Finding may help personalize treatment of early-stage breast cancer

In the journal Oncotarget, Brazilian researchers describe a set of 26 genes with the potential to become biomarkers of aggressiveness in ductal carcinoma in situ.

By Karina Toledo  |  Agência FAPESP – In a study recently published in the journal Oncotarget, researchers at the A.C. Camargo Cancer Center in São Paulo, Brazil, describe a set of 26 genes with the potential to become biomarkers of aggressiveness for a subtype of breast cancer known as ductal carcinoma in situ (DCIS).

The study was supported by FAPESP. The principal investigator was Dirce Carraro, a researcher at A.C.Camargo Cancer Center’s Laboratory of Genomics & Molecular Biology.

“If the discovery is confirmed by further research, it will help personalize treatment of the disease and avoid aggressive clinical approaches such as surgery and radiotherapy for patients with lesions considered indolent,” Carraro said in an interview with Agência FAPESP. 

Cases are classified as DCIS when tumor cells of epithelial origin remain confined within the lining of the breast milk duct. Currently, there are no biomarkers capable of distinguishing DCIS lesions with the true potential to progress to invasive disease by spreading to adjacent tissue and causing metastasis from lesions likely to remain relatively stable and indolent in the long term such that they do not threaten the life of the patient.

In general, DCIS is considered non-invasive. In 98% of conventionally treated patients, this disease neither recurs nor progresses to invasive cancer. There is no consensus on the proportion that would progress if left untreated. To make sure DCIS neither recurs nor progresses, oncologists recommend patients with DCIS undergo the most aggressive forms of treatment, including surgical removal of the lesions and, in some cases, radiotherapy. 

Experts believe, however, that treatment could be less aggressive for a significant proportion of DCIS patients and that some cases could possibly be left untreated.

“About 30% of breast lesions detected are diagnosed as DCIS today,” Carraro said. “The percentage will tend to rise as diagnostic methods improve. The lesions are small and non-palpable but can be detected during breast screening. Hence the ongoing effort to find biomarkers that help with determining prognosis and planning treatment.”

Methodology

For approximately ten years, the research team at the Laboratory of Genomics & Molecular Biology has been studying how gene expression is altered in lesions that eventually invade tissue adjacent to the breast milk duct. In a paper published in 2008, Carraro’s group suggested that this modification in gene expression occurs even before any morphological change takes place, i.e., before tumor cells leave their original site to invade neighboring tissue.

The new study published in Oncotarget – with the collaboration of researchers at São Paulo’s Syrian-Lebanese Hospital and the University of São Paulo’s Ribeirão Preto Medical School (FMRO-USP) – reinforces this hypothesis and presents a list of 26 genes that are typically expressed at lower levels in lesions considered more aggressive: AZGP1, CAMP, EDN1, EPOR, GRB10, INPP1, MAPK8, P4HB, RARRES3, ALSM1, ANAPC13, ARHGAP9, CHRNB1, CPN3, CTTNBP2NL, HLTF, LSM4, RABEPK, REC8, UTP20, CLNS1A, FCGR3A, POSTN, SAA1, SLC37A1 and TFF1.

To reach these conclusions, the group evaluated gene expression in 63 frozen tumor samples obtained from the Biobank at the A.C. Camargo Cancer Center. The samples were divided into three groups representing the different stages of progression of the disease.

The initial stage was represented by samples taken from lesions still entirely confined to the milk duct. The most advanced stage was mimicked by the use of malignant cells that had escaped to adjacent tissue. To represent the intermediate stage, the researchers used the so-called “in situ component” (tumor cells present in the duct) from lesions that had become invasive.

The expression of more than 5,000 genes was compared across all three groups using methods considered extremely thorough, such as microarray analysis. “We found a large difference in the expression of these 26 genes when we compared the initial and intermediate stages,” Carraro said. “The difference between the intermediate and advanced stages was smaller, reinforcing the idea that the change in gene expression precedes the change in morphology.”

Carraro and her group believe that this panel of 26 genes represents a gene expression signature associated with DCIS progression. A new study is under way to evaluate the clinical usefulness of this discovery and whether it can be used to identify lesions requiring more aggressive treatment, this time using tumor samples collected for biopsy and later stored in paraffin.

“We’ll analyze only samples of purely in situ lesions, with cells still confined to the milk duct,” Carraro said. “It will be a retrospective analysis, in the sense that we’ll know what happened to the patients who donated the tissue samples. The goal is to find out whether by analyzing the expression of these 26 genes in tumor cells we can identify the cases in which the lesions turned out to be aggressive. We’ll need a large number of samples to do that.”

Other studies in progress are designed to clarify the functional roles played by these 26 genes in the spread of tumor cells. According to Carraro, preliminary data from functional assays with the gene ANAPC13 support the hypothesis that it may serve as a biomarker of aggressiveness, depending on its expression level and its location in tumor cells. “We observed in vitro that when expression of this gene decreases in tumor cell cytoplasm, the cells’ capacity for invasion and migration increases,” she said.

The article “Epithelial cells captured from ductal carcinoma in situ reveal a gene expression signature associated with progression to invasive breast cancer” (DOI: 10.18632/oncotarget.12352) can be read at: impactjournals.com/oncotarget/index.php?journal=oncotarget&page=article&op=view&path[]=12352&pubmed-linkout=1.