The cocaine and amphetamine-regulated transcript (CART) is an independent prognostic factor in lymph node-negative breast cancer and predicts tamoxifen response


Session type:

Darran O'Connor1, Donal Brennan1, Henriette Laursen1, Sharon McGee1, Sarah McCarthy1, Elton Rexhepaj1, Michael J Duffy3, Aedin Culhane7, Stephen Hewitt2, Finian Martin1, Goran Landberg4, Lisa Ryden4, Rene Bernards6, Robert Millikan5, Karin Jirstrom4, William Gallagher1

1University College Dublin, Conway Institute, Dublin, Ireland, 2National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA, 3St. Vincent's University Hospital, Dublin, Ireland, 4Lund University, Malmo, Sweden, 5University of North Carolina, Chapel Hill, North Carolina, USA, 6Netherlands Cancer Institute, Amsterdam, Netherlands, 7Dana Farber Cancer Institute, Boston, Massachusetts, USA


Proffered paper presentation

Previously, we reported a bioinformatic re-analysis of the Vant Veer et al. breast cancer progression dataset that identified CART as a putative indicator of poor prognosis. Subsequently, we examined CART expression by tissue microarray analysis in two independent breast cancer cohorts comprising over 1000 patients. Cohort I (post-menopausal) consisted of 512 consecutive breast cancer cases diagnosed between 1988 and 1992. Cohort II consisted of 564 pre-menopausal women with primary breast cancer who enrolled in a multi-centre randomised tamoxifen trial between 1984 and 1991.

Automated image analysis of CART expression in cohort I demonstrated a highly significant association between high CART expression and poor overall survival (OS) (p=0.0002). Subset analysis of CART expression in cohort I showed that high CART expression was more specifically a marker of poor prognosis in estrogen receptor (ER)-positive (p=0.0003) and lymph node-negative tumours (p=0.0025). In cohort II, subset analysis revealed a highly significant association between high CART expression and poor recurrence-free survival (RFS) (p=0.008) and OS (p=0.013) in lymph node-negative patients. Since half of cohort II received no tamoxifen, the untreated arm of cohort II was used to investigate the true prognostic potential of CART in the absence of any adjuvant treatment bias. High levels of CART were associated with a decreased RFS (p < 0.001) in untreated lymph node negative patients, and multivariate Cox regression analysis of RFS in this subgroup revealed that CART was an independent prognostic factor (HR 4.81, 95% CI 1.44 - 16.09, p = 0.011).

Additionally, ectopic expression of CART in ER+ MCF7 breast cancer cells, or treatment with recombinant CART peptide, resulted in the transcriptional activation of ER via the MAPK pathway evidenced by phosphorylation of ER at S118, activation of an ERE-luciferase reporter construct and up-regulation of the endogenous ER target gene PgR. Pre-treatment of MCF7 cells with the MEK inhibitor U0126 or mutation of Ser 118 to alanine prevented CART-mediated activation of ER transcriptional activity. As ligand-independent activation of ER through MAPK phosphorylation at S118 is a postulated mechanism of tamoxifen resistance, we examined the effect of CART expression on tamoxifen-mediated apoptosis and found that ectopic expression of CART in two ER+ breast cancer cell lines protected them from tamoxifen-mediated cell death (p=0.025). Furthermore, high CART expression was associated with poor response to tamoxifen in cohort I (p=0.002) and stratifying patients from the randomised tamoxifen trial in cohort II according to CART expression demonstrated a loss of response to tamoxifen in patients expressing high levels of CART.