New study identifies key protein biomarkers for early detection of pancreatic cancer

By Dr. Chinta SidharthanReviewed by Susha Cheriyedath, M.Sc.May 19 2024

In a recent study published in the journal Cell Genomics, a team of researchers in China conducted a case-control study to analyze a large panel of serum proteins to identify protein biomarkers for incident pancreatic cancer. They used a Mendelian randomization approach to assess the potential causal effects of these proteins in pancreatic cancer.

Study: Identification of biomarkers and potential therapeutic targets for pancreatic cancer by proteomic analysis in two prospective cohorts. Image Credit: Sebastian Kaulitzki / Shutterstock

Background

Pancreatic cancer has the third highest cancer-related mortality, with late diagnosis cases having a five-year survival rate of only 10%. However, detecting the cancer in its early stages can increase the survival rates by almost 24% to 37%. Given that pancreatic cancer is also a slowly progressing form of cancer, with an average span of 11.7 years between the initial and invasive stages, there is sufficient time to detect the cancer in the early stages.

Traditional cancer biomarkers, which include the carcinoembryonic antigen and carbohydrate antigens 19-9, 125, and 242, have shown varying specificity for pancreatic cancer. Inflammatory proteins such as tumor necrosis factor (TNF), C-reactive protein (CRP), and interleukin-6 (IL-6) have also not shown significant associations with incident pancreatic cancer.

Analysis of the blood proteome, which consists of all the circulating proteins secreted by normally functioning cells and tissues as well as those that are damaged, is a promising method of detecting cancers since aberrations in the circulating proteins often indicate the development of lesions and tumors in the body.

About the study

In the present study, the researchers investigated a prospective cohort to identify and evaluate serum protein biomarkers that could be used to detect pancreatic cancer. The participants consisted of 44 pairs of elderly individuals with pancreatic cancer, and their healthy controls were matched by age, sex, blood-draw date, and hospital. The mean age of the participants was 68.48 years, and 45% of the study population was male. The follow-up data for the cohort spanned approximately 5.7 years.

Circulating proteins were measured from fasting serum samples using a technique based on proximity extension assay. Close to 1,500 proteins were measured and quantified using a Normalized Protein eXpression or NPX value. Various baseline characteristics such as smoking status, alcohol consumption levels, education levels, glycemic indices, and body mass index were compared between the pancreatic cancer cases and matched healthy controls to identify categorical variables.

The protein expression values were standardized, and each protein's odds ratio was calculated. Additionally, the ribonucleic acid (RNA) data obtained from the Genotype-Tissue Expression (GTEx) project was used to examine each protein’s gene expression profile in 54 tissues. A sensitivity analysis was also conducted after stratifying the data by sex and adjusting for the occurrence of type 2 diabetes.

The plasma proteomics dataset obtained from the United Kingdom (U.K) Biobank Pharma Proteomics Project (PPP) was used for a replication analysis for major protein biomarkers. Additionally, a two-sample Mendelian randomization approach was used to evaluate potential causal effects that the identified proteins could have in pancreatic cancer.

Data from a large-scale quantitative trait locus study of proteins and various genome-wide association studies of pancreatic cancer among European populations was used for the Mendelian randomization analysis.

Results

The study identified four proteins — phospholipase A2 group IB (PLA2G1B), tumor necrosis factor (TNF), and regenerating family members (REG) 1A and 1B  — to be associated with incident pancreatic cancer. Of these, REG1A and REG1B were replicated using the U.K. Biobank PPP dataset. Furthermore, the Mendelian randomization analysis using the genome-wide association and quantitative trait locus data found causal effects of REG1A and REG1B in pancreatic cancer.

Colocalization analysis for the REG1 proteins found moderate evidence supporting pancreatic cancer and the REG1 proteins sharing a causal variant. Furthermore, the Mendelian randomization analysis found no evidence of other causal variants confounding the association between the REG1 proteins and pancreatic cancer.

The REG1 proteins were also found to be similarly elevated for incident lung and esophageal cancers. These proteins are part of the REG1 family of proteins found clustered in tandem on the short arm of chromosome 2. These proteins are synthesized in the β cells of the islets of Langerhans in the pancreas and are involved in diabetes development and islet cell regeneration.

The researchers hypothesized that given its role in pancreatic function, tumors or lesions in the pancreas would stimulate the proliferation of the β cells, leading to aberrant secretion of REG1 proteins. Furthermore, the c-type lectin domain found on the REG1 proteins could also bind to the carbohydrates on tumor cell surfaces and promote malignant growth.

Conclusions

To summarize, the researchers examined circulating proteins to identify potential biomarkers for pancreatic cancer. The findings reported two proteins — REG1A and REG1B — which were found to have causal effects in pancreatic cancer and were elevated in cases of lung and esophageal cancer as well. The results highlight the potential for REG1A and REG1B proteins to be used for early detection and large-scale screening for pancreatic cancer.