Mursla Bio has published peer-reviewed data showing that isolating liver-derived extracellular vesicles can reveal disease-associated proteins and microRNAs missed by conventional liquid biopsy analysis.

Tissue-specific analysis reveals more proteins

The study, published in the Journal of Extracellular Vesicles, evaluated NEXPLOR, Mursla Bio’s technology for enriching tissue-specific extracellular vesicles (TS-EVs) from blood plasma.

Extracellular vesicles are particles released by cells that carry proteins, RNA and other biological material. Analysing them could help researchers identify molecular changes associated with disease without taking a tissue biopsy.

However, plasma contains vesicles from multiple tissues, making it difficult to identify which organs are responsible for particular molecular signals.

Researchers compared hepatocyte-derived extracellular vesicles (h-EVs) with bulk EV preparations from the same plasma samples. The samples, each measuring less than 2ml, came from patients with hepatocellular carcinoma (HCC), patients with cirrhosis and healthy donors.

The tissue-specific fraction identified 57% more proteins than bulk EV preparations and showed better reproducibility.

Bulk EV samples were dominated by proteins secreted by the liver into the bloodstream, including albumin. In contrast, the enriched h-EVs contained more proteins associated with the internal structure and function of hepatocytes, including membrane, cytosolic and mitochondrial proteins.

The researchers identified 86 proteins that were either detected exclusively in the h-EV fraction or significantly increased in liver disease compared with healthy donors. Bulk EV analysis identified four unique proteins and no significant differential signal.

The findings suggest that isolating vesicles according to their tissue of origin can provide a more detailed picture of disease-associated molecular changes.

Liver cancer signals distinguished from cirrhosis

The h-EV analysis identified proteins associated with processes shared by cirrhosis and HCC, including fibrosis, altered fat metabolism and inflammation.

Comparing samples from patients with HCC against those from patients with cirrhosis also revealed proteins associated with HCC and a wider network of related proteins. The corresponding bulk EV analysis did not identify the same signals.

Small-RNA sequencing provided a second line of evidence. Although the h-EV fraction produced fewer differentially expressed microRNAs than bulk EV preparations, the tissue-specific microRNAs were associated with HCC signalling pathways. MicroRNAs identified in bulk EVs were instead associated with a more general inflammatory response.

The results indicate that the number of detected biomarkers alone may not determine their usefulness. Their biological context, including the tissue from which they originate, could be equally important.

Dr Tomás Dias, chief scientific officer at Mursla Bio, said: “To our knowledge, this is the most comprehensive validation to date of a platform for tissue-specific EV isolation from human plasma.”

Wider applications remain to be established

The hepatocyte panel is the first developed using the company’s tissue-specific EV platform, with additional panels in development. Mursla Bio says the technology could support applications including disease surveillance, molecular residual disease monitoring, patient stratification and biomarker discovery for drug development.

Its EvoLiver liver cancer surveillance test uses the platform and received US Food and Drug Administration Breakthrough Device designation in April 2025. The designation is intended to support development and regulatory review; it does not constitute marketing approval.

Pierre Arsène, founder and chief executive officer of Mursla Bio, said: “Our thesis is that biological context matters as much as analytical depth.”

The published findings support further investigation of tissue-specific EV analysis as a way to identify disease-associated molecular signals in blood. However, the study does not by itself establish the clinical performance of a diagnostic test or demonstrate improved patient outcomes.

Further research will be needed to determine how reliably the approach can distinguish diseases in clinical settings and how it performs across larger patient populations.