Alpha fetoprotein DNA prime and adenovirus boost immunization of two hepatocellular cancer patients

Abstract

Background: Alpha fetoprotein (AFP) is an oncofetal antigen over-expressed by many hepatocellular cancers (HCC). We previously demonstrated that HLA-A2-restricted epitopes derived from AFP are immunogenic in vitro and in vivo despite high circulating levels of this oncofetal antigen. In order to test a more broadly applicable, HLA-unrestricted, inexpensive, cell-free vaccine platform capable of activating tumor antigen-specific CD8+ and CD4+ T cells, we tested full length AFP in a plasmid DNA construct in combination with an AFP-expressing replication-deficient adenovirus (AdV) in a prime-boost vaccine strategy.

Methods: HCC patients who had an AFP+ tumor and previous treatment for HCC were screened and two patients received vaccination with three plasmid DNA injections followed by a single AdV injection, all delivered intramuscularly (i.m.).

Results: The vaccine was well tolerated and safe. Both patients showed immunologic evidence of immunization. The first patient had a weak AFP-specific T cell response, a strong AdV-specific cellular response and recurred with an AFP-expressing HCC at nine months. The second patient developed a strong AFP-specific CD8+ and CD4+ cellular response and an AdV neutralizing antibody response, and recurred at 18 months without an increase in serum AFP.

Conclusions: The AFP DNA prime-AdV boost vaccine was safe and immunogenic. Circulating anti-AdV neutralizing antibodies at baseline did not prohibit the development of AFP-specific cellular immunity. The patient who developed CD8+ and CD4+ AFP-specific T cell immunity had more favorable progression-free survival. The observations with these two patients support development of this vaccine strategy in a larger clinical trial.

Trial registration: ClinicalTrials.gov: NCT00093548.

Figures

Figure 1

IFNγ ELISPOT assay testing CD8+ and CD4+ T cell responses to AFP and AdV. Patient PBMC from all time points were thawed and batch tested by direct IFNγ ELISPOT for recognition of autologous DC alone (background), DC transduced with AdVLacZ (for AdV responses) or co-cultured with AFP protein. Responses were also tested to HLA-A2 restricted peptides pulsed onto T2 cells (not shown). The spot counts after background (T cells + DC alone) per 105 T cells is shown. A: Patient 7 responses are shown. Thin arrows indicate blood draws one month after each plasmid DNA injection and the thick arrow shows the blood draw one month after the AdV boostb. B: Patient 8 responses are shown as in 1A.

Figure 2

Serum anti-AdV neutralizing antibodies. The MFI of AdVeGFP-transduced A549 cells incubated with dilutions of serum is shown, with positive and negative assay controls. A: Results from patient 7 show no evidence of anti-AdV neutralizing antibody induction. Baseline levels show no inhibition of AdVeGFP transduction of A549 cells. B: Results from patient 8 show that the AdV boost induced increased titers of anti-AdV neutralizing antibodies. Patient 8 shows positive baseline neutralizing antibodies as well.

Figure 3

Total anti-adenovirus antibodies. The total AdV ELISA results show no increase in total anti-AdV antibodies associated with AdvhAFP vaccination for either patient (7, 8). Also shown are HD controls (HD1, HD2) and serum from other HCC patients screened for the trial.

Figure 4

Multiplex serum cytokine, chemokine and growth factor assessment. The levels of a broad array of analytes were tested by Luminex assay in the serum of patients 7 and 8. The levels of the 12 most highly detected analytes are graphed on a log scale in pg/ml.