Abstract 1360P
Background
Next generation sequencing has identified some rare EGFR fusions in lung cancer, such as EGFR-RAD51 and EGFR-SEPT14, which are known to respond to EGFR-TKIs. However, prevalence of EGFR fusions in a large population is unknown yet. Herein, we analyzed a large-scale dataset comprehensively to explore the mutation characteristics of EGFR fusions. This study offered an estimate of the proportion of patients with EGFR fusion who could benefit from therapies.
Methods
We retrospectively evaluated the frequency of EGFR fusions in total of 15,576 Chinese lung cancer patients whose tumor specimen and/or liquid biopsy samples underwent genomic profiling between 2017 and 2019 by a targeted sequencing consisting of all exons and flanking intronic regions of EGFR. The co-mutations were analyzed simultaneously.
Results
We found two pattern of EGFR fusions, EGFR-partner (13 pts) and partner-EGFR (14 pts), in 0.17% (27/15576) tumors. Only 9 of the 27 EGFR fusions maintained a full kinase domain (fKD) of EGFR, that may benefit from EGFR-TKIs. For EGFR-partner fusion, previous identified EGFR-RAD51 and EGFR-LANCL2 with fKD of EGFR were found. The break points of both fusions were in intron 24 of EGFR. In the form of partner-EGFR fusion, seven novel activating EGFR fusions were discovered; the partner were ATAD2B, LETM1, SLC37A1, BRAF, APP, LANCL2 and TCF12; and the break points of EGFR were respectively in intron 1,2,4,5,10,15 and 16. The wide distributions of break points suggest that introns should be included as much as possible when designing DNA panels to detect EGFR fusions. Regardless of EGFR fusion forms, fusions with partial or no kinase domain were more likely to co-mutate with other driver mutations, such as EGFR and TP53 hotspots and EGFR amplifications.
Conclusions
Our data showed 0.17% lung cancer patients had EGFR fusions. 1/3 of fusions, including seven novel EGFR fusions, may maintain a functional EGFR and could potentially benefit from EGFR-TKIs. Moreover, we demonstrated a wide distribution of EGFR break points in these fusions, which offered strategies in designing DNA panels for detecting EGFR-fusions.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.