Immunotherapies that harness an individual’s immune system are at the forefront of cancer therapeutics with curative potential. Cancer vaccines that generate de novo T cell responses have been very successful at providing prophylactic immunity to prevent certain forms of cancer (e.g. Gardisil). However, eliminating pre-existing or residual cancer cells after initial front-line therapies using cancer vaccines has had limited success. Our laboratory has studied biomaterial-based cancer vaccines, which provide both controlled release of bioactive molecules and a niche for dendritic cell (DC) recruitment. We previously used macroporous scaffolds delivering granulocyte-macrophage colony-stimulating factor (GM-CSF) and TLR9 agonist CpG to recruit and activate DCs in the presence of tumor antigen [1,2]. Activated DCs traffic to draining lymph nodes and prime CD8+ T cells against this antigen. These biomaterial vaccines have shown greatly improved antitumor efficacy over bolus vaccines (containing the same components without the biomaterial). In mouse models of solid tumors, we have demonstrated that these vaccines have potent prophylactic efficacy and can synergize with other clinically approved immunotherapies to eradicate established tumors .
In our recent work in Nature Biomedical Engineering, we extended this system to treat acute myeloid leukemia (AML), a blood cancer. We used alginate-polyethylene glycol (PEG) cryogels: hydrogels polymerized at -20˚C to form ice crystals, leaving a macroporous network once thawed. These were prepared by pipetting the chilled polymer mixture mixed with initiator into prepared molds. First, we tested whether this vaccine could prevent AML engraftment when administered as a prophylactic. We subcutaneously injected mice with cryogels delivering GM-CSF, CpG, and antigen in the form of either AML cell lysates or an AML-associated antigen peptide. Both of these vaccines generated robust antitumor CD8+ T cell responses and prevented AML expansion better than the bolus vaccine control.
We next assessed the cryogel in a therapeutic setting in leukemia-bearing mice. Here, the vaccine alone was not 100% effective, but when combined with inductive chemotherapy it effectively eradicated disease and prevented relapse. The bone marrow of mice treated with this combination showed no residual leukemia cells and contained anti-AML CD8+ T cells capable of preventing AML engraftment when transferred into naïve recipient mice. We repeated this survival experiment in a double-blinded manner, and to be thorough decided to include an additional experimental group: an antigen-free cryogel vaccine containing only GM-CSF and CpG (Fig. 1). We were surprised to find that this group had identical survival as mice treated with the full, antigen-containing vaccine. As we were expecting the vaccine efficacy to depend on antigen uptake by recruited DCs, we repeated this experiment again and observed the same result. Clearly another mechanism was at play, and chemotherapy was critical in this outcome.
Figure 1. Schematic of the combination therapy using chemotherapy and an antigen-free cryogel vaccine to eradicate AML from diseased mice (images from BioRender).
To broadly investigate potential mechanisms behind this result, we delivered antigen-free vaccines into mice bearing AML and harvested cryogels, lymph nodes, and bone marrow at multiple timepoints. We observed that AML cells accumulated in the gel scaffolds and draining lymph nodes over time. Importantly, these cells expressed high levels of apoptotic markers, indicating cell death in these compartments where DCs are already concentrated and available to take advantage of released AML antigens. Chemotherapy was critical: in addition to antigen generation it debulked the systemic tumor burden to give immunity a fighting chance and depleted regulatory T cells to boost CD8+ T cell function in the bone marrow.
The promising results we observed with the cryogel vaccine in AML could lead to human vaccines capable of preventing relapse in leukemia patients previously treated with chemotherapy, and in extending this approach to other hematologic cancers. Translating these findings to solid tumors may prove more of a challenge, but this chemo-immunotherapy combination could potentially generate immune responses against circulating metastatic cells using a similar mechanism. With further development, this antigen-free cancer vaccine could provide personalized yet off-the-shelf cancer therapy.
Our paper: Shah, N.J. & Najibi, A.J., et al. A biomaterial-based vaccine eliciting durable tumour-specific responses against acute myeloid leukaemia. Nature Biomedical Engineering (2020). doi: https://doi.org/10.1038/s41551-019-0503-3.
 Ali, O.A., et al. Infection-mimicking materials to program dendritic cells in situ. Nature Materials (2009). doi:10.1038/NMAT2357.
 Li, A.W., et al. A facile approach to enhance antigen response for personalized cancer vaccination. Nature Materials (2018). doi: https://doi.org/10.1038/s41563-018-0028-2.