© 2005 Nature Publishing Group
The discovery of cancer stem cells in solid tumours has
changed our view of carcinogenesis and chemotherapy.
One of the unique features of the bone-marrow stem
cells that are required for normal haematopoiesis is their
capacity for self-renewal.In the haematopoietic system,
there are three different populations of multipotent
progenitors — stem cells with a capacity for long-term
renewal, stem cells with a capacity for short-term
renewal,and multipotent progenitors that cannot renew
but differentiate into the varied lineages in the bone mar-
row1–3.The multipotent progenitors and their derived
lineages undergo rapid cell division,allowing them to
populate the marrow.The factors that determine the
self-renewing capacity of a cell,and how cancer cells
acquire this ability,are not yet understood.
Pluripotent stem cells that possess both self-renewal
capabilities and the ability to generate an organ-specific,
differentiated repertoire of cells exist in organs other
than the haematopoietic system and these can be stud-
ied to gain better insight into the stem-cell biology of a
tumour. The concept of organ stem cells is difficult
when one considers the many different cell types and
functions of an organ,but emerging evidence indicates
such pluripotent stem cells exist.In the normal mam-
mary gland,for example,three cell lineages have been
described — myoepithelial cells that form a basal cell
layer,ductal epithelial cells,and milk-producing alveolar
cells4.Although transplantation studies in mice have
demonstrated that most mammary cells have a limited
capacity for self-renewal,clonal populations that can
recapitulate the entire functional repertoire of the
gland have been identified5.In an elegant study,human
mammary epithelial cells derived from reduction
mammoplasties were used to generate non-adherent
spheroids (designated mammospheres) in cell culture
and demonstrate the presence of the three mammary
cell lineages.More importantly,the cells in the mam-
mospheres were clonally derived,providing evidence
for a single pluripotent stem cell4. These same
approaches are being used to isolate and characterize
breast cancerstem cells.
In the haematopoietic system as well as in other
normal tissues, the normal stem cell must be both
self-renewing and pluripotent.Although stem cells
can self-renew,they are generally quiescent,spending
most of their time in G0.Because stem cells can repair
their DNA as they self-renew,they have the potential
to accumulate mutations acquired after exposure to
carcinogens. If tumours arise from stem cells, the
accumulation of these mutations might be what we
have come to recognize as the ‘multistep process of
carcinogenesis’. So do cancer stem cells arise from
normal stem cells,or do they arise from differentiated
cells that acquire self-renewal capacity,or both? Does
the innate resistance of normal stem cells to radiation
and toxins contribute to the failure of some cancer
TUMOUR STEM CELLS AND
Michael Dean*,Tito Fojo‡and Susan Bates‡
Abstract | The contribution of tumorigenic stem cells to haematopoietic cancers has been
established for some time, and cells possessing stem-cell properties have been described in
several solid tumours. Although chemotherapy kills most cells in a tumour, it is believed to leave
tumour stem cells behind, which might be an important mechanism of resistance. For example,
the ATP-binding cassette (ABC) drug transporters have been shown to protect cancer stem cells
from chemotherapeutic agents. Gaining a better insight into the mechanisms of stem-cell
resistance to chemotherapy might therefore lead to new therapeutic targets and better
NATURE REVIEWS | CANCER
VOLUME 5 | APRIL 2005 | 275
Correspondence to M.D.
R E V I E W S
© 2005 Nature Publishing Group
Malignant germ-cell tumours
that exhibit cell phenotypes that
are derived from more than one
of the three primary germ-cell
276 | APRIL 2005 | VOLUME 5
R E V I E W S
‘Only’7 of these autotransplants resulted in tumour
growth at the injection site.Furthermore,studies of
acute myelogenous leukaemia have shown that only
0.1–1% of all cells have leukaemia-initiating activity9.
These leukaemia-initiating cells have many markers
and properties of normal haematopoietic stem cells10,11.
So it is believed that leukaemia arises from a stem cell
that becomes transformed and gives rise to a large pop-
ulation of clones that proliferate but cannot self-renew
or fully differentiate12. Similar populations of self-
renewing cells,such as those that carry the chromoso-
mal translocation t(9;22)(q34;q11),which forms the
BCR–ABL fusion gene, have also been identified in
patients with chronic lymphocytic leukaemia and
chronic myelogenous leukaemia(CML)13.
Evidence for the existence of a pluripotent cell in
solid tumours includes clinical observations with
human TERATOCARCINOMAS,an experiment of nature in
which differentiated tissues such as muscle and bone
can appear in the tumour mass14–16,and from the obser-
vation that mouse teratocarcinoma cells can produce a
normal mouse17.Instead of haematopoietic markers,
stem cells identified from solid tumours usually express
organ-specific markers.In eight of nine human breast
cancer samples,for example,a tumorigenic stem-cell
population was found that expressed the unique cell-
surface marker profile CD44+CD24–/lowLin– (REFS 4,18).
This population was enriched 50- to 100-fold with cells
able to form tumours in mice.The resulting tumours
possess the phenotypic heterogeneity found in the
original tumour population,including both tumori-
genic and non-tumorigenic cells. In another study,
overexpression of the WNT family of genes,important
regulators of normal cell development,led to expansion
of the mammary-stem-cell pool and cancer susceptibil-
ity19.Finally,stem cells with a capacity to self-renew and
undergo pluripotential differentiation have been iso-
lated from human central-nervous-system tumours20–22.
These cells were reported to express CD133— a cell-
surface antigen known originally as a marker of
haematopoietic stem cells and later observed as a
marker ofstem cells in other normal tissues23–26.
The exact origin ofpluripotent stem cells in tumours
might vary.They could arise from the malignant trans-
formation of a normal stem cell that has accumulated
oncogenic insults over time.Alternatively,the original
therapies? How can we exploit our knowledge of
stem-cell biology to specifically target these cells and
Cancer stem cells
Cells with stem-cell qualities have been identified in
malignancies ofhaematopoietic origin and in some solid
tumours. The existence of such a population would
imply that the stem cell represents the cell of origin for
the tumour,as illustrated in FIG.1.One can predict that
such cancer stem cells represent only a small fraction ofa
tumour,as they possess the capability to regenerate a
tumour,and most cancer cells lack this regenerative capa-
bility.For example,when plated in soft agar or injected
into mice, most tumour cells do not give rise to
colonies6,7. Similarly, in experiments performed in
humans in the 1950s,unthinkable by today’s ethical stan-
dards,35 patients had an estimated one billion of their
own tumour cells injected into their thigh or forearm8.
• Stem-cell populations have been identified in a range ofhaematopoietic and solid tumours,and might represent the
cell oforigin ofthese tumours.
• Normal and cancer stem cells express high levels ofATP-binding cassette (ABC) transporters,such as ABCB1,which
encodes P-glycoprotein,and the half-transporter ABCG2,which was originally identified in mitoxantrone-resistant cells.
• The drug-transporting property ofstem cells conferred by ABC transporters is the basis for the ‘side-population’
phenotype that arises from the exclusion ofthe fluorescent dye Hoechst 33342.
• Cancer stem cells are likely to share many ofthe properties ofnormal stem cells that provide for a long lifespan,
including relative quiescence,resistance to drugs and toxins through the expression ofseveral ABC transporters,an
active DNA-repair capacity and a resistance to apoptosis.Therefore,tumours might have a built-in population of
drug-resistant pluripotent cells that can survive chemotherapy and repopulate the tumour.
↓ Cell death
↑ Immune evasion
Mutation in stem cell
Additional mutation(s) in stem cell
or in progenitor cells
Figure 1 | Cancer stem cells and tumour progression. Normal stem cells give rise to
multipotent progenitor cells, committed progenitors and mature, differentiated cells. Mutations in
a stem cell give rise to a stem cell with aberrant proliferation and result in a pre-malignant lesion.
Additional mutations lead to the acquisition of further increased proliferation, decreased
apoptosis, evasion of the immune system, and further expansion of the stem-cell compartment
that is typical of malignant tumours.
© 2005 Nature Publishing Group
NATURE REVIEWS | CANCER
VOLUME 5 | APRIL 2005 | 277
R E V I E W S
long lifespan such as relative quiescence,resistance to
drugs and toxins through the expression of several ATP-
binding cassette (ABC) transporters,an active DNA-
repair capacity,and a resistance to apoptosis.It follows
that cancer stem cells might also possess these resistance
mechanisms.The paradigm that drug resistance origi-
nates in the stem-cell phenotype might stimulate new
strategies for the development ofanticancer therapies.
Drug transporters in stem cells
Stem cells have many properties that separate them
from mature,differentiated cells.In addition to their
ability to self-renew and differentiate,they are quies-
cent,dividing infrequently.They also require specific
environments comprising other cells, stroma and
growth factors for their survival28.One particularly
intriguing property of stem cells is that they express
high levels of specific ABC drug transporters. For
example,haematopoietic stem cells express high levels
of ABCG2,but the gene is turned off in most commit-
ted progenitor and mature blood cells29.The two ABC-
transporter-encoding genes that have been studied
most extensively in stem cells are ABCB1, which
encodes P-glycoprotein30,and ABCG2 (REFS 29,31–34).
Along with ABCC1,they represent the three principal
MULTIDRUG-RESISTANCE genes that have been identified in
tumour cells.These genes,members of the ABC-trans-
porter superfamily,are promiscuous transporters of
both hydrophobic and hydrophilic compounds30,35
(TABLE 1). These transporters also have important roles
in normal physiology in the transport of drugs across
the placenta and the intestine (more accurately,the
retention of drugs in the intestinal lumen),and are
important components of the blood–brain and
blood–testis barriers. By using the energy of ATP
hydrolysis, these transporters actively efflux drugs
from cells, serving to protect them from cytotoxic
agents35–37. Mice deficient in either Abcg2, Abcb1 or
Abcc1 are viable, fertile and have normal stem-cell
compartments36,38,39.This indicates that none of these
tumour cell could be a more differentiated cell that
develops the capacity for continual self-renewal,thus
acquiring the properties of a stem cell27.Distinguishing
between these two might be difficult.Evidence that cells
other than stem cells can acquire the ability to undergo
self-renewal has been recently provided in studies
examining the progression of CML13. The chronic
phase of the disease occurs when a stem cell acquires
the expression of the BCR–ABL fusion protein,leading
to increased proliferation of cells within the granulo-
cyte–macrophage progenitor pool and their down-
stream progeny.It is hypothesized that progression to
BLAST CRISIS follows additional genetic or epigenetic
events that confer progenitor cells with the capacity to
self-renew, making them indistinguishable from a
leukaemic stem cell.Further proof is needed to confirm
that progression to blast crisis occurs at the level of the
progenitor pool,but the proposal that the stem-cell
compartment is not rigidly defined is attractive and
suggests a degree of plasticity in cancer.
Cancer stem cells (with either inherent or acquired
capabilities for self-renewal) give rise to cells that lack
long-term self-renewal capability but retain a finite abil-
ity to divide.In normal physiology,this would be called
‘differentiation’,as the cell acquires traits specific to its
place in the tissue.But in cancer,cells lack the ability to
undergo differentiation into phenotypically mature cells.
A limited amount of differentiation often does occur,
giving rise to the well-known histopathological and mol-
ecular distinctions between tumours.In fact,the further
along this pathway the cancer cell travels,the more dif-
ferentiated and the more like its normal counterpart it
becomes,accordingly demonstrating a slower growth
rate.Where the so-called ‘de-differentiated’tumours fit
along this continuum is uncertain,but it is possible that
self-renewal might be a property that represents a higher
Therefore,the cancer stem cell shares many proper-
ties of the normal stem cell.It is generally accepted that
normal stem cells show properties that provide for a
In patients with chronic
term describes the progression
of the disease to an acute
advanced phase,evidenced by an
increased number of immature
white blood cells in the
Simultaneous resistance to
several structurally unrelated
drugs that do not necessarily
have a common mechanism of
Table 1 | ABC transporters involved in drug resistance
Chemotherapeutic drugs effluxed by transporter
Colchicine, doxorubicin, etoposide,vinblastine, paclitaxel
Doxorubicin, daunorubicin, vincristine, etoposide,
colchicine, camptothecins, methotrexate
Vinblastine, cisplatin, doxorubicin, methotrexate
6-mercaptopurine, 6-thioguanine and
6-mercaptopurine, 6-thioguanine and metabolites
Mitoxantrone, topotecan, doxorubicin, daunorubicin,
irinotecan, imatinib, methotrexate
Other drugs and substrates
PMEA, cAMP, cGMP
PMEA, cAMP, cGMP
PMEA, cAMP, cGMP
Pheophorbide A, Hoechst
ABC, ATP-binding cassette; BCRP, breast cancer resistance protein; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanine
monophosphate; MDR, multidrug resistance; MRP, multidrug-resistance-associated protein; MXR, mitoxantrone resistance protein;
© 2005 Nature Publishing Group
278 | APRIL 2005 | VOLUME 5
R E V I E W S
addition,in a range of cell lines,differentiating agents
induce expression of ABCB1,inhibit cell growth,and
increase the expression of markers of maturation53,54.
Additional limitations exist in using cancer cell lines
cultured in vitro to study stem-cell biology and drug
resistance.Although SP cells and cells with stem-cell
properties have been reported in cultured cell lines,it is
difficult to reconcile the hypothesis that only a small
fraction of cells in culture possess stem-cell characteris-
tics with the rapid doubling time of cells in culture.
Current paradigms envision a small stem-cell compart-
ment possessing cells with the capacity for perpetual
self-renewal existing alongside a much larger prolifera-
tive compartment with cells that have a finite ability to
proliferate before presumably arresting and/or undergo-
ing apoptosis. These paradigms can explain the low
cloning efficiency of most cell lines,their inefficiency at
colony formation in soft agar,and their limited tumori-
genicity.However,none of these models can explain
how the stem cells remain a constant fraction of the
total population,if indeed they do.Any proposal will
require stem cells to divide slowly,and must recognize
that in a cell line derived from a solid tumour the num-
ber of cells undergoing apoptosis is relatively small.One
possibility is that there is an interchange ofcells between
a proliferative compartment and the stem-cell pool.
That such an interchange might occur is not improba-
ble,as the cell line almost certainly originated from a
stem cell with a proliferative advantage.As discussed
above,plasticity has already been proposed in CML.
Drug resistance in cancer cells
Cancer cells can acquire resistance to chemotherapy by a
range of mechanisms,including the mutation or over-
expression of the drug target,inactivation of the drug,
or elimination of the drug from the cell. Typically,
tumours that recur after an initial response to
chemotherapy are resistant to multiple drugs (they are
multidrug resistant).In the conventional view of drug
resistance,one or several cells in the tumour population
acquire genetic changes that confer drug resistance
(FIG.2a).These cells have a selective advantage that allows
them to overtake the population of tumour cells follow-
ing cancer chemotherapy. Based on the tumour-
stem-cell concept,an alternative model posits that the
cancer stem cells are naturally resistant to chemotherapy
through their quiescence,their capacity for DNA repair,
and ABC-transporter expression (FIG.2b).As a result,at
least some of the tumour stem cells can survive
chemotherapy and support regrowth of the tumour.In
a third model ofacquired resistance,drug-resistant vari-
ants of the tumour stem cell or its close descendants
arise,producing a population of multidrug-resistant
tumour cells that can be found in many patients who
have recurrence oftheir cancer following chemotherapy
(FIG.2c).The same mechanisms that allow stem cells to
accumulate mutations over time,producing the long-
term consequences of exposure to irradiation or car-
cinogens,would then allow cancer stem cells to accu-
mulate mutations that confer drug resistance to their
abnormally developing offspring1. As an example,
genes are required for stem-cell growth or maintenance.
However,these knockout mice are more sensitive to the
effects of drugs such as vinblastine,ivermectin,topote-
can and mitoxantrone,consistent with a role for these
ABC transporters in protecting cells from toxins.
The drug-transporting property of stem cells con-
ferred by these ABC transporters is an important marker
in the isolation and analysis ofhaematopoietic stem cells.
Most cells accumulate the fluorescent dyes Hoechst
33342 and rhodamine 123,but stem cells do not,as these
compounds are effluxed by ABCG2and ABCB1,respec-
tively.Because they don’t accumulate these fluorescent
dyes,stem cells can be sorted by collecting cells that con-
tain only a low level of Hoechst 33342 fluorescence.
These cells are referred to as ‘dull cells’or ‘side popula-
tion’(SP) cells.The term side population was coined
because during flow-cytometry analysis, SP cells are
visualized as a negatively stained ‘side population’to one
side ofthe majority ofcells on a density dot plot.A large
fraction ofhaematopoietic stem cells are found in the SP
fraction40and when isolated from mice and transplanted
into irradiated mice,small numbers ofthese SP cells can
reconstitute the bone marrow,demonstrating that these
cells are pluripotent.SP cells can be isolated from many
tissues including the brain,breast,lung,heart,pancreas,
testes,skin and liver,and these cells might represent lin-
eage-specific stem cells40–48.Hoechst-33342 staining of
bone marrow from ABCG2-null mice fails to detect SP
cells.However,the lack ofstaining for SP cells occurs not
because these cells are absent,but because the lack of
ABCG2 expression allows these cells to accumulate
Hoechst dye and become fluorescent.
SP cells in tumours and cell lines
Once it was recognized that stem cells were predomi-
nantly found in the SP fraction,it became possible to
sort and purify stem cells from virtually any popula-
tion of cells or tissue.SP cells were identified in 15 of
23 neuroblastoma samples and in neuroblastoma,
breast cancer, lung cancer and glioblastoma cell
lines49.Furthermore,analysis of several cell lines that
had been maintained in culture for long periods of
time demonstrated a small population of SP cells.In
the rat glioma C6 cell line, a population of SP cells
was seperated from a population of non-SP cells.
Through the use of growth factors, investigators
maintained these cells in culture, and showed that
only the SP cells gave rise to both populations and
produced cells with both neuronal and glial markers
that were tumorigenic in mice50. This latter study
provided strong evidence that in this cell line the SP
population reflected a population with a capacity for
self-renewal and limited maturation. However, this
isolation approach is imperfect as the SP compart-
ment is composed of stem and non-stem cells, and
some stem cells are not in the SP fraction38. For
example, non-stem-cell tumour cells often express
ABCG2 andABCB1.These genes are highly expressed
in drug-resistant cells,and histopathological studies
have reported increased expression of the ABCB1
transporter in more differentiated tumours51,52. In
© 2005 Nature Publishing Group
NATURE REVIEWS | CANCER
VOLUME 5 | APRIL 2005 | 279
R E V I E W S
cancer stem cell persists in the committed,abnormally
developing progenitors that comprise the proliferative
pool ofcancer cells.
So in the cancer-stem-cell model of drug resistance,
tumours have a built-in population of drug-resistant
pluripotent cells that can survive chemotherapy and
regrow.Again,a parallel with normal stem cells can be
found in stem-cell-driven recovery of normal tissues
following chemotherapy.The rapid relapse observed
with some tumours, at times within one cycle of
chemotherapy, has a normal-tissue parallel in the
repopulation of the bone marrow by normal
haematopoietic stem cells and the recovery of the
mucosa of the gastrointestinal tract,both of which usu-
ally occur within one 3-week cycle.Similarly,tumour
recurrences that occur months to years after an original
response to chemotherapy can be modelled on the
slower recovery that is observed with hair follicles57,58
Although it is therapeutically attractive,the hypothe-
sis that the intrinsic properties of stem cells alone
provide the basis for drug resistance might be too sim-
plistic.Recent studies of imatinib (Glivec) resistance in
patients with leukaemia provide an example of how
ABC-transporter-mediated efflux in stem cells could
facilitate,but not be solely responsible for,the acquisi-
tion of acquired mechanisms of drug resistance.
Imatinib has been recently shown to be both a substrate
and inhibitor ofABCG2,making it susceptible to efflux
by a stem cellthat expresses this ABC transporter59–61.
The initial studies that reported imatinib-resistant
leukaemia cells described ‘acquired’mutations in the
kinase domain of ABL in patients with CML or with
acute lymphoblastic leukaemia associated with
t(9;22)(q34;q11).These findings indicate that although
the expression of drug transporters by the cancer stem
cell might provide some level of drug resistance, an
acquired mutation in ABL could confer higher levels of
drug resistance.Although these mutations might have
arisen during therapy,their existence before the admin-
istration ofimatinib has not been excluded.Indeed,pre-
existing mutations that confer resistance to imatinib
have also been described in a subset of patients62,63.
These findings are reminiscent of the Goldie–Coldman
hypothesis, proposed more than 20 years ago, that a
small percentage of cells in a population harbouring
intrinsic mutations confer drug resistance64. The
Goldie–Coldman hypothesis would theorize that the
cell acquiring the mutation is the stem cell.
Although the expression of ABC transporters
could render stem cells resistant to drugs,it is not the
sole determinant of resistance, as the DNA-repair
capacity of the cell and the reluctance to enter apopto-
sis could be equally or more important. Generally
regarded as quiescent and non-dividing, stem cells
would be expected to be inherently refractory to drugs
that target either the cell cycle or rapidly dividing cells.
To the extent that quiescence is an important mecha-
nism of drug resistance in stem cells,agents will have
to be developed that are effective in non-dividing
cells.For example,studies with imatinib have shown
that blocking BCR–ABL-positive cells at the G1/S
genetic alterations such as those that upregulate ABCB1
expression in human leukaemia and lymphoma cells
could have originated in the stem cell55,56.In a final ‘intrin-
sic resistance’model,both the stem cells and the variably
differentiated cells are inherently drug resistant,so thera-
pies have little or no effect,resulting in tumour growth
(FIG.2d).An example ofthe latter is an intrinsically resis-
tant cancer such as renal-cell cancer,in which ABCB1is
expressed in all cells and contributes to chemotherapy
tolerance.In this case,the resistance phenotype of the
Tumour stem cell
Figure 2 | Models of tumour drug resistance. a | In the conventional model of tumour-cell drug
resistance, rare cells with genetic alterations that confer multidrug resistance (MDR) form a drug-
resistant clone (yellow). Following chemotherapy, these cells survive and proliferate, forming a
recurrent tumour that is composed of offspring of the drug-resistant clone. b | In the cancer-stem-
cell model, drug resistance can be mediated by stem cells. In this model, tumours contain a small
population of tumour stem cells (red) and their differentiated offspring, which are committed to a
particular lineage (blue). Following chemotherapy, the committed cells are killed, but the stem cells,
which express drug transporters, survive. These cells repopulate the tumour, resulting in a
heterogeneous tumour composed of stem cells and committed but variably differentiated
offspring. c | In the ‘acquired resistance’ stem-cell model, the tumour stem cells (red), which
express drug transporters, survive the therapy, whereas the committed but variably differentiated
cells are killed. Mutation(s) in the surviving tumour stem cells (yellow) and their descendants
(purple) can arise (by mechanisms such as point mutations, gene activation or gene amplification),
conferring a drug-resistant phenotype. As in model a, the stem cell with the aquired mutations
could be present in the population before therapy. d | In the ‘intrinsic resistance’ model, both the
stem cells (yellow) and the variably differentiated cells (purple) are inherently drug resistant, so
therapies have little or no effect, resulting in tumour growth.
© 2005 Nature Publishing Group
280 | APRIL 2005 | VOLUME 5
R E V I E W S
As cancer stem cells express drug transporters that
make them resistant to many chemotherapy agents,
anticancer strategies should include efforts to target
these cells with their special properties.Clinical studies
have attempted to overcome drug resistance through
combination therapies in which a cytotoxic drug was
given along with an ABC-transporter inhibitor.In a
new paradigm,transport inhibitors might be thought
of as ‘tumour stem cell sensitizing agents’that allow
the most crucial and most drug-resistant cells in a
tumour to be destroyed.Skeptics could argue persua-
sively that ABCB1 inhibitors have shown very limited
effectiveness in clinical trials.However,one could reply
that clinical trials with these inhibitors have not
focused on targeting cancer stem cells.Rather,they
have determined response rates by measuring the
reduction in size of tumours that express a particular
drug transporter (usually ABCB1).If the stem cells are
the main mediators of drug resistance,however,ABC
inhibitors would not necessarily reduce tumour bur-
den immediately,but efficacy could be observed using
alternative end points,such as the frequency of relapse
or the time to relapse.A skeptic would counter that
these effects would surely have been reported in the
trials conducted so far,if ABCB1 inhibitors did indeed
destroy cancer stem cells.However,it is possible that
the cytotoxic drugs or ABC inhibitors testedwere inef-
ficient in killing cancer stem cells.An inhibitor of drug
transport might be most beneficial when combined
with an anticancer agent that specifically targets
the stem cells, such as imatinib, which targets the
leukaemia stem cells that carry the BCR–ABL fusion
protein.Another potential reason that clinical trials
involving drug-transport inhibitors have not proven
successful is that the wrong transporter was inhibited.
boundary in vitro had no significant impact on the
ability of imatinib to induce apoptosis,indicating that
imatinib is effective in non-dividing cells65–68.
Overcoming drug resistance
By inhibiting the main transporters of chemotherapy
drugs, it was thought that drug resistance could be
avoided and tumour cells eliminated.Therefore,much
effort has been devoted to the development ofinhibitors
of ABC transporters. First-generation compounds
included drugs identified as ABCB1 inhibitors,such as
verapamil and cyclosporine,that were in clinical use to
treat other diseases.These inhibitors were combined
with a range of chemotherapy regimens for many can-
cers30As the results were not convincing,subsequent
clinical trials were attempted with second-generation
inhibitors such as PSC 833 and VX-710 (TABLE 2).The
results of these trials were largely negative,failing in
some cases because of pharmacokinetic interaction
between the chemotherapeutic agent and the ABCB1
inhibitor.These studies might also have failed because of
the presence of additional transporters,such as ABCC1
and ABCG2,that were not targeted by the inhibitor.Yet,
although the results of these trials were negative,correl-
ative studies did show that transport by ABCB1 could
be inhibited.Efflux activity was assessed with a radionu-
clide-imaging agent (99mTc-Sestamibi),confirming that
some human tumours have ABCB1 activity that can be
suppressed with the ABCB1 inhibitors VX 710,PSC 833
and tariquidar (XR9576)69–72. The increased 99mTc-
Sestamibi retention in the entire tumour following
treatment with tariquidar indicates that the transporter-
expressing phenotype of the cancer stem cell persists in
the committed,abnormally developing progenitors that
comprise the proliferative pool ofcancer cells.
Table 2 | ABCB1 inhibitors
InhibitorLimitationsToxicityCancer testedClinical benefit
in clinical trials?
VerapamilLow potencyHypotensionMultiple myeloma
92QuinidineLow potency Gastrointestinal
Valspodar (PSC833) Pharmacokinetic
Biricodar (VX710)Not known NoneNot known97–100
Zosuquidar (LY-335979) Not known
ONT-093 (OC-144-093) Not known
*Two randomized trials in acute mylogenous leukaemia (AML) demonstrated no overall benefit95,96; however, one trial indicated a benefit
in the subset of patients with functional drug-transport activity96. NSCLC, non-small-cell lung cancer; SCLC, small-cell lung cancer.
105 Not known
© 2005 Nature Publishing Group
NATURE REVIEWS | CANCER
VOLUME 5 | APRIL 2005 | 281
R E V I E W S
The identification of potent,specific and non-toxic
inhibitors of ABCB1,ABCG2 and ABCC1 is required
before the full effects of blocking these transporters can
be determined. However, this might be difficult to
accomplish in vivowithout the destruction of normal
stem cells — especially of haematopoietic stem cells —
that depend on the expression of drug transporters to
survive drug therapy.Stem-cell-driven tissue repopula-
tion not only mediates regrowth of tumours,but also
mediates regrowth of normal tissues in the adult,
including the bone marrow,gastrointestinal tract and
hair follicles.Whether a ‘therapeutic window’exists that
would allow the destruction of cancer stem cells but not
normal stem cells remains to be determined.
One cannot deny the appeal of explaining the pool of
drug-resistant cells and the problem of chemotherapy
resistance in terms of the existance of a relatively quies-
cent stem-cell population armed with multiple drug
transporters.But,how does this modelfit in the context
ofthe clinical problem ofdrug resistance? Unfortunately,
for most drug-resistant cancers,including kidney,pan-
creatic and colon cancers,the problem is not that a few
cells survive but, rather, that only a few cells die in
response to chemotherapy.For this vexing problem,the
stem-cell model of drug resistance at present has little
applicability.But for cancers that respond to chemother-
apy with an apparent clinical complete response,only to
relapse months or years later,this stem-cell model of
drug resistance has more appeal.Admittedly,students of
this problem will quickly point out that this hypothesis is
not new;that the only thing new is that the term stem
cell is used to describe those cells that were previously
refered to as resting in G0 (REFS 76,77).
To get the best results,future stem-cell models will
need to have various qualities.First,they will need to
define stem cells by tumorigenicity or clonogenicity
(that is,their capacity for long-term self-renewal),and
Most of the studies evaluating cells with the SP phe-
notype have shown that stem cells overexpress
ABCG2, rather than ABCB1, which has been the
transporter targeted in most clinical studies49. To
properly evaluate the latter possibility it will be
important to develop an inhibitor for ABCG2. The
compound fumitremorgin C (FTC) is a natural prod-
uct that specifically inhibits ABCG2 (REF.73).However
this compound is toxic to cells,as well as to mice,and
is not thought to be suitable for clinical studies.
Chemically synthesized derivatives of FTC such as
Ko143 have been developed,and several of these show
high specificity and low toxicity74.In mice,these com-
pounds have been shown to sensitize mouse tumour
cells to drugs. Studies with Ko143 have also shown
that inhibition of ABCG2 allows for a greater absorp-
tion of certain drugs across the intestine74.In addi-
tion,the compound GF120918 is an ABCB1 inhibitor
that has been shown to inhibit ABCG2 in vitro and
apparently also in vivo75.
Box 1 | New therapeutic opportunities
Administration ofABCG2 inhibitors either before or during chemotherapy might help eliminate tumour stem cells.
Two compounds (GF120918 and tariquidar) that inhibit both ABCG2 and ABCB1 are already approved for clinical
studies.Additional ABCG2 inhibitors are in development.
Antibodies against ABCG2 or other stem-cell markers might be useful in killing tumour stem cells.These antibodies
could be used to deliver toxins or radioisotopes.The antibodies might also be used in diagnostics to detect tumours,
visualize metastasis,or monitor therapy response or relapse.
Stem-cell self-renewal and survival requires signalling from a range ofmolecules through specific cell-surface receptors.
A potential stem-cell inhibitor is cyclopamine,a compound that inhibits the Hedgehog–Patched receptor signalling
protein Smoothened.Inhibiting such receptors and signalling molecules might preferentially inhibit tumour stem cells.
Several clinical protocols involve the activation ofa patient’s immune cells against his/her cancer cells,or the transplant
ofbone-marrow stem cells from a donor to kill the tumour cells in the recipient.Purified tumour stem cells from a
patient could be lethally irradiated and used to ‘immunize’the patient or to activate the donor’s immune cells against the
tumour stem cells.
Cancer prone Abcg2–/–
Figure 3 | Mouse models for testing tumour-stem-cell therapies. Numerous cancer-prone
mouse strains exist and they can be crossed with mouse strains with disruptions in the Abcg2
gene. Tumours and tumour stem cells (red) that arise in these mice would not express ABCG2,
so therapies could be tested either in the Abcg2–/–(blue) or a wild-type strain (pink). This would
allow for the true impact of complete ABCG2 suppression to be assessed, as the tumour cells
would lack the gene encoding this protein. If the absence of ABCG2 improves the
chemotherapy, this would give justification for further experiments with ABCG2 inhibitors.
Experiments with Abcg2+/+tumours are more complicated, as the inhibition of ABCG2 could lead
to severe toxicity of the normal stem cells of the mouse.
© 2005 Nature Publishing Group
282 | APRIL 2005 | VOLUME 5
R E V I E W S
be answered (BOX 1).Excellent approaches have been
developed to isolate and grow stem cells.Initial analysis
of gene expression in these cells reveals many genes that
are either over- or underexpressed in tumour stem
cells. Isolation of stem cells from different types of
tumour will allow determination of the similarity or
differences of these molecular profiles among different
stem cells. This could lead to improved diagnostic
tools to detect pre-malignant lesions and tumours,as
well as targeted therapies,such as antibodies,directed
against tumour stem cells.Agents that suppress stem-
cell growth might be useful chemopreventive agents in
individuals with a high cancer risk,much as we inhibit
oestrogen production or action in those at risk of
Abcb1-,Abcg2-and Abcc1-null mice have been gener-
ated,and all are viable.If the transporters encoded by
these genes are required for the protection ofstem cells,
then mice that lack these genes might have a higher sus-
ceptibility to tumorigenesis from certain mutagenic
chemicals.Such studies could lead to the development of
interesting new cancer models.To better assess the role of
the ABCG2 and ABCB1 transporters in chemotherapy,it
is possible to study tumours that develop in mice that lack
these genes.This will allow the isolation of cancer stem
cells that lack these transporters and the direct testing of
the ability ofcurrent and future drugs to kill cancer stem
cells (FIG.3).Mice with conditional knockouts of these
genes can be developed to allow the tumour stem cells to
form in the presence of the transporters,and the genes
can then be deleted before the start ofthe chemotherapy.
One additional pathway important for the growth
and differentiation of stem cells is the Hedgehog–
Patched (HH–PTCH) pathway. Studies of the
HH–PTCH pathway in tumours provide support for
the importance of tumour stem cells in cancer78(BOX 2),
indicating that proliferation of normal stem cells is reg-
ulated by signals from surrounding normal cells.
Transformation of these stem cells can lead to a pre-
malignant stem cell with abnormal HH expression or
deficient PTCH activity.Such cells can grow in an unre-
strained manner, leading to local proliferation.
Additional genetic events give rise to a tumour stem cell
that can generate more tumour stem cells as well as
mature tumour cells. This model leads to specific
hypotheses that can be tested as well as new avenues for
therapeutics.For example,is the HH overexpression
that is seen in many tumour cell lines a property of the
stem cells or of all the cells in the population? Do drugs
like cyclopamine,which targets the HH–PTCH path-
way,slow tumour growth by inhibiting stem cells? Can
ABCG2 inhibitors be combined with these drugs to
provide higher levels of chemotherapy drugs to tumour
stem cells,without toxicity to normal stem cells?
The existence of tumour stem cells might underlie
the intractable nature of many human cancers,explain-
ing why conventional cancer therapy fails in many
patients.Normal stem cells evolved in the presence of
radiation and multiple environmental toxins,and their
ability to expel these toxins is essential for survival in
adverse conditions.Although the existence of tumour
not merely by presence of the SP phenotype.Second,
they will need to reconcile the observation that many
normal tissues and well-differentiated tumours have
high levels of the same ABC transporters found in stem
cells.Third,they will have to explain how over time,a
repopulated tumour acquires increasing drug resistance.
Fourth,they will need to delineate the ‘plasticity’of cells
in a tumour (that is,the extent to which cells down-
stream of stem cells can acquire the capacity to self
renew).Last,they will have to address the problem of
intrinsic resistance,where all cells are refractory to drug
therapy, not just the small population of stem cells.
Theoretical considerations aside,an exciting aspect of
these new lines of inquiry into cancer is that there are
many tools at hand to quickly allow crucial questions to
Box 2 | Hedgehog signalling and cancer
The Hedgehog molecules (SHH,IHH and DHH) are important signalling proteins in
the development ofembryonic stem cells and in the differentiation ofmany tissues79.
Hedgehog (HH) binds to the cell-surface receptor Patched (PTCH) and signals through
the Smoothened (SMO) and GLI proteins.This pathway has a clear role in tumour
formation in patients with nevoid basal-cell carcinoma syndrome,in which PTCH
mutations have been described80–82.Additional members ofthe HH pathway have also
been found to be tumour suppressors or oncogenes78.Recently,components ofthe
HH–PTCH pathway have been shown to be disrupted or overexpressed in a large
number oftumours,including sporadic medulloblastomas,breast,prostate,stomach,
colon and pancreatic cancers83–87.
Most sporadic medulloblastomas have either germline PTCHmutations or PTCH
silencing through methylation.Treatment ofmedulloblastomas with the SMO-
inhibitory compound cyclopamine resulted in reduced proliferation and changes in gene
expression consistent with differentiation77.Small-cell lung tumour cell lines show high
expression ofSHH,and their growth can be inhibited by SHH antibodies or
cyclopamine85.Similarly high levels ofHH expression and HH–PTCH pathway
activation have been found in oesophageal,stomach,pancreatic,prostate and biliary
tumours and in cell lines.Treatment with cyclopamine led to regression ofpancreatic and
prostatic tumours in mice,providing a model system for therapeutic development84,87.
HH overexpression could lead to the unregulated growth oftissue stem cells (see
figure).This would result in a pre-malignant lesion in which abnormal stem-cell growth
drives hyperproliferation.These unregulated stem cells would be the target for genetic
events that drive the stem cells into the formation oftumour stem cells.Continued
evolution ofthe tumour stem cells could occur to give rise to metastatic cells or further
In the above figure,normal stem cells (blue) undergo transformation to a stem cell
with abnormal HH–PTCH signalling.This cell (orange) proliferates abnormally,but is
pre-malignant.Subsequent genetic events give rise to a tumour stem cell (red) that can
generate additional stem cells with abnormal signalling and tumour cells that are
committed but incompletely differentiated (purple)68.
Normal tissue HH–PTCH activationTumour
Normal stem cell
Normal tissue cell
Pre-malignant stem cell
Tumour stem cell
Committed tumour cell
© 2005 Nature Publishing Group
NATURE REVIEWS | CANCER
VOLUME 5 | APRIL 2005 | 283
R E V I E W S
1. Reya, T., Morrison, S. J., Clarke, M. F. & Weissman, I. L.
Stem cells, cancer, and cancer stem cells. Nature 414,
Al-Hajj, M., Becker, M. W., Wicha, M., Weissman, I. &
Clarke, M. F. Therapeutic implications of cancer stem cells.
Curr. Opin. Genet. Dev. 14, 43–47 (2004).
Morrison, S. J., Wandycz, A. M., Hemmati, H. D., Wright, D. E.
& Weissman, I. L. Identification of a lineage of multipotent
hematopoietic progenitors. Development 124, 1929–1939
Dontu, G., Al-Hajj, M., Abdallah, W. M., Clarke, M. F. &
Wicha, M. S. Stem cells in normal breast development
and breast cancer. Cell Prolif. 36 (Suppl. 1), 59–72
Kordon, E. C. & Smith, G. H. An entire functional mammary
gland may comprise the progeny from a single cell.
Development 125, 1921–1930 (1998).
Price, J. E. & Tarin, D. Low incidence of tumourigenicity in
agarose colonies from spontaneous murine mammary
tumours. Differentiation 41, 202–207 (1989).
Gioanni, J. et al. In vitro clonogenicity in relation to kinetic
and clinicopathological features of breast cancer. Bull.
Cancer 75, 285–290 (1988).
Southam, C. M. & Brunschwig, A. Quantitative studies of
autotransplantation of human cancer. Cancer 14, 971–978
Lapidot, T. et al. A cell initiating human acute myeloid
leukaemia after transplantation into SCID mice. Nature 367,
Original report demonstrating the existence of stem
cells in leukaemia.
10. Bonnet, D. & Dick, J. E. Human acute myeloid leukemia is
organized as a hierarchy that originates from a primitive
hematopoietic cell. Nature Med. 3, 730–737 (1997).
11. Hope, K. J., Jin, L. & Dick, J. E. Acute myeloid leukemia
originates from a hierarchy of leukemic stem cell classes that
differ in self-renewal capacity. Nature Immunol. 5, 738–43
12. Dick, J. E. Breast cancer stem cells revealed. Proc. Natl
Acad. Sci. USA 100, 3547–3549 (2003).
13. Jamieson, C. H. et al. Granulocyte–macrophage progenitors
as candidate leukemic stem cells in blast-crisis CML.
N. Engl. J. Med. 351, 657–667 (2004).
14. Andrews, P. W. From teratocarcinomas to embryonic stem
cells. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 405–417
15. Damjanov, I. Teratocarcinoma stem cells. Cancer Surv. 9,
16. Sell, S. Stem cell origin of cancer and differentiation therapy.
Crit. Rev. Oncol. Hematol. 51, 1–28 (2004).
17. Mintz, B. & Illmensee, K. Normal genetically mosaic mice
produced from malignant teratocarcinoma cells. Proc. Natl
Acad. Sci. USA 72, 3585–3589 (1975).
18. Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J.
& Clarke, M. F. Prospective identification of tumorigenic
breast cancer cells. Proc. Natl Acad. Sci. USA 100,
Demonstrates a population of cells in breast cancers
possessing markers in common with normal breast
stem cells. Small numbers of these cancer stem cells
could generate tumours in mice, whereas the non-
cancer stem cells could not.
19. Liu, B. Y., McDermott, S. P., Khwaja, S. S. & Alexander, C. M.
The transforming activity of Wnt effectors correlates with
their ability to induce the accumulation of mammary
progenitor cells. Proc. Natl Acad. Sci. USA 101, 4158–4163
20. Singh, S. K. et al. Identification of human brain tumour
initiating cells. Nature 432, 396–401 (2004).
21. Hemmati, H. D. et al. Cancerous stem cells can arise from
pediatric brain tumors. Proc. Natl Acad. Sci. USA 100,
Describes the isolation of tumour stem cells from
paediatric brain cancers.
22. Singh, S. K., Clarke, I. D., Hide, T. & Dirks, P. B. Cancer stem
cells in nervous system tumors. Oncogene 23, 7267–7273
23. Singh, S. K. et al. Identification of a cancer stem cell in
human brain tumors. Cancer Res. 63, 5821–5828
24. Richardson, G. D. et al. CD133, a novel marker for human
prostatic epithelial stem cells. J. Cell Sci. 117, 3539–3545
25. Yu, S., Zhang, J. Z., Zhao, C. L., Zhang, H. Y. & Xu, Q.
Isolation and characterization of the CD133+precursors
from the ventricular zone of human fetal brain by magnetic
affinity cell sorting. Biotechnol. Lett. 26, 1131–1136 (2004).
26. Wu, X. et al. Tissue-engineered microvessels on three-
dimensional biodegradable scaffolds using human
endothelial progenitor cells. Am. J. Physiol. Heart Circ.
Physiol. 287, H480– H487 (2004).
27. Cozzio, A. et al. Similar MLL-associated leukemias arising
from self-renewing stem cells and short-lived myeloid
progenitors. Genes Dev. 17, 3029–3035 (2003).
28. Blanpain, C., Lowry, W. E., Geoghegan, A., Polak, L. &
Fuchs, E. Self-Renewal, multipotency, and the existence of
two cell populations within an epithelial stem cell niche. Cell
118, 635–648 (2004).
29. Scharenberg, C. W., Harkey, M. A. & Torok-Storb, B. The
ABCG2 transporter is an efficient Hoechst 33342 efflux
pump and is preferentially expressed by immature human
hematopoietic progenitors. Blood 99, 507–512 (2002).
30. Gottesman, M. M., Fojo, T. & Bates, S. E. Multidrug
resistance in cancer: role of ATP-dependent transporters.
Nature Rev. Cancer 2, 48–58 (2002).
31. Kim, M. et al. The multidrug resistance transporter ABCG2
(breast cancer resistance protein 1) effluxes Hoechst 33342
and is overexpressed in hematopoietic stem cells. Clin.
Cancer Res. 8, 22–28 (2002).
32. Allikmets, R., Schriml, L. M., Hutchinson, A., Romano-
Spica, V. & Dean, M. A human placenta-specific ATP-
binding cassette gene (ABCP) on chromosome 4q22 that is
involved in multidrug resistance. Cancer Res. 58,
33. Miyake, K. et al. Molecular cloning of cDNAs which are
highly overexpressed in mitoxantrone-resistant cells:
demonstration of homology to ABC transport genes.
Cancer Res. 59, 8–13 (1999).
34. Doyle, L. A. et al. A multidrug resistance transporter from
human MCF-7 breast cancer cells. Proc. Natl Acad. Sci.
USA 95, 15665–15670 (1998).
35. Dean, M., Rzhetsky, A. & Allikmets, R. The human ATP-
binding cassette (ABC) transporter superfamily. Genome
Res. 11, 1156–1166 (2001).
36. Schinkel, A. H. et al. Disruption of the mouse Mdr1a P-
glycoprotein gene leads to a deficiency in the blood-brain
barrier and to increased sensitivity to drugs. Cell 77,
37. Zhou, S., Zong, Y., Lu, T. & Sorrentino, B. P. Hematopoietic
cells from mice that are deficient in both Bcrp1/Abcg2 and
Mdr1a/1b develop normally but are sensitized to
mitoxantrone. Biotechniques 35, 1248–1252 (2003).
38. Zhou, S. et al. Bcrp1 gene expression is required for
normal numbers of side population stem cells in mice, and
confers relative protection to mitoxantrone in
hematopoietic cells in vivo. Proc. Natl Acad. Sci. USA 99,
39. Jonker, J. W. et al. The breast cancer resistance protein
protects against a major chlorophyll-derived dietary
phototoxin and protoporphyria. Proc. Natl Acad. Sci. USA
99, 15649–15654 (2002).
40. Goodell, M. A., Brose, K., Paradis, G., Conner, A. S. &
Mulligan, R. C. Isolation and functional properties of murine
hematopoietic stem cells that are replicating in vivo. J. Exp.
Med. 183, 1797–1806 (1996).
41. Summer, R. et al. Side population cells and Bcrp1
expression in lung. Am. J. Physiol. Lung Cell Mol. Physiol.
285, L97–L104 (2003).
42. Alvi, A. J. et al. Functional and molecular characterisation of
mammary side population cells. Breast Cancer Res. 5,
43. Zhou, S. et al. The ABC transporter Bcrp1/ABCG2 is
expressed in a wide variety of stem cells and is a molecular
determinant of the side-population phenotype. Nature Med.
7, 1028–1034 (2001).
Shows that ABCG2 is highly expressed in normal
stem cells and is responsible for the low retention of
fluorescent dyes used to identify the SP population.
44. Lassalle, B. et al. ’Side Population’ cells in adult mouse
testis express Bcrp1 gene and are enriched in
spermatogonia and germinal stem cells. Development 131,
45. Asakura, A. & Rudnicki, M. A. Side population cells from
diverse adult tissues are capable of in vitro hematopoietic
differentiation. Exp. Hematol. 30, 1339–1345 (2002).
46. Martin, C. M. et al. Persistent expression of the ATP-binding
cassette transporter, Abcg2, identifies cardiac SP cells in the
developing and adult heart. Dev. Biol. 265, 262–275 (2004).
47. Lechner, A., Leech, C. A., Abraham, E. J., Nolan, A. L. &
Habener, J. F. Nestin-positive progenitor cells derived from
adult human pancreatic islets of Langerhans contain side
population (SP) cells defined by expression of the ABCG2
(BCRP1) ATP-binding cassette transporter. Biochem.
Biophys. Res. Commun. 293, 670–674 (2002).
48. Terunuma, A., Jackson, K. L., Kapoor, V., Telford, W. G. &
Vogel, J. C. Side population keratinocytes resembling bone
marrow side population stem cells are distinct from label-
retaining keratinocyte stem cells. J. Invest. Dermatol. 121,
49. Hirschmann-Jax, C. et al. A distinct ‘side population’ of cells
with high drug efflux capacity in human tumor cells. Proc.
Natl Acad. Sci. USA 101, 14228–14233 (2004).
50. Kondo, T., Setoguchi, T. & Taga, T. Persistence of a small
subpopulation of cancer stem-like cells in the C6 glioma cell
line. Proc. Natl Acad. Sci. USA 101, 781–786 (2004).
Shows that long-established cancer cell lines contain
a small population of SP cells that behave as cancer
stem cells. Like cancer stem cells isolated from brain
tumours, these SP cells can give rise to the multiple
differentiated cell types found in the cell line.
51. Mizoguchi, T. et al. Expression of the MDR1 gene in human
gastric and colorectal carcinomas. J. Natl Cancer Inst. 82,
52. Nishiyama, K. et al. Expression of the multidrug transporter,
P-glycoprotein, in renal and transitional cell carcinomas.
Cancer 71, 3611–3619 (1993).
53. Bates, S. E. et al. Expression of a drug resistance gene in
human neuroblastoma cell lines: modulation by retinoic
acid-induced differentiation. Mol. Cell. Biol. 9, 4337–4344
54. Mickley, L. A. et al. Modulation of the expression of a multidrug
resistance gene (mdr-1/P-glycoprotein) by differentiating
agents. J. Biol. Chem.264, 18031–18040 (1989).
55. Knutsen, T. et al. Cytogenetic and molecular
characterization of random chromosomal rearrangements
activating the drug resistance gene, MDR1/
P-glycoprotein, in drug-selected cell lines and patients
with drug refractory ALL. Genes Chromosom. Cancer 23,
56. Mickley, L. A., Spengler, B. A., Knutsen, T. A., Biedler, J. L. &
Fojo, T. Gene rearrangement: a novel mechanism for MDR-1
gene activation. J. Clin. Invest. 99, 1947–1957 (1997).
57. Sato, N., Leopold, P. L. & Crystal, R. G. Effect of adenovirus-
mediated expression of Sonic hedgehog gene on hair
regrowth in mice with chemotherapy-induced alopecia.
J. Natl Cancer Inst. 93, 1858–1864 (2001).
58. Cotsarelis, G. & Millar, S. E. Towards a molecular
understanding of hair loss and its treatment. Trends Mol.
Med. 7, 293–301 (2001).
59. Houghton, P. J. et al. Imatinib mesylate is a potent inhibitor
of the ABCG2 (BCRP) transporter and reverses resistance
to topotecan and SN-38 in vitro. Cancer Res. 64,
60. Ozvegy-Laczka, C. et al. High-affinity interaction of tyrosine
kinase inhibitors with the ABCG2 multidrug transporter. Mol.
Pharmacol. 65, 1485–1495 (2004).
61. Burger, H. et al. Imatinib mesylate (STI571) is a substrate for
the breast cancer resistance protein (BCRP)/ABCG2 drug
pump. Blood 104, 2940–2942 (2004).
62. Roche-Lestienne, C. et al. Several types of mutations of the
Abl gene can be found in chronic myeloid leukemia patients
resistant to STI571, and they can pre-exist to the onset of
treatment. Blood 100, 1014–1018 (2002).
63. Hofmann, W. K. et al. Presence of the BCR–ABL mutation
Glu255Lys prior to STI571 (imatinib) treatment in patients
with Ph+acute lymphoblastic leukemia. Blood 102,
64. Goldie, J. H. & Coldman, A. J. A mathematic model for
relating the drug sensitivity of tumors to their spontaneous
mutation rate. Cancer Treat. Rep. 63, 1727–1733 (1979).
stem cells helps explain some previously recognized
phenomena,they are far from being fully understood.
The models advanced must be considered as ‘works in
progress’that will ultimately require adjustments and
refinements.However,the fact that we can now identify,
purify and propagate cancer stem cells allows the devel-
opment of new strategies to improve targeted therapies
© 2005 Nature Publishing Group
284 | APRIL 2005 | VOLUME 5
R E V I E W S
65. Graham, S. M. et al. Primitive, quiescent, Philadelphia-
positive stem cells from patients with chronic myeloid
leukemia are insensitive to STI571 in vitro. Blood 99,
66. Holtz, M. S. et al. Imatinib mesylate (STI571) inhibits growth
of primitive malignant progenitors in chronic myelogenous
leukemia through reversal of abnormally increased
proliferation. Blood 99, 3792–3800 (2002).
67. Paterson, S. C., Smith, K. D., Holyoake, T. L. & Jorgensen,
H. G. Is there a cloud in the silver lining for imatinib? Br. J.
Cancer 88, 983–987 (2003).
68. La Rosee, P., Shen, L., Stoffregen, E. P., Deininger, M. &
Druker, B. J. No correlation between the proliferative status
of Bcr–Abl positive cell lines and the proapoptotic activity of
imatinib mesylate (Gleevec/Glivec). Hematol. J. 4, 413–419
69. Agrawal, M. et al. Increased 99mTc-sestamibi accumulation
in normal liver and drug-resistant tumors after the
administration of the glycoprotein inhibitor, XR9576. Clin.
Cancer Res. 9, 650–656 (2003).
70. Bakker, M. et al. 99mTc-Sestamibi scanning with SDZ PSC
833 as a functional detection method for resistance
modulation in patients with solid tumours. AntiCancer Res.
19, 2349–2353 (1999).
71. Bates, S. E. et al. A phase I/II study of infusional vinblastine
with the P-glycoprotein antagonist valspodar (PSC 833) in
renal cell carcinoma. Clin. Cancer Res. 10, 4724–4733
72. Peck, R. A. et al. Phase I and pharmacokinetic study of the
novel MDR1 and MRP1 inhibitor biricodar administered
alone and in combination with doxorubicin. J. Clin. Oncol.
19, 3130–3141 (2001).
73. Rabindran, S. K., Ross, D. D., Doyle, L. A., Yang, W. &
Greenberger, L. M. Fumitremorgin C reverses multidrug
resistance in cells transfected with the breast cancer
resistance protein. Cancer Res. 60, 47–50 (2000).
74. Allen, J. D. et al. Potent and specific inhibition of the breast
cancer resistance protein multidrug transporter in vitro and
in mouse intestine by a novel analogue of fumitremorgin C.
Mol. Cancer Ther. 1, 417–425 (2002).
75. Cisternino, S., Mercier, C., Bourasset, F., Roux, F. &
Scherrmann, J. M. Expression, up-regulation, and transport
activity of the multidrug-resistance protein Abcg2 at the mouse
blood–brain barrier. Cancer Res.64, 3296–3301 (2004).
76. Weisenthal, L. M. & Lippman, M. E. Clonogenic and
nonclonogenic in vitro chemosensitivity assays. Cancer
Treat. Rep. 69, 615–632 (1985).
77. Secchi, G. C. Cancer chemotherapy. Ann. Ital. Med. Int. 5,
78. Dean, M. Towards a unified model of tumor suppression:
lessons learned from the human patched gene. Biochimica
et Biophysica Acta 1332, M43–M52 (1997).
79. Nusslein-Volhard, C. & Wieschaus, E. Mutations affecting
segment number and polarity in Drosophila. Nature 287,
80. Hahn, H. et al. A mammalian patched homolog is expressed
in target tissues of sonic hedgehog and maps to a region
associated with developmental anomalies. J. Biol. Chem.
271, 12125–12128 (1996).
81. Hahn, H. et al. Mutations of the human homolog of
Drosophila patched in the nevoid basal cell carcinoma
syndrome. Cell 85, 841–851 (1996).
Positional cloning of PTCH as the tumour-suppressor
gene responsible for the nevoid basal-cell carcinoma
syndrome. Established a role for the HH–PTCH
pathway in cancer.
82. Johnson, R. L. et al. Human homolog of patched, a
candidate gene for the Basal Cell Nevus Syndrome. Science
272, 1668–1671 (1996).
83. Berman, D. M. et al. Medulloblastoma growth inhibition by
hedgehog pathway blockade. Science297, 1559–1561
84. Berman, D. M. et al. Widespread requirement for Hedgehog
ligand stimulation in growth of digestive tract tumours.
Nature 425, 846–851 (2003).
85. Watkins, D. N. et al. Hedgehog signalling within airway
epithelial progenitors and in small-cell lung cancer. Nature
422, 313–317 (2003).
86. Thayer, S. P. et al. Hedgehog is an early and late mediator of
pancreatic cancer tumorigenesis. Nature425, 851–856
87. Karhadkar, S. S. et al. Hedgehog signalling in prostate
regeneration, neoplasia and metastasis. Nature 431,
88. Dalton, W. S. et al. A phase III randomized study of oral
verapamil as a chemosensitizer to reverse drug resistance in
patients with refractory myeloma. A Southwest Oncology
Group study. Cancer 75, 815–820 (1995).
89. Belpomme, D. et al. Verapamil increases the survival of
patients with anthracycline-resistant metastatic breast
carcinoma. Ann. Oncol. 11, 1471–1476 (2000).
90. Millward, M. J. et al. Oral verapamil with chemotherapy for
advanced non-small cell lung cancer: a randomised study.
Br. J. Cancer 67, 1031–1035 (1993).
91. Milroy, R. A randomised clinical study of verapamil in
addition to combination chemotherapy in small cell lung
cancer. West of Scotland Lung Cancer Research Group,
and the Aberdeen Oncology Group. Br. J. Cancer 68,
92. Wishart, G. C. et al. Quinidine as a resistance modulator of
epirubicin in advanced breast cancer: mature results of a
placebo-controlled randomized trial. J. Clin. Oncol. 12,
93. Sonneveld, P. et al. Cyclosporin A combined with vincristine,
doxorubicin and dexamethasone (VAD) compared with VAD
alone in patients with advanced refractory multiple myeloma:
an EORTC-HOVON randomized phase III study (06914). Br.
J. Haematol. 115, 895–902 (2001).
94. List, A. F. et al. Benefit of cyclosporine modulation of drug
resistance in patients with poor-risk acute myeloid leukemia:
a Southwest Oncology Group study. Blood 98, 3212–3220
95. Greenberg, P. L. et al. Mitoxantrone, etoposide, and
cytarabine with or without valspodar in patients with
relapsed or refractory acute myeloid leukemia and high-risk
myelodysplastic syndrome: a phase III trial (E2995). J. Clin.
Oncol. 22, 1078–1086 (2004).
96. Baer, M. R. et al. Phase 3 study of the multidrug resistance
modulator PSC-833 in previously untreated patients 60 years
of age and older with acute myeloid leukemia: Cancer and
Leukemia Group B Study 9720. Blood100, 1224–1232
97. Rago, R. P. et al. Safety and efficacy of the MDR inhibitor
Incel (biricodar, VX-710) in combination with
mitoxantrone and prednisone in hormone-refractory
prostate cancer. Cancer Chemother. Pharmacol. 51,
98. Seiden, M. V. et al. A phase II study of the MDR inhibitor
biricodar (INCEL, VX-710) and paclitaxel in women with
advanced ovarian cancer refractory to paclitaxel therapy.
Gynecol. Oncol. 86, 302–310
99. Toppmeyer, D. et al. Safety and efficacy of the multidrug
resistance inhibitor Incel (biricodar; VX-710) in
combination with paclitaxel for advanced breast cancer
refractory to paclitaxel. Clin. Cancer Res. 8, 670–678
100. Bramwell, V. H. et al. Safety and efficacy of the multidrug-
resistance inhibitor biricodar (VX-710) with concurrent
doxorubicin in patients with anthracycline-resistant
advanced soft tissue sarcoma. Clin. Cancer Res. 8,
101. Sparreboom, A. et al. Clinical pharmacokinetics of
doxorubicin in combination with GF120918, a potent
inhibitor of MDR1 P-glycoprotein. Anticancer Drugs 10,
102. van Zuylen, L. et al. The orally administered P-glycoprotein
inhibitor R101933 does not alter the plasma
pharmacokinetics of docetaxel. Clin. Cancer Res. 6,
103. Sandler, A. et al. A Phase I trial of a potent P-glycoprotein
inhibitor, zosuquidar trihydrochloride (LY335979),
administered intravenously in combination with doxorubicin
in patients with advanced malignancy. Clin. Cancer Res. 10,
104. Guns, E. S., Denyssevych, T., Dixon, R., Bally, M. B. &
Mayer, L. Drug interaction studies between paclitaxel
(Taxol) and OC144-093 — a new modulator of MDR in
cancer chemotherapy. Eur. J. Drug Metab. Pharmacokinet.
27, 119–126 (2002).
105. Oldham, R. K., Barnett, D. & Ramos, Z. A phase II study of
paclitaxel/CBT–1, an MDR modulator. Proc. Am. Soc. Clin.
Oncol. 22, 148 (2003).
Competing interests statement
The authors declare no competing financial interests.
The following terms in this article are linked online to:
ABCB1 | ABCC1 | ABCG2 | ABCG2 | ABL | BCR | CD133 | HH |
National Cancer Institute: http://cancer.gov/
breast cancer | chronic myelogenous leukaemia | renal-cell cancer
nevoid basal-cell carcinoma syndrome
Access to this interactive links box is free online.