Fernando Pastor

Agonists of Co-stimulation in Cancer Immunotherapy Directed Against CD137, OX40, GITR, CD27, CD28, and ICOS.

T and natural killer (NK) lymphocytes are considered the main effector players in the immune response against tumors. Full activation of T and NK lymphocytes requires the coordinated participation of several surface receptors that meet their cognate ligands through structured transient cell-to-cell interactions known as immune synapses. In the case of T cells, the main route of stimulation is driven by antigens as recognized in the form of short polypeptides associated with major histocompatibility complex (MHC) antigen-presenting molecules. However, the functional outcome of T-cell stimulation towards clonal expansion and effector function acquisition is contingent on the contact of additional surface receptor-ligand pairs and on the actions of cytokines in the milieu. While some of those interactions are inhibitory, others are activating and are collectively termed co-stimulatory receptors. The best studied belong to either the immunoglobulin superfamily or the tumor necrosis factor-receptor (TNFR) family. Co-stimulatory receptors include surface moieties that are constitutively expressed on resting lymphocytes such as CD28 or CD27 and others whose expression is induced upon recent previous antigen priming, ie, CD137, GITR, OX40, and ICOS. Ligation of these glycoproteins with agonist antibodies actively conveys activating signals to the lymphocyte. Those signals, acting through a potentiation of the cellular immune response, give rise to anti-tumor effects in mouse models. Anti-CD137 antibodies are undergoing clinical trials with evidence of clinical activity and anti-OX40 monoclonal antibodies (mAbs) induce interesting immunomodulation effects in humans. Antibodies anti-CD27 and GITR have recently entered clinical trials. The inherent dangers of these immunomodulation strategies are the precipitation of excessive systemic inflammation or/and invigorating silent autoimmunity. Agonist antibodies, recombinant forms of the natural ligands, and polynucleotide-based aptamers constitute the pharmacologic tools to manipulate such receptors. Preclinical data suggest that the greatest potential of these agents is achieved in combined treatment strategies.

CD28 Aptamers as Powerful Immune Response Modulators

CD28 is one of the main costimulatory receptors responsible for the proper activation of T lymphocytes. We have isolated two aptamers that bind to the CD28 receptor. As a monomer, one of them interfered with the binding of CD28 to its ligand (B7), precluding the costimulatory signal, whereas the other one was inactive. However, dimerization of any of the anti-CD28 aptamers was sufficient to provide an artificial costimulatory signal. No antibody has featured a dual function (i.e., the ability to work as agonist and antagonist) to date. Two different agonistic structures were engineered for each anti-CD28 aptamer. One showed remarkably improved costimulatory properties, surpassing the agonistic effect of an anti-CD28 antibody. Moreover, we showed in vivo that the CD28 agonistic aptamer is capable of enhancing the cellular immune response against a lymphoma idiotype and of prolonging survival of mice which receive the aptamer together with an idiotype vaccine. The CD28 aptamers described in this work could be used to modulate the immune response either blocking the interaction with B7 or enhancing vaccine-induced immune responses in cancer immunotherapy.

Use of Oligonucleotide Aptamer Ligands to Modulate the Function of Immune Receptors

The paucity of costimulation at the tumor site compromises the ability of tumor-specific T cells to eliminate the tumor. The recent U.S. Food and Drug Administration approval of ipilumimab, an antibody that blocks the inhibitory action of CTLA-4, and clinical trials targeting 4-1BB and PD-1 or PD-L1, have underscored the therapeutic potential of using immunomodulatory antibodies to stimulate protective immunity in human patients. Nonetheless, systemic administration of immunomodulatory antibodies has been associated with dose-limiting autoimmune pathologies, conceivably reflecting also the activation of resident autoreactive T cells. Arguably, targeting immunomodulatory ligands to the disseminated tumor lesions of the patient would reduce such drug-associated toxicities. We have recently developed a new class of inhibitory (CTLA-4) and agonistic (4-1BB and OX-40) ligands composed of short oligonucleotide (ODN) aptamers that exhibited bioactivities comparable or superior to that of antibodies. To reduce toxicity, the immunomodulatory aptamers were targeted to the tumor by conjugation to a second aptamer that bound to a product expressed on the surface of the tumor cell, the targeting aptamer, generating a bispecific aptamer conjugate analogous to bispecific antibodies. In a proof-of-concept study in mice, we have shown that an agonistic 4-1BB–binding aptamer conjugated to a prostate-specific membrane antigen (PSMA)–binding aptamer led to the inhibition of PSMA-expressing tumors, was more effective than, and synergized with, vaccination, and exhibited a superior therapeutic index compared with nontargeted costimulation with 4-1BB antibodies or 4-1BB aptamers. The cell-free chemically synthesized ODN aptamers offer significant advantages over antibodies in terms of synthesis, cost, as well as conjugation chemistry needed to generate bispecific ligand fusions. Clin Cancer Res; 19(5); 1054–62. ©2012 AACR.

Anti-Inflammatory Cytokines Induce Lipopolysaccharide Tolerance in Human Monocytes Without Modifying Toll-Like Receptor 4 Membrane Expression

Toll‐like receptor 4 (TLR4) participates in innate immunity by detecting lipopolysaccharides (LPS) of Gram‐negative bacterial cell walls. TLR4 macrophage expression in mice is modulated by LPS. This fact constitutes, at least partially, the molecular basis for LPS tolerance. Very recently, the effect of interferon‐γ (IFN‐γ), a pro‐inflammatory cytokine, has been described on TLR4 membrane expression of human monocytes. IFN‐γ up‐regulates TLR4 expression and antagonizes the LPS‐induced TLR4 down‐regulation. These data prompted us to study the expression of membrane TLR4 in human mono‐ cytes in which LPS tolerance was induced by LPS and by anti‐inflammatory cytokines [interleukin‐10 (IL‐10) and transforming growth factor β1 (TGFβ1)]. Data concerning this latter model, and more specifically, the effect of anti‐inflammatory cytokines over TLR4 expression, are not available at present. We show here that membrane TLR4 expression in human monocytes falls after LPS exposure. The effect was prolonged for 12 h, but then expression returned to normal levels. The incubation of human monocytes with IL‐10, TGFβ1 or a mixture of both induces no alterations in membrane TLR4 expression. However, these cytokines are able to substitute the tolerizing LPS exposure in order to induce LPS tolerance. Our data help to achieve a better understanding of the way cytokines control the cellular expression of TLR.

Identification of TIM3 2’-fluoro oligonucleotide aptamer by HT-SELEX for cancer immunotherapy

TIM3 belongs to a family of receptors that are involved in T-cell exhaustion and Treg functions. The development of new therapeutic agents to block this type of receptors is opening a new avenue in cancer immunotherapy. There are currently several clinical trials ongoing to combine different immune-checkpoint blockades to improve the outcome of cancer patients. Among these combinations we should underline PD1:PDL1 axis and TIM3 blockade, which have shown very promising results in preclinical settings. Most of these types of therapeutic agents are protein cell-derived products, which, although broadly used in clinical settings, are still subject to important limitations. In this work we identify by HT-SELEX TIM3 non-antigenic oligonucleotide aptamers (TIM3Apt) that bind with high affinity and specificity to the extracellular motives of TIM3 on the cell surface. The TIM3Apt1 in its monomeric form displays a potent antagonist capacity on TIM3-expressing lymphocytes, determining the increase of IFN-γ secretion. In colon carcinoma tumor-bearing mice, the combinatorial treatment of TIM3Apt1 and PDL1-antibody blockade is synergistic with a remarkable antitumor effect. Immunotherapeutic aptamers could represent an attractive alternative to monoclonal antibodies, as they exhibit important advantages; namely, lower antigenicity, being chemically synthesized agents with a lower price of manufacture, providing higher malleability, and antidote availability.

Clinical Safety and Immunogenicity of Tumor-Targeted, Plant-Made Id-KLH Conjugate Vaccines for Follicular Lymphoma

We report the first evaluation of plant-made conjugate vaccines for targeted treatment of B-cell follicular lymphoma (FL) in a Phase I safety and immunogenicity clinical study. Each recombinant personalized immunogen consisted of a tumor-derived, plant-produced idiotypic antibody (Ab) hybrid comprising the hypervariable regions of the tumor-associated light and heavy Ab chains, genetically grafted onto a common human IgG1 scaffold. Each immunogen was produced in Nicotiana benthamiana plants using twin magnICON vectors expressing the light and heavy chains of the idiotypic Ab. Each purified Ab was chemically linked to the carrier protein keyhole limpet hemocyanin (KLH) to form a conjugate vaccine. The vaccines were administered to FL patients over a series of ≥6 subcutaneous injections in conjunction with the adjuvant Leukine (GM-CSF). The 27 patients enrolled in the study had previously received non-anti-CD20 cytoreductive therapy followed by ≥4 months of immune recovery prior to first vaccination. Of 11 patients who became evaluable at study conclusion, 82% (9/11) displayed a vaccine-induced, idiotype-specific cellular and/or humoral immune response. No patients showed serious adverse events (SAE) related to vaccination. The fully scalable plant-based manufacturing process yields safe and immunogenic personalized FL vaccines that can be produced within weeks of obtaining patient biopsies.

Prolonged idiotypic vaccination against follicular lymphoma.

During the last 2 decades, idiotypic vaccination has provided proof of principle of biological efficacy, clinical efficacy and clinical benefit in small follicular lymphoma trials. However, with the exception of anecdotal reports, most patients have received no more than 10 doses of their customised idiotype (Id) vaccine. Therefore, it is not known whether prolonged usage of idiotypic vaccination is safe. Since 2002, 18 previously treated patients with follicular lymphoma have received extended idiotypic vaccination at our institution outside clinical trials. Vaccination was provided as a compassionate alternative to no further treatment, and was meant to be stopped only upon complete consumption of the available patient- and tumor-specific vaccine [Id-keyhole limpet hemocyanin + granulocyte-macrophage colony-stimulating factor (Id-KLH + GM-CSF)], or in case of disease relapse or any serious non-local toxicity. So far, 18 patients have received an average of 18 doses of Id vaccine (median: 17; mean: 18; range: 10–31). Eleven patients are still actively receiving idiotypic vaccination: some of them are now over more than 6 years. Toxicity has been systematically negligible and mostly local. No patient has abandoned the vaccination program because of toxicity. Prolonged idiotypic vaccination with the soluble protein Id-KLH + GM-CSF formulation is safe and well tolerated.

MRP1-CD28 bi-specific oligonucleotide aptamers: target costimulation to drug-resistant melanoma cancer stem cells.

In this work we show a clinically feasible strategy to convert in situ the own tumor into an endogenous vaccine by coating the melanoma cancerous cells with CD28 costimulatory ligands. This therapeutic approach is aimed at targeting T-cell costimulation to chemotherapy-resistant tumors which are refractory and been considered as untreatable cancers. These tumors are usually defined by an enrichment of cancer stem cells and characterized by the higher expression of chemotherapy-resistant proteins. In this work we develop the first aptamer that targets chemotherapy-resistant tumors expressing MRP1 through a novel combinatorial peptide-cell SELEX. With the use of the MRP1 aptamer we engineer a MRP1-CD28 bivalent aptamer that is able to bind MRP1-expressing tumors and deliver the CD28 costimulatory signal to tumor-infiltrating lymphocytes. The bi-specific aptamer is able to enhance costimulation in chemotherapy-resistant tumors. Melanoma-bearing mice systemically treated with MRP1-CD28 bivalent aptamer show reduced growth, thus proving an improved mice survival. Besides, we have designed a technically feasible and translational whole-cell vaccine (Aptvax). Disaggregated cells from tumors can be directly decorated with costimulatory ligand aptamers to generate the vaccine Aptvax. CD28Aptvax made of irradiated tumor cells coated with the CD28-agonistic aptamer attached to MRP1 elicits a strong tumor- cell immune response against melanoma tumors reducing tumor growth.

Targeting inhibition of Foxp3 by a CD28 2'-Fluro oligonucleotide aptamer conjugated to P60-peptide enhances active cancer immunotherapy.

The specific inhibition of Treg function has long been a major technical challenge in cancer immunotherapy. So far no single cell-surface marker has been identified that could be used to distinguish Treg cells from other lymphocytes. The only available specific marker mostly expressed in Treg is Foxp3, which is an intracellular transcription factor. A targeting molecule able to penetrate the membrane and inhibit Foxp3 within the cell is needed. P60-peptide is able to do that, but due to lack of target specificity, the doses are extremely high. In this study we have shown as a proof of concept that P60 Foxp3 inhibitor peptide can be conjugated with a CD28 targeting aptamer to deliver the peptide to CD28-expressing cells. The AptCD28-P60 construct is a clinically feasible reagent that improves the efficacy of the unconjugated P60 peptide very significantly. This approach was used to inhibit Treg function in a vaccination context, and it has shown a significant improvement in the induced immune response, entailing a lower tumor load in an antigen-specific cancer vaccine protocol.

Idiotype vaccines for human B-cell malignancies.

After twenty years of use in humans, customized idiotypic vaccination yet remains a non-approved, experimental therapeutic option for patients with lymphoma and myeloma. Potentially applicable to all B-cell malignancies whose cells express a clonal immunoglobulin or its epitopes on their surface, this treatment is designed to prevent disease recurrence or progression. Mostly used in follicular lymphoma patients so far, idiotype vaccines have clearly shown biological efficacy, clinical efficacy and clinical benefit in this setting, although no study aiming at regulatory approval of the procedure has been able to meet its main clinical endpoints. In mantle cell lymphoma, only biological efficacy has been proven for idiotypic vaccination, while in multiple myeloma a limited number of studies support the notion of biological and perhaps even clinical efficacy, although no credible evidence of clinical benefit has still emerged. Idiotype vaccines have been produced and administered in a number of substantially different manners. Therefore, the results of most clinical trials cannot be easily compared, and even less pooled together in meaningful meta-analyses. A more creative and yet scientifically sound way to design clinical trials of customized active immunotherapies will be key to the future development of idiotype vaccines, particularly considering that we currently lack any clinical or biological indicator to possibly predict which patients are more likely to respond to idiotypic vaccination from an immunologic point of view. This review aims at summarizing the multifaceted success achieved by idiotype vaccines, as well as at outlining the challenges awaiting them in the near future: how to improve feasibility, immunogenicity and efficacy, as well as how to confirm benefit and gain regulatory approval.