Enhanced Detection of EGFRvIII in Tumors: A Comparative Study of Split Read- and Read Depth-Based DNA Sequencing Approaches.

Epidermal growth factor receptor variant (EGFRv)-III, a common oncogenic variant in glioblastoma and other solid tumors, results from an in-frame deletion of exons 2 to 7 from the EGFR gene. Detection of EGFRvIII is crucial for understanding tumor biology, guiding targeted therapies, and developing personalized treatment strategies. In this study, two detection approaches based on DNA next-generation sequencing-read depth (RD)-based and split read (SR)-based detection-were compared to evaluate their sensitivity and accuracy in identifying EGFRvIII. Thirty-one tumor samples, including glioblastoma and pancreatic adenocarcinoma, were analyzed using both methods. The SR-based method detected EGFRvIII in 20 of 31 samples, while the RD-based method identified it in only 12 samples (P < 0.001), demonstrating that the SR-based method had significantly higher sensitivity. RNA next-generation sequencing confirmed EGFRvIII expression in most SR-positive cases. Additionally, the SR-based method identified multiple breakpoints in several tumors, revealing intratumor heterogeneity and the subclonal origins of EGFRvIII. The RD-based method was prone to false negatives, particularly in cases with high EGFR amplification or low tumor cell percentage, where copy number variations could be masked by background noise. The findings highlight the sensitivity and accuracy of SR-based detection in identifying EGFRvIII and capturing intratumor heterogeneity. SR-based analysis is recommended as the method for EGFRvIII detection in both clinical and research settings.