A first-in-class platform pioneering immune cross-dressing for the treatment of solid tumors.
We engineer dendritic cell progenitors to do it themselves, with no antigen selected in advance and nothing loaded onto the cells beforehand.
The receptor forces the dendritic cell to take up vesicles shed by the cancer cell. The cell then carries the tumor's own antigen complexes on its surface. That mechanism is called cross-dressing.
Three approaches dominate the field. Each works, and each runs into the same wall in solid tumors. The immune system never sees enough of what makes a particular tumor different.
Objective response rates run at roughly 15–25% across solid tumors. A few cancer types do considerably better; most tumors do not respond at all.
Transformative in blood cancers. In solid tumors it is limited by poor infiltration, short persistence, loss of the target antigen and T cell exhaustion.
More than 300 trials have shown they are safe and immunogenic, but efficacy has been modest. They use monocyte-derived cells loaded outside the body with a small, pre-selected set of antigens.
The product is made from the patient's own blood. CD34+ cells are cultured for seven days into dendritic cell progenitors, modified with mRNA delivered in lipid nanoparticles, and returned to the patient.
Progenitors engraft and differentiate in the body into type 1 and 2 conventional dendritic cells (cDC1 and cDC2), the subsets specialized for presenting tumor antigen to CD8+ and CD4+ T cells. In head-to-head preclinical comparisons they outperformed monocyte-derived dendritic cells on tumor control.
No conditioning regimen and no lymphodepletion are required, and repeat dosing is feasible.
The progenitors engraft in the tumor and locally release a cytokine payload. IL-12 activates NK and T cells and drives interferon-γ programs that reprogram the tumor microenvironment across several cell types. FLT3L expands the patient's own network of conventional dendritic cells. Delivered as mRNA, expression is transient and self-limiting rather than constitutive.
The EVIR (extracellular-vesicle–internalizing receptor) forces the dendritic cell to internalize tumor-derived extracellular vesicles, and it activates the cell as that material arrives.
The first version of this chimeric receptor, called EVIR, was published in Nature Methods in 2018. The company takes its name from it.
A conventional dendritic cell vaccine breaks antigens down and re-presents them, which means somebody has to decide in advance which antigens matter. Cross-dressing, which the recent literature also calls MHC-dressing, works the other way round. The dendritic cell acquires peptide–MHC class I complexes that the cancer cell has already assembled, and displays them directly. Nothing has to be identified, synthesized or chosen. And because the cell keeps sampling in the patient’s body, it follows the tumor as the tumor changes.
| Aspect | Conventional dendritic cell vaccines | The EVIR approach |
|---|---|---|
| Antigen source | Selected antigens are loaded onto the cells outside the body. | Acquired inside the body from tumor-derived vesicles. |
| Antigen breadth | Fixed at the point of manufacture. Antigen-negative clones escape. | The full antigenic surface of the tumor, sampled continuously. |
| Cell type | Monocyte-derived dendritic cells, with limited engraftment and migration. | Progenitors that generate type 1 and 2 conventional dendritic cells in vivo. |
| Microenvironment | Little effect on the tumor microenvironment. | IL-12-driven interferon-γ response reprograms it across cell types. |
The De Palma laboratory built the first receptor designed to enforce this mechanism for tumor antigen transfer, in 2018, when cross-dressing was a neglected pathway in tumor immunology. In April 2026, Kenneth Murphy's group at Washington University reported in Nature that cross-dressing supplies a substantial part of CD8+ T cell priming after mRNA–lipid nanoparticle vaccination, and that it depends on interferon signaling. That is independent confirmation of the biology this platform was built on.
EVIR Therapeutics is preclinical. The platform has been validated in four peer-reviewed studies across six tumor models, and work is now directed at a first study in patients with recurrent glioblastoma.
Receptor, cell type and payloads defined and published.
Efficacy across several tumor models. Safety profile characterized.
mRNA–lipid nanoparticle product, manufacturing and regulatory package. Current focus.
Planned in recurrent glioblastoma. Not yet open.
Three patent families, licensed exclusively from EPFL, covering the receptor, the cell type and the payloads. The first family, on the receptor itself, is granted in multiple jurisdictions.
The receptor, the cell type and the payloads have each been described in peer-reviewed studies. The four below were carried out by the De Palma laboratory at EPFL and its collaborating groups.
First description of the EVIR receptor. Dendritic cells modified with it take up cancer-cell vesicles selectively and present the associated antigens to T cells, primarily through MHC class I recycling and cross-dressing.
The progenitor platform. Progenitors engineered to produce IL-12 and FLT3L suppressed tumor growth in skin, liver and brain cancer models with no antigen loading and no myeloablative conditioning. Human progenitors were produced by the same method.
Receptor and progenitors combined. Head-to-head superiority over monocyte-derived cells, a favorable safety profile, improved response to PD-1 blockade in a resistant melanoma model, and cross-dressing shown with human cells.
An advanced version of the EVIR receptor. Antigen uptake is coupled to activation of the cell. Treated mice showed expanded low-frequency T cell clones and delayed tumor growth in immunotherapy-resistant melanoma, with no antigen loading and no ex vivo maturation.
Murphy's group shows that dendritic cells acquiring peptide–MHC class I complexes from non-hematopoietic cells account for a substantial part of CD8+ T cell priming, in a type I interferon-dependent manner. That is the pathway our receptor was designed to exploit.
His laboratory described the EVIR and the dendritic cell progenitor platform.
Tumor immunologist. He pioneered personalized dendritic cell vaccines for malignant glioma and leads investigational immunotherapy trials in glioma patients at UCLA.
Neurosurgical oncologist specializing in malignant brain tumors, with a research focus on immunotherapy for glioma.