| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Peter MacCallum Cancer Institute (R.M.L.M., N.C., J.D., P.P.), Melbourne 3002; and St. Vincent’s Institute of Medical Research (V.G., P.W.M.H.), Melbourne 3065, Australia
Address all correspondence and requests for reprints to: Robin Murray, M.D., Peter MacCallum Cancer Institute, Locked Bag 1, A’Beckett Street, Victoria 8006, Australia.
Abstract
PTH and ionized calcium levels were measured in 131 patients with advanced prostate cancer, all of whom had received at least first-line hormone therapy. Patients were classified into those in remission, those with stable disease, or those with progressive disease according to their prostate-specific antigen response and their clinical status.
Thirty-four percent of all patients had PTH levels above the upper level of normal for controls of similar age (7.0 pmol/liter), and in 44% of these patients this was associated with a normal ionized calcium. Patients with proven bone metastases had significantly higher PTH levels than those without. (7.3 ± 0.5 vs. 4.3 ± 0.4 pmol/liter, P < 0.0005).
There was evidence for a difference in the PTH levels between the three response groups. The PTH levels tended to be higher in patients with progressive disease. Thirty-seven of 65 patients (57%) with both progressive disease and proven bone metastases had elevated PTH levels. Mean levels of urinary deoxypyridinoline and cAMP were significantly greater in patients with high PTH than in those with a normal PTH.
Treatment with oral calcium supplements in 32 patients with a high PTH seemed to have only a transient effect on elevated PTH or low ionized calcium levels.
These data show that secondary hyperparathyroidism occurs frequently in patients with advanced prostate cancer, particularly in those with both progressive disease and bone metastases. The increased PTH levels are associated with an increase in bone resorption markers. These findings raise important questions about the role of PTH in progression of prostatic cancer in bone and the potential limitations of the use of bisphosphonates in patients with a raised PTH or low serum calcium.
PROSTATE CANCER IS the most common cancer in men and a leading cause of cancer deaths, usually associated with widespread bony metastases. Men with prostate cancer and bony metastases face a bleak future with increasing incapacity, debility, bone pain, and, not infrequently, spinal cord compression and paraplegia.
Metastatic deposits in bone have been found at autopsy in 60–85% of patients diagnosed with prostate cancer (1). In general, the characteristic feature of bone metastases in prostate cancer is an osteoblastic reaction rather than the osteolytic reaction usually seen with metastases from breast, bowel, or lung cancer. Some studies have suggested, however, that there is an osteolytic component associated with osteoclastic bone destruction in conjunction with this osteoblastic reaction (2, 3, 4, 5, 6). The relationship between these osteoblastic and osteolytic reactions and the spread of prostate cancer through bone is not clear.
It is known that high levels of PTH are associated with increased bone resorption and degradation of the bone matrix (7, 8). There are isolated reports of low serum calcium (9, 10, 11, 12, 13) and high PTH levels (14, 15, 16) in a number of studies involving small numbers of patients with advanced prostate cancer. The extent of these abnormalities and their pathophysiological significance is not well defined.
In this study of 131 patients with advanced prostatic cancer we report on the relationships between PTH, ionized calcium, urinary deoxypyridinoline (DPD), and cAMP in patients in remission (R), with stable (S) or progressive disease (PD) and with or without bone metastases (BM).
Patients and Methods
Between January 1998 and October 1998, 146 patients with proven advanced prostatic cancer attended the Endocrine Department at Peter MacCallum Cancer Institute. All patients had blood taken for measurement of serum PTH, ionized calcium, creatinine, prostatic-specific antigen (PSA), alkaline phosphatase (ALP), and 25 hydroxyvitamin D (25-OHD) on at least one occasion during the study period. When possible, urine samples were also collected for measurement of urinary cAMP and DPD. Patients with impaired renal function (serum creatinine, >0.13 mmol/liter), vitamin D deficiency (25-OHD, <25 mmol/liter), and elevated PTHrP (>2 pmol/liter), and patients receiving calcium or vitamin D supplements or who had received bisphosphonate in the previous 3 months were excluded. For each patient, the results from the first occasion within the study period for which measurements of all the variables were available were used for analyses. That is, the data set contained only one set of measurements per patient.
The study was approved by the Research and Ethics Committees of the Peter MacCallum Cancer Institute, and all patients and controls gave informed consent.
Before estimation of PTH patients were classified as in R, with S, or with PD according to their PSA response or their clinical status if the PSA response was considered to be unreliable because of recent radiotherapy.
PD was defined as a progressive increase in PSA of greater than or equal to 25% above the nadir. R was defined as a progressive decrease in PSA of at least 50%, whereas S was defined as a decrease of less than 50% or an increase of less than 25%.
The presence or absence of BM was determined by isotope bone scan within 3 months of the sampling date.
Blood and urine samples were taken at the time of clinic visits.
The normal range for PTH was determined in 108 male controls of similar
age (median age, 73 yr; range, 55–86) to the study cohort who all had
normal serum creatinine (
0.13 mmol/liter) and no evidence of
prostatic cancer. The controls were men who were attending Returned
Service League Clubs and who volunteered for the study.
Blood samples for PTH measurement were collected into 10-ml EDTA tubes, centrifuged, and plasma was stored frozen at -20 C until assay.
Intact PTH was measured by a two-site chemiluminescent enzyme immunometric assay on the Immulite Automated Immunoassay system (Diagnostic Products, Los Angeles, CA). For the purpose of the study, a high PTH was defined as a value that was greater than 7.0 pmol/liter. PTHrP was measured by an N-terminal RIA (17).
Ionized calcium was determined using a Bayer 850 Blood Gas Analyzer (normal range, 1.13–1.30 mmol/liter; Bayer Corp., Terrytown, NY). For the purpose of the study, a low ionized calcium was defined as a value of less than 1.13 mmol/liter.
25-OHD (25-hydroxycalciferol) was determined in serum by RIA following solvent extraction (25-Hydroxyvitamin D 125I RIA kit; DiaSorin, Inc., Stillwater, MN) (reference range, 25–108 nmol/liter).
PSA was determined in serum by a two-site sandwich immunoassay using a chemiluminescent label (Bayer Corp.ACS:180 PSA2) (reference range, 0–4.0 µg/liter).
Urinary DPD was measured by a competitive direct chemiluminescent immunoassay on the ACS:180 Automated Immunoassay analyser (Chiron Corp., East Walpole, MAs).
Urinary cAMP was measured by RIA using an antiserum supplied by Dr. P.
Marley (Department of Pharmacology, University of Melbourne, Melbourne,
Australia). Urine samples were diluted 1:100 in 50 mM
sodium acetate, 1 mM theophylline buffer (pH 5), and100
µl were assayed. One hundred microliters of cAMP standard (adenosine
3',5' cyclic phosphate) were diluted to give 0–5000 pmol/ml. One
hundred microliters of antibody and 100 µl iodinated cAMP (2'-o
monosuccinyladenosine 3',5' cyclic monophosphate tyrosylmethylester)
were added to assay tubes. All tubes were vortexed and incubated at 4 C
overnight. One milliliter of charcoal mixture was then added to all
tubes. The tubes were vortexed and incubated at 15 min before
centrifugation for 10 min at 4 C. Supernatants were aspirated, and
pellets were counted in a Pardard Cobra Auto-
counter. This assay
has a detection limit of 2.5 pmol/ml. The intra-assay and interassay
coefficients of variation were 5% and 10.8%, respectively.
Statistical methods
Clinical characteristics and biochemical parameters were compared across the response categories using a nonparametric ANOVA test for trend (Jonckheere-Terpstra test) for continuous data and the Cochran-Armitage test for trend for ordinal categorical data.
A test of the equality of the slopes in Fig. 1
, A and B, was performed by regressing
the logarithm of PTH level on the response status adjusting for BM
status (absent of present). The natural logarithm of PTH was used in
analyses as the distribution of PTH was highly skewed to the right.
|
, A and B, was performed by regressing
the logarithm of PTH level on ionized calcium adjusting for BM status.
The test was repeated after including a term for the response status
and an interaction term for the response status and BM status in the
model.
|
Ninety-five percent confidence intervals for percentages were obtained based on the exact binomial distribution. All statistical tests carried out were two-sided. No formal adjustments were made for multiple comparisons, with the exception discussed above. The analyses were carried out using StatXact (CYTEL Software Corp., Cambridge, MA) and SPSS software (SPSS, Inc., Chicago, IL). Graphs were plotted using the SPSS software and the S-PLUS statistical package (MathSoft, Inc., Seattle, WA).
Results
Of the 146 patients attending the Endocrine Department at Peter MacCallum Cancer Institute between January 1998 and October 1998, 131 patients satisfied the eligibility criteria. The results of 25-OHD measurements were excluded for 12 patients because of a laboratory error. Measurement of PTH and urinary DPD and urinary cAMP was available in 51 and 57 patients, respectively.
The mean (±2 SD) PTH in controls was 3.8 ± 3.2 pmol/liter. For the study, the upper limit of normal PTH was defined as 7.0 (mean + 2 SD) pmol/liter.
Summaries of the clinical characteristics and biochemical parameters
for all patients and by response status are shown in Table 1. Forty-five patients (34%; 95%
confidence interval, 26–43%) had an elevated PTH level. There was
evidence for a difference in the PTH levels between the three response
groups (P < 0.00005). The PTH level tended to be
higher in patients with PD. The levels of other biochemical parameters
that were significantly different between the three response groups
were PSA (P < 0.00005), Ca++
(P = 0.0022), ALP (P < 0.00005), DPD
(P = 0.0072), and cAMP (P = 0.022).
|
PTH levels for each of the response groups subdivided according to
their BM status are shown in Fig. 1
, A and B, respectively. The
relationships between PTH and response status in the BM+ and BM-
groups were different (P = 0.001): there was a
significant increase in the PTH level from the R group to the S group
to the PD group in patients with BM, whereas there was no significant
relationship between the PTH level and response status in patients with
no BM.
There was a significant inverse relationship between PTH and ionized
calcium (Table 2; P =
0.0001, Fisher’s exact test). Fifty-eight percent of patients who had
a low ionized calcium had an elevated PTH level whereas only 23% of
patients with normal ionized calcium had an elevated PTH. The
relationships between PTH and ionized calcium in patients according to
their BM status are shown in Fig. 2, A and B, respectively. The
correlation between PTH and ionized calcium seemed to be stronger in
the BM+ group (Pearson correlation coefficient, r = -0.48)
compared with the BM- group (r = -0.12). However, the result of
a test of the equality of the slopes in the plots using a regression
model was not statistically significant (P = 0.14).
This conclusion remained unaltered after allowing for the response
status (P = 0.20).
|
). Urinary cAMP and urinary
DPD were greater in patients with a raised PTH (P =
0.023 and P = 0.001, respectively, t tests;
Table 3).
|
. The plot suggests ionized calcium
increases to a plateau level after 1 month of treatment and decreased
by 3 months. However, none of the results of testing for the
differences in Ca++ levels over the 3-month
period were statistically significant after adjusting for multiple
testing. The plot also suggests that PTH decreases to a plateau level
at 1 month and increased by 3 months. The results of testing the PTH
levels at 1 and 2 months relative to the pretreatment level were
statistically significant (P = 0.002 and
P = 0.009, respectively). These significant findings,
however, need to be interpreted with caution as the possible effect of
calcium supplements is confounded with the phenomenon of "regression
toward the mean": patients who received calcium supplements all had
high PTH levels at pretreatment and would be expected to have lower PTH
levels at subsequent measurement even when calcium supplements were not
given.
|
It did not appear that the type of treatment had any effect on PTH levels.
Discussion
This study shows that secondary hyperparathyroidism occurs frequently in patients with advanced prostate cancer, with an incidence of 57% in patients with PD and BM. Patients who had other known possible causes of secondary hyperparathyroidism, such as renal failure or vitamin D deficiency, were excluded from this series. Our finding that elevated PTH was associated with high urinary cAMP and with high urinary DPD suggests that PTH was biologically active and that this increase was responsible, in part at least, for increased bone breakdown.
It has been known for many years that hypocalcemia can occur in patients with osteoblastic metastases from prostate cancer (9, 10, 11, 12, 13). In 1962, Ludwig (16) postulated the following sequence: osteoblastic metastases cause increased deposition of calcium and phosphate in bone, tending to decrease serum concentrations of both ions. The resulting hypocalcemia stimulates PTH secretion. Secondary hyperparathyroidism then causes a further decrease in serum phosphate concentration that, in some instances, ultimately reaches hypophosphatemic levels. In our series, patients with BM had a lower mean ionized calcium level than those without. Forty-three (33%) patients had a low ionized calcium, and, of these, 35 (81%) were in the group with BM.
Elevated PTH levels, usually in association with a low corrected calcium, have also been previously reported in prostate cancer. Minisola et al. (12) in 1987 reported that 2 of 14 patients with BM from prostate cancer had elevated PTH levels. Rico et al. (14) reported high levels in 2 of 20 patients. Charhon et al. (15) reported that serum PTH was significantly increased in 14 patients with osteosclerotic BM compared with age-matched controls. In our series of 131 patients, 45 (34%) had elevated PTH levels, and in 20 (44%) patients this was associated with a normal ionized calcium. This may represent a compensated state, with serum ionized calcium being maintained within the normal range by increased circulating PTH.
Mean PTH levels were higher in patients with BM and in those with PD with a strikingly high incidence (57%) occurring in those who had both PD and BM. Our findings are consistent with the hypothesis that osteoblastic metastases in prostate cancer are the primary phenomenon inducing hypocalcemia and compensatory hyperparathyroidism in these patients.
PTH is the principal hormonal agent that controls bone resorption, and elevated levels are associated with increased bone turnover, secondary to an increase in numbers of and activity of osteoclasts. High levels of PTH are associated with increased bone resorption and degradation of the bone matrix (7, 8). Evidence of increased bone resorption with elevated urinary pryidinoline and DPD (19, 20, 21, 22) levels has been reported in patients with prostate cancer. In one study, bone resorption markers were high in patients with active cancer but not in those with controlled disease. Histomorphological studies have also confirmed increased bone resorption in metastatic prostate cancer (14). In our patients there was an association of PTH levels with urinary DPD excretion, consistent with a PTH-driven generalized increase in bone resorption. High PTH levels could also explain the observation reported by Urwin et al. (5) that in patients with BM from prostate cancer there was histological evidence of increased bone resorption at sites distant from skeletal metastases.
A number of substances, including proteases released by tumors, are thought to result in extracellular matrix breakdown and lysis facilitating tumor invasiveness (8, 23). The capacity to destroy mineralized matrix, however, requires the involvement of the osteoclast (24). The growth of prostate cancer is often accelerated once it has spread to bone, suggesting that the bone microenvironment may provide a proliferation-stimulating factor for metastatic prostate cancer cells (23). Substances have been suggested as possible prostate cancer-stimulating factors, including bone fibroblast-derived factor and transferrin.
A systemic "vicious cycle" could occur in prostate cancer patients with increased PTH causing bone matrix degradation and release of proliferation-stimulating substances, leading to stimulation and progression of the prostate cancer and, in turn, further deposition of calcium in sclerotic metastases with a subsequent further elevation of PTH. A similar local mechanism has been postulated for the progression of breast cancer in bone, with PTHrP causing osteoclastic bone resorption, leading to the release of growth factors (in particular TGFß) and stimulation of cancer cell proliferation (23). Whereas PTHrP has been shown to be present in some prostatic tissue (25), circulating levels are generally undetectable in patients with prostatic cancer—only 1 of 100 of our patients had a detectable level.
Three patients who seemed to be in R also had high levels of PTH in our series. PD may not always be reflected by a rising PSA, however, these patients were both biochemically and clinically in R. Evidence of BM was not an absolute prerequisite for a high PTH in our patients, but it is possible that the one patient who had a high PTH but did not have proven BM may have had occult BM, which were not yet apparent on bone scan.
Supplementation with oral calcium in 32 of our patients had only a
transient effect on ionized calcium and elevated PTH levels. Ionized
calcium seemed to increase to a plateau level after 1 month and to
decrease again by 3 months. However, none of the changes in
Ca++ were statistically significant after
adjusting for multiple testing. PTH values fell significantly (but not
to normal) to a plateau level at 1 month but had returned to
pretreatment by 3 months (Fig. 3). These findings suggest that large
amounts of calcium may be necessary to significantly raise serum
calcium and lower serum PTH in these patients and clearly have
significant implications for current and planned studies evaluating the
effect of bisphosphonates in patients with prostate cancer. The use of
bisphosphonates in such patients would be expected to result in further
falls in serum calcium and, thus, further increases in PTH, as has been
reported in the treatment of hypercalcemia of malignancy
(26). This could, in turn, cause increased bone lysis and
limit the therapeutic effectiveness of bisphosphonates unless calcium
supplements were given in doses sufficient to maintain PTH in the
normal range.
Acknowledgments
We thank Kally Yuen for hard work and expert advice in carrying out the statistical analysis. We also thank Bruce Ruxton (President of the Victorian Branch of the Returned Services League) for help in organizing the volunteers who gave blood as controls.
Footnotes
Abbreviations: ALP, Alkaline phosphatase; BM, bone metastases; DPD, deoxypryidinoline; 25-OHD, 25-hydroxyvitamin D; PD, progressive disease; PSA, prostate-specific antigen; R, remission; S, stable disease.
Received October 13, 2000.
Accepted May 24, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  G. G. Schwartz Prostate Cancer, Serum Parathyroid Hormone, and the Progression of Skeletal Metastases Cancer Epidemiol. Biomarkers Prev., March 1, 2008; 17(3): 478 - 483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. G. Schwartz, M. C. Hall, D. Stindt, S. Patton, J. Lovato, and F. M. Torti Phase I/II Study of 19-nor-1{alpha}-25-Dihydroxyvitamin D2 (Paricalcitol) in Advanced, Androgen-Insensitive Prostate Cancer Clin. Cancer Res., December 15, 2005; 11(24): 8680 - 8685. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Berruti, M. Tucci, C. Terrone, R. M. Scarpa, A. Angeli, and L. Dogliotti Re: A Randomized, Placebo-Controlled Trial of Zoledronic Acid in Patients With Hormone-Refractory Metastatic Prostate Carcinoma J Natl Cancer Inst, February 19, 2003; 95(4): 332 - 333. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vieth Dairy products, calcium, and prostate cancer risk in the Physicians' Health Study Am. J. Clinical Nutrition, August 1, 2002; 76(2): 490 - 491. [Full Text]
|
|
|
|
 |
|
 A. Berruti, L. Dogliotti, M. Tucci, R. M. Scarpa, and A. Angeli Hyperparathyroidism Due to the So-Called Bone Hunger Syndrome in Prostate Cancer Patients J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1910 - 1911. [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 |
