Autologous dendritic cells transfected with prostate-specific antigen RNA stimulate CTL responses against metastatic prostate tumors

Abstract

Autologous dendritic cells (DCs) transfected with mRNA encoding prostate-specific antigen (PSA) are able to stimulate potent, T cell-mediated antitumor immune responses in vitro. A phase I trial was performed to evaluate this strategy for safety, feasibility, and efficacy to induce T cell responses against the self-protein PSA in patients with metastatic prostate cancer. In 13 study subjects, escalating doses of PSA mRNA-transfected DCs were administered with no evidence of dose-limiting toxicity or adverse effects, including autoimmunity. Induction of PSA-specific T cell responses was consistently detected in all patients, suggesting in vivo bioactivity of the vaccine. Vaccination was further associated with a significant decrease in the log slope PSA in six of seven subjects; three patients that could be analyzed exhibited a transient molecular clearance of circulating tumor cells. The demonstration of vaccine safety, successful in vivo induction of PSA-specific immunity, and impact on surrogate clinical endpoints provides a scientific rationale for further clinical investigation of RNA-transfected DCs in the treatment of human cancer.

Figures

Figure 1

Incubation of immature DCs with PSA RNA provides a maturation stimulus. (a) Upregulation of CD83 expression by immature DCs in the presence of increasing PSA mRNA concentrations. (b–e) Immature DCs were incubated for 45 minutes in the presence of 20 μg/ml of PSA mRNA and analyzed after 48 hours by flow cytometry for expression of the DC-specific marker CD1a and the DC maturation marker CD83, respectively. (b) Staining with isotype immunoglobulins IgG2b-FITC and IgG1-PE (isotype control). (c) DCs incubated in the absence of PSA mRNA (untreated). (d) DCs incubated with PSA mRNA that had been treated for 1 hour at 37°C with 10 μg/ml RNase A (PSA-RNA + RNase A), and (e) DCs incubated in the presence of PSA mRNA. One representative experiment out of three independent experiments is presented.

Figure 2

In vivo induction of PSA-specific T cell responses. PBMCs obtained at baseline (PSA, KK pre) and after three vaccination cycles (PSA, KK post) using PSA RNA–transfected DCs were cultured overnight with PSA or kallikrein protein (5 μg/ml). The specific T cell frequencies in each of the evaluable patients treated at a low dose level (107 cells; patient 2, 3, and 5), medium dose level (3 × 107 cells; patient 6, 7, and 9), or high dose level (5 × 107 cells; patient 11 and 13) is expressed as the number of spot-forming cells per 5 × 105 PBMCs seeded in each well.

Figure 3

Functional properties of the in vivo–generated PSA-specific CTLs. PBMCs derived from study patients at baseline (white bars) or after immunotherapy (black bars) were stimulated twice with PSA RNA (a) or GFP RNA–transfected DCs (b). Using standard 51Cr microcytotoxicity assays, lytic activities against PSA RNA– or GFP RNA–transfected DC targets were calculated. (c) Experiments in which MHC class I– or CD8-specific mAb’s were used to inhibit recognition and lysis of PSA RNA-loaded DC targets. The addition of these mAb but not isotype control mAb (Control mulgG1) during incubation of effector and target cells (40:1 E/T ratio) resulted in significant inhibition of PSA-specific cytolytic activity, suggesting that the observed responses are predominately mediated by MHC class I–restricted CD8 CTL.

Figure 4

PSA velocities after vaccination with PSA RNA–transfected DCs. Pretherapy (pre slope) and post-treatment (post slope) serum PSA kinetics of study candidates vaccinated with PSA RNA–transfected DCs were calculated as described previously (13). A linear regression model was used to obtain estimates on the change of serum PSA over time, and the differences between pre- and post-therapy log slope estimates were computed for each patient. (a–c) The results on the serum PSA kinetics in all seven patients who were eligible for analysis before (pre) and after (post) vaccine therapy. In one of seven patients available for analysis (a), a decrease in the log slope PSA was calculated upon initiation of therapy. In five of the seven evaluable patients we could demonstrate a significant reduction of the log slope PSA (patient 3, 5, 7, 9, 10) (b), whereas one patient (patient 2) exhibited unchanged (c) PSA velocities after treatment with PSA RNA-pulsed DCs. (d) Pre- (white bars) and post-therapy (black bars) log slope PSA values for the entire group. Asterisk indicates statistically significant differences between the pre- and post-treatment slope.

Figure 5

Molecular clearance of circulating tumor cells after vaccination. Real-time PCR was used to quantitatively assess the kinetics of circulating tumor cells at baseline (week 0), during vaccination (week 2, 4, 6), and after vaccination (weeks 8, 10, 12) in three study patients using primers and probes specific for PSA (a) or the epithelial marker EpCAM (b). PSA or EpCAM mRNA was amplified from total RNA extracted from 107 PBMCs, and the corresponding copy numbers were quantitated and subsequently normalized to β-actin RNA copy numbers amplified from the same PBMC sample. To further improve the sensitivity of this assay, we also determined the average mRNA copy number of each marker amplified from PBMCs of ten age-matched healthy male volunteers (dotted lines). In three patients analyzed, vaccination with PSA RNA–transfected DCs led to a complete but temporary clearance of blood-borne circulating tumor cells, an effect that was not predicted by serum PSA levels.