SUMMARY: The American Cancer Society estimates that in 2024, about 66,440 people will be diagnosed with pancreatic cancer and 51,750 people will die of the disease. Detecting cancer at early stages can significantly increase survival rates and outcomes.
Several multi-cancer early detection tests are being developed that involve blood-based circulating cell-free tumor DNA (cfDNA) in the plasma, to track hundreds of patient-specific mutations, to detect Minimal Residual Disease (MRD) , as well as detection of abnormal methylation patterns, followed by machine learning approaches, to differentiate between cancer and non-cancer for detecting clinically significant, late-stage (III and IV) cancers. Early detection of cancer, however, is the key to improving survival. This is particularly relevant for certain cancer types. Pancreatic Ductal AdenoCarcinoma (PDAC) is one of the deadliest cancers, and a leading cause of all cancer-related deaths in the United States and is typically detected when the disease is advanced. However, when detected at Stage I, survival rates can be as high as 80%. Even though serum CA19-9 is intended as an aid in the management of patients with confirmed pancreatic cancer for serial monitoring of their response to therapy and disease progression, it is NOT recommended by the FDA for screening, as it may be elevated in several benign conditions. Currently, there are no general screening strategies to detect asymptomatic, early-stage PDAC and there is therefore a significant unmet need in this patient group.
Exosomes are 30-150 nm-sized Extracellular Vesicles (EVs) secreted by multiple different cell types and released by tumors into the bloodstream. They mediate intercellular signaling by shuttling mRNAs and microRNAs between distant cells and tissues and therefore carry functional protein biomarkers representing the tumor proteome. MicroRNAs play a role in regulating gene expression and have been implicated in various cancer types due to their stability and specific expression patterns associated with tumor presence. Exosomes retain the cytoplasmic content of the cell from which they were shed, essentially replicating the biology of their tissue of origin. Exosomes represent one potential approach for more sensitive detection of cancer-related biomarkers from blood.
The researchers developed an exosome-based liquid biopsy signature as a diagnostic tool to detect pancreatic cancer through the analysis of circulating biomarkers in blood samples. This signature utilizes two primary types of biomarkers – Eight microRNAs unique to exosomes shed from pancreatic cancer cells, combined with five cell-free DNA (cfDNA) markers such as mutations or alterations in DNA methylation patterns found in the blood of patients with pancreatic cancer.
In previous work, researchers had tested an exosome-based liquid biopsy signature on a cohort of 95 individuals from either the United States or Japan. This initial study reported an impressive 98% detection rate for pancreatic cancer. Building on this success, the latest research aimed to validate this liquid biopsy approach in larger, more diverse populations across multiple institutions and countries.
The most recent study involved a comprehensive, multi-cohort evaluation to assess the performance of the exosome-based liquid biopsy signature. The study enrolled participants from four countries, each contributing both pancreatic cancer patients and healthy donors:
• Japan: 150 individuals with pancreatic cancer and 102 healthy donors
• United States: 139 individuals with pancreatic cancer and 193 healthy donors
• South Korea: 184 individuals with pancreatic cancer and 86 healthy donors
• China: 50 individuals with pancreatic cancer and 80 healthy donors
The research methodology involved training the liquid biopsy signature using data from the Japanese cohort. Machine learning algorithms were likely used to train and validate the diagnostic model, optimizing its sensitivity and specificity. After training the signature on the Japanese cohort, the researchers validated its performance using data from cohorts in the United States, South Korea, and China. Each of these cohorts included a mix of pancreatic cancer patients and healthy controls, allowing for an assessment of the signature’s performance across diverse populations.
The performance of the liquid biopsy signature across different cohorts was as follows:
• United States Cohort: The test detected 93% of pancreatic cancers.
• South Korean Cohort: The test detected 91% of pancreatic cancers.
• Chinese Cohort: The test detected 88% of pancreatic cancers.
These validation results demonstrated the signatures robustness and effectiveness in detecting pancreatic cancer across different ethnic and geographic backgrounds, although there was a noted decrease in detection rates with different cohorts.
To enhance the diagnostic accuracy, the researchers combined their exosome-based signature with the traditional pancreatic cancer marker CA 19-9, a carbohydrate antigen often elevated in pancreatic cancer but is less sensitive for detecting early-stage disease. The combination of the exosome-based signature with CA 19-9 improved the detection rate of early-stage pancreatic cancers (Stage I and II) to 97% in the U.S. cohort. This suggests that the combination approach offers a more comprehensive evaluation than CA 19-9 alone.
In conclusion, the exosome-based liquid biopsy signature represents a significant advancement in the early detection of pancreatic cancer. By leveraging the unique properties of exosomal microRNAs and cfDNA, the researchers have developed a diagnostic tool with high sensitivity and specificity. The successful validation in multiple cohorts underscores its potential utility in clinical practice. This assay may be helpful for early detection of pancreatic cancer in certain groups with a high risk for pancreatic cancer such as a family history of pancreatic cancer, germline BRCA mutations, new-onset diabetes, chronic pancreatitis or pancreas precancer lesion such as intraductal papillary mucinous neoplasm. However, addressing the limitations and further optimizing the technology will be crucial for its widespread adoption and effectiveness in diverse populations.
An exosome-based liquid biopsy for non-invasive, early detection of patients with pancreatic ductal adenocarcinoma: A multicenter and prospective study. Xu C, Xu Y, Han H, et al. AACR Annual Meeting 2024. Abstract 3899. Presented April 8, 2024.