Abstract P1-06-09: Relationships between breast cancer subtypes and expression of autophagy related genes

Background:

Breast cancer (BC) is a heterogeneous disease and can be classified according to the expression of four genes: estrogen receptor (ER), progesterone receptor (PGR), human epidermal growth factor receptor 2 (HER2) and marker of proliferation KI67. Four groups of BC have been described: Luminal A (ER+ or PGR+, KI67- HER2-), Luminal B (ER+ or PGR+, KI67+ or HER2+), HER2 (ER- HER2+) and triple negative (ER- PGR- HER2-). Autophagy is a lysosomal degradation pathway which plays a crucial dual role in tumorigenesis, producing pro-survival or pro-death activity. Altered autophagy has been observed in BC but no study has described transcript level variations of "autophagy genes" according to sub-groups. In order to further explore those alterations, we analyzed gene expression of 40 "autophagy genes" in normal and tumor cells using public transcriptomic data.

Samples and Methods:

5497 transcriptomes were obtained from raw data downloaded from public databases. Two distinct Affymetrix series were built after GC-RMA normalization: one from HG-U133A arrays (n=2806) and one from HG-U133 plus2 arrays (n=2691). Each series was standardized with the Aroma R package that was designed to normalize multicenter extremely large Affymetrix data sets. The same computations were performed in the two data set in order to cross-validate the results. Samples were classified with a reduced number of genes (such as for St Gallen classification) and with classifiers using 50 or more than 300 genes (PAM50, CIT and IntClust centroids).

Results:

In each of the two series, there were about 500 Basal, 300 HER2, 1800 Luminal tumors and 100 normal epithelial breast cells. Positivity or negativity of four genes (ER, PGR, KI67, HER2) produced robust classifications concordant with PAM50 classification based upon centroids. Within the 40 "autophagy genes" studied, at least 10 genes were significantly correlated with one or two sub-groups: ATG3 and ATG9A were associated with ER- tumors including basal group; ATG2B , BECN1 and ULK2 were positively correlated with ER expression and luminal subtypes; ULK1 and RAB24 were associated with ER+HER2+ tumors and MAP1LC3B and ATG5 with ER-HER2+ subgroup; GABARAPL1 was more expressed in normal breast tissue than in cancer cells.

Discussion:

Exploring autophagy in large-scale transcriptome data, we confirmed a previous result showing that GABARAPL1 expression is reduced in breast cancer cells compared with the normal epithelial cells. Interestingly, transcript levels of "autophagy genes"were not evenly distributed among the different tumor subtypes. For instance, some genes were preferentially expressed in ER+ subtypes and others in ER- tumors, suggesting that autophagy might play different roles in the different subgroups, providing different potential targets for therapy.

Citation Format: Feugeas J-P, Belmiloudi S, Boyer-Guittaut M, Peixoto P, Hervouet E. Relationships between breast cancer subtypes and expression of autophagy related genes [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-06-09.