BCG therapy is a clinically successful therapy in the treatment of bladder cancer but its acceptance is hampered by hazards and side effects related to the use of viable mycobacteria. The aim of this study was to evaluate the anti-tumor potential of the well-defined, non-viable mycobacterial antigen PstS1.
To this end we first tested the potential of recombinant PstS1 to activate human PBMC cultures. In this series of experiments PstS1 induced strong cytotoxicity, proliferation and IFN-γ production in human PBMC. Ten μg/ml PstS1 were found to be optimal for activation of human PBMC. We used BCG lyophilisate as a reference stimulus in these experiments as we have previously described BCG as a potent activator of human PBMC functions [
27]. PstS1 followed a similar time kinetic as BCG in the stimulation of human PBMC, albeit activation of PBMC functions was somewhat lower with PstS1 compared with BCG. PstS1 was previously shown to induce profound cellular immunity in different vaccination studies [
28‐
30]. Therefore, our results further confirm the potential of PstS1 to function as a potent inducer of T-helper 1 and cell-mediated immune responses.
While IFN-γ is a key cytokine in TH1 immune responses produced by activated lymphocytes, IL-12 is mainly produced by monocytes/macrophages as well as dendritic cells and important in the early phase of cellular immunity [
31]. IL-12 is a potent inducer of anti-tumor immunity [
32] and acts together with IFN-γ and TNF-α in a positive feedback loop [
25,
33,
34]. Although BCG mycobacteria are relatively weak inducers of bioactive IL-12p70 in human dendritic cells in the absence of additional costimulation like CD154 or IFN-γ (figure
3b), IL-12 has been shown to be essential for therapeutic efficacy in experimental BCG immunotherapy of bladder cancer [
24]. Given the importance of dendritic cells for the initiation of anti-tumor immune responses [
35] and as a source of IL-12 and TNF-α we tested the capacity of recombinant PstS1 to activate human monocyte-derived dendritic cells. Our results clearly identified PstS1 as a potent stimulator of human DCs as it provoked upregulation of CD83 and CD86 surface expression as well as induction of cytokines. Of interest is the expression of high amounts of bioactive IL-12 after stimulation of DCs with PstS1 as only small amounts of this cytokine are produced after challenge with BCG or other mycobacterial antigens (figure
3 and unpublished observations).
The encouraging
in vitro experiments then prompted us to evaluate the immunotherapeutic potential of PstS1 in a well-established murine model of experimental bladder cancer [
20]. The effect of prior sensitization or responsiveness to mycobacterial antigens for the effectiveness and outcome of subsequent immunotherapy with BCG is still a matter of intensive debate [
21‐
23]. Taking this into account we took advantage of the fact that we had in hand a well-defined, recombinant antigen already evaluated in vaccination studies against mycobacterial infections and included a prime-boost treatment regimen into our experimental immunotherapy protocol. The main objective of our series of
in vivo experiments was to determine the anti-tumor potential of local instillations of recombinant PstS1 and to compare the effect of intravesical PstS1 in sensitized and non-sensitized mice. Because adsorption of PstS1 to L-particles has previously been described as very efficient in inducing specific cellular immunity to this antigen [
29], we adopted this method of sensitization for our treatment protocol. Using a protocol of four weekly instillations of PstS1 into the bladder (adopted from the treatment schedule in use for immunotherapy with BCG) we observed a strong therapeutic effect of PstS1 instillations. Intravesical PstS1 significantly prolonged survival of mice and induced systemic immune responses and cellular infiltration of the bladder with different subpopulations of leukocytes. The anti-tumor effect of PstS1 was already evident after intravesical instillation of PstS1 only but was further enhanced after s.c. injection of empty L-particles (Fig.
5). Unexpectedly, sensitization of mice with PstS1-loaded particles almost completely abrogated the therapeutic effect of intravesical PstS1 (Fig.
4). In order to test whether the negative effect of prior sensitization was specific for PstS1 or could also be induced by s.c. injection of L-particles loaded with an irrelevant antigen, we compared the therapeutic effect of intravesical PstS1 combined with a) s.c. sensitization with BSA-loaded L-particles, b) s.c. injection of empty L-particles and c) s.c. injection of PBS only (Fig.
5). Clearly, the injection of BSA-loaded particles did not influence the therapeutic effect of intravesical PstS1 indicating that only PstS1-specific previous sensitization is detrimental (Fig.
5a and
5c). We analyzed the systemic serum antibody response, activation of splenocytes and the local cellular infiltration of the bladder wall in sensitized and non-sensitized mice (Fig.
6) in order to understand and explain the remarkable therapeutic potential of intravesical PstS1 as well as the negative effect of prior specific sensitization. Interestingly, even without sensitization, instillation of PstS1 into the bladder provoked a systemic anti-PstS1 response visualized by murine anti-PstS1 serum antibodies and splenocyte proliferation after rechallenge
in vitro. In addition to this systemic immune activation, intravesical PstS1 also induced the local influx of lymphocytes, macrophages and granulocytes into the bladder. While only a minimal influx of CD8-cells could be observed after instillation of PstS1, a considerable infiltration of the bladder with granulocytes, macrophages and CD4-cells was noted. Although prior specific sensitization with PstS1 completely abrogated the anti-tumor effect of intravesical PstS1, the cellular infiltration of the bladder wall remained essentially unchanged when comparing sensitized and non-sensitized animals. In the splenocyte restimulation assay an enhanced response of two out of four mice was noted in the group of mice which received s.c. PstS1 followed by intravesical PstS1 compared with the two groups which received either treatment alone (Fig.
6). This indicates that the combination of s.c. sensitization and intravesical treatment indeed augmented the systemic immune response at least in some animals. Surprisingly, this enhanced systemic immune response coincided with an abrogation of tumor-therapeutic efficacy. A possible explanation for this phenomenon could be that the anti-PstS1 antibodies, which were induced after s.c. priming would bind to the recombinant PstS1 shortly after instillation into the bladder and thereby neutralize its function and anti-tumor effect. However, as mentioned earlier, even in the PstS1-specific prime-boost treatment regimen an unchanged cellular infiltration of the bladder wall was noted. Alternatively, the sensitization might render the subsequent local immunotherapeutic immune response insufficient, because the host immune system is still actively responding to the priming at the onset of immunotherapy. This situation might be called "immune exhaustion" making the host unable to mount a sufficient local anti-tumor immune response while still responding to the specific priming. Further experiments are needed to clarify whether a modification of the time schedule of such a "prime-immunotherapy" protocol could prevent the negative effects observed in this study and might even enhance immunotherapy with PstS1.