Podium Presentation in Room 2 on Wednesday at 14:40 (Chair: Michael Chen)
Authors: Andrew DelaCourt (1), Alyson Black (1), Peggi Angel (1), Richard Drake (1), Yujin Hoshida (2), Anand Mehta (1)
Hepatocellular Carcinoma (HCC) is the second leading cause of cancer deaths globally, and the incidence rate in the US is predicted to exceed 50,000 patients by 2021. Recent work has identified significant changes in N-linked glycosylation directly in HCC tissue by MALDI glycan imaging. However, there was significant heterogeneity between HCC tissues, suggesting a potential correlation between the glycan changes and specific molecular subtypes of HCC. Therefore, this work will analyze HCC tissues, categorized by subtype, by MALDI-IMS in an effort to fit glycosylation patterns within different classes of HCC.
This project aims to combine the analysis of glycan information in tumor tissue and genetic tumor information, which has not been done previously.
In order to analyze the glycosylation of HCC tissue samples, MALDI-IMS imaging was utilized, which uniquely allows for the analysis of spatially mapped N-glycans to paraffin embedded tissues. The sample set of HCC tissues consisted of Hoshida HCC subtypes 1, 2, 3.1 and 3.2 tissues, with there being 53 total samples. Tissues were prepared through a previously published protocol that involved antigen retrieval, spraying of PNGase F PrimeTM with a TM-Sprayer, then spraying of alpha-cyano-4-hydroxycinnamic acid matrix. This protocol allows for N-glycans to be cleaved while retaining their spatial location on the tissue. Each tissue is then imaged on a Bruker MALDI solariX FT-ICR, and data is analyzed using flexImaging and SCiLS software.
The primary goal of this work was to determine if differences within N-glycan profiles of HCC tissues are related to the differences between Hoshida genetic subtypes. Preliminary data showed that HCC tissues contain increased fucosylation of glycoproteins, along with other glycan expression differences, yet there was significant heterogeneity within analyzed HCC tissues regarding the specific glycan structures that were expressed. With HCC tissues that were categorized into different Hoshida subtypes, glycan MALDI-IMS data showed trends that could approximately separate the subtypes. Subtype 1, which is less differentiated and is associated with poor survival, showed more extreme tumor-associated glycan expression compared to the surrounding normal and cirrhotic tissue than the other subtypes. Using the co-localization feature from SCiLS, subtype 1 tissues showed a significantly higher number of unique glycan structures that associated with the tumor tissue over surrounding normal and cirrhotic tissue. Subtype 1 showed an average of 18.00 ± 8.48 tumor-associated glycans, subtype 2 with 4.25 ± 2.21, and subtype 3 with 2.25 ± 1.71. While the identity of these glycan structures varied from tissue to tissue, trends emerged, particularly regarding how branched and fucosylated glycans associate to tumor tissue with each subtype. Overall, this data shows that some of the heterogeneity in glycan expression of HCC tumors may be attributable to differences between genetic subtypes of HCC.
HCC tissues exhibit glycan expression that is noticeably different from normal or cirrhotic tissue, yet there is significant variation between HCC tissues regarding both the number diversity of glycan structures that are observed. Using the Hoshida classification system, some of that variation can be explained, as differing tumor subtypes appear to express differing numbers of independent, tumor-associated glycan structures, specifically with a high expression in subtype 1.
|Planning to mention or discuss specific products or technology of the company(ies) listed above:||