Abstract
Triple Negative Breast Cancer (TNBC) accounts for 10-20% of all breast cancer diagnoses. Compared to other subtypes, TNBC is more frequently diagnosed in younger women, lacks markers for many target therapies and has worse clinical outcomes.
Fn14, a member of the tumour necrosis factor receptor superfamily, is aberrantly overexpressed in 74-87% of invasive breast cancers, but not in normal mammary epithelium, and promotes invasion and metastasis. Fn14 is therefore a promising translational biomarker and drug target.
The Cunliffe lab utilised CRISPR-Cas9 technology to knock out Fn14 expression in the TNBC cell line MDAMB231, generating an Fn14-null isogenic derivative called MDAMB231 J18 (J18). Unexpectedly, J18 cells exhibited a more invasive phenotype than MDAMB231. Comparative RNAseq analysis of MDAMB231 and J18 cells showed enrichment of KRAS signalling as well as the urokinase plasminogen system (uPAR/uPa), a mediator of tumour invasion, multidrug resistance and poorer patient outcomes. As MDAMB231 cells contain an oncogenic G13D KRAS mutation, we postulated that the more invasive phenotype observed in J18 cells was due to dominant KRAS signalling in the absence of Fn14.
Through an investigation of Fn14 expression in publicly available TNBC transcriptome data annotated with patient survival, we identified high Fn14 expression significantly predicted distant metastasis-free survival (p<0.0039) in TNBC (classified by immunohistochemistry), with stronger associations in lymph node positive cases (p<0.00013). Additional statistically significant associations were seen within TNBC Pietenpol subtypes that warrant further investigation. Our in vitro studies comparing MDAMB231 with J18 firstly confirmed upregulation of both uPa (1.44-fold) and uPAR (1.60-fold) proteins by western blotting, validating the previous RNAseq observation. Imaging of both cell lines in 2D and 3D showed that J18 had more morphologic features of invasion and motility including longer filopodia and frequent lamellipodia. Migration assays using Scratch-wound and Boyden Chamber techniques, and invasion assays in Matrigel-coated Boyden Chambers did not show increases in malignant behaviour in J18, however assays were conducted only once each. Lastly, we investigated in triplicate whether J18s had altered sensitivity to Doxorubicin or Trametinib (MEK1/2 inhibitor). While J18s showed no significant increase in resistance to Doxorubicin, they exhibited markedly elevated resistance to Trametinib (IC50 7.69nM WT vs 39.95nM J18).
Together, this study provides novel prognostic insight into the role of Fn14 in TNBC. Furthermore, our data suggests that future targeted inhibition of Fn14 in TNBC cases harbouring oncogenic KRAS mutations may not benefit from inclusion of RAS/RAF/MEK-targeted therapy.