Authors
Jeanne Tie, Sherif Mohamed El-Refai, Takayuki Yoshino, Salvatore Siena, Sara Lonardi, Andrea Sartore-Bianchi, Yoshiaki Nakamura, Hideaki Bando, Takao Fujisawa, Sheau Wen Lok, Hui-Li Wong, Kunal C. Kadakia, Elizabeth Mauer, Mohamed E. Salem
Background:
ERBB2 is a rapidly emerging therapeutic target for a subset of colorectal cancer (CRC) harboring oncogenic alterations in this gene. Oncogenic ERBB3 mutations have been reported in various cancers including CRC, but little is known about its functional impact. Optimal targeting of this pathway requires understanding of the genomic context in which somatic ERBB2/3 alterations (ERBB2/3-alt) occur in a real-world CRC population.
Methods:
We analyzed 7,688 de-identified records from CRC patients that underwent next generation sequencing with the Tempus|xT assay (DNA-seq of 648 genes at 500x coverage, full transcriptome RNA-seq). We assessed the prevalence and association of ERBB2/3-alt with demographics, co-occurring alterations, immuno-oncology biomarkers (microsatellite instability [MSI], tumor mutational burden [TMB], PD-L1 expression), and consensus molecular subtype (CMS, available for the subgroup with primary biopsies and RNA data [n = 2,686]).
Results:
Overall, 5% (376/7688) of tumors harbored an ERBB2 or ERBB3-alt. 1.9% (n = 143) ERBB2-amplified, 1.3% (n = 97) ERBB2-mutated, 0.9% (n = 72) ERBB3-mutated, < 1% other combinations (excluded from analyses). There were no significant differences in baseline demographics (e.g., age of onset, race and gender) between groups. Patients with ERBB2/3-alt were more likely to be MSI-high and TMB-high (Table). There was a trend towards higher prevalence of positive PD-L1 in ERBB3-alt vs ERBB3-WT tumors. We observed significant differences in co-occurring alterations among ERBB2/3-alt and WT groups (Table). CMS classification did not identify significant differences by ERBB2-alt or ERBB3-alt; ERBB2-alt compared to WT (CMS1: 8.6% vs 13%; CMS2: 24% vs 26%; CMS3: 20% vs 17%; CMS4: 30% vs 34%; p = 0.12) and ERBB3-alt compared to WT (CMS1: 24% vs 13%; CMS2: 12% vs 26%; CMS3: 16% vs 17%; CMS4: 40% vs 34%; p = 0.3).
Conclusions:
ERBB2/3 mutated CRC are more frequently MSI-H, TMB-high and KRAS mutated than ERBB2/3-WT tumors. Correlation of ERBB2/3 alterations with other genomic alterations including BRAF, TP53, CDK12, PIK3CA, and TOP2A will help advance the clinical development of HER2-targeted therapies.
VIEW THE POSTER
VIEW THE PUBLICATION