Evaluation of hybrid capture-based targeted and metagenomic next-generation sequencing for pathogenic microorganism detection in infectious keratitis

Abstract

Objective: This study aimed to compare the performance of hybrid capture-based targeted next-generation sequencing (hc-tNGS) and metagenomic next-generation sequencing (mNGS) in detecting the causative pathogens of infectious keratitis. Methods: A total of 60 patients with clinically diagnosed infectious keratitis were enrolled between January and December 2024. Corneal scraping samples were analyzed using hc-tNGS and mNGS. Detection rates, pathogen spectra, normalized reads, turnaround time (TAT), and costs were compared between the two techniques. Results: hc-tNGS exhibited a significantly higher overall detection rate than mNGS (86.7% versus 73.3%, P < 0.001). In particular, hc-tNGS detected 29 pathogens (13 bacteria, 9 viruses, and 7 fungi), whereas mNGS detected 22 pathogens (9 bacteria, 7 viruses, and 6 fungi). Furthermore, hc-tNGS detected additional low-abundance pathogens in 17 mNGS-positive patients (28.3%, 17/60) and 8 mNGS-negative patients (11.3%, 8/60). The normalized reads for viruses, bacteria, and fungi in hc-tNGS were 57.2-, 2.7-, and 3.3-fold higher than those in mNGS, respectively (P < 0.001, P = 0.003, and P = 0.028). Moreover, hc-tNGS reduced TAT by 11.3% (18.0 versus 20.3 h) and costs by 22.4–48.8%. The median of sequencing data size of mNGS was 29.8 million (29.8 M) reads, which was significantly higher than that of tNGS (1.5 M, P < 0.001). Conclusion: hc-tNGS demonstrates superior performance and cost-effectiveness in detecting potential pathogens of infectious keratitis, especially low-abundance pathogens, whereas mNGS remains valuable for detecting novel pathogens. Owing to its enhanced performance, faster TAT, and reduced costs, hc-tNGS is a promising clinical tool for pathogen detection.