Title: Stage-associated differences in the serum N- and O-glycan profiles of patients with non-small cell lung cancer
Journal: Clinical Proteomics
Published: 2019
Background
Lung cancer is the leading cause of cancer-related death in China and globally. In China, lung cancer has seen a significant increase in death rates from 2005 to 2016, with non-small cell lung cancer (NSCLC) accounting for 85% of all cases. Early diagnosis is critical for improving survival, as survival rates are significantly higher in the early stages of the disease. Low-dose computed tomography (LDCT) is a screening tool, but it has limitations, including a high false-positive rate. As a result, there is a growing need for non-invasive biomarkers to improve early detection. Currently, serum markers like CEA and CYFRA21-1 are used, but their sensitivity remains low in the early stages of NSCLC. Abnormal glycosylation patterns, commonly observed in cancers, have emerged as potential biomarkers. This study focuses on investigating serum N- and O-glycan profiles in NSCLC patients to identify possible novel biomarkers using lectin microarray analysis.
Materials & Methods
Subject Recruitment and Sample Collection:
Serum samples were collected from 62 newly diagnosed primary NSCLC patients (28 adenocarcinoma and 34 SCC) at the First Affiliated Hospital of Xi'an Jiaotong University. Inclusion criteria for NSCLC patients included newly diagnosed, histologically confirmed cases, with no prior treatment. Healthy controls were recruited from a health checkup center, confirming no malignancy through various tumor biomarker tests and CT scans. Serum samples were stored at -80°C until analysis.
Serum Protein Labeling:
Adenocarcinoma and SCC patients were grouped by disease stages (Stage I/II, III, and IV) based on the AJCC staging system. Serum from each stage group was pooled and labeled with Cy3 fluorescence dye, followed by separation from excess dye using Sephadex G-25 columns. The labeled proteins were quantified and stored at -20°C.
Lectin Microarrays and Data Analysis:
Lectin microarrays were produced using 37 lectins with different specificities for N- and O-linked glycans. Cy3-labeled serum proteins were applied to the microarrays, incubated, and scanned using a Genepix 4000B scanner. Data analysis was conducted using Genepix 3.0 software, with background subtraction and normalization. The median values of each lectin from repeated blocks were averaged and used for further analysis.
Statistical Analysis:
Statistical analyses were performed using SPSS version 20.0. Differences between groups were evaluated using one-way ANOVA or Student's t-tests. Hierarchical clustering analysis (HCA) was conducted using Expander 7.1 software. P values <0.05 were considered statistically significant.
Results
Serum Glycopatterns in Patients with NSCLC and Healthy Controls
The study analyzed serum glycopatterns in 62 newly diagnosed NSCLC patients (28 with adenocarcinoma and 34 with SCC) and healthy controls. The serum proteins were labeled with Cy3 and analyzed using lectin microarrays. The lectin signals were categorized into three groups: (1) increased normalized fluorescent intensities (NFIs), (2) decreased NFIs, and (3) stable NFIs. Significant alterations were identified between NSCLC and healthy controls, as shown in Fig. 1a and summarized in Table S2.
Adenocarcinoma and Healthy Controls
A total of 18 lectins showed significant glycan expression differences between adenocarcinoma patients and controls. Among these, 14 lectins, including AAL, GSL-I, and SBA, exhibited increased NFIs in adenocarcinoma groups. Specific lectins such as Jacalin, UEA-I, and DBA demonstrated increased NFIs in Stage III/IV adenocarcinoma. Conversely, 4 lectins, such as LTL and RCA-120, showed decreased NFIs in adenocarcinoma patients. These differences were especially noticeable in early-stage adenocarcinoma. Further analysis revealed that the NFIs of some lectins, such as AAL, increased progressively with cancer stage (from Stage I/II to IV), suggesting a potential correlation with tumor burden.
SCC and Healthy Controls
For SCC, 16 lectins exhibited significant differences in glycan expression compared to healthy controls. Nine lectins, including HHL and PHA-E+L, showed higher NFIs in SCC patients. Seven lectins, such as Jacalin and EEL, showed lower NFIs in SCC. These differences were also observed at early stages of SCC, though stage-related changes were not as pronounced as in adenocarcinoma.
Heat map and hierarchical clustering analysis of the normalized data of 37 lectins in the 7 groups.
Discussion
The results indicate significant alterations in the serum glycopatterns of NSCLC patients, with the most prominent changes in early-stage adenocarcinoma. Specific glycan modifications, including core fucosylation and Tn/T antigen expressions, were identified as potential biomarkers for early detection. Core fucosylation, identified by AAL, was notably increased in Stage I/II adenocarcinoma and may serve as a potential early diagnostic biomarker. Tn and T antigens were also upregulated in adenocarcinoma, potentially serving as markers to distinguish adenocarcinoma from SCC. High-mannose glycosylation, identified by HHL, was particularly elevated in SCC patients.
The study highlights the potential of lectin microarrays as a tool for glycan-based serological cancer biomarker discovery. Despite limitations such as the use of pooled serum samples, these findings suggest that glycan profiling may provide valuable diagnostic insights for NSCLC detection, particularly in early stages. Future research is needed to further validate these findings and explore their diagnostic and prognostic significance.
Reference
- Liang, Yiqian, et al. "Stage-associated differences in the serum N-and O-glycan profiles of patients with non-small cell lung cancer." Clinical Proteomics 16 (2019): 1-10. https://doi.org/10.1186/s12014-019-9240-6
