Exosomes are nanovesicles originating from late endosomal compartments and secreted by most living cells in ex vivo cell culture conditions. The interest in exosomes was rekindled when B-cell and dendritic cell-derived exosomes were shown to mediate MHC-dependent immune responses. Despite limited understanding of exosome biogenesis and physiological relevance, accumulating evidence points to their bioactivity culminating in clinical applications in cancer. This review focuses on the preclinical studies exploiting the immunogenicity of dendritic cell-derived exosomes (Dex) and will elaborate on the past and future vaccination trials conducted using Dex strategy in melanoma and non-small cell lung cancer patients. Cancer Res; 70(4); 1281–5
Exosomes were first described as vesicles secreted by reticulocytes allowing the elimination of obsolete molecules such as transferrin receptors (1, 2). Intense research ensued when B-cell-derived exosomes were shown to stimulate MHC class II-dependent CD4+ T cells in vitro (3, 4). Since then, many studies highlighted that dendritic cell (DC)-derived exosomes (Dex) could modulate immune responses (4, 5), either directly by exposing MHC and costimulatory molecules or indirectly by conveying internal components to surrounding cells (6). Importantly, exosomes seem to also transfer nucleic acids such as mRNA and microRNA and may represent a new mechanism of genetic exchange between cells (4, 7). Despite significant advances achieved in delineating their protein (8) and lipid composition (9) and their biogenesis (10), their physiopathological relevance remains unclear. Two studies on placental exosomes indicated that such vesicles exhibiting T-cell inhibitory potential through various mechanisms (such as the role of FasL; ref. 11; and ULBP molecules; ref. 12) may be associated with term delivery during pregnancy (11). An elegant work pointed out that delivering antigens in vivo through small secreted vesicles such as exosomes is more immunogenic than the mere delivery of its soluble form in tumor models (13). The increasing knowledge about the biological effects of exosomes provides exciting prospects for exosome development in therapy. Here, we will focus on Dex and their capacity to induce immune responses in tumor-bearing hosts, on the possibility of enhancing Dex bioactivity (“second generation” Dex), and finally we will introduce our future phase II clinical trial testing the clinical efficacy of second generation Dex.
Dex as Cell-free Vaccines for Cancer Treatment
In 1998, we showed in mouse models that Dex pulsed with peptide acid eluted from a variety of different tumors mediated tumor growth retardation in an MHC- and CD8+ T-cell-dependent manner (5). Since then, the mechanisms implicated in the bioactivity of Dex in vitro and in vivo were further dissected. Indeed, most lines of experimental evidence converge toward the demonstration of the indirect capacity of exosomal MHC class I and II molecules to trigger CD8+ and CD4+ T-cell activation. Dex can transfer functional peptide-loaded MHC class I and II complexes to DCs in vitro (6, 14). In vivo studies of exosome transfer indicated that Dex require ex vivo generated (14) or host DC (15). We next analyzed the molecular mechanisms involved in the transfer of exosomal MHC molecules to host APC and the role played by the maturation status of DC for the Dex-mediated immunogenicity in vivo. The analysis of the Dex protein composition revealed that Dex are enriched in a variety of proteins that could potentially dock their membrane to that of host APC. Indeed, exosomal milk fat globule-EGF factor VIII/MFG-E8 (theoretically binding to the integrins αvβ5 and αvβ3; refs. 8, 16, 17), as well as Mac-1 α-chain (CD11b) and tetraspanins (CD9, CD63, CD82, CD81), were tested for their potential role in the immunogenicity of Dex (at least when mouse bone marrow-derived DC were used as recipients; ref. 18).8
8Théry C. Unpublished data.
9Zitvogel L. Unpublished data.
First, Dex produced by imDex express low level of molecules implicated in direct T-cell activation or DC targeting such as CD40, CD80, CD86, and ICAM-1, respectively (19–23). Secondly, imDex were poorly immunogenic in cancer-bearing mice and patients. In HHD2 mice, Dex loaded with Mart-1 peptides required TLR9L adjuvants to mount Mart-1-specific CD8+ T-cell activation leading to tumor regression (15). In the first clinical trials using imDex as cell-free vaccines in advanced melanoma (24) and lung cancer-bearing patients (25), we failed to detect vaccine-specific T-cell responses while observing potent Dex-related NK cell activation.
Hence, Dex not only mediate bioactivities on T lymphocytes but also modulate in the innate arm of immune responses. Although tumor-derived exosomes (Tex) could inhibit NK cells through blockade of IL-2-mediated NK cell activation leading to tumor escape (26) or through down regulation of natural killer group 2 member D (NKG2D) on peripheral blood lymphocytes leading to decreased T-cell cytotoxicity (27, 28), Dex can directly trigger NK cell activation in mice and cancer patients. Indeed, we showed that Dex secreted from bone marrow-derived or monocyte-derived DC harbor functional membrane bound NKG2D ligands and interleukin (IL)-15Rα. Importantly, the first phase I clinical study revealed that Dex vaccines significantly augmented circulating NK cell numbers and NKG2D-dependent functions in the majority of melanoma patients (29).
Modulation of Dex Immunogenicity: Second Generation Dex
Over the past 10 years, many groups have been interested in finding a way of using and improving exosomes as immunological tools (Fig. 1A). Because DC represent a promising cell-based strategy to elicit or enhance antitumor immune responses, investigators started to think about the pros and cons of Dex versus DC therapy. Dex are unarguably stable vesicles harboring defined protein and lipid contents that could be tailor-manufactured from genetically modified and GMP-certified cell lines. Furthermore, on the basis of the two first feasibility trials, we generated about 10 vaccines of NK cell-stimulating Dex from one leukapheresis in most individuals, suggesting that Dex may at least complement DC-based therapies of cancer and represent a valuable strategy to boost DC-mediated T- and NK-cell responses. Therefore, the question arose of how to influence Dex bioactivity.
Although the issues of dose and route were studied by some investigators (15, 24, 25, 30), such considerations may be best addressed in future trials aimed at monitoring Dex bioactivity in humans.
In contrast, many studies aimed at changing the molecular composition of exosomes were conducted in order to influence their immune properties. Thus, between 2005 and 2007, Robbins and colleagues attempted to generate tolerogeneic Dex using immature DC treated or transduced with adenoviral vector encoding IL-10, IL-4, or FasL, which were able to suppress inflammation in a murine model of delayed-type hypersensibility and reduce the severity of established collagen-induced arthritis (31–35). Likewise, Dex bearing donor-MHC antigens delayed chronic rejection when transferred into a rat model of cardiac allograft (36), and when produced by TGF-β1- and IL-10-treated DC, Dex were efficient inducers of immune tolerance in a murine skin transplantation model (37). More recently, Dex secreted from DC overexpressing indoleamine 2,3-dioxygenase (IDO) reduced inflammation in a model of rheumatoid arthritis (32).
Likewise, Dex can be directly or indirectly (via DC) loaded with desired immunogens to redirect and gear the immune response. Aline and colleagues showed that mice immunized with Dex loaded with Toxoplasma gondii antigen were protected against T. gondii chronic infection (38). Humoral responses directed against diphtheria toxin were induced by DT-pulsed Dex in naïve mice (39). Dex produced by DC pulsed with acid-eluted tumor peptides reduced tumor growth in mice (5). The question of whether Dex is more potent at raising CD4+ and/or CD8+ T-cell responses when antigens are directly versus indirectly loaded onto exosomes has been addressed by us and others. In an MHC class II-dependent TCR transgenic T-cell model, Qazi and colleagues showed that B- and T-cell responses can only be triggered in vivo when Dex are indirectly loaded through DC pulsing and concluded that in such conditions, Dex exert potent antigen-specific Th1 responses in vivo (40). When comparing the MHC class II-dependent immunogenicity of Dex secreted from immature versus mature DC in skin transplantation models, we established the superiority of LPS-matured DC-derived exosomes, which was accounted for by their enrichment in ICAM-1 and CD86 molecules (20). Our group has implemented these results with human cells and generated Dex from mature DC cultures. Such mDex offer advantages compared with imDex. First, mDex exhibit increased expression of MHC II molecules, CD40, ICAM-1, IL-15Rα, and NKG2D ligands. Second, mDex were indirectly loaded with MHC class I and II peptide antigens through maturing DC leading to potent T-cell triggering in the absence of recipient DC.10
10Chaput N. Unpublished data.
Clinical Trial Using Second Generation Dex in Non–Small Cell Lung Cancer Patients
A phase II clinical trial testing the clinical benefit of Dex as maintenance immunotherapy in patients bearing inoperable (stage IIIB to IV) non small cell lung cancer (NSCLC) responded to or stabilized after induction chemotherapy was launched in November 2009 at the Gustave Roussy and Curie Institutes (Fig. 1B). Dex will be purified from autologous maturing MD-DC loaded with HLA-DP*04-restricted (MAGE-3) and HLA-A*02-restricted peptides (NY-ESO-1, MAGE-1, MAGE-3, MART-1). Patients will first receive four cycles of platinum-based chemotherapy. Based on our preclinical (41) and clinical data (42) showing that metronomic cyclophosphamide (CTX) facilitates Dex-mediated T-cell priming and restores T- and NK-cell functions in end stage patients, HLA-A*02 responders will be eligible for the induction immunotherapy on the basis of the combination of a 3-week oral therapy with low dose CTX (42) followed by four weekly intradermal Dex injections. Continuation immunotherapy will maintain Dex vaccines every 2 weeks for 6 weeks. Forty-one patients will be enrolled between November 2009 and October 2011. The primary objective is to ameliorate the rate of progression-free survival at 4 months postchemotherapy. Secondary objectives are the clinical efficacy of Dex (assessed as overall survival, objective response rates), the biomarkers of efficacy (NK activation, restoration of NKG2D expression, and peptide vaccine-specific T-cell responses), and the safety of mDex in this cohort.
Dex are very promising vaccines in various physiopathological contexts. It is now possible to manipulate the cells from which Dex originate to modulate Dex bioactivity. Dex are able to activate adaptive (6, 15, 40, 41) and innate immunity (29). In addition, preclinical studies showed that peptide vaccines have a higher antitumor efficacy when carried by exosomes (15, 41). However, Dex are vesicles derived from DCs and therefore, their production requires a clinical cell therapy unit qualified for production according to good manufacturing practices with skilled and experienced employees. Moreover, DC differentiation involves growth factors and maturation agents that make the process expensive. However, Dex are produced in large quantities allowing several vaccines. Finally, Dex are very stable and can be cryo-preserved more than 6 months at −80°C with a phenotype and function preserved. This stability gives them an advantage over DCs. Research and development of Dex is timely because phase III trials show the clinical benefit of peptides and antigen-loaded DC in melanoma and prostate cancer, respectively. Dex could substitute or boost other strategies of immunotherapy or be used as a maintenance vaccine. However, future prospects should aim at designing and engineering synthetic exosomes for broader therapeutic indications as has been recently proposed (43).
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant Support: Institut National de la Santé et de la Recherche Médicale (INSERM), Institut Gustave Roussy, Institut Curie, l'Institut National du Cancer, la Fondation Bettencourt Schueller, and donations (Elisabeth Badinter, Agnès b.), Fondation de France.