Breast cancer is the most common type of cancer among women in Norway, and the search for new diagnostic tools and better treatments is continuous. Researchers at the MR Centre at NTNU have mapped the metabolic profiles of different types of breast cancer and hope that this can lead to better prognostic tools.
Maria Tunset Grinde at the MR Center has in her PhD thesis shown that high resolution magic angle spinning magnetic resonance spectroscopy (HR-MAS MRS) and gene expression analysis reflect different characteristics of breast cancer.
“I believe that HR-MAS MRS, and HR-MAS MRS in combination with gene expression analysis, can become a tool in the future. Some patients do not respond very well to treatment – but why? Perhaps MRS combined with gene expression analysis can add extra information about this,” Grinde says.
The aim of the first of the three studies making up the thesis was to map the prognostic factors in breast cancer tissue with HR-MAS MRS and multivariate data analysis.
The researchers analysed 150 of 200 tissue samples from breast cancer patients to find the metabolic markers for oestrogen receptor positive and negative samples. The remaining 50 samples were used to test if the researcher would be able to categorise these correctly, based on the data from the first 150 samples.
“We did pretty well,” Grinde says. “Of the 50 samples, we managed to place 44 in the right group. That means that HR-MAS MRS could be a method for classifying patient samples with different oestrogen receptor status, and that oestrogen receptor positive and negative samples have different metabolic profiles.”
Labelled sugar gives answers
To map the metabolic processes of the aggressive breast cancer called basal-like and the less aggressive luminal-like, Grinde and the other researchers looked at the glycolytic activity – that is the transformation of glucose into lactate. The prognoses in the luminal-like model are usually better than with basal-like breast cancer.
The study is based on two different animal models, one for each of these two types of breast cancer, in which tumour tissue from breast cancer patients was implanted directly into the mice. The mice were given 13C-labelled glucose before the tissue samples were collected. The glucose had then been transformed into the metabolites lactate and alanine.
As expected, both tumour models had high glycolytic activity. In previous studies this has been associated with aggressive types of cancer. However, when the researchers looked at the amount of glucose left in the tumour cells, less glucose was found in the luminal-like breast cancer.
“The results were perhaps not exactly as we expected – and this means that glycolytic activity not necessarily is associated with aggressiveness.
The findings were confirmed by the gene expression analysis, which showed that most of the genes regulating glycolysis were higher expressed in the luminal-like samples.
“We thought it was quite interesting that there is a good correlation between the metabolites and the gene data. I think this is something that could contribute to finding potential targets for cancer treatments in the future,” Grinde says.
Tissue samples give new results
In the third and last study, tissue from both breast cancer patients and xenograft models were analysed, with a particular focus on the choline metabolism. Choline metabolites play an important role in the synthesis of cell membranes and cancer cells typically have higher levels of cholines.
Previous cell cultures studies have indicated that choline and phosphocholine (PCho) are associated with aggressiveness. But in the samples from the xenograft models with human tissue, it looked as if it was rather glycerophosphocholine (GPC) which was associated with aggressiveness. The PCho/GPC ratio was high in the luminal-like models and lower in the basal-like xenograft models.
“It is interesting that there is a difference between cell culture studies and tissue. Cells in cultures often stem from a cancer cell that has been specifically grown, and adapted to that environment. Tissue from patients and these xenograft models are more heterogeneous as cancer tumours normally are.”
Gene expression analysis on the tissue was also performed, which showed a good correlation between the choline metabolites and genes regulating choline metabolism. A very good correlation between the samples from the breast cancer patients and the xenograft models was also found.
The methods for analysing the metabolism in cancer cells are under constant development, and Grinde informs us that a method called hyperpolarised MRS is being tested clinically on prostate cancer patients in San Francisco. This method gives a signal strength which is 10,000 times greater compared to traditional 13C MRS.
“There is an incredible signal boost with this technique and it can be used to measure glycolytic activity in patients. This is something we will see more of in the future,” Grinde believes.
Maria Tunset Grinde will defend her PhD thesis “Characterization of breast cancer using MR metabolomics and gene expression analysis” on 24. October 2012, at 12.15pm in the Auditorium ØHA11 at Øya, Trondheim. The trial lecture will be given at 10.15am at the same location.
- Multivariate modeling and prediction of breast cancer prognostic factors using MR metabolomics. Giskeødegård GF et.al.
- 13C high-resolution-magic angle spinning MRS reveals differences in glucose metabolism between two breast cancer xenograft models with different gene expression patterns. Grinde MT et.al.
High resolution magic angle spinning magnetic resonance spectroscopy (HR-MAS MRS) is a method for studying the metabolism in tissue samples. With traditional MRS the tissue must be extracted, but with HR-MAS MRS the tissue remains intact and it can therefore be used for other analyses such as histopathology and gene expression analysis.
13C HR-MAS MRS
With MRS one typically study signals from the protons, but it is also possible to look at the signals from other MR-sensitive nuclei such as carbon-13 (13C). 13C MRS is less sensitive than proton (1H) MRS as only 1% of the carbon isotopes are 13C. 13C-MRS is used mostly for metabolism studies of 13C-labelled metabolites such as glucose. The glucose is absorbed by the tumour cell and is transformed into lactate and alanine containing a 13C isotope.
Metabolites are molecules taking part in, or made from, the metabolism in the cells.
Xenograft models are models where tissue or cells are implanted from one species (for example a human being) to another (for example mice). Most xenograft models used in cancer research are based on cell cultures. The xenograft models used in this study are based on tissue taken directly from breast cancer patients and implanted into mice.
This post is also available in: Norwegian Bokmål