Novel Machine Learning Identifies 5 Asthma Phenotypes Using Cluster Analysis of Real-World Data.

BACKGROUND: Asthma classification into different subphenotypes is important to guide personalized therapy and improve outcomes. OBJECTIVES: To further explore asthma heterogeneity through determination of multiple patient groups by using novel machine learning (ML) approaches and large-scale real-world data. METHODS: We used electronic health records of patients with asthma followed at the Cleveland Clinic between 2010 and 2021. We used k-prototype unsupervised ML to develop a clustering model where predictors were age, sex, race, body mass index, prebronchodilator and postbronchodilator spirometry measurements, and the usage of inhaled/systemic steroids. We applied elbow and silhouette plots to select the optimal number of clusters. These clusters were then evaluated through LightGBM's supervised ML approach on their cross-validated F1 score to support their distinctiveness. RESULTS: Data from 13,498 patients with asthma with available postbronchodilator spirometry measurements were extracted to identify 5 stable clusters. Cluster 1 included a young nonsevere asthma population with normal lung function and higher frequency of acute exacerbation (0.8 /patient-year). Cluster 2 had the highest body mass index (mean ± SD, 44.44 ± 7.83 kg/m2), and the highest proportion of females (77.5%) and Blacks (28.9%). Cluster 3 comprised patients with normal lung function. Cluster 4 included patients with lower percent of predicted FEV1 of 77.03 (12.79) and poor response to bronchodilators. Cluster 5 had the lowest percent of predicted FEV1 of 68.08 (15.02), the highest postbronchodilator reversibility, and the highest proportion of severe asthma (44.9%) and blood eosinophilia (>300 cells/µL) (34.8%). CONCLUSIONS: Using real-world data and unsupervised ML, we classified asthma into 5 clinically important subphenotypes where group-specific asthma treatment and management strategies can be designed and deployed.

Prevalence of adverse events varies with the different oral isotretinoin brands in acne treatment: a retrospective observational study.

Oral isotretinoin remains the most effective treatment for acne. The aim of this retrospective single-center cohort study was to estimate the prevalence of adverse events with the different oral isotretinoin brands used in acne treatment. The population consisted of all patients who consulted for acne between January 2015 and January 2020. The inclusion criterion was the initiation of treatment with oral isotretinoin. The exclusion criteria were the use of two or more brands during the same course of treatment and previous treatment with oral isotretinoin. Statistical analysis was carried out using Chi-square and Mann-Whitney tests. We analyzed 468 patients of whom 68.6% were female. The median age was 21 years. The median weight was 65 kg. The treatment was Roaccutane®, Curacné®, Acnotren®, Isosupra®, Contracné®, or Acnogen® in 44.2%, 28%, 14.5%, 10.5%, 1.7% and 0.4% of cases, respectively. Xerosis was the most frequently reported side effect regardless of the brand. The highest frequencies of hypercholesterolemia (25.6%) and eczema (13%) were noted with Roaccutane®; hypertriglyceridemia (16.8%), epistaxis (9.9%) and fatigue (3.1%) with Curacné®; excessive sweating (4.1%) and headache (4.1%) with Isosupra®; and abnormal liver function tests (11%) with Acnotren®. We found a significant correlation mainly between abnormal ASAT and Acnotren® (p = 0.009), hypercholesterolemia and Roaccutane® [OR = 1.652 (95% CI 1.056-2.585)], hypertriglyceridemia and higher body weight (p = 0.004). Factors related to the drug brand and to characteristics of acne patients could explain the variability in the prevalence of some adverse events.