Molecular Portraits of Cancer : Discovery of Biomarker Signatures using Affinity Proteomics

Abstract

The use of antibodies as capture molecules in assays is a common practice. By either printing monoclonal single chain fragment variables (scFvs) on a solid support – recombinant antibody microarray – or attaching them to magnetic beads – Global Proteome Survey (GPS) – we can specifically capture and measure target proteins of interest. Using the recombinant antibody microarray, we target predominantly intact immunoregualtory proteins, and measure their expression patterns in clinical samples, such as cancer samples. With GPS, we target peptides derived from digested (cancer) samples to identify and quantify a variety of proteins. While antibody microarrays use one antibody per targeted protein, GPS targets peptides, where one antibody can bind hundreds of different proteins – i.e. one antibody per several target proteins. The latter is accomplished by designing antibodies against a short 4 to6 amino acid long peptide motif shared among these proteins. In this thesis, both of these methods have been used to decipher cancer biomarker signatures.
In paper I, the antibody microarray was used to examine the immunosignature of pancreatic ductal adenocarcinoma (PDAC), and how the profile of the immune response differed between cancer patients compared to both healthy and benign controls. We successfully identified an immunosignature of four to ten scFv antibodies that could classify PDAC from controls with high specificity and sensitivity.
In paper II we investigated with the antibody microarray how the immune response evolved with, and shaped, emerging cancer cells in patients later diagnosed with breast cancer using antibody microarray. By determining protein expression profiles for cases and controls, we identified several deregulated immunological proteins that reflected evolving breast cancer, up to two years before diagnosis.
In papers III and IV, we used both the antibody microarrays and GPS to decipher biomarker signatures to differentiate between histological grades in breast cancer tumors. With molecular signatures capable of classifying grade, we could also visualize the heterogeneity of intermediate grade 2 tumors. Grade 2 tumors have little clinical prognostic value, and a majority of samples are classified as grade 2. Results from paper IV showed that it might be possible to reclassify many of these samples as either grade 1 or 3, giving clinicians further information for optimal treatment selection.
In conclusion, in this thesis I have demonstrated the use of profiling parts of the immune response as a tool for surveying and classifying disease, by using antibodies as specific binders to capture and measure low and high abundant proteins.

Abstract (Swedish)

The use of antibodies as capture molecules in assays is a common practice. By either printing monoclonal single chain fragment variables (scFvs) on a solid support – recombinant antibody microarray – or attaching them to magnetic beads – Global Proteome Survey (GPS) – we can specifically capture and measure target proteins of interest. Using the recombinant antibody microarray, we target
predominantly intact immunoregualtory proteins, and measure their expression patterns in clinical samples, such as cancer samples. With GPS, we target peptides derived from digested (cancer) samples to identify and quantify a variety of proteins. While antibody microarrays use one antibody per targeted
protein, GPS targets peptides, where one antibody can bind hundreds of different proteins – i.e. one antibody per several target proteins. The latter is accomplished by designing antibodies against a short 4 to 6 amino acid long peptide motif shared among these proteins. In this thesis, both of these methods have been used to decipher cancer biomarker signatures.
In paper I, the antibody microarray was used to examine the immunosignature of pancreatic ductal adenocarcinoma (PDAC), and how the profile of the immune response differed between cancer patients compared to both healthy and benign controls. We successfully identified an immunosignature of four to ten scFv antibodies that could classify PDAC from controls with high specificity and sensitivity.
In paper II we investigated with the antibody microarray how the immune response evolved with, and shaped, emerging cancer cells in patients later diagnosed with breast cancer using antibody microarray. By determining protein expression profiles for cases and controls, we identified several deregulated
immunological proteins that reflected evolving breast cancer, up to two years before diagnosis.
In papers III and IV, we used both the antibody microarrays and GPS to decipher biomarker signatures to differentiate between histological grades in breast cancer tumors. With molecular signatures capable of classifying grade, we could also visualize the heterogeneity of intermediate grade 2 tumors. Grade 2
tumors have little clinical prognostic value, and a majority of samples are classified as grade 2. Results from paper IV showed that it might be possible to reclassify many of these samples as either grade 1 or 3, giving clinicians further information for optimal treatment selection.
In conclusion, in this thesis I have demonstrated the use of profiling parts of the immune response as a tool for surveying and classifying disease, by using antibodies as specific binders to capture and measure low and high abundant proteins.

• author: Skoog, Petter
• supervisor: Christer Wingren; Carl Borrebaeck
• opponent: Docent Schwenk, Jochen, Kungliga Tekniska högskolan, Solna
• organization: Department of Immunotechnology
• publishing date: 2017-06-09
• type: Thesis
• publication status: published
• subject: Natural Sciences; Immunology
• keywords: Breast Cancer, Pancreatic cancer, antibody microarray, Immunosignature, Global Proteome Survey, Affinity proteomics, Biomarker
• edition: 1
• pages: 74 pages
• defense location: Hörsalen, Medicon Village, Scheelevägen 2, Lund University, Faculty of Engineering.
• defense date: 2017-06-09 09:15:00
• ISBN: 978-91-7753-266-8
• ISBN: 978-91-7753-265-1
• language: English
• LU publication?: yes
• id: 3d09fb9a-8337-4791-a028-7b23bc005a21
• date added to LUP: 2017-05-10 15:51:32
• date last changed: 2018-11-21 21:31:56