Colorectal cancer prognosis: Imaging mass spectrometry of tissue microarrays
- Published: Mar 16, 2017
- Author: Steve Down
- Channels: Base Peak
Colorectal cancer screening
Cancer is one of the biggest killers in the world, second only to heart disease, and the World Health Organization reported that the 8.8 million deaths caused by cancer in 2015 are expected to increase by 70% over the next two decades. The top cancer killer is lung cancer, facilitated by smoking, which killed about 1.69 million people in 2015. This is more than double the next most prolific cancers with liver, colorectal and stomach cancers accounting for about 0.75 million deaths each.
One of the keys to survival from cancer is early diagnosis but this is not always possible because many types are not discovered until the later stages when they are far more difficult to eradicate. This is often the case with colorectal cancer for which the five-year survival rate has been calculated to be 65% by the American Society of Clinical Oncology. However, this is an average figure. For patients with localised stage colorectal cancer, it rises to 90% but if it has spread to surrounding tissues or organs and/or the regional lymph nodes, the 5-year survival rate is 71%. Further spreading reduces this to about 11%.
Colorectal cancer has been the subject of many screening studies in order to try and identify compounds in the body that signal its presence. Despite the many candidates unearthed, few have been validated in large patient sets to allow their progression into clinical practice. The paucity of sufficient tissue samples with which to test the biomarkers remains one major obstacle to overcome.
Scientists in Germany have been using a system involving tissue microarrays in combination with imaging mass spectrometry on a MALDI instrument to produce the large data sets that are needed. Ronald Simon and colleagues from University Medical Center Hamburg-Eppendorf, have worked with cancer of the prostate, esophagus, bladder and kidney and, in each case, have identified mass spectrometry peaks that correlated with adverse tumour phenotypes. The technique was also deemed to be highly reproducible, a six-month interval between measurements on the same microarray producing matching results.
The next cancer to come under their microscope was colorectal cancer in a new study published in Journal of Mass Spectrometry. A microarray of tissue samples from an earlier study on colon cancer prognosis was reused for this study. The tissues had been collected from 522 patients who had undergone surgery and were fixed in formalin and paraffin-embedded in a microarray. In practice, 173 of these were rejected due to the absence of cancer or spoiled samples.
The microarray blocks were cut into thin sections and mounted on glass slides. Following treatment to remove the paraffin, the proteins present in the tissue were digested with trypsin and the sections were covered in a layer of matrix solution for analysis by matrix-assisted laser-desorption/ionisation mass spectrometry, scanning back and forth across each sample to produce 2D representations.
The microarray sections were then washed and stained so that the mass spectral peaks and intensities could be related to the histological structures. Peaks linked to epithelial structures were classified as positive for tumour cells and all peaks were assigned as negative, weak or positive for colorectal cancer.
Mass spectral markers
The research team searched for mass spectral peaks in the range m/z 615-1955 in which short peptides would be present and found 27 that could be linked to epithelial structures on the tissue. They were surprised to find that 15 of these signals were strongly associated with clinical outcome, seven being linked to a good prognosis and eight to a poor prognosis.
In addition, 18 m/z values were associated with the aggressiveness of the tumour in terms of its stage, grade or nodal status, and 21 others were linked to cell proliferation. When five particular mass spectrometric signals were combined in a multivariate analysis, they provided a three-step classifier that was better than univariate analysis.
Many of these significant m/z values were independent of the tumour stage or spreading of the tumour to the lymph nodes. So, even though the nature of most of the molecules remains unclear at present, it appears that a number of them can be exploited to determine the aggressiveness of colorectal cancer.
The researchers suggested that the combination of tissue microarrays with MALDI imaging mass spectrometry could be used routinely in labs for assessing and monitoring patients with this type of cancer. It would provide better reproducibility than the current DNA /RNA analyses and immunohistochemical tests which are known to have poor interlaboratory reproducibility.
Journal of Mass Spectrometry 2017, 52, 165-173: "MALDI imaging mass spectrometry reveals multiple clinically relevant masses in colorectal cancer using large-scale tissue microarrays"
Article by Steve Down
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.
Follow us on Twitter!