Third International Conference on Cancer Vaccines/Adjuvants/Delivery for the Next Decade (CVADD 2009)

Abstract

The Third International Conference on Cancer Vaccines/Adjuvants/Delivery for the Next Decade (CVADD), chaired by Malcolm S Mitchell, was hosted in Dublin, Ireland, on 11–13 November 2009. The conference was intended to aid the translation of basic research into clinical progress and to bridge the gap between academia and industrial innovation. In total, 44 speakers gave presentations on preclinical and clinical vaccine development, mostly focusing on cancer vaccination. The present report does not represent a comprehensive review of all topics covered, but highlights selected points of particular novelty and interest. Within the field of cancer vaccines, the considerable discrepancy between high immune response rates and limited clinical effects has led to increased focus on how vaccines may best be applied to maximize their clinical impact. This challenge emerged at the forefront of the conference, reflected by fruitful discussions on adjuvants, delivery systems, strategies for countering tumor tolerance and on combination with conventional therapy. The studies presented at CVADD 2009 support the argument for combining active immunization with agents countering tumor tolerance and with conventional cancer treatment, while also pointing to a need for improved knowledge on how to develop these multimodal regimes.

Keywords::

• cancer vaccination
• chemotherapy
• CVADD
• human
• immune response
• meeting report
• vaccine adjuvant

The Third International Conference on Cancer Vaccines/Adjuvants/Delivery for the Next Decade (CVADD) 2009 started with an introductory lecture by the conference chair, Malcolm S Mitchell (University of Texas at El Paso, TX, USA), addressing key challenges within cancer vaccine development. Mitchell emphasized that tumor antigens are only slightly different from ‘self’ antigens, resulting in an entirely different situation from prophylactic vaccination against microbes. Reviewing key developments over the last 40 years, he pointed out several examples of vaccine trials with encouraging clinical outcomes. Mitchell stressed that observed clinical benefits may be real and should not be overlooked, even if the chosen p value is not met. As an example, he cited the Phase III trial by Hersey et al. in stage IIb/III melanoma patients [1]. The treated subjects in this trial had substantially increased median survival compared with controls (12.5 vs 7.3 years), but the difference fell narrowly short of statistical significance (p = 0.68). Other speakers at the CVADD meeting highlighted that clinical benefits reaching statistical significance were demonstrated in three studies reported at the American Association for Cancer Research or American Society of Clinical Oncology conferences in 2009 [2–4]. The results from these studies, evaluating vaccine therapy against B-cell lymphoma, malignant melanoma and prostate cancer, may signal a shift from the previously sparse clinical effects from immunization against cancer. On the other hand, the general picture still tells of limited efficacy from vaccination and points to the need for improved strategies. At CVADD 2009, a number of approaches were discussed on how to counter tumor tolerance, polarize the immune response, improve vaccine delivery or combine vaccination with other treatment modalities. The present report provides an overview of interesting results presented at the conference, starting with an update on regulatory immune cells and their role in tumor tolerance.

Regulatory T cells & myeloid-derived suppressor cells

Therapeutic cancer vaccines do not encounter a virgin environment, but rather need to overcome established tumor tolerance. Regulatory immune cells are considered as key players, transforming the tumor microenvironment into a tolerogenic zone and counteracting vaccine effects. In mouse models, CD4⁺CD25^highFoxp3⁺ natural T-regulatory cells (nTregs) and IL-10⁺TGF-β⁺ type 1 regulatory cells (Tr1 cells) have been shown to be highly efficient suppressors of vaccine responses. In humans, however, even effector T cells may express Foxp3, and the classification and functional importance of different Treg subsets remain unclear. Theresa Whiteside (University of Pittsburgh Cancer Institute, PA, USA) presented data at CVADD 2009 from several studies in humans, describing three main Treg subsets:

• • CD4⁺CD25^highFoxp3⁺ nTregs that mediate tolerance to ‘self’ by cell contact-dependent suppression

• • CD4⁺CD25^-CD132⁺IL-10⁺TGF-β 1⁺ Tr1 cells, exerting suppression through IL-10 and TGF-β, are inducible upon antigen encounter amid the influence of IL-10 and tolerogenic dendritic cells (DCs)

• • CD39⁺CD73⁺ nTreg or Tr1 cells that hydrolyze ATP and exert suppression via adenosine

Whiteside reported increased numbers of CD4⁺CD25^highFoxp3⁺ Tregs in the circulation of melanoma and ovarian cancer patients and of CD39⁺ Tregs in subjects with late-stage compared with early-stage head and neck squamous cell carcinoma (HNSCC). Interestingly, data from HNSCC patients suggest that tumor-infiltrating lymphocytes are enriched in Tregs compared with peripheral blood mononuclear cells (PBMCs) and that Tregs from tumors exhibit enhanced suppressor capacity [5]. Increased suppressor function of peripheral CD4⁺CD25^highFoxp3⁺ T cells from HNSCC patients correlated with N-stage (lymph node involvement), but not with T-stage (primary tumor extension) disease. J Nesselhut (University Hospital of Ulm, Germany) reported an increase of CD25⁺CD39⁺ and CD25⁺CD127^-CD4⁺ T cells in the blood of cancer patients compared with healthy donors. Treatment of patients with metronomic chemotherapy induced a downregulation of these cells. Further, Nesselhut reported the identification of a distinct CD2^highCD86⁺ lymphocyte subpopulation in patients with advanced-stage disease. The latter cells were downregulated upon treatment with a tetradecapeptide that is known to attenuate inflammatory cytokine levels.

The role of myeloid-derived suppressor cells (MDSCs) in tumor tolerance was brought to attention by several speakers. Rolf Kiessling (Karolinska Institute, Stockholm, Sweden) pointed out that MDSCs are characterized by elevated levels of arginase I and Stat3, high oxidative stress and coexpression of phenotypic markers for mature and immature DCs. From melanoma patients, he reported that MDSC frequencies in blood correlate with disease stage. Viktor Umansky (German Cancer Research Center, Heidelberg, Germany) presented data from studies in ret transgenic mice with spontaneous melanoma, where MDSCs accumulate in tumor and induce ζ-chain downregulation in nonregulatory tumor-infiltrating lymphocytes. Therapeutic intervention with Viagra™ increased the survival of mice amid inhibition of MDSC suppression, partial correction of ζ-chain expression and T-cell accumulation in tumors.

Vaccine adjuvants & strategies for polarizing the vaccine response

Antigen-specific responses induced through vaccination may not necessarily promote immunity. On the contrary, vaccination in a tolerogenic environment may rather induce anergy, tolerance or an irrelevant immunoactivation. In cancer immunization, type 1/Th1 responses are generally regarded as desirable. A series of different strategies for polarizing the immune response were discussed at the meeting. Walter J Storkus (University of Pittsburgh School of Medicine, PA, USA) presented an elegant study with DCs genetically engineered to express high levels of the transcription factor T-bet [6]. The results indicate that T-bet DCs were superior in inducing type 1 differentiation of naive T cells, while control DCs were equally efficient in promoting type 1 responses from memory T cells. Surprisingly, DC T-bets exhibited a reduced capacity for secreting IL-12 family member cytokines, otherwise known to be important for inducing type 1 immunity. Storkus concluded that the data pointed to a novel mechanism for induction of type 1 immunity, independent of IL-12, and that T-bet DCs may be suitable for immunization against cancer or infectious disease.

Kingston Mills and Neil Marshall (Trinity College Dublin, Ireland) have investigated a series of interesting approaches for enhancing the efficacy of DC-based vaccines through the manipulation of signaling pathways in the DCs. In experiments with antigen-primed and Toll-like receptor-activated DCs, they observed that co-incubation with p38 MAP kinase, or PI3K β or σ inhibitors suppressed tumor cell-induced IL-10 secretion, while maintaining IL-12 levels. These DCs promoted Th1 responses in vivo, but not Tregs, and had considerable therapeutic efficacy against tumors in mice. The data from Mills and Marshall also suggest that DC modulation with siRNA against TGF-β may be useful for countering Treg stimulation.

Andrew G Sikora (Mount Sinai School of Medicine, NY, USA) and Willem W Overwijk (MD Anderson Cancer Center, Houston, USA) reported interesting findings from mouse studies on adjuvants for cancer vaccination. Sikora presented a study evaluating IFN-α as adjuvant for antimelanoma peptides [7]. IFN-α synergized with peptide vaccination in a dose-dependent manner by boosting relative and absolute numbers of gp100-specific T cells that suppressed B16 melanoma growth. IFN-α treatment also increased the long-term maintenance of CD8⁺ T cells with an effector memory phenotype. The latter effect was dependent on IL-15 and on T-cell expression of the IFN-α receptor. Overwijk questioned whether the widely used incomplete Freund’s adjuvant (IFA) represents a suitable adjuvant for eliciting durable responses. In a mouse study, he observed that vaccination with gp100 in IFA induced short-term expansion of antigen-specific cells, but long-term tolerance. Moreover, the data suggested that persistent depots of IFA/gp100 induced eventual depletion of antigen-specific CD8⁺ T cells.

DNA-based cancer vaccination

The DNA vaccine sessions at CVADD 2009 included a series of talks presenting data from clinical trials, timely illustrating that DNA-based vaccines are increasingly overcoming safety hurdles and moving into clinical testing. Douglas G McNeel (University of Wisconsin, WI, USA) presented a clinical trial evaluating a DNA vaccine encoding prostatic acid phosphatase (PAP) in patients with noncastrate, non-metastatic prostate cancer [8]. No significant adverse events were recorded at the three dose levels evaluated (100, 500 and 1500 µg). Three out of 22 patients developed PAP-specific IFN-γ-secreting CD8⁺ T cells as measured by ELISPOT, and nine out of 22 subjects had augmented PAP-specific CD4⁺ or CD8⁺ T-cell proliferation. Further, the immunoassays indicated induction of antigen-specific Tregs. Median prostate-specific antigen (PSA) doubling time increased from 6.5 to 8.5 months (p = 0.033). The investigators conclude that further clinical evaluation is warranted and plan a multicenter, randomized Phase II trial with the lowest vaccine dose.

Richard Harrop (Oxford BioMedica, Oxford, UK) presented data on the clinical development of TroVax, a poxvirus DNA vaccine based on the tumor-associated antigen 5T4. Taking together nine completed Phase I/II trails in patients with prostate, renal or colorectal cancer, 97% of patients receiving at least three TroVax vaccinations have mounted cellular and/or humoral 5T4-specific immune responses (antibody: 94%; proliferation: 88%; ELISPOT: 46%) [9]. The treatment was considered to be well tolerated, and a correlation between immune response and clinical outcome was observed in seven out of nine Phase I/II trials. However, no significant survival advantage (primary end point) has been observed in a recent Phase III trial in renal cancer patients, evaluating TroVax combined with sunitinib, IL-2 or IFN-α. A possible clinical advantage was detected in certain subgroups (normal hematological prestudy levels, and an antibody response against 5T4) and the developers consider further studies based on selecting patients who are likely to respond.

Ronald Rooke (Transgene SA, Strasbourg Cedex, France) reported the development of TG4010, a cancer vaccine targeting mucin-1-expressing tumors [10]. TG4010 is a recombinant viral vector, based on an attenuated strain of vaccinia virus, expressing both MUC1 and IL-2. Rooke provided an update on a randomized Phase IIb trial evaluating TG4010 as an adjunct to chemotherapy in 148 patients with advanced non-small-cell lung cancer. The progression-free survival after 6 months was 44% in the TG4010 plus chemotherapy arm, compared with 26% in the control chemotherapy arm. After a median follow-up of 17.1 months, an overall survival advantage of 39.4 versus 22.7% was observed. These apparent clinical benefits did not reach statistical significance (p > 0.05), but the chosen primary end point of 40% progression-free survival after 6 months was met. Exploratory biomarker analyses identified patients with normal baseline levels of activated natural killer cells as a possible target group. This subpopulation had a favorable clinical outcome after TG4010 therapy compared with controls on all relevant readouts, including a 6-month increase in median survival (17.1 vs 11.3 months).

Four presentations at CVADD 2009 outlined results from clinical trials utilizing electroporation for delivering plasmid-based DNA vaccines. These studies represent the first trials to explore the approach of in vivo DNA electroporation. The presentations from Christian Ottensmeier (University of Southampton, Southampton, UK), Niranjan Sardesai (Inovio, PA, USA), Iacob Mathiesen (Inovio) and Drew Hannaman (Ichor, San Diego, CA, USA) were consistent in showing acceptable tolerability and safety of the applied vaccines and electroporation devices. The reports from Ottensmeier and Hannaman also supported previous data from animal models indicating that delivery by electroporation yields improved immune responses compared with regular plasmid DNA injection. Ottensmeier showed that electroporation substantially enhanced the humoral immune response against the tetanus toxin component of the vaccine, reaching statistical significance within the group of vaccinated patients [11]. Furthermore, the results suggested augmented numbers of CD8⁺ T cells specific for the vaccine antigen prostate-specific membrane antigen (PSMA)27. Sardesai provided an update on an ongoing trial in cervical carcinoma patients evaluating a DNA construct encoding human papillomavirus E16/E18. The results from the first patients indicate vaccine-specific cellular or humoral responses in three out of five or four out of five subjects, respectively. Mathiesen reported results from a clinical trial in hepatitis C patients, using electroporation to deliver a plasmid-based vaccine encoding NS3/4 (collaboration with Matti Sällberg and Ola Weiland [Tripep AB/Karolinska Institute, Sweden]). The results suggest that the vaccine is able to induce transient T-cell responses and reductions in viral load. In all, the data from these four trials suggest that in vivo electroporation is safe, clinically applicable and may enhance T-cell responses after DNA vaccination.

Peptide vaccination combined with chemotherapy

From a clinical point of view, one of the most striking developments within cancer vaccine research is the shift towards combining vaccines with conventional therapies. At CVADD 2009, the presentations clearly showed that previous scepticism towards combing vaccines with chemotherapy is being replaced by discussions on how to combine them. The TG4010 trials described above incorporated vaccination in standard cisplatin and gemcitabin schedules. Giorgio Parmiani (San Raffaele Scientific and University Institute, Milan, Italy) and Hans W Nijman (University Medical Center Groningen, The Netherlands) both presented data from peptide vaccine trials applying low-dose cyclophosphamide. Pioneering studies from the early 1980s have suggested that metronomic cyclophosphamide may enhance immune responses [12]. More recent studies have attributed this phenomenon to a suppressive effect on CD4⁺CD25⁺ Tregs [13,14]. By contrast, the independent studies from Parmiani and Nijman presented at CVADD 2009 were consistent in indicating that low-dose cyclophosphamide did not alter the Treg frequency in PBMCs. The conflicting evidence on this issue points to the complexities of evaluating Tregs in humans. The discussions at the meeting further highlighted our sparse knowledge on how different chemotherapeutics influence Treg function and accumulation in tumors, as well as the influence on other regulatory cell subsets, such as MDSCs.

The cyclophosphamide plus vaccine study presented by Parmiani investigated vaccination of prostate cancer patients (n = 20) with a cocktail of three peptides derived from survivin or PSMA. The results showed induction/boosting of antigen specific T-cell-mediated immunity in the majority of vaccinated patients, amid negligble toxicity. Interestingly, T-cell responses against a PSMA peptide exhibited cross-reactivity against tumor cells and association with stabilization in PSA levels. Parmiani also discussed data from a Phase III trial comparing vitespen, an autologous tumor-derived heat shock protein gp96 peptide complex vaccine, with physician’s choice of treatment for stage IV melanoma [15]. Here, intention-to-treat analysis showed no overall survival benefit. Exploratory landmark analyses revealed that a subgroup, that is, M1a/b substage patients receiving at least ten vaccinations, exhibited an extended median survival compared with the control arm (20.9 vs 12.8 months; p = 0.03). Parmiani concluded that the results warrant a new randomized Phase III trial testing M1a/b melanoma patients receiving at least ten vaccinations.

Immune response in clinical responders after immunotherapy

The majority of immunological responders in cancer vaccine trials experince no detectable clinical benefit. The discrepancy between immunological and clinical responses highlights the need to improve discrimination between useful and pointless immunoactivation. Jianda Yuan (Memorial Sloan-Kettering Cancer Center, NY, USA) and Jon A Kyte (Oslo University Hospital, Norway) have taken the approach to investigate immune responses in patients with clinical response after immunotherapy. At CVADD 2009, Yuan presented data from melanoma patients with clinical responses after therapy with ipilimumab (anti-CTLA4 monoclonal antibody). Five out of eight patients with evidence of clinical benefit had antibodies against NY-ESO-1, compared with none of seven clinical nonresponders [16]. All five NY-ESO-1 seropositive patients had detectable CD4⁺ and CD8⁺ T cells recognizing NY-ESO-1 epitopes. Among five clinical nonresponders analyzed, only one had a NY-ESO-1 CD4⁺ T-cell response. Yuan further highlighted three patients that had experienced tumor regression after receiving both ipilimumab and cancer vaccination. Here, tetramer assays revealed an increased frequency of tumor-specific T cells recognizing gp100 or tyrosinase. Overall, the results provide additional arguments for the attractive strategy of combining CTLA4 blockade with vaccine therapy.

Kyte reported data from advanced cancer patients with unexpected long-term survival after vaccination with telomerase, RAS or TGFβRII peptides [17]. The surviving individuals harbored durable tumor-specific T-cell memory (up to 9 years observation time). Moreover, their responses exhibited several features of possible advantage, including combined T-helper and cytotoxic functionality, diverse HLA restrictions and recognition of naturally processed antigens. Kyte also pointed out high IFN-γ/IL-10-ratios, possibly favoring immunity over tolerance, and polyfunctional cytokine profiles. Interestingly, the cytokine patterns did not follow a Th1/Th2 delineation. Most IFN-γ^high/IL-4^low/IL-10^low cultures included high concentrations of the hallmark Th2 cytokines IL-5 and IL-13. This mixed Th1/Th2 profile was independent of tumor form/vaccination protocol and also applied to patients without a clinical response. Moreover, the unconventional Th1/Th2 pattern did not reflect a mixture of Th1 and Th2 clones, but was observed in 24 out of 25 T-cell clones confirmed to be monoclonal by TCR clonotype mapping. Overall, the cytokine patterns suggest that cytokine profiling in cancer vaccine trials should not rely on a Th1/Th2 paradigm, but rather assess polyfunctionality and the balance between key cytokines.

Conclusion

The knowledge of basic tumor immunology has taken giant strides over the last decades, pointing out attractive strategies for cancer vaccination, but also uncovering the difficulties of defeating a moving and heterogenous tumor target. It has proven to be far easier to design promising therapies in experimental models, where the number of variables are restricted, than to work out how to master the biological complexity within a patient. The CVADD 2009 meeting adressed several of the major challenges facing clinical vaccine development. There was widespread support for combining active immunization with chemotherapy and with strategies countering tumor tolerance. On the other hand, the discussions revealed that current knowledge in humans is sparse on how to manipulate tolerance and design multimodal regimes. The war against cancer looks unlikely to be won with a magic vaccine bullet, but rather calls for a multimodal arsenal of drugs, adjuvants and delivery methods. In this context, the CVADD meetings provide a useful platform for bringing experts in different aspects of vaccine development and application together. It remains to be seen whether the exchange of ideas facilitated by these and other meetings may eventually lead to the successful translation of basic biological and technical advances into effective medicine.

Acknowledgements

The author would like to express his sincere gratitude to the speakers at CVADD 2009 for valuable feedback during preparation of the manuscript. Special thanks to Iacob Mathiesen, Malcolm S Mitchell and Christian Ottensmeier for generous assistance.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.