utility of trajectory analysis of disease course as a method for determining treatment efficacy.

Rituximab-induced hypogammaglobulinemia and infection risk in pediatric patients.

BACKGROUND: Rituximab is a B-cell depleting agent used in B-cell malignancies and autoimmune diseases. A subset of adult patients may develop prolonged and symptomatic hypogammaglobulinemia following rituximab treatment. However, this phenomenon has not been well delineated in the pediatric population. OBJECTIVES: This study sought to determine the prevalence, risk factors, and clinical significance of hypogammaglobulinemia following rituximab therapy in children. METHODS: This was a multicenter, retrospective cohort study that extracted clinical and immunological data from pediatric patients who received rituximab. RESULTS: The cohort comprised 207 patients (median age, 12.0 years). Compared to baseline values, there was a significant increase in hypogammaglobulinemia post-rituximab therapy, with an increase in prevalence of hypo-IgG (28.7%-42.6%; P = .009), hypo-IgA (11.1%-20.4%; P = .02), and hypo-IgM (20.0%-62.0%; P < .0001). Additionally, low IgG levels at any time post-rituximab therapy were associated with a higher risk of serious infections (34.4% vs 18.9%; odds ratio, 2.3; 95% CI, 1.1-4.8; P = .03). Persistent IgG hypogammaglobulinemia was observed in 27 of 101 evaluable patients (26.7%). Significant risk factors for persistent IgG hypogammaglobulinemia included low IgG and IgA levels pre-rituximab therapy. Nine patients (4.3%) within the study were subsequently diagnosed with a primary immunodeficiency, 7 of which received rituximab for autoimmune cytopenias. CONCLUSIONS: Hypogammaglobulinemia post-rituximab treatment is frequently diagnosed within the pediatric population. Low IgG levels are associated with a significant increase in serious infections, and underlying primary immunodeficiencies are relatively common in children receiving rituximab, thus highlighting the importance of immunologic monitoring both before and after rituximab therapy.

Risk factors for suicide attempt in children, adolescents, and young adults hospitalized for mental health disorders.

BACKGROUND: Suicide is a leading cause of death among children, adolescents, and young adults (AYA), and mental health disorders are a major contributing factor. Yet, suicidal behaviors among children and AYA with mental health concerns remain understudied and age-specific risk factors are poorly understood. We examined the risk factors for suicide attempt in children and AYA with mental health disorders across three age groups: pre-adolescent children (aged ≤ 12), adolescents (aged 13-17), and young adults (aged 18-25). METHODS: A cross-sectional study of children and AYA hospitalized for a mental health disorder (n = 18,018) at a private hospital system with 141 facilities across the United States (year 2014). RESULTS: Suicide attempts six months prior to hospitalization were reported in 12.1% (n = 177) pre-adolescent children, 22% (n = 1476) adolescents, and 17.9% (n = 1766) young adults. Evidence of psychological trauma was present in 55.4% of pre-adolescent children, 51.2% of adolescents, and 44.5% of young adults. Predictors for suicide attempt observed across all three age groups included the following: female sex, depressive disorder, and being a victim of bullying. Risk factors for suicide attempt specific to pre-adolescent children included being uninsured and having an unsafe home or school environment. Among AYA, suicide attempt was associated with non-Hispanic white, family history of suicide, emotional traumas, and other traumatic experiences. Alcohol use disorder was also a significant predictor of suicide attempt in young adults. CONCLUSIONS: Suicide attempts among children and AYA admitted to a hospital with mental health concerns are highly prevalent. Socioeconomic stressors appeared to be an important contributing factor of suicidal behavior in pre-adolescent children but not in older AYA. Effective suicide prevention strategies targeting children and AYA would need to consider age-specific risk factors.

Socioeconomic status and survival in patients with pulmonary hypertension.

In a publicly funded healthcare system, no evidence of survival disparities across socioeconomic classes among patients with pulmonary hypertension was observed, underscoring the importance of eliminating financial barriers to medical care and treatment https://bit.ly/2Eb1ju2.

Claims-Based Algorithms for Identifying Patients With Pulmonary Hypertension: A Comparison of Decision Rules and Machine-Learning Approaches.

Background Real-world healthcare data are an important resource for epidemiologic research. However, accurate identification of patient cohorts-a crucial first step underpinning the validity of research results-remains a challenge. We developed and evaluated claims-based case ascertainment algorithms for pulmonary hypertension (PH), comparing conventional decision rules with state-of-the-art machine-learning approaches. Methods and Results We analyzed an electronic health record-Medicare linked database from two large academic tertiary care hospitals (years 2007-2013). Electronic health record charts were reviewed to form a gold standard cohort of patients with (n=386) and without PH (n=164). Using health encounter data captured in Medicare claims (including patients' demographics, diagnoses, medications, and procedures), we developed and compared 2 approaches for identifying patients with PH: decision rules and machine-learning algorithms using penalized lasso regression, random forest, and gradient boosting machine. The most optimal rule-based algorithm-having ≥3 PH-related healthcare encounters and having undergone right heart catheterization-attained an area under the receiver operating characteristic curve of 0.64 (sensitivity, 0.75; specificity, 0.48). All 3 machine-learning algorithms outperformed the most optimal rule-based algorithm (P<0.001). A model derived from the random forest algorithm achieved an area under the receiver operating characteristic curve of 0.88 (sensitivity, 0.87; specificity, 0.70), and gradient boosting machine achieved comparable results (area under the receiver operating characteristic curve, 0.85; sensitivity, 0.87; specificity, 0.70). Penalized lasso regression achieved an area under the receiver operating characteristic curve of 0.73 (sensitivity, 0.70; specificity, 0.68). Conclusions Research-grade case identification algorithms for PH can be derived and rigorously validated using machine-learning algorithms. Simple decision rules commonly applied in published literature performed poorly; more complex rule-based algorithms may potentially address the limitation of this approach. PH research using claims data would be considerably strengthened through the use of validated algorithms for cohort ascertainment.

Geographical Distribution, Incidence, Malignancies, and Outcome of 136 Eastern Slavic Patients With Nijmegen Breakage Syndrome and NBN Founder Variant c.657_661del5.

Nijmegen breakage syndrome (NBS) is a DNA repair disorder characterized by combined immunodeficiency and a high predisposition to lymphoid malignancies. The majority of NBS patients are identified with a homozygous five base pair deletion in the Nibrin (NBN) gene (c.657_661del5, p.K219fsX19) with a founder effect observed in Caucasian European populations, especially of Slavic origin. We present here an analysis of a cohort of 136 NBS patients of Eastern Slav origin across Belarus, Ukraine, Russia, and Latvia with a focus on understanding the geographic distribution, incidence of malignancy, and treatment outcomes of this cohort. Our analysis shows that Belarus had the highest prevalence of NBS (2.3 per 1,000,000), followed by Ukraine (1.3 per 1,000,000), and Russia (0.7 per 1,000,000). Of note, the highest concentration of NBS cases was observed in the western regions of Belarus and Ukraine, where NBS prevalence exceeds 20 cases per 1,000,000 people, suggesting the presence of an "Eastern Slavic NBS hot spot." The median age at diagnosis of this cohort ranged from 4 to 5 years, and delay in diagnosis was more pervasive in smaller cities and rural regions. A total of 62 (45%) patients developed malignancies, more commonly in males than females (55.2 vs. 34.2%; p=0.017). In 27 patients, NBS was diagnosed following the onset of malignancies (mean age: 8 years). Malignancies were mostly of lymphoid origin and predominantly non-Hodgkin lymphoma (NHL) (n=42, 68%); 38% of patients had diffuse large B-cell lymphoma. The 20-year overall survival rate of patients with malignancy was 24%. However, females with cancer experienced poorer event-free survival rates than males (16.6% vs. 46.8%, p=0.036). Of 136 NBS patients, 13 underwent hematopoietic stem cell transplantation (HSCT) with an overall survival of 3.5 years following treatment (range: 1 to 14 years). Indications for HSCT included malignancy (n=7) and immunodeficiency (n=6). Overall, 9% of patients in this cohort reached adulthood. Adult survivors reported diminished quality of life with significant physical and cognitive impairments. Our study highlights the need to improve timely diagnosis and clinical management of NBS among Eastern Slavs. Genetic counseling and screening should be offered to individuals with a family history of NBS, especially in hot spot regions.

Racial and Ethnic Differences in Pediatric Pulmonary Hypertension: An Analysis of the Pediatric Pulmonary Hypertension Network Registry.

OBJECTIVE: To investigate racial and ethnic differences in pulmonary hypertension subtypes and survival differences in a pediatric population. STUDY DESIGN: This was a retrospective analysis of a cohort of patients with pulmonary hypertension (aged ≤18 years) enrolled in the Pediatric Pulmonary Hypertension Network registry between 2014 and 2018, comprising patients at eight Pediatric Centers throughout North America (n = 1417). RESULTS: Among children diagnosed after the neonatal period, pulmonary arterial hypertension was more prevalent among Asians (OR, 1.83; 95% CI, 1.21-2.79; P = .0045), lung disease-associated pulmonary hypertension among blacks (OR, 2.09; 95% CI, 1.48-2.95; P < .0001), idiopathic pulmonary arterial hypertension among whites (OR, 1.58; 95% CI, 1.06-2.41; P = .0289), and pulmonary veno-occlusive disease among Hispanics (OR, 6.11; 95% CI, 1.34-31.3; P = .0184). Among neonates, persistent pulmonary hypertension of the newborn (OR, 4.07; 95% CI, 1.54-10.0; P = .0029) and bronchopulmonary dysplasia (OR, 8.11; 95% CI, 3.28-19.8; P < .0001) were more prevalent among blacks, and congenital diaphragmatic hernia was more prevalent among whites (OR, 2.29; 95% CI, 1.25-4.18; P = .0070). An increased mortality risk was observed among blacks (HR, 1.99; 95% CI, 1.03-3.84; P = .0396), driven primarily by the heightened mortality risk among those with lung disease-associated pulmonary hypertension (HR, 2.84; 95% CI, 1.15-7.04; P = .0241). CONCLUSIONS: We found significant racial variability in the prevalence of pulmonary hypertension subtypes and survival outcomes among children with pulmonary hypertension. Given the substantial burden of this disease, further studies to validate phenotypic differences and to understand the underlying causes of survival disparities between racial and ethnic groups are warranted.

A Bayesian Network Approach to Disease Subtype Discovery.

Human diseases are historically categorized into groups based on the specific organ or tissue affected. Over the past two decades, advances in high-throughput genomic and proteomic technologies have generated substantial evidence demonstrating that many diseases are in fact markedly heterogeneous, comprising multiple clinically and molecularly distinct subtypes that simply share an anatomical location. Here, a Bayesian network analysis is applied to study comorbidity patterns that define disease subtypes in pediatric pulmonary hypertension. The analysis relearned established subtypes, thus validating the approach, and identified rare subtypes that are difficult to discern through clinical observations, providing impetus for deeper investigation of the disease subtypes that will enrich current disease classifications. Further advances linking disease subtypes to therapeutic response, disease outcomes, as well as the molecular profiles of individual subtypes will provide impetus for the development of more effective and targeted therapies.

Delay in reviewing test results prolongs hospital length of stay: a retrospective cohort study.

BACKGROUND: Failure in the timely follow-up of test results has been widely documented, contributing to delayed medical care. Yet, the impact of delay in reviewing test results on hospital length of stay (LOS) has not been studied. We examine the relationship between laboratory tests review time and hospital LOS. METHODS: A retrospective cohort study of inpatients admitted to a metropolitan teaching hospital in Sydney, Australia, between 2011 and 2012 (n = 5804). Generalized linear models were developed to examine the relationship between hospital LOS and cumulative clinician read time (CRT), defined as the time taken by clinicians to review laboratory test results performed during an inpatient stay after they were reported in the computerized test reporting system. The models were adjusted for patients' age, sex, and disease severity (measured by the Charlson Comorbidity index), the number of test panels performed, the number of unreviewed tests pre-discharge, and the cumulative laboratory turnaround time (LTAT) of tests performed during an inpatient stay. RESULTS: Cumulative CRT is significantly associated with prolonged LOS, with each day of delay in reviewing test results increasing the likelihood of prolonged LOS by 13.2% (p < 0.0001). Restricting the analysis to tests with abnormal results strengthened the relationship between cumulative CRT and prolonged LOS, with each day of delay in reviewing test results increasing the likelihood of delayed discharge by 33.6% (p < 0.0001). Increasing age, disease severity and total number of tests were also significantly associated with prolonged LOS. Increasing number of unreviewed tests was negatively associated with prolonged LOS. CONCLUSIONS: Reducing unnecessary hospital LOS has become a critical health policy goal as healthcare costs escalate. Preventing delay in reviewing test results represents an important opportunity to address potentially avoidable hospital stays and unnecessary resource utilization.