1228 Immune escape through oncogene-mediated editing of the HLA-ligandome

Dustin McCurry, Emerie E Danson, Xiaohong Leng, Jae‐Hyun Lee, Zhongqi Ge, Rishi Pasamurti, Thomas Köhnke, Oren Pasvolsky, Vinhkhang N. Nguyen, Shuqiang Li, Kenneth J. Livak, Elham Azizi, Robert J. Soiffer, Ravindra Majeti, Jerome Ritz, Sachet A. Shukla, Scott B. Ficarro, Catherine J. Wu, Derin B. Keskin, Jeffrey J. Molldrem, Pavan BachireddyRegular 2025 · DOI

TL;DR

Researchers have discovered that the oncogene ASXL1 doesn't just help cancer grow—it actively "edits" the library of protein fragments (the HLA-ligandome) displayed on the cell surface. By altering the cell's internal protein-shredding machinery, this mutation allows leukemia cells to hide from the immune system, providing a key mechanism for resistance to immunotherapy.

What Problem Does This Paper Address?

The immune system identifies cancer cells by inspecting small protein fragments, called peptides, displayed on the cell surface by Human Leukocyte Antigen (HLA) molecules. Think of HLA as a "billboard" that tells T-cells what is happening inside the cell.

While scientists knew that cancer often "turns down the volume" by reducing the total number of HLA molecules (quantitative escape), it wasn't clear if cancer also "changes the message" by altering which specific peptides are displayed (qualitative escape). This study investigates how specific cancer-driving mutations (oncogenes) reshape this "ligandome" to drive resistance to Donor Lymphocyte Infusion (DLI), a common immunotherapy for leukemia.

Methods

The team employed a multi-omic approach to bridge the gap between patient data and molecular mechanisms:

  • Patient Profiling: They performed whole-exome sequencing (WES) and single-cell RNA sequencing (scRNA-seq) on samples from Chronic Myeloid Leukemia (CML) patients who either responded or failed to respond to DLI.
  • Genome Editing: To isolate the effect of a single mutation, they used CRISPR-Cas9 to create "isogenic" cell lines (K562 and U-937)—identical cells where the only difference was the presence or absence of the ASXL1 mutation.
  • Immunopeptidomics: They used mass spectrometry to physically identify the peptides presented on HLA molecules and utilized computational tools like NetMHCpan and MHCflurry to predict binding affinities.
  • Functional Assays: They measured T-cell killing effectiveness and used flow cytometry to quantify surface HLA levels.

Datasets

  • Clinical Cohort: Patient samples from the Dana-Farber Cancer Institute and MD Anderson Cancer Center.
  • Cell Lines: Standard myeloid leukemia models (K562 and U-937) modified for controlled experimentation.
  • Bioinformatics Tools: MuTect2 (for mutation calling) and Harmony (for integrating single-cell data across different batches).

Key Insights and Ideas

  • ASXL1 as a Resistance Marker: The mutation ASXL1m was found in 78% of patients who failed to respond to immunotherapy, but in 0% of those who did respond. This strongly nominates it as a primary driver of drug resistance.
  • The "Protein Shredder" Connection: The researchers found that ASXL1m reduces the levels of immunoproteasome components (PSMB8, 9, and 10). These are the enzymes responsible for cutting proteins into the "right-sized" peptides for HLA presentation.
  • Altered "Anchor" Residues: The mutation didn't just reduce peptide numbers; it changed the type of peptides. Specifically, it altered the amino acids at key "anchor" positions (2 and 9) that allow a peptide to lock into the HLA molecule.
  • Beyond Quantity: Even when the researchers forced the cells to express more HLA (using Interferon-gamma), the ASXL1m cells still showed lower peptide diversity. This proves the escape is qualitative, not just quantitative.

Key Results

  • Immune Evasion: Cells with ASXL1m showed significantly reduced HLA-I expression and were significantly more resistant to CD8+ T-cell killing (p < 0.0001).
  • Reduced Diversity: Immunopeptidomics confirmed that ASXL1m leads to a significant drop in the variety of peptides presented to the immune system (p < 0.026).
  • Computational Validation: Predictions showed that the peptides remaining on ASXL1m cells had lower binding affinities and lower "presentation ranks," making them less likely to trigger an immune response.

Limitations and Future Work

While this study provides a groundbreaking first look at oncogene-mediated ligandome editing, there are areas for further exploration:

  • Scope of Mutations: The study focuses heavily on ASXL1. It remains to be seen how many other common oncogenes (like TP53 or KRAS) employ similar "editing" strategies.
  • Therapeutic Reversal: The authors identified the immunoproteasome as the culprit. Future work could investigate whether drugs that activate or bypass this machinery could "restore" the ligandome and make cancer cells visible to the immune system again.
  • Clinical Application: While the correlation with DLI resistance is strong, prospective studies are needed to determine if ASXL1 status can be used as a definitive biomarker to guide treatment decisions in the clinic.

Summary updated

BibTeX
@inproceedings{mccurry2025immune,
  author = {Dustin McCurry and Emerie E Danson and Xiaohong Leng and Jae‐Hyun Lee and Zhongqi Ge and Rishi Pasamurti and Thomas Köhnke and Oren Pasvolsky and Vinhkhang N. Nguyen and Shuqiang Li and Kenneth J. Livak and Elham Azizi and Robert J. Soiffer and Ravindra Majeti and Jerome Ritz and Sachet A. Shukla and Scott B. Ficarro and Catherine J. Wu and Derin B. Keskin and Jeffrey J. Molldrem and Pavan Bachireddy},
  title = {{1228 Immune escape through oncogene-mediated editing of the HLA-ligandome}},
  booktitle = {Regular},
  year = {2025},
  doi = {10.1136/jitc-2025-sitc2025.1228},
}
Figure