The development and progression of malignant tumors is a complex process involving tumor cells and their microenvironment, including local (paracrine) and systemic (endocrine) signaling and local as well as systemic cell recruitment. It has become clear that cancers do not show a linear growth and progression, since periods of a “steady state” might appear and interrupt exponential tumor growth. In such instances, fully transformed tumor cells do not expand into a clinically detectable cancer, possibly due to lack of stimulatory or permissive signals, or due to active inhibitory mechanisms. This state of dormancy might apply to early stages of primary cancers, to remnants of primary tumors which would be “seeds” for disease recurrences, and to “dormant” micrometastases which could, after a latency period, be “reactivated” and evolve into clinically manifested disease. Several mechanisms are involved in this “lack of tumor progression”.

Studies about tumor dormancy have increased exponentially since the 1970s, especially since the introduction of experimental models, allowing mechanistic studies of the clinical phenomenon of “latency periods” in cancer progression (Fig. 1). The purpose of this special issue is to shed some light on different forms of tumor dormancy and the regulatory mechanisms involved, and to present up-to-date clinical and experimental evidence in support of various hypotheses proposed to explain this phenomenon: from cell cycle arrest to hormonal dependence, lack of tumor neovascularization (angiogenesis), immune surveillance, and the role of the microenvironment. The authors report on observations of human cancers as well as experimental studies of tumor progression and dormancy regulation. This might be relevant for the development and implementation of novel diagnostic and therapeutic strategies.


Figure 1. Number of PubMed-cited papers on the subject of tumor dormancy by year of publication.

Download figure to PowerPoint

Goss et al. describe the problem of tumor dormancy from the clinical point of view, pointing at the difficulties in predicting the probability of tumor recurrences after primary treatment, apparently due to remaining and dormant tumor cells. The authors ask the following key questions: 1. Are micrometastases already present at the time of primary tumor detection, and how can this be predicted? 2. If micrometastases are present, are they in a state of true non-proliferating dormancy or a state of clinical dormancy in which cell division and cell death are balanced? 3. Can micrometastases arise from residual disease at the primary tumor site or regional nodes after primary therapy, or is spread from one site of micrometastasis to another site possible? 4. What are the growth characteristics of micometastases—do they grow exponentially or are there intermittent periods of dormancy? 5. Finally, what therapeutic interventions may be effective, and are there predictive markers for selecting patients most able to benefit from these? Possible treatments and trials aimed at detecting the presence of dormant tumor cells locally are discussed, for instance the MAI17 and the TEACH trials. Experimental data are also described, relating to single cell dormancy, epigenetic regulation of dormancy, role of stem cells in dormancy, and improved detection of small lesions by magnetic resonance imaging.

Naumov et al. focus on the importance of angiogenesis in the progression of malignant tumors and indicate that repression of tumor vascularization might be one key mechanism of tumor dormancy. Various molecular mechanisms are discussed, and the authors also indicate that novel systemic angiogenesis markers might be used to improve early tumor detection, possibly at the stage of dormancy.

Horak and collaborators discuss the role of metastasis suppressor genes (MSGs) in tumors and how these may be related to tumor dormancy, especially KISS1, KAI1, MKK4/7 and Nm23-H. Of these, KISS1 appears to trigger dormancy in solitary, metastatic tumor cells by growth arrest at the secondary site, whereas KAI1 induces growth arrest prior to extravasation by binding a vascular endothelial cell surface marker. MKK4, MKK7 and Nm23-H1 appear to promote dormancy of micrometastatic colonies after disseminated tumor cells have undergone several rounds of proliferation. Some of the important questions include: What are the signaling pathways mediating metastatic dormancy? How do tumor cells overcome dormancy? Can tumor cells return to dormancy after they have become proliferative? Can the re-expression of MSGs or mimicking MSG signaling shunt tumor cells into a dormant state, or have they developed mechanisms sufficient to overcome the suppressive effects of MSGs?

Allgayer & Aguirre-Ghiso describe how the urokinase receptor (u-PAR) is a critical regulator of invasion, intravasation, and metastasis. μ-PAR is a key player in regulating the shift between single cell tumor dormancy and proliferation, including interactions with fibronectin and integrins. With relevance for the progression of minimal residual disease, the authors discuss the hypothesis that u-PAR might be an essential molecule in bone marrow disseminated tumor cells (DTCs) for long-term survival during dormancy, and in reactivation of their proliferation years after primary treatment.

In his review, Udagawa discusses different mechanisms of tumor dormancy, including cell cycle withdrawal, immune surveillance, and blocked angiogenesis. He also comments on the fact that cancer stem cells and angiogenic cells show some interesting parallels. In an experimental model of human liposarcoma, the angiogenic subclones functionally resemble tumor stem cells with respect to the capacity to expand in immune-deficient mice. Non-angiogenic clones resemble terminally differentiated cells.

Felsher reports on the phenomenon of oncogene addiction and its possible relevance for tumor dormancy. In some tumors, oncogene inactivation results in the elimination of all or almost all tumor cells through apoptosis by oncogene addiction. In other cases, oncogene inactivation predominantly results in terminal differentiation or cellular senescence of tumor cells. Thus, oncogene inactivation can result in a state of tumor dormancy.

Kang & Watnick focus on the intercellular signaling that takes place between tumor cells and the surrounding tumor-associated stroma. Also, these mechanisms have not been extensively studied with regard to the regulation of angiogenesis and tumor dormancy. They describe the importance of tumor-mediated paracrine regulation of stromal Tsp-1 and VEGF expression, and how this is involved in tumor angiogenesis and dormancy. They comment on the significance of PDGF, hormones and nuclear receptors.

Favaro et al. discuss the importance of angiogenesis in the growth promotion of micrometastases and the significance of their microenvironment. They discuss the cross-talk between different cell types in the so-called vascular niche, the vascularized microenvironment feeding or supporting tumor cells. This process includes cell populations recruited from distant sites, such as bone marrow, and mechanisms by which local processes, such as inflammation and trauma, might reactivate dormant tumor cells by inducing an angiogenic spike.

Rak and coauthors comment on the importance of age and the coagulation system for tumor dormancy. They discuss some of the vascular properties of cancer stem cells (CSC) relevant to tumour dormancy and progression, including: (1) the role of CSCs in regulating tumor vascular supply, i.e. the onset and maintenance of tumour angiogenesis; (2) the consequences of changing vascular demand (vascular dependence) of CSCs and their progeny; (3) the interplay between CSCs and the vascular system during the process of metastasis, and especially (4) the impact of the coagulation system on the properties of CSCs and their niches.

Pontier & Muller discuss the roles of integrins in the subclinical persistence and dormancy of breast cancer cells, and the relevance of stem cell regulation. Especially, the importance of Wnt, Notch and c-Myc is described.

Quesnel describes different roles of the immune system in tumor dormancy. An equilibrium between immune responses and tumor cells might lead to long-term tumor dormancy. Experimental models have shown that dormant tumor cells may overexpress B7-H1 and B7.1 and inhibit CTL-mediated lysis. These cells resist apoptosis by methylating SOCS1 and by paracrine production of cytokines. Thus, the delicate equilibrium between tumor cells and the host immune response may be affected by immune escape mechanisms, resulting from active suppression of adaptive immunity, and intrinsic resistance to apoptosis.

In their extensive review, Fukumura & Jain show how imaging of tumor vessels and their interactions with the microenvironment might advance studies of tumor progression and states of dormancy. By intravital microscopy, host-tumor interactions and treatment effects might be examined. Thus, restoring the balance of pro- and anti-angiogenic factors in tumors may ‘‘normalize” tumor vasculature and improve its function.

Hoffman focuses on the visualization of dormancy, proliferation, and death of tumor cells by using fluorescent proteins and high-powered imaging technology to visualize the nuclear-cytoplasmic dynamics of these phenomena in vivo.

In their review, Retsky and co-workers present epidemiological data suggesting the presence of tumor dormancy after breast surgery. To explain the bimodal relapse patterns observed in breast cancer data, the authors propose that metastatic breast cancer growth commonly includes periods of temporary dormancy at both the single cell stage and the avascular micrometastasis stage. They suggest that surgery to remove a primary tumor might terminate dormancy, resulting in accelerated relapses. Such statistical findings might provide a link between experimental studies of tumor dormancy and clinical efforts to improve patient outcome.

Fehm et al. also discuss the concept of tumor dormancy in the setting of breast cancer biology and clinical observations. Tumor cells might disseminate at an early stage, and a subset of these disseminated tumor cells may persist in a state of dormancy. Based on cell culture and animal models, dormancy can occur at two different stages. Single dormant cells are defined as cells with lack of proliferation and apoptosis, whereas the micrometastasis model defines tumor cell dormancy as a state of balanced apoptosis and proliferation of micrometastasis, resulting in no net increase of tumor mass. In this review, the authors summarize findings about different factors for tumor cell dormancy and potential therapeutic implications that may help to reduce metastatic relapse in cancer patients.

Wikman and coauthors focus on the relationship between disseminated tumor cells (DTC) and clinical disease progress in cancer patients. Most likely, the ability of a dormant DTC to be “activated” is a complex process involving: 1. somatic aberrations in the tumor cells, 2. the interaction of DTCs with the new microenvironment at the secondary site, and 3. hereditary components of the host (i.e., cancer patient). The occurrence, detection and characterization of DTCs are described, as well as tumor-host interactions important for tumor dormancy.

In summary, current evidence indicates that the concept of tumor dormancy is important to understand discontinuous disease progress as observed in human cancer. As Weinberg points out, tumor dormancy has “many faces” that represent different mechanisms by which tumors do not progress. This knowledge is fundamental for early tumor detection as well as for patient treatment and management.


  1. Top of page
  2. Acknowledgments

We would like to thank the following persons for excellent help during the work with this special issue of APMIS on Tumor Dormancy: Christine Møller and Hanne Freno at the APMIS editorial office in Copenhagen; Wendy Foss, Pauline Breen, Kristin Johnson and other staff members at the Vascular Biology Program, Children's Hospital Boston and Harvard Medical School, Boston; staff members at The Gade Institute, University of Bergen, and Haukeland University Hospital, Bergen.