Repositioned to kill stem cells

Article metrics


Chemotherapy-resistant cancer stem cells make it hard to cure many forms of the disease. Repositioning an existing drug to tackle this problem could significantly improve treatment for one form of leukaemia. See Letter p.380

In most cases of chronic myeloid leukaemia (CML), a daily oral medication can rapidly transform a progressive and ultimately fatal cancer into a chronic but manageable condition. But this is not a cure. The persistence of quiescent (dormant, non-cycling) and thus drug-resistant leukaemic stem cells (LSCs) poses an unmet clinical challenge, and any attempt to cure CML must target the eradication of these cells. In this issue, Prost et al.1 (page 380) present provocative preclinical and early clinical findings demonstrating that a drug currently used for diabetes therapy can be repositioned to target a pathway that controls quiescence in LSCs, causing the gradual erosion of this cellular pool.

The cause of CML is a mutation in a normal blood stem cell involving an exchange of genetic material between chromosomes 9 and 22. This translocation creates a cancer-driving gene known as BCR–ABL1, which produces a protein with enhanced activity as a tyrosine kinase enzyme, leading to uncontrolled cell proliferation. BCR–ABL1 has been shown to be sufficient to drive the development of leukaemia in mouse models2, and the discovery of this protein led to the development of tyrosine kinase inhibitors (TKIs) for CML treatment.

In the past two decades, TKIs have dramatically improved the outcome for people with this cancer. Most of those who present with early disease respond rapidly to TKI therapy and go into long-lasting remission. However, TKIs fail to eradicate LSCs, the cells that initiate and maintain CML, and these drug-resistant cells can drive relapse, or evolve to cause further forms of TKI resistance and more-aggressive disease. As a result, people on life-long TKI therapy are exposed to associated, often serious, side effects and may cease to respond to the treatment at any time. Furthermore, the significantly improved survival for those taking TKIs means that the prevalence of CML is increasing each year, with inherent social and economic implications.

Several potential mechanisms to explain the insensitivity of LSCs to TKIs have been proposed, including cellular quiescence. Prost et al. report that quiescence in LSCs is regulated by a pathway involving the receptor PPARγ, the transcription factors STAT5 and HIF2α, and the protein CITED2, which is known3 to regulate blood stem-cell quiescence (Fig. 1). A particular strength of the study was the use of primary blood stem cells (expressing the marker CD34) from people with CML to dissect the pathway and confirm the role of each component in regulating LSC quiescence.

Figure 1: Targeting leukaemic stem cells in chronic myeloid leukaemia.

Prost et al.1 describe a molecular pathway, involving the receptor PPARγ, the transcription factors STAT5 and HIF2α, and the regulatory protein CITED2, that induces leukaemic stem cells (LSCs) to enter a dormant (quiescent) state. They also show that the drug pioglitazone, approved for diabetes treatment, activates PPARγ to block this pathway, and can kill these cells when used in conjunction with tyrosine kinase inhibitors (TKIs), which inhibit the protein BCR–ABL1 and thus STAT5, and which are the standard therapy against active (cycling) leukaemic cells. Several drugs used to treat other diseases, such as axitinib, arsenic trioxide and hydroxychloroquine, have also been repositioned to treat chronic myeloid leukaemia, but these have different mechanisms of action.

The authors go on to show that combining imatinib, the standard TKI used to manage CML, with the antidiabetic agent pioglitazone, which activates PPARγ, blocks this pathway in CML cells. The synergistic effects of the drugs reduce STAT5 expression and activity, downregulate HIF2α and CITED2 expression, and trigger the death of quiescent LSCs. Although the mechanism by which LSCs are killed in response to this drug combination is not clear, they are probably either killed directly or driven to exit quiescence, which may lead to their eradication by the TKI. The authors also demonstrate that the compound JQ1, a bromodomain inhibitor with broad activity that includes the suppression of STAT5 activity, is as effective as pioglitazone (in combination with imatinib). Although this finding supports a role for the STAT5 pathway in LSC quiescence, the door is still open for studies of other agents that may target LSCs through this or alternative pathways.

Collectively, these results strengthen the concept that cancer stem cells exhibit vulnerabilities in otherwise normal molecular pathways that may be targeted in a selective manner to obtain a cure. Earlier work demonstrated that CML stem-cell quiescence is in part maintained by the promyelocytic leukaemia tumour-suppressor protein, which can be targeted by arsenic trioxide4, and that the cellular process of autophagy functions as a survival pathway for CML stem cells that can be targeted by repositioning the antimalarial agent hydroxychloroquine5 (Fig. 1). Both of these approaches are currently under investigation in the clinic.

Prost et al. also tested the addition of pioglitazone to imatinib therapy in three people with CML, and found that they converted from having demonstrable residual leukaemia to being disease-free. The effect lasted for months to years after pioglitazone treatment ceased. These data provided a strong rationale for a phase II clinical trial, which started in July 2009 (ACTIM EudraCT 2009-011675-79). Although the interim results from this trial are encouraging, the study is non-randomized, so it will be difficult to ascertain definitively that improved response rates are driven by pioglitazone.

Despite the need for further clinical testing of this combination therapy, Prost et al. have demonstrated the substantial potential for drug repositioning in CML research. Their results follow a recent report6 in which axitinib, a TKI approved for the treatment of drug-resistant renal-cell cancer, was repositioned to tackle TKI resistance in CML. Using drugs that have already been approved for other purposes can shorten the drug-development pathway by 5–10 years and reduce risks and costs.

Although drug repositioning can be rather serendipitous, Prost and colleagues had a tangible rationale that PPARγ activators such as pioglitazone warranted investigation in CML on the basis of their observation7 of the drugs' activity against a cell-line model of the disease. Already around 30% of drugs newly approved for a particular treatment have been repositioned from another therapy, and such hypothesis-driven repositioning strategies are likely to become more common in cancer drug discovery. This figure is set to rise further as our understanding of cellular pathways and processes increases and we include innovative computational approaches to facilitate disease-, drug- and treatment-oriented drug repositioning. It is clear that repositioning will increasingly help the fast-tracking of drugs into the clinic. As demonstrated by Prost and colleagues, this could soon signal the beginning of the end for stem-cell quiescence in CML and other cancers.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Prost, S. et al. Nature 525, 380–383 (2015).

  2. 2

    Daley, G. Q., Van Etten, R. A. & Baltimore, D. Science 247, 824–830 (1990).

  3. 3

    Kranc, K. R. et al. Cell Stem Cell 5, 659–665 (2009).

  4. 4

    Ito, K. et al. Nature 453, 1072–1078 (2008).

  5. 5

    Bellodi, C. et al. J. Clin. Invest. 119, 1109–1123 (2009).

  6. 6

    Pemovska, T. et al. Nature 519, 102–105 (2015).

  7. 7

    Prost, S. et al. J. Clin. Invest. 118, 1765–1775 (2008).

Download references

Author information

Correspondence to Tessa Holyoake or David Vetrie.

Related links

Related links

Related links in Nature Research

Cancer: Persistence of leukaemic ancestors

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Holyoake, T., Vetrie, D. Repositioned to kill stem cells. Nature 525, 328–329 (2015) doi:10.1038/nature15213

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.