News & Views | Published:

A needle in the 'cancer vaccine' haystack

Nature Medicine volume 16, pages 854856 (2010) | Download Citation

Vaccines can prevent infections by several pathogens. Success, however, has been limited for other chronic diseases, reflecting a constraint for effectively manipulating the human immune system. The results from four studies describe a novel dendritic cell (DC) subset in humans that may be crucial for the design of vaccines against cancer and other chronic diseases.

Vaccines rely on the targeting and activation of DCs, potent antigen-presenting cells of the immune system, to promote protective immunity1,2. Upon acquisition of antigens, DCs can instruct the type of immune response to be induced. Several subsets have been described in mice, each endowed with distinct functions. DCs are generally subdivided into lymphoid tissue–resident conventional DCs (cDCs), type 1 interferon–producing plasmacytoid DCs (pDCs) and 'migratory' DCs such as Langerhans cells in the skin. cDCs can be further differentiated into CD8α+ and CD8α DCs.

In mice, splenic CD8α+ DCs have emerged as the most potent DC subset for cross-presentation, through their ability to phagocytose dying infected cells or cancer cells and present exogenous antigens onto major histocompatibility complex type I molecules to elicit cytotoxic CD8+ T cell (CTL) responses. Splenic CD8α+ DCs, therefore, represent a promising target in vaccination strategies3, as cross-presentation is essential for eradication of cancers, viruses and other pathogens. A large effort has been made to find the human counterpart of mouse CD8α+ DCs, and new findings suggest that human DCs expressing blood DC antigen-3 (also known as CD141) are the human equivalent of the mouse CD8α+ subset4,5,6,7.

Although rare in human blood, BDCA3+ DCs are also found in lymph nodes, tonsils, bone marrow and spleen, where they localize to the T cell areas4,5,6,7. This human DC subset shares several phenotypic and functional properties with mouse CD8α+ DCs (Fig. 1a). Both specifically express DC NK lectin group receptor-1 (DNGR-1), a C-type lectin that can act as a sensor of necrotic cells8,9, allowing these DCs to phagocytose dead cells, process captured antigens and cross-present them to CD8+ T cells8. The chemokine receptor XCR1 also seems to be selectively expressed on BDCA3+ DCs4,5. In fact, release of the XCR1 ligand XCL1 by activated natural killer cells and CD8+ T cells at the site of infection may serve as a specific signal to recruit BDCA3+ DCs10. Both human and mouse DCs also express basic leucine zipper transcriptional factor ATF-like-3 and interferon regulatory factor-8 (refs. 6,7), transcription factors required for mouse CD8α+ DC development, but not interferon regulatory factor-4, which is involved in the development of other mouse DC subtypes.

Figure 1: Alternative antitumor vaccine strategies based on a previously undescribed BDCA3-expressing DC subset.
Figure 1

(a) Upon TLR3 stimulation, BDCA3+ DCs recognize necrotic tumor cells via DNGR-1, phagocytose dying cells and cross-present tumor antigens to CD8+ T cells, inducing CTL responses. They also secrete interferon-β and IL-12, cytokines that induce TH1 responses, which in turn enhance CTL responses crucial for antitumor immunity. (b) DC-based vaccines rely on two main approaches: ex vivo–generated DC vaccines loaded with tumor-associated antigens (peptides, proteins, irradiated tumor cells, RNA encoding tumor-associated antigens (TAAs) and bacterial and viral vectors, left); and specific DC targeting with DC-specific antibodies fused with tumor-associated antigens (right). Recent findings characterizing the BDCA3+ DC subset enable rational approaches to isolate, expand and load BDCA3+ DCs ex vivo or to specifically target tumor-associated antigens to these DCs with antibodies specific to cell surface receptors such as DNGR1. MoDCs, monocyte-derived DCs. Image: Katie Vicari

Human BDCA3+ DCs, as well as mouse CD8α+ DCs, express high levels of the pattern recognition receptor Toll-like receptor 3 (TLR3) and respond to poly I:C, a surrogate double-stranded RNA and agonist of TLR3 that has been shown to enhance cross-presentation in vitro and in vivo11. Upon TLR3 stimulation, BDCA3+ DCs produce interleukin-12 (IL-12) and interferon-β, cytokines known to enhance T helper type 1 (TH1) and CTL responses6,7. Notably, they efficiently internalize material from dead cells and cross-present cell-associated and soluble antigens to CD8+ T cells, properties essential for cell-mediated antitumor immunity4,5,6,7. BDCA3+ DCs do not express TLR7, but, unlike mouse CD8α+ DCs, they express the closely related TLR8. Given these specific features (similarity to mouse CD8α+ DCs, IL-12 expression, ability to cross-present and TLR responsiveness) BDCA3+ DCs may be key natural 'adjuvants' for therapeutic cancer or chronic viral vaccines.

Clinical trials have shown that DC-based vaccines can directly induce specific antitumor immune responses. But this vaccine strategy has fallen short of demonstrating a significant decrease in the progression of cancer. Provenge (Dendreon, also known as sipuleucel-T)12, a prostate cancer vaccine composed of white cells partially enriched with blood DCs pulsed with a prostate cancer–associated antigen, is the only approved cell-based immune therapy. Although patients given Provenge showed only a small improvement in overall survival in a clinical trial, it was statistically significant, pointing to the therapeutic potential of DC-based vaccines and highlighting the need to improve current DC-based vaccination strategies.

DC-based immunotherapies require optimization at several levels: the maturation stimuli used, the type and form of antigen to be administered, the number of DCs to inject and the frequency, route and site of the injection. Questions remain as to whether the classical ex vivo–generated DCs widely used in DC-based vaccines are the most effective means of inducing clinically significant antitumor immunity.

Although most clinical trials use monocyte-derived DCs obtained after exposure of monocytes to granulocyte-macrophage colony–stimulating factor and IL-4, DCs generated under alternative culture conditions such as granulocyte-macrophage colony–stimulating factor and IL-15—which have properties characteristic of Langerhans cells—may induce more efficient CTL responses13. Cytokine cocktails used to activate DCs are also crucial for the quality of the elicited T cell responses. It seems, however, that the standard combination of cytokines—tumor necrosis factor-α, IL-1β, IL-6 and prostaglandin E2—used in clinical trials may not induce optimal antitumor responses14,15 and suggests that activation of alternative DC subsets can lead to improved clinical efficacy of DC vaccines.

The authors found that BDCA3+ DCs produce higher levels of IL-12 than BDCA1+ DCs, another myeloid blood DC subset, and are more efficient at cross-presentation than monocyte-derived DCs, BDCA1+ DCs or pDCs4,5,6,7. Poulin et al.7 showed that BDCA3+ DCs are more effective at cross-presenting tumor antigens, such as long melan-A–derived peptides (that do not require prior human leukocyte antigen matching) to CD8+ T cells as compared with monocyte-derived DCs, which may translate clinically to improved immune responses in patients.

Further studies are needed to define the immunogenic potential of BDCA3+ DCs before they can be used for vaccination. Human DC subsets need to be directly compared, after activation with various stimuli, for their capacity to induce CTL and TH1 responses. In vivo, the immunogenicity of BDCA3+ DCs and monocyte-derived DCs might be tested in a tumor-bearing humanized mouse model for their ability to induce tumor rejection. It is also necessary to define which DC subsets migrate into tumors and whether tumor-derived immunosuppressive factors modify their function. Several attributes of the immune response elicited by BDCA3+ DC vaccines, such as the migratory capacity and avidity of T cells induced within the tumor lesion, also need to be evaluated.

Looking forward, the main challenge will be to develop an efficient way to generate large numbers of autologous BDCA3+ DCs ex vivo. Although Poulin et al.7 were able to generate BDCA3+ DCs from cord blood hematopoietic stem cells, their yield was too low to be practical for using this method to generate enough cells for vaccination. Alternative vaccination strategies such as the delivery of tumor antigens in vivo to specific DC subsets could also be employed (Fig. 1b). Given the restricted expression of DNGR-1 on BDCA3+ DCs and data showing that a DNGR-1–targeted antigen is efficiently cross-presented to CD8+ T cells in mice8, this receptor might be used to target tumor-associated antigens to endogenous BDCA3+ DCs along with activation signals such as poly I:C. In the same way, the selective XCR1 expression on only these DCs makes this chemokine receptor another interesting target for the delivery of vaccines to cross-presenting DCs4.

Studies in humans and mice have emphasized that different DC subsets are endowed with specialized functions, and a good vaccine should target these subsets in a coordinated way. The current studies provide the basis for a new approach relying on BDCA3+ DCs as antitumor and antiviral vaccines, as they seem to be a key subset for cross-presentation of cell- associated antigens. Further characterization of these DCs will enable rational approaches to target them to improve vaccine efficacy.


  1. 1.

    & Immunotherapy 2, 37–56 (2010).

  2. 2.

    & Nature 449, 419–426 (2007).

  3. 3.

    & Immunol. Rev. 234, 18–31 (2010).

  4. 4.

    et al. J. Exp. Med. 207, 1273–1281 (2010).

  5. 5.

    et al. J. Exp. Med. 207, 1283–1292 (2010).

  6. 6.

    et al. J. Exp. Med. 207, 1247–1260 (2010).

  7. 7.

    et al. J. Exp. Med. 207, 1261–1271 (2010).

  8. 8.

    et al. J. Clin. Invest. 118, 2098–2110 (2008).

  9. 9.

    et al. Nature 458, 899–903 (2009).

  10. 10.

    et al. Immunity 31, 823–833 (2009).

  11. 11.

    et al. Nature 433, 887–892 (2005).

  12. 12.

    et al. Nat. Rev. Drug. Discov. 9, 513–514 (2010).

  13. 13.

    et al. Eur. J. Immunol. 37, 1678–1690 (2007).

  14. 14.

    et al. Cancer Res. 64, 5934–5937 (2004).

  15. 15.

    et al. Cancer Res. 68, 9441–9450 (2008).

Download references

Author information


  1. Anne Gallois and Nina Bhardwaj are at the New York University School of Medicine, New York University Langone Medical Center, Cancer Institute, New York, New York, USA.

    • Anne Gallois
    •  & Nina Bhardwaj


  1. Search for Anne Gallois in:

  2. Search for Nina Bhardwaj in:

Competing interests

The authors declare competing financial interests: details accompany the full-text HTML version of the paper at

Corresponding authors

Correspondence to Anne Gallois or Nina Bhardwaj.

About this article

Publication history



Further reading

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing