“Genomic and phenotypic characterization of Investigator Global Assessment (IGA) scale based endotypes in atopic dermatitis”

A. Supplemental Text: Methods
B. The attached supplement folder contains
1. Table 1. Baseline demographic and clinical characteristics
2. EPIONE Profile and Design
3. Table 2 Efficacy Measures at Week 8
4. S Figure 2. Tradipitant-treatment Improves Itch and Sleep in Mild Atopic Dermatitis. A. Tradipitant-treatment improved worst itch in mild AD. B. Improvement in itch was observed after one full day of tradipitant treatment C. Tradipitant-treatment improved sleep disturbance in mild AD. D. Improvement in sleep was seen after two full days of tradipitant treatment. E. A greater proportion of mild AD patients achieved success of four points or greater on WI-NRS and at least a 50% improvement on SCORAD. A-D. P-values are from MMRM analysis. E. P-values are from Fisher’s exact test. *p<0.05
5. S Figure 3. 3D scatter plot across baseline clinical measurements and EOS.
6. S Figure 4a. Differences in clinical feature space across IGA severity
7. S Figure 4b. 3D scatter and 2D scatter plot of baseline clinical features
8. S Figure 5. PCA analysis of WGS samples
9. S Figure 5b. Distribution of EOS across genotypes for rs200005614
10. S Figure 6. allelic frequencies of LOF FLG variants across IGA groups and in the control population
11. S Figure 7. Individual cytokine distributions across IGA groups
12. S Figure 8a. Decision tree-based classification all features and confusion matrix
13. S Figure 8b. Decision tree-based classification (lab features) and confusion matrix, class labels are severity endotypes
14. S Figure 8c. Decision tree-based classification and confusion matrix. Classification with class labels being status of FLG LOF mutations. This ultimately is presenting the same data from another perspective yet confirming the association of carrier FLG LOF status with increased AD severity.
15. S Figure 8d. Decision tree-based on 8c. FLG based class labels
16. S Figure 9. Correlations between EASI and IGA (top) and SCORAD and EASI (bottom)

A. 补充文本：研究方法
B. 附带的补充文件夹包含以下内容：
1. 表1：基线人口统计学与临床特征
2. EPIONE方案与研究设计
3. 表2：第8周疗效评估指标
4. 补充图2：特拉匹坦治疗改善轻度特应性皮炎（AD）的瘙痒与睡眠。具体细节如下：
A. 特拉匹坦治疗可改善轻度特应性皮炎患者的最严重瘙痒症状；
B. 仅接受1天完整疗程的特拉匹坦治疗后，即可观察到瘙痒症状改善；
C. 特拉匹坦治疗可缓解轻度特应性皮炎患者的睡眠障碍；
D. 仅接受2天完整疗程的特拉匹坦治疗后，即可观察到睡眠质量改善；
E. 更高比例的轻度特应性皮炎患者在最差瘙痒数字评分量表（WI-NRS）中达到4分及以上的改善，且在特应性皮炎面积和严重程度指数（SCORAD）中实现至少50%的病情缓解。
注：A-D项的P值来自混合效应重复测量（MMRM）分析；E项的P值来自Fisher精确检验，*p<0.05。
5. 补充图3：基线临床指标与嗜酸性粒细胞（EOS）的三维散点图
6. 补充图4a：不同研究者整体评分（IGA）严重程度分组间的临床特征空间差异
7. 补充图4b：基线临床特征的三维散点图与二维散点图
8. 补充图5：全基因组测序（WGS）样本的主成分分析（PCA）
9. 补充图5b：rs200005614位点不同基因型的嗜酸性粒细胞（EOS）分布情况
10. 补充图6：不同IGA严重程度分组及对照人群中功能丧失（LOF）型丝聚蛋白基因（FLG）变异的等位基因频率分布
11. 补充图7：不同IGA严重程度分组间各细胞因子的个体分布情况
12. 补充图8a：基于全特征集的决策树分类模型与混淆矩阵
13. 补充图8b：基于实验室特征的决策树分类模型与混淆矩阵，分类标签为病情严重程度内表型
14. 补充图8c：基于决策树的分类模型与混淆矩阵，分类标签为丝聚蛋白基因功能丧失（FLG LOF）突变携带状态。本部分从另一视角呈现了相同数据，同时验证了FLG LOF携带状态与特应性皮炎病情严重程度升高的关联关系。
15. 补充图8d：基于FLG分类标签的决策树模型（同8c）
16. 补充图9：湿疹面积和严重程度指数（EASI）与研究者整体评分（IGA）的相关性（上图），以及特应性皮炎面积和严重程度指数（SCORAD）与湿疹面积和严重程度指数（EASI）的相关性（下图）