Review
10.1517/13543784.14.6.557 © 2005 Ashley Publications Ltd ISSN 1354-3784 557
Ashley Publications
www.ashley-pub.com
Monthly Focus: Oncologic
Peroxisome proliferator-activated
receptor-γ ligands for the
treatment of breast cancer
Martin H Fenner† & Elena Elstner
†Charité School of Medicine, Department of Oncology and Haematology, Humboldt University,
Schumannstr. 20/21, 10117 Berlin, Germany
Pioglitazone and rosiglitazone are thiazolidinediones used for the treatment
of Type 2 diabetes mellitus. They modulate glucose and fat metabolism,
mainly by binding to the nuclear hormone receptor peroxisome proliferator-
activated receptor (PPAR)-γ. PPAR-γ signalling is involved in a number of
other disease conditions including cancer. In breast cancer cells, PPAR-γ lig-
ands inhibit proliferation and induce apoptosis both in vitro and in vivo.
PPAR-γ ligands also inhibit tumour angiogenesis and invasion. The only pub-
lished clinical trial using a PPAR-γ ligand in patients with metastatic breast
cancer failed to show any clinical benefits. The mechanism of action of the
thiazolidinediones in breast cancer cells is not fully understood but involves
interactions with other nuclear hormone receptors, transcriptional co-activa-
tors and repressors as well as PPAR-γ-independent effects. A better under-
standing of these mechanisms will be needed before PPAR-γ ligands may be
useful in the treatment of breast cancer patients.
Keywords: apoptosis, breast cancer, differentiation, nuclear hormone receptor, proliferation,
PPAR-γ, thiazolidinedione
Expert Opin. Investig. Drugs (2005) 14(6):557-568
1. Introduction
The antidiabetic drugs pioglitazone and rosiglitazone mediate their positive effects
on glucose and lipid metabolism mainly by binding to the transcription factor per-
oxisome proliferator-activated receptor (PPAR)-γ. PPAR-γ signalling is also involved
in a variety of disease conditions including inflammation and cancer [1]. The role of
PPAR-γ in many cancers, including liposarcoma, prostate cancer and colon cancer,
has been reviewed elsewhere [2-7]. In this review, the focus will be on the preclinical
and clinical data of PPAR-γ ligand treatment in breast cancer.
2. Peroxisome proliferator-activated receptors
The peroxisome is an organelle present in most eukaryotic cells and is involved in
many important metabolic processes including the β-oxidation of long- and very
long-chain fatty acids. Peroxisome proliferators are chemicals that increase the vol-
ume density of peroxisomes and of peroxisomal fatty-acid β-oxidation activity.
The peroxisome proliferators are a heterogeneous group of substances that also
include the fibrate class of hypolipidaemic drugs. Peroxisome proliferation
induces hypolipidaemia, but prolonged administration in rodents can also lead to
hepatocyte hyperplasia and ultimately hepatocellular carcinoma [8]. PPAR-α was
cloned as the main molecular target of peroxisome proliferators in 1990 [9].
PPAR-α is a ligand-activated transcription factor of the nuclear hormone receptor
superfamily. The gene for acyl coenzyme A oxidase, the most widely used marker
1. Introduction
2. PPARs
3. PPAR-γ ligands
4. PPAR-γ receptor/ligand
interactions
5. Other effects mediated by
PPAR-γ ligands
6. PPAR-γ ligands in breast cancer
7. Interactions of PPAR-γ with
other signal transduction
pathways in breast cancer
8. PPAR-γ in animal models of
breast cancer
9. Clinical trials in breast cancer
10. Expert opinion

Peroxisome proliferator-activated receptor-γ ligands for the treatment of breast cancer
558 Expert Opin. Investig. Drugs (2005) 14(6)
of peroxisome proliferation, was the first gene shown to con-
tain a peroxisome proliferating response element (PPRE) in
its promoter [10]. Heterodimers of PPAR-α and the retinoid
X receptor (RXR) bind to this PPRE, which consists of two
repeats of the consensus sequence AGGTCA spaced by one
nucleotide [11]. Bexarotene and other specific ligands for the
RXR are called rexinoids; some retinoids such as isotretinoin
are ligands for both the retinoic acid receptor (RAR) and
RXR. Treatment with fibrates or isotretinoin enhances the
transcriptional activity of the acyl coenzyme A oxidase pro-
moter. It is now clear that the lipid-lowering effects of
fibrates such as bezafibrate are mainly mediated via
activation of PPAR-α [12].
A second member of the PPAR family, now called
PPAR-δ, was cloned in 1992 [13]. PPAR-δ is the least studied
PPAR but probably plays an important role in skin wound
healing and colon cancer. PPAR-δ heterozygous mutant
mice show an increased keratinocyte proliferative response
and a delay during the wound-healing process [14]. PPAR-δ
expression is elevated in colorectal cancer and repressed by
adenomatous polyposis in colon (APC) in colorectal cancer
cells. This repression is mediated by β-catenin/transcription
factor (Tcf )-4-responsive elements in the PPAR-δ promoter
[15]. The genetic disruption of PPAR-δ decreases the tumori-
genicity of human colon cancer cells [16]. In contrast, colon
polyp formation is significantly enhanced in mice nul-
lizygous for PPAR-δ in both the multiple intestinal neoplasia
(Min) mutant and chemically induced mouse models of
colon cancer [17].
PPAR-γ, the third member of the PPAR family, was identi-
fied in 1994 as an enhancer binding to the 5′-flanking region
of the adipocyte P2 (aP2) gene [18]. A year later, rosiglitazone,
a member of the thiazolidinedione (TZD) class of anti-
diabetic drugs, was identified as a high-affinity ligand for
PPAR-γ [19]. In 1982, Takeda Pharmaceutical had synthesised
ciglitazone, the prototype TZD, in the search for active com-
pounds in the KKAy mouse model of diabetes [20]. The
related compounds pioglitazone, troglitazone and rosiglita-
zone were synthesised in the 1980s. It soon became apparent
that the main mechanism of action of the TZDs in diabetes is
via PPAR-γ transactivation. PPAR-γ not only has beneficial
effects on glucose metabolism but is also important for adi-
pocyte differentiation. The promoters of aP2 and a number of
other genes involved in adipocyte differentiation, including
acyl coenzyme A synthetase and lipoprotein lipase, contain a
PPRE [21]. Forced overexpression of PPAR-γ and another adi-
pogenic transcription factor, CCAAT/enhancer-binding pro-
tein (C/EBP), leads to the transdifferentiation of myocytes
into mature adipocytes [22].
3. Peroxisome proliferator-activated receptor-γ
ligands
PPAR-γ is activated by several lipophilic ligands, including
polyunsaturated fatty acids and arachidonic acid derivatives.
The cyclopentone 15-deoxy-Ε12,14-prostaglandin J2 (15d-PGJ2)
is probably the most potent endogenous PPAR-γ ligand [23,24].
The positive effects of the TZDs on glucose and lipid metabo-
lism led to the approval of pioglitazone (Takeda/Eli Lilly), tro-
glitazone (Sankyo/Parke-Davis) and rosiglitazone
(GlaxoSmithKline) for the treatment of diabetes mellitus in
many countries between 1997 and 2000. The PPAR-γ binding
affinity of the TZDs correlates with the half-maximal effective
dose (ED50) for reduced plasma glucose in db/db mice, and ros-
iglitazone has a higher binding affinity than either troglitazone
or pioglitazone [25].
Although all PPAR-γ ligands currently in clinical use are
TZDs, most of the newer PPAR-γ ligands in development do
not belong to the TZD class of drugs. CDDO and
CDDO-Im (Reata Discovery) are novel synthetic triterpe-
noids that inhibit the proliferation of breast cancer cell lines
[26]. They bind and transactivate PPAR-γ but the anticancer
activity is partly independent of PPAR-γ. C-substituted di-
indolylmethanes inhibit carcinogen-induced rat mammary
tumour growth, inhibit MCF-7 cell proliferation, induce
apoptosis and downregulate cyclin D1 and oestrogen recep-
tor-α (ER-α) expression in breast cancer cells [27]. Ly-293111
(Eli Lilly) is a leukotriene B4 receptor antagonist and PPAR-γ
ligand that is currently in Phase II clinical trials in pancreatic
cancer [28,201]. CS-7017 (Sankyo) inhibits the growth of thy-
roid carcinoma cell lines [29]. Spirolaxine (Sigma-Tau) inhibits
the proliferation of various tumour cell lines [29]. A Phase II
clinical trial in chronic lymphocytic leukaemia has been initi-
ated in 2004 for SDX-101 (Salmedix), the (R)-enantiomer of
etodolac. (R)-etodolac modulates PPAR-γ function, probably
by binding to RXR [30].
As expected from other nuclear hormone receptors,
PPAR-γ ligands with partial agonist activity were discovered
[31]. The maximal transcriptional activation of PPAR-γ target
genes by these partial agonists is much lower than with strong
agonists such as rosiglitazone. Isaglitazone (MCC-555) was
the first PPAR-γ ligand with partial agonist activity [32]. It
only has 10% of the affinity for PPAR-γ compared with ros-
iglitazone and has partial agonist or antagonist activity,
depending on the cell type or DNA binding site. L-746604 is
a partial PPAR-γ agonist with strong binding affinity for a
PPAR-γ reporter construct but with maximal transcriptional
activation of only 25% of the activation achieved by TZDs
such as troglitazone [33]. GW-0072 was identified as a high-
affinity PPAR-γ ligand that was a weak partial agonist of
PPAR-γ transactivation. In cell culture, GW-0072 was a
potent antagonist of adipocyte differentiation [34]. The
cyclooxygenase inhibitor indomethacin also acts as a partial
agonist of PPAR-γ transactivation [31]. The angiotensin type 1
receptor antagonist telmisartan (GlaxoSmithKline) also has
PPAR-γ partial agonist activity [35]. In contrast to irbesartan,
telmisartan does not recruit the co-activator transforming
growth factor-β1-induced anti-apoptotic factor (TIF)-2 [36].
In order to better treat patients with both Type 2 diabetes
mellitus and dyslipidaemia, dual PPAR-α/γ agonists were