Objective
The purpose of this study was to explore the potential of toll-like receptor-3 stimulation, with polyI:C 12 U (poly[l].poly[C 12 ,U]; rintatolimod [Ampligen; Hemispherx Biopharma, Philadelphia, PA]) to enhance bioactivity of cancer immunotherapies.
Study Design
Several models of immune activation were assessed with polyI:C 12 U at concentrations that were achieved clinically. Dendritic cell maturation and antigen-specific immune responses were evaluated in vitro and in a murine model. The potential for polyI:C 12 U to enhance antibody-dependent cellular cytotoxicity against tumor was also evaluated.
Results
Dendritic cells are matured and T-cell stimulation is enhanced in the presence of polyI:C 12 U. In addition, polyI:C 12 U induced the release of proinflammatory chemokines and cytokines. Prostate-specific antigen-specific T-cell and antibody responses were enhanced significantly in a BALB/c prostate-specific antigen transgenic mouse model. Finally, rituximab-mediated antibody-dependent cellular cytotoxicity against tumor targets was improved significantly by the addition of polyI:C 12 U.
Conclusion
PolyI:C 12 U shows promise as a potential agent for selective enhancement of effect with currently available and future cancer immunotherapies.
Immunotherapy of cancer holds the promise of disease-specific intervention without the toxicity that is associated with traditional therapeutic modalities. This category of treatment theoretically is ideal to supplement current standards of care to improve outcomes. Unfortunately, definitive demonstration of therapeutic benefit for most immunotherapeutic strategies is still pending. Obstacles to success include our natural immune regulatory processes and tumor escape mechanisms that have yet to be defined fully and surmounted. We recently reported the outcome of an advanced ovarian cancer mono-immunotherapy phase III study in a population that is expected to mobilize induced immunity successfully. Adequate tumor-specific immune induction in the maintenance setting was not generated to have a clinical effect. Interestingly, we also reported results of a phase II study, with the same agent, that demonstrated immune adjuvant effects of concomitant chemotherapy, with a suggestion that the schedule of chemoimmunotherapy was critical to the quality of the induced immune response. These paired observations along with recent advances in the appreciation of immune regulation and of adjuvant strategies prompted us to revisit approaches to immune mobilization in some preclinical model systems.
The immune response to microbial infection includes an important series of related receptor-signal transduction pathways that are known as the toll-like receptor (TLR) family. The appreciation of the toll-like receptor systems have only recently been fully elucidated. These receptors serve as principal mediators of innate immune response to infection and serve to enhance subsequent adaptive immunity that includes antibody-specific and cellular immune responses. Different TLRs respond to different microbial components (such as lipopolysaccharide, peptidoglycan-associated lipoprotein, cytosine-guanine–rich bacterial DNA (CPG), and intracellular RNA species that are associated with viral replication). Classic adjuvants that are used traditionally in research immunology contain microbial extracts and can activate multiple TLRs. Although potent in stimulating adaptive immunity, toxicity such as large painful inflammatory lesions at injection sites have prevented clinical use in humans. TLRs are expressed on antigen-processing cells of the immune system and include both intracellular and cell-surface elements. Two predominant intracellular signaling pathways MyD88-dependent and MyD88-independent have been characterized, which has led to signal transduction and gene transcription through IRF3 and NF-κB, respectively. Most microbial TLR agonists activate the MyD88 pathway by way of TLR2, TLR4, TLR5, and TLR9. TLR4 is the lipopolysaccharide receptor, and TLR9 is the CpG receptor. Exposure to these agents is also associated with toxicity. Polyribosinic:polyribocytidic acid (polyI:C) is a double-stranded RNA molecule that stimulates the TLR3 pathway. This molecule was developed originally as an interferon inducer before the TLR pathways were characterized and is also associated with clinical toxicity that includes nephropathy and coagulopathy. To reduce toxicity, a modified form of polyI:C, polyI:C 12 U (rintatolimod, polyI:polyC 12 U, Ampligen; Hemispherx Biopharma, Philadelphia, PA), was developed to substitute uridine for cytosine at a ratio of 1:12 in the enzymatic synthesis of the single-stranded polyribocytidic strand. This product is metabolized more quickly than polyI:C and has greater specificity for the MyD88 independent pathway. The compound recently has been studied in hundreds of patients as an experimental therapy for chronic fatigue syndrome and is currently the subject of research as an adjuvant for enhancing efficacy of flu vaccines. Based on the extensive body of clinical experience and apparent adjuvant properties, we elected to evaluate it for its ability to enhance bioactivity of putative cancer immunotherapies.
Oregovomab alters antigen uptake and processing of cancer antigen 125 (CA125) by dendritic cells and enhances specific immunity to the tumor antigen. The complexity of the CA125 genomic structure has precluded development of relevant CA125 transgenic models to date; notably, oregovomab does not mediate antibody-dependent cellular cytotoxicity in human effector cells. The immune effector pathways likely to be used by cancer immunotherapies more generally include dendritic cell-mediated antigen presentation, CD4 and CD8-specific T-cell activation, and antibody-dependent cellular cytotoxicity. The direct targeting antibody rituximab is a prime example of cancer immunotherapy that is mediated by antibody-dependent cellular cytotoxicity. We assessed the activity of polyI:C 12 U in human in vitro cell assays that model these effector pathways and also in a prostate-specific antigen transgenic mouse model that is characterized by murine tolerance to the human prostate-specific antigen transgene. These studies, in conjunction with our previous observations regarding the important interaction of immunotherapy with chemotherapy, suggest that combining chemotherapy with cancer immunotherapy and TLR stimulation should be explored, with more specific modeling to be tailored to the specific antigen targets that are under clinical evaluation.
Materials and Methods
Generation of dendritic cells
Dendritic cells were generated by culturing human peripheral blood mononuclear cells in the presence of supplemental interleukin-4 (IL4) and granulocyte macrophage stimulating factor (GMCSF) for 1 week, as previously described. The Stanford Institutional Review Board approved this research study (IRB protocol number 13063). The activation state of cells from the primary cell culture and 24 hours after the addition of the maturation cocktail (tumor necrosis factor–α/interferon-α [TNF-α/IFN-α]; 10 ng/mL, 50 U/mL) or polyI:C 12 U (25 μg/mL) was assessed by expression of multiple dendritic cell markers. Expression was analyzed by staining with labeled antibodies (HLA-DR-FITC [clone HA58] and CD32-PE [clone FLI8.6]; BD Biosciences Pharmingen, San Diego, CA; CD80-FITC [clone 2D10], CD86-FITC [clone IT2.2], CD40-FITC [clone HB14], and HLA-ABC-PE/Cy5 [clone W6/32]; BioLegend, San Diego, CA) followed by flow cytometry. Dendritic cells preferentially uptake antigen when they are immature and present antigen to T-cells when they have matured.
Cytokine profile
Supernatant from a dendritic cell culture that had been matured with TNF-α/IFN-α or polyI:C 12 U were compared in a Multiplex Assay (Biorad, Hercules, CA) that was specific for IL-4, -6, -10, and -12(p70), IFN-γ, macrophage chemoattractant protein-1 (MCP-1), macrophage inflammatory protein 1-alpha (MIP-1α), and TNF-α. This panel reflects T-cell modulating cytokines and chemokines that are expected to influence the character of subsequent T-cell responses.
Antigen-specific T-cell response
An antigen presentation assay that had been adapted from a method reported previously was used to assess the impact of polyI:C 12 U on T-cell responses to the tumor antigen CA125 in the presence or absence of oregovomab, which is the monoclonal antibody that is specific for CA125. Oregovomab was used previously to augment cross-presentation that led to enriched CD4 + and CD8 + specific T-cell immunity against tumor antigen in patients with ovarian cancer. Briefly, human peripheral blood lymphocytes were exposed to 2 rounds of in vitro sensitization by autologous cultured dendritic cells in the presence of varying concentrations of CA125 and antibody (oregovomab). At the third round of in vitro sensitization, polyI:C 12 U (25 μg/mL) and lymphocyte-depleted peripheral blood mononuclear cells were added to selected cultures. After the third round of sensitization, the T cells were treated with brefeldin A to prevent the secretion of synthesized cytokines. The next day, cells were harvested and stained extracellularly with anti-CD3-FITC and anti-CD8-phycoerithrin/Cy5. Cells were then washed, fixed, permeabilized, stained intracellularly with anti–IFN-γ-phycoerithrin and analyzed by flow cytometry. The CD8+ IFN-γ–producing population was defined as CD3+, CD8+, and IFN-γ+; the CD4+ IFN-γ–producing population was defined as CD3+, CD8-, and IFN-γ+.
Prostate-specific antigen transgenic mouse immunization
In a protocol that was approved by the local institutional animal welfare committee, mice were vaccinated subcutaneously at weeks 1, 3, 5, and 7 with 2 μg purified prostate-specific antigen (Maine Biotechnology Services, Inc, Portland, ME) in the presence or absence of a human chimeric antiprostate-specific antibody (10 μg) and polyI:C 12 U (50 μg). Murine antibody response to prostate-specific antigen was assessed at weeks 0, 4, and 8. T-cell responses to prostate-specific antigen were assessed by T-cell proliferation in response to prostate-specific antigen from splenocytes that were harvested at week 8.
Assessment of antibody-dependent cellular cytotoxity
Antibody-dependent cellular cytotoxity activity was determined by standard 4-hour 51 Cr-release assay. 51 Cr-labeled target cells (10 6 SUDHL4 cells/mL) were incubated with media alone or in the presence of rituximab (10 μg/mL; MabThera, Roche, Italy) at 37°C for 30 minutes. Unbound antibody was washed off, and the cells were plated at 5000 cells/well. PolyI:C 12 U was added to the wells at the indicated concentrations. Effector cells (peripheral blood mononuclear cells from healthy donors) were then added to the plates at indicated effector-to-target ratios. After a 4-hour incubation, supernatants were removed and measured with the use of a gamma counter (Wallac Microbeta 1450 Liquid Scintillation and Luminescence Counter; PerkinElmer, Waltham, MA). All groups were studied in triplicate. The percentage of specific cell lysis was determined by the following equation: % lysis = 100 × (ER-SR)/(MR-SR), where ER = experimental release, SR = spontaneous release, and MR = maximal release. Spontaneous release and maximum release were determined from wells that contained target cells that had been incubated in medium alone or in 1% Triton X-100, respectively.
Statistical methods
In all graphs, error bars represent standard deviations. When possible, probability values were evaluated by the 2-tailed Student t test.
Results
Direct effect on dendritic cells in vitro
To determine how polyI:C 12 U treatment affects human peripheral blood mononuclear cells that are derived myeloid dendritic cells, we examined the expression of a number of dendritic cell markers on polyI:C 12 U-treated dendritic cells, compared with immature or mature dendritic cells. On maturation, dendritic cells typically down-regulate surface expression of Fc receptors (CD64 and CD32) and up-regulate both MHC class I (HLA-ABC) and MHC class II (HLA-DR) molecules and mediators of secondary signals (CD86, CD80, and CD40), which corresponds to the transition from antigen uptake to antigen presentation in an inflammatory reaction. Figure 1 , A shows the change in surface marker intensity with maturation relative to TNF-α/IFN-α (10 ng/mL, 50 U/mL) and the negative media control for a range of polyI:C 12 U concentrations. For the most part, human peripheral blood mononuclear cells that were derived dendritic cells responded similarly when exposed to the standard cocktail of TNF-α/IFN-α or to polyI:C 12 U. For example, the up-regulation of CD86 and MHC molecules, which are central to antigen presentation and immune-cell activation, are comparable. Likewise, CD40, which is important for dendritic cell maturation and the differentiation of T helper type 2 (T h 2) cells, is up-regulated similarly on TNF-α/IFN-α– and polyI:C 12 U-treated cells. With respect to the FcγRs, the expression of FcγRI (CD64) is unchanged by either treatment, but the expression of FcγRIIa (CD32) appears to be down-regulated, in particular, by polyI:C 12 U treatment. Also, CD83, which is a marker of dendritic cell maturation, is more affected by treatment with TNF-α/IFN-α than with polyI:C 12 U. Thus, signals with polyI:C 12 U are in large part analogous to those of maturation signals; however, there are several differences, which suggest that polyI:C 12 U and TNF-α/IFN-α signals may result in unique changes to dendritic cell function.
