Prostate Cancer

A review by Alam and McNeel on DNA vaccination against prostate cancer focused on preclinical models, recent clinical trial results, and efforts to improve vaccine potency . Described is the pTVG-HP vaccine that encodes the full-length human prostatic acid phosphatase (PAP), a protein whose expression is essentially restricted to normal and malignant prostate tissue. In 2009, the group published results of a dose-escalation trial . Twenty-two patients with stage D0 adenocarcinoma of the prostate were immunized intradermally (i.d.) with plasmid DNA (100, 500, or 1500 μg) and granulocyte–macrophage colony-stimulating factor (GM-CSF; 200 μg) coadministered as an adjuvant at 14-day intervals for six cycles. At the end of 24 weeks, no significant adverse events were reported and no PAP-specific antibodies were detected. PAP-specific IFN-γ-secreting CD8 ⁺ T cells (the primary immunological endpoint of the trial) were observed in 1 patient from each dosing cohort. Threefold (or greater) increases in PAP-specific CD4 ⁺ and CD8 ⁺ proliferative T cell responses were observed in 6 and 3 patients, respectively. No complete prostate-specific antigen (PSA) responses were observed, and no PSA values declined by greater than 50%. However, PSA doubling time, a sign of recurrent disease, significantly increased ( p =0.033) from a median 6.5 months pretreatment to 8.5 months on-treatment. The group concluded that the vaccine was safe and, based on increased PSA doubling times, may have slowed tumor growth rates. However, the detected immunological responses were of low magnitude.

In a subsequent 2010 publication with a view toward optimizing vaccination schedules for future clinical trial, the same group reported more detailed longitudinal immune analysis using cryopreserved samples from the original study . Specifically, this group sought to determine if immune responses required multiple immunizations and if responses were delayed, durable, and/or associated with changes in PSA doubling time. In addition, to determine if elicited responses could be further augmented, the trial was amended to permit additional booster immunizations in two subjects with stable disease who had demonstrated evidence of a PAP-specific CD8 ⁺ T cell response. These subjects were treated at monthly intervals with the same pTVG-HP (100 μg) vaccine, once again coadministered with GM-CSF (200 μg) as adjuvant. Although patient numbers were small, the results suggested that several immunizations were indeed necessary to elicit PAP-specific T cells. Responses tended to increase with the number of immunizations, with the majority (three out of four patients) of significant PAP-specific T cell responses detectable only after six immunizations. These observations indicated a benefit to continued immunizations, at least in the cancer setting, rather than induction of tolerance. Immune responses were durable in some patients but seemed to wane over time in others and could be augmented by booster immunizations. Durable responses were highly associated with a greater than twofold increase in PSA doubling time ( p =0.001). Of particular interest was the suggestion that a DNA vaccine targeting the PAP antigen could potentially be investigated in combination with Dendreon Corporation’s Provenge to maintain or augment long-term PAP-specific T cell immunity.

Melanoma

The glycoprotein gp100 is a differentiation antigen and a structural component of the melanosome expressed predominantly by melanoma cells and normal melanocytes. In 2009, Yuan and colleagues reported results from a phase I randomized crossover trial of plasmid DNA encoding xenogeneic (mouse) or human gp100 . Melanoma patients were injected (via the Biojector 2000 jet delivery device) intramuscularly (i.m.) once every 3 weeks for 9 weeks with plasmid (100, 500, or 1500 μg) containing either mouse or human DNA. After the first three injections, patients were then immunized three times with a similar injection containing DNA from the other species. Eighteen patients were evaluated immunologically at crossover and after the final immunization. The study was not powered to detect differences in progression-free survival. Instead, the investigators found no dose-limiting toxicities. The most common grade 1/2 toxicity was an injection site reaction in 12 patients. They reported epitope-specific CD8 ⁺ T cell responses in 5 patients. As for the pTVG-HP vaccine, the incidence (5/18 patients) of induced immunity was low.

Breast Cancer

In 2010, Norell and colleagues reported results from a pilot clinical trial of a full-length HER-2/neu vaccine in patients with metastatic breast cancer . HER-2/neu is a protooncogene overexpressed not only in breast cancer but also in a number of other malignancies, including ovarian, cervical, and renal carcinoma. Eight patients were given three cycles (4 weeks apart) of treatment with a plasmid construct encoding a full-length signaling-deficient version of HER-2/neu. Low doses of GM-CSF i.d. (at the injection site) and interleukin-2 (IL-2) subcutaneously (s.c.; into the abdominal area) were given daily for 3 days (starting 2 days prior to the immunization) or for 4 days (starting 24 hr after the immunization), respectively. In each cycle, the plasmid was administered both i.m. and i.d. (270 and 30 μg, respectively). Six patients finished the study. One patient was withdrawn after one cycle due to severe erysipelas at the location of a skin metastasis, and another patient withdrew due to disease progression. Acute toxicity, autoimmunity, or cardiotoxicity were not observed. Neither epitope-specific proliferative responses nor increased numbers of specific IFN-γ-producing T cells were observed 10 days after the final immunization. However, three patients exhibited dramatically increased numbers of IFN-γ-producing T cells as well as strong humoral responses at 22-month follow-up. The group concluded that the vaccine was safe, well tolerated, and could induce long-lasting cellular and humoral immune responses against HER-2/neu in approximately 50% of patients with advanced breast cancer.

Colorectal Cancer

Carcinoembryonic antigen (CEA) is another tumor-associated antigen that is overexpressed in breast, lung, gastric, pancreatic, ovarian, and colorectal carcinomas . Staff and colleagues reported phase I safety results in 10 patients with colorectal cancer . The group tested a plasmid DNA vaccine construct encoding a modified CEA gene fused to a promiscuous T helper epitope of tetanus toxoid. In this open-label study, plasmid was delivered (using the Biojector 2000 jet injection device) on weeks 0, 2, and 6 either i.d. (2 mg) or i.m. (8 mg). GM-CSF (150 μg) was used as adjuvant and delivered (on days −1, 0, 1, and 2, relative to the immunization) with a fine-needle syringe (i.d./s.c.) immediately following treatments. All patients received a single i.v. dose of cyclophosphamide (300 mg/m ² ) 3 days prior to the initial immunization to selectively reduce regulatory T cells numbers to augment induction of immune responses . The treatment was well tolerated, and no signs of autoimmunity were detected. Immunologic results were not reported.

Cervical Cancer

Most cervical cancers are causally associated with ‘‘high-risk’’ human papilloma virus (HPV) types, especially 16, 18, and closely related subtypes. Cervical cancer is an attractive target because the two HPV proteins, E6 and E7, are oncogenic and disrupt the cell cycle and are also non-self-proteins, thus precluding the issue of tolerance . Various aspects of this topic, including clinical updates on specific therapeutic DNA-based candidates, have been reviewed by Lin and colleagues and more broadly by Su and colleagues . In 2009, Trimble and colleagues reported on a phase I trial of an HPV DNA vaccine candidate expressing the HPV16 E7 protein, mutated at residues 24 and 26 to reduce the protein’s oncogenic potential and linked to both signaling sequences to allow protein secretion and to sequences coding HSP70 to improve immunogenicity . The open-label, dose-escalation study reported on 15 women given 0.5, 1.0 ( n =3 per group), or 3.0 mg ( n =9) by i.m. injection administered on the day of enrollment (day 0) and on weeks 4 and 8. Immunogenicity was measured on week 0 (preexisting), on week 15, and at 6 months (post-immunization). Histological analysis was performed on weeks 0 and 15. Dose-limiting toxicities were not observed. The vaccine did not induce specific antibody responses, nor did it boost preexisting ones. With use of overnight ELISPOT assay, baseline E7-specific IFN-γ T cell responses were detected in 3 patients, and at week 15 these were found to be increased in one, stable in another, and reduced in the third. Four patients without detectable preexisting immunity had measurable responses post-vaccination. None showed any increase in either E6 or E7 after vaccination. Six months after vaccination, responses to E7 were detected in 5 patients in the highest dose cohort, revealing a slower than expected kinetics of response to the immunization. Because responses were low frequency, a second round of testing was performed using a cycle of in vitro stimulation (IVS) (IL-2 added on days 3 and 7). Using IVS, baseline responses were now detected in 5 patients, and increased post-vaccination responses were observed in 8 (in the middle- and high-dose cohorts). Responses measured by either method were not sustainable. Complete histological regression in 3 patients (all in the highest dose cohort) was reported. Although not significant, the response rate was somewhat higher than what might have been expected in a similar, unvaccinated cohort. It is important to note that responses to viral proteins such as E7 should not have central tolerance compared to tumor-associated “self” antigens. Thus, it may be that DNA vaccination may induce unusual T cell immune responses and/or that formulations, adjuvants, etc. may still be suboptimal.

Multiple Indications

Cancer-testis antigens are proteins normally expressed only in germ cells of testis and abnormally in a wide variety of human cancers . In 2009, Gnjatic and colleagues reported on clinical and immune responses induced by a DNA vaccine against the full-length cancer-testis antigen NY-ESO-1 . Sixteen patients with advanced cancers (prostate, non-small-cell lung, or esophageal) were divided into three cohorts and immunized (Powderject particle-mediated delivery system) with doses ranging from 1 to 8 μg administered according to various schedules, including monthly (e.g., weeks 1, 5, and 9) and cluster dosing (e.g., days 1, 3, and 5 for 8 weeks). Only minimal side effects (erythema at the vaccination site) were noted. Clinical responses were limited to stable disease in 5 patients. Fifteen of 16 patients developed NY-ESO-1-specific T cell responses, detectable at various time points by ELISIPOT. All had CD4 ⁺ T cell responses, and 5 had CD8 ⁺ T cell responses. Two patients had positive DTH skin responses, one of which did not have any detectable T cell responses by ELISPOT. Preexisting antibody responses were detected in 1 patient and remained detectable for the trial period. Interestingly, depletion of regulatory T cells from negative cultures resulted in detectable antigen-specific CD4 ⁺ responses, suggesting that naturally occurring regulatory T cells were acting to suppress antigen-specific effector T cell responses induced by vaccination.

Summary

The previously discussed preliminary studies of single-antigen vaccines suggest that the approaches used were generally only weakly and transiently immunogenic. No data were provided on the persistent expression of plasmid DNA at the site of injection, but the incidence of immunity at time points distant from active immunization begs the question of persistent antigen expression continuing to augment immunity. Most of the vaccine studies cited previously, particularly those immunizing against “self” antigens, utilized full-length proteins or large sequences derived from full-length proteins.

Melanoma

Considering that use of plasmid DNA vaccines alone had shown only modest immune responses and little clinical efficacy in cancer patients, Dangoor and colleagues tried a different (“heterogeneous prime-boost”) approach . The group evaluated rising doses and varying regimens of a plasmid-based DNA (pDNA) immunization followed by booster immunizations with recombinant modified vaccinia virus Ankara (MVA), both expressing a string of seven epitopes from five melanoma antigens. A previous phase I study of the same approach in melanoma patients had demonstrated safety but inadequate immunogenicity . In the subsequent multicenter, nonrandomized, open-label, dose-escalation study, the dose was escalated (20-fold) and the dosing period extended (by increasing the interval between primary pDNA immunizations and using additional, less frequent MVA boosters) . The group tested the safety and cellular immunogenicity of the vaccination in 41 late-stage metastatic melanoma patients randomized into seven groups (5–8 patients per group) that differed with respect to dose and number of pDNA and MVA immunizations. Primary objectives were to assess tolerability and immune responses generated by the various regiments of primary and booster immunizations. Secondary endpoints included tumor response, time to progression, and survival, although group sizes were intended to assess the primary objectives.

The group reported a trend toward increasing toxicity (mainly injection site reactions) with rising dose. Subsequent immunizations were generally better tolerated. They concluded that the toxicity of the vaccine was low because the only significant toxicities observed were flu-like symptoms and injection site reactions. Importantly, despite high doses and extended immunization periods, no autoimmune events and no changes in double-stranded DNA-specific antibodies were observed. T cell responder rates were dose dependent, with specific cytotoxic T cell responses (to at least one epitope) detected in 71% of patients by evaluation with tetramer staining. IFN-γ responses were detected (by ELISPOT) in 32%. Eight patients showed clinical benefit, including 1 with partial response, 5 with stable disease, and 2 with mixed responses. Seven of these demonstrated specific cytotoxic immune responses, which were associated with a median 8-week increase in time to progression and a 71-week increase in survival compared to nonresponders. In summary, T cell responder rates were 50% with low-dose pDNA/MVA or MVA alone, rising to more than 90% with high-dose pDNA/MVA, representing a marked improvement in immunogenicity compared to the other DNA-based vaccine studies previously discussed.

Cervical Cancer

Like the E7 protein, the E6 protein of the high-risk HPV-16 serotype is also associated with disruption of the cell cycle and is oncogenic. Because both are required to maintain the transformed state, together they present an attractive therapeutic target . Such a product is Amolimogene bepiplasmid , an immunotherapeutic microparticle encapsulated DNA-based vaccine encoding T cell epitopes from E6 and E7 proteins of “high-risk” HPV types. It includes the complete HPV-encoding sequences contained within its precursor ZYC101, which was studied in an early phase I trial , as well as other regions encoding segments of HPV-16 and HPV-18 E6 and E7 viral proteins . In a 2004 multicenter, double-blind, randomized, placebo-controlled phase II trial in subjects with high-grade cervical neoplasia, it was found to be safe, although it was not significantly superior to placebo except in a predefined group of women younger than 25 years of age. In that group, disease resolution was significantly higher in the combined ZYC101-treated groups compared to placebo (70 vs. 23%, respectively; p =0.007) .

In 2011, Matijevic and colleagues reported on peripheral blood CD8 ⁺ T cell responses in subjects enrolled in the 2004 phase II clinical study . Based on MHC type and archived sample availability, peripheral blood mononuclear cells (PBMCs) from 26 patients from the phase II clinical trial were selected for analysis. These patients had received i.m. injections (100–200 μg) 3 weeks apart, and their PBMC samples, processed from blood obtained at baseline (week 0) and at 14 and 26 weeks later, were stored in freezing medium at −80°C. Eight patients had received the 100-μg dose, 13 had received 200 μg, and 5 had received placebo. Using a 3-week in vitro immunization and expansion assay system, enhanced (at least twofold greater than pretreatment) HPV-16- and HPV-18-specific CD8 ⁺ T cell responses were detected in 11 (42.3%) of the amolimogene-immunized patients at week 14. However, the group reported that in most cases, these responses were lower or undetectable at 26 weeks. Interestingly, they also demonstrated that T cell responses could be cross-reactive to epitopes from HPV-6 and -11.