Genomic Profiling of HER2-Positive Gastric Cancer: PI3K/Akt/mTOR Pathway as Predictor of Outcomes in HER2-Positive Advanced Gastric Cancer Treated with Trastuzumab

Abstract

Background: HER2-positive gastric cancer (GC) affects 7%-34% of patients with GC. Trastuzumab-based first-line treatment has become the standard of care for HER2-positive advanced gastric cancer (AGC). However, there are no clinically validated biomarkers for resistance to HER2-targeted therapies. Upregulation of PI3K pathway and tyrosine kinase receptor (TKR) alterations have been noted as molecular mechanisms of resistance in breast cancer. Our study aimed to perform a molecular characterization of HER2-positive AGC and investigate the role of PI3K/Akt/mTOR signaling pathway activation and TKR gene copy number (GCN) gains as predictive biomarkers in HER2-positive AGC treated with trastuzumab.

Patients and methods: Forty-two HER2-positive GC samples from patients treated with trastuzumab-based first-line chemotherapy were selected. DNA samples were sequenced. PTEN and MET immunohistochemistry were also performed.

Results: Concurrent genetic alterations were detected in 97.1% of HER2-positive AGC. We found activation of PI3K/Akt/mTOR pathway in 52.4% of patients and TKR GCN gains in 38.1%. TKR GCN gains did not correlate with overall survival (OS) or progression-free survival (PFS). Multivariate Cox models showed that PI3K/Akt/mTOR activation negatively affects the effectiveness of trastuzumab-based chemotherapy in terms of OS and PFS.

Conclusion: Our results provide for the first time a detailed molecular profile of concurrent genetic alterations in HER2-positive AGC. PI3K pathway activation could be used as a predictive marker of worse outcome in this patient population. In addition, gains in copy number of other TKR genes in this subgroup may also influence the survival benefit obtained with trastuzumab.

Implications for practice: This article reports, for the first time, a detailed molecular profile of genomic alterations in patients with HER2-positive advanced gastric cancer (AGC). PI3K/Akt/mTOR signaling pathway activation seems to have a differentially negative effect on overall survival and progression-free survival in AGC treated with trastuzumab-based chemotherapy. Combining different targeted agents could be a successful therapeutic strategy to improve the prognosis of HER2-positive AGC.

Keywords: Gastric cancer; HER2; Next‐generation sequencing; PI3K; Trastuzumab; Tyrosine kinase receptors.

Figure 1.

Study design flow chart. *, All tumor samples were centrally retested for HER2 status using IHC and dc‐SISH, regardless of the local laboratory results. HER2 FISH was performed when dc‐SISH failed or showed a HER2‐CEP17 ratio <4 to confirm that we were only including patients with bona fide amplification. We defined the HER2‐positive study population candidate for trastuzumab treatment (according to U.S. Food and Drug Administration criteria) as meeting any of the following criteria: IHC 3+; IHC 2+ and dc‐SISH HER2‐CEP17 ratio ≥4; IHC 2+ and dc‐SISH HER2‐CEP17 ratio ≥2, confirmed by FISH (HER2‐CEP17 ratio ≥2). **, IHC of MET and PTEN. Abbreviations: AGC, advanced gastric cancer; dc‐SISH, dual‐color silver‐enhanced in situ hybridization; FISH, fluorescence in situ hybridization; IHC, immunohistochemistry; NGS, next‐generation sequencing.

Figure 2.

Tile plot of concurrent genetic alterations in HER2‐positive gastric cancer. Alterations considered included mutation (blue), deletion (green), and amplification (red). DNA samples were sequenced using the Ion Torrent Personal Genome Machine (PGM) system (ThermoFisher Scientific, Waltham, MA). Data from the PGM runs were processed using Ion Torrent platform‐specific pipeline software Torrent Suite, version 2 (ThermoFisher Scientific). For a sequence variant to be considered relevant, sequencing coverage of 250× and an allele frequency of at least 5% were used as minimum requirements in this study. Sequencing data were analyzed with Ion Reporter software version 4.2 to detect any copy number alteration of the genes included in the panel used. For this purpose, 10 normal gastric samples were selected.

Figure 3.

PI3K/Akt/mTOR signaling pathway alterations and tyrosine kinase receptor (TKR) gene copy number gains. (A): Frequency of activating alterations in the PI3K/Akt/mTOR signaling cascade, indicated by the presence of PI3KCA mutation, PI3KCA copy number gain, and/or loss of PTEN expression. (B): Frequency of TKR gene (EGFR, FGFR, MET) copy number gains. Copy number variations and mutations were determined using the Ion Torrent Personal Genome Machine system (ThermoFisher Scientific, Waltham, MA) and Ion AmpliSeq Cancer Hotspot Panel, version 2 (ThermoFisher Scientific). PTEN protein status was determined by immunohistochemistry using the PTEN (D4.3) XP rabbit antibody.

Figure 4.

Kaplan‐Meier curves for overall survival (OS) and progression‐free survival (PFS) according to PI3K/Akt/mTOR pathway activation and TKR gene copy number. (A): OS according to PI3K/Akt/mTOR pathway status. (B): PFS according to PI3K/Akt/mTOR pathway status. (C): OS according to TKR gene copy number gains. (D): PFS according to TKR gene copy number gains. PI3K/Akt/mTOR pathway activation was defined as PIK3CA mutation, PIK3CA gene copy number gain, and/or loss of expression of PTEN. Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio; TKR, tyrosine kinase receptor.