Borrowing Viral Immune Memory to Fight Cancer: PROTEXI Vaccine Awakens Cold Tumors in Mice

Most people’s immune systems remember the coronavirus from infection or vaccination. Researchers are now trying to “lend” this ready-made immune memory to cancer treatment: first reactivating helper T cells familiar with the virus, then having them help killer T cells recognize tumors. A preclinical study published in *Nature Communications* showed that a dendritic cell vaccine called PROTEXI slowed tumor growth and prolonged survival in multiple mouse models.

Dendritic cells collect and present antigens, showing T cells what to attack. Existing cancer vaccines, however, are often limited by the weak immunogenicity of tumor antigens and the difficulty of obtaining tumor fragments capable of reliably activating CD4⁺ helper T cells. PROTEXI therefore uses a mixed design in which the same batch of dendritic cells presents both tumor antigens and CD4⁺ T-cell epitopes from the SARS-CoV-2 spike protein. This harnesses viral memory to provide stronger helper signals, which in turn activate tumor-recognizing CD8⁺ cytotoxic T cells.

In B16F10 melanoma and 4T1 breast cancer mouse models, PROTEXI inhibited tumor growth, increased survival, and prompted more T cells, natural killer cells, and antigen-presenting cells to enter “cold tumors” that initially contained few immune cells. The study also observed epitope spreading: the immune response was no longer directed solely at the antigens introduced by the vaccine but gradually extended to other tumor antigens. When researchers removed CD4⁺ T cells, the tumor-suppressing effect disappeared, supporting the role of helper T cells as the central component of the design.

To more closely approximate human immunity, the team also created humanized mice using peripheral blood immune cells from people who had received COVID-19 vaccines, then implanted human melanoma. PROTEXI loaded with spike protein epitopes and the tumor antigens PRAME and MAGE-A3 reduced tumor burden and significantly increased T-cell responses to the spike protein. However, responses to PRAME and MAGE-A3 showed only an upward trend and did not reach statistical significance.

The study also tested combination treatments. In a model that responded poorly to immune checkpoint therapy, combining PROTEXI with an anti-PD-1 antibody enhanced intratumoral T-cell responses. Another set of experiments added the TGF-β type I receptor inhibitor vactosertib. The paper showed that the combination delayed tumor progression and kept one-quarter of the mice tumor-free for more than 50 days. MedPacto separately stated that tumors in the combination group were less than one-quarter the volume of those in the PROTEXI-only group, and that the survival rate on day 25 of treatment increased from approximately 50% to more than 75%. These precise comparisons come from the company’s description of the animal data and cannot be regarded as predictions of clinical efficacy.

The study still has clear limitations. Most mechanistic experiments used mouse antigens or surrogate epitopes to simulate pre-existing viral memory, while the humanized model used cells from only a small number of healthy donors and could not reproduce the full condition of cancer patients after chemotherapy, aging, or immunosuppression. Differences in individuals’ HLA types, the strength of their spike protein memory, and the tumor antigens available may also affect whether the vaccine works.

PROTEXI therefore currently demonstrates a feasible immunological concept, not a cancer vaccine that is ready for use. In addition to confirming manufacturing consistency, dosing safety, and suitable patient populations, the next step must answer two key questions in human trials: whether viral memory can be reliably converted into tumor killing, and whether, when combined with anti-PD-1 or a TGF-β inhibitor, the benefits are sufficient to offset the additional toxicity and treatment complexity.