| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Department of Medicine, University of Texas Health Science Center at San Antonio, San Antonio, Texas 78284-7877
 |
Abstract |
|---|
 Top Abstract I. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
|
I. Introduction |
|---|
|
TopAbstractI. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
Since the time of Paget, it has become clear that cancer affects bone in several ways: 1) indirectly through elaboration of factors that act systemically on target organs of bone and kidney to disrupt normal calcium homeostasis; 2) locally and directly via secondary spread of tumor to bone; and 3) via direct involvement by primary bone tumors. As primary bone tumors comprise a small minority of all tumors affecting bone, this review will focus on the aspects of cancer and bone related to the former only.
The three most common neoplasms in humans, breast, prostate, and lung cancer, frequently affect the skeleton. Since the majority of patients dying of cancer have bone involvement either through metastatic spread or as a result of the systemic effects of tumor-produced factors on bone and kidney, this is not a trivial problem. In 1996 alone, the estimated number of new cancer cases in men included 317,000 cases of prostate cancer and 98,900 cases of lung cancer while 184,300 cases of breast cancer and 78,100 cases of lung cancer were diagnosed in women (2). Furthermore, in the same year, prostate and lung cancer were responsible for 41,400 and 94,400 deaths, respectively, in men while breast and lung cancer deaths in women totaled 44,300 and 64,300 individuals, respectively (2). Despite advances in cancer therapy, cancer statistics indicate that the mortality rate of lung cancer is still rising for women, even though 1996 is the first year that it has leveled off for men. Additionally, the age-adjusted death rates of prostate cancer continue to rise, and although 1996 was the first year that a slight decrease in mortality due to breast cancer was observed, the age-adjusted death rate for breast cancer remains similar to that of 1930 (2). Thus, to improve therapy and prevention, it is important to understand the pathophysiology of the effects of cancer on bone as it will be a continued source of morbidity for years to come. Although the topic is an expansive one, this review will attempt to detail scientific advances in this area regarding the pathophysiology of the effects of cancer on bone.
|
II. Normal Calcium and Bone Homeostasis |
|---|
|
TopAbstractI. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
A. Bone remodeling
Bone is unique among target tissues affected by cancer as it is
being continually remodeled under the influence of systemic hormones
and local bone-derived growth factors. Bone consists of two physically
and biologically distinctive structures. The outer cortical bone is
hard mineralized matrix in which cellular and metabolic activities are
relatively low. Cortical bone makes up 85% of the total bone in the
body and is most abundant in the long bones of the appendicular
skeleton. The volume of cortical bone is regulated by the formation of
periosteal bone, by remodeling within Haversian systems, and by
endosteal bone resorption. Cancellous or trabecular bone constitutes
the remaining 15% of the skeleton and is most abundant in the
vertebral bodies. The adult skeleton is in a dynamic state as the
coordinated actions of osteoclasts and osteoblasts on trabecular
surfaces and in Haversian systems result in continual bone resorption
and formation. The normal mineralization of bone matrix is contingent
upon adequate amounts of vitamin D, calcium, and phosphate. The
mineralized bone matrix contains abundant amounts of growth factors,
transforming growth factor-ß (TGFß) and insulin-like growth factor
II (IGF-II) comprising the majority (3, 4). Such growth factors are
released from the bone matrix as a result of osteoclastic bone
resorption (5), a component of the normal remodeling process necessary
to maintain the structural integrity of bone. The inner portion of bone
consists of multicellular bone marrow in which hematopoietic stem
cells, stromal cells, and immune cells reside. The hematopoietic stem
cells have the potential to differentiate into the blood-forming
elements and bone-resorbing osteoclasts, while the stromal cells
support the differentiation of the hematopoietic cells as well as from
bone-producing osteoblasts. Cells in the bone marrow, stromal and
immune in particular, produce cytokines and growth factors that mediate
cell-to-cell interactions in autocrine, paracrine, and/or juxtacrine
fashions (6). Thus, tumor secretion of hormones that act systemically
on bone may disrupt normal calcium homeostasis and bone remodeling to
result in hypercalcemia and bone loss while tumor-produced phosphaturic
factors may result in osteomalacia. Likewise, once cancer cells arrest
in bone, the high concentrations of growth factors and cytokines in the
bone microenvironment provide a fertile soil on which the cells can
grow. Furthermore, when the tumor cells stimulate osteoclastic bone
resorption, this bone microenvironment is even more enriched with
bone-derived growth factors that enhance survival of the cancer and
similarly disrupt normal bone remodeling to result in bone destruction.
B. Calcium homeostasis
Blood-ionized calcium concentrations are remarkably stable in
normal individuals due to a complex regulatory system involving the
actions of three calciotropic hormones on the target organs of bone,
gut, and kidney. Calcium exchanged between the extracellular fluid and
these target organs normally remains in zero balance (Fig. 1
). Normal calcium homeostasis is
dependent on the interactions of PTH,
1,25-(OH)2D3, and calcitonin on these organs to
maintain the ionized calcium concentration within a very narrow range.
Regulation of normal calcium homeostasis has been extensively reviewed
by Chattopadhyay et al. (7) as well as by Parfitt (8, 9).
|
) and represents a simple negative
feedback loop (10). Similar to other endocrine hormones, such as those
secreted by the anterior pituitary, PTH is secreted in a pulsatile
fashion in the normal state as well as in states of primary
hyperparathyroidism (12). The calcium-sensing receptor that mediates
this negative feedback has been cloned from bovine parathyroid cells
(13) and is mutated in the disorders of familial benign hypocalciuric
hypercalcemia (13, 14, 15, 16, 17, 18, 19) and autosomal dominant hypocalcemia (20, 21).
Active vitamin D metabolites decrease PTH synthesis in vitro
and in vivo (22, 23) as well.
|
Metabolism of PTH is complex, and the intact and biologically active peptide has a half-life of less than 4 min (29). Intact PTH is cleared rapidly by kidney and liver (30, 31, 32, 33, 34). Carboxy-terminal fragments circulate significantly longer than the intact hormone, mainly because they are cleared exclusively by glomerular filtration (35, 36). Highly sensitive and specific immunoradiometric assays for intact PTH are now widely available (37) and are extremely useful when employed in the differential diagnosis of hypercalcemia.
2. Calcitonin. Plasma-ionized calcium concentration is the most important regulator of calcitonin secretion (38). Increases in plasma-ionized calcium result in an increase in calcitonin secretion and, conversely, a fall in the ambient calcium concentration inhibits calcitonin secretion. These changes are likely mediated through the calcium-sensing receptor, as the parafollicular cells of the thyroid gland express the same calcium-sensing receptor that is expressed in the parathyroid and kidney (39). Gastrointestinal peptide hormones, gastrin in particular, are potent calcitonin secretagogues. Although the physiological significance of this observation remains unclear, it is the basis for the pentagastrin stimulation test, a provocative test to determine the capacity of a patient to secrete calcitonin (40).
The precise biological role of calcitonin in the overall schema of calcium homeostasis is uncertain. Calcitonin directly inhibits osteoclastic bone resorption (41), and the effect is rapid, occurring within minutes of administration. This inhibition is accompanied by the production of cAMP (42), as well as an increase in cytosolic calcium (43) in the osteoclast, and results in contraction of the osteoclast cell membrane (44). These effects are transient and likely have little role in chronic calcium homeostasis. Clinical observations support this since neither calcitonin-deficient patients (athyroid) nor patients with medullary thyroid cancer and excess calcitonin production experience alterations in calcium homeostasis. The calcitonin receptor (45) is a G protein-coupled receptor with seven-transmembrane domains that is structurally similar to the PTH/PTHrP and secretin receptors. The half-life of calcitonin is measured in minutes and metabolism occurs predominantly in the kidney (38). Clinical abnormalities of calcitonin secretion include medullary thyroid carcinoma, small cell lung cancer, and carcinoids and islet cell tumors of the pancreas.
3. Calcitriol. The steroid hormone calcitriol or
1,25-(OH)2D3 is the major biologically active
metabolite of the vitamin D sterol family. Vitamin D precursor
(previtamin D3) is either ingested in the diet or
synthesized in the skin from 7-dehydrocholesterol through exposure to
sunlight (46, 47). Hydroxylation occurs in the liver at the C-25
position to form 25-hydroxyvitamin D [25(OH)D], the precursor of the
more potent metabolite, 1,25-(OH)2D3. 25(OH)D
is hydroxylated at the C-1 position in the kidney by 1
-hydroxylase,
a complex cytochrome P450 mitochondrial enzyme system
located in the proximal nephron (48), to form
1,25-(OH)2D3 (49, 50, 51). The renal
1-hydroxylation of 25(OH)D is the major recognized control point in
vitamin D metabolism, responding to ambient phosphorus, PTH, and
calcium concentrations. PTH and low serum phosphate concentrations
independently increase 1,25-(OH)2D3 production,
while hypercalcemia and 1,25-(OH)2D3 inhibit
renal 1-hydroxylase activity. Under physiological conditions, the
kidney is the sole source of 1,25-(OH)2D3. The
other known important extrarenal sites of
1,25-(OH)2D3 production are the placenta and
granulomatous tissue (52, 53, 54). The half-life of
1,25-(OH)2D3 in the circulation is
approximately 5 h in humans. Fifteen percent is excreted as
urinary metabolites and 50% as fecal metabolites.
1,25-(OH)2D3 increases plasma calcium and
phosphate concentrations by increasing the absorption of calcium and
phosphate from the gastrointestinal tract (51). It also increases bone
resorption (55) and enhances the capacity for PTH to promote renal
tubular calcium reabsorption in the nephron. It is a powerful
differentiation agent for committed osteoclast precursors (56, 57),
causing their maturation to multinucleated cells that are capable of
resorbing bone. Thus, 1,25-(OH)2D3 ensures a
supply of calcium and phosphate available at bone surfaces for the
mineralization of bone matrix. Deficiency of
1,25-(OH)2D3 or of 1-hydroxylase results in
osteomalacia or rickets, as does resistance to
1,25-(OH)2D3, caused by mutations in the
vitamin D receptor (58, 59, 60, 61). Although the function of other vitamin D
metabolites has been unclear, recent evidence from mice deficient in
the 24-hydroxylase gene indicate that such metabolites have a role in
normal bone metabolism. Deficiency of 24-hydroxylase results in lack of
the vitamin D metabolites hydroxylated at the 24 position and abnormal
bone structure consisting of accumulation of osteoid at sites of
intramem-branous ossification (62).
C. Defenses against hyper- and hypocalcemia
The normal physiological defenses against hypercalcemia and
hypocalcemia are listed in Table 1. The
majority of these defense mechanisms are mediated through the hormonal
actions of PTH and 1,25-(OH)2D3. Although the
role of endogenous calcitonin is relatively modest in comparison to PTH
and 1,25-(OH)2D3, pharmacological calcitonin
therapy can be beneficial as discussed later.
|
In the converse situation, a rise in ionized calcium concentration results in decreased PTH secretion from the parathyroid glands. Thus, renal tubular calcium reabsorption is decreased, as is osteoclastic bone resorption. Synthesis of 1,25-(OH)2D3 and, subsequently, gastrointestinal absorption of dietary calcium and phosphate are decreased. Thus, the normal response to increases in ionized calcium is an increase in renal calcium excretion and a decrease in intestinal absorption of calcium.
In general, these hormonal responses are more effective in protecting against hypocalcemia than hypercalcemia. Perturbations in these mechanisms, as exemplified by excessive increases in bone resorption, deficiencies or excess of PTH or 1,25-(OH)2D3, and defects in renal capacity to handle calcium and phosphate, will result in either hypercalcemia or hypocalcemia.
|
III. Humoral Mechanisms by Which Solid Tumors Affect the Skeleton |
|---|
|
TopAbstractI. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
1. Clinical features of hypercalcemia. The clinical features
of hypercalcemia are listed in Table 2.
Symptoms may vary in individual patients and are related both to the
absolute concentration of serum calcium and to the rate of rise in
serum calcium. Symptoms also reflect the underlying cause of the
hypercalcemia as well as intercurrent medical conditions. In older or
critically ill patients, symptoms of hypercalcemia may be more
prominent with relatively small increases in serum calcium
concentration. Hypercalcemia most often results in neuromuscular,
gastrointestinal, and renal manifestations. Severe hypercalcemia is
likely the result of a vicious cycle. The hypercalcemic effects of
anorexia, nausea, vomiting, and impaired renal concentrating ability
lead to dehydration and, subsequently, altered mental status. This, in
turn, may promote immobilization and lead to worsening hypercalcemia.
In addition to the symptoms of hypercalcemia, clinical features of
hypercalcemia of malignancy include signs and symptoms of the
underlying cancer. Generally, the cancer is well advanced when
hypercalcemia occurs, and the prognosis is poor. Survival beyond 6
months is uncommon (64, 65, 66).
|
(67). Hypercalcemia occurring in the
setting of malignancy may be due to 1) humoral factors secreted by
tumors that act systemically on target organs of bone, kidney, and
intestine to disrupt normal calcium homeostasis; 2) local factors
secreted by tumors in bone, either metastatic or hematological, which
directly stimulate osteoclastic bone resorption; and 3) coexisting
primary hyperparathyroidism.
|
a. PTHrP.
Although cancer has been associated with
hypercalcemia since the 1920s, it wasn’t until 1941 that Fuller
Albright first proposed the syndrome of ectopic hormone production in a
patient with hypercalcemia, renal carcinoma, and a solitary bone
metastasis (68). As the patient had normal parathyroid glands at the
time of the neck operation, Albright postulated that the tumor
produced PTH but when he had Collip assay the tumor for PTH, it
was not detected. Thereafter, the syndrome was often referred to as
"pseudohyperparathyroidism" or "ectopic hyperparathyroidism."
When more sensitive PTH assays became available, it was clear that the
offending factor was not PTH, but rather an immunologically distinct
factor that had PTH-like biological activity (69).
Biochemical features of malignancy-associated hypercalcemia share some similarities to those seen in primary hyperparathyroidism (1oHPT). Specifically, in addition to the hypercalcemia, hypophosphatemia can occur (depending on the patient’s renal function) as well as hypercalciuria, hyperphosphaturia, and increased excretion of nephrogenous cAMP (69). Plasma-intact PTH concentrations are suppressed in this syndrome except in rare cases of ectopic PTH production or concomitant primary hyperparathyroidism. Plasma 1,25-(OH)2D3 concentrations are generally low except in certain hematological malignancies when the humoral mediator is 1,25-(OH)2D3.
Almost 50 yr after Albright’s observations, this PTHrP was purified
from human lung cancer (70), breast cancer (71), and renal cell
carcinoma (72) simultaneously by several independent groups and was
cloned shortly thereafter (73). It is now evident that PTHrP, and not
PTH, is a major mediator of humoral hypercalcemia of malignancy (74, 75), although four cases of authentic tumor-produced PTH have been
reported (76, 77, 78, 79). PTHrP has 70% homology to the first 13 amino acids
of the N-terminal portion of PTH (73), binds to PTH receptors (80), and
shares similar biological activity to PTH (81). Specifically, it
stimulates adenylate cyclase in renal and bone systems (71, 72, 81, 82, 83), increases renal tubular reabsorption of calcium and
osteoclastic bone resorption (82, 83), decreases renal phosphate uptake
(81, 82, 84), and stimulates 1-hydroxylase (81). PTHrP has been
found in a variety of tumor types associated with hypercalcemia
including squamous, breast, and renal carcinoma (85, 86). Although the
majority of squamous cell carcinomas produce PTHrP (87), the capacity
to cause hypercalcemia may depend on the level of PTHrP gene
expression, which in turn may be determined by differential
transcription of the PTHrP gene promoters (88). The regulation of PTHrP
is complex, and factors such as PRL (89), epidermal growth factor (EGF)
(90, 91, 92, 93), insulin (91), IGF-I (91, 94) and II (91), TGF (95), TGFß
(93, 96, 97, 98), angiotensin II (99), stretch (100), and the
src protooncogene (101) have been shown to increase
expression, while glucocorticoids (91, 94, 102, 103) and
1,25-(OH)2D3 (90, 91) decrease it. Estrogen has
been shown to increase PTHrP expression in uterine tissue, and
in vitro studies suggest that an estrogen response
element is present in the PTHrP gene (104, 105). Mutations in codons
248 and 273 of the p53 tumor suppressor gene repress PTHrP gene
expression in some squamous cell carcinomas (106). The cell death
inhibitor, Bcl-2, is downstream in a signaling pathway that is required
for normal skeletal development (107).
The human PTHrP gene is much larger and more complex than the human PTH gene. It spans approximately 15 kb of genomic DNA and has nine exons and three promoters. Three PTHrP isoforms of 139, 141, or 173 amino acids as well as multiple PTHrP mRNA species exist (93). There is considerable sequence homology across species up to amino acid 111 (108). Cell line-specific utilization of the promoters and of the 3'-alternative splicing pathways among bone, breast, kidney, and lung cell lines have been demonstrated (93). In this study, dexamethasone decreased while EGF and TGFß increased abundance of each of the alternative mRNA species. Furthermore, EGF treatment increased transcription from promoters 1 and 2 and stabilized exon VII- and IX-containing transcripts in various cell lines (93).
Like PTH and other endocrine peptides, PTHrP undergoes endoproteolytic posttranslational processing that results in several secretory forms: 1) an amino-terminal PTHrP-(1–36); 2) a midregion species that begins at amino acid 38 that has an undefined carboxyl terminus (109, 110); and 3) a carboxyl-terminal species that is recognized by an antibody directed against the 109–138 region (110, 111, 112). The preponderance and arrangement of basic residues in the protein sequence suggest that members of the subtilisin family of endoproteases, such as furin (113), PC 1/3, PC-2, PACE-4, and PC8 (114), are responsible for such processing (109, 115, 116). Posttranslational modification of PTHrP also occurs as glycosylation of an amino-terminal PTHrP species produced by keratinocytes has been reported (117). The subject of posttranslational processing of PTHrP, as well as the receptor and signal transduction pathways employed by the mature secretory forms of PTHrP, has been extensively reviewed by Orloff et al. (115). Regulation of PTHrP secretion may be cell specific as PTHrP expressed in neuroendocrine cells is secreted in a regulated fashion as compared with a constitutive secretion when expressed in nonneuroendocrine cell types such as squamous cell carcinoma (116). Although PTHrP mediates its calcemic effects through the classic PTH/PTHrP receptor, there is evidence for a separate PTHrP receptor (118). However, the function of such a receptor remains unclear.
PTHrP has been detected in a variety of tumor types as well as in normal tissue (85, 86). The widespread expression of PTHrP in normal tissue was the first evidence that the hormone had a role in normal physiology. In addition to the PTH-like effects, emerging work testifies to the fact that PTHrP has an important role in normal physiology. Such a topic is beyond the scope of this review, but suffice it to say that PTHrP appears to be important in 1) the regulation of cartilage differentiation and bone formation through endochondral ossification (119, 120, 121); 2) growth and differentiation of skin (122), mammary gland (123, 124), and pancreatic islets (125); 3) cardiovascular function (126); 4) transepithelial calcium transport in the distal nephron, mammary epithelia, and the placenta (124, 127); 5) relaxation of smooth muscle in uterus, bladder, arteries, stomach, and ileum (99, 128, 129, 130); and 6) host immune function (131, 132, 133). Bone perichondrial cell production of PTHrP regulates cartilage cell differentiation and has been linked to expression of Indian Hedgehog gene (134). Indian Hedgehog protein expressed by prehypertrophic cartilage cells inhibited cartilage differentiation, and this inhibitory effect was mediated by PTHrP. The normal physiological functions of PTHrP have been extensively reviewed elsewhere (135, 136).
The role of PTHrP in normal breast physiology (137, 138) sheds light on its potential importance in the pathophysiology of hypercalcemia and bone metastasis associated with breast cancer, which is discussed in later portions of this review. PTHrP is expressed in lactating mammary tissue (139) and secreted into milk at concentrations 10,000–100,000 times greater than plasma concentrations of humans with malignancy-associated hypercalcemia (140, 141, 142, 143, 144). Increased plasma PTHrP concentrations have been documented in at least two patients with the rare syndrome of lactational hypercalcemia (145, 146, 147) in addition to some breast-feeding mothers (148). Thus, PTHrP may be responsible for mobilizing calcium from maternal bone for use in milk production, and it may mediate lactation-associated bone loss.
In addition to the diverse and accumulating normal physiological functions of PTHrP, it likely has a multifunctional role in cancer as well. Such identified functions include 1) mediating hypercalcemia; 2) aiding in the development and progression of osteolytic bone metastasis in breast cancer; 3) regulating growth of cancer cells (149, 150, 151); and 4) acting as a cell survival factor (107).
The hypercalcemia of malignancy syndrome was the first identified consequence of the PTHrP effects in cancer. In this syndrome, tumor-produced PTHrP interacts with PTH receptors in bone and kidney to cause hypercalcemia, osteoclast-mediated bone resorption, and increased nephrogenous cAMP and phosphate excretion. The PTH-like properties of PTHrP, and specifically increasing osteoclastic bone resorption and renal tubular calcium reabsorption, appear to be responsible for the hypercalcemia. Approximately 80% of hypercalcemic patients with solid tumors have detectable or increased plasma PTHrP concentrations (111). Plasma PTHrP concentrations, as measured by a sensitive two-site immunoradiometric assay, are low or undetectable in the plasma of normals (152), but as is the situation with PTH, the C-terminal fragment is increased in patients with chronic renal failure (111). In fact, the plasma C-terminal PTHrP concentration increases as the glomerular filtration rate decreases (112).
i. Humoral hypercalcemia of malignancy (HHM) vs.
1oHPT.Despite similarities between HHM and
1oHPT and the similar biological actions of PTHrP and PTH,
respectively, unexplained differences between these syndromes exist.
First, patients with PTHrP-mediated HHM have low serum concentrations
of 1,25-(OH)2D3 compared with patients with
1oHPT (153), even though both proteins stimulate renal
1-hydroxylase activity. Clinical studies in which normal humans
received short-term infusions of PTHrP-(1–34) (154) or PTHrP-(1–36)
(155) revealed increased serum 1,25-dihydroxyvitamin D concentrations
comparable to those who had received a similar infusion of PTH-(1–34).
Animal studies have revealed similar findings (156, 157). Female nude
mice infused with synthetic PTHrP-(1–40) for 7 days developed
hypercalcemia, hypophosphatemia, and increased serum
1,25-dihydroxyvitamin D concentrations (157). Likewise, male nude mice
bearing Chinese hamster ovarian (CHO) cell tumors transfected with the
cDNA for human prepro-PTHrP or prepro-PTH developed similar
hypercalcemia and increased plasma concentrations of
1,25-dihydroxyvitamin D when compared with control animals bearing
untransfected CHO tumors (158). Additionally, similar increases in
blood ionized calcium and 1,25-dihydroxyvitamin D concentrations were
observed in nude mice bearing CHO tumors that were engineered to
secrete PTHrP mutants truncated at the carboxyl terminus (156).
Second, human studies using either quantitative bone histomorphometry (159) or biochemical markers of bone turnover (160) have demonstrated that although patients with either HHM or 1oHPT have increased osteoclastic bone resorption, many patients with HHM do not have the coupled increase in osteoblastic bone formation that those with 1oHPT have. Serum osteocalcin concentrations, a marker for bone formation, were significantly increased in patients with 1oHPT compared with normals (161). In the same study, serum osteocalcin concentrations in hypercalcemic patients with bone metastasis were significantly lower compared with those of normal controls, while normocalcemic patients with bone metastases had values similar to normal humans. These osteocalcin concentrations correlated with histomorphometric parameters of bone formation but not bone resorption (161). In the studies by Fraher et al. (154) and Everhart-Caye et al. (155), in which normal humans received infusions of PTH or PTHrP, bone histomorphometry or biochemical markers of bone turnover were not measured. Such studies done with PTHrP infusions in rodents have revealed increased osteoclastic bone resorption, as well as increased bone formation, as assessed by dynamic bone histomorphometry (157). In contrast, nude mice bearing a PTHrP-secreting human squamous cell carcinoma demonstrated increased bone resorption and decreased bone formation as assessed by dynamic bone histomorphometry (162). Thus, whether PTHrP alone is responsible for the uncoupling of bone formation from bone resorption is an issue that remains controversial.
Unlike the metabolic acidosis seen in patients with primary
hyperparathyroidism, patients with malignancy-associated hypercalcemia
often have a metabolic alkalosis with a low plasma chloride and high
plasma bicarbonate concentration. Although many explanations have been
postulated for the discrepancies between HHM and 1oHPT,
such as differences between the pulsatile secretion of PTH and the
presumed continuous secretion of PTHrP, suppression of bone formation
and 1-hydroxylase activity by other tumor-associated factors,
biologically active PTHrP fragments, or hypercalcemia per se
(163), the reasons for these differences have not been adequately
elucidated.
ii. Modulation of PTHrP effects by other tumor-associated
factors. Regardless of the reasons for the clinical differences
between HHM and 1oHPT, there is clear evidence that other
tumor-produced factors can modulate the end-organ effects of PTHrP as
well as its secretion from tumors. Using an in vivo model of
PTH and PTHrP-mediated hypercalcemia, Uy et al. (164)
demonstrated that both proteins, when produced by tumors in which the
corresponding genes were transfected and then inoculated into nude
mice, caused similar hypercalcemia as well as increases in osteoclastic
bone resorption, more committed marrow mononuclear osteoclast
precursors, and mature osteoclasts. No stimulatory effects were seen on
the multipotent osteoclast precursors, the granulocyte/macrophage
colony-forming unit. In a similar model system, IL-6 potentiated the
hypercalcemia and bone resorption mediated by PTHrP in vivo
by stimulating production of early osteoclast precursors (165).
Likewise, TGF has been shown to enhance the hypercalcemic effects of
PTHrP in an animal model of malignancy-associated hypercalcemia (166)
as well as to modulate the renal and bone effects of PTHrP (167, 168).
Sato et al. (169) demonstrated that IL-1 and PTHrP may
have synergistic effects in vivo, and others have shown that
IL-1 may modulate the renal effects of PTHrP (170). Finally, Uy
et al. have demonstrated that tumor necrosis factor-
(TNF) enhanced the hypercalcemic effect of PTHrP by increasing the
pool of committed osteoclast progenitors with a subsequent increase in
osteoclastic bone resorption. Bone formation parameters in these nude
mice indicate that TNF did not inhibit the new bone formation
stimulated by PTHrP (171).
Such tumor-associated factors also appear to be important regulators of
PTHrP expression and secretion by tumors. EGF has been shown to
stimulate PTHrP expression in a keratinocyte cell line (172) as well as
a mammary epithelial line (91) while TGF enhanced PTHrP expression
in a human squamous cell carcinoma of the lung (173). Interleukin-6
(IL-6), TNF, IGF-I, and IGF-II increased the production of PTHrP
in vitro by a human squamous cell carcinoma (94). TGFß,
which is abundant in bone, released in active form by resorbing bone
and expressed by some breast cancers (174, 175) and cancer-associated
stromal cells (176), has been shown to enhance secretion of and
stabilize the message for PTHrP in a renal cell carcinoma (96) as well
as in a squamous cell carcinoma (97, 98). Other data (177, 178)
demonstrate that this relationship also exists in a human breast
adenocarcinoma cell line, MDA-MB-231.
iii. PTHrP in hypercalcemia associated with breast cancer. Hypercalcemia in breast cancer represents a special situation. Although it is clear that the predominant way in which breast cancer affects bone is through metastatic mechanisms, there is sufficient evidence to support the notion that breast cancers may secrete factors that act systemically to stimulate osteoclastic bone resorption and to increase renal tubular reabsorption of calcium (179, 180, 181). Hypercalcemia is associated with breast cancer, occurring in approximately 10% of afflicted women during the course of their disease (182). It is likely more common in those with advanced breast cancer. Osteoclast-mediated skeletal destruction by metastatic tumor is a major mechanism responsible for hypercalcemia, as increased osteoclastic bone resorption has been documented histologically in areas surrounding breast cancer metastases (183, 184, 185, 186). However, humoral mechanisms may contribute in 10–60% of cases of breast cancer-associated hypercalcemia as evidenced by increased nephrogenous cAMP and plasma PTHrP in some patients (186, 187, 188, 189, 190).
PTHrP is clearly a significant factor in mediating hypercalcemia in breast cancer (191). Since PTHrP is expressed in normal breast tissue and appears to play an important role in normal breast physiology, its overproduction in breast cancer is not surprising. One of the three tumors from which PTHrP was originally purified was a breast cancer from a patient with humoral hypercalcemia of malignancy (70). PTHrP was detected by immunohistochemical staining in 60% of 102 invasive breast tumors removed from normocalcemic women, but not in normal breast tissue (192). At least four other studies have confirmed these percentages (193, 194, 195, 196), and one of these has demonstrated immunoreactive PTHrP within the cytoplasm of lobular and ductal epithelial cells in normal and fibrocystic breast tissues (193). Furthermore, 65–92% of hypercalcemic breast cancer patients (with and without bone metastasis) had detectable plasma PTHrP concentrations by RIA similar to those documented in patients with humoral hypercalcemia of malignancy due to nonbreast tumors (180, 195). Not only is PTHrP an important mediator of hypercalcemia in breast cancer, it may have a significant role in the pathophysiology of breast cancer metastasis to bone as evidenced by the clinical studies indicating that PTHrP expression by the primary breast cancer is more commonly associated with the development of bone metastasis and hypercalcemia (195). This topic will be discussed in a later section.
iv. PTHrP in hypercalcemia associated with hematological malignancies. The mechanisms responsible for hypercalcemia associated with hematological malignancies are multifactorial and include secretion of local bone-active cytokines, such as IL-6, IL-1, and lymphotoxin or TNFß, from tumor in bone or from systemic effects of tumor-produced factors such as 1,25-(OH)2D3 (discussed below). Recent data from a clinical study of 76 patients with various hematological malignancies demonstrate that PTHrP also may be an important pathogenetic factor in the development of hypercalcemia in some patients (197). In this study, eight of the 14 hypercalcemic patients had non-Hodgkin’s lymphoma and, of these, 62% had significant increases in plasma PTHrP concentrations. The serum 1,25-(OH)2D3 concentrations, when measured, were low in the hypercalcemic non-Hodgkin’s lymphoma patients who had increased plasma PTHrP concentrations (197). Additionally, one of two hypercalcemic patients with Hodgkin’s disease and one of four hypercalcemic patients with multiple myeloma had increased plasma PTHrP concentrations. Also of interest in this study is the fact that several normocalcemic patients with non-Hodgkin’s lymphoma, Hodgkin’s lymphoma, multiple myeloma, and Waldenstrom’s macroglobulinemia had increased plasma PTHrP concentrations as measured by an amino-terminal PTHrP assay (197). Using a sensitive two-site immunoradiometric assay, other investigators have noted increased plasma PTHrP concentrations in patients with adult T cell leukemia and B cell lymphoma (198). Finally, circulating concentrations of PTHrP, comparable to those in humoral hypercalcemia of malignancy, were present in two of four hypercalcemic patients with non-Hodgkin’s lymphoma, in three of nine with myeloma (199), and in a patient with myeloid blast crisis of chronic myeloid leukemia (200). Thus, the humoral mediators in the hypercalcemia associated with hematological malignancies include both 1,25-(OH)2D3 and PTHrP.
b. 1,25-(OH)2D3.
In the setting of
hypercalcemia, serum concentrations of
1,25-(OH)2D3 are normally suppressed unless an
autonomous source of PTH is the cause, such as a parathyroid adenoma.
Lack of 1,25-(OH)2D3 suppression in this
situation is evidence of disordered regulation of
1,25-(OH)2D3 synthesis and indicates extrarenal
production such as that observed in the hypercalcemia associated with
granulomatous disease. Less commonly, tumors may secrete other humoral
factors responsible for hypercalcemia. A major mediator of
hypercalcemia in Hodgkin’s disease, non-Hodgkin’s lymphoma, and other
hematological malignancies appears to be extrarenal production of
1,25-(OH)2D3 (201). The mechanism is similar to
that observed in hypercalcemia associated with granulomatous disease in
which activated macrophages within the granuloma synthesize
1,25-(OH)2D3 (52, 202, 203). In this scenario,
patients usually have increased plasma
1,25-(OH)2D3 concentrations in addition to low
or normal plasma PTH and urinary cAMP concentrations (204) in the
absence of bone involvement. In similar studies, affected patients have
also been shown to have increased fasting urinary calcium excretion
(204) as well as increased intestinal calcium (47Ca)
absorption (205). Increased 1,25-(OH)2D3
concentrations were noted in 12 of 22 hypercalcemic patients with
non-Hodgkin’s lymphoma. In addition, 71% of 22 normocalcemic patients
with non-Hodgkin’s lymphoma were hypercalciuric, and 18% had
increased serum 1,25-(OH)2D3 concentrations.
These findings led the investigators to conclude that dysregulated
1,25-(OH)2D3 production is common in patients
with diffuse large cell lymphoma (201).
Thus, the mechanisms responsible for hypercalcemia in this setting
appear to be multifactorial and include increased intestinal calcium
absorption as well as increased osteoclastic bone resorption.
Additionally, many of the reported patients had altered renal function,
a finding that suggests that impaired renal calcium clearance may also
be contributing to the hypercalcemia in certain patients. The low serum
PTH and urinary cAMP concentrations indicate that neither PTH nor PTHrP
mediates the hypercalcemia in this setting. Prostaglandins, when
measured, have been low, and selected patients had no calcium-lowering
effect from indomethacin therapy (199). It is likely that the lymphoma
tissue itself hydroxylates 25-hydroxyvitamin D to the active
1,25-(OH)2D3 similar to the situation in
hypercalcemia associated with granulomatous disease (52, 203). One
-hydroxylase activity has been demonstrated in human T cell
lymphotrophic virus type I-transformed lymphocytes (206). None of the
reported patients with 1,25-(OH)2D3-mediated
hypercalcemia had concomitant granulomatous disease, and hypercalcemia
often improved with medical or surgical therapy that resulted in a
decrease in serum 1,25-(OH)2D3 concentrations.
Recurrence of hypercalcemia and increased plasma
1,25-(OH)2D3 concentrations has been documented
with recurrence of disease (207).
c. PTH.
After Fuller Albright’s observations in 1941, it was
postulated that malignancy-associated hypercalcemia was due to tumor
production of PTH. This notion was strengthened by the early PTH RIA
data in hypercalcemic patients with malignancy, which suggested that
tumors produced factors recognized in these PTH RIAs (208, 209, 210, 211).
Although, for many years, malignancy-associated hypercalcemia was
attributed to ectopic tumor-produced PTH, it is now clear that PTHrP is
responsible for most cases. Analysis of 13 human and three animal
nonparathyroid tumors of diverse origin associated with hypercalcemia
did not detect PTH RNA transcripts (212). Since that time, four cases
of authentic tumor-produced PTH have been convincingly demonstrated in
a small cell carcinoma of the lung (78), an ovarian cancer (77), a
widely metastatic primitive neuroectodermal tumor (76), and a thymoma
(79). Molecular analysis of the ovarian carcinoma revealed both DNA
amplification and rearrangement in the upstream regulatory region of
the PTH gene (77). Interestingly, the primitive neuroectodermal tumor
produced both PTH and PTHrP that resulted in severe hypercalcemia (76).
These reported patients did not have coexisting primary
hyperparathyroidism since the parathyroid glands were normal at the
time of neck exploration or at autopsy in all cases. However, the fact
remains that ectopic production of PTH is a rare event, and it is
clearly documented that most patients with malignancy-associated
hypercalcemia have suppressed plasma PTH concentrations (69). It should
be emphasized that the most likely cause of hypercalcemia in the
setting of malignancy that is associated with a normal or increased
serum PTH concentration is coexisting primary hyperparathyroidism.
d. Other tumor-associated factors.
There is accumulating
evidence that solid tumors may produce other factors, alone or in
combination with PTHrP, that have the capacity to stimulate
osteoclastic bone resorption and cause hypercalcemia (213). These
factors include IL-1, IL-6, TGF, and tumor necrosis factor (TNF).
Administration of IL-1 injections to mice caused mild hypercalcemia
(214, 215), and this IL-1-induced hypercalcemia has been effectively
blocked by the IL-1 receptor antagonist (216). Mice bearing CHO tumors
transfected with the cDNA for IL-6 developed mild hypercalcemia (217)
as did mice bearing a renal carcinoma that cosecreted IL-6 and PTHrP
(218). Human TGF and TNF have been demonstrated to stimulate
osteoclastic bone resorption in vitro and cause
hypercalcemia in vivo (219, 220, 221, 222, 223, 224). TNF also caused
hypercalciuria, without an increase in nephrogenous cAMP, and increased
osteoclastic bone resorption in vivo in a mouse model
(225). In addition, as noted in the previous section, some of these
factors have been shown to modulate the end-organ effects of PTHrP on
bone and kidney. In some instances, factors such as TGF, IL-1, IL-6,
and TNF enhance the hypercalcemic effects of PTHrP. The ability of IL-6
to enhance PTHrP-mediated hypercalcemia appears to be due to increased
production of the early osteoclast precursor, granulocyte macrophage
colony forming units, by IL-6 in combination with increased production
of the more committed osteoclast precursors stimulated by PTHrP (165).
Figure 3 summarizes the known effect of
various tumor-produced factors on stages of the osteoclast lineage as
determined in bone marrow cultures from mice treated with respective
factors (164, 165, 171, 226).
|
3. Treatment of hypercalcemia associated with malignancy. Treatment of hypercalcemia due to malignancy should always involve treating the underlying tumor. Unfortunately, since this is often not effective or cannot be accomplished with the rapidity needed when the patient is faced with life-threatening hypercalcemia, therapy should also be directed against the mechanisms responsible for the hypercalcemia. In essentially all patients with hypercalcemia of malignancy, there is an increase in osteoclastic bone resorption, and in many there is also an increase in renal tubular calcium reabsorption, even in malignancies that are not associated with PTHrP production (230). Medical therapy is therefore aimed at inhibiting bone resorption and promoting renal calcium excretion. Because hypercalcemia associated with cancer is often accompanied by dehydration, volume expansion with isotonic saline is essential. This serves to increase the glomerular filtration rate and reduces the fractional reabsorption of both sodium and calcium. Since hydration alone will normalize serum calcium concentrations only transiently (231), inhibitors of bone resorption such as the bisphosphonates or calcitonin should be administered as well. When possible, mechanism-specific treatment should be attempted (232). Glucocorticoids, for example, are more effective in reducing the serum calcium concentration in hematological malignancies and 1,25-(OH)2D3-mediated hyper-calcemia than in solid tumors. Dietary calcium restriction is ineffective in reducing serum calcium concentrations except in cases of vitamin D-mediated hypercalcemia. In these cases, dietary calcium should be restricted to 400 mg daily until the underlying disorder is corrected. It is not desirable or advantageous to reduce calcium intake in hypercalcemia due to malignancy.
Bisphosphonates, analogs of pyrophosphate, have become the most useful antiresorptive agents among the currently available armamentarium for the treatment of hypercalcemia. They have a high affinity for hydroxyapatite in bone and concentrate in areas of high bone turnover. The mechanisms by which bisphosphonates inhibit bone resorption are not clearly understood, but potentially include induction of osteoclast apoptosis, inhibition of osteoclast formation and recruitment, or stimulation of osteoblasts to produce an inhibitor of osteoclast formation (233, 234). Another mechanism by which bisphosphonates might affect bone resorption is by decreasing the function of the osteoclast with respect to attachment and ruffled border formation (233). Recent in vitro findings using rodent marrow cultures suggest that tyrosine phosphatase activity is important in osteoclast formation and function and is a potential molecular target of bisphosphonate action (235). Bisphosphonates also inhibit axenic growth of amoebe of the slime mold Dictyostelium discoideum, and this property of growth inhibition paralleled the potency of inhibition of bone resorption (236). These findings indicate that bisphophonates may have a mechanism of action that is similar in both the osteoclast and Dictyostelium discoideum.
Bisphosphonates vary in potency but, in general, are poorly absorbed and are most effective in treating hypercalcemia when given intravenously. Bisphosphonates are concentrated in bone and remain there until the bone is resorbed. Etidronate, the first available bisphosphonate in the United States, is the least potent. Intravenous etidronate, given in doses of 7.5 mg/kg iv over 3 consecutive days normalized calcium concentration in 30%–40% of patients (237, 238, 239). Oral etidronate is generally ineffective in treating hypercalcemia (229), and at sustained dosages of 25 mg/kg per day for more than 6 months, it can cause bone mineralization defects (234). Etidronate can also cause hyperphosphatemia which, in addition to hypercalcemia, may lead to a high calcium-phosphate solubility product.
Pamidronate is a potent aminobisphosphonate available for the treatment of hypercalcemia of malignancy. The drug combines high efficacy with low toxicity profile and thus has become the current bisphosphonate of choice for the treatment of hypercalcemia of malignancy. It is highly effective in normalizing serum calcium concentrations and, when used in dosages recommended for hypercalcemia of malignancy, is not associated with bone mineralization defects. Pamidronate, administered as a single 24-h infusion, normalized serum calcium concentrations in 30% of patients who received 30 mg, 61% of patients who received 60 mg, and 100% of patients who received 90 mg (66). Successful therapy with bisphosphonates is associated with an increase in the plasma PTH and 1,25-(OH)2D3 concentrations as well as a decrease in the biochemical markers of bone resorption (66, 153, 190). Clinical studies of pamidronate treatment in patients with hypercalcemia of malignancy indicate that the calcium-lowering response to bisphosphonates correlates positively with the presence of bone metastases (240, 241, 242) and correlates negatively with plasma PTHrP concentrations (65, 240, 241). Such a relationship has also been demonstrated with clodronate, an oral bisphosphonate (243). This is presumably due to the effects of PTHrP to increase renal tubular reabsorption of calcium, which are not blocked by bisphosphonates. Nonetheless, pamidronate compares favorably to other inhibitors of bone resorption such as plicamycin, calcitonin, and gallium nitrate and is well tolerated. Pamidro-nate should be delivered as an intravenous infusion over 4–24 h. Clinical studies using 90 mg infusion of pamidronate over 4 h indicate that the mean time to achieve normocalcemia is approximately 4 days while the mean duration of normocalcemia is 28 days (244). Similarly, intravenous pamidronate, 60 mg every 2 weeks, maintained normocalcemia in a majority of patients with malignancy-associated hypercalcemia (245). An effective method for achieving more rapid reduction of the serum calcium is to use the combination of calcitonin and pamidronate (246). Calcitonin acts rapidly to lower the serum calcium, although usually its effects are only transient. Although escape from calcitonin therapy may occur within 48 h, by that time pamidronate is beginning to exert its maximal effects. Calcitonin can be administered either intramuscularly or subcutaneously every 12 h in doses of 200–400 MRC units. Reported side effects of pamidronate include transient low-grade fever and asymptomatic mild hypocalcemia (66). Bone mineralization defects have been reported only in patients receiving high-dose pamidronate at weekly intervals for the treatment of Paget’s disease (247). It is possible that other bisphosphonates, such as alendronate, risedronate, and tiludronate, will be effective oral therapy for hypercalcemia of malignancy. Due to its propensity to cause mouth ulcers, oral pamidronate, although effective, is not likely to be approved for such use in the United States.
Calcitonin inhibits osteoclastic bone resorption and renal tubular reabsorption of calcium. The main advantages of calcitonin are its rapid onset of action and its relative lack of serious side effects. Unfortunately, calcitonin alone only transiently normalizes the calcium concentration in patients with hypercalcemia of malignancy. Tachyphylaxis, probably due to down-regulation of calcitonin receptors, frequently develops with calcitonin administration, although this can be delayed with concomitant glucocorticoid treatment (248). However, calcitonin use can be particularly effective in the setting of severe hypercalcemia while waiting for the more sustained hypocalcemic effect of administered bisphosphonates to occur. Calcitonin use with bisphosphonates lowers calcium concentrations more quickly and effectively than either alone (249). Although human calcitonin is available, salmon calcitonin is generally used. If salmon calcitonin is used, a test dose of 1 MRC unit should be administered first, since rare anaphylactic reactions have been reported (250).
Plicamycin, or mithramycin, an antineoplastic agent used in the treatment of certain embryonal cancers (251), is also a potent inhibitor of bone resorption. Plicamycin inhibits DNA-dependent RNA synthesis (252) in tumor cells by binding to the promoter regions on DNA, thus preventing transcription (253). Presumably, osteoclastic bone resorption is inhibited by this mechanism as well. The dosage used to treat hypercalcemia (25 µg/kg) is one-tenth of the usual chemotherapeutic dose and should be infused over 4 h. Although plicamycin is almost invariably effective in lowering serum calcium concentrations, its considerable toxicity has limited its use in more recent years as more potent bisphosphonates have become available. Plicamycin has serious hepato- and nephrotoxicity in addition to local irritation and thrombocytopenic effects, which can limit its use in cancer patients as well as in those with renal impairment.
Gallium nitrate is another antineoplastic agent that, like plicamycin, was found to induce hypocalcemia in normocalcemic cancer patients receiving it (254). It inhibits osteoclastic bone resorption and appears to be more effective in lowering serum calcium concentration than calcitonin (255) and etidronate. Gallium is administered as a continuous infusion over 5 days, making it somewhat less convenient than some other antihypercalcemic agents. Gallium is excreted unchanged by the kidneys and has significant nephrotoxicity. Thus, it should not be administered to patients with renal impairment or to those receiving other nephrotoxic drugs (256).
About 30% of patients treated with glucocorticoids for hypercalcemia associated with nonparathyroid malignancy respond with a fall in calcium concentration (229). However, the response is often not complete and the responsiveness to glucocorticoids is unpredictable (229). Glucocorticoids are most effective in hypercalcemic patients with hematological malignancies, multiple myeloma in particular, as well as in vitamin D-associated disorders such as lymphomas. In hematological malignancies, glucocorticoids inhibit osteoclastic bone resorption by decreasing tumor production of locally active cytokines in addition to having direct tumorolytic effects (257). Glucocorticoids in dosage equivalents of prednisone, 40 to 60 mg daily, should be given for 10 days. If the calcium has not decreased in this period, glucocorticoids should be discontinued. Long-term adverse effects of glucocorticoids, such as osteopenia and Cushing’s syndrome, occur with continued administration over several months. This is usually not a consideration in patients with widespread malignancy, who have a very limited prognosis.
B. Oncogenic osteomalacia
Oncogenic osteomalacia is a rare tumor-associated disorder, first
recognized in 1947 (258, 259), that is characterized by
hypophosphatemia, phosphaturia, normocalcemia, and osteomalacia in the
absence of a family history of rickets, heavy metal poisoning, or
Fanconi’s syndrome. The serum alkaline phosphatase concentration is
increased and the serum 1,25-(OH)2D3
concentration is decreased. Affected patients typically present with
bone pain, proximal muscle weakness, and fractures. The disorder may
manifest as rickets if it occurs before fusion of the growth plate.
Tumors associated with this disorder are generally of mesenchymal
origin, small, and benign, although it has occasionally been associated
with malignant tumors. Reported tumor types include sclerosing
hemangioma (260, 261), paraganglioma (262), prostate cancer (263, 264),
oat cell carcinoma of the lung (265), fibrous dysplasia,
hemangiopericytoma, osteosarcoma, chondroblastoma, chondromyxoid
fibroma, malignant fibrous histiocytoma, giant cell tumor (266), and a
metaphyseal fibrous defect (267). Oncogenic osteomalacia associated
with metastatic prostate cancer comprises about 10% of all reported
cases (264). Regardless of the origin, tumors causing oncogenic
osteomalacia are often small and difficult to locate. Some reported
locations include the groin, nasopharynx, and the popliteal region.
Biochemical abnormalities resolve after complete tumor resection and
recur with tumor regrowth. Serum phosphate concentrations increase
immediately in the postoperative period while alkaline phosphatase
concentrations may take more than 1 yr to normalize with healing of the
osteomalacia (260, 262). In one patient, the bone mineral density
measurement increased from a preoperative value of 0.627
g/cm2 to 1.097 g/cm2 198 days postoperative
(262).
Phosphaturia may contribute to the osteomalacia in affected patients as does the apparent decrease in plasma 1,25-(OH)2D3 concentrations, both of which presumably lower the concentrations of available phosphate ions at the mineralizing bone site. The responsible phosphaturic factor has not been identified, but does not appear to be PTH or PTHrP since calcium and nephrogenous cAMP concentrations are normal in affected patients. Conditioned medium from cell culture of oncogenic osteomalacia tumors has been shown to inhibit phosphate uptake in cultured epithelial opossum kidney cells (260, 262). In one report, there was no measurable immunoactive PTH or PTHrP in the conditioned media from a paraganglioma (262). Another report found that conditioned media from a hemangioma inhibited phosphate uptake in opossum kidney cells without increasing cellular concentrations of cAMP. The media contained PTH-like immunoreactivity without PTHrP immunoreactivity, and the inhibition of phosphate transport was not blocked by a PTH antagonist (260). This putative factor appeared to be heat sensitive and of a molecular mass between 8 and 25 kDa (260).
Serum 1,25-(OH)2D3 concentrations are low in
patients with oncogenic osteomalacia, despite the presence of
hypophosphatemia, which normally increases
1,25-(OH)2D3 production by stimulating renal
1-hydroxylase activity independent of PTH. Additionally, most
reported cases of oncogenic osteomalacia have normal serum
25-hydroxyvitamin D concentrations. Deficient production of
1,25-(OH)2D3 could be a contributing factor to
the pathogenesis of oncogenic osteomalacia in these patients as the
clinical and biochemical abnormalities improve during calcitriol
therapy in some patients. The pathophysiology of the vitamin D
derangement is not well understood, but the clinical features have led
investigators to hypothesize that tumor-produced factors inhibit
1-hydroxylase activity. In one study, tumor extracts from a
hemangiopericytoma inhibited the formation of
1,25-(OH)2D3, and transplantation of this tumor
into athymic mice resulted in renal phosphate wasting and decreased
1,25-(OH)2D3 concentrations (268).
C. Tumor lysis syndrome
Another disorder of calcium homeostasis associated with malignancy
is the tumor lysis syndrome, which may occur as a consequence of
successful therapy of neoplastic disease (269). The syndrome often
occurs during therapy of hematological malignancies, particularly
high-grade lymphomas, in which a large number of tumor cells are lysed
in a short period of time. It has also been reported during therapy of
small cell carcinoma of the lung (270), breast cancer, and
medulloblastoma as well as during immunotherapy for sarcoma (271) and
therapy with TNF and monoclonal antibody against GD3 ganglioside in
metastatic melanoma (272). Rare reports of spontaneous tumor lysis have
also been reported (273). The release of tumoral intracellular ions,
such as phosphate and potassium, into the extracellular fluid result in
hyperphosphatemia. Hypocalcemia, hyperuricema (due to uric acid release
from lysed cells), and renal failure are a consequence.
|
IV. Local Mechanisms by Which Tumors Affect the Skeleton |
|---|
|
TopAbstractI. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
A. Clinical manifestations
Both solid tumors and hematological malignancies frequently affect
the most vascular areas of the skeleton, specifically in the red bone
marrow of the axial skeleton, the proximal ends of the long bones,
ribs, and the vertebral bodies. The most common way in which cancer
affects the skeleton is directly through local tumor-mediated
stimulation of osteoclastic bone resorption. Such osteolytic bone
lesions are typical of breast and lung carcinoma as well as
hematological malignancies such as multiple myeloma. Although breast
cancer cells have been shown to resorb bone directly in
vitro (278), most evidence [scanning electron microscopic
examination of adjacent bone surfaces (184) and response to osteoclast
inhibitors] is consistent with the notion that factors secreted by
cancer cells can activate osteoclasts locally. This is illustrated in
Fig. 4, a photomicrograph of an
osteolytic lesion due to the human breast cancer cell line MDA-MB-231.
Thus, local and systemic effects of cancer on bone are mediated through
one common final pathway, the osteoclast. The resulting osteolytic bone
destruction can lead to pain, pathological fractures, nerve compression
syndromes, and hypercalcemia.
|
Hypercalcemia associated with metastatic bone disease or hematological malignancies has been referred to as local osteolytic hypercalcemia, as clinical studies have demonstrated that many patients with bone involvement and hypercalcemia did not have the increase in plasma PTHrP or nephrogeneous cAMP concentration observed in patients with humoral hypercalcemia of malignancy (69, 111). The mechanism of hypercalcemia in this situation was postulated to be local tumor production of factors that stimulate osteoclastic bone resorption, such as TNFß, IL-6, and IL-1 as in the case of myeloma. However, it is now clear that humoral mediators of hypercalcemia, such as PTHrP, may mediate local osteolysis, even in the absence of hypercalcemia and increased plasma PTHrP concentrations (282). Additionally, if tumor burden in bone is great enough, tumor-produced factors in bone may be produced in enough quantity locally to reach the systemic circulation and have effects on sites distant from affected bone.
B. Pathophysiology of the metastatic process to bone
Breast and prostate cancer are the most common malignancies in
which bone metastases occur. Breast cancer is most often associated
with osteolytic metastasis while osteoblastic metastases are more often
manifest in prostate cancer. Mixed osteolytic and osteoblastic lesions
are often evident in both breast and prostate cancer. The remainder of
the review will focus on general principles of metastasis to bone,
followed by mechanisms specific to osteolytic and osteoblastic
metastasis, citing examples from current research in breast and
prostate cancer, respectively. This will be followed by a review of
bone involvement in myeloma. The reader should understand that breast
and prostate cancer, although predominantly lytic and blastic,
respectively, often have components of both osteolysis and
osteosclerosis. In this review, delineation of osteolytic and
osteoblastic mechanisms of bone metastasis to breast and prostate
cancer, respectively, is by no means meant to be exclusive. It is
likely that both mechanisms are often operative in the same patient.
1. Anatomical. Tumor metastasis to bone is not a random event, but rather a result of anatomical factors, tumor cell phenotype, and suitability of the metastatic site for tumor growth. Blood flow from the primary site is a significant determinant of the site of metastasis. Studies by Batson (283) describe in detail a low-pressure, high-volume system of valveless vertebral veins that communicate between the spine and intercostal veins independently of the pulmonary, caval, or portal systems. Batson accessed this plexus in cadavers via dye injection into the dorsal vein of the penis, an integral part of the prostatic venous plexus. Through these extensive injection studies of the prostatic plexus and venules of the breast in male and female cadavers, as well as in animals, Batson described this vertebral vein system as 1) consisting of the epidural veins, the perivertebral veins, the veins of the thoraco-abdominal wall, the veins of the head and neck, and the veins of the walls of blood vessels of the extremities; 2) valveless vessels that carry blood under low pressure; 3) subject to arrest and reversal of blood flow; 4) parallels, connects with, and provides bypasses for the portal, pulmonary, and caval systems. This plexus may serve as a major channel by which certain malignancies, such as prostate and breast cancer, metastasize to bone.
This concept that the vertebral system of veins acts as a direct conduit in the spread of prostatic carcinoma to the skeletal system was refuted by Dodds et al. (284), who analyzed 99mtechnetium bone scans in patients with skeletal metastases from assorted primary tumors. They found that the distribution of metastases was virtually identical in patients with prostatic and nonprostatic tumors. Of the patients with prostatic carcinoma, 25% had bone scan lesions exclusively outside the region of the sacrum, pelvis, and lumbar spine. The distribution of skeletal metastases from prostatic carcinoma did not support the concept that the vertebral veins have a substantial role in the dissemination of this tumor.
2. Seed and soil. Regardless of whether or not blood flow or
anatomic considerations are important determinants of the site of
metastasis, they are not the only ones. The distribution of metastases
to various organs are not predicted by anatomic considerations alone in
approximately 40% of tumors (285). Thus, other determinants of the
site of metastasis, such as properties of both the tumor cell and the
metastatic site, are important. Metastasis is an extremely complex
event that involves a cascade of linked sequential events that must be
completed before a tumor cell successfully establishes a secondary
tumor in bone (Fig. 5). Specifically, a
tumor cell must 1) detach from the primary site; 2) enter tumor
vasculature to reach the circulation; 3) survive host immune response
and physical forces in the circulation; 4) arrest in distant capillary
bed; 5) escape the capillary bed; and 6) proliferate in the metastatic
site. The events involved in entering the tumor vasculature are similar
to those involved with exiting the vasculature in the bone marrow
cavity. These include 1) attachment of tumor cells to the basement
membrane; 2) tumor cell secretion of proteolytic enzymes to disrupt the
basement membrane; and 3) migration of tumor cells through the basement
membrane. Attachment of tumor cells to basement membranes and to other
cells are mediated through cell adhesion molecules such as laminin and
E-cadherin. Tumor cell secretion of substances such as
metalloproteinases facilitate disruption of the basement membranes and
enhance invasion. Inherent tumor cell motility or motility in response
to chemotactic stimuli are also important factors for tumor cell
invasion to the secondary site.
|
Why is breast cancer one of the limited primary tumors to display osteotropism? Paget, during his observations of breast cancer in 1889, proposed the "seed and soil" hypothesis to explain this phenomenon. "When a plant goes to seed, its seeds are carried in all directions; but they can only grow if they fall on congenial soil" (1). In essence, the microenvironment of the organ to which the cancer cells metastasize may serve as a fertile soil on which the cancer cells (or seeds) may grow. Although this concept was proposed over a century ago, it remains a basic principle in the field of cancer metastasis at the present time. Thus, breast cancer cells possess certain properties that enable them to grow in bone, and the bone microenvironment provides a fertile soil on which to grow.
C. Local tumor syndromes in bone
Understanding the pathophysiology of bone metastasis has been a
slow process scientifically, as very few useful animal models of
spontaneous bone metastasis exist. Thus, various techniques of
experimental bone metastasis have been used throughout the years and
include injection of tumor cells directly into 1) the intramedullary
cavity (286); 2) abdominal aorta (287); 3) tail vein with inferior vena
cava occlusion (288); 4) left upper thigh muscle (289); 5) left
thoracic artery with renal artery occlusion; and 6) left cardiac
ventricle (290, 291). A complete review of the advantages and
disadvantages of these models has been extensively discussed by Orr
et al. (292).
1. Osteolytic metastases. Cancer metastatic to bone often causes bone destruction or osteolysis. Although several tumor types, such as prostate, lung, renal cell, and thyroid, are associated with osteolytic lesions, breast cancer is the most common. A comprehensive review of more than 500 patients dying of breast cancer revealed that 69% had bone metastasis, and bone was the most common site of first distant relapse (182). In those patients with disease confined to the skeleton, the median survival was 24 months compared with 3 months in those patients whose first relapse occurred in the liver. For these reasons, the following discussion on the pathophysiology of cancer-mediated osteolysis will focus on breast cancer.
a. Breast cancer cells as the seed.
A variety of common
characteristics are necessary for tumor cells to possess the metastatic
phenotype. Such properties include 1) the production of proteolytic
enzymes necessary for detachment from the primary site, invasion into
surrounding soft tissues, intravasation, extravasation, and bone matrix
degradation; 2) expression or loss of cell adhesion molecules essential
for detachment from the primary site and arrest at a metastatic site;
3) migratory activity to travel in the circulation; 4) escape from the
host immune surveillance to survive; and 5) capacity to respond to a
chemoattractant. Although these properties are common to tumor cells
metastasizing to any organ, they are insufficient to explain the
propensity of breast cancer to metastasize to bone. Therefore, it is
likely that breast cancer cells have additional characteristics that
are specifically required for causing metastases in bone.
Since bone is mainly composed of a hard mineralized tissue, it is more resistant to destruction than other soft tissues. Thus, in order for cancer cells to grow in bone, they must possess the capacity to cause bone destruction. Histological review of breast cancer metastatic to bone reveals that tumor cells are adjacent to osteoclasts resorbing bones (184, 185, 186) and indicate that breast cancer cells possess the capacity to stimulate osteoclastic bone resorption. Breast cancer cells may either induce osteoclastic differentiation of hematopoietic stem cells, activate mature osteoclasts already present in bone, or do both, through releasing soluble mediators or via cell-to-cell contact. Clinical and experimental evidence indicates that tumor-produced PTHrP is a major candidate factor responsible for the osteoclastic bone resorption present at sites of breast cancer metastatic to bone (293, 294, 295). PTHrP has been detected by immunohistochemistry (293) and in situ hybridization (294) in 92% of breast cancer metastases in bone compared with only 17% of similar metastases to nonbone sites, an observation that prompted speculation that production of PTHrP as a bone-resorbing agent may contribute to the ability of breast cancers to grow as bone metastases. Bundred and colleagues (195) found positive immunohistochemical staining for PTHrP in 56% of 155 primary breast tumors from normocalcemic women, and PTHrP expression was positively correlated to the development of bone metastases and hypercalcemic episodes. PTHrP expression was detected by RT-PCR in 37 of 38 primary breast cancers, and subsequent development of bone metastases was associated with a higher PTHrP expression (295). Finally, PTHrP was detected by immunohistochemistry in 83% of patients who developed bone metastases compared with 38% in those who developed lung metastases and 38% in those without recurrence (196). There have been no consistent correlations between PTHrP expression in the primary breast tumor and standard prognostic factors, recurrence, or survival. The only significant and consistent correlations have been between PTHrP positivity and the development of bone metastases and hypercalcemia.
These clinical observations have been extended by using a mouse model
of bone metastases (290, 296) in which inoculation of a human breast
cancer cell line, MDA-MB-231 (297), into the left cardiac ventricle
reliably causes osteolytic metastases. MDA-MB-231 cells produce low
amounts of PTHrP in vitro and when the cells were
engineered to overexpress PTHrP, by transfection with the cDNA for
human prepro-PTHrP, an increase in the number of osteolytic metastases
was observed (178). In contrast, when mice were treated with monoclonal
antibodies directed against the 1–34 region of PTHrP, before
inoculation with parental MDA-MB-231 cells, the number and size of
observed osteolytic lesions were dramatically less than similar animals
treated with control (Fig. 6). Mice with
established osteolytic metastases due to MDA-MB-231, treated with the
antibody, had a decrease in the rate of progression of metastases when
compared with mice that received a control injection (282, 298).
Similar findings have been demonstrated in this model using a human
lung squamous cell carcinoma (299). Taken together, these data strongly
suggest that PTHrP expression by breast cancer cells is important for
the development and progression of breast cancer metastases in bone. It
stands to reason, then, that production of other osteoclast-stimulating
factors should potentiate the development of bone metastases as well.
|
Cancer cell expression of factors affecting motility are important in the general metastatic process (302) as well in those processes specific to bone metastasis. Once tumor cells arrive in the bone marrow sinusoids, they must possess the capacity to move through those sinusoids to the bone tissue. Autocrine motility factor (303, 304), Thymosin ß15, and possibly the small heat shock protein 27 (Hsp27) have emerged as potential factors controlling cell motility. Thymosin ß15 increases cell motility, and when its production was decreased by expression of antisense constructs, as recently reported by Bao et al. (305), metastases were prevented in the Dunning rat prostate adenocarcinoma model. Similarly, overexpression of Hsp27 in MDA-MB-231 cells decreased cell motility in vitro and bone metastasis in mice (306).
Another important property of the breast cancer seed that enables it to
establish growth in bone resides in the adhesion molecules.
Experimental evidence supports the notion that tumor cell surface
expression of such molecules mediates targeting to bone and the
resultant development of bone metastasis. For example, bone marrow
stromal cells express the vascular cell adhesion molecule-1 (VCAM-1), a
ligand for 4ß1 integrin (307). Tumor cells
expressing 4ß1 integrin may preferentially
adhere to bone marrow stromal cells to establish bone metastasis. CHO
cells transfected with 4ß1 caused bone and
lung metastases when inoculated intravenously into nude mice compared
with only lung metastases in mice similarly inoculated with
untransfected CHO cells (308). In that report, bone metastases were
inhibited by antibodies against 4ß1 or
VCAM-1. Similar expression of 3ß1,
6ß1, or vß1
did not induce bone metastases (308). Although many breast cancer cells
express the vß3 integrin receptor that
binds the bone matrix protein, osteopontin, a potential avenue for the
development of bone metastasis, MDA-MB-231 cell populations with
high-level expression of the vß3 integrin
were less likely to cause bone metastasis than those cells expressing
low amounts of vß3 in the mouse model of
bone metastasis (309). Bone sialoprotein peptides containing RGD
sequences have been shown to decrease MDA-MB-231 cell adhesion to
extracellular bone matrix in vitro (310). Finally, tumor
cell expression of CD44 may mediate binding to osteopontin via
RGD-independent mechanisms (311). Such observations illustrate the
complex and multifactorial nature of the mechanisms underlying the
metastatic process.
Taken together, tumor cell expression of osteolytic factors, adhesion molecules, and motility factors significantly impact the ability of the tumor cell, or seed, to develop and grow as bone metastases.
b. Bone microenvironment as the soil.
Bone is unique among
metastatic target tissues since it undergoes continual remodeling under
the influence of systemic hormones and local bone-derived growth
factors. Mineralized bone matrix is a repository for growth factors, of
which TGFß and IGF-II constitute the majority (3). As described
earlier, these growth factors are released from the bone matrix as a
result of normal osteoclastic bone resorption (5), a part of the normal
remodeling process necessary for maintenance of the structural
integrity of bone. The hematopoietic stem cells in the bone marrow can
differentiate into bone-resorbing osteoclasts. Other cells in the bone
marrow, stromal and immune cells in particular, produce cytokines and
growth factors that may potentiate tumor cell growth or expression of
osteolytic factors. Thus, once breast cancer cells arrest in bone, the
high concentrations of growth factors and cytokines in the bone
microenvironment provide a fertile soil on which the cells can grow.
Such host cytokines may also enhance osteoclastic bone resorption
stimulated by tumor-produced factors such as PTHrP. Furthermore, when
the tumor cells stimulate osteoclastic bone resorption, this bone
microenvironment is even more enriched with bone-derived growth factors
that enhance survival of the cancer. Finally, bone-derived TGFß may
have an important role as a chemoattractant for breast cancer cells.
A large body of indirect evidence to support the concept that bone is a fertile soil, further enriched by the process of osteoclastic bone resorption, has accumulated in studies using bisphosphonates in the treatment of bone metastases. It is already clear from clinical studies that the use of bisphosphonates, potent inhibitors of bone resorption, significantly reduces skeletal morbidity in advanced breast cancer (312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323). In a recent multicenter trial that consisted of more than 700 patients with stage IV breast cancer with two or more predominantly lytic lesions, with at least one lesion that was 1 cm or greater in diameter, treatment with pamidronate 90 mg iv every 3–4 weeks for 12 months in conjunction with chemotherapy or hormonal therapy resulted in a significant reduction in skeletal complications and bone pain compared with the control group (321). Bisphosphonates have also been shown to decrease the number of bone metastases as well as tumor burden in animal models (296, 324, 325). Thus, by decreasing osteoclastic bone resorption, the bone microenvironment is a less fertile soil for the growth of tumor.
TGFß, which is present in high concentrations in the bone microenvironment and expressed by some breast cancers (177, 178) and cancer-associated stromal cells (176), has been shown to enhance secretion of and stabilize the mRNA for PTHrP in a renal cell carcinoma (96), a squamous cell carcinoma (97, 98), and a human breast adenocarcinoma, MDA-MB-231 (177, 178). In fact, of the known growth factors present in the mineralized bone matrix other than TGFß, such as IGF-I and -II, fibroblast growth factors (FGFs) 1 and 2, bone morphogenetic proteins (BMPs) and platelet-derived growth factor, only TGFß has been shown to significantly stimulate PTHrP secretion from the human breast cancer cell line, MDA-MB-231 (178). The fact that TGFß is abundant in bone (3) and can enhance PTHrP expression by cancer cells makes it an important candidate factor in the establishment and progression of breast cancer metastases to bone. TGFß is a member of a large superfamily of proteins that are important regulators of bone cell activity (326). Multiple isoforms of TGFß exist in mammals and appear to control cell proliferation and differentiation in many human cell types (327). The prototype of these isoforms, TGFß1, is highly expressed by differentiated osteoblasts and osteoclasts, is stored in bone matrix, and is released in active form during osteoclastic bone resorption (5). The effects of TGFß include stimulation of cell proliferation of mesenchymal cells, growth inhibition of epithelial cells, synthesis of extracellular matrix proteins, and enhancement of cell adhesion. These effects of TGFß are mediated through complex receptor interactions (328). TGFß binds to the type II receptor, and this complex recruits and phosphorylates the type I receptor, which in turn initiates signal transduction mediated by the recently identified Smad protein family (328). The effects of TGFß on cancer cells are complex and variable (329). In some cancer cells, TGFß inhibits growth, while in others growth is stimulated (330). It likely has effects on apoptosis as well.
Further evidence for the role of bone-derived TGFß in the development
and progression of breast cancer metastasis to bone has been
demonstrated in the same in vivo animal model of
osteolysis described above. Since TGFß increases PTHrP expression by
MDA-MB-231 cells in vitro, this cell line was
transfected with a cDNA encoding a TGFß type II receptor lacking a
cytoplasmic domain (TßRII
cyt) (331). This receptor binds TGFß,
but since it cannot phosphorylate the type I receptor, signal
transduction is not initiated and it acts in a dominant-negative
fashion to block the biological effects of TGFß (332). Stable clones
expressing TßRIIcyt did not increase PTHrP secretion in response
to TGFß stimulation compared with controls of untransfected
MDA-MB-231 cells or those transfected with the empty vector. Mice
inoculated into the left cardiac ventricle with MDA-MB-231 cells
expressing TßRIIcyt had fewer osteolytic lesions as well as a
smaller area of osteolytic lesions by radiography and histomorphometry
compared with the controls of parental cells or those transfected with
the empty vector (333). These data indicate that TGFß responsiveness
of this human breast cancer cell line is important for the expression
of PTHrP in bone and the development of osteolytic bone metastasis
in vivo.
In the mouse model of bone metastasis in which human tumor cells
inoculated into the left cardiac ventricle cause bone metastasis,
metastasis to the calvariae rarely occur due to the relatively low rate
of bone turnover at this site compared with other bones. To increase
the rate of bone turnover in the calvariae, Sasaki et al.
(334) injected recombinant IL-1 subcutaneously over the calvariae of
nude mice for 3 days. Upon completion of these treatments, MDA-MB-231
cells were inoculated into the left cardiac ventricle of female nude
mice. Four weeks after tumor cell inoculation, IL-1-treated mice had
obvious metastatic tumor deposits in the calvariae compared with none
in the control-treated mice. Further experiments demonstrated that
pretreatment of the mice injected with IL-1 with the bisphosphonate,
risedronate, profoundly diminished the development of metastatic tumor
deposits in the calvariae (334). These data suggest that growth factors
released from bone matrix may potentiate tumor cell growth in bone
metastases. However, these findings do not exclude some other
alteration in the bone matrix or microenvironment that may be enhanced
due to increased bone turnover.
Growth factors released from resorbing bone likely have significant effects on tumor cell growth as well. Experimental evidence suggest that IGFs may be important in this regard. Culture supernatants from resorbing neonatal mouse calvariae strongly increased the proliferation of MDA-MB-231 breast cancer cells in culture (335). Inhibition of bone resorption by adding risedronate to the calvarial organ cultures blocked the subsequent breast cancer cell proliferation. Additionally, neutralizing antibodies to the IGF-I receptor markedly impaired the growth-stimulating effects of the resorbing bone culture supernatants on the tumor cells (335). These results strongly suggest that IGFs are released from bone during bone resorption and promote breast cancer cell proliferation.
Another property of bone that may explain the predilection of certain
tumor types to grow in bone is chemotactic attraction of circulating
cancer cells. Culture supernatants of resorbing bone stimulate
chemotactic movement of breast cancer cells in a Boyden chamber assay
(336). Bone matrix factors such as TGFß, type I collagen and its
fragments, osteocalcin, and IGFs have been shown to stimulate
chemotaxis of breast cancer cells (337). Most recently, IGF-I was shown
to stimulate vß5 integrin-mediated
chemotactic migration of human breast cancer cell lines (338).
c. Tumor cell-bone interactions.
In addition to the properties
of breast cancer as a seed and the bone microenvironment as a soil,
there are likely complex interactions between the bone microenvironment
and the tumor cell as well as between and within tumor cells that
influence osteoclast activation. Both bone-derived and tumor-associated
factors have been shown to increase PTHrP expression by tumor cells as
well as modulate the end-organ effects of PTHrP. Thus, such factors may
enhance the ability of tumor cells to activate osteoclasts and promote
bone destruction.
Other tumor-associated factors in addition to bone-derived growth
factors may be important regulators of PTHrP expression in breast
cancer metastatic to bone. As indicated in the previous section on
hypercalcemia, many tumor-associated factors, such as EGF (90, 91),
TGF (166), IL-6 (165), TNF, IGF-I, and IGF-II (91), have the
potential not only to enhance tumor production of PTHrP but to modulate
its end-organ effects on bone as well.
d. Implications of PTHrP status in breast cancer.
These
findings have important implications for breast cancer effects on the
skeleton. First, breast cancers expressing PTHrP may affect the
skeleton through humoral and osteolytic mechanisms. Second, the effects
of PTHrP on bone may be enhanced if the breast cancer expresses other
bone-active factors, such as TGF or IL-6, in addition to PTHrP.
Finally, growth of breast cancer cells in bone may be enhanced if the
tumor cells express PTHrP or other bone-resorbing factors. TGFß, as
well as other bone-derived growth factors, increase PTHrP expression by
breast cancer cells in bone, so that TGFß-responsive tumors may
preferentially grow in bone. Thus, enhanced osteoclastic bone
resorption causes increased release of TGFß and other growth factors
into the bone microenvironment. The result is 1) greater PTHrP
expression by the breast cancer cells; 2) enhanced growth of the cancer
cells; and 3) chemoattraction of more tumor cells by bone-derived
factors. A cycle is thus established, as illustrated in Fig. 7, which ends in bone destruction and the
other consequences of lytic bone metastases. The clinical finding of
increased PTHrP expression in bone compared with other sites supports
the notion that production of PTHrP as a bone-resorbing agent may
contribute to the ability of breast cancers to grow as bone metastases
and/or that the bone microenvironment enhances production of PTHrP.
These and other reasons may, in part, explain the propensity of breast
cancers to metastasize to bone and the alacrity with which breast
cancer grows in bone.
|
2. Osteoblastic metastasis. Osteoblastic metastases occur most commonly in prostate cancer and less so in breast cancer. Rarely, osteoblastic bone lesions have been described in other malignancies such as an osteosclerotic variant of myeloma (279), colon cancer (339), astrocytoma (340), glioblastoma multiforme (341), thymoma (342), carcinoid (343), nasopharyngeal carcinoma (344), leptomeningial gliomatosis (345), Zollinger-Ellison syndrome (346), and cervical carcinoma (347). Similar to the pathophysiology of breast cancer-mediated osteolysis, the seed and soil hypothesis applies to this situation as well in that tumor cells secrete factors that stimulate bone formation, and the bone microenvironment readily supports the growth of prostate cancer cells (348).
Prostate cancer is relatively unique in its ability to form osteoblastic bone metastases, and there is much speculation on the mechanisms involved. Understanding the pathophysiology of prostate cancer-mediated osteoblastic metastasis has been limited, in part, to the paucity of in vivo models that adequately reproduce the spectrum of human disease, including the osteoblastic phenotype. However, over the past few years several models have been reported that may provide insight into the mechanisms responsible for the osteoblastic metastasis. Thalmann et al. (349) reported that an androgen-independent clone of the LNCaP human prostate cancer cell line, when inoculated subcutaneously or orthotopically into castrated male nude mice, spontaneously metastasized to bone in 11–50% of mice. Histological evidence of new bone formation was observed at the site of bone metastasis. Greenberg et al. (350) developed a transgenic mouse model of spontaneous prostate cancer, in which the simian virus 40 (SV40) large tumor T antigen is driven by the rat probasin promoter to target the dorsolateral epithelium of the prostate (TRAMP mice). In this model, 100% of mice develop distinct pathology in the dorsolateral epithelium of the prostate by 10 weeks of age that range from mild intraepithelial hyperplasia to large multinodular malignant neoplasia (350). Distant metastases occur as early as 12 weeks in common sites of periaortic lymph nodes and lungs and less common sites of kidney, adrenal gland, and bone (351). As reported by Gingrich et al. (351), one of these TRAMP mice developed paraplegia and was found to have tumor metastatic to the spinal canal. Bone pathology revealed osteoclastic bone resorption and new bone formation at vertebral sites adjacent to the spinal metastasis. Finally, in a model similar to the one described previously for breast cancer metastasis to bone, described in a later section of this review, rat prostate cancer cells inoculated into the left cardiac ventricle of syngeneic rats caused osteoblastic bone metastasis (352). Further investigation of these animal models should provide significant insight into the pathophysiology of prostate cancer metastasis to bone.
a. Prostate cancer as the seed.
As prostate cancer is more
frequently associated with osteoblastic metastases, prostate cancer
cells must possess properties different from those of other tumor types
commonly associated with osteolytic metastases. The histomorphometric
studies of Charhon et al. (353) indicate that osteoblastic
metastases are likely due to soluble factors that are produced by
metastatic prostate cancer cells that stimulate bone formation.
Osteoblast-stimulating activity produced by prostate cancer has been
described by a number of investigators. Conditioned media from
Xenopus oocytes injected with total RNA from the human
prostate cancer cell line PC3 stimulated both mitogenesis and alkaline
phosphatase activity in osteosarcoma cells with the osteoblast
phenotype (354). In fetal rat calvarial cells, PC-3-conditioned media
stimulated osteoblast proliferation (355). Koutsilieris et
al. (356) found that extracts of prostate cancer tissue and normal
prostate tissue stimulated proliferation of bone cells.
b. Osteoblastic factors.
Such data indicate that prostate
cancer is a source of osteoblast-stimulating activity. The following
tumor products have been proposed to be important in the genesis of the
osteoblastic response to tumor cells in bone.
i. TGFß. TGFß is secreted by osteoblasts in a latent biologically inactive form that is incorporated into bone extracellular matrix. TGFß is synthesized as latent high molecular mass complexes, composed of TGFß, the amino-terminal portion of the TGFß precursor, and the latent TGFß-binding protein (LTBP). Osteoblasts not only produce TGFß but they also possess high-affinity receptors for it (357), providing the opportunity for autocrine stimulation of osteoblast replication. Latent TGFß can be activated by a number of agents including acid pH, or by proteases such as plasmin or cathepsin D (358, 359). TGFß1 and -2 are homologous disulfide-linked homodimers of 25 kDa that have powerful effects on bone.
The local function of TGFß may be very important in contributing to
the differentiated activity of osteoblasts. It stimulates collagen
synthesis and regulates gene expression of mRNA for pro-I (I)
collagen, osteonectin, alkaline phosphatase, fibronectin (360),
osteopontin (361), and osteocalcin (362). TGFß increases the
abundance of matrix proteins by stimulating their synthesis and
inhibiting their degradation. It is a potent stimulator of collagen and
fibronectin synthesis and secretion in fibroblasts and osteoblasts
(363, 364), acting by increasing the mRNA for collagen and fibronectin
(365). TGFß also inhibits degradation of matrix proteins by
decreasing the synthesis of matrix-degrading enzymes, as well as
increasing the synthesis of protease inhibitors (366). TGFß promotes
the differentiation of cells of the osteoblast lineage toward the
mature osteoblast and the formation of new bone.
TGFß, injected subcutaneously adjacent to bone surfaces, causes a profound increase in new bone formation (367, 368, 369). When TGFß is administered by injection over the calvariae of mice daily for 3 days, bone width is increased 40% over the next month (368). This is initially woven bone, but it is later replaced by lamellar bone. Similar effects are seen when TGFß is injected or infused directly into the marrow cavity of the femur.
TGFß may also affect osteoclastic bone resorption as indicated by recent studies using a transgenic mouse model in which active TGFß2 overexpression was targeted to osteoblasts through the use of an osteocalcin promoter (370). This osteoblast-specific overexpression of TGFß2 resulted in progressive bone loss associated with increases in osteoblastic matrix deposition and osteoclastic bone resorption.
TGFß, isoforms 1 and 2 in particular, are produced by prostate cancer. TGFß2 is produced in abundant amounts in the human prostate cancer cell line PC3 (371). Although most studies in primary prostate indicate that TGFß is produced by prostate cancer cells, one study demonstrated immunohistochemical localization of TGFß in peritumoral fibroblasts (372). TGFß expression in human prostate cancer tissue appears to be greater than in normal prostate or benign prostatic hypertrophy (373). Studies from histopathologically verified human prostate cancer indicate that TGFß is produced without associating with the LTBP whereas in normal and benign prostatic hyperplasia tissues, TGFß may be produced in a complex associated with LTBP (374). Other investigators have found that prostate cancer cells secrete TGFß1 in the latent form, and the prostate cancer cells themselves further activate approximately 50% to the bioactive form (373). Other studies have shown that overexpression of TGFß1 in the rat prostate cancer cell line MATLyLu was associated with enhanced growth, viability, and aggressiveness in vivo (375). Although there is little evidence in the literature to demonstrate a direct relationship between tumor-produced TGFß and the development of osteoblastic metastasis, the fact that TGFß is produced by human prostate cancer coupled with its profound effects on bone formation have obvious implications in the pathophysiology of prostate cancer-mediated osteoblastic metastasis.
ii. IGF-I and -II. The IGF system is a fairly complex one and consists of two ligands, IGF-I and IGF-II, two receptors, and six binding proteins. The topic of IGFs, their binding proteins, and their biological actions has been extensively reviewed by Jones and Clemmons (376). Most of the cellular effects of the IGFs are mediated by binding of the peptides to the IGF-I receptor. The affinity of the IGF-I receptor for IGF-II is 2- to 15-fold lower than for IGF-I. The IGF-II/cation-independent mannose 6-phosphate receptor binds IGF-II with a 500-fold greater affinity than IGF-I. IGFs mediate their biological effects via interaction with their respective receptors, and these receptor interactions are affected by the presence of IGF-binding proteins. Six IGF-binding proteins (IGFBPs) have been identified, and binding of these proteins to IGFs can enhance or inhibit the biological effects of IGFs. IGF-I and -II are weak bone cell mitogens, but have clear and potent stimulatory effects on the differentiated function of the osteoblast, as evidenced by an increase in osteocalcin and type I collagen synthesis in osteoblasts. As a result, IGFs increase bone matrix apposition rates and bone formation. IGFs also decrease collagen degradation and the expression of interstitial collagenase, functions that suggest a role in the preservation of bone matrix. IGFs enhance bone formation in vivo, and mice with null mutation of type I IGF receptor have delayed skeletal development and ossification (377). The anabolic properties of IGF-I and -II, their inhibitory actions on matrix degradation, and their abundance in bone tissue suggest that these factors play a central role in the maintenance of bone mass (378).
Regulation of IGF-I in bone is further complicated by the production of IGFBPs by osteoblasts, which express all six IGFBPs. Binding of IGF to one of these binding proteins can inhibit or potentiate the biological effect of IGF. Binding to IGFBP-1, for example, decreases the biological activity of IGF-I. Conversely, IGFBP-5 has been shown to increase bone formation and, thus, appears to enhance the effect of IGF-I. This system is further complicated by the observation that growth factors such as TGFß, platelet-derived growth factor, FGF, and BMP-2 inhibit synthesis of IGFBP-5 in bone cell cultures (379, 380) while IGF-I and retinoic acid increase it (381). Thus, it appears that the effect of IGFs on bone is anabolic, and that local regulation of IGFs in bone is highly complex.
IGFs are potent mitogens for the growth of human prostate cancer cells, and primary cultures of prostate epithelial cells have been demonstrated to express all aspects of a functional IGF system: IGFs, IGF receptors, and IGFBPs (382). Human seminal fluid contains IGF-I and -II, IGFBP-2 and -4, as well as IGFBP-3 fragments and IGFBP-3 protease activity (383). This IGFBP-3 protease activity in seminal fluid has been attributed to prostate-specific antigen (PSA) (384) while production of other proteases such as urokinase receptor and cathepsin D have been demonstrated in prostate cancer (385, 386). IGFBP-2 appears to be the main binding protein produced by prostate cancer cells and, accordingly, clinical studies have demonstrated serum concentrations of IGFBP-2 to be increased and IGFBP-3 to be decreased in patients with prostate cancer (387, 388). Furthermore, significant positive correlations between serum concentrations of IGFBP-2 and PSA as well as between IGFBP-2 and tumor stage have been observed in men with prostate cancer (387, 388). Although at least one of these studies included patients with bone metastases, neither report comments on whether there was a significant correlation between IGFBP-2, PSA, and the presence of bone metastases. Immunohistochemistry and in situ hybridization in prostate tissue containing benign epithelium, high-grade prostate intraepithelial neoplasia, and adenocarcinoma indicate that mRNA and immunostaining intensity for IGFBP-2 progressively increased from benign prostate tissue to malignant adenocarcinoma whereas the immunostaining intensity for IGFBP-3 was increased in prostate intraepithelial neoplasia compared with normal, but decreased in malignant, cells (389). These authors conclude that the decreased expression of IGFBP-3 in malignant prostate tissue may be due to pre- and/or posttranslational mechanisms, including proteolysis, and that these observations correlate with serum changes of IGFBPs described in men with prostate cancer.
There is accumulating evidence that prostate cancers produce a variety of proteases, such as PSA, urokinase type plasminogen activator, and cathepsin D, that may be responsible for dissociating IGF-I and IGF-II from respective binding proteins to result in enhanced effects on not only tumor growth, but also, in the case of prostate cancer metastatic to bone, mitogenic effects on osteoblasts. PAIII cell-conditioned media has been shown to contain a 35-kDa proteinase capable of digesting IGFBPs that may serve to increase the bioavailability of osteoblast-derived IGFs (390). In addition to these proteolytic effects to activate growth factors, these proteases may be mitogenic for tumor cells as well.
Based on the above observations of the presence of an intact IGF system (including IGFBP proteases) in prostate cancer, the mitogenic effect of IGFs on prostate cancer, as well as on osteoblasts, and the positive correlations between serum IGFBP-2 and PSA, it is conceivable that local production of IGFs by prostate cancer in bone may mediate the osteoblastic response so characteristic of prostate cancer metastatic to bone. Unfortunately, the data described above are associations at best, and a direct causal role has yet to be proven.
iii. Proteases.
1. PSA. PSA is a serine protease, single-chain glycoprotein that has trypsin-like and chymotrypsin-like enzymatic activity (391). As PSA was initially believed to be produced exclusively by prostate epithelial cells, it has been extensively used as a marker for prostate cancer (392). The three clinical diseases associated with an increased serum PSA concentration are prostate cancer, benign prostatic hypertrophy, and acute bacterial prostatitis (393). In patients with prostate cancer, the serum PSA concentration is a valuable biological marker for diagnosis, prognosis, and management. The pretreatment serum PSA concentration has been shown to be a significant predictor of disease outcome after radiation therapy for local and regional prostate cancer (394). Androgenic hormones increase the production of PSA via transcriptional regulation (395). Serum PSA concentrations have been shown to correlate significantly with the presence of bone metastases by radionuclide scanning (396). In a large clinical study of 521 men with newly diagnosed and untreated prostate cancer, only one of 306 patients with a serum PSA concentration of less than 20 ng/ml had a positive bone scan (397). Serum PSA concentration proved to be the best predictor of bone scan findings when compared with tumor grade, local clinical stage, acid phosphatase, and prostatic acid phosphatase (397, 398, 399). Thus, in a newly diagnosed patient with prostate cancer, a serum PSA concentration of less than 10 ng/ml, and no skeletal symptoms, a bone scan may not be necessary (396) although others recommend measurement of PSA in conjunction with bone-specific alkaline phosphatase (400). Immunoreactive PSA has recently been demonstrated in 27% of 174 primary breast cancers (401) even though it was once believed to be an exclusive product of prostate epithelium. Breast-derived PSA was identical to PSA derived from prostate (402), and PSA has been shown to be produced at the ovarian metastatic site of a breast cancer (403). Furthermore, in a larger study of breast tumor cytosols from women and men, a positive correlation between immunoreactive PSA and progesterone receptor was observed (404).
The function of PSA in prostate cancer is unclear, but its proteolytic activity may prove to be important in the genesis of osteoblastic response to prostate tumor in bone. PSA has been shown to proteolyze IGFBP-3 into at least seven fragments with molecular masses of 13 kDa to 26 kDa with at least five different proteolytic recognition sites in this binding protein for PSA (405). Three of the five proteolytic sites were consistent with a kallikrein-like enzymatic activity while two of the sites were consistent with a chymotryptic-like enzymatic activity. Furthermore, some of the IGFBP-3 fragments retained the ability to bind IGF (405). Additionally, PSA has been shown to stimulate osteoblast proliferation at concentrations of 2.5 ng/ml possibly through activation of latent TGFß (406). Thus, it is tempting to speculate that PSA-induced proteolytic cleavage of IGFBP-IGF complex results in locally active IGF at the site of prostate cancer metastatic to bone to stimulate the osteoblastic response. Furthermore, recent evidence demonstrates that PSA also cleaves PTHrP-(1–141) at the carboxyl-terminal phenylalanine 23 and inactivates the biological effects of PTHrP to stimulate cAMP production in an osteoblast cell line (407). This may have important implications for the predominantly osteoblastic phenotype observed in prostate cancer. The fact that breast cancers also express PSA is equally interesting. Metastatic breast cancer to bone is one of the few other carcinomas associated with osteoblastic metastases, albeit at a much lower frequency than observed with prostate cancer.
2. Urokinase type plasminogen activator (uPA). Urokinase-type plasminogen activator is a member of the serine protease family that also includes tissue-type plasminogen activator (tPA). These proteins are expressed in normal cells, and the major function of tPA is related to intravascular thrombolysis while uPA is involved in proteolysis during cell migration and tissue remodeling. Although both tPA and uPA have been identified in malignant tissue, uPA appears to have a more prominent role in malignancy by promoting tumor cell migration and invasion by activating plasminogen to plasmin which, in turn, cleaves extracellular matrix components of laminin, fibronectin, and collagen.
uPA has been isolated from several prostate cancer cell lines that promote new bone formation in vivo. The rat prostate PA III tumor line causes new bone formation when inoculated over the scapula of rats and athymic nude mice (408). Conditioned media from PA III cells stimulated proliferation of osteoblasts in vitro. uPa expression by human PC-3 prostate cancer cells is increased by EGF and trans-retinoic acid and decreased by dexamethasone (409). In an experiment to demonstrate the influence of uPA on the nature of prostate cancer metastasis, Achbarou et al. (352) used gene transfer techniques to overexpress uPA in the rat prostate cancer cell line, Mat LyLu, by 5-fold compared with the same cells expressing empty vector. A separate Mat LyLu cell line that expressed uPA mRNA in the antisense orientation had 3-fold reduction in uPA mRNA compared with the empty vector cells. The uPA-overexpressing, underexpressing, and parental cell lines were compared in a rat model of bone metastases in which tumor cells inoculated into the left cardiac ventricle of inbred male Copenhagen rats cause bone metastasis. Rats inoculated with the uPA-overexpressing cell line developed hind limb paralysis sooner than rats inoculated with empty vector Mat LyLu cells. Similarly, rats inoculated with the uPA antisense-expressing Mat LyLu cells developed hind limb paralysis later that rats inoculated with parental or uPA-overexpressing Mat LyLu (352). Histological assessment of the sites of tumor metastasis indicated that more metastatic tumor was present sooner in both skeletal and nonskeletal sites of the rats inoculated with the uPA-overexpressing Mat LyLu cell line, compared with those inoculated with the empty vector or antisense cell line. Furthermore, histological analysis of bone indicated that although both osteolytic and osteoblastic lesions were present in both control and experimental rats, the osteoblastic response was the predominant feature in rats bearing the uPA-overexpressing Mat LyLu cells.
iv. FGFs. Both acidic and basic FGFs, now known as FGF-1 and -2, respectively, are present in mineralized bone matrix and stimulate the replication of cells in the skeletal system, but do not increase the differentiated function of the osteoblast. Therefore, they may play an important role in bone repair where bone cell mitogenesis may be necessary (378). FGFs enhance TGFß expression in cells with the osteoblast phenotype and have powerful stimulatory effects on bone formation in vivo. When injected locally over the calvariae of mice, FGF causes a 50% increase in bone thickness. When administered to ovariectomized rats, FGF blocked the associated bone loss and also increased trabecular connectivity and bone microarchitecture (410).
Prostate cancer cells express large amounts of both FGF-1 and -2 (411, 412). Not only have various prostate cancer cell lines been demonstrated to produce FGF-1 and -2 (413, 414, 415) as well as FGF receptor, but other tumor-produced FGF-like polypeptides have been demonstrated as well (415). An extended amino-terminal form of FGF-2 was purified from a human amnion tumor by its ability to stimulate proliferation of the osteoblast cell line MG-63 (416). This tumor has been reported to cause bone formation in vivo when inoculated into nude mice. Other data suggest that FGF-2 inhibits osteoclast formation via stromal cells and osteoblasts (417). Although this evidence supports the notion that FGFs may mediate the predominantly osteoblastic phenotype of metastasis in patients with prostate cancer, like TGFß, there are presently no direct associations between tumor-produced FGFs and osteoblastic metastasis.
v. BMPs. BMPs are bone-derived polypeptides and, with the exception of BMP-1, are members of the extended TGFß superfamily. At least 15 members are currently recognized, and the list is growing. BMP-2 through BMP-8 share some TGFß-related gene sequences. BMPs are synthesized by bone cells locally and stimulate the formation of ectopic bone when injected intraperitoneally or subcutaneously into rodents (418). BMPs stimulate the replication and differentiation of normal cells of the osteoblast lineage and, in contrast to TGFß, enhance the expression of the differentiated osteoblastic phenotype (378, 419). BMP-1, -2, -3, -4, and -6 are temporally expressed in primary cultures of fetal rat calvarial cells (420). BMP-2, -4, and -7 have been shown to induce differentiation of primitive mesenchymal cells into bone when implanted into subcutaneous tissue (421). BMP-2 accelerates differentiation in primary cultures of fetal rat calvarial cells as demonstrated by an increase in expression of alkaline phosphatase and osteocalcin (421). BMP-3 decreases osteoclastic bone resorption and is chemotactic for monocytes. BMP-7 (osteogenic protein-1) suppresses cell proliferation and stimulates the expression of markers characteristic of the osteoblast phenotype in rat osteosarcoma cells but stimulates growth and differentiation in rat calvarial cultures (422). In vivo, human BMP-7 was capable of inducing new bone formation in the rat subcutaneous bone induction model (423). Recently, overexpression of BMP-4 in lymphocytes was described in association with the disabling ectopic osteogenesis of fibrodysplasia ossificans progressiva (424).
Normal and neoplastic prostate tissue express BMP-2, -3, -4, and -6 mRNA. The predominant form in normal human prostate tissue was shown to be BMP-4. While this pattern was observed in human prostate cancer cell lines, PC-3 and DU-145, PC-3 also expressed BMP-2 and -3 in large amounts. The rat prostate cancer PAIII expressed predominantly BMP-3 mRNA (425). PAIII is a cell line derived from a strain of rats, Lobund-Wistar, that has a 10% frequency of spontaneous prostate adenocarcinoma (426). PAIII stimulates new bone formation in this strain of rats, as well as in nude mice, when inoculated over the scapula. Rat BMP-3 was isolated from PAIII cells (427), and transfection of the PAIII tumor cells with a BMP-3 antisense construct somewhat reduced the osteoblastic response (428). Thus, biologically active BMPs expressed by prostate tumor in bone may contribute to the new bone formation at metastatic tumor sites in bone.
vi. Endothelin-1 (ET-1). ET-1 is the most recent factor
implicated in the genesis of osteoblastic metastases. It is a potent
vasoconstrictor and was originally purified from endothelial cells
(429). Prostatic epithelium produces ET-1, and high-affinity ET-1
receptors are present throughout the prostate gland (430). ET-1
concentrations in seminal fluid are 500 times greater than those in
plasma. ET-1 stimulates mitogenesis in osteoblasts, and osteoblasts
have high-affinity receptors for ET-1 (431, 432). Osteoclastic bone
resorption and osteoclast motility are decreased by ET-1 as well (433).
Moreover, mean plasma endothelin concentrations in men with advanced,
hormone-refractory prostate cancer with bone metastases were
significantly higher than those concentrations in men with
organ-confined prostate cancer or normal controls (417). However, these
endothelin measurements were not correlated to tumor burden in bone and
did not correlate with serum PSA concentrations. Human prostate cancer
cell lines, DU-145, LNCaP, PC3, PPC-1, and TSU, have been shown to
express ET-1 by RT-PCR. Finally, in vivo, ET-1 stimulated
BMP-induced bone formation as assessed by alkaline phosphatase activity
in a rat model of matrix-induced bone formation (434). Additionally,
IL-6, but not estrogen, tamoxifen, TGFß, TNF, 
-interferon, or
IL-1, stimulated ET-1 production from human breast cancer cells MCF-7
and ZR-75–1 (435). This is of interest since breast cancer is
occasionally associated with osteoblastic metastasis.
Thus, the mechanisms responsible for the predominantly osteoblastic
phenotype of prostate cancer metastatic to bone is complex and likely
is the result of multiple tumor-produced factors on normal bone
remodeling. Figure 8 is a schematic model
based on available data from the literature that identify potential
tumor-bone interactions.
|
2ß1 integrins (438). TGFß
has also been shown to stimulate cell motility of the MATLyLu, an
in vitro observation that suggests that bone-derived
TGFß may be an important chemotactic factor in prostate cancer (439).
EGF, secreted by MG-63 bone cells stimulated chemomigration of the
TSU-pr1 prostate cancer line in Boyden chambers (440). 3. Hematogenous. Multiple myeloma is a plasma cell malignancy that is almost invariably associated with destructive bone lesions, either in the form of diffuse osteopenia or localized osteolysis throughout the skeleton (441). Eighty percent of patients with myeloma first present with pain, which can be related to the bone disease, and bone complications are the most obvious clinical feature in most patients (441). The bone lesions of myeloma may be diffuse or localized and comprise three types. In the majority of patients (more than 95%), they are osteolytic. These lesions occur predominantly in those bones that are rich in red marrow, e.g., the axial skeleton, and are associated with increased osteoclast activity adjacent to sites of myeloma cell accumulation. This suggests that myeloma cells produce locally active soluble factors that stimulate the remaining osteoclasts to resorb bone (442). In some patients, the bone loss is more generalized and its appearance more closely resembles that of osteoporosis. In these patients the myeloma cells tend to be more diffusely spread throughout the axial skeleton. Some patients may have a combination of these two pictures, i.e., osteopenia of vertebral bodies but discrete osteolytic bone lesions of the skull. Myeloma bone disease is always an important differential diagnosis in the patient who presents with apparent osteoporosis. The bone disease of myeloma tends to be steadily progressive in most patients and can be used as one of the parameters to monitor the course of the disease. For example, obvious progression of the bone lesions or the appearance of new discrete lesions indicates that the disease is active. On the other hand, vertebral body collapse can occur in patients in remission because of the weakened state of the skeleton, and bone pain itself is not a reliable indicator of the state of the disease activity.
The type of osteolytic disease that occurs in myeloma may be quite different from the bone disease associated with other types of malignancy, such as carcinoma of the breast. In myeloma bone disease there is often no increase in bone formation or osteoblast activity. The reason for this complete uncoupling of bone formation and bone resorption is unknown. It is paralleled clinically by the absence of an increase in the markers of bone formation, which are frequently present in other types of osteolytic bone disease due to malignancy, such as serum osteocalcin and alkaline phosphatase (161). In addition, the radionuclide bone scan shows no evidence of increased isotope uptake at the site of bone lesions.
A small number of patients with myeloma bone disease present with an entirely different picture of diffuse osteosclerosis (279, 443). Osteosclerotic myeloma often occurs as a part of a syndrome of polyneuropathy and is associated with the cutaneous and endocrine features that comprise POEM’s syndrome (polyneuropathy, organomegaly, endocrinopathy, M-protein, and skin changes) (444).
Thus, the majority of patients with myeloma have a crippling form of
bone disease associated with intractable pain, susceptibility to
fracture upon trivial injury, and nerve compression syndromes (most
frequently spinal cord compression), associated with vertebral body
collapse. About 30% of patients develop hypercalcemia at some stage
during the course of the disease, usually in association with impaired
renal function (441). The pathophysiology of bone destruction that is
so characteristic of myeloma bone disease remains unclear. Cultures of
myeloma cells in vitro produce several osteoclast activating
factors such as IL-6, IL-1ß, and TNFß or lymphotoxin (445, 446, 447),
which have been implicated. Lymphotoxin (or TNFß) is a powerful
bone-resorbing cytokine that is produced by lymphoid cell lines
in vivo. It is expressed by many B cell lines in culture,
including cell lines derived from patients with myeloma (447). It is
capable of causing hypercalcemia in vivo and has effects on
bone resorption that are indistinguishable from those of TNF, namely
an increase in osteoclast formation and activity, associated with
impaired osteoblast differentiation and increased osteoblast precursor
proliferation (224). Although produced by many cell lines derived from
patients with myeloma, it has not been detected in freshly isolated
marrow cells from these patients.
The second cytokine that has been implicated in the bone lesions in myeloma is IL-1ß (445, 446). IL-1ß is also a powerful stimulator of osteoclastic bone resorption, which increases osteoclast formation and osteoblast proliferation and can cause hypercalcemia in vivo. Freshly isolated marrow cells from patients with myeloma, which contain myeloma cells plus other marrow cells, have been shown to contain IL-1ß in the conditioned medium harvested from these cells. Moreover, bone-resorbing activity present in these conditioned media can be neutralized by antibodies to IL-1ß.
Finally, IL-6 is a cytokine that has an important growth-regulatory role in many patients with myeloma (448). IL-6 concentrations in bone marrow and in plasma correlate with the stage of disease (449, 450, 451). It may be produced by the myeloma cells or other stromal cells in the marrow microenvironment. Endogenous IL-6 production is 2–30 times greater in patients with myeloma than in normals (450). Administration of a murine-human chimeric anti-IL-6 antibody to patients with multiple myeloma resistant to second-line chemotherapy suppressed this IL-6 production, but did not prevent infection-induced IL-6 production (450). Such results suggest that this IL-6 antibody inhibited a positive feedback IL-6-dependent loop.
IL-6 is not a powerful bone-resorbing factor in its own right, but is capable of enhancing the effects of other factors on bone resorption, presumably by increasing generation of precursors for cells in the osteoclast lineage (165). Myeloma cells occasionally produce PTHrP and since IL-6 may potentiate the osteoclastic bone resorption mediated by PTHrP, such cytokine interactions have important implications in the genesis of the bone destruction associated with myeloma.
Just as there may be a vicious cycle in the bone microenvironment
between the tumor cells and bone-derived growth factors in breast
cancer, there may be a similar relationship between the process of bone
resorption and myeloma cell behavior (Fig. 9). In this latter case, the responsible
mediator may be IL-6, which is the major growth factor for myeloma
cells. In myeloma, stimulation of bone resorption may lead to IL-6
generation (452), which in turn may be responsible for maintaining
aggressive growth of the malignant cells. Primary tumor cells from
myeloma patients induced IL-6 secretion by adherent cells from
long-term bone marrow cultures (451). Interestingly, this IL-6
production did not occur when tissue-culture inserts, which prevented
direct contact between the adherent cells from bone marrow and the
myeloma cells, were used in the wells. Such findings imply an important
role for cell-cell contact in mediating the bone marrow cell induction
of IL-6 by myeloma cells. Binding of myeloma cells to adherent cells
from long-term bone marrow cultures was partly inhibited by antibodies
against the adhesion molecules, very late antigen-4 (VLA-4), CD44, and
lymphocyte function-associated antigen 1 (LFA-1), as was IL-6
production (451). These data indicate that one source of IL-6 may be
the normal cells in the bone marrow and that adhesion molecules
expressed on myeloma cells may mediate the induction of IL-6.
|
At this point in time, it is not possible to say which is the most important cytokine involved in myeloma. The bone disease may be due to the combination and interaction of a number of cytokines and other molecules working in parallel, derived from both myeloma as well as from the normal host cells. Several animal models of myeloma bone disease have been developed recently that should provide significant insight into the pathophysiology of this disease (462, 463).
Until recently, the management of myeloma bone disease comprised treatment of the underlying malignancy, management of bone pain with analgesics, radiation therapy if the bone disease was localized, and cautious treatment of hypercalcemia when it occurred because it is likely accompanied by impaired glomerular filtration. Recently, this has changed with the availability of more potent bisphosphonates. The Food and Drug Administration approved the use of pamidronate for myeloma bone disease in 1995. Pamidronate has been shown to reduce skeletal events associated with myeloma, including the need for radiation therapy and pathological fractures. It also probably reduces the number of episodes of hypercalcemia. When given in doses of 90 mg by intravenous infusion over 4 h monthly for 9 months, it has been shown to be very effective in patients with advanced disease, and the number of skeletal events is almost halved in patients taking pamidronate. Moreover, an objective assessment of quality of life in treated patients has shown a beneficial effect of pamidronate (464). Similar studies in Europe have also suggested that potent bisphosphonates such as clodronate are effective in improving performance status, reducing bone pain, vertebral fractures, and progression of osteolytic bone lesions, as well as preventing hypercalcemic episodes. Animal studies using a mouse model of myeloma with the bisphosphonate, risedronate, reveal similar findings (465). Previous clinical studies with less potent bisphosphonates in patients with myeloma have been less successful as oral etidronate was ineffective (466) and oral clodronate inhibited the progression of osteolytic bone lesions but did not reduce bone pain or fracture rate (467). Clinical studies in patients with myeloma indicate that adjuvant treatment with low-dose gallium nitrate attenuates the rate of bone loss as well as the associated bone pain (468). There is no definitive evidence as yet for a beneficial effect of inhibition of bone resorption on tumor burden. Bisphosphonates seem to be most effective in patients with minimal disease, but are appropriate for almost all patients.
As with solid tumors that cause destructive bone lesions and hypercalcemia, hypercalcemia is also frequent in myeloma, occurring in about 30% of all patients sometime during the course of the disease. Hypercalcemia is always associated with bone destruction and often with impaired renal function. When hypercalcemia occurs in patients with myeloma, it is usually accompanied by impaired renal function, which limits the availability of specific therapies for treatment of the hypercalcemia. Under these circumstances, the most appropriate medical agent to use is a new generation bisphosphonate, although experience with these drugs is limited in patients with markedly impaired glomerular filtration. An alternative is the use of a combination of calcitonin and glucocorticoids, which has been shown to be very effective in reducing hypercalcemia in patients with myeloma (67, 248). However, since calcitonin must be given by injection, this combination is not as convenient as the new generation bisphosphonates.
In myeloma, bone destruction occurs as a consequence of osteoclastic bone resorption, as osteoclasts accumulate on bone-resorbing surfaces adjacent to myeloma cells. In an early clinical study of patients with myeloma, biopsy samples indicate that active osteoclastic bone resorption correlated with the presence of greater than 20% myeloma cells in the adjacent marrow cell population (442, 469). The striking feature in myeloma is that there is a marked increase in osteoclastic bone resorption, usually without manifestations of increased bone formation (469). These abnormalities in bone formation were confirmed by transiliac bone biopsies from 118 patients with myeloma in which quantitative bone histomorphometry demonstrated that osteoid seams were reduced in thickness and had a lowered calcification rate (470). This is in contrast to breast cancer, where although the bone lesions are mainly destructive, there is usually a slight increase in bone formation and an increase in serum alkaline phosphatase and radionuclide uptake at sites of osteolytic deposits associated with increased osteoblast activity. The mechanism of this uncoupling of bone formation from bone resorption is not known but is the subject of intense study.
D. Therapy of tumor in bone
Most patients with bone metastasis are normocalcemic. In a
majority of breast cancer patients with bone metastases, local
osteolysis occurs without hypercalcemia (268), increases in
nephrogenous cAMP (69), or increases in PTHrP (111). Osteolytic bone
lesions are most frequent in patients with carcinoma of the breast,
carcinoma of the lung, and myeloma, the same malignancies that are
associated with hypercalcemia. However, there are also other solid
tumors in which hypercalcemia is rare but osteolytic bone lesions are
relatively frequent. These include patients with carcinoma of the
thyroid. These patients suffer considerably because of their bone
lesions, which cause intractable pain, pathological fracture after
trivial injury, nerve compression syndrome such as spinal cord
compression, and propensity to develop hypercalcemia. Until recently,
therapy of tumor in bone was directed against tumor cells for breast
and prostate cancer as well as myeloma and other malignancies. This
usually involved chemo- or hormonal therapy, local field irradiation,
radionuclide therapy, or surgery (274, 471, 472). The advent of
bisphosphonates has changed this perspective somewhat in that it has
added therapy directed against the osteoclast to our current
armamentarium of anticancer drugs.
As metastatic bone disease is mediated by osteoclastic bone resorption, and factors that stimulate osteoclastic bone resorption, such as PTHrP, enhance bone destruction by tumor, it is logical to consider therapy with inhibitors of bone resorption to prevent the development of bone metastasis or to delay their progression. Other mechanisms by which bisphosphonates exert their effects to decrease bone metastases may involve tumor cell adhesion to bone. In vitro studies demonstrate that a number of bisphosphonates decrease attachment of MDA-MB-231 breast cancer cells to extracellular bone matrix. Of interest is the fact that the effect of these bisphosphonates to decrease tumor adhesion positively correlated to the antiresorptive potency (473).
Several prospective, double-blind placebo-controlled trials have been published documenting the efficacy of the bisphosphonate, pamidronate, in decreasing the skeletal complications associated with breast cancer (321) and myeloma (464). For patients with myeloma, the Food and Drug Administration (FDA) has recently approved pamidronate for use in patients with osteolytic lesions who are not hypercalcemic. This is based on a recent study (464), which shows that intravenous pamidronate given every 4 weeks for nine cycles in almost 400 patients with myeloma caused a significant reduction in skeletal complications (defined as pathological fracture, requirement for radiation to bone or surgery, or spinal cord compression), decreased the occurrence of new pathological fractures, and prevented development of hypercalcemia. In addition, this treatment alleviated bone pain and improved quality of life. There was a suggestion in these patients that there may have also been a beneficial effect on overall survival. Pamidronate is therefore now being widely used early in the course of myeloma since it is a relatively nontoxic drug and may have a beneficial effect not only on bone complications. There is no definitive evidence as yet for a beneficial effect on tumor burden or survival, and this will require careful controlled studies in more extended numbers of patients.
Clinical studies have been ongoing for 20 yr in normocalcemic patients with solid tumors and osteolytic bone metastases. All of the available evidence from these studies suggests that drugs that decrease bone resorption, such as the potent bisphosphonates, have a beneficial effect on skeletal complications, including pain and pathological fracture, prevention of hypercalcemia, and improved quality of life. In a recent multicenter trial which consisted of more than 700 patients with stage IV breast cancer with two or more predominantly lytic lesions, with at least one lesion that was 1 cm or greater in diameter, treatment with pamidronate 90 mg iv every 3 to 4 weeks for 12 months in conjunction with chemotherapy or hormonal therapy resulted in a significant reduction in skeletal complications and bone pain compared with the control group (321). However, there may be an added beneficial effect of the bisphosphonates that is even more important. In experimental studies in which human breast cancer cells are inoculated into the left ventricle of the nude mouse, Sasaki et al. (296) have shown that bisphosphonates such as risedronate and ibandronate not only prevent the development of skeletal complications and bone metastases, but they also reduce tumor burden in bone. This likely occurs because the bisphosphonates make bone a less favorable environment for the growth of tumor cells by reducing bone turnover and decreasing the supply of local bone-derived growth factors that also act as tumor growth factors in the bone microenvironment. It is apparent from clinical studies that the use of bisphosphonates reduces significant skeletal morbidity in advanced breast cancer (312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323). These data suggest that drugs that inhibit bone resorption may be useful adjuvant therapy in patients with malignant disease by preventing the growth of tumor cells in the skeleton.
Studies using bisphosphonates in the treatment of prostate cancer metastatic to bone are fewer and less impressive. Clodronate treatment of men with advanced prostate cancer not only resulted in decreased osteoclastic bone resorption, as assessed histomorphometrically, but also in osteomalacia. The authors explained the transient relief of bone pain in the clodronate group to this resultant osteomalacia (474). In a small study of breast and prostate cancer patients with osteosclerotic lesions, treatment with pamidronate resulted in a decrease in bone pain. This response was mostly predicted by a decrease in the urinary marker of bone resorption, deoxypyridinoline (475).
Despite the encouraging results presented in these and other studies, several important questions remain regarding the use of bisphosphonates for treatment of tumor in bone. 1) Will bisphosphonates be useful as adjuvant therapy in tumor types other than breast cancer and myeloma? 2) Will bisphosphonate treatment in cancer improve survival? 3) Will bisphosphonates be beneficial in prevention of bone metastasis if therapy is initiated before the development of bone metastasis in patients with limited disease? 4) Will bisphosphonate therapy in cancer prove to be cost effective? 5) Is there a role for bisphosphonate therapy in osteoblastic metastasis? Although animal studies suggest that the answers to these questions may already be obvious, only prospective trials in humans will provide us with the definitive answers.
|
Footnotes |
|---|
1 Supported by NIH Grants CA-69158 and AR-01899 (to T.A.G.) and
RR-01346, AR-07464, and CA-40035 (to G.R.M) as well as Grant
DAMD17–94-J-4213 from the Department of Defense, U.S. Army (to
T.A.G.). 
|
References |
|---|
|
TopAbstractI. IntroductionII. Normal Calcium and...III. Humoral Mechanisms by...IV. Local Mechanisms by...References |
|---|
and interleukin-4.
Endocrinology 135:351–358[Abstract]
enhance hypercalcemia in
vivo and bone resorption in vitro. J Clin
Endocrinol Metab 77:40–45[Abstract]
and parathyroid
hormone-related protein on phosphate transport in renal cells. Am
J Physiol 259:F929–F935
of phosphate transport in
osteoblast-like cells. J Bone Miner Res 6:1235–1241[Medline]
and 1ß
cause hypercalcemia in normal mice. Proc Natl Acad Sci USA 85:5235–5239 stimulates bone resorption and inhibits formation
in vitro. Proc Natl Acad Sci USA 83:2228–223225-dihydroxycholecalciferol deficiency: association with a giant
cell tumor of bone. J Clin Invest 60:1046–1053
4ß1 integrin/VCAM-1
inhibits osteoblastic differentiation. J Bone Miner Res 11[Suppl
1]:S117 (Abstract 90)
4ß1 into tumor cells. Am
J Pathol 148:55–61[Abstract]
vß3 integrin.
J Bone Miner Res 11[Suppl 1]:S98 (Abstract 15)
stimulates mouse
osteoblast-like cells (MC3T3–E1) to produce macrophage-colony
stimulating activity and prostaglandin E2. Biochem Biophys Res Commun 145:323–329[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  Y.-C. Chen, D. M. Sosnoski, U. H. Gandhi, L. J. Novinger, K. S. Prabhu, and A. M. Mastro Selenium modifies the osteoblast inflammatory stress response to bone metastatic breast cancer Carcinogenesis, November 1, 2009; 30(11): 1941 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Gilmore, R. M. Gonterman, K. Menon, G. Lorch, D. J. Riese II, A. Robling, and J. Foley Reconstitution of Amphiregulin-Epidermal Growth Factor Receptor Signaling in Lung Squamous Cell Carcinomas Activates PTHrP Gene Expression and Contributes to Cancer-Mediated Diseases of the Bone Mol. Cancer Res., October 1, 2009; 7(10): 1714 - 1728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Bedard, J.-J. Body, and M. J. Piccart-Gebhart Sowing the Soil for Cure? Results of the ABCSG-12 Trial Open a New Chapter in the Evolving Adjuvant Bisphosphonate Story in Early Breast Cancer J. Clin. Oncol., September 1, 2009; 27(25): 4043 - 4046. [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Idris, I. R. Greig, E. Bassonga-Landao, S. H. Ralston, and R. J. van 't Hof Identification of Novel Biphenyl Carboxylic Acid Derivatives as Novel Antiresorptive Agents that Do Not Impair Parathyroid Hormone-Induced Bone Formation Endocrinology, January 1, 2009; 150(1): 5 - 13. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lorenzo, M. Horowitz, and Y. Choi Osteoimmunology: Interactions of the Bone and Immune System Endocr. Rev., June 1, 2008; 29(4): 403 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. R. Mundy and J. R. Edwards PTH-Related Peptide (PTHrP) in Hypercalcemia J. Am. Soc. Nephrol., April 1, 2008; 19(4): 672 - 675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Langley and I. J. Fidler Tumor Cell-Organ Microenvironment Interactions in the Pathogenesis of Cancer Metastasis Endocr. Rev., May 1, 2007; 28(3): 297 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Van Sant, G. Wang, M. G. Anderson, O. J. Trask, R. Lesniewski, and D. Semizarov Endothelin signaling in osteoblasts: global genome view and implication of the calcineurin/NFAT pathway Mol. Cancer Ther., January 1, 2007; 6(1): 253 - 261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Marguiles, V. S. Klimberg, S. Bhattacharrya, D. Gaddy, and L. J. Suva Genomics and Proteomics of Bone Cancer. Clin. Cancer Res., October 15, 2006; 12(20): 6217s - 6221s. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. S. Bruland, S. Nilsson, D. R. Fisher, and R. H. Larsen High-Linear Energy Transfer Irradiation Targeted to Skeletal Metastases by the {alpha}-Emitter 223Ra: Adjuvant or Alternative to Conventional Modalities? Clin. Cancer Res., October 15, 2006; 12(20): 6250s - 6257s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Leeming, G. Delling, M. Koizumi, K. Henriksen, M. A. Karsdal, B. Li, P. Qvist, L. B. Tanko, and I. Byrjalsen Alpha CTX as a Biomarker of Skeletal Invasion of Breast Cancer: Immunolocalization and the Load Dependency of Urinary Excretion. Cancer Epidemiol. Biomarkers Prev., July 1, 2006; 15(7): 1392 - 1395. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Normanno and W. J Gullick Epidermal growth factor receptor tyrosine kinase inhibitors and bone metastases: different mechanisms of action for a novel therapeutic application? Endocr. Relat. Cancer, March 1, 2006; 13(1): 3 - 6. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Angelucci, G. L. Gravina, N. Rucci, D. Millimaggi, C. Festuccia, P. Muzi, A. Teti, C. Vicentini, and M. Bologna Suppression of EGF-R signaling reduces the incidence of prostate cancer metastasis in nude mice. Endocr. Relat. Cancer, March 1, 2006; 13(1): 197 - 210. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Muguruma, S. Yano, S. Kakiuchi, H. Uehara, M. Kawatani, H. Osada, and S. Sone Reveromycin A Inhibits Osteolytic Bone Metastasis of Small-Cell Lung Cancer Cells, SBC-5, through an Antiosteoclastic Activity Clin. Cancer Res., December 15, 2005; 11(24): 8822 - 8828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Miwa, A. Mizokami, E. T. Keller, R. Taichman, J. Zhang, and M. Namiki The Bisphosphonate YM529 Inhibits Osteolytic and Osteoblastic Changes and CXCR-4-Induced Invasion in Prostate Cancer Cancer Res., October 1, 2005; 65(19): 8818 - 8825. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Nilsson, R. H. Larsen, S. D. Fossa, L. Balteskard, K. W. Borch, J.-E. Westlin, G. Salberg, and O. S. Bruland First Clinical Experience with {alpha}-Emitting Radium-223 in the Treatment of Skeletal Metastases Clin. Cancer Res., June 15, 2005; 11(12): 4451 - 4459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Onuma, Y. Azuma, H. Saito, T. Tsunenari, T. Watanabe, M. Hirabayashi, K. Sato, H. Yamada-Okabe, and E. Ogata Increased Renal Calcium Reabsorption by Parathyroid Hormone-Related Protein Is a Causative Factor in the Development of Humoral Hypercalcemia of Malignancy Refractory to Osteoclastic Bone Resorption Inhibitors Clin. Cancer Res., June 1, 2005; 11(11): 4198 - 4203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Tfelt-Hansen, N. Chattopadhyay, S. Yano, D. Kanuparthi, P. Rooney, P. Schwarz, and E. M. Brown Calcium-Sensing Receptor Induces Proliferation through p38 Mitogen-Activated Protein Kinase and Phosphatidylinositol 3-Kinase But Not Extracellularly Regulated Kinase in a Model of Humoral Hypercalcemia of Malignancy Endocrinology, March 1, 2004; 145(3): 1211 - 1217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Niida, M. Kawahara, Y. Ishizuka, Y. Ikeda, T. Kondo, T. Hibi, Y. Suzuki, K. Ikeda, and N. Taniguchi {gamma}-Glutamyltranspeptidase Stimulates Receptor Activator of Nuclear Factor-{kappa}B Ligand Expression Independent of Its Enzymatic Activity and Serves as a Pathological Bone-resorbing Factor J. Biol. Chem., February 13, 2004; 279(7): 5752 - 5756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Alvarez, M. Westmore, R. J. S. Galvin, C. L. Clapp, E. L. Considine, S. J. Smith, K. Keyes, P. W. Iversen, D. M. Delafuente, S. Sulaimon, et al. Properties of Bisphosphonates in the 13762 Rat Mammary Carcinoma Model of Tumor-Induced Bone Resorption Clin. Cancer Res., November 15, 2003; 9(15): 5705 - 5713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Peyruchaud, C.-M. Serre, R. NicAmhlaoibh, P. Fournier, and P. Clezardin Angiostatin Inhibits Bone Metastasis Formation in Nude Mice through a Direct Anti-osteoclastic Activity J. Biol. Chem., November 14, 2003; 278(46): 45826 - 45832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Yin, K. S. Mohammad, S. M. Kakonen, S. Harris, J. R. Wu-Wong, J. L. Wessale, R. J. Padley, I. R. Garrett, J. M. Chirgwin, and T. A. Guise From the Cover: A causal role for endothelin-1 in the pathogenesis of osteoblastic bone metastases PNAS, September 16, 2003; 100(19): 10954 - 10959. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Morris and H. I. Scher Clinical Approaches to Osseous Metastases in Prostate Cancer Oncologist, April 1, 2003; 8(2): 161 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Fournier, S. Boissier, S. Filleur, J. Guglielmi, F. Cabon, M. Colombel, and P. Clezardin Bisphosphonates Inhibit Angiogenesis in Vitro and Testosterone-stimulated Vascular Regrowth in the Ventral Prostate in Castrated Rats Cancer Res., November 15, 2002; 62(22): 6538 - 6544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Y. W. Bourguignon, P. A. Singleton, H. Zhu, and B. Zhou Hyaluronan Promotes Signaling Interaction between CD44 and the Transforming Growth Factor beta Receptor I in Metastatic Breast Tumor Cells J. Biol. Chem., October 11, 2002; 277(42): 39703 - 39712. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Abe, K. Hiura, J. Wilde, K. Moriyama, T. Hashimoto, S. Ozaki, S. Wakatsuki, M. Kosaka, S. Kido, D. Inoue, et al. Role for macrophage inflammatory protein (MIP)-1alpha and MIP-1beta in the development of osteolytic lesions in multiple myeloma Blood, August 28, 2002; 100(6): 2195 - 2202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Sanders, N. Chattopadhyay, O. Kifor, T. Yamaguchi, and E. M. Brown Ca2+-sensing receptor expression and PTHrP secretion in PC-3 human prostate cancer cells Am J Physiol Endocrinol Metab, December 1, 2001; 281(6): E1267 - E1274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. O. Oyajobi, D. M. Anderson, K. Traianedes, P. J. Williams, T. Yoneda, and G. R. Mundy Therapeutic Efficacy of a Soluble Receptor Activator of Nuclear Factor {{kappa}}B-IgG Fc Fusion Protein in Suppressing Bone Resorption and Hypercalcemia in a Model of Humoral Hypercalcemia of Malignancy Cancer Res., March 1, 2001; 61(6): 2572 - 2578. [Abstract] [Full Text]
|
|
|
|
 |
|
 E. M. Brown and R. J. MacLeod Extracellular Calcium Sensing and Extracellular Calcium Signaling Physiol Rev, January 1, 2001; 81(1): 239 - 297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Sanders, N. Chattopadhyay, O. Kifor, T. Yamaguchi, R. R. Butters, and E. M. Brown Extracellular Calcium-Sensing Receptor Expression and Its Potential Role in Regulating Parathyroid Hormone-Related Peptide Secretion in Human Breast Cancer Cell Lines Endocrinology, December 1, 2000; 141(12): 4357 - 4364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Pederson, B. Winding, N. T. Foged, T. C. Spelsberg, and M. J. Oursler Identification of Breast Cancer Cell Line-derived Paracrine Factors That Stimulate Osteoclast Activity Cancer Res., November 1, 1999; 59(22): 5849 - 5855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Thomas, T. A. Guise, J. J. Yin, J. Elliott, N. J. Horwood, T. J. Martin, and M. T. Gillespie Breast Cancer Cells Interact with Osteoblasts to Support Osteoclast Formation Endocrinology, October 1, 1999; 140(10): 4451 - 4458. [Abstract] [Full Text]
|
|
|
|
 |
|
 W. C. Dougall, M. Glaccum, K. Charrier, K. Rohrbach, K. Brasel, T. De Smedt, E. Daro, J. Smith, M. E. Tometsko, C. R. Maliszewski, et al. RANK is essential for osteoclast and lymph node development Genes & Dev., September 15, 1999; 13(18): 2412 - 2424. [Abstract] [Full Text]
|
|
|
|
 |
|
 L. M. Demers Biochemical Markers in the Management of Patients with Metastatic Bone Disease Clin. Chem., August 1, 1999; 45(8): 1131 - 1132. [Full Text] [PDF]
|
|
|
|
|
|
A. Berruti, L. Dogliotti, G. Gorzegno, M. Torta, M. Tampellini, M. Tucci, S. Cerutti, M. Mosca Frezet, M. Stivanello, G. Sacchetto, et al. Differential Patterns of Bone Turnover in Relation to Bone Pain and Disease Extent in Bone in Cancer Patients with Skeletal Metastases Clin. Chem., August 1, 1999; 45(8): 1240 - 1247. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
