Abstract
Approximately 17% of breast cancer diagnoses are classified as positive for the human
epidermal growth factor receptor 2 (HER2+) due to gene amplification or activating
mutations in the ERBB2 oncogene. Prognosis for patients with the HER2+ subtype was poor
until the addition of targeted anti-HER2 therapies, such as trastuzumab, pertuzumab and
lapatinib into clinical treatment. However, a significant proportion of patients develop
resistance to anti-HER2 therapies by mechanisms not well understood.
The Cunliffe laboratory previously shown 83% of HER2+ breast cancers aberrantly express
Fibroblast Growth Factor-inducible 14 (Fn14). Fn14, a cell surface receptor and member of
the tumour necrosis factor receptor superfamily, is not expressed in normal mammary
epithelium, thus a potential biomarker to exploit in HER2+ disease. Fn14 has been reported
to promote invasion and migration across various cancer types however, its mechanistic role
in HER2+ breast cancer remains unexplored. We aimed to investigate this association in the
HER2+ AU565 breast cancer cell line.
We successfully developed a reproducible transient Fn14-overexpression model in AU565
cells, confirming an 8-fold average increase in Fn14 protein relative to empty vector control
(p=0.0065) by western blotting. Using this model, we then showed an Fn14-mediated 5% and
25% reduction in HER2 and EGFR protein expression, respectively. Further investigation
into the activation of the downstream targets of HER2 and EGFR, including the MAPK
(ERK1/2) and PI3K pathways, showed reductions of 64% and 27%, respectively; suggesting
that, when transiently overexpressed, Fn14 engages in antagonistic crosstalk with HER2 and
EGFR.
An in silico analysis of lapatinib sensitivity data (https://rocplot.org/) indicated a trend
between high Fn14 expression and increased sensitivity to lapatinib (p = 0.19). Using this
information and the knowledge that lapatinib inhibits both HER2 and EGFR signalling, we
next investigated whether our observed Fn14-mediated antagonism alters lapatinib sensitivity
in our AU565 model. A 72-hour drug sensitivity assay demonstrated fewer viable cells at the
1.0X IC50 concentration of lapatinib compared to the vehicle control (IC50 9.2nM for HER2,
p=0.052). This finding is consistent with our observed Fn14-mediated reduction in HER2 and
EGFR signalling.
Lastly, in vitro Boyden chamber invasion and migration assays show transient Fn14
overexpression resulted in a 9-fold increase in invasion and a 1.2-fold increase in migration,
consistent with the pro-malignant properties of Fn14.
Our results show Fn14 can functionally engage in crosstalk with HER2 and EGFR signalling,
offering clinically relevant insights into lapatinib sensitivity in HER2+ patients, warranting
further mechanistic investigation in stable models.