📄 medicine

Detecting tumor-associated peptides from the plasma-derived soluble immunopeptidome in human hepatocellular carcinoma patients: a pilot study

This pilot study demonstrates that plasma-derived soluble HLA-I peptides in hepatocellular carcinoma patients contain tumor-associated antigens shared with tumor tissue, suggesting that sHLA-I profiling is a promising minimally invasive strategy for identifying actionable targets and tumor-specific signatures in HCC.

Original authors: Amy L. Kessler, Yannick S. Rakké, Lisa Bokhout, Robbie J. Luijten, Wouter A.S. Doff, Karel Bezstarosti, Roel F.A. Pieterman, Monique T.A. de Beijer, Jan N.M. IJzermans, Michail Doukas, Dave Sprengers

Published 2026-07-07

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Amy L. Kessler, Yannick S. Rakké, Lisa Bokhout, Robbie J. Luijten, Wouter A.S. Doff, Karel Bezstarosti, Roel F.A. Pieterman, Monique T.A. de Beijer, Jan N.M. IJzermans, Michail Doukas, Dave Sprengers, Jeroen A.A. Demmers, Sonja I. Buschow

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine your body is a bustling city, and your immune system is the police force patrolling the streets. To do their job, the police need to know what the "criminals" (cancer cells) look like. They do this by checking ID cards carried on the surface of every cell. These ID cards are called HLA molecules, and they display tiny fragments of proteins (peptides) from inside the cell. If a cell is cancerous, its ID card might display a "wanted poster" piece of a tumor protein.

Usually, to see these wanted posters, doctors have to perform surgery to cut out a piece of the tumor and examine it directly. This is like sending a detective into a dangerous building to look for clues. It's invasive, and sometimes there isn't enough building left to examine.

The Big Idea: The "Liquid" Clue
This study asks a simple question: Can we find these "wanted posters" floating in the blood instead?

When cells talk to each other or when they die, they shed these ID cards into the bloodstream. These floating ID cards are called soluble HLA (sHLA). The researchers hypothesized that in patients with liver cancer (Hepatocellular Carcinoma, or HCC), the blood would be full of these floating cards carrying clues about the tumor, acting like a "liquid biopsy."

How They Did It (The Detective Work)

The Collection: They took blood samples from two groups of people: those with liver cancer and those with severe liver disease but no cancer.
The Fishing Net: They used a special magnetic "net" (antibodies) to fish out all the floating HLA cards from the blood plasma.
The Cleanup: The blood is a messy place. It's full of common proteins (like albumin) that break down into tiny pieces. These pieces are like trash that clogs the net. The researchers developed a smart filter to throw away the trash and keep only the specific "wanted poster" pieces that fit the shape of an HLA card.
The Analysis: They used a high-tech machine (Mass Spectrometry) to read the chemical code of every piece they caught. They then used a computer to predict which pieces fit the specific "lock" (HLA type) of each patient.

What They Found

The Cleanup Worked: Before cleaning, the data was messy. After removing the "trash," they found thousands of unique, high-quality peptide pieces. Most of these were the perfect size to be real HLA ID cards.
The Cancer Connection: They compared the floating cards from cancer patients against those from non-cancer patients. They found that the cancer patients had a unique set of cards that the healthy (or non-cancer sick) people didn't have.
Matching the Source: They also had tissue samples from the actual tumors of some patients. They found that about 17 different tumor-associated antigens (the "wanted posters") were present in both the actual tumor tissue and the floating blood cards.
Specific Hits: Nine of these "wanted posters" were found only in the blood of cancer patients, not in the non-cancer group. Five of these matched the actual tumor tissue perfectly. One of these was a protein called ESM1, which is known to be high in various cancers.

The Limitations (The Fine Print)
The authors are careful to say this is a pilot study (a small test run).

Sensitivity: They didn't find every tumor clue. The technology isn't sensitive enough yet to catch everything. It's like having a metal detector that finds big coins but misses small ones.
No New Mutations: They didn't find any "neoantigens" (brand new mutations caused by the cancer's DNA errors). This is because they didn't have the full genetic map of the tumors to compare against.
Size: The group of people they studied was small. They need to test many more people to be sure this works for everyone.

The Bottom Line
This study proves that it is possible to catch tumor clues floating in the blood of liver cancer patients. It shows that the blood contains a "signature" of the tumor that overlaps with what is found in the actual tissue. While this isn't a ready-made medical test for doctors to use tomorrow, it opens the door for a future where a simple blood draw could help identify specific targets for immunotherapy or monitor the disease, without needing to cut into the patient.

Technical Summary: Detecting Tumor-Associated Peptides from Plasma-Derived Soluble Immunopeptidome in Hepatocellular Carcinoma

Problem Statement
Hepatocellular carcinoma (HCC) currently exhibits limited response to immune checkpoint inhibitor (ICI) therapy, necessitating the development of antigen-based immunotherapies and improved disease monitoring strategies. While identifying targetable T-cell epitopes via liquid chromatography–mass spectrometry (LC-MS/MS) of membranous HLA (mHLA) is standard, this approach is constrained by tissue scarcity and inter/intra-patient heterogeneity. The authors hypothesize that plasma-derived soluble HLA class I (sHLA-I) peptides could serve as a minimally invasive "liquid biopsy" source that reflects tumor presence and antigen presentation, potentially overcoming the limitations of tissue-based profiling. However, prior to this study, the sHLA-I cargo had not been interrogated specifically for HCC.

Methodology
This pilot study utilized a cohort of HCC patients ( $n=5$ ) and non-HCC controls with end-stage liver disease ( $n=5$ ) from Erasmus MC University Medical Center.

Sample Collection: Paired tumor and adjacent liver tissues were obtained via surgical resection or transplantation. Plasma was collected, platelet-depleted, and stored at -80°C.
Immunopurification: Soluble HLA-I complexes were isolated from 5 mL of plasma using pan-HLA-I affinity beads (clone W6/32). Immunoprecipitation efficiency was verified via sandwich ELISA.
Mass Spectrometry: Peptides were eluted, desalted, and analyzed using LC-MS/MS (Orbitrap Eclipse Tribrid) in data-dependent acquisition (DDA) mode.
Data Processing & Filtering: Raw MS data were searched against human and hepatitis B reference proteomes. A critical step involved a structured decision tree to filter out peptides derived from abundant plasma proteins (contaminants), retaining only those likely to be bona fide HLA ligands.
Bioinformatics: Identified peptides were subjected to patient-specific HLA-binding prediction (NetMHCpan 4.1, $\le$ 2% binding rank). The study compared sHLA-I peptidomes against: (1) paired mHLA-I peptidomes from tumor and adjacent tissues, and (2) sHLA-I/mHLA-I from non-HCC controls. Tumor-associated antigens (TAAs) were identified by cross-referencing with established databases.

Key Results

Peptidome Quality: After contaminant filtering, the plasma-derived sHLA-I peptidomes contained between 1,553 and 9,421 unique peptides per sample. The data showed a strong enrichment for canonical 8–11-mer lengths and a high proportion of peptides predicted to bind patient-specific HLA-I alleles.
Source Proteins: The HCC sHLA-I peptidome was derived from 3,907 source proteins. Of these, 815 proteins were unique to HCC patients (not found in non-HCC controls). Protein interaction analysis revealed that these HCC-unique source proteins formed a significantly connected network centered around tumor-associated proteins.
Tumor-Associated Antigen (TAA) Overlap:
- Comparison of HCC patient sHLA-I and mHLA-I (tumor/adjacent) revealed 20 peptides from 17 distinct TAAs shared between the liquid and tissue sources.
- When filtering for peptides unique to HCC (absent in non-HCC controls), 9 TAAs were identified in the sHLA-I cargo.
- Five of these nine TAAs overlapped with peptides found in tumor mHLA-I. Notable examples included ESM1, MAGED4, and H3F3A.
- Specific peptides from DCD2 and ESM1 were identified only in tumor tissue mHLA-I and not in adjacent tissue, yet their corresponding peptides were detected in the plasma sHLA-I of HCC patients.
Quantitative Levels: Unlike findings in other cancers, sHLA-I levels in HCC patients were not significantly elevated compared to non-HCC controls, suggesting sHLA-I levels may reflect hepatocellular disease burden rather than tumor-specific biology in this context.

Significance and Claims
The authors present this work as a proof-of-concept demonstrating that the plasma-derived sHLA-I immunopeptidome in HCC patients contains peptides derived from tumor-associated antigens, some of which are also presented on tumor tissue.

Minimally Invasive Utility: The study supports the potential of sHLA-I profiling as a non-invasive strategy to identify actionable targets and tumor-specific signatures for HCC.
Hypothesis-Generating: The authors explicitly state that due to the small, heterogeneous cohort and analytical sensitivity limits, the study is hypothesis-generating. They note that the low overlap between tumor and sHLA TAA cargo may be due to sensitivity constraints.
Future Directions: The paper concludes that establishing clinical utility requires future improvements in analytical sensitivity and throughput, as well as validation in larger, more diverse patient cohorts. It suggests that combining sHLA-I immunopeptidomics with circulating tumor DNA (ctDNA) could enhance prognostic screening and neoantigen identification, though this combination was not tested in the current study.

More like this