PROTEINS: A POTENTIAL ANTICANCER TARGET
PROTEINS AND THEIR DISCOVERY
The molecular chaperone Hsp90 was initially identified
as a highly conserved protein whose expression is induced by cellular stress.
Hsp were initially described in 1962. Although their function was unclear
initially, subsequent research demonstrated a cytoprotective effect consistent
with their function as molecular chaperones for several cellular proteins. Hsp are
highly abundant in normal cells, constituting about 1-2% of total cellular
proteins under non-stress conditions. Their function is essential for normal
cell viability and growth. Hsp90 exerts its chaperone effect by correcting conformation,
activity, intracellular localization, and proteolytic disposal of multiple
proteins involved in cell growth, differentiation and survival. In tumor cells
growing in a hypoxic, low pH, nutrient depleted environment, Hsp consists of
4-6% of total proteins and that prevent destabilization of normal proteins. Out
of hundreds of such client proteins, particularly important are several key
oncogenic proteins involved in signal transduction pathways leading to
proliferation, cell cycle progression, apoptosis, angiogenesis and metastasis.
Few examples of such oncoproteins include ERBB2, BCR-ABL, AKT/PKB, CRAF, CDK4,
PLK-1, MET, mutant p53, HIF-1α, steroid hormone receptors, survivin and
telomerase hTERT. The interaction of Hsp90 with multiple oncogenic pathways
makes it an ideal target for anticancer therapy. Initial studies revealed Hsp90
to be essential for normal cell viability and growth suggesting it was not a
suitable potential therapeutic target. However, studies with the naturally
occurring Hsp90 inhibitor GA demonstrated consistent in vitro antitumor
activity, supporting further research for potential anticancer agent targeting Hsp.
Subsequently, several other Hsp inhibitors emerged as therapeutic agents
candidates and are currently undergoing clinical development.
AND FUNCTIONS OF Hsp
The Hsp90 family is comprised of 5 isoforms that
differ in their cellular localization. Hsp90α and Hsp90β are the two major
cytoplasmic isoforms and they share about 85% homology in their protein sequences.
GRP94 (glucose regulated protein 94) is localized in endoplasmic reticulum,
TRAP1 (tumor necrosis factor receptor associated protein 1) in the mitochondria,
and Hsp90N, the most recently discovered member of this family, is associated with
RAF . All Hsp90 isoforms have similar structures except that the
NH2-terminal ATP/ADP binding domain is absent in Hsp90N. Hsp90α and Hsp90β are the
most studied family members.
Hsp90 is a highly abundant 90 kDa protein that exists
as a homodimer. X-ray crystallography played an essential role in establishing the
details of the molecular structure of the Hsp90. Each Hsp90 monomer contains
three functional domains: an NH2-terminal ATP/ADP binding domain, a middle
domain which is involved in client protein binding, and a COOH terminal
dimerization domain. The chaperon activity of Hsp90 is dependent on its
NH2-terminal dimerization, which stimulates the ATPase activity. This process is
regulated by a number of accessory proteins known as co-chaperones. Initially,
the client protein interacts with an Hsp70/Hsp40/HIP complex. This complex containing
Hsp70 is linked by an adaptor protein HOP/p60, which interacts with the
C-terminals of both Hsp90 and Hsp70 via its TPR (tetrapeptide repeat) domain.
HOP/p60 can only bind to ADP-bound-Hsp90, which has an open conformational
structure and high affinity for hydrophobic substrates. Subsequently Hsp90
undergoes a transient dimerization of the NH2-terminal domain. This conformational
change is initiated by exchange of ADP for ATP by Hsp90 and leads to
dissociation of Hsp90/Hsp40/HIP and HOP and ATP dependent association of other
co-chaperones such as p50 (CDC37), p23 and immunophilins. The protein p50
(CDC37) helps incorporate kinase clients into Hsp90. In contrast, p23 assists Hsp90
to stay in ATP bound form. This final complex enables the client protein to
express its functions.
Inhibition of ATP binding to Hsp90 blocks the
formation of this final complex and leads to degradation of client protein by
the proteosome. This is facilitated by E3-ubiquitin ligase and carboxyl
terminus of Hsc70-interacting protein (CHIP), which is a TPR protein that can
interact with both Hsp70 and Hsp90.
Hsp90 IN CANCER
Despite the in-depth understanding of the role of Hsp
in normal cell growth and proliferation, until recently Hsp90 was not considered
a viable cancer drug target. It has an essential role in promoting normal cell
viability and growth and no mutations or overexpression in tumor had been
reported. Further studies indicated that Hsp90 may still be an anticancer
target because it is involved in maturation and stabilization of a wide range of
client proteins crucial for oncogenesis and malignant transformation. The
unfavorable tumor environment characterized by hypoxia, acidosis and
nutritional deprivation tends to destabilize the cellular proteins thereby
making them more dependent on Hsp90 activity. Furthermore in tumors where Hsp90
is present as multi-chaperon complex with unusually high affinity for ATP, in
contrast to normal cells where Hsp90 is present in uncomplexed latent state.
Similarly, Hsp90 inhibitors selectively kill the tumor cells compared to normal
cells as Hsp90 derived from tumor cells has 100-fold higher binding affinity for
17-allylaminogeldanamycin (17-AAG) (another Hsp90 inhibitor) compared to Hsp90
from normal cells. Tumor specific accumulation of Hsp inhibitors has been
reported for several Hsp90 inhibitors and selectivity, this is not due to
structural or physic-chemical properties of the compound but to the properties
of Hsp90 itself.
Several client proteins of Hsp90 have critical roles
in one or more signaling pathways responsible for malignant transformation and
proliferation and will be discussed below.
Client proteins involved in pro-survival growth signaling
Environmental growth signals that trigger cell
proliferation and growth are relayed via transmembrane receptors and associated
tyrosine kinases. Several of these receptors and tyrosine kinases are client
proteins for Hsp90.
Epidermal growth factor receptor (EGFR) is a
transmembrane tyrosine kinase receptor responsible for normal cell growth and
proliferation. Overexpression of EGFR and mutations in the kinase domain are frequently
found in various epithelial tumors. In lung cancer, EGFR mutations are found in
about 10-25% of patients and these correlate with response to EGFR tyrosine
kinase inhibitors (TKI) such as gefitinib and erlotinib. Resistances to these agents
pose treatment challenges. EGFR is a client protein for Hsp90, and degradation of
EGFR is induced by Hsp90 inhibitors in gefitinib resistant xenografts cells. The
cells with wild type EGFR are less sensitive to Hsp90 inhibition compared to
cells with mutated EGFR. Based on these observations, Hsp90 inhibitors may be
an important therapeutic option in selected lung cancer patients who develop
resistance to EGFR inhibitors.
ErbB2/Her-2 is another transmembrane receptor with
tyrosine kinase activity and it is overexpressed in about 30% of breast cancers
and about 15% of prostate cancers. Its expression is associated with poor
prognosis. Hsp90 is required for normal activity of Her-2 and inhibition of
Hsp90 has been associated with significant decrease in Her-2 levels in cell
culture and in prostate xenograft models. This suggests that Hsp90 inhibitors may
have clinical utility in Her-2 overexpressing tumors, such as breast and
Another client protein of Hsp90 is c-kit, a
transmembrane receptor exhibiting gain-of-function mutations in the tyrosine
kinase domain in about 90% of the patients with GIST. This mutation leads to activation
of multiple downstream signaling pathways and tumorigenesis. Therapeutic
inhibition of c-kit by imatinib mesylate leads to clinical responses in about 80%
of the patients with metastatic disease. Primary resistance to imatinib is
observed in about 10-20% of patients and secondary resistance due to additional
mutations after imatinib therapy also occurs with high frequency. Some of these
patients can still respond to other multi-targeted tyrosine kinases inhibitors such
as sunitinib, but overall the activity of these agents is limited due to the
large variety of c-kit mutations. Therefore, degradation of the resistant c-kit
following Hsp90 inhibition may be a rational strategy to address this issue.
Hsp90 plays a role in protein stabilization and activation of c-kit. Inhibition
of Hsp90 has been shown
to cause degradation of both wild type c-kit and
imatinib-resistant mutant forms of ckit. This was further validated in a study
of Hsp90 inhibition in imatinib-sensitive GIST cell lines (GIST882),
imatinib-resistant GIST cell lines dependent on c-kit (GIST430 and GIST48), and
GIST cell lines independent of c-kit oncoprotein (GIST62). Hsp90 inhibition resulted
in the significant reduction of c-kit expression, a decrease in downstream signaling
proteins, and in the inhibition of cell proliferation and survival. This inhibition
of cell proliferation was observed only in the cells that were imatinib
sensitive or dependent on c-kit, however in cells resistant to imatinib but not
dependent on c-kit, it had no effect. Based on this preclinical evidence, Hsp90
inhibitors are currently in clinical evaluation in GIST.
Insulin like growth factor-I receptor (IGF-IR) is a
transmembrane receptor protein and its activity is controlled by IGF-I and
IGF-II ligands. Ligand binding leads to autophosphorylation of IGF-IR and
activation of downstream prosurvival signaling pathways. Degradation of IGF-IR
occurs following Hsp90 inhibition in some breast cancer cell lines, leading to
blockade of downstream signal transduction through the Akt and
mitogen-activated protein (MAP) kinase pathways . Similar observations have
been made in pancreatic cancer cell lines and xenografts where inhibition of
Hsp90 caused disruption of IGF-I and IL-6-induced proangiogenic signaling, and inhibition
of tumor proliferation. These studies suggest alternating mechanism of IGF-IR
inhibition by Hsp90 inhibitors.
FLT3 is another receptor tyrosine kinase that has been
widely studied in hematological malignancies. It regulates proliferation, survival,
and differentiation of hematopoietic cells, and is expressed in 20% of the patients
with acute myeloid leukemia (AML). In AML, internal tandem duplication in the
FLT3 juxtamembrane domain is common and is usually associated with leucocytosis
and poor prognosis. Evidence suggests that FLT3 is an Hsp90 client as inhibition
Hsp90 causes loss of kinase activity of FLT3 and decreased downstream signaling
through MAP kinase and AKT. This result in apoptosis of leukemic cells
expressing duplicated FLT3. Thus Hsp inhibitors may have potential utility in the
treatment of patients with FLT3 positive leukemias.
The aberrant fusion protein BCR-ABL (p210Bcr-Abl) is
another Hsp90 client found in virtually all chronic myeloid leukemias (CML) and
in some acute leukemias. The tyrosine kinase inhibitor imatinib is extremely effective
in initial treatment of chronic phase CML, but a significant proportion of
these patients will ultimately develop resistance due to changes in the BCR-ABL
tyrosine kinase. Resistance can arise from secondary gene amplification, or
from a variety of specific point mutations in the kinase domain. One study
suggests that imatinib-resistant CML cells containing BCR-ABL point mutations
remained sensitive to Hsp90 inhibitors. Degradation of both wild type and mutant
BCR-ABL was noticed following Hsp90 inhibition, but more potent activity occurred
in CML cells with the resistant mutant form of BCR-ABL. This indicates that Hsp90
inhibitors may have therapeutic potential in the treatment of patients with CML
resistant to imatinib. Besides the above mentioned oncoproteins that are mainly
associated with the cell membrane, elements of the downstream signaling pathways
are also client proteins for Hsp90. The activation of membrane receptors and
tyrosine kinases such as EGFR, IGF-IR initiate a series of signaling events
through two major signaling pathways: the phospatidylinositol 3kinase-AKT (PI3K-AKT)
and the Ras-Raf-extracellular signal related kinase (Ras-Raf-ERK) pathways. These
pathways converge at the level of DNA transcription leading to the expression
of proteins responsible for cell proliferation and growth. The activity of
PI3K-AKT pathway leads to activation of pro-survival transcription nuclear factor-kB
(NF-kB). High activity of AKT has been noticed in breast, prostate, lung, pancreatic,
ovarian, and colorectal cancers and has been associated with resistance to
chemotherapy and other targeted therapies. Recent studies suggest that Hsp90 ensures
the conformation, stability and activity of AKT. Inhibition of Hsp90 has been
shown to decrease the levels and activity of AKT. The RasRaf-ERK signaling pathway
is critical for cell proliferation and survival. Mutations in Ras or Raf are
common in human malignancies, and mutation of the B-Raf isoform is found in 70%
of malignant melanomas and 36% of papillary thyroid cancers. Studies have
suggested that Raf and the downstream mitogen-activated protein (MAP) kinase,
MEK, are both clients for Hsp90, and inhibition of Hsp90 has shown to deplete their
levels in tumor cells. Inhibition of Hsp90 provides a unique strategy to block both
PI3K-AKT and Ras-Raf-ERK pro-survival pathways simultaneously.
Steroid receptors are present in the intracellular
compartment where they bind the steroid ligands, translocate to the nucleus, and
initiate their transcriptional activity ultimately promotes cell growth. The estrogen,
progesterone and androgen steroid receptors are of particular importance for
human malignancies. In the unstimulated state, estrogen and progesterone
receptors exist in the nucleus as multimolecular complexes consisting of Hsp90 and
other chaperones. High expression of Hsp90 has been associated with poor
prognosis in patients with breast cancer. Upon ligand binding, estrogen receptor
protein dissociates from the complex and initiates gene transcription.
Inhibition of Hsp90 has shown to destabilize the multimolecular complex and
depletes the levels of these receptors in breast cancer cell lines. Similar
results were also shown in the hormone receptor-positive breast cancer
xenografts. This suggests that Hsp90 inhibitors may be an effective therapeutic
option for hormone-receptor positive breast cancers, especially for those
patients who have developed resistance to tamoxifen and aromatase inhibitors.
Similarly, androgen deprivation is mainstay of therapy for prostate cancer.
Evidence suggests that androgen receptors are clients for Hsp90. Inhibition of
Hsp90 can cause degradation of androgen receptors in both hormone-sensitive and
hormonerefractory prostate cancer cell lines and xenografts. The simultaneous inhibition
of Her-2/neu and AKT in these cells may provide additional benefit in these
refractory cells. This provides the basis for exploring Hsp90 inhibitors in
patients with prostate cancer, especially those refractory to hormone therapy.
Apoptosis is a physiological, highly regulated process
leading to programmed cell death through the activation of a family of cystein
aspartic acids proteases known as caspases. Dysregulation of apoptosis can lead
to abnormal development and oncogenesis. The cascade of caspases is activated by
diverse apoptotic stimuli from outside and inside of the cell. There are two
main regulatory mechanisms to control apoptosis. The first pathway is initiated
by the activation of specific death receptors expressed at the cell surface including
Fas, tumor necrotic factor receptor 1 (TNFR1) and death receptor 3 (DR3)
leading to activation of caspase -8 and other downstream “effector” caspases.
In the second pathway, a variety of intracellular and extracellular death
stimuli trigger the release of cytochrome-c from mitochondria which in turn binds
Apaf-1, which then promotes the activation of caspase-9.
Hsp90 inhibits cytochrome-c mediated oligomerization
of Apaf-1 and blocks the activation of procaspase-9, thereby preventing apoptosis
in human leukemic cells. The death domain kinase, receptor interacting protein (RIP),
is one of the major components of the tumor necrosis factor receptor 1 (TNFR1)
complex and it plays an essential role in TNF-mediated nuclear factor kappaB (NF-kappaB)
activetion. The activation of NF-kappaB protects cells against TNF-induced apoptosis.
RIP is Hsp90 client protein and Hsp90 inhibition leads to inhibition of RIP and
subsequent apoptosis. Another pathway through which Hsp90 facilitates
activation of NF-kappaB is via the IkappaB kinase (IKK) complex. IKK forms a complex
with Cdc37 and Hsp90 that is required for normal activity. Inhibition of Hsp90
may cause apoptosis by disrupting IKK complex. In addition, Hsp90 may also block
apoptosis through PI3K/AKT pathway. Indeed, phosphorylated AKT can inactivate the
Bcl-2 family protein Bad and caspase-9 leading to the inhibition of apoptosis.
In conclusion, Hsp90 has many molecular targets in the
apoptotic pathway that may be critical in the survival of malignant cells.
Thus, inhibition of Hsp90 may prove to be a useful therapeutic strategy aiming
to restore apoptosis either alone or in combination with other proapoptotic
Tumor angiogenesis is essential for tumor growth and
proliferation. One of the main pro-angiogenic factors is the vascular endothelial
growth factor (VEGF), which is produced by cancer cells and binds to VEGF receptors
on endothelial cells, leading to endothelial cell proliferation and migration.
Hypoxia inducible factor-1α (HIF-1α) is a nuclear transcription factor that
activates numerous target genes involved in promotion of angiogenesis secondary
to hypoxia including the VEGF gene. Hsp90 activity may be necessary for HIF-1α
function since inhibition of Hsp90 decreases HIF-1α transcriptional activity.
Additionally, Hsp90 is required for the stable expression of all three VEGF
receptors, and Hsp90 inhibition can reduce expression of these VEGF receptors,
thereby blocking angiogenesis. Thus, due to its important role in VEGF
signaling, Hsp90 may be an attractive target for inhibiting tumor angiogenesis.
Tissue invasion and metastasis
Malignant cells tend to locally invade the normal
tissues and can metastasize distantly. This process is complex and is
coordinated by the interaction of several key proteins and tyrosine kinases. One
of the best studies of these protein kinases is the proto-oncogene c-Met which
functions as the receptor for hepatocyte growth factor/scatter factor (HGF/SF).
Dysregulation of c-Met plays a key role in tumor invasion and metastasis. C-Met
gene amplification and activating mutations have been identified in a variety of
primary human cancers. C-Met is an Hsp90 client and inhibition of Hsp90 leads
to destabilization of c-Met and inhibition of downstream signaling, thereby
interfering with cell growth in cell lines that overexpress c-Met.
Similarly, focal adhesion kinase (FAK) is a nonreceptor
tyrosine kinase involved in adhesion-mediated signal transduction, and the
level of FAK expression is related to the invasiveness of some malignant tumors.
Inhibition of Hsp90 can cause degradation of FAK and inhibition of FAK
Another protein that plays an important role in tumor
invasion and metastasis is matrix metalloproteinase-2 (MMP2). MMP2 “digests”
the extracellular matrix, thus allowing tumor cells to migrate to blood vessels
and lymphatics leading to distant metastases. It was demonstrated that Hsp90α isoform
is secreted to extracellular surface through unknown mechanism.
Interaction of Hsp90 with cell cycle regulatory proteins
Cyclin-dependent kinases (CDKs) are a group of protein
kinases involved in regulation of the cell cycle. CDK4 is the catalytic subunit
of the protein kinase complex that is critical for progression of cell cycle through
G1 phase. CDK4 inhibits the activity of retinoblastoma (RB) tumor suppressor
protein and it facilitates the transition of cells through the G1 phase of cell
cycle. CDK4 amplification and activating mutations have been found in human
malignancies such as glioblastoma multiforme and melanomas. CDK4 has been
reported to be a client protein for Hsp90, and inhibition of Hsp90 causes degradation
of CDK4 and inhibition of cellular proliferation by inducing G1 arrest in some tumor
cell lines. Hsp90 inhibition also limits CDK4 activity by downregulating cyclin-D,
which is an integral component of CDK4 complex. Similarly, CDK2 which controls
DNA synthesis and replication in S-G2 cell cycle phase and CDK7, which
activates CDK2, are also Hsp90 clients. Thus, the modulation of these CDKs by
Hsp90 inhibitors suggests a potential role in the control of cell cycle
progression in cancer cells.
Telomeres are a specific DNA sequence repeats found in
condensed form at the 3’ end of DNA and are postulated to provide stability to
chromosomes. Telomeres are shortened after each replicative cell cycle a process
that ultimately affects the integrity of chromosomes and leads to cell death.
Telomerase is an enzyme responsible for maintenance of telomeres and it is
upregulated in cancer cells. Telomerase activity promotes cell immortality and allows
cells to multiply continuously. Hsp90 is essential for activity of telomerase. Hsp90
has been shown to interact with the catalytic subunit of telomerase and inhibition
of Hsp90 has been shown to inhibit activity of telomerase in vitro. Hsp90 assists
in loading of telomerase onto the telomere and thereby stabilize the functional
structure of telomerase. Inhibition of Hsp90 may be a potential mechanism for inhibiting
cell cycle progression in cancer cells with excessive telomerase activity.
DEVELOPMENT OF Hsp90 INHIBITORS
Naturally occurring Hsp90 inhibitors such as GA,
radiciol, and novobiocin have been used extensively in preclinical studies to
assess Hsp90 inhibitory activity. Their clinical use is limited because of
their poor pharmacological properties including low aqueous solubility, chemical
instability, and off-target toxicities. Preclinical toxicity assessment of GA
in mice and beagle dogs showed that GA is not suitable for clinical use in
humans due to its significant hepatic toxicity. This led to search for newer
Hsp90 inhibitors with improved toxicity profiles. Various second generation Hsp90
inhibitors that are GA analogs have been developed and are being clinically evaluated.
Additional Hsp90 inhibitors with novel non-GA related scaffolds structures have
also been developed and entered clinical testing.
SECOND GENERATION Hsp90 INHIBITORS
17-allylamino-17-demethoxygeldanamycin (17AAG) is a
carbon-17 substituted derivative of GA with similar activity against Hsp90 but
with a considerably less toxic profile.
17AAG was the first Hsp90 inhibitor to be tested
clinically. In several phase I pharmacological studies utilizing different
schedules the toxicity of 17AAG was dose and schedule dependent. Main side-effects
included nausea, vomiting, fatigue, myalgia, anorexia and transaminitis. Major
dose limiting toxicities were fatigue, pancreatitis and hepatic toxicity.
Hepatic toxicity was more common with a daily administration schedule. Hematological
toxicities were minimal.
Further improvement in the formulation of 17AAG led to
development of a cremophor-containing preparation known as tanespimycin (also known
as KOS-953, (KosanBiosciences Inc, Hayward, California). Phase I and Phase I/II
studies in combination with trastuzumab in advanced solid tumors (primarily breast
cancer), and with botezomib in multiple myeloma, showed favorable safety
profiles and suggested preliminary antitumor activity. A separate Phase II study
of cremophor-free tanespimycin in women with HER2/neu positive advanced breast
cancer showed response rates of 24% and clinical benefit in 57% of patients.
CNF-1010 (Biogen Idec, Inc, Cambridge, MA) is an
oilin-water nanoemulsion of 17AAG which was evaluated in solid tumor patients
in a Phase I study and demonstrated acceptable toxicity and pharmacokinetic profiles.
Main side effects were nausea, vomiting, diarrhea, fatigue, anemia,
hyperbilirubinemia and hyperglycemia. Preliminary indications of biological activity
(mainly inhibition of Hsp90 function) were also observed.
Retaspimycin (IPI-504) (Infinity Pharmaceuticals, Inc.
Cambridge, MA) is a highly water soluble hydroquinone hydrochloride derivative
of 17-AAG with favorable pharmacological properties. A Phase I study in
patients with gastrointestinal stromal tumors (GIST) refractory to imatinib and
sunitinib showed significant decreases in PET avidity and stabilization of
disease suggestive for antitumor activity, with acceptable toxicity when
administered intravenously (i.v.). A Phase III study of IPI-504 is currently
undergoing in TKI refractory GIST patients. A Phase II study of IPI-504 in
patients with advanced NSCLC who had progressed to other prior tyrosine kinase therapy
showed RECIST partial responses in both wild type and mutated EGFR kinase patients.
Based on the single agent activity of retaspimycin in NSCLC, a Phase Ib study
of retaspimycin in combination with docetaxel was performed in patients with
advanced solid tumors and showed acceptable toxicity profiles and no
pharmacological interaction between these two drugs. This combination is being
is another GA analog also known as avespimycin (KOS-1022: Kosan Biosciences Inc,
Hayward, California). It is 3 to 5 fold more potent than 17AAG and it is water
soluble and available in oral and i.v. preparations. In Phase I studies in
refractory hematological malignancies and solid tumor patients and in
combination with transtuzumab in patients with metastatic breast cancer 17-DMAG
showed acceptable toxicity profiles, predictable pharmacokinetics, and evidence
of clinical activity. However, further development of this drug was halted in
favor of KOS-953.
Although the clinical activity of above mentioned GA
analogs provides proof of relevant Hsp90 target inhibition, the dose-limiting hepatotoxicity
remains a concern in the further development of these drugs. The hepatotoxicity
has been attributed to the benzoquinone moiety in the 17AAG molecule.
Furthermore resistance to 17AAG secondary to the high expression of
P-glycoprotein (P-gp) or due to the loss or mutation of the NAD(P)H/quinone oxidoreductase-1(NQO1)
gene for enzyme required for the bio-reduction of 17AAG to more potent
hydroquinone are also potential concerns. This led to development of agents that
retain Hsp90 inhibitory activity but are structurally different in benzoquinone
portion of the molecule thereby avoiding the above mentioned drawbacks of the
Hsp90 INHIBITORS WITH NOVEL SCAFFOLD STRUCTURES
Several novel Hsp90 inhibitors have been developed
based on diverse chemical scaffolds and are currently under clinical
investigation. These agents are reported to have improved stability, pharmacological
profiles, and do not appear to have the drug resistance issues seen with 17 AAG.
CNF-2024 (BIIB021) (Biogen Idec, Inc. Cambridge, MA) is
a fully synthetic orally available purin-scaffold Hsp90 inhibitor. In a Phase I
study in patients with advanced solid tumors and chronic lymphocytic leukemia (CLL),
patients were enrolled into two sequential dose cohorts of BIIB021 po daily for
3 weeks followed by 1 week off (CLL) or twice weekly for 3 weeks followed by 1
week off (solid tumors). The drug was well tolerated and demonstrated biological
activity as shown by changes in pharmacodynamic (PD) markers such as an
increase in Hsp70 expression and by the inhibition of the extracellular domain of
HER2/neu. Major toxicities were noticed as fatigue, hyponatremia, hypoglycemia and
abnormal liver function tests. Dose limiting toxicities were grade-3 syncope
and grade-3 dizziness at dose level of 800 mg twice weekly. A Phase II study of
CNF-2024 in patients with TKI refractory/intolerant GIST is currently ongoing
to assess the activity of this agent.
AUY922 (Novartis; Basel, Switzerland) is a novel
pyrazole scaffold, isoxazole-based synthetic Hsp90 inhibitor. In a Phase I
study, AUY922 demonstrated a favorable safety profile when administered intravenously
to patients with advanced solid tumors. MTD was established at 40 mg/m, with
main DLTs being anorexia, fatigue and diarrhea. PD analysis demonstrated dose
proportional induction of Hsp70 in peripheral blood mononuclear cells indicative
for biological activity. A Phase I/II study of AUY922 in patients with advanced
solid tumors and HER2/neu positive breast cancer is ongoing.
SNX-5422 is a 6, 7-dihydro-indazol-4-one scaffold,
highly potent and orally administered pro-drug that is converted into the
active metabolite, SNX-2122. STA-9090 (Synta) is another synthetic Hsp90
inhibitor administered intravenously. Both of these agents are being assessed in
various Phase I studies in patients with advanced solid tumors and refractory hematological
AND FUTURE DIRECTIONS
The development of Hsp inhibitors remains an active
area of oncology drug development. Although these are highly targeted
anticancer therapies, the biological impact on numerous important cellular
pathways has complicated their clinical testing. Further studies need to clarify
the precise Hsp90 client proteins responsible for antitumor activity and to identify
biomarkers for better patient selection. Development of further Hsp90 targeting
agents with clear clinical efficacy and minimal systemic toxicities is still an
evolving goal. Nonetheless Hsp90 remains a highly promising and scientifically fascinating
target for the treatment of human malignancies.