Analysis of mechanisms of drug resistance to trastuzumab in HER2 overexpressing breast cancer based on gene expression changes in selected cell lines

Funding: PRELUDIUM, NCN
Project period: 2016 - 2019
Project leader: Anna Rusek

Targeted therapy approach was a milestone in cancer molecular medicine. Targeted therapeutics have been established successfully for several cancer types and, in many cases, significantly increase average patient survival. One of the best examples is humanized monoclonal antibody – trastuzumab, a drug against specific breast cancer type, targeting HER2 receptor. Available on market since 1998 as HerceptinTM, combined with a standard chemotherapy, significantly improve overall survival among breast cancer patients. Nevertheless, most of them eventually develop trastuzumab resistance.

The main aim of proposed study is to get insights into molecular mechanisms underlying the development of trastuzumab resistance. We are planning implement a new approach to investigate the evolution of resistance in time, using high throughput method for transcriptome changes investigation combined with computational modelling, in order to explain the process. We expect to identify significant molecular contributors and reveal signalling pathways involved in development of Herceptin resistance. Longitudinal approach will allow us to monitor the stability of observed changes during the process of evolution. It will give the unique opportunity to identify the molecular mechanisms, which rapidly change, pinpointing the most significant pathways engaged in trastuzumab resistance, especially at the early stage.

In order to achieve this goal, two HER2 positive breast cancer cell lines have been chosen as the initial biological model for the study. SKBR3 and BT474 cell lines are confirmed to be trastuzumab sensitive and commonly known to be able to develop Herceptin resistance. These parental cell lines will serve as controls and as a starting point to create resistant cell lines. Proposed biological model aims at creating conditions that mimic clinical situation during development of Herceptin resistance. In order to achieve sufficient similarity, it is planned to use long term, constant, relatively small dose of the drug assessed as 10% - 20% of IC50. The final effect should not exceed 10 - 20 fold resistance. Based on proliferation rate and morphology characteristics, the most appropriate 6 time points will be chosenfor subsequent expression microarray experiments.Taking into consideration the availability of the equipment, as well as experimental design of the study (common control for all time point samples), we decided for Agilent Technologies One Colour Sure Print G3 Human GE 8x60K v2 Microarray. Among other advantages, chosen array provides more than 50 000 biological features for detection, including long intergenic non-coding RNAs, which allows to get enough information to build more integrated picture of studied process. At the first stage of the data analysis we will conduct per gene analysis of gene expression levels. We will account for the multiple testing problem by standard techniques.At this stage of the analysis, we will identify significantly down- and upregulated genes related with trastuzumab resistance phenotype at all time points of the process. Analysis of expression changes (increasing/ decreasing trends in expression intensity) for particular molecules along all the process, will allow to indicate most appropriate set of candidates for subsequent modelling. Once, we obtain a list of genes characterized by significant expression changes during resistance evolution we will use KEGG database to group these genes into pathways. Functional scoring methods (summarized p-value per gene set) will be applied to select few significant pathways using before-after differences on the expression levels. Then we will analyse these selected pathways longitudinally. At the next step we plan to investigate signal network topology with respect to possible mechanisms of drug action and drug resistance.

Breast cancer remains a huge burden for the society. It is the second most common cancer worldwide and the first most common in Europe. Nearly 1.7 million new breast cancer cases were diagnosed worldwide in 2012. Approximately 15% to 20% of them overproduce the growth-promoting HER2 protein. These tumours tend to grow faster, spread more aggressively, are associated with poorer short-term prognosis, and generally are more likely to recur as well are much harder to treat. Trastuzumab led to a 37% relative improvement in overall survival in patients with breast cancer characterized by HER2 overexpression and/or amplification. Unfortunately, about 70% do not respond the treatment due to primary or secondary Herceptin resistance, which makes this therapy much less efficient than it could be.

We are strongly convinced, that the model we will propose, enrich the knowledge in basic science by explaining the mechanisms of trastuzumab resistance development in time at molecular level. This knowledge will serve as source of information for future studies concerning Herceptin resistance and how to overcome this phenomena. Moreover, the longitudinal approach of the study can be extend in a multilevel manner. Even though, undoubtedly transcriptomic variations are most obvious and visible, lying in the centre of molecular biology of the cell, possibly being cause and/or effect of other changes; in future perspective it is reasonable to investigate other levels of investigation (DNA, epigenetic, 3D nuclear, microRNA, protein, and metabolic) in order to build full picture of studied process.