The objective of this project is to develop novel treatments, using expression-targeted gene therapy and minicell technology, to replace current methodologies used in the clinic for the treatment of carcinomas, especially transitional cell carcinoma of the bladder. The expression-targeted gene therapy procedure involves cancer-specific DNA elements (promoters) to drive the expression of engineered suicide genes to induce apoptosis in cancer cells. Minicells, a kind of bacterial derivative , prevent tumor recurrence and growth through targeted toxicity and an induced immune response that is similar to that induced by Bacille Calmette-Guerin (BCG), but without the risk of infection due to lack of chromosomes. The osteopontin promoter (popn) was selected via currently accepted methods by comparing endogenous gene expression between normal and cancerous cells. The opn gene is expressed in far greater amounts in cancer cells, so it was reasoned that the opn promoter would be more active in cancer cells as well. Reporter constructs using popn were transfected into both cancerous and normal cell types, with maximum Popn-driven reporter intensity in the cancer cells showing up as strong (102.7%) compared to Pcmv-driven positive controls. Popn-driven reporter intensity was reduced by ~90% in the non-cancer cells. Further enhancements to targeting and expression were obtained through the incorporation of single-nucleotide polymorphisms (SNPs) in the promoter sequence. Further investigations to confirm a correlation between endogenous opn mRNA levels and Popn-driven reporter expression produced a surprising lack of correlation (R2=0.24). However, taking into account opn mRNA splicing variants, a strong negative correlation was determined between mRNA levels of the variant opn-a and Popn-driven transgene activity (R2=0.95). Three novel cancer-specific promoter pran, pbrms1 and pmcm5 were identified through a new screening logic. The activities of those promoters were verified to be much higher in the tested cancer cell lines than the current gold standard used to target gene expression to cancer cells: the promoter of human telomerase reverse transcriptase (phTERT). A constitutively active, apoptosis-inducing analog of caspase 3, referred to as Reverse Caspase3 (RevCasp3), was engineered via gene recombination and cloned into expression-targeted plasmid constructs. These constructs showed excellent activity in inducing apoptosis within the cancer cells tested. Moreover, Pran-RevCasp3 constructs were shown to have significant, cancer-specific killing action within both human and murine cell in vitro. The therapeutic effects of minicell constructs known as VAX-IP were tested within our orthotopic, murine model of transitional cell carcinoma of the bladder. In trials focused on the prevention of tumor growth and tumor implantation, bell-shaped curves were produced by data reflecting the relation between drug dose and tumor burden. The median and average bladder weights, used as a surrogate for tumor burden, decreased with increasing doses of VAX-IP minicells administered via intravesical, transurethral delivery. Activity was lost at high doses of VAX-IP minicells. Compared with the sham-treated group, 1x108 VAX-IP minicells, delivered at 24 hours post-surgery with repeated administrations every 7 days for a total of four treatments, yielded a significant survival advantage to the treated animals (P=0.03). / acase@tulane.edu

Background: Myrmecia pilosula (the jack jumper ant, JJA) is the principal cause of ant venom anaphylaxis in Australia. Whereas honeybee and wasp venom allergy can be treated by venom immunotherapy (VIT), no such treatment is available for ant sting allergy. In addition, information on the natural history of JJA sting allergy is required to identify those most likely to benefit from immunotherapy. The main objectives of this research were to establish: (i) the prevalence, natural history and determinants of reaction severity for JJA allergy, and; (ii) the efficacy and tolerability of JJA VIT. Methods: A search of the Royal Hobart Hospital (RHH) forensic register, a random telephone survey, and a review of emergency department (ED) presentations were performed. Three hundred eighty-eight JJA allergic volunteers were assessed, including serum venom-specific IgE RAST, and then followed up for accidental stings over a 4-year period. Finally, a randomised double-blind, placebo-controlled, crossover trial of JJA VIT was performed. Laboratory parameters measured during the trial were; leukocyte stimulation index (SI), IL-4 production, IgE RAST, histamine release test (HRT), leukotriene release test (LRT) and basophil activation test (BAT). Intradermal venom skin testing (VST) was also performed at trial entry. Findings: The prevalence of JJA sting allergy was 2.7% in the Tasmanian population, compared to 1.4% for honeybee. People aged 35 or older had a greater risk of both sting allergy and hypotensive reactions. Four deaths were identified, all in adults with significant comorbidities. During follow-up, 79 (70%) of 113 accidental jack jumper stings caused systemic reactions. Only prior worst reaction severity predicted the severity of follow-up reactions, with the majority of people experiencing similar or less severe reactions when stung again. Sixty-eight otherwise healthy JJA allergic adult volunteers were enrolled in the clinical trial. Systemic reactions to therapy were recorded in 34% during VIT. Objectively defined systemic reactions to sting challenges arose in 1/35 after VIT (mild self-limiting urticaria only) versus 21/29 in the placebo group. Treatment with oxygen, intravenous adrenaline infusion and volume resuscitation was effective and well tolerated. Hypotension was always accompanied by a relative bradycardia, which was severe and treated with atropine in two patients. In the placebo group, only VST and HRT were predictive of sting challenge results. Although IgE RAST, leukocyte SI and IL-4 production, LRT and BAT all correlated well with VST, they did not predict sting challenge outcome. After successful VIT, venom-induced leukocyte IL-4 production tended to fall, whereas IgE RAST increased and a natural decline in HRT reactivity was reversed. Interpretation: VIT is highly effective in prevention of JJA sting anaphylaxis and is likely to be of most benefit to people with a history of severe systemic reactions, which usually occur in people aged over 35. Neurocardiogenic mechanisms &/or direct cardiac effects may be important factors in some anaphylaxis deaths. Systemic reactions to immunotherapy are common and require immediate access to resuscitation facilities. The HRT warrants further investigation as a test for selecting those most likely to benefit from VIT. None of the tests evaluated appear to be reliable markers of successful VIT.

<p>Bacillus Calmette Guérin (BCG), or attenuated Mycobacterium bovis, is the gold standard of immunotherapy in the clinic to treat superficial bladder cancer. However, setbacks remain due to a high recurrence rate, side effects, and BCG-refractory disease. In this thesis, we explored the use of novel immunotherapeutic agents such as CpG oligodeoxynucleotides (CpG ODNs) or synthetic ODNs containing unmethylated CpG dinucleotides. Since unmethylated CpG motifs are predominant in bacterial but not vertebrate DNA, they function as a “danger signal” leading to a potent immune response.</p><p>To be able to test various immunotherapeutic agents, we optimized subcutaneous (s.c.), metastatic, and orthotopic models using the murine bladder-49 (MB49) cancer cell line. In the orthotopic model, we show that poly-L-lysine promotes MB49 attachment to the bladder leading to 100% tumor take. In addition, Clorpactin (sodium oxychlorosene) potently enhances adenoviral transduction in the bladder.</p><p>Utilizing the MB49 model, we compare CpG ODNs with BCG and demonstrate the increased efficacy of CpG ODNs which could cure both s.c. and aggressive orthotopic bladder cancer. In our model, type B ODNs were most optimal and the antitumor response required T cells in order to induce regression and tumor-specific immunity. We also combined CpG ODNs with adenoviral vectors (Ad) expressing the immunostimulatory molecules CD40L, TRANCE, lymphotactin, IL2 or IL15. However, we show that CpG ODNs are effective as a monotherapy and adenoviral vectors did not enhance the effect.</p><p>AdCD40L was also used to genetically modify human dendritic cells (DCs). AdCD40L-transduced DCs not only had a higher and prolonged expression of the Th1 cytokine IL12 compared to TNFα-matured DCs, but CD40L-activated DCs could also resist the suppressive effects of IL10 and TGFβ. Since TNFα is commonly used in clinical DC vaccination protocols and because tumors often secrete immunosuppressive cytokines, these data have important implications for optimizing cancer immunotherapy.</p>

During homeostatic expansion, peripheral T cells proliferate in response to lymphopenia, in the absence of cognate antigen or costimulatory signals. Host CD8 reconstitution following hematopoietic cell transplantation (HCT) involves the de novo-generation of T cells in addition to the homeostatic expansion of mature donor T cells present in the graft and donor lymphocyte infusion as well as host T cells that survive conditioning. Although it is well appreciated that CD8 homeostatic expansion contributes to host CD8 reconstitution following HCT, the factors governing the extent of the contribution by donor CD8 homeostatic expansion to host reconstitution have not been precisely and systematically examined. The ex vivo generation of memory CD8 T cells specific for an epitope of the immunodominant minor histocompatibility antigen H60 was demonstrated. By adapting a previously described culture system designed to generate large numbers of transgenic memory CD8 T cells, memory CD8 cells (TM) specific for a physiological antigen were elicited from a heterogeneous population of CD8 T cells. The ex vivo-generated antigen-specific memory CD8 cells were then expanded under conditions of lymphopenia in ablatively conditioned syngeneic transplant recipients and found to persist greater than 2 months post-transplant. These findings support the notion that the transplantation of small numbers ex vivo-generated memory CD8 cells, specific for a physiologically relevant antigen, can help restore host immune function following HCT. Transplant conditions were found to modulate the contribution of homeostatically expanded donor CD8 TM to the host CD8 compartment. Varying the conditioning intensity, timing of infusion, and infusion dose affected the kinetics of expansion as well as the homeostatic set-point. In my HCT model, delayed infusions of 3 weeks and transplantation of small numbers of donor CD8 T cells resulted in significant contributions to host CD8 T cell compartment. These findings could be applied clinically to enhance the effectiveness of clinical immunotherapy in restoring host immune function post-transplant. Moreover, transplantation of varying doses of donor CD8 TM demonstrated there was maximal donor contribution to host CD8 reconstitution. Finally, the homeostatic expansion, persistence, and function of transgenic memory and naïve CD8 populations were investigated following transplantation into ablatively conditioned syngeneic recipients. Both donor CD8 populations underwent 2 weeks of expansion post-transplant. The naïve CD8 population reproducibly achieved higher homeostatic numbers than the memory CD8 population. By one month post-HCT, the transplanted naïve CD8 populations also exhibited a memory CD8 phenotype. Moreover, the CD8 TN→TM population mediated an effective response to a primary challenge, comparable to the antigen-experienced memory CD8 population. Thus, regardless of the state of differentiation of the donor CD8 population at the time of transplant, donor CD8 homeostatic expansion contributes a memory CD8 cells population, able to mediating effective immune responses, to host reconstitution.

Activation of immunity to self-antigens is the goal in cancer immunotherapy, whereas blocking such responses is the goal in autoimmune disease. Thus, it is not surprising that investigation into cancer immunotherapy might also produce insights for the treatment of autoimmune disease. Heat shock protein, gp96, based therapies lead to the robust activation of CD8+ cytotoxic T cells that can slow tumor growth in 60-70% of mice, but only lead to the elimination of tumors in 30-40% of animals. The primary goal of the current studies was to understand why vaccination with a secreted gp96 vaccine was not efficacious in a larger proportion of animals, and identify combination therapies that enhanced the anti-tumor activity of gp96-Ig vaccination. It was found that in mice bearing established tumors, some mice responded well to vaccination with gp96-Ig, and that the induction of CD8+ T cells was found to correlate with tumor rejection; indicating that the proportion of mice that failed to reject tumors had established mechanisms of tumor-mediated suppression of anti-tumor immunity. The mechanism of this suppression was found to differ between various tumor models, so combination therapy sought to amplify CD8+ T cell responses directly, rather than by indirectly inhibiting suppressive factors induced by established tumors. It was found that antibody-based therapies leading to the stimulation of TNFRSF25, a powerful T cell co-stimulatory receptor, caused synergistic expansion of tumor-specific T cells when given in combination with gp96-Ig vaccination and led to enhanced rejection of multiple tumor types. Interestingly, TNFRSF25 agonistic antibodies were also found to directly stimulate the proliferation of natural CD4+FoxP3+ regulatory T cells. This activity was found to be beneficial in the prevention of allergic lung inflammation when administered prior to antigen challenge. These studies have therefore identified the conditions required for successful tumor elimination following gp96-Ig vaccination, and discovered that a TNFRSF25 agonistic antibody may be used to enhance anti-tumor immunity induced by gp96-Ig. These studies have also identified TNFRSF25 stimulation as the most powerful, and physiologically relevant, method to selectively induce Treg proliferation in vivo ever discovered, with important consequences for the treatment of autoimmune inflammation.

Thesis (Ph. D.)--University of Washington, 2006. / Vita. Includes bibliographical references (leaves 122-135).

Bacillus Calmette Guérin (BCG), or attenuated Mycobacterium bovis, is the gold standard of immunotherapy in the clinic to treat superficial bladder cancer. However, setbacks remain due to a high recurrence rate, side effects, and BCG-refractory disease. In this thesis, we explored the use of novel immunotherapeutic agents such as CpG oligodeoxynucleotides (CpG ODNs) or synthetic ODNs containing unmethylated CpG dinucleotides. Since unmethylated CpG motifs are predominant in bacterial but not vertebrate DNA, they function as a “danger signal” leading to a potent immune response. To be able to test various immunotherapeutic agents, we optimized subcutaneous (s.c.), metastatic, and orthotopic models using the murine bladder-49 (MB49) cancer cell line. In the orthotopic model, we show that poly-L-lysine promotes MB49 attachment to the bladder leading to 100% tumor take. In addition, Clorpactin (sodium oxychlorosene) potently enhances adenoviral transduction in the bladder. Utilizing the MB49 model, we compare CpG ODNs with BCG and demonstrate the increased efficacy of CpG ODNs which could cure both s.c. and aggressive orthotopic bladder cancer. In our model, type B ODNs were most optimal and the antitumor response required T cells in order to induce regression and tumor-specific immunity. We also combined CpG ODNs with adenoviral vectors (Ad) expressing the immunostimulatory molecules CD40L, TRANCE, lymphotactin, IL2 or IL15. However, we show that CpG ODNs are effective as a monotherapy and adenoviral vectors did not enhance the effect. AdCD40L was also used to genetically modify human dendritic cells (DCs). AdCD40L-transduced DCs not only had a higher and prolonged expression of the Th1 cytokine IL12 compared to TNFα-matured DCs, but CD40L-activated DCs could also resist the suppressive effects of IL10 and TGFβ. Since TNFα is commonly used in clinical DC vaccination protocols and because tumors often secrete immunosuppressive cytokines, these data have important implications for optimizing cancer immunotherapy.

The work presented herein focuses on the activation of the adaptive immune system in order to develop T cell-based immunotherapy for viral infections and cancer. The main goal was to identify and activate viral or tumoral antigen-specific T cells by using different identification, isolation and stimulation techniques. One such approach was that we modified dendritic cells (DCs) with an adenoviral vector encoding the full length pp65 antigen from cytomegalovirus (CMV). Through strategic modification techniques we demonstrate that it is possible to obtain DCs presenting antigen-specific peptides both on major histocompatibility complex (MHC) class I and MHC class II molecules for simultaneous CD8+ and CD4+ T cell activation. We also demonstrate that it is possible to generate prostate antigen-specific CD8+ T cells from a naïve repertoire of T cells by using DCs and HLA-A2-restricted peptides derived from prostate tumor-associated antigens or by using an adenoviral vector encoding the full length prostate tumor-associated antigen STEAP. We further demonstrate that CD8+ T cells directed against several prostate-specific peptide epitopes can be found in peripheral blood and in the prostate tumor area of prostate cancer patients. Furthermore, we have characterized a number of prostate-derived cell lines in terms of HLA haplotype and tumor-association antigen expression. We concluded that our methods for generating T cells restricted to CMV antigen have the ability to be applied for adoptive T cell transfer to patients with CMV disease and that prostate antigen-specific T cells can be found within prostate cancer patients, which enables future development of immunotherapeutic strategies for prostate cancer.

This thesis comprises studies on murine and human models of bladder carcinoma with the aim to develop novel immunogene therapies. On the basis of the results presented in this thesis, a clinical trial is underway. The potential of activating the immune system to combat cancer has long intrigued immunologists. Research has now been intensified and clinically effective treatments are beginning to materialize. We evaluated the induction of anti-tumor responses by inserting immunomodulating genes into tumor cells with adenovectors. Human biopsies and cell lines were positive for adenovirus attachment receptors, and cell lines were easily transduced. CD40L modified cells efficiently induced maturation of dendritic cell (DC). Phenotypical changes of AdCD40L transduced cells, such as increased apoptotic rate, upregulated MHC-I, Fas and TNFR may further strengthen the anti-tumor response. CD40L modified murine bladder cancer cells activated systemic immunity upon vaccination and in situ injections of AdCD40L inhibited tumor progression. Cytotoxic assays revealed the presence of cytotoxic T cells (CTLs) in vaccinated mice. Many tumors have developed ways to evade the immune system. Bladder carcinoma is associated with immune escape mechanisms like IL10 production. We demonstrated that immunosuppression by IL10 inhibited CTL function and that IL10 suppression may be reverted by AdCD40L therapy. In conclusion, AdCD40L therapy induces systemic immunity and inhibits tumor progression in murine models. The immunological mechanisms involve maturation of nearby DCs and CTL induction. AdCD40L therapy is effective despite immune escape mechanisms, e.g. IL10 secretion. The thesis argues for using AdCD40L immunogene therapy as a treatment of bladder carcinoma.

A foundation for the control of tumors is presented, based upon the formulation of a realistic, knowledge-based mathematical model of the interaction between tumor cells and the immune system. The parametric control variables relevant to the latest experimental data, e.g., the sigmoidal dose-response relationship and Michaelis-Menten dynamics, are also considered. The model consists of 12 states, each composed of first-order, nonlinear differential equations based on cellular kinetics and each of which can be modeled bilinearly. In recent years a great deal of clinical progress has been achieved in the use of optimal controls to improve cancer therapy patient care. For this study, a cancer immunotherapy problem is investigated in which the aim is to minimize the tumor burden at the end of the treatment period, while penalizing excessive administration of interleukin-2 as a limit of toxicity. The optimal solution developed for this investigation is a mixture of an initially large dose of interleukin-2, followed by a gradually decreased dosage and a continuing infusion to maintain the tumor cell population at its allowable limit. Sensitivity analysis is applied to an investigation of the influences of system parameters. It has been found that the immune system is influenced greatly by several parameters such as macrophage level, tumor killing rate, tumor growth rate, and IL-2 level. The simulation results suggest that parametric control variables are important in the destruction of tumors and that the application of exacerbation theory is a good method of tumor control. / Graduation date: 1991