Applicability of a microRNA-based Dynamic Risk Score (DRS) for type 1 diabetes

Identifying biomarkers of functional β-cell loss is critical in risk stratification for Type 1 Diabetes (T1D). We report a microRNA-based dynamic (responsive to environment) risk score developed using multi-center, multi-ethnic/country (“multi-context”) cohorts. Discovery (wet-lab and dry-lab) analysis identified 50 microRNAs that were measured across n=2,204 individuals from four contexts (4C=AUS/Australia, DNK/Denmark, HKG/Hong Kong SAR China, IND/India). A 4-context, microRNA-based dynamic risk score (DRS4C) was generated, which effectively stratified individuals with/without T1D. Generative artificial intelligence (GAI) was used to create an enhanced (e)DRS4C, that offered high AUC (0.84) on an independent multi-context Validation-set (n=662) and most accurately predicted future exogenous-insulin requirement at one-hour of islet transplantation in Canada (CAN) recipients. In a clinical trial assessing an emerging T1D therapy, baseline microRNA signature, but not the clinical characteristics, stratified 1-year response to Imatinib. This study harnessed ML and GAI approaches, identifying and validating a microRNA-based DRS for T1D stratification and treatment efficacy prediction. This capsule presents the code for stratifying controls and T1D study participants as well as for predicting outcomes of T1D therapy.

鉴定功能性β细胞丢失的生物标志物，对于1型糖尿病（Type 1 Diabetes, T1D）的风险分层至关重要。本研究报道了一款基于微RNA（microRNA）的动态（可响应环境变化）风险评分，该评分依托多中心、多民族/多国家（即"多情境"）队列开发而成。通过湿实验与干实验相结合的发现性分析，我们在来自4个情境（4C=澳大利亚AUS、丹麦DNK、中国香港特别行政区HKG、印度IND）的2204名受试者中完成了50种微RNA的检测。我们构建了基于4个情境的微RNA动态风险评分（DRS4C），该评分可有效区分伴/不伴1型糖尿病的受试者。随后借助生成式人工智能（Generative Artificial Intelligence, GAI）构建了增强型（e）DRS4C，该模型在独立多情境验证集（n=662）中曲线下面积（AUC）达0.84，并可精准预测加拿大（CAN）胰岛移植受者术后1小时的未来外源胰岛素需求量。在一项评估新型1型糖尿病治疗方案的临床试验中，基线微RNA表达特征（而非临床特征）可有效区分受试者对伊马替尼的1年治疗应答情况。本研究借助机器学习（Machine Learning, ML）与生成式人工智能技术，鉴定并验证了一款基于微RNA的动态风险评分，可用于1型糖尿病的风险分层与治疗疗效预测。本代码包提供了用于区分对照组与1型糖尿病研究受试者的代码，以及预测1型糖尿病治疗结局的代码。