124P - Incidence of activating frameshift and nonsense mutations in clinically actionable oncogenes

Background

Comprehensive molecular diagnostics plays a key role in precision medicine and is commonly used to screen for activating hotspot mutations, amplifications and gene fusions of oncogenes for guiding targeted therapies. However, several studies have shown that specific frameshift and nonsense mutations in such oncogenes (e.g. FGFR2, PIK3CA) can also result in oncogenic activation, but are rarely searched for or classified as oncogenic in routine diagnostics. As such noncanonical mutations in known targetable oncogenes are potential biomarkers for targeted therapies in, for example, clinical trials, we investigated their pan-cancer incidence.

Methods

The Hartwig Medical Foundation database containing whole genome sequencing (WGS) data of ˜8000 patients with metastatic disease was used to identify noncanonical genomic alterations in oncogenes. Clinically actionable oncogenes were examined for structural variants, nonsense and frameshift mutations that potentially cause aberration of the C-terminus of the protein.

Results

We identified frameshift and nonsense mutations in the C-terminus of PIK3CA, FGFR2, EGFR and ERBB2 (or HER2) in 30 patients. Interestingly, EGFR and ERBB2 are receptor tyrosine kinases of the ErbB family and are reported with complex genomic rearrangements at the C-terminus that are clinically actionable. In 7992 patients, 26 (0.33%) ERBB2 nonsense or frameshift mutations were identified, of which 19 (73%) are potentially truncating the C-terminus similarly to reported ERBB2 gene fusions. Like these fusions, the noncanonical mutations in ERBB2 were predominantly found in gastrointestinal, lung and breast cancers.

Conclusions

This study identified noncanonical truncating mutations in several clinically actionable oncogenes, which are typically not covered or not classified in routine next-generation sequencing (NGS) panels. Despite the relative low incidence, identification of these biomarkers is warranted to consider all possible treatment options and to further investigate the clinical benefit.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Hartwig Medical Foundation.

Disclosure

All authors have declared no conflicts of interest.