Twin-screw extruded lipid implants for vaccine delivery

In recent years there has been a considerable focus on the
development of subunit vaccines, preferred over traditional
vaccines for reasons of safety and purity. However, subunit
vaccines are less immunogenic than attenuated vaccines and need
therefor multiple administrations in combination with
immunostimulatory adjuvants, in order to induce immunity. The
sustained release of a vaccine together with the release of an
adjuvant is a potential alternative to giving multiple doses. The
aim of this thesis was to manufacture lipid implants for vaccine
delivery by twin-screw (tsc) extrusion and evaluate the potency of
these lipid systems to stimulate an immune response in vivo. To
accomplish this, lipid implants consisting of cholesterol, soybean
lecithin, and Dynasan 114 (D114) were prepared. Different
formulations were evaluated for their extrudability before adding
the model antigen ovalbumin (OVA) and the adjuvant Quil-A (QA) to
the formulation. Investigating the release behaviour of OVA and QA
showed that mainly cholesterol influences the release behaviour of
OVA, increasing the fraction of cholesterol slows down the release
of OVA. To further slow down the release of OVA from the implants,
they were cured at different temperature resulting in an even
longer OVA release. Furthermore, the addition of QA to the implants
influenced the release behaviour of OVA and vice versa. The
investigation of the implant polymorphism after the extrusion
process as well as during storage showed good stability. To combine
the advantage of particulate delivery and sustained release,
preformed liposomes were incorporated into the implants prior to
extrusion. For the analysis of the immune response, two sets of
animal experiments in mice were performed, one evaluating the
kinetics of the release of the model antigen in vivo, a second one
to evaluate the immune response in vivo. Evaluation of these data
indicated a correlation between the in vitro and in vivo release
behaviour of OVA. Furthermore, immune responses similar to those
induced by two booster injections, consisting of OVA and alum could
be achieved using implant formulations containing QA. These results
further emphasized the importance of adjuvant in the formulation.
The incorporation of preformed liposomes into the implants on the
other hand did not lead to an improved outcome. In a second part of
this work, an in vivo tumour study was prepared, using the TRP2
peptide as active ingredient. Due to the use of this expensive
peptide, a transfer to a different extruder was necessary. The
influence that a change of the production device has on the
implants characteristics was investigated. Once the formulation was
adapted to the new extruder, implants containing TRP2 and QA were
produced. The in vitro release of TRP2 proved to be very slow, much
different from the OVA release. Furthermore, the preparation of
vesicular phospholipid gels (VPGs) as an alternative lipid delivery
system for TRP2 was investigated. The TRP2 release from the VPGs
was also slow and incomplete. Both formulations were used in an in
vivo tumour growth study. Mice were injected with B16F10luc2
melanoma cells, 6 days later formulations were administered. VPGs
showed adverse reactions in the mouse and are therefore not s
suitable delivery system. TRP2 implants showed a slow delay in the
start of tumour growth, but were not more potent that TRP2 in PBS
injections given to the mice. The very slow in vitro release data
of TRP2 brought up the question about interactions between the
lipid implants and the peptide influencing the release. Choosing
peptides of different size and hydropathy, an investigation of
their release behaviour and interaction with the implants was
conducted. In conclusion, lipid implants were well tolerated and
offer a great potential as sustained release delivery system for
vaccines. They allow releasing the active component and the
adjuvant together, enabling to achieve a strong immune response.