News & Views | Published:


Immune pact with the enemy

Nature volume 450, pages 803804 (06 December 2007) | Download Citation

Progress comes from the latest investigations into a long-standing question in immunology — the role of the immune system in maintaining small, potentially cancerous lesions in a state of dormancy.

The current belief about treating cancer is that tumour cells need to be eradicated as quickly as possible, so as to halt tumour growth and spread, and to prevent or delay the death of the patient. The startling results of Koebel et al.1 (page 903 of this issue) demonstrate that considering cancer as a fatal disease is not always appropriate*.

The authors show in mice that induction of cancer by methylcholanthrene (MCA), a tar component of the kind found in cigarette smoke, causes an initial wave of deadly tumours affecting many animals, after which the surviving mice show no evidence of growing tumours. Deceptively, however, dormant tumours still exist in these apparently healthy mice, and these are kept in check by the immune system. This state can be fatally disrupted by suppression of the immune system, allowing the dormant tumours to wrest themselves from immune control and grow, spread and kill their host.

The concept of tumour dormancy, or latency, has a long history, and is well documented in animal studies and in anecdotal clinical observations2,3,4,5. Also, it has long been suspected that immunosuppression can activate dormant tumours. For example, a cancer — malignant melanoma — developed in the recipient of a kidney transplant from a donor who had been treated for melanoma 16 years earlier, and was considered cured5. The cancer in the recipient was found to be of donor origin: apparently, the immunosuppression required to prevent the recipient's rejection of the kidney triggered regrowth of the cancer that had been kept under immune control in the donor for all those years. In none of these previous reports, however, was the dormant state actually visualized, and the component of the immune system involved in maintaining dormancy was not identified.

Koebel et al.1 take strides forward in both respects. After the initial wave of MCA-induced deadly tumours, dormant lesions in immunocompetent mice — those with a fully functional immune system — developed into progressive tumours only after treatment that resulted in the depletion of immune-system cells known as T lymphocytes (T cells), or neutralization of the cytokines interleukin-12 or interferon-γ, which are involved in adaptive immunity. Depletion of cells called natural killer cells, which are more broadly acting but less specific immune agents, had no effect. These results point to highly specific, adaptive T-cell immunity as the component of the immune system that maintains dormancy. Interestingly, in a different tumour model, immunization with tumour cells can generate antibodies that also contribute to dormancy4.

Koebel et al. also found that many cells in the stable, dormant lesions showed morphological features reminiscent of those in progressively growing, MCA-induced cancers. Like the growing cancers, the stable lesions were infiltrated by immune cells, including T cells, indicating that they were immunogenic (that is, they were being recognized by the immune system). But there was a much lower percentage of proliferating cells and an increased incidence of cell death. Transient culture of cells from dormant lesions yielded atypical fibroblast-like cells that grew out as tumours when injected into immunodeficient, but not immunocompetent, mice. Even in some immunocompetent animals, however, stable lesions occasionally escaped from dormancy and became cancerous. But these lesions could do so only if they had lost their immunogenicity, as indicated by their subsequent ability to grow in immunocompetent host animals.

Thus, Koebel and colleagues' work for the first time characterizes a state of tumour dormancy. The hallmarks of this state are stable lesions of transformed immunogenic cells, which are controlled by the host's adaptive immune system in a condition dubbed 'equilibrium' because of its dynamic nature. Obviously, this is a precarious situation — loss of either immunocompetence or immunogenicity can lead to tumour outgrowth, as the authors show.

The implications of this work are far-reaching. First and foremost, the description and visualization of dormant lesions offers an opportunity to characterize their molecular signatures, as determined by their gene-expression profiles, and to compare these signatures with those of the lesions that became cancerous even in immunocompetent hosts. Indeed, such understanding may lead to the development of new treatments, including non-immune-drug interventions, to turn overt cancers into less aggressive, stable lesions. Second, as Koebel et al.1 point out, cancer immunotherapy can aim not only at complete tumour eradication, but also at establishing tumour equilibrium by encouraging the production of interferon-γ-producing, tumour-specific T cells. Indeed, spontaneous T-cell infiltration into human cancers is now increasingly recognized as a favourable prognostic sign, independently of other indicators6,7,8.

Third, this model of dormancy has striking parallels with the chronic infection caused by Mycobacterium tuberculosis, one of the world's most successful pathogens. Typically, an asymptomatic or latent infection is established, which can last for decades before the pathogen is reactivated and clinical tuberculosis ensues. Such processes often coincide with a phase of immune suppression9,10. Indeed, M. tuberculosis is thought to use special bacterial gene products to maintain latency, and it is tempting to speculate that dormant tumours use similar tricks to avoid being eradicated.

Fourth, a more intense search for dormant tumours is warranted — particularly for those tumours induced by chemicals, such as may be present in cigarette smokers, given that Koebel and colleagues' mouse system mimics that situation. Obviously, dormant tumours in smokers would pose a threat, because they can awaken and become overt cancer, at which time it is usually too late for effective therapy11. It may be difficult to detect truly dormant lung cancers. But perhaps patients with breast cancer offer another route for investigation: up to 22 years after undergoing a mastectomy, one-third of patients reportedly have evidence of circulating 'tumour' cells without any evidence of disease3. Are these cells also kept in check by immune responses, or are they controlled by other mechanisms?

A final, unwelcome, thought prompted by the new results concerns the treatment of cancer patients with immunosuppressive chemotherapy or irradiation. A downside of such treatment could be the escape of dormant tumour cells from immune control. Dormant cells themselves are likely to be less susceptible to these treatments, which primarily target rapidly dividing cells.


  1. 1.

    This News & Views article and the paper concerned1 were published online on 18 November 2007.


  1. 1.

    et al. Nature 450, 903–907 (2007).

  2. 2.

    et al. Breast Dis. 26, 87–98 (2007).

  3. 3.

    et al. Cancer Biol. Ther. 6, 1–9 (2007).

  4. 4.

    & Semin. Cancer Biol. 11, 277–283 (2001).

  5. 5.

    et al. N. Engl. J. Med. 348, 567–568 (2003).

  6. 6.

    et al. Science 313, 1960–1964 (2006).

  7. 7.

    et al. Proc. Natl Acad. Sci. USA 102, 18538–18543 (2005).

  8. 8.

    et al. Cancer Res. 67, 354–361 (2007).

  9. 9.

    Trends Microbiol. 15, 7–13 (2007).

  10. 10.

    et al. Microbes Infect. 8, 2052–2060 (2006).

  11. 11.

    J. Clin. Invest. 117, 3436–3444 (2007).

Download references

Author information


  1. Cornelis J. M. Melief is in the Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, Albinusdreef 2, PO Box 9600, 2300 RC Leiden, the Netherlands.

    • Cornelis J. M. Melief


  1. Search for Cornelis J. M. Melief in:

About this article

Publication history



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.

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