Background

Dual specificity tyrosine-phosphorylation-regulated kinase 2 (DYRK2) belongs to a family of CMGC kinases that function as modulators of different downstream pathways that allow cells to cope with hypoxia, DNA damage and various stress signals. Additionally, DYRK2 has been implicated in various human cancers with both pro- and anti-tumour roles, which are probably cancer type- and cell type-dependent. Furthermore, studies show that DYRK2 is involved in epithelial-mesenchymal transition, hence suggesting a role in tumour metastasis. The current study investigates the prognostic role of DYRK2 in breast cancer and investigates its potential as a novel therapeutic target.

Methods

Immunohistochemistry was employed to investigate if nuclear expression of DYRK2 was associated with clinical outcome measures in a cohort of 715 patients. Expression was determined using the weighted histoscore method. Antibody specificity was confirmed in paraffin embedded cell pellets +/- DYRK2 silencing. Cell counts in parental and CRISPR-mediated DYRK2 knocked-out MDA-MB-468 and MDA-MB-231 cells (ER, PR, HER2, AR negative) were measured using Alamar Blue; NSGTMmice (n=8) were injected subcutaneously with MDA-MDB-231 with or without DYRK2 depletion to assess tumour growth in vivo.

Results

In a cohort of 715 patients, median follow-up was 160 months with 155 breast cancer deaths and 135 deaths due to other causes. The majority of patients were over 50 years of age (71%), had ductal carcinoma (88%), tumours <20mm in size (56%) and node negative disease (57%). 489 patients had ER positive disease, 226 had ER negative disease and of these 148 had TN (triple-negative) disease. DYRK2 expression was observed in the cell cytoplasm and nucleus and ranged from 3 to 200 weighted histoscore units (WHS) and ROC analysis was used to determine cut-offs, tumours with a cytoplasmic and nuclear WHS <145 were classified as low expression and tumours with a cytoplasmic and nuclear WHS >145 were classified as high expression. In the full cohort (p=0.087) and ER negative (p=0.066) cohort DYRK2 was not associated with cancer specific survival. However in TN disease high DYRK2 expression was associated with cancer specific survival (p=0.012, mean survival 145 months versus 107 months). This was potentiated in patients with ER, PR, HER2, AR negative disease (p=0.005, mean survival 166 months versus 100 months) and independent in multivariate analysis with age, histological tumour type, tumour size tumour grad, nodal status, ki67 index, chemotherapy, radiotherapy and recurrence (p=0.13, HR 3.920). Following this observation, patients with ER, AR negative disease were investigated and again high DYRK2 expression was associated with cancer specific survival (p=0.0003, mean survival 163 months versus 86 months) and was independent when combined in multivariate analysis (p=0.001, HR 4.154).

To investigate if DYRK2 was a potential target in TN breast cancer, the effect of silencing DYRK2 was investigated. CRISPR-mediated DYRK2 depletion impeded cell proliferation in TN cell-lines and markedly reduced tumour burden in mouse MDA-MDB-231 xenografts (p<0.0001).

Conclusions

Our studies indicate that DYRK2 is indeed a potential therapeutic target for patients with TN breast cancer or ER, AR negative breast cancer.

Citation Format: Edwards J, Baillie G, Quinn J, Monreno R, Banerjee S, Tomkinson N, MacKay S, De La Vega L. DYRK2 is a novel therapeutic target in ER negative breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-10-10.