Dual angiopoietin-2 and VEGFA inhibition elicits antitumor immunity that is enhanced by PD-1 checkpoint blockade

Pathological angiogenesis is a hallmark of cancer and a therapeutic target. Vascular endothelial growth factor A (VEGFA) and angiopoietin-2 (ANGPT2; also known as ANG2) are proangiogenic cytokines that sustain tumor angiogenesis and limit antitumor immunity. We show that combined ANGPT2 and VEGFA blockade by a bispecific antibody (A2V) provided superior therapeutic benefits, as compared to the single agents, in both genetically engineered and transplant tumor models, including metastatic breast cancer (MMTV-PyMT), pancreatic neuroendocrine tumor (RIP1-Tag2), and melanoma. Mechanistically, A2V promoted vascular regression, tumor necrosis, and antigen presentation by intratumoral phagocytes. A2V also normalized the remaining blood vessels and facilitated the extravasation and perivascular accumulation of activated, interferon-γ (IFNγ)–expressing CD8+ cytotoxic T lymphocytes (CTLs). Whereas the antitumoral activity of A2V was, at least partly, CTL-dependent, perivascular T cells concurrently up-regulated the expression of the immune checkpoint ligand programmed cell death ligand 1 (PD-L1) in tumor endothelial cells. IFNγ neutralization blunted this adaptive response, and PD-1 blockade improved tumor control by A2V in different cancer models. These findings position immune cells as key effectors of antiangiogenic therapy and support the rationale for cotargeting angiogenesis and immune checkpoints in cancer therapy.

Anti-angiogenic immunotherapy for lung cancer

Ece Kadioglu

Lung cancer is the leading cause of cancer-associated deaths worldwide. Platinum-based chemotherapy is the most common therapeutic approach in advanced non-small cell lung cancer (NSCLC), particularly for tumors that carry non-druggable activating mutations. However, these tumors are generally not eradicated when diagnosed at an advanced stage and are frequently resistant to chemotherapy. Therefore, there is an urgent medical need for new treatments for this frequent cancer type. Immunotherapies are treatments based on increasing the strength of immune responses against cancer. In particular, monoclonal antibodies that block inhibitory "immune-checkpoints" expressed on T cells (e.g., programmed cell death protein 1, or PD1) show clinical efficacy in increasing tumor types. However, these therapies improve survival in only a minority of NSCLC patients, and ongoing efforts are being made for increasing their efficacy through combinatorial treatments. In this regard, growing evidence indicates that tumor angiogenesis is closely associated with immunosuppression. Pro-angiogenic factors, such as vascular endothelial growth factor A (VEGFA) and angiopoietin 2 (ANG2), cause abnormalities and dysfunctions in tumor blood vessels that impair T cell extravasation and anti-tumoral activity. Recent findings in preclinical cancer models show that reprogramming the features of tumor-associated blood vessels by anti-angiogenic drugs may help to increase intratumoral T cell abundance and sensitize refractory tumors to immunotherapy. Nevertheless, the potential of anti-angiogenics in the context of lung cancer immunotherapy is poorly understood. Therefore, the main objective of my project was to audit combinations of anti-angiogenic therapy and immune checkpoint blockade in a genetically engineered mouse model of NSCLC. I first studied the anti-tumoral activity of dual VEGFA and ANG2 inhibition using a bispecific antibody (A2V) in the KrasLSL-G12D/+;p53fl/fl (KP) NSCLC model. I found that A2V, but not single VEGFA or ANG2 blockade, had robust tumor-inhibitory activity in KP mice, which was at least comparable to standard-of-care chemotherapy. However, A2V only moderately increased cytotoxic T cells in the tumors. To potentially increase the immunogenicity of KP tumors, I also generated two KP mouse variants by either co-expressing a surrogate neoantigen or genetically deleting a DNA mismatch repair enzyme in the cancer cells. However, the therapeutic efficacy of A2V was not further improved in spite of a slightly more activated immune response. I next investigated combination therapies involving A2V and PD1 blockade. In contrast to previous findings in other cancer types, the addition of PD1 antibodies deteriorated the efficacy of A2V, likely through enhancing immunosuppressive mechanisms sustained, at least in part, by regulatory T cells and T cell-antagonizing cytokines. In summary, my work supports the potential clinical efficacy of combined VEGFA and ANG2 blockade for lung cancer therapy but provided little evidence for its synergy in association with anti-PD1 therapy.

EPFL2019

Therapeutic Immunomodulation of the Lymphoid-like Tumor Environment Using Nanoparticulate Formulations

Iraklis Kourtis

Cancer is a leading cause of death worldwide. Understanding the underlying mechanisms that lead to tumor escape and evasion is pivotal for the design of new therapeutic applications. This thesis takes an immunobioengineering approach, and navigates through identifying the immunological parameters relevant in the tumor microenvironment, proposing and developing a strategy in response to address these parameters using nanosized drug and antigen carriers. It is widely accepted that changes in the immunological equilibriumof the tumor microenvironment are crucial for the progression, dissemination and metastasis of a developing tumor. Motivated by tumors' natural secretion of CCL21, a chemokine known for dendritic (DC) migration towards the lymphatics, and for attracting DC and T cells in the lymph node (LN), we perturbed the physiological secretion of CCL21 of B16-F10 melanomas, deriving both a knock-down (CCL21low) and an over-expressing (CCL21high) variant. Interestingly, in immunocompetent mice, growth in CCL21-secreting tumors had an advantage over the CCL21low sub-clones and was CCR7 dependent. CCL21low tumors harbored more antigen specific T cells, while CCL21-secreting, more T regulatory cells (Tregs), which also correlated with subsequent biochemical polarization. Moreover, CCL21-secreting tumors formed a reticular fibroblastic (FRC) network (ERTR7+gp38+CCL21+) reminiscent of the lymphoid tissue of the paracortex (T cell zone) within LNs. The lymphoidlike stroma was orchestrated by the recruitment of CCR7+ adult lymphoid tissue inducers (LTis), responsible for LN formation during embryogenesis, present only in the CCL21-secreting tumors. In mice lacking LTis (Rorc(γt)-deficient) control tumors displayed impaired growth, suggesting that tumor growth is stroma dependent. In addition, the stroma of control or CCL21high tumors was infiltrated by more myeloid suppressor cells (MDSCs) than in CCL21low variant. MDSCs are a heterogeneous population of immature myeloid precursor cells that repress T cell activation and antigen presentation in chronic inflammation and cancer, hampering the efficacy of anti-tumoral vaccinations. CCL21's ability to protect allografts was further demonstrated in non-syngeneic recipients and in ectopically CCL21 transduced βTC tumor models. By mimicking secondary lymphoid organs, tumors may hi-jack and modulate the immune response from immunogenic to tolerogenic to allow for growth and metastasis. Intrigued by the excessive stroma infiltration of MDSCs in CCL21-secreting tumors, we performed a detailed spatiotemporal biodistribution analysis of ultrasmall (sub 50 nm) nanoparticles (NPs) in naïve and tumor bearing mice. Intradermally administered NPs rapidly drained to the LNs and spleen and persistently targetedmajor phagocytic populations that play a key role in hosting an immune response in naïve and tumor bearing mice. Furthermore, tumor-induced myeloid compartments, mainlyMDSCs, were highly targeted naturally within the tumormicroenvironment and spleen (tumor macroenvironment) 12 h post administration, without the use of site specific ligand. Following the MDSCs' natural preference towards phagocytosing NPs, and taking advantage of their excessive reductive cytoplasm, we designed a macromolecular pro-drug formulation that bears a cytotoxic payload and can be released in highly reductive microenvironments. Specifically, we engineered a reducible delivery system of 6-tioguanine (TG) in various formulations. Remarkably, bothMDSC compartments were ablated, monocytic ∼20 fold (CD11b+Ly6c+), polymorphonuclear ∼100 fold (CD11b+Ly6g+), for at least 7 d after administration in tumor bearing mice. As the bioidstribution revealed that apart from MDSCs, other antigen presenting cells (APCs) have phagocytosed NPs, we observed only a ∼2 fold decrease in APCs, suggesting that the rest could be used for vaccination. Taking into account that cancer vaccines and cancer immunotherapy rely on the efficient antigen presentation and education of naïve T cells by professional APCs, we explored the ability of NPs to deliver antigens in the appropriate sub-cellular compartment for antigen presentation and activation of T cells. Using the model antigen ovalbumin, we illustrated that exogenous antigen coupled to NPs by reducible bonds were internalized by DCs in vivo and in vitro. The conjugated antigen was preferentially released from the NPs with a reducible formulation, and was successfully presented in the major histocompatibility complex (MHC) I and II, which lead to activation and proliferation of antigen specific CD8+ and CD4+ T cells in vivo, respectively. Taking together, the nanoparticle platformcould be used efficiently for both inducing cytotoxicity and invoking an immune response. From discovery of new suppressive niches to therapeutical applications, we envisioned a two prong strategy that combined I) a modulatory regime to upset the balance of theMDSC populations both within the tumor microenvoronent and systemically with II) a cancer vaccine with the potential to significantly improve anti-tumoral therapeutic applications. This thesis demonstrated that such an approach is both possible and probable.

EPFL2012

Targeting and Modulating Tumor-Associated Sites with Novel Nanoparticle-Based Immunotherapies

Laura Jeanbart

Cancer is a global disease and a leading cause of death worldwide. While surgery, radiotherapy and chemotherapy comprise the classical tools to eradicate tumors, they do not cure cancer, cannot prevent metastases, lead to side effects, and most importantly they do not instruct the patientʼs immune system to recognize and fight tumor cells. Since the discovery of the immune systemʼs implication in cancer, immunotherapies, which aim at using and educating the immune system to fight and reject cancer, have come to complement classical tumor-debulking methods. A major goal of immunotherapy is to induce or activate effector cytotoxic CD8+ T lymphocytes that can infiltrate and kill the tumor while overcoming immune suppressive mechanisms, which impede their efficacy. Moreover, cancer immunotherapies aim at inducing memory to tumor antigens to prevent relapse or metastases, which traditional therapies cannot do. Dendritic cell (DC) targeting and activation are key for inducing adaptive immunity. Our laboratory has developed synthetic nanoparticles (NPs) capable of draining through lymphatics to target skin-draining lymph nodes (LNs) and being phagocytosed by resident antigen-presenting cells (APCs) upon intradermal injection. We developed a reproducible method to conjugate tumor antigens to NPs and co-delivered them with CpG-conjugated NPs. We found that a NP vaccine composed of a single melanoma-derived peptide (TRP- 2180-188) led to dramatic tumor growth delay in melanoma and was more efficient than a NP vaccine containing several tumor antigens from whole tumor cell lysate (TL). These findings contradicted our hypothesis that immunization with TL might lead to a broader T cell response and hence improved therapeutic outcomes. Additionally, we found that the dose of CpG could modulate the magnitude of the therapeutic outcomes in both single epitope and multiple antigen based NP-vaccines. Considering the efficacy of the developed NP vaccines, we used them as a tool to ask a fundamental and extremely relevant question regarding the influence of vaccine administration site on clinical efficacy: is it therapeutically more beneficial to target a tumor- draining lymph node (tdLN), which is immune suppressed but tumor antigen-primed, or a non-tdLN, which is neither suppressed nor primed by the tumor? We found that targeting the tdLN with NP vaccines led to significantly enhanced therapeutic outcomes in two different tumor models over targeting a non-tdLN, and that efficacy relied on NP conjugation of antigen and adjuvant. This suggested that antigen drainage from the tumor might be available to APCs in the tdLN and we further hypothesized that therapies that lead to immunogenic tumor cell death might synergize with a tdLN-targeted adjuvant approach. While cytotoxic modalities, such as chemotherapy and radiotherapy, have traditionally been used for their ability to kill tumor cells, tumor cells shed antigens and debris upon cell death that drain to the tdLN. Consistent with our hypothesis, targeting an adjuvant to the tdLN enhanced therapeutic benefits of chemo- and radio- therapy, while targeting a non-tdLN did not. We hypothesized that targeting the tdLN with NP vaccines might switch the tdLN microenvironment from a suppressive to an immunogenic one. This suggested that direct targeting of active suppressive mechanisms in the tumor might further improve immunotherapy. We thus engineered a nano-sized micellar carrier loaded with 6-thioguanine (MC-6TG), a cytotoxic drug used in leukemia, and explored its use in targeting T cell- impairing myeloid-derived suppressor cells (MDSCs) in order to enhance the efficacy of therapeutic vaccines. MC-6TG entirely eradicated both Ly6c+ monocytic and Ly6g+ granulocytic MDSCs for up to 7 d in the blood, tumor and tdLN of tumor-bearing mice. Since MC-6TG also targeted certain subsets of macrophages and DCs, depleting MDSCs did not have an impact on the efficacy of a NP vaccine, which relies on APCs for efficacy. However, MC-6TG synergized with adoptively transferred CD8+ T cells, significantly delaying tumor growth and enhancing survival in a murine model of implantable melanoma. Overall, this thesis describes the design and implementation of a novel approach to immunotherapy: we engineered NP-based therapeutic vaccines, optimized their delivery route to enhance efficacy, and simultaneously tackled immune suppressive cells in the tumor microenvironment to allow optimal therapeutic outcomes. The technologies and methods developed in this work are particularly relevant to clinical oncology and have the potential to benefit cancer patients by improving cancer therapy efficacy.

EPFL2013