Management of Advanced/Metastatic Prostate Cancer: 2000 Update
Management of Advanced/Metastatic Prostate Cancer: 2000 Update
The clinical presentation of prostate
cancer has been evolving over the past several years, in part due to increased
public awareness of the disease and the availability of prostate-specific
antigen (PSA) as a screening serologic test in the late 1980s. Currently,
approximately 75% of prostate cancer patients present with clinically localized
disease, compared with about 50% in the mid-1980s.
Today, fewer patients with prostate cancer present with an
abnormal digital rectal exam. In many current series, the most common category
at initial presentation is T1C disease. This represents patients who have
elevated serum PSA levels without associated nodularities within the prostate
gland on rectal exam but are found to have prostate cancer on biopsy. On the
other hand, the incidence of metastatic prostate cancer has almost halved over
the past 15 years, so that about 10% to 15% of patients currently present with
clinical metastatic disease.
Although both prostate cancer incidence and mortality have begun
to decline in recent years, 37,000 men in the United States still died from this
disease in 1999, making it the second leading cause of US cancer deaths in men.
Invariably, prostate cancer deaths are due to progressive, metastatic disease
that has failed initial therapies.
Dramatic progress has been made in understanding the molecular
and biochemical pathways involved in the development and progression of prostate
cancer. Before discussing treatment approaches, we will describe a few examples
of the evolving biological principles, which can be grouped broadly into several
categories that are not mutually exclusive. Many of the following principles are
being incorporated into new treatment strategies.
In terms of normal physiology and malignant transformation, the
important interactions between the prostate epithelium and the underlying stroma
within the prostate gland are being increasingly recognized.[2,3] Dynamic
interactions that normally occur between the stroma, endothelial cells,
extracellular matrix, and prostate epithelium can be altered as prostate cancer
progresses and metastasizes. Potential alterations in intergrin-mediated
cell-cell and cell-extracellular matrix interactions/signaling represent one
A family of enzymes called matrix metalloproteases (MMPs), as
well as tissue inhibitors of MMPs, are involved in the physiologic remodeling of
the extracellular matrix. Perturbations in this remodeling process appear to
be an important step in tumor growth and invasion. Formation of new blood
vessels is a necessary step for the initial tumors to continue increasing in
size and subsequently metastasize. Prostate cancer cells can assume
angiogenic potential by secreting growth factors such as vascular endothelial
growth factor (VEGF) and basic fibroblast growth factor (bFGF). In turn,
endothelial cells recruited during angiogenesis can stimulate tumor growth by
secreting growth factors and cytokines such as interleukin (IL)-1 and IL-6.[6-8]
Changes in Cell Signaling
It is now well recognized that changes in intra- and
intercellular signaling response to growth factors, cytokines, cell-adhesion
molecules, and other ligands are a fundamental aspect of tumor pathogenesis. We
are now discovering that significant cross-talk can also occur between the
various signaling cascades. The erbB family of receptor tyrosine kinases, which
mediate signaling functions of epidermal growth factor (EGF)-like growth factors
and other ligands (heregulins), are among the most frequently implicated cell
surface receptors in human cancers.
HER2a member of the erbB family of receptor tyrosine kinasesis
overexpressed in a proportion of patients with prostate cancer, and presumably
contributes to altered cell functions. Although, to date, no direct ligands
for HER2 have been found, it appears to modulate signaling via dimerization with
other members of the erbB family. Studies suggest that IL-6 signaling in
prostate cancer cells via the IL-6 cytokine receptor requires direct
interactions of the latter with HER2, thus implicating cross-talk between
cytokine receptors and tyrosine kinase receptors.
Androgen-receptor signaling is fundamental to both normal and
malignant prostate physiology. Recent work suggests that ligand (ie,
androgen)-independent cross-talk can occur between HER2 and androgen receptors
during prostate cancer progression. Androgen-independent cross-talk has also
been shown to occur between androgen receptors and the protein kinase A (PKA)
signal-transduction pathways in prostate cancer cells. In particular, PSA
gene expression can be mediated via PKA-dependent phosphorylation of the
androgen receptor/coactivator(s) in an androgen-independent manner.
In addition to examples of the altered cell signaling noted
above, other metabolic/biochemical pathways can be modified as a consequence of
malignant transformation. In this regard, a key metabolic pathway involving
citrate metabolism has been implicated in prostate cancer. Studies have shown
that normal as well as benign hyperplastic prostate tissues accumulate very high
levels of citrate and zinc. On the other hand, malignant prostate tissue
cannot accumulate zinc or citrate.
Although the mechanisms of differential zinc uptake/transport in
the normal prostate gland vs prostate cancer have yet to be clarified, recent
work has shown that the high levels of zinc in the normal prostate inhibit the
enzyme m-aconitase, which prevents citrate from being oxidized, thus resulting
in the high citrate levels in the normal prostate. In contrast, the low
accumulation of zinc in prostate cancer cannot inhibit m-aconitase, thereby
further metabolizing citrate and leading to low levels of citrate in malignant
tissue. Metabolic products of arachidonic acidgenerated by the action of
12-lipoxygenasecan activate downstream targets like protein kinase C (PKC).
These products have been implicated in prostate cancer progression and
Calcium within cells can serve as an important signaling
molecule that modulates many cellular processes. For example, a rise in calcium
within the cytosol of cells can occur in response to activation of cell surface
receptors, and this, in turn, can trigger a variety of biological functions,
including muscle contraction, gene transcription, cell-cycle progression, and
In prostate cancer, disruption of intracellular calcium
homeostasis is a prominent feature of hormone ablative therapy, which remains
the cornerstone of treatment for metastatic disease. In particular, androgen
deprivation is associated with an increase in cytosolic calcium levels, which
normally are tightly regulated. This, in turn, triggers the apoptotic program
that results in cell death.
It has been suggested that a failure to generate such an
increase in cytosolic calcium in response to hormone ablation may be one of the
mechanisms responsible for the hormone-resistant phenotype that often occurs in
prostate cancer, resulting in the eventual failure of androgen ablative therapy
in this disease. On the other hand, apoptosis can be induced in
hormone-insensitive prostate cancer cells if elevated levels of cytosolic
calcium can be generated and sustained for several hours.
Ultimately, genetic alterations form the molecular basis for
many of the phenotypic changes in cell biochemistry, cell signaling, and
cell-cell interactions that occur as a consequence of malignant transformation.
Multiple chromosomal changes have been identified in prostate cancer, and
recently, a correlation has been found between tumor grade/tumor invasion and
frequency of loss in genetic heterozygosity.[22-24] Examples of genes whose
functions may be modified or inactivated during transformation include p53,
PTEN, and glutathione S-transferase pi.[25-27]
Another important aspect of prostate tumor biology is the
emergence of clinical resistance to initial hormone ablative therapy. Although a
majority of patients with metastatic prostate cancer respond to androgen
deprivation (see below), most ultimately do not, and consequently,
hormone-resistant prostate cancer (which is relatively resistant to chemotherapy
Whether androgen-dependent and androgen-independent prostate
cancer cells are present at the outset, or only androgen-dependent cells are
present initially but androgen-independent cells are selected for during hormone
ablation, remains an unresolved issue. However, some preclinical models favor
the former hypothesis.
Several mechanisms have been implicated in the development of
androgen independence, including changes within the androgen receptor and
overexpression of the antiapoptotic protein, bcl-2.[29-31] Overexpression and
modification of the latter can partly account for the pleiotropic resistance to
cytotoxic agents seen in hormone-resistant prostate cancer.
Evaluating the effectiveness of any cancer therapy requires a
definition of parameters that can be used to evaluate response. This has been a
particularly difficult problem in advanced prostate cancer, because in a
majority of patients with metastasis, the dominant sites of involvement are the
boneswhich are not readily amenable to the classic criteria of response to
therapy. These difficulties are reflected in earlier definitions of response
used by the National Prostate Cancer Project (NPCP) in the 1970s, where disease
"stabilization" was included in the overall response category.
Ever since PSA screening became available, it has been
incorporated in most clinical trials as one marker of disease progression or
response to treatment. Several reports have indicated a utility to using a 50%
or greater decline in PSA posttherapy as a marker of clinical benefit and
possibly prolonged survival.[33,34] However, in most studies, definitions vary
regarding the incorporation of PSA as a measure of disease progression or
response to therapy. Hence, the Prostate-Specific Antigen (PSA) Working Group
recently set a series of guidelines defining (1) the different categories of men
with metastatic, androgen-independent prostate cancer who might be eligible for
clinical trials; and (2) the criteria of disease progression and response
(including changes in PSA) to be used in evaluating patients in trials.
Several investigators have sounded a note of caution, however,
with respect to using PSA as a marker of response: Some of the newer,
"nontraditional" therapies in current clinical testing (including
certain differentiation agents, antiangiogenesis drugs, and growth-factor
modulators) may actually upregulate PSA gene expression.[36-38] Another aspect
regarding the evaluation of therapies is the use of palliative end points, such
as quality-of-life (QOL) measures and pain control. These parameters have
been incorporated into many of the trials now evaluating the role of
chemotherapy in androgen-independent prostate cancer.
Since prostate cancer is primarily dependent on androgens for
growth, the initial treatment for advanced/metastatic disease continues to be
suppression of testicular androgen production. This therapy was originally
described by Huggins 60 years ago in the form of surgical castration (ie,
bilateral orchiectomy), which effectively removes 90% of circulating
testosterone from the bloodstream.[39a]
Another way of nonsurgically suppressing testicular androgens is
via analogs of gonadotropin-releasing hormones [GnRH]), which have been
available since the 1970s. The GnRH analogs (leuprolide [Lupron], goserelin
[Zoladex]) are supra-agonists that, upon binding to the luteinizing hormone-releasing
hormone (LHRH) receptor within the pituitary gland, cause an initial surge of
LH, and thus, testosterone, release. However, the continued receptor occupancy
causes internalization, degradation, and desensitization of the LHRH receptor.
This, in turn, leads to decreased testicular androgen synthesis, and hence
castrate levels of testosterone within 3 to 4 weeks of drug administration.
Due to the more favorable toxicity profile of the GnRH analogs
compared to diethylstilbestrol (Stilphostrol), the GnRH analogsdespite their
expensehave become the treatment of choice for achieving medical castration
in the United States. Newer long-acting pure antagonists of GnRH (which avoid
the initial LH/testosterone surge) are undergoing clinical trials.
Effective nonsurgical castration can also be achieved with
estrogens like diethylstilbestrol (administered at 3 to 5 mg/d). However, due to
the increased risk of cardiovascular toxicities, this form of therapy has fallen
out of favor, at least in the United States.
Approximately 80% of patients with advanced prostate cancer
respond initially to either medical or surgical castration. Substantial
responses occur in a majority of patients with soft-tissue disease, including
normalization of elevated PSA levels in up to 70% of patients as well as
stabilization and improvement of bone lesions in a significant proportion. Those
with pain resulting from bone metastasis can achieve almost immediate pain
relief upon surgical castration.
Although gonadal androgen suppression removes about 90% of
circulating testosterone, other androgens (that are primarily adrenally derived)
can also potentially exert tumor-promoting effects. Hence, attempts have been
made to suppress the action of the remaining circulating androgens via the use
of steroidal (cyproterone, megestrol) or nonsteroidal antiandrogens
(bicalutamide [Casodex], flutamide [Eulexin], nilutamide [Nilandron]).
Particularly with the availability of the nonsteroidal
antiandrogens (which act by blocking the peripheral androgen receptor),
substantial effort and resources have been devoted to determining whether total
androgen suppression (achieved by medical or surgical castration plus the use of
antiandrogens) is better than gonadal androgen suppression with respect to
response rates and survival in patients with advanced prostate cancer.[42,43]
Despite a plethora of prospective, randomized trials addressing this issue (27
such trials conducted to date), any significant advantages of total androgen
suppression over gonadal androgen suppression have not been clearly
demonstrated. If any survival advantages are seen with total androgen
suppression, they are likely to occur primarily in patients with metastatic
disease who have minimal tumor burden.
Peripheral androgen blockade refers to the use of antiandrogens
in conjunction with inhibitors of 5-alpha reductase (finasteride [Proscar]), an
enzyme that converts testosterone to its more active form of
dihydrotestosterone. With this approach, serum testosterone levels are less
likely to be suppressed to castrate levels, and therefore, some side effects
associated with gonadal androgen suppression/total androgen suppression (ie, hot
flashes, osteoporosis, anemia, muscle weakness, and impotence) are likely to be
Although peripheral androgen blockade can provide control of
advanced prostate cancer, questions remain regarding its overall efficacy.
Recent randomized trials comparing antiandrogen monotherapy with gonadal
androgen suppression or total androgen suppression confirm the superiority of
the latter approaches with respect to survival in patients with metastatic
prostate cancer.[45,46] Therefore, gonadal androgen suppression plus or minus
antiandrogen therapy remains the most effective first-line treatment in patients
with advanced prostate cancer.