Integrative analytical and statistical framework for optimization of multiplex qPCR detection of TGFBI mutations in refractive surgery candidates.

内容摘要

Refractive surgery can unmask or accelerate transforming growth factor-β-induced (TGFBI)-related corneal dystrophies that are undetectable by routine slit-lamp examination, creating a clear need for a rapid, standardized, preoperative genetic screening. We developed a multiplex, allele-specific real-time quantitative polymerase chain reaction (qPCR) panel targeting five high-frequency TGFBI hotspots (R124C/L/H, R555W/Q) and built a statistics-driven analytical framework to optimize assay decisions. Receiver operating characteristic (ROC) analysis defined locus-specific cycle threshold (CT) cut-offs that were harmonized to a single decision threshold (CT=36) to simplify deployment. Analytical sensitivity was established by Probit modeling of serial two-fold dilutions, and confirmed by ≥20 replicates per level. In a 158-sample validation set (38 mutation-positive; 120 negative), qPCR agreed perfectly with Sanger sequencing (Cohen's kappa coefficient (κ)=1.0). Probit analysis yielded locus-specific limit of detection (LoD) values ranging from 0.035 to 0.200 ng/µL; at 0.200 ng/µL, the detection rate was over 95%. Repeatability and intermediate precision were high (CT coefficient of variation (CV) 0.34%-1.21%). No cross-reactivity was observed against non-target TGFBI variants or other ophthalmic genes, and interference from blood, oral flora/rinse, or toothpaste produced small, bounded shifts (approximately -7.8% to +2.8%). Calibration with serial dilutions demonstrated linear CT-log(copy) relationships suitable for routine quality control. Prospective screening of 10 055 refractive surgery candidates identified six TGFBI carriers (0.06%) harboring R124H (including one homozygote), R124L, R124C, or R555W mutation, all confirmed by Sanger sequencing. This study established a clinically applicable, statistically optimized multiplex qPCR platform that integrated ROC-derived cut-offs and Probit-defined LoD with rigorous evaluations of precision, specificity, and robustness, enabling large-scale population implementation. Positive screening results guide clinical decision-making through a standardized post-screening workflow, and the targeted hotspot screening strategy serves as a cost-effective first-tier high-throughput approach for preoperative risk assessment. The framework provides a transparent, reproducible path to standardize preoperative TGFBI screening and reduce iatrogenic risk in refractive surgery candidates. TGFBI基因突变是导致角膜营养不良的重要病因，也是屈光手术术前必须排查的遗传风险因素。然而，目前仍缺乏稳定、灵敏且可标准化的多重检测体系。本研究构建了一套整合分析与统计框架，用于优化针对TGFBI高频突变位点的多重等位基因特异性定量聚合酶链反应（qPCR）检测方法，并系统评估其临床应用价值。通过条件优化、阈值确定、重复性与特异性验证，建立了可同时检测R124C/L/H、R555W/Q等关键突变的多重qPCR体系。统计分析结果表明，该方法扩增效率良好、重复性高且检出限低，在梯度稀释样本中表现出稳定的定量能力，且无明显非特异性扩增。受试者工作特征曲线（ROC）与概率单位（Probit）分析进一步验证了体系的准确性与可靠性，能够有效检出微量模板中突变序列。将该体系应用于大规模屈光手术候选者术前筛查，成功检出了低频TGFBI突变携带者，且结果与Sanger测序完全一致。上述结果证实，基于整合分析与统计优化的多重qPCR体系，能够高效、可靠地用于TGFBI突变筛查，为屈光手术提供标准化的术前遗传风险评估工具，也为其他遗传病位点的多重检测优化提供了可借鉴的统计与方法学框架。.