Biomaterial scaffold-based vaccines sustain robust immune responses through the lymph nodes

Abstract

Emergent vaccine therapies aim to engineer the multi-scale interactions that naturally occur within lymph nodes (LNs), from cellular behavior to tissue dynamics. Therapeutic cancer vaccines have demonstrated safety and immunogenicity in the clinic, but efficacy remains modest, and antigen selection poses a challenge. To improve the potency of the immune response, biomaterial scaffold-based vaccines have been developed to concentrate and activate antigen-presenting cells alongside tumor antigen. In this thesis, we explored the ability of biomaterial scaffold-based vaccines to elicit robust, long-lived, and effective adaptive immune responses against solid and liquid tumors, including melanoma, breast cancer, and leukemia, even without a loaded vaccine antigen. We also applied these vaccines in infectious disease (E. coli) and contraception (targeting a reproductive hormone), tested the impact of vaccine delivery location, and explored the biophysical remodeling of draining LNs after immunization. Biomaterial scaffold-based vaccines formulated as poly(lactide-co-glycolide) matrices, alginate cryogels, and self-assembling mesoporous silica (MPS) rods all generated robust antibody titers that were maintained over a year after single-shot immunization. The cytokine response and antibody subclass production could be tuned by altering the adjuvant formulation. Furthermore, scaffold vaccine delivery proximity to the draining LN did not affect dendritic cell recruitment or LN migration, CD8+ T cell or antibody responses, or therapeutic efficacy, suggesting flexibility in delivery location. The presence of a tumor also did not affect dendritic cell recruitment or phenotype. A unique approach to “antigen-free” vaccination demonstrated efficacy in diverse cancer models, leveraging chemotherapy to harness antigen generated in situ by immunogenic tumor cell death. Across cancer types, exposure to chemotherapy altered the expression of immune-interacting molecules (e.g. MHC-I, PD-L1) on live tumor cells. In a leukemia model, systemic chemotherapy in combination with an antigen-free gel vaccine debulked systemic leukemia burden, evoked antigen spreading, and increased leukemia-specific CD8+ T cells in the bone marrow. Dying AML cells accumulated in gels and LNs, likely supplying antigen to activated antigen-presenting cells in these environments. In a triple-negative breast cancer model, a gel-based vaccine delivering local chemotherapy prevented tumor recurrence after surgical resection and provided long-term protection against metastases. Through noninvasive ultrasound imaging of inguinal LNs, we observed a dramatic (~7-fold) expansion of MPS vaccine draining LNs which was maintained for several weeks before contracting over the following months. LN expansion correlated with vaccine outcomes, and a variety of cellular, mechanical, and matrix responses occurred within the LNs during this response. Notably, an inflammatory monocyte population with antigen-presenting potential was highly enriched in MPS vaccinated mouse LNs, and remained elevated throughout LN expansion. Applying these principles, we developed an off-the-shelf approach to boost bolus vaccine efficacy by first “jump-starting” LN expansion to better engage immune responses.