Cancer immunotherapy

Dendritic-cell vaccines on the move

Vaccines that induce an antitumour immune response are disappointingly ineffective in treating patients with cancer. Pre-conditioning the vaccination site to induce inflammation might provide a way to improve this therapy. See Letter p.366

Dendritic cells (DCs) are often called nature's adjuvants because of the way in which they help to initiate an immune response. Found throughout the body, the cells acquire and process antigens (the molecules recognized and bound by antibodies) from pathogens and tumours. They then migrate to lymph nodes and activate T cells, which in turn induce protective immune responses. These properties have driven attempts to develop vaccines containing DCs loaded with tumour antigens, with the aim of inducing antitumour immune responses in patients with cancer1. But this strategy has fallen short of expectations. In this issue, Mitchell et al.2 (page 366) show how simply improving DC migration to lymph nodes dramatically enhances antitumour responses in humans and mice, pointing to a way to optimize the use of DC vaccines.

There is a general consensus that DC vaccines can safely induce long-lasting antitumour immune responses. These vaccinations have produced encouraging, if modest, clinical results in some patients with advanced cancers3. For instance, the vaccine sipuleucel-T (the only cell-based cancer vaccine approved for use in the United States) increases median survival times by four months in patients with prostate cancer4. But several factors might be limiting the efficacy of DC vaccines: the source and type of DCs used; the site and frequency of injection; and the ability of DCs to migrate to lymph nodes. Moreover, the injected DCs may not themselves directly instigate an immune response, but instead might act indirectly through DCs already present in the lymph node5.

Less than 5% of cells in a DC vaccine reach the lymph nodes6. In mice, DC migration can be improved either by injecting activated DCs or by pre-conditioning the vaccination site in the skin with the inflammatory molecule TNF-α (ref. 7). Mitchell and colleagues therefore investigated whether pre-conditioning the DC vaccine site to generate local inflammatory responses might enhance DC migration in humans. To do this, they used a tetanus/diphtheria (Td) toxoid vaccine. Most people have been exposed to this toxoid during childhood vaccinations, and re-exposure activates a subset of T cells called memory CD4+ T cells that recognize only the Td antigen and mount a strong and rapid inflammatory immune response in its presence.

Glioblastoma multiforme (GBM) is an aggressive brain tumour in which cells specifically express pp65, an antigen from a common herpesvirus called cytomegalovirus, making pp65 an attractive target for immunotherapy. The authors treated patients who had GBM (and who had had their tumours removed and were receiving chemotherapy) by pre-conditioning them either with Td or, as a control, with DCs that had not been treated with pp65. All patients were then vaccinated with pp65-loaded DCs and given monthly DC vaccinations until their tumours regrew.

More DCs accumulated in the lymph nodes of patients who received Td treatment than of those who did not. Furthermore, these patients survived longer on average than those treated with DCs — three out of six survived past the end of the trial, almost twice as long as expected with standard treatment. Mitchell and co-workers report that Td-treated patients had longer-lasting, higher levels of pp65-specific T-cell responses than controls, a possible reason for their improved outcomes. However, this is only a small study, and so the relationship between elevated T-cell responses and clinical outcome remains unclear and warrants investigation in a larger patient cohort.

The authors then validated their results in mice that had been previously exposed to Td, using ovalbumin as a tumour antigen. They found that, as in humans, increased trafficking of antigen-loaded DCs to the lymph nodes depended on Td-specific memory CD4+ T cells, because migration was less efficient in mice not previously exposed to Td. Furthermore, the efficiency with which DCs migrated to lymph nodes increased on both sides of the body following pre-conditioning, even though the animals were pre-conditioned on only one side. This systemic effect seems to be mediated in both humans and mice by CCL3, a chemokine found in blood serum that was rapidly induced following Td pre-conditioning (chemokines are signalling proteins that guide cell migration).

The researchers demonstrated that both CCL3 expression and Td-specific memory CD4+ T-cell responses are required for DC migration. They propose that CCL3 functions by upregulating another chemokine, CCL21, in the skin and lymph nodes. CCL21 binds to and activates the chemokine-receptor protein CCR7, an interaction that is key for the efficient migration of activated DCs and T cells8,9. In support of their hypothesis, Mitchell and colleagues demonstrated that the efficacy of ovalbumin-loaded vaccines in a mouse model of melanoma depends on Td pre-conditioning and production of CCL3 and CCL21 (Fig. 1).

Figure 1: Optimizing dendritic-cell vaccines.

a, Dendritic cells (DCs) migrate to lymph nodes to present tumour-derived antigens to two types of immune cell: memory CD4+ T cells that have previously been exposed to that antigen and so elicit strong, rapid immune responses; and CD8+ T cells, which kill cells that express the antigen. This property is exploited to treat patients with cancer using vaccinations containing DCs loaded with pp65, which acts as a tumour antigen in this setting. However, migration of injected DCs to the lymph nodes is inefficient. b, Mitchell et al.2 report that pre-conditioning the DC vaccination site with a tetanus/diphtheria (Td) toxoid vaccine improves DC migration by inducing inflammatory immune responses mediated by Td-recognizing CD4+ T cells and generation of the protein CCL3. This protein upregulates production of the protein CCL21, which promotes DC and T-cell migration into lymph nodes. CCL3 may also recruit CD8+ T cells to sites where DCs and CD4+ T cells interact.

These results expand our understanding of the mechanisms underlying successful DC vaccines in several ways. Because both DCs and CD4+ T cells produce CCL3 following antigen recognition10, one could speculate that the interaction between these cell types explains the fact that pre-conditioning depends on host CCL3 and activation of Td-specific memory CD4+ T cells. Moreover, CCL3 has other roles in the immune response that may further enhance DC vaccines: mobilizing DC precursors, regulating DC maturation and migration, and recruiting CD8+ T cells, which kill cancer cells, to sites of interaction between DCs and CD4+ T cells11. Finally, regulatory T cells, which inhibit the induction of tumour-specific immune responses, can decrease production of CCL3 (ref. 12), and pre-conditioning might elevate CCL3 to sufficient levels to overcome this effect.

DC vaccines loaded with patient-derived tumour preparations are being tested in large-scale phase III trials for renal cancer and GBM. Regardless of the outcome of these trials, Mitchell et al. provide a tantalizing approach to improving suboptimal DC vaccines. However, the increased efficiency that the authors observe may also be partially attributable to other factors: the use of a tumour antigen that provokes a stronger immune response than those used previously13; the fact that the authors vaccinated patients whose tumours had already been removed; or the fact that the patients also received chemotherapy, which may deplete regulatory T cells14. Furthermore, it will be key to identify which patients' tumours actually expressed the pp65 protein targeted by the vaccine. Pre-conditioning and activating DC vaccines more potently15, combining vaccines with agents to counteract factors that inhibit tumour-specific immune responses16, and targeting DC vaccines directly to the tumour17 are all approaches with the potential to further enhance vaccine efficacy.

Footnote 1


  1. 1.

    See all news & views


  1. 1

    Sabado, R. L. & Bhardwaj, N. Ann. NY Acad. Sci. 1284, 31–45 (2013).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Mitchell, D. A. et al. Nature 519, 366–369 (2015).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Palucka, K. & Banchereau, J. Immunity 39, 38–48 (2013).

    CAS  Article  Google Scholar 

  4. 4

    Kantoff, P. W. et al. N. Engl. J. Med. 363, 411–422 (2010).

    CAS  Article  Google Scholar 

  5. 5

    Allan, R. S. et al. Immunity 25, 153–162 (2006).

    CAS  Article  Google Scholar 

  6. 6

    de Vries, I. J. M. et al. Cancer Res. 63, 12–17 (2003).

    CAS  PubMed  Google Scholar 

  7. 7

    Martín-Fontecha, A. et al. J. Exp. Med. 198, 615–621 (2003).

    Article  Google Scholar 

  8. 8

    Sallusto, F. & Lanzavecchia, A. Immunol Rev. 177, 134–140 (2000).

    CAS  Article  Google Scholar 

  9. 9

    Griffith, J. W., Sokol, C. L. & Luster, A. D. Annu. Rev. Immunol. 32, 659–702 (2014).

    CAS  Article  Google Scholar 

  10. 10

    Luther, S. A. & Cyster, J. G. Nature Immunol. 2, 102–107 (2001).

    CAS  Article  Google Scholar 

  11. 11

    Menten, P., Wuyts, A. & Van Damme, J. Cytokine Growth Factor Rev. 13, 455–481 (2002).

    CAS  Article  Google Scholar 

  12. 12

    Morlacchi, S. et al. J. Immunol. 186, 6807–6814 (2011).

    CAS  Article  Google Scholar 

  13. 13

    Nair, S. K., Sampson, J. H. & Mitchell, D. A. OncoImmunology 3, e29289 (2014).

    Article  Google Scholar 

  14. 14

    Ridolfi, L. et al. J. Transl. Med. 11, 135 (2013).

    CAS  Article  Google Scholar 

  15. 15

    Bogunovic, D. et al. Cancer Res. 71, 5467–5476 (2011).

    CAS  Article  Google Scholar 

  16. 16

    Ribas, A. et al. Clin. Cancer Res. 15, 6267–6276 (2009).

    CAS  Article  Google Scholar 

  17. 17

    Vera, M. et al. Mol. Ther. 12, 950–959 (2005).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Nina Bhardwaj.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sabado, R., Bhardwaj, N. Dendritic-cell vaccines on the move. Nature 519, 300–301 (2015).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.