Mistaken identity

Cancer is really a stem cell disease, argue George Murphy and Malcolm R Alison

"Mrs Smith, I'm afraid you have cancer." The words die on your lips as you wait to see how she reacts-a bolt from the blue or just official confirmation of a long nursed suspicion? She's petrified-no diagnosis scares her more, with some justification: one in four Europeans will die of some form of cancer, despite mean five year survival rates of more than 45%.^w1 But what if you could treat cancer, any cancer, as simply and effectively as a bacterial infection? Such a world may be closer than you think, and is coming from a surprising direction-stem cells.

Dynamic tissues

Tissues such as the gut epithelium and skin are constantly regenerated from tissue specific stem cells. Moreover, even those tissues which are not routinely renewed, such as neurones, appear to have previously unrecognised populations of stem cells that may be involved in maintenance and repair.^w2

So, human tissues are not static collections of similar cells but are arranged instead as a hierarchy,^w3 based on the cells' replicative potential and functional abilities. At the top of this hierarchy are stem cells (see box 1). This small population of self renewing cells generates the next group-transit amplifying cells.

These cells, as their name suggests, divide rapidly for a short time, before producing mature, fully (terminally) differentiated cells, which cannot reproduce further. The advantage of this is that it allows for a massive increase in cell numbers while protecting the "template" DNA of the stem cells from excessive copying, with the accompanying risk of mutation (see figure). This explosive replicative potential is controlled by the niche that the stem cells reside in, through a variety of signalling pathways.

Is cancer a biological system?

This change in the way we view normal tissues is mirrored by our changing idea of cancer. Although it was noted in the 1930s that a single transplanted malignant leucocyte could produce leukaemia,^w4 it took the development of techniques of modern molecular biology to characterise the cell involved. This raises the question-are all leukaemic cells the same, or was this cell different to most?

Rationally, we might expect cancer to develop from cells high up the hierarchy of potential. We know that it is an acquired genetic disease requiring multiple mutations to develop. However, most transit amplifying and terminally differentiated cells are too short lived to gain such mutations. Unless an initial mutation allowed them to escape their usual fate, cancer is much more likely to develop from stem cells.

Cancer is also more likely to adopt the normal proliferative pathway, rather than creating a new system of its own.

In other words, we might expect cancer to develop from mutated stem cells and to depend on a small population of malignant stem cells to make most of the cells in the tumour. If this is so, then rather than aiming chemotherapy at the bulk of the tumour cells, as at present, we should target the cancer stem cells. So far so logical, but what is the evidence to support this?

SL-ICs, dyes, and videotape

The first evidence came from leukaemias, which are some of the easier cancers to study. Patients with chronic myelogenous leukaemia have the same chromosomal abnormality in all their leukaemic cells; this suggests that all the leukaemic cells are descended from one cell that initially mutated, and that original cell could rapidly expand to form the leukaemic cell population. This common origin is called clonality, and fits with the known, stepwise clinical progression of chronic myelogenous leukaemia. Evidence of clonality has also been found in acute myelo­genous leukaemia, by comparing the pattern of X linked gene inactivation in the leukaemic cells with the normal haematopoietic cells.^w5

The strongest evidence has been the discovery that only a small subset of leukaemic cells can proliferate extensively, both in vitro^w6 and in vivo.^w7 For acute myelogenous leukaemia, this subset has been serially purified and transplanted through successive immunosuppressed (non-obese diabetic, NOD, and severe combined immunodeficient, SCID) mice, forming leukaemias with the same multiple cell types as the original cancer. This subset has been named SCID leukaemia initiating cells (SL-ICs), and in common with normal haematopoietic stem cells always has CD34+ and CD38- surface antigens, although other markers distinguish the two. No other leukaemic cell type could form new leukaemias.^w7 This clearly shows that SL-ICs have both the function and the physical markers of stem cells, and are responsible for initiating and maintaining leukaemia. However, it does not prove whether SL-ICs develop from normal stem cells or from some other cell type.

Epithelial cells are the origin of more than 90% of human cancers,^w8 including all of the big four in Britain-breast, lung, bowel, and prostate). As mentioned above, many epithelia are constantly renewed from stem cells; unsurprisingly, stem cells have also been implicated in their cancers.

In breast cancer, as with acute myelogenous leukaemia, only a few human breast tumour cells could produce new breast tumours in immunosuppressed mice; this was true for a variety of types of breast cancer. These cancer stem cells had low CD44+ and CD24- surface markers, and as few as 100 cells could produce a new tumour. These daughter tumours had the same variety of cell types as the parent tumour, and the stem cells could be extracted by cell sorting (see box 2) and cause new tumours in other mice. By contrast, tens of thousands of tumour cells with different surface antigens were unable to form new tumours.^w10

More recently, it has been shown that a single mouse mammary stem cell can generate a complete functional mammary gland. A connection between these normal stem cells and their cancerous cousins is suggested by the expanded number of stem cells in premalignant breast lesions.^w11

Similar tumour analysis experiments have found cancer stem cells in adenocarcinoma of the bowel^w12 and in both adult and paediatric forms of the brain tumours medulloblastoma and glioblastoma.^w13 A slightly different model, using a viral vector to activate oncogenes in genetically modified mice, has shown that the normal stem cell population expands and mutates in adenocarcinoma of the lung.^w14 Similarly, infection of prostatic stem cells with an oncogenic lentivirus caused carcinoma in situ; no other prostatic cells were transformed.^w15 Both these experiments show normal stem cells becoming cancer stem cells in an experimental system, although this has yet to be shown in clinical lesions.

The role of cancer stem cells in metastasis is another interesting aspect of the field, although this is more poorly understood. It has been suggested that only cancer stem cells can successfully metastasise to distant sites. If so, this would explain the presence of "micrometastases" (distant tumour cells that fail to grow), which must be more differentiated cells that have escaped into the bloodstream.^w16 It would also be consistent with the known ability of normal haematopoietic stem cells to traffic around the body, implanting and growing at distant sites.^w3 This subversion of a normal pathway might explain recent data that indicates that metastasis is an early genetic event in the development of tumours.^w17

So, in summary, there is strong evidence that many of the most globally prevalent cancers are dependent on a stem cell population for their growth and survival, and there is moderate evidence for a number of other cancers. It also seems probable that these cancer stem cells develop from normal stem cells, and possibly that these cells are the sole cause of clinically important metastasis, although the evidence here is weaker. But what does this mean for their treatment?

Magic bullets

Promises of a wonder cure for the illness of the age should provoke suspicion, but despite the relatively early stage of the research, there are definite grounds for optimism. Most current chemotherapy targets rapidly dividing cells, but stem cells rarely divide, leaving the rapid replication to expendable transit amplifying cells. This missed target effect has been modelled mathematically for treatment of chronic myelogenous leukaemia with imatinib and predicts the observed clinical course of the illness. Imatinib initially kills all leukaemic cells except the stem cells, but once treatment is stopped or resistance develops, the stem cells cause a rapid relapse of the disease.^w18

If we can pinpoint the cancer stem cells, then we should be able to rationally design treatments that will kill them, with minimal damage to healthy tissues. The most obvious targets are the surface antigens on individual cancer stem cells; several have been found that set them apart from normal stem cells. These antigens could be targeted by a monoclonal antibody, in a similar fashion to trastuzumab (Herceptin). Such a magic bullet could be as well targeted as an antibiotic.

An alternative approach might exploit the niche environment, which nurtures and controls stem cells. Communication between the niche and normal stem cells is necessary and involves direct physical inter­action and local signalling, through classical pathways such as Wnt, Notch, and JAK-STAT. Also, in some cases the niche receives direct neural input, and stem cells can respond to local metabolic factors, such as the concentration of Ca2+ ions.^w19 A clearer knowledge of how these regulatory systems normally function and how they are subverted in particular cancers may provide new therapeutic approaches.

Treatment of the future

So, what may await future patients such as Mrs Smith? We can imagine a world where the initial stages are similar, as she presents and a differential diagnosis of cancer is made. Currently, a biopsy would be taken, although how this is done may vary. However, this would do more than confirm the cancer and identify its stage. The key aim will be to get samples of the cancer stem cells for analysis; not just of their genotype but also of the surface antigens they display. In future, this may allow treatments such as highly targeted monoclonal antibodies to be prescribed,^w20 leading to a complete cure. Also, such a cure could be accomplished without needing either invasive surgery or harmful radiation and with a minimum of drug side effects. Such a happy outcome may seem distant as you tread the wards, but it is plausible and rationally based, and it is coming.

Competing interests: None declared.

George Murphy, final year medical student, Bart's and the London, Queen Mary's School of Medicine and Dentistry, London
Email: grfm2@cantab.net
Malcolm R Alison, professor of stem cell biology,, Centre for Diabetes and Metabolic Medicine, Queen Mary's School of Medicine and Dentistry


studentBMJ 2007;15:1-44 January ISSN 0966-6494

1. Cancer in the EU. London: Cancer Research UK, 2004. http://info.cancerresearchuk.org/cancerstats/geographic/cancerineu/survival/?a=5441 (accessed 12 Aug 2006).
2. Adams GB, Scadden DT. The hematopoietic stem cell in its place. Nature Immunol 2006;7:333-7.
3. Alison MR, Poulsom R, Forbes S, Wright NA. An introduction to stem cells. J Pathol 2002;197:419-23.
4. Furth J, Kahn MC. The transmission of leukaemia of mice with a single cell. Am J Cancer 1937;31:276-82.
5. Huntly BJP, Gilliland. Leukaemia stem cells and the evolution of cancer-stem-cell research. Nat Rev Cancer 2005;5:311-21.
6. Hamburger AW, Salmon SE. Primary bioassay of human tumor stem cells. Science 1977;197:461-3.
7. Wang JCY, Dick JE. Cancer stem cells: lessons from leukaemia. Trends Cell Biol 2005;15:494-501.
8. Miller SJ, Lavker RM, Sun T. Interpreting epithelial cancer biology in the context of stem cells: tumor properties and therapeutic implications. Biochim Biophys Acta 2005;1756:25-52.
9. Hirschmann-Jax C, Foster AE, Wulf GG, Goodell MA, Brenner MK. A distinct side population of cells in human tumour cells. Cell Cycle 2005;4:203-5.
10. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 2004;100:3983-8.
11. Shackleton M, Vaillant F, Simpson KJ, Stingl J, Smyth GK, Asselin-Labat M-L, et al. Generation of a functional mammary gland from a single stem cell. Nature 2006;439:84-8.
12. Haraguchi N, Utsunomiya T, Inoue H, Tanaka F, Mimori K, Barnard GF, et al. Characterization of a side population of cancer cells from human gastrointestinal system. Stem Cells 2006;3:506-13.
13. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, et al. Identification of human brain tumour initiating cells. Nature 2004;432:396-401.
14. Bender Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I, Vogel S, et al. Identification of bronchoalveolar stem cells in normal lung and lung cancer. Cell 2005;121:823-35.
15. Lam JS, Reiter RE. Stem cells in prostate and prostate cancer development. Urol Oncol 2006;24:131-40.
16. Waterworth A. Introducing the concept of breast cancer stem cells. Breast Cancer Res 2004;6:53-4.
17. Weigelt B, Peterse JL, van’t Veer LJ. Breast cancer metastasis: markers and models. Nat Rev Cancer 2005;5:591-602.
18. Michor F, Hughes TP, Iwasa Y, Branford S, Shah NP, Sawyers CL, et al. Dynamics of chronic myeloid leukaemia. Nature 2005;435:1267-70.
19. Scadden DT. The stem-cell niche as an entity of action. Nature 2006;441:1075-9.
20. Cheson BD. Monoclonal antibody therapy for B-cell malignancies. Semin Oncol 2006;33(suppl):s2-14.