Prognostic value of serum markers for prostate cancer

ORIGINAL ARTICLE

Prognostic value of serum markers for prostate cancer

ULF-HAKAN STENMAN1, PER-ANDERS ABRAHAMSSON2, GUNNAR AUS3,

HANS LILJA4, CHRIS BANGMA5, FREDDIE C. HAMDY6, LAURENT BOCCON-GIBOD7

& PETER EKMAN8

1Department of Clinical Chemistry, Helsinki University Central Hospital, Helsinki University, Helsinki, Finland,2Department of Urology, Malmo University Hospital, Malmo, Sweden, 3Department of Urology, Goteborg University,

Goteborg, Sweden, 4Department of Pathology, Memorial Sloan�/Kettering Cancer Center, New York, USA, 5Department

of Urology, Erasmus University, Rotterdam, The Netherlands, 6Academic Urology Unit, Royal Hallamshire Hospital,

Sheffield, UK, 7Service d’Urologie, Hopital Bichat, Paris, France, 8Department of Urology, Karolinska University Hospital,

Stockholm, Sweden

AbstractThe incidence of prostate cancer has increased dramatically during the last 10�/15 years and it is now the commonest cancerin males in developed countries. The increase is mainly caused by the increasing use of opportunistic screening or case-finding based on the use of prostate-specific antigen (PSA) testing in serum. With this approach, prostate cancer is detected5�/10 years before giving rise to symptoms and on average 17 years before causing the death of the patient. While this has ledto detection of prostate cancer at a potentially curable stage, it has also led to substantial overdiagnosis, i.e. detection ofcancers that would not surface clinically in the absence of screening. A major challenge is thus to identify the cases that needto be treated while avoiding diagnosing patients who will not benefit from being diagnosed and who will only suffer from thestigma of being a cancer patient. It would be useful to have prognostic markers that could predict which patients need to bediagnosed and which do not. Ideally, it should be possible to measure these markers using non-invasive techniques, i.e. bymeans of serum or urine tests. As it is very useful for both early diagnosis and monitoring of prostate cancer, PSA isconsidered the most valuable marker available for any tumor. Although the prognostic value of PSA is limited, measurementof the proportion of free PSA has improved the identification of patients with aggressive disease. Furthermore, the rate ofincrease in serum PSA reflects tumor growth rate and prognosis but, due to substantial physiological variation in serumPSA, reliable estimation of the rate of PSA increase requires follow-up for at least 2 years. Algorithms based on thecombined use of free and total PSA and prostate volume in logistic regression and neural networks can improve thediagnostic accuracy for prostate cancer, and assays for minor subfractions of PSA and other new markers may provideadditional prognostic information. Markers of neuroendocrine differentiation are useful for the monitoring of androgen-independent disease and various bone markers are useful in patients with metastatic disease.

Key Words: Diagnosis, follow-up, prognosis, prostate cancer, serum markers, treatment

Introduction

One of the most typical characteristics of prostate

cancer is the high prevalence of subclinical disease. It

has been shown in autopsy studies [1,2] that :/30�/

60% of males aged 50�/70 years have at least a

microscopic cancer. In North America and Europe,

:/15�/30% of these are diagnosed clinically while

4�/7%, which corresponds to 2�/3.5% of all men, die

from this disease [3,4]. Prostate cancer is thus a

serious health problem. In order to reduce the

morbidity and mortality of prostate cancer, regular

screening based on determination of prostate-

specific antigen (PSA) in serum is advocated in

some countries, although this policy is not recom-

mended in others. However, mass screening for

prostate cancer is not practiced as a public

health policy anywhere in the world. Ongoing

controlled screening studies [5] are expected to

Correspondence: Ulf-Hakan Stenman, Helsinki University Central Hospital, BP 63, Helsinki University, FIN-00014 Helsinki, Finland. E-mail:

[email protected]

Scandinavian Journal of Urology and Nephrology Supplement, 2005; 216: 64�/81

ISSN 0300-8886 print/ISSN 1651-2537 online # 2005 Taylor & Francis

DOI: 10.1080/03008880510030941

show whether screening is effective and medically

justifiable.

PSA is the landmark marker for the management

of prostate cancer in the diagnosis, staging and

follow-up of the disease. When used for early

detection, prostate cancer is diagnosed long before

it gives rise to symptoms. In retrospective serum

bank studies, the average time from an increase in

serum PSA to clinical diagnosis of prostate cancer

was found to be 7�/10 years [6�/8], while the lead

time estimated from a screening study was 11.2 years

[9]. With a cut-off of 4 mg/l, the median time from

PSA increase to death from prostate cancer is 17

years [8]. If the cut-off is lowered to 3 mg/l, the time

to clinical symptoms and death can be expected to

increase by several years.

The increasing use of PSA testing has led to a

rapid increase in the incidence of prostate cancer [4];

in some countries, e.g. Finland, the increase has

been more than two-fold over a 10-year period [10].

When PSA testing was introduced, an initial rapid

increase in incidence was caused by a ‘‘harvesting’’

effect as a result of the detection of cases that would

otherwise have surfaced only within the subsequent

5�/10 years. This happened in the USA in the early

1990s: the age-adjusted incidence peaked around

1992, after which it decreased. Since 1995 the

incidence has continued to rise faster than before

the PSA era [4]. In many European countries, which

have had a relatively low incidence of prostate

cancer, the incidence is now approaching that in

the USA. Today, :/17% of all men in the USA will

be diagnosed with prostate cancer and 3.4% will die

of it [3,4]. Opportunistic screening causes this rising

incidence and while only some of the screening-

detected cancers are likely to threaten the life of the

patient, the course of an individual prostate cancer

cannot be predicted reliably [11]. Even without

treatment, only a minor proportion of screening-

detected patients die within 10 years and 20�/30%

survive for several decades with little or no disease

progression [8,12�/14]. Thus, if PSA screening had

not been used, a large proportion of the patients

diagnosed today would never have been detected

[15]. The indiscriminate use of PSA testing in

elderly men and subsequent prostate biopsies will

result in substantial overdiagnosis and overtreatment

of an otherwise potentially non-life-threatening con-

dition [16].

An increase in the incidence of prostate cancer is

often accompanied by an increase in mortality due to

attribution bias, i.e. the cause of death is erroneously

ascribed to prostate cancer once a man has been

diagnosed with the disease. This explains the in-

crease in prostate cancer mortality a few years after

the peak incidence in the USA in 1992 [17,18].

Therefore, it is likely that the decrease in prostate

cancer mortality after the peak in 1992�/93 can be

partly explained by the simultaneous decrease in

incidence. The increasing use of endocrine therapy

in early patients with recurrent disease is also likely

to reduce mortality, but treatments with curative

intent for early-stage disease may have started to play

a role in the recent decrease in mortality [19].

About 25�/35% of patients who are diagnosed

when PSA is in the range 4�/10 mg/l will experience a

PSA relapse after treatment with curative intent [20]

and therefore there is a trend to further lower the

cut-off to 2.5�/3 mg/l [21�/24]. While this will

probably reduce the number of patients with incur-

able disease, it will also increase overdiagnosis [8,9].

It is therefore highly desirable to reduce this problem

to an acceptable level by using algorithms to predict

which men are likely to benefit from early detection

[25]. Programs predicting overdiagnosis and lead

time show that if screening is applied to men aged 75

years, two out of three cases will not benefit from

having been diagnosed [9]. So far these methods

have not been used to reduce the problem of

overdiagnosis.

Because the benefits of early detection are unclear

and radical therapy is not always curative and not

without side-effects it is important to develop better

prognostic algorithms in order to avoid unnecessary

diagnosis and treatment in men who do not benefit

from being diagnosed. These efforts must be ba-

lanced against the need to detect those at risk of

developing aggressive disease and dying from pros-

tate cancer at an even earlier stage than is possible

today. Histological grade, in combination with

clinical and pathological stage, is the most important

prognostic factor and this is dealt with in another

paper in this supplement. Markers in biological

fluids also have prognostic value and the aim of

this article is to describe the prognostic value of these

markers in the management of prostate cancer.

Serum markers for prostate cancer

PSA

The chemistry and biology of PSA and its various

molecular forms. PSA is a 30�/33-kDa protease

belonging to the kallikrein family, which comprises

15 serine proteases encoded by a cluster of genes on

chromosome 19q3 [26]. The genes are numbered

KLK1�/15 and the corresponding proteins hK1�/15.

PSA (or hK3) is closely related to another prostate-

specific protease, hK2, with which it shares 80%

homology. PSA is produced as a preproenzyme

comprising a signal peptide that is removed during

synthesis. The secreted proenzyme, proPSA, con-

Prognostic value of serum markers for prostate cancer 65

tains 244 amino acids, including a seven-amino acid

activation peptide [27] which is split off by a trypsin-

like enzyme after secretion. The activating enzyme

can be hK2 [28] or trypsin, which is also expressed

in the prostate [29]. Mature PSA contains 237

amino acids and a carbohydrate chain linked to

serine 45 and its molecular weight is 28 430 Da [30].

When isolated from seminal fluid, PSA has been

completely activated and 30�/40% of it has been

partially degraded by proteolytic cleavage or ‘‘nick-

ing’’ after Arg85, Lys145 and Lys182 [31,32].

In the prostate, secretion of PSA is directed into

the prostatic ducts and its function in semen is to

digest the gel formed by semenogelins and fibronec-

tin after ejaculation [33]. Normally only a minor

fraction of PSA escapes into the circulation but when

the tissue architecture in a prostatic cancer is de-

ranged and the tumor loses contact with the pro-

static ducts, the secretion is directed into the

extracellular space and directly into the circulation

[25,34]. This explains why a prostatic cancer pro-

duces :/30-fold higher serum concentrations of PSA

per gram of tissue than the normal prostate and 10

times more than benign prostatic hyperplastic

(BPH) tissue [35], although expression of PSA is

lower in cancer than in benign prostatic tissue [36].

Most of the PSA detected by immunoassays in

male serum occurs in a complex with a1-anti-

chymotrypsin (PSA�/ACT), while 5�/35% is free.

In patients with prostate cancer, the proportion of

PSA�/ACT is higher and that of free PSA lower than

in healthy men and patients with BPH [37,38]. The

utility of the determination of PSA�/ACT [39] or

free PSA and especially the proportion of free PSA

(often called the F/T ratio or F/T-PSA) has been

extensively documented [40�/43], and assays for free

PSA are now widely available. Assay of PSA�/ACT is

hampered by technical problems that cause a vari-

able non-specific background [40], although this can

be avoided by the use of monoclonal antibodies

specific for ACT in complex with PSA. However, for

methodological reasons determination of the ratio of

PSA�/ACT to total PSA is more demanding than

that of free to total PSA [44]. This problem can be

controlled by simultaneous assay of both forms in

the same tube using a double-label assay, but such

assays are not commercially available [39,44]. PSA�/

ACTassays that do not utilize this approach have not

been found to provide an advantage over total PSA.

About 1�/2% of PSA in male serum occurs in

complex with a1-protease inhibitor (API; also called

a1-antitrypsin) and 5�/10% in complex with a2-

macroglobulin (A2M) [37,45,46]. Somewhat sur-

prisingly, the proportion of these complexes is higher

in BPH than in prostate cancer serum and assay of

each form has been shown to improve cancer

specificity when used together with free and total

PSA [46,47]. Assays for these complexes are de-

manding and not generally available. PSA�/A2M is

not detected by conventional immunoassays, but

PSA�/ACT and PSA�/API are both detected by

assays for total PSA and an assay for ‘‘complexed

PSA’’ or cPSA [48]. This assay measures PSA�/ACT

and PSA�/API together, which is theoretically dis-

advantageous because PSA�/ACT increases while

PSA�/API decreases in prostate cancer. However,

the contribution of PSA�/API is small and determi-

nation of cPSA has been found to provide a

moderate advantage over total PSA [48,49], but

not over F/T-PSA [50].

The mechanisms causing a higher proportion of

cPSA in prostate cancer than BPH are only partially

understood. The most likely explanation is that PSA

reaches the circulation by different routes from

cancerous and benign prostatic tissues. To reach

the circulation from benign tissue, PSA has to leak

out through the extracellular space, where it is

subject to degradation [34], and a large proportion

of the PSA recovered from extracellular fluid of BPH

tissue has been found to be partially degraded or

‘‘nicked’’ by proteolytic cleavage [51,52]. It is also

possible that some PSA diffuses back into the

circulation after being activated and cleaved in the

prostatic ducts. A majority of the PSA released from

a prostate cancer is thought to reach the circulation

in an active form, which can then form complexes

with inhibitors, and this could explain the high

proportion of PSA�/ACT and the low proportion of

free PSA in the serum of prostate cancer patients

[25]. Isolation and characterization of PSA from

patient sera has confirmed that a larger proportion of

PSA is nicked in sera from BPH than from that in

cancer patients [53,54]. Recent studies with anti-

bodies to various forms of free PSA have confirmed

these observations. Thus an assay specific for PSA

cleaved after Lys182, called B-PSA, has shown

increased specificity for BPH [55], while an assay

specific for PSA that is not nicked at Lys145 has

appeared to be more specific for prostate cancer

[56]. Assays detecting various proPSAs have also

been shown to improve cancer specificity [57].

Taken together, these findings are in agreement

with the notion that the processing of PSA is

different during release from benign and malignant

prostatic tissue [25]. Specific determination of

various forms of free PSA in serum is a potential

way of further improving the cancer specificity of

PSA, but so far these methods are not generally

available.

A number of physiological and pathological fac-

tors affect the concentrations of PSA and its

66 U.-H. Stenman et al.

subfractions in serum. A summary of these effects is

shown in Table I.

Other serum markers

While PSA is a very sensitive marker for prostate

cancer, it is prostate- rather than cancer-specific and

thus most moderately elevated PSA values can be

caused by benign diseases: mainly BPH, and occa-

sionally prostatitis. Therefore much effort has been

devoted to the development of markers that could

improve the cancer specificity of PSA.

hK2. The expression of hK2 is highly prostate-

specific, but while the expression of PSA tends to

decrease with increasing tumor grade, that of hK2

increases or remains constant. Thus, in some studies

[58�/62], the ratios of hK2 to PSA and F/T-PSA

have been found to improve the identification of

aggressive prostate cancer. However, in other studies

[63,64] a clinical utility of hK2 was not observed.

Because serum concentrations of hK2 are 50�/100-

fold lower than those of PSA, determination of hK2

is quite demanding and, at low levels, where

discrimination between indolent and aggressive can-

cer is important, the correlation with tumor behavior

is weak [65,66]. The variability in assay performance

may be responsible for the inconsistent results

reported in some studies [67]. The clinical value of

hK2 remains to be determined but, if its association

with tumor aggressiveness is proven, serum measure-

ments might be useful in the selection of high-

risk patients for biopsy and aggressive treatment

strategies.

Prostatic acid phosphatase. Prostatic acid phosphatase

(PAP) was one of the first tumor markers to be used

clinically. With the advent of PSA testing, and its

ready availability, PAP does not provide any useful

additional information in the context of prostate

cancer and it is presently not used clinically [68].

Markers reflecting neuroendocrine differentiation. Data

on the prognostic significance of neuroendocrine

(NE) tissue markers in prostate cancer are conflict-

ing. Some researchers have shown a significant

correlation between NE differentiation (NED),

tumor grade and poor prognosis, but other groups

failed to demonstrate a significant association be-

tween the number of NE tumor cells, tumor grade

and prognosis. Despite these limitations, NE tumor

cells may have biological significance for determin-

ing prostate cancer behavior. Controversial data in

terms of the prognostic value of NE tissue markers

may be explained by a number of factors, including

differences in patient cohorts, variable methodologi-

cal approaches, limited amounts of tissue sample

obtained for analysis and unequal distribution of NE

tumor cells [69,70]. Measurement of NE markers in

serum may therefore represent a more accurate

determinant of the entire NE tumor cell population

in primary and metastatic lesions compared with NE

tissue markers.

Table I. Mechanisms causing changes in serum PSAa.

Mechanism PSA elevation PSA depression

Intra-individual variation 30% 30%

Prostate cancer Up to 10 000-fold

BPH Up to 10-fold

Prostatitis Up to 50-fold

Impaired kidney function

(severe decrease in GFR)

Moderate elevation of free

PSA and% free PSA

Urethral instrumentation Up to 10-fold

Urinary retention Up to 10-fold

Ejaculation NS

DRE NS

Prostatic needle biopsy Up to 50-fold

Change in prostate volume Up to 5-fold

Asian diet Up to 25%

Relief of urinary retention Up to 4-fold

Finasteride About 50%

GnRH antagonists/surgical

castration

Up to 100-fold

aDRE, transrectal ultrasonography, urethral instrumentation and prostate biopsy cause a spike in free PSA levels but total PSA levels remain

largely unchanged following the procedure. Free PSA levels return to baseline within a few days.

GFR�/glomerular filtration rate; GnRH�/gonadotropin-releasing hormone.

Prognostic value of serum markers for prostate cancer 67

Correlations between serum levels of chromogra-

nin A (CgA) and neuron-specific enolase (NSE) and

disease progression and androgen independence

have been demonstrated in several studies [71�/73].

Interestingly, NE markers do not appear to be

suppressed by androgen ablation [69,71�/73]. Ka-

miya et al. [74] demonstrated that high serum NSE

levels are associated with poor prognosis in patients

with metastatic prostate cancer and concluded that

serum NSE measurements may be of prognostic

help in patients treated with androgen suppression.

Markers reflecting the risk of developing cancer

A number of hormones and hormone-binding pro-

teins affect the proliferation of both benign and

malignant prostatic cells. While these are not con-

sidered tumor markers, they have a potential role for

evaluating the risk of developing prostate cancer.

Insulin-like growth factor 1 and insulin-like growth

factor binding protein 3. Insulin-like growth factor 1

(IGF-1) stimulates cell proliferation and increased

serum levels of IGF-1 have been shown to be

associated with several cancers [75]. High serum

concentrations of IGF-1 have been reported to be

associated with the future development of prostate

cancer in serum bank studies [76,77], but not with

the detection of prostate cancer in screening studies

[78]. This discrepancy may be explained by the

evolution of prostate cancer diagnosis over the past

two decades. Before the PSA era, a large proportion

of prostate cancers were diagnosed incidentally

following transurethral resection of the prostate for

presumed BPH associated with bladder outflow

symptoms [79]. Interestingly, acromegaly patients,

who have high concentrations of growth hormone,

and therefore also of IGF-1, develop BPH at a very

early age [80]. One could therefore speculate that

the association between IGF-1 and prostate cancer

could be caused by the incidental detection of

prostate cancers in BPH patients. However, in a

recent study [77], high IGF concentrations were

found to be associated with later detection of

prostate cancer in men aged B/60 years, suggesting

that a high serum IGF-1 level is important at the

early stages of prostate cancer development. There is

sufficient evidence to suggest that IGF-1 may be

associated with both BPH and prostate cancer but,

in a screening setting, determination of IGF-1 does

not appear to improve the diagnostic accuracy

obtained with free and total PSA [78,81].

In plasma, most IGF-1 is bound to several binding

proteins (BPs), of which IGFBP-3 is the major one.

IGF-1 bound to IGFBP-3 is not biologically active

and therefore the ratio of IGF-1 to IGFBP-3 is

thought to be important. Expression of both IGF-1

and IGFPB-3 is regulated by growth hormone but

the ratio between them may vary. However, adjust-

ment for IGFBP-3 has not been found to modulate

the impact of IGF-1 on the risk of prostate cancer

[77].

Testosterone. Testosterone is necessary for the devel-

opment of the prostate [82] and eunuchs do not

develop prostate cancer [83]. The concentration of

testosterone in serum has been thought to be a

prognostic factor, but results of epidemiological

studies are controversial. In a recent large serum

bank study, testosterone levels in the lowest and

highest quintiles were associated with a reduced risk

of developing prostate cancer (p-value for trend�/

0.05). However, after adjustment for Sex Hormone

Binding Globulin (SHBG), these correlations were

no longer significant. Variations in serum testoster-

one do not therefore appear to affect the risk of

prostate cancer [84].

Leptin. The association between serum leptin con-

centration and prostate cancer is of potential interest

because of the known epidemiological risk factors of

Western lifestyle and obesity. Preliminary work

suggested a positive correlation [85], but this finding

was not confirmed in a later study [86].

Markers of bone metabolism

In the advanced stages of prostate cancer the clinical

picture is dominated by the problem of bone

metastases, which occur in 85% of patients. They

represent the most important cause of morbidity: the

pain caused by metastases requires substantial an-

algesia and the resulting complications include

pathological fractures and spinal compression. The

development of skeletal metastases occurs at a

constant rate of :/8%/year in patients with advanced

disease, reaching 40% at 5 years. Skeletal metastases

in prostate cancer are predominantly osteosclerotic,

although histological and biochemical studies indi-

cate associated osteoclastic activity. Early recogni-

tion of metastatic spread to bone is critical in the

clinical management of these patients. A significant

proportion of men staged as having clinically loca-

lized disease have occult circulating prostate cancer

cells, which are undetectable using conventional

methods, but their significance in terms of their

metastatic potential is unclear. Furthermore, it is

now well recognized that androgen suppression will

lead to loss of bone mineral density and osteoporosis

in a significant proportion of men. Most of these

68 U.-H. Stenman et al.

changes will occur in the first 12 months of

hormonal treatment and bone protection strategies

are being developed, using established agents such as

bisphosphonates. Monitoring of such changes in the

skeleton is paramount in order to reduce the

morbidity associated with treatment and the devel-

opment of late complications.

At present, serum PSA measurement and bone

scintigraphy are used to stage and monitor disease

status, with bone scanning remaining the most

sensitive technique for the detection of skeletal

metastasis. However, bone scans are expensive and

time-consuming, and some discrepancies may occur

due to lack of specificity. Novel tests to predict the

metastatic potential of prostate cancer at diagnosis

and its progression after treatment are urgently

needed.

Markers of bone turnover. The major structural

protein in bone is type I collagen, which is synthe-

sized by osteoblasts and accounts for :/90% of the

organic matrix. Type I collagen is initially synthe-

sized as a pro-collagen, and during post-translational

modification in the extracellular space, the amino

(P1NP) and carboxyl (P1CP) propeptides are re-

moved by specific peptidases. When collagen is

broken down during bone resorption, the collagen

crosslinks cannot be degraded and are therefore

excreted and filtered by the kidney. Fifty to 60%

are excreted bound to type 1 collagen fragments

linked to the C- or N-terminal telopeptides, and

recently developed assays can detect these C-term-

inal telopeptides (CTX) in the serum. Various

breakdown products of type I collagen, such as

urinary N-telopeptide, urinary a/bC-telopeptides

and urinary deoxypyridinoline, are markers of

bone resorption. In contrast, serum total alkaline

phosphatase, bone-specific alkaline phosphatase,

osteocalcin and type I collagen carboxy-terminal

propeptide are markers of bone formation [87].

Bone turnover markers reflect osteoblast activity

during bone formation, and osteoclast activity dur-

ing bone resorption. The levels of these markers

increase in patients with metastatic bone disease.

Prostate cancer is recognized for its propensity to

metastasize to bone marrow and bone resorption and

formation factors have been studied in prostate

cancer for many years. Gutman et al. [88] investi-

gated general phosphatase activity levels in prostate

cancer, and Huggins and Hodges [89] were the first

to correlate total alkaline phosphatase (AP) serum

levels with the presence of bony metastases in

prostate cancer.

Both categories of markers may be increased in

prostate cancer metastasis, largely because of the

mixed nature of the bony lesions. Repair processes

involved in healing these lesions may also explain the

raised levels of these factors, with decreased speci-

ficity. Type I collagen of bone is strengthened by

specific molecular crosslinks that provide rigidity.

Crosslinks of mature type I collagen in bone are the

pyridinium crosslinks pyridinoline and deoxypyridi-

noline (DPD). DPD is formed by the enzymatic

action of lysyl oxidase on the amino acid lysine.

During bone resorption, this crosslink product is

released into the blood and subsequently excreted in

the urine. No further metabolism of this breakdown

product occurs before excretion; therefore, it is

thought to represent a direct measure of bone

resorption [90]. Urinary DPD has a higher positive

predictive value for bony metastases than either AP

or PSA [90].

Levels of the bone formation marker P1NP and

the bone resorption marker CTX exhibit significant

changes in patients with bone metastases [91�/93]

and these correlate with the extent of bone involve-

ment. The carboxy-terminal pyridinoline cross-

linked telopeptide of type I collagen (ICTP) is

another degradation product of type I collagen,

which reflects the extent of disease and response to

hormonal treatment in patients with bone metastases

[94,95]. The serum levels of type I collagen (ICTP)

and carboxyterminal propeptide of type I procolla-

gen (PICP) correlate with the presence of bone

metastases and reflect response to treatment during

androgen suppression [96]. The serum levels of AP

and urinary DPD reflect events secondary to pros-

tate cancer-induced bone disease. Urinary DPD

concentrations predict independently skeletal-re-

lated events (SREs) [97].

In a study by Garnero et al. [93], urinary non-

isomerized (a CTX) and b-isomerized (b CTX) type

I collagen C-telopeptides (CTX) and a new assay for

serum CTX were used to assess bone resorption

following bisphosphonate treatment in patients with

skeletal metastases from prostate cancer. A signifi-

cant reduction in these markers, but not in bone

formation markers, was observed, suggesting that

resorption markers may be used to monitor the

effects of bisphosphonate treatment in these pa-

tients. Because the new and potent bisphosphonate

zoledronic acid appears to delay SREs in men with

metastatic prostate cancer [98,99], these markers

may be of critical value in targeting bisphosphonate

treatment in the future.

Osteoprotegerin. Osteoclast differentiation has been

reported to be regulated by a complex signaling

system involving receptor activator of nuclear factor

(NF)kb(RANK), osteoprotegerin and RANK ligand

Prognostic value of serum markers for prostate cancer 69

(RANKL). In healthy bone, the binding of RANKL

to the receptor activator of RANK stimulates osteo-

clast formation and activation. Excessive bone re-

sorption is controlled by osteoprotegerin (OPG), a

soluble lipoprotein present in serum. OPG acts as a

decoy receptor for RANKL, neutralizing its interac-

tion with RANK and thereby suppressing osteoclas-

togenesis. Increased levels of OPG have been

observed during bone turnover and patients with

advanced prostate cancer have higher levels of serum

OPG than those with earlier-stage disease [100�/

102]. This suggests that increased expression of

OPG may correlate with progression of prostate

cancer to bone. In a recent study, Jung et al. [101]

evaluated 10 serum markers of bone turnover in a

case mix of patients with prostate cancer and

matched controls. Of all markers of bone formation

and resorption measured, OPG and tartrate-resis-

tant acid phosphatase isoenzyme 5b (TRAP) were

the most significant variables predicting bone me-

tastases, with an overall accuracy of 93%.

Alkaline phosphatase. AP is associated with osteoblast

maturation and activity. Although its precise role is

not known, it seems to be important in the initiation

of new bone mineralization [103]. The rate of

remodeling of the bone matrix is higher in metastatic

prostate cancer patients and measurement of this

rate is an indirect marker of metastatic disease. It can

be assessed either by measuring a prominent enzy-

matic activity of the bone-forming (osteoblast) or

-resorbing (osteoclast) cells, or by measuring bone-

matrix components that are released into the circu-

lation during formation or resorption [104]. Patients

with a serum AP flare after orchidectomy may

benefit from early chemotherapy [105]. Whilst

bone AP is a reliable indicator of bone turnover,

and was thought to be an independent prognostic

marker in metastatic prostate cancer, more recent

data [101,106] suggest conflicting evidence regard-

ing its independent value.

The usefulness of markers of bone turnover in the

management of prostate cancer is complicated by

the fact that ageing, hormonal therapy and other

benign bone pathologies, such as Paget’s disease of

the bone, all affect bone turnover. Indeed, it has

been shown in various studies [107] that prostate

cancer patients have decreased bone mass and

increased bone turnover following androgen depri-

vation therapy. Osteoblastic lesions that respond to

therapy in the form of hormones, chemotherapy or

radiation tend to show declining values of both

formation and resorption markers [108].

Bisphosphonates are bone-matrix stabilizing

agents which have been shown to reduce the overall

progression of bony involvement in numerous can-

cers and the incidence of SREs [98,99]. It has

been indicated in some studies [104,109] that a

high level of resorption markers after treatment may

reflect resistance of disease to treatment with bi-

sphosphonates. This may be of value in targeting

treatment in patients likely to benefit from bisphos-

phonates.

Table II summarizes markers that are of potential

value in the management of prostate cancer, and also

shows recommendations for the type of sample to be

used.

Use of serum markers to evaluate the

probability of finding prostate cancer at biopsy

Prevalence of prostate cancer in relation to serum PSA

concentration

The probability of prostate cancer detection in men

with low PSA levels was recently studied in the

Prostate Cancer Prevention Trial. In that study,

2950 men aged ]/62 years (estimated median age

69 years) with a PSA level ofB/4 mg/l were biopsied

[110]. Prostate cancer was found in 15% of all men,

in 9% of those with PSAB/1 mg/l, in 17% when PSA

was 1�/2 mg/l, in 24% when PSA was 2�/3 mg/l and in

27% when PSA was 3�/4 mg/l (Table III). Based on

the distribution of PSA values in 55�/67-year-old

men participating in the Finnish prostate cancer

screening study and the frequency of positive biop-

sies in various studies, it can be calculated that

prostate cancer will be found in 16% of all men aged

60�/75 years should they undergo prostate biopsies.

Interestingly, this is similar to the lifetime probability

of being diagnosed with prostate cancer in the USA,

which is presently 17% [4], and corresponds to 25�/

30% of the known prostate cancer prevalence

detected by means of systematic histological exam-

ination of prostates in autopsy studies [2]. It is

notable that :/25% of all the cancers are found in

men with PSAB/1 mg/l, 57% when PSA isB/2 mg/l,

73% when PSA isB/3 mg/l and 81% when PSA isB/4

mg/l (Table III). Thus, most prevalent prostate

cancers occur in men with what have been inter-

preted as ‘‘normal’’ PSA values in recent years and

it appears that, as far as the true prevalence of

the disease is concerned, those cancers found by

PSA-based screening alone represent the tip of a

large iceberg, albeit there is a significant association

with the risk of developing clinical disease (see

below). However, :/10�/12% of the tumors found

in men with PSAB/2 mg/l are high-grade cancers

(Table IV).

70 U.-H. Stenman et al.

Sensitivity of PSA for detection of clinically relevant

prostate cancer

The calculation above shows that only 20�/25% of

the cancers detected by systematic biopsy will be

detected by one round of PSA-based screening using

a cut-off of 3�/4 mg/l. Thus 70�/75% of the poten-

tially biopsy-detectable cancers remain occult. Be-

cause the clinical relevance of occult cancers is not

clear, it is important to estimate the sensitivity of

PSA for detecting whether prostate cancer will

develop during the lifetime of the patient. In a

screening setting, :/8�/12% of men aged 50�/70

years will have a PSA level of�/2.5�/4 mg/l and, in

20�/30% of these, a prostate cancer is detected by

biopsy [22,23,111,112]. The detection rate is in-

creased by repeating the screening. Over a 7-year

period, the cumulative risk of prostate cancer

diagnosis in men with an initial total PSA value of

3�/4 mg/l was 33%, for those with a PSA value of 4�/7

mg/l it was 39%, for those with a PSA value of 7�/10

mg/l it was 50% and for those with a PSA value of�/

10 mg/l it was 77% (Table V) [22]. Thus elevated

serum PSA is a strong indication of the later

development of prostate cancer.

Another estimate of sensitivity can be obtained by

calculating the frequency of interval cancers detected

in a screening program, i.e. the cancers diagnosed

between the screening rounds. With a 4-year screen-

ing interval, the test sensitivity calculated for PSA

with a cut-off level of 4 mg/l was estimated to be�/

90% in the Finnish screening program [24] and 86%

in the Dutch program [113]. Thus nearly all cancers

diagnosed within 4 years after the initiation of

screening were identified by means of elevated PSA

in the initial screening round.

Serum bank studies provide information on how

PSA identifies men who will develop clinical prostate

cancer within a long follow-up period. In a study

Table II. Summary of potentially useful serum markers for prostate cancer.

Disease markers and detection

techniques under investigation Sample type Method Comments

ProPSA, intact-fPSA,

‘‘nicked’’-fPSA,

B-PSA

Plasma

or serum

Research use

immunoassays

Subfractions of fPSA: proportion relative to total PSA may

help to discriminate between

prostate cancer and BPH [53,55�/58]

PSA�/ACT, PSA�/API,

PSA�/AMG

Plasma

or serum

Research use

immunoassays

Complexes between PSA and protease inhibitors. Absolute

concentrations and proportion

of total PSA improve

discrimination between prostate

cancer and BPH [38,40,45�/47]

hK2, free hK2, pro-hK2 Serum Research use

immunoassay

Homologous to PSA, serum

concentrations 50�/100-fold lower.

Elevated in prostate cancer and BPH [61],

potentially useful to detect

extracapsular extension [63]

IGF-1, IGFBP-3 Serum Commercial

immunoassays

available

Elevated levels are associated with later development of

prostate cancer; not useful for

early diagnosis and screening of

prostate cancer [77�/79].

Markers of bone resorption:

CTX;

tartrate-resistant acid

phosphatase; type I

collagen cross-linked

N-telopeptides

Serum,

urine

Commercial

immunoassays

available

Elevated levels are associated with the presence of skeletal

metastases, and can be more

sensitive than PSA.

Elevated levels after treatment

with hormones or

bisphosphonates can predict resistance to

therapy and progression [88�/96]

Markers of bone turnover/

formation:

bone-specific AP;

osteocalcin; type I

procollagen C-propeptides

Serum,

urine

Commercial

immunoassays

available

Elevated levels are associated with the presence of skeletal

metastases, and can predict their development. Levels do not

appear to be affected by bisphosphonates [88�/96]

Other markers of bone activity

(OPG)

Serum Commercial

immunoassays

available

Elevated levels are associated with the presence of skeletal

metastases, disease progression and hormone-refractory

disease [101�/103]

Markers of neuro-endocrine

differentiation: CgA; NSE

Serum Commercial

immunoassays

available

Elevated levels are associated with disease progression, poor

prognosis and

hormone-refractory disease [70�/75]

Prognostic value of serum markers for prostate cancer 71

based on 21 000 men [114], the sensitivity and

specificity for the detection of prostate cancer

developing during a 20-year follow-up period were

44% and 94%, respectively with a PSA cut-off level

of 4 mg/l. In men agedB/65 years at the initial

measurement, the sensitivity and specificity were

even higher, at 93% and 96%, respectively, and

sensitivity was 100% for cancers detected within 5

years. Samples for the study were collected between

1968 and 1976, and follow-up continued until 1991,

i.e. before the PSA era. Thus the results reflect the

development of clinical disease, rather than inciden-

tal cancers detected by means of opportunistic

screening. Other studies with long-term follow-up

[115,116] show that the risk of developing prostate

cancer relative to that of a reference group of men

with a PSA level ofB/1 mg/l increases gradually, with

the increase in PSA level being 20�/40-fold of those

with a PSA level of 4�/10 mg/l and�/100-fold of those

with a level above 10 mg/l. Taken together, these

results show that PSA is a very sensitive marker for

the identification of men who develop clinically

relevant prostate cancer during their lifetime.

Frequency of prostate cancers not expressing PSA

A number of case studies indicate that some

hormone-refractory cancers do not express PSA,

but little information is available on the frequency of

PSA-negative primary tumors. In a study [117] of

PSA in patients who relapsed after radical prosta-

tectomy, none of the 304 cases relapsed without an

increase in PSA. In summary, few primary prostate

cancers do not express PSA, and lack of expression

appears to correlate with an aggressive phenotype of

the disease.

Rate of increase in serum PSA as an indicator of tumor

growth

The growth rate of a prostate cancer correlates with

the rate of increase in serum PSA, which can be

expressed as the doubling time (DT) or the absolute

increase in PSA concentration per year. The DT is

calculated using the formula DT�/log2/[(logPSA2�/

logPSA1)/(T2�/T1)], where PSA1 and PSA2 are the

PSA values measured at times T1 and T2, respec-

tively. Based on the increase in serum PSA of

patients monitored with watchful waiting and in

serum bank studies, the average DT of prostate

cancer has been estimated to be 2�/3 years [6,8,118].

Because tumors are thought to grow exponentially,

the DT provides a more physiological measure of

tumor growth than the absolute increase in concen-

tration (micrograms per liter per year). Long-term

monitoring of serum PSA prior to the diagnosis of

prostate cancer indicates that the PSA DT (or PSA

velocity) reliably identifies men who will develop

detectable prostate cancer [6]. Although initially it

was thought that changes in serum PSA over the 12

Table IV. Probability of finding prostate cancer and the proportion of high-grade prostate cancer at biopsy in relation to serum PSA

concentration [110].

Serum PSA (mg/l) n Positive biopsies; n (%) High-grade cancers; n (%)

0�/0.5 486 32 (6.6) 4/32 (12.5)

0.6�/1.0 791 80 (10.1) 8/80 (10.0)

1.1�/2.0 998 170 (17.0) 20/170 (11.8)

2.1�/3.0 482 115 (23.9) 22/115 (19.1)

3.1�/4.0 193 52 (25.0) 13/52 (25.0)

Table III. Prevalence of prostate cancer detectable by systematic biopsy in men with various concentrations of PSAa.

PSA (mg/l)

No. of men at

risk per 1000

Positive

biopsies (%)

No. of cancers

per 1000 men

Proportion of all

cancers found (%)

0.0�/1.0 462 8.7 40.5 25.5

1.1�/2.0 294 10.1 50.1 31.5

2.1�/3.0 104 17.0 25.0 15.7

3.0�/4.0 52 23.9 13.8 8.7

4.0�/6.0 47 28.0 13.0 8.2

6.1�/10 27 30.0 8.0 5.0

�/10 14 59.0 8.6 5.4

All 159.0

aThe distribution of PSA values is based on data from the Finnish prostate cancer screening study. The frequency of positive biopsies in the

PSA range 0�/4 mg/l is from the study of Thompson et al. [110] and the frequency at higher values represents average values from several

studies in patients aged�/60 years.

72 U.-H. Stenman et al.

months before therapy were not associated with

adverse pathological findings or biochemical recur-

rence [119], the results of a more recent study [20]

suggests that an increase in serum PSA of ]/2 mg/l

over the year prior to radical treatment is a strong

predictor of death from prostate cancer.

Prediction of stage and grade

The stage of prostate cancer is strongly related to

serum PSA concentration [35,120,121]. This is used

to evaluate stage before biopsy as well as the

probability of a positive bone scan [122,123]. The

diagnostic accuracy of PSA can be improved by

measuring the two major forms of circulating PSA.

This can be achieved by measuring free and total

PSA separately, and calculating their ratio (F/T-

PSA). A low F/T-PSA ratio is a strong predictor of

prostate cancer, while a high ratio is an indicator of

BPH as the cause of the elevation in total PSA [37].

In a screening setting, this can reduce the number of

false-positive results by 20�/30%, whilst maintaining

a sensitivity for prostate cancer detection of 90�/95%

[41,124,125].

Tumor grade is also associated with serum PSA,

and a low F/T-PSA has been found to be significantly

associated with high-grade and -stage prostate can-

cer [126,127]. In patients undergoing radical pros-

tatectomy for clinically localized disease, a low F/T-

PSA ratio was found to be a strong independent risk

factor of extracapsular tumor extension (stage pT3)

[128]. Determination of hK2 in serum has also been

found to facilitate identification of aggressive disease

[61], but this has not been confirmed in other

studies [64].

The serum PSA level of a man without prostate

cancer is related to prostate volume [129], the ratio

of which is called the PSA density [130]. Because

prostate cancer releases more PSA into the circula-

tion per gram of tissue than the benign gland, PSA

density increases in cancer and provides an improve-

ment in cancer specificity similar to that obtained

using F/T-PSA. A high PSA density was also found

to predict advanced disease [126,131]. In a multi-

variate analysis [132], both the proportion of free

PSA and prostate volume were independent factors

predicting the presence of prostate cancer at biopsy.

An advantage of the F/T-PSA ratio is that it can be

determined before clinical examination and used to

decide whether the patient needs to be referred to a

urologist [131]. BPH is mostly located in the

transition zone (TZ) of the prostate, and BPH tissue

produces three times more PSA in the circulation

than normal prostatic tissue [35]. It has been shown

[133] that dividing the PSA value by the TZ volume

(PSA TZ density) provides a further improvement in

cancer specificity, but no advantage over PSA

density was observed in a screening study [134].

This may be explained by the known interobserver

variability in measuring the TZ volume accurately in

asymptomatic men with a small prostate.

Combined use of multiple diagnostic variables

The diagnostic accuracy provided by total and free

PSA can be further improved by estimating the

combined impact of marker data and the results of

clinical examinations using either logistic regression

(LR) or artificial neural networks (ANN). The latter

have a greater capacity to take non-linear relation-

ships between variables into account than LR. An

advantage of LR is that it provides an estimate of the

relative contribution of each variable to the diag-

nostic power of the algorithm. It has been shown in

some studies [132,135] that ANN provide a greater

improvement in diagnostic accuracy than LR but, in

a recent large multicenter study [136], ANN and LR

were found to perform equally. Several useful

diagnostic algorithms [132,135�/137] have been

established that can help the clinician to decide

whether it is necessary to refer the patient for

urological consultation.

An abnormal result at digital rectal examination

(DRE) has traditionally been used as a sign of

prostate cancer prompting a prostate biopsy. How-

ever, depending on the proportion of free PSA the

probability that a nodule is associated with cancer at

biopsy may vary 10-fold in men with the same

concentration of total PSA. In a multivariate analy-

sis, a positive DRE is a prognostic factor indepen-

dent of serum PSA and F/T-PSA which, however,

are stronger predictors of prostate cancer [132]. In a

screening setting primarily based on PSA, DRE is of

limited utility and is not cost-effective [138].

Table V. Cumulative risk of prostate cancer in relation to a single PSA measurement�/3 ng/ml [22].

PSA (ng/ml) No. of men Percentage of population No. of cancers Detection rate (%)

3.0�/3.99 267 4.6 89 33.3

4.0�/6.99 265 4.5 103 38.7

7.0�/9.99 60 1.0 30 50.0

]/10 69 1.2 53 76.8

Prognostic value of serum markers for prostate cancer 73

The relative impacts of PSA and the proportion of

free PSA are visualized in Figure 1. A low proportion

of free PSA (7%) is associated with a 10-fold higher

probability of finding prostate cancer compared with

35% free PSA when the total PSA level is 4 mg/l. This

demonstrates the strong impact of free PSA on

prostate cancer diagnosis. The impact of DRE

findings and prostate volume can be estimated using

an Excel formula that can be downloaded from a

website (www.finne.info). The formula is based on

an algorithm which, in addition to free and total

PSA, utilizes prostate volume and the result of DRE

(positive or negative). A positive DRE increases the

probability of prostate cancer at biopsy by about

two-fold and an increase in prostate volume from 25

to 50 ml decreases the risk by :/50% [132,139].

Monitoring of tumor growth during watchful waiting

With a DT of 2 years, a typical screening-detected

tumor with a volume of 1 ml will increase in volume

to 8 ml after 4 years, 64 ml after 8 years, 256 ml after

12 years and 1 l after 16 years. These figures are in

agreement with the observed time of 5�/10 years

from PSA increase to the appearance of clinical

symptoms, and the median time of 17 years until

cancer death observed in a serum bank study

performed during a time when radical prostatectomy

and radiotherapy were not used [8]. Interestingly, a

sharp increase in cancer-specific mortality has been

noted between 15 and 21 years after diagnosis in

patients with initially early-stage low-grade disease

that was not treated with curative intent [140].

When measured before therapy, the PSA DT is�/

2 years in most patients [118], indicating that the

majority of the tumors are fairly indolent and grow

slowly. Patients with tumors judged to be non-

aggressive are often followed up with watchful wait-

ing, and monitoring of serum PSA is an important

part of this strategy. A high risk of progression has

been associated with a DT ofB/2 years [12] and an

absolute increase in serum PSA of�/0.1 mg/l/year

[141]. However, there is considerable physiological

variation in serum PSA over time [142], and

spurious increases in serum PSA during watchful

waiting may be misinterpreted as indicating tumor

growth. Patients on watchful waiting often have a

DT of�/5 years but, depending on the time intervals

used, the median DT may vary between 6 and 12

years in the same patient [143]. Therefore, reliable

estimates of the DT and PSA increase require long-

term follow-up [144].

Several prognostic algorithms for estimating prog-

nosis after biopsy have been established and were

recently reviewed [145]. Tumor grade and stage are

major factors predicting prognosis but the concen-

tration of serum PSA is also an important variable in

these algorithms. Grade alone is a very strong

predictor of cancer-specific death [146] and the

addition of PSA and stage further enhances the

predictive power. Serum PSA concentrations ofB/

10, 10�/20 and�/20 mg/l are used to define prostate

cancers with good, intermediate and poor prognoses,

respectively [147]. The prognostic value of tissue

markers is dealt with in a separate article in this

supplement.

Identification of relapse after radical therapy

After therapy, the patient is monitored by assay of

serum PSA, usually at 3-month intervals during the

first year, every 6 months during the second year and

once-yearly thereafter. After radical prostatectomy of

patients with clinically organ-confined prostate can-

cer, serum PSA reaches undetectable levels within a

couple of weeks. About 35% of these patients

experience a biochemical relapse, i.e. an increase in

serum PSA. Depending on the assay used the cut-off

level used has varied between 0.07 and 0.2 mg/l. In

more than half of cases the relapse is detected within

2 years, but occasionally�/10 years later [117,148].

The median DT of serum PSA after relapse is 7�/12

months and a short DT is a highly significant

indicator of metastatic disease and a stronger in-

dicator of mortality than Gleason score and other

parameters measured at the time of diagnosis

[117,148�/151]. Because PSA DT is a reliable

marker of survival after a relapse, it appears to be a

useful surrogate marker for clinical trials of oncolo-

35302520151050

10

20

30

40

50

60

70

80

PSA 4 µg/lPSA 7 µg/lPSA 10 µg/lPSA 20 µg/l

F/T PSA (%)

Pro

babi

lity

of f

indi

ng p

rost

ate

canc

er o

n bi

opsy

(%

)

Figure 1. Probability of finding prostate cancer at biopsy on the

basis of various concentrations of total and free PSA. The graph is

based on data from the Finnish prostate cancer screening trial

[132]. The values for total and free PSA were determined using

the Wallac AutoDelfia dual-label assay. Because of variation

between different assays, the results are not necessarily valid for

results based on other assays. However, the graph demonstrates

the strong impact of the proportion of free PSA on prostate cancer

risk.

74 U.-H. Stenman et al.

gic drugs [152]. Early salvage radiotherapy improves

the prognosis for patients who relapse after radical

prostatectomy, even in those with a short PSA DT

[153], and frequent monitoring of PSA facilitates

decision making for adjuvant therapy. A low, stable

serum PSA level after radical surgery may be caused

by imperfect surgical technique and residual benign

prostatic tissue at the apex of the prostate, which

does not require additional treatment.

After radiotherapy, serum PSA decreases slowly

and, even in cured patients, serum PSA remains

detectable because benign prostatic tissue is only

partially affected. The nadir value is sometimes

reached several years after therapy, but a short

time to reach the nadir value and a low serum PSA

value are strong indicators of good prognosis [154].

About 30�/40% of patients experience a relapse as

revealed by increasing serum PSA levels. In most

cases endocrine therapy induces a remission, but

hormone resistance typically develops within 2�/4

years. Monitoring of serum PSA facilitates the early

detection of patients who are likely to benefit from

adjuvant endocrine therapy, but patients with a very

slow increase in serum PSA may not benefit from

androgen ablation. The development of metastatic

disease and mortality can be significantly reduced by

radiotherapy combined with adjuvant androgen

suppression for 3 years [155], and recently [156] a

similar effect was achieved with only 6 months of

adjuvant endocrine therapy.

Androgen ablation

PSA expression is under androgen control, and

androgen ablation mostly results in a rapid decrease

in serum PSA, accompanied by a 90% reduction in

the number of epithelial prostatic cells [157], as well

as objective and subjective relief of symptoms.

Although downregulation of PSA expression is not

invariably associated with a reduction in tumor

growth, serum PSA measurement remains a reliable

indicator of tumor progression. The rate of decrease

in PSA concentration during androgen suppression

correlates with the response to therapy and its

predictive power can be further increased by com-

paring the slope of the decrease with the slope of the

increasing level before therapy. A steeper decrease

than increase is a sign of a good response [151].

Tumors that develop androgen independence may

lose PSA expression, and these mostly show NE

differentiation and express NE markers. Elevated

concentrations of NE markers in serum have been

found to correlate with distant metastases but not

with local disease progression [158], and the number

of CgA-positive NE tumor cells correlates with

serum CgA concentration [159]. The two most

commonly used NE markers, CgA and NSE, may

be useful in terms of diagnosis and prognosis in

prostate cancer patients. Moreover, serum measure-

ment of NE markers may offer complementary

information with respect to PSA. CgA is superior

to NSE and could be useful in the follow-up of

patients with advanced disease [160]. There is a

significant correlation between serum CgA and the

extent of NE features, as reflected by the Gleason

score and the stage of the disease. However, no

correlation has been found between serum CgA and

PSA in either localized or metastatic disease [161].

Thus, NE markers may be useful for evaluating

prognosis in prostate cancer patients, especially in

the hormone-refractory state of the disease. The

prognostic value of NE serum markers needs to be

further evaluated in large-scale studies.

Bone markers are also useful for monitoring the

therapy of patients with bone metastases. The recent

development of specific and sensitive biochemical

markers, reflecting the overall rate of bone formation

and resorption, has improved the non-invasive

assessment of bone turnover abnormalities in pa-

tients with prostate cancer. Immunoassays for bone-

specific AP and type I collagen propeptides are

currently the most sensitive markers for assessing

bone formation. The best indices of bone resorption

are the immunoassays for the pyridinoline crosslinks

and related peptides, which can be measured in

urine and more recently in serum.

The most sensitive markers of bone formation and

resorption are markedly increased in patients with

bone metastases, and the levels correlate with the

extent of bone involvement. However, their sensitiv-

ity remains limited, suggesting that they cannot be

used, as yet, as a surrogate for bone scintigraphy in

the diagnosis of bone involvement. A few authors

have suggested that the measurement of bone

markers may be useful in the assessment of response

to endocrine therapy, although available data indi-

cate a lower sensitivity than that of PSA. Additional

longitudinal studies are required to assess the

potential use of changes in bone markers, especially

to identify patients who relapse during treatment

and, more specifically, those who progress to skeletal

metastases. Bone markers are likely to become a

useful and objective tool to monitor bisphosphonate

treatment and individualize therapeutic regimens.

Despite the noticeable stage migration in prostate

cancers detected in the Western world due to the

widespread use of PSA testing, there is an increasing

tendency for patients to receive early androgen

suppression. This is known to lead to loss of bone

mineral density, which can be monitored using

biochemical markers of bone turnover. In addition,

the previous long-standing conception that the life

Prognostic value of serum markers for prostate cancer 75

expectancy of patients with hormone-refractory dis-

ease tends to be in the region of 9�/12 months has

been superseded by recent data [162], and it is

increasingly recognized that these men now live

significantly longer. This increases the demands on

physicians to manage quality of life and provide

adequate palliation to these patients; hence the

potential growing role of bone markers in the

management of advanced prostate cancer.

Conclusions

The widespread use of serum PSA testing has led to

a dramatic increase in the incidence of prostate

cancer. While this facilitates the detection and

treatment of prostate cancer at an early and poten-

tially curable stage, it has not yet led to any

significant decrease in mortality. Because most

prostate cancers grow slowly and cause death on

average 17 years after the first increase in PSA, a

possible reduction in mortality can be expected only

after 5�/10 more years. Meanwhile, many patients

with slowly growing tumors are overtreated but

about one-third of those treated with curative intent

relapse. It is, therefore, important to develop prog-

nostic methods to facilitate even earlier detection of

aggressive tumors while avoiding the detection of

prostate cancer in men who do not benefit from

treatment. These methods should preferably be non-

invasive, i.e. based on determination of markers in

blood or urine. Although it is unlikely that the

perfect serum marker will ever be discovered, it is

obvious that the available markers, e.g. free and total

PSA, can be used much more efficiently to optimize

diagnosis and the selection of treatment.

References

[1] Franks L. Latent carcinoma of the prostate. J Pathol

Bacteriol 1954;68:603�/16.

[2] Sakr WA, Grignon DJ, Haas GP, Heilbrun LK, Pontes JE,

Crissman JD. Age and racial distribution of prostatic

intraepithelial neoplasia. Eur Urol 1996;30:138�/44.

[3] Wilson SS, Crawford ED. Screening for prostate cancer.

Clin Prostate Cancer 2004;3:21�/5.

[4] Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A,

Ward E, et al. Cancer statistics, 2004. CA Cancer J Clin

2004;54:8�/29.

[5] Auvinen A, Rietbergen JB, Denis LJ, Schroder FH, Prorok

PC. Prospective evaluation plan for randomised trials of

prostate cancer screening. The International Prostate

Cancer Screening Trial Evaluation Group. J Med Screen

1996;3:97�/104.

[6] Carter HB, Pearson JD, Metter EJ, Brant LJ, Chan DW,

Andres R, et al. Longitudinal evaluation of prostate-specific

antigen levels in men with and without prostate disease.

JAMA 1992;267:2215�/20.

[7] Pearson JD, Carter HB. Natural history of changes in

prostate specific antigen in early stage prostate cancer. J

Urol 1994;152(5 Pt 2):1743�/8.

[8] Stenman U-H, Hakama M, Knekt P, Aromaa A, Teppo L,

Leinonen J. Serum concentrations of prostate specific

antigen and its complex with a1-antichymotrypsin before

diagnosis of prostate cancer. Lancet 1994;344:1594�/8.

[9] Draisma G, Boer R, Otto SJ, van der Cruijsen IW, Damhuis

RA, Schroder FH, et al. Lead times and overdetection due

to prostate-specific antigen screening: estimates from the

European Randomized Study of Screening for Prostate

Cancer. J Natl Cancer Inst 2003;95:868�/78.

[10] Finnish Cancer Registry. Mean annual numbers of new

cancer cases in 1955�/2001, by primary site and period,

MALES. Available from: www.cancerregistry.fi/v2001/

v20010003i.html

[11] Hoedemaeker RF, Van der Kwast TH, Schroder FH. The

clinical significance of a small focus of well-differentiated

carcinoma at prostate biopsy. BJU Int 2003;92(Suppl

2):92�/6.

[12] Klotz L. Expectant management with selective delayed

intervention for favorable risk prostate cancer. Urol Oncol

2002;7:175�/9.

[13] Carter HB, Walsh PC, Landis P, Epstein JI. Expectant

management of nonpalpable prostate cancer with curative

intent: preliminary results. J Urol 2002;167:1231�/4.

[14] Allaf ME, Carter HB. Update on watchful waiting for

prostate cancer. Curr Opin Urol 2004;14:171�/5.

[15] Zappa M, Ciatto S, Bonardi R, Mazzotta A. Overdiagnosis

of prostate carcinoma by screening: an estimate based on

the results of the Florence Screening Pilot Study. Ann

Oncol 1998;9:1297�/300.

[16] Draisma G, De Koning HJ. MISCAN: estimating lead-time

and over-detection by simulation. BJU Int 2003;92(Suppl

2):106�/11.

[17] Feuer EJ, Merrill RM, Hankey BF. Cancer surveillance

series: interpreting trends in prostate cancer. Part II. Cause

of death misclassification and the recent rise and fall in

prostate cancer mortality. J Natl Cancer Inst

1999;91:1025�/32.

[18] Etzioni R, Penson DF, Legler JM, di Tommaso D, Boer R,

Gann PH, et al. Overdiagnosis due to prostate-specific

antigen screening: lessons from U. S. prostate cancer

incidence trends. J Natl Cancer Inst 2002;94:981�/90.

[19] Holmberg L, Bill-Axelson A, Helgesen F, Salo JO, Folmerz

P, Haggman M, et al. A randomized trial comparing radical

prostatectomy with watchful waiting in early prostate

cancer. N Engl J Med 2002;347:781�/9.

[20] D’Amico AV, Chen MH, Roehl KA, Catalona WJ. Pre-

operative PSA velocity and the risk of death from prostate

cancer after radical prostatectomy. N Engl J Med

2004;351:125�/35.

[21] Catalona WJ, Ramos CG, Carvalhal GF, Yan Y. Lowering

PSA cutoffs to enhance detection of curable prostate

cancer. Urology 2000;55:791�/5.

[22] Hugosson J, Aus G, Lilja H, Lodding P, Pihl CG. Results of

a randomized, population-based study of biennial screening

using serum prostate-specific antigen measurement to

detect prostate carcinoma. Cancer 2004;100:1397�/405.

[23] Schroder FH, Roobol-Bouts M, Vis AN, van der Kwast T,

Kranse R. Prostate-specific antigen-based early detection of

prostate cancer�/-validation of screening without rectal

examination. Urology 2001;57:83�/90.

[24] Auvinen A, Maattanen L, Finne P, Stenman UH, Aro J,

Juusela H, et al. Test sensitivity of prostate-specific antigen

in the Finnish randomised prostate cancer screening trial.

Int J Cancer 2004;111:940�/3.

76 U.-H. Stenman et al.

[25] Stenman U-H, Leinonen J, Zhang WM, Finne P. Prostate-

specific antigen. Semin Cancer Biol 1999;9:83�/93.

[26] Diamandis EP, Yousef GM, Clements J, Ashworth LK,

Yoshida S, Egelrud T, et al. New nomenclature for the

human tissue kallikrein gene family. Clin Chem 2000;

46:1855�/8.

[27] Lundwall A. Characterization of the gene for prostate-

specific antigen, a human glandular kallikrein. Biochem

Biophys Res Commun 1989;161:1151�/9.

[28] Lovgren J, Rajakoski K, Karp M, Lundwall A, Lilja H.

Activation of the zymogen form of prostate-specific antigen

by human glandular kallikrein 2. Biochem Biophys Res

Commun 1997;238:549�/55.

[29] Paju A, Bjartell A, Zhang WM, Nordling S, Borgstrom A,

Hansson J, et al. Expression and characterization of

trypsinogen produced in the human male genital tract.

Am J Pathol 2000;157:2011�/21.

[30] Belanger A, van Halbeek H, Graves HC, Grandbois K,

Stamey TA, Huang L, et al. Molecular mass and carbohy-

drate structure of prostate specific antigen: studies for

establishment of an international PSA standard. Prostate

1995;27:187�/97.

[31] Christensson A, Laurell CB, Lilja H. Enzymatic activity of

prostate-specific antigen and its reactions with extracellular

serine proteinase inhibitors. Eur J Biochem 1990;194:755�/

63.

[32] Zhang W-M, Leinonen J, Kalkkinen N, Dowell B, Stenman

U-H. Purification and characterization of different mole-

cular forms of prostate specific antigen in human seminal

fluid. Clin Chem 1995;41:1567�/73.

[33] Lilja H. A kallikrein-like serine protease in prostatic fluid

cleaves the predominant seminal vesicle protein. J Clin

Invest 1985;76:1899�/903.

[34] Stenman UH. Prostate-specific antigen, clinical use and

staging: an overview. Br J Urol 1997;1:53�/60.

[35] Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS,

Redwine E. Prostate-specific antigen as a serum marker for

adenocarcinoma of the prostate. N Engl J Med 1987;317:

909�/16.

[36] Abrahamsson PA, Lilja H, Falkmer S, Wadstrom LB.

Immunohistochemical distribution of the three predomi-

nant secretory proteins in the parenchyma of hyperplastic

and neoplastic prostate glands. Prostate 1988;12:39�/46.

[37] Stenman UH, Leinonen J, Alfthan H, Rannikko S, Tuhka-

nen K, Alfthan O. A complex between prostate-specific

antigen and alpha 1-antichymotrypsin is the major form of

prostate-specific antigen in serum of patients with prostatic

cancer: assay of the complex improves clinical sensitivity for

cancer. Cancer Res 1991;51:222�/6.

[38] Lilja H, Christensson A, Dahlen U, Matikainen MT,

Nilsson O, Pettersson K, et al. Prostate-specific antigen in

serum occurs predominantly in complex with alpha 1-

antichymotrypsin. Clin Chem 1991;37:1618�/25.

[39] Leinonen J, Lovgren T, Vornanen T, Stenman U-H.

Double-label time-resolved immunofluorometric assay of

prostate-specific antigen and the complex between pros-

tate-specific antigen with a-1-antichymotrypsin. Clin Chem

1993;39:2098�/103.

[40] Pettersson K, Piironen T, Seppala M, Liukkonen L,

Christensson A, Matikainen MT, et al. Free and complexed

prostate-specific antigen (PSA): in vitro stability, epitope

map, and development of immunofluorometric assays for

specific and sensitive detection of free PSA and PSA-alpha

1-antichymotrypsin complex. Clin Chem 1995;41:1480�/8.

[41] Catalona WJ, Smith DS, Wolfert RL, Wang TJ, Ritten-

house HG, Ratliff TL, et al. Evaluation of percentage of

free serum prostate-specific antigen to improve specificity of

prostate cancer screening. JAMA 1995;274:1214�/20.

[42] Stenman U-H, Leinonen J, Zhang W-M, Finne P, Wu P.

The clinical importance of free prostate-specific antigen

(PSA). Curr Opin Urol 1998;8:393�/9.

[43] Lilja H, Haese A, Bjork T, Friedrich MG, Piironen T,

Pettersson K, et al. Significance and metabolism of

complexed and noncomplexed prostate specific antigen

forms, and human glandular kallikrein 2 in clinically

localized prostate cancer before and after radical prosta-

tectomy. J Urol 1999;162:2029�/34; discussion 2034�/5.

[44] Zhu L, Leinonen J, Zhang WM, Finne P, Stenman UH.

Dual-label immunoassay for simultaneous measurement of

prostate-specific antigen (PSA)-alpha(1)-antichymotrypsin

complex together with free or total PSA. Clin Chem

2003;49:97�/103.

[45] Zhang WM, Leinonen J, Kalkkinen N, Stenman U-H.

Prostate-specific antigen forms a complex with and cleaves

alpha 1-protease inhibitor in vitro. Prostate 1997;33:87�/

96.

[46] Zhang W-M, Finne P, Leinonen J, Vesalainen S, Nordling

S, Rannikko S, et al. Characterization and immunological

determination of the complex between prostate-specific

antigen and alpha2-macroglobulin. Clin Chem 1998;44:

2471�/9.

[47] Finne P, Zhang W-M, Auvinen A, Leinonen J, Maattanen

L, Rannikko S, et al. Use of the complex between prostate-

specific antigen and a1-protease inhibitor in screening for

prostate cancer. J Urol 2000;164:1956�/60.

[48] Allard WJ, Zhou Z, Yeung KK. Novel immunoassay for the

measurement of complexed prostate-specific antigen in

serum. Clin Chem 1998;44(6 Pt 1):1216�/23.

[49] Horninger W, Cheli CD, Babaian RJ, Fritsche HA, Lepor

H, Taneja SS, et al. Complexed prostate-specific antigen for

early detection of prostate cancer in men with serum

prostate-specific antigen levels of 2 to 4 nanograms per

milliliter. Urology 2002;60(4 Suppl 1):31�/5.

[50] Lein M, Kwiatkowski M, Semjonow A, Luboldt HJ,

Hammerer P, Stephan C, et al. A multicenter clinical trial

on the use of complexed prostate specific antigen in low

prostate specific antigen concentrations. J Urol 2003;170(4

Pt 1):1175�/9.

[51] Chen Z, Chen H, Stamey TA. Prostate specific antigen in

benign prostatic hyperplasia: purification and characteriza-

tion. J Urol 1997;157:2166�/70.

[52] Mikolajczyk SD, Millar LS, Marker KM, Wang TJ, Ritten-

house HG, Marks LS, et al. Seminal plasma contains

‘‘BPSA,’’ a molecular form of prostate-specific antigen that

is associated with benign prostatic hyperplasia. Prostate

2000;45:271�/6.

[53] Hilz H, Noldus J, Hammerer P, Buck F, Luck M, Huland

H. Molecular heterogeneity of free PSA in sera of patients

with benign and malignant prostate tumors. Eur Urol

1999;36:286�/92.

[54] Peter J, Unverzagt C, Krogh TN, Vorm O, Hoesel W.

Identification of precursor forms of free prostate-specific

antigen in serum of prostate cancer patients by immuno-

sorption and mass spectrometry. Cancer Res 2001;61:957�/

62.

[55] Linton HJ, Marks LS, Millar LS, Knott CL, Rittenhouse

HG, Mikolajczyk SD. Benign prostate-specific antigen

(BPSA) in serum is increased in benign prostate disease.

Clin Chem 2003;49:253�/9.

[56] Nurmikko P, Pettersson K, Piironen T, Hugosson J, Lilja

H. Discrimination of prostate cancer from benign disease

by plasma measurement of intact, free prostate-specific

Prognostic value of serum markers for prostate cancer 77

antigen lacking an internal cleavage site at Lys145-Lys146.

Clin Chem 2001;47:1415�/23.

[57] Mikolajczyk SD, Millar LS, Wang TJ, Rittenhouse HG,

Marks LS, Song W, et al. A precursor form of prostate-

specific antigen is more highly elevated in prostate cancer

compared with benign transition zone prostate tissue.

Cancer Res 2000;60:756�/9.

[58] Piironen T, Lovgren J, Karp M, Eerola R, Lundwall A,

Dowell B, et al. Immunofluorometric assay for sensitive and

specific measurement of human prostatic glandular kallik-

rein (hK2) in serum. Clin Chem 1996;42:1034�/41.

[59] Darson MF, Pacelli A, Roche P, Rittenhouse HG, Wolfert

RL, Saeid MS, et al. Human glandular kallikrein 2

expression in prostate adenocarcinoma and lymph node

metastases. Urology 1999;53:939�/44.

[60] Becker C, Piironen T, Pettersson K, Bjork T, Wojno KJ,

Oesterling JE, et al. Discrimination of men with prostate

cancer from those with benign disease by measurements of

human glandular kallikrein 2 (HK2) in serum. J Urol

2000;163:311�/6.

[61] Recker F, Kwiatkowski MK, Piironen T, Pettersson K,

Huber A, Lummen G, et al. Human glandular kallikrein as

a tool to improve discrimination of poorly differentiated

and non-organ-confined prostate cancer compared with

prostate-specific antigen. Urology 2000;55:481�/5.

[62] Haese A, Graefen M, Steuber T, Becker C, Pettersson K,

Piironen T, et al. Human glandular kallikrein 2 levels in

serum for discrimination of pathologically organ-confined

from locally-advanced prostate cancer in total PSA-levels

below 10 ng/ml. Prostate 2001;49:101�/9.

[63] Partin AW, Catalona WJ, Finlay JA, Darte C, Tindall DJ,

Young CY, et al. Use of human glandular kallikrein 2 for the

detection of prostate cancer: preliminary analysis. Urology

1999;54:839�/45.

[64] Bangma CH, Wildhagen MF, Yurdakul G, Schroder FH,

Blijenberg BG. The value of (�/7, �/5)pro-prostate-specific

antigen and human kallikrein-2 as serum markers for

grading prostate cancer. BJU Int 2004;93:720�/4.

[65] Haese A, Vaisanen V, Finlay JA, Pettersson K, Rittenhouse

HG, Partin AW, et al. Standardization of two immunoas-

says for human glandular kallikrein 2. Clin Chem 2003;49:

601�/10.

[66] Blijenberg BG, Wildhagen MF, Bangma CH, Finlay JA,

Vaisanen V, Schroder FH. Comparison of two assays for

human kallikrein 2. Clin Chem 2003;49:243�/7.

[67] Stenman UH. Standardization as a private enterprise. Clin

Chem 2003;49:535�/6.

[68] Stamey TA, Ekman PE, Blankenstein MA, Cooper EH,

Kontturi M, Lilja H, et al. Tumor markers. Consensus

Conference on Diagnosis and Prognostic Parameters in

Localized Prostate Cancer. Stockholm, Sweden, May 12�/

13, 1993. Scand J Urol Nephrol Suppl 1994;162:73�/87.

[69] Aprikian AG, Cordon-Cardo C, Fair WR, Reuter VE.

Characterization of neuroendocrine differentiation in hu-

man benign prostate and prostatic adenocarcinoma. Cancer

1993;71:3952�/65.

[70] Bubendorf L, Sauter G, Moch H, Schmid HP, Gasser TC,

Jordan P, et al. Ki67 labelling index: an independent

predictor of progression in prostate cancer treated by

radical prostatectomy. J Pathol 1996;178:437�/41.

[71] Kadmon D, Thompson TC, Lynch GR, Scardino PT.

Elevated plasma chromogranin-A concentrations in pro-

static carcinoma. J Urol 1991;146:358�/61.

[72] Tarle M, Rados N. Investigation on serum neurone-specific

enolase in prostate cancer diagnosis and monitoring:

comparative study of a multiple tumor marker assay.

Prostate 1991;19:23�/33.

[73] Abrahamsson PA, Falkmer S, Falt K, Grimelius L. The

course of neuroendocrine differentiation in prostatic carci-

nomas. An immunohistochemical study testing chromogra-

nin A as an ‘‘endocrine marker’’. Pathol Res Pract 1989;

185:373�/80.

[74] Kamiya N, Akakura K, Suzuki H, Isshiki S, Komiya A,

Ueda T, et al. Pretreatment serum level of neuron specific

enolase (NSE) as a prognostic factor in metastatic prostate

cancer patients treated with endocrine therapy. Eur Urol

2003;44:309�/14; discussion 314.

[75] Yu H, Rohan T. Role of the insulin-like growth factor family

in cancer development and progression. J Natl Cancer Inst

2000;92:1472�/89.

[76] Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J,

Wilkinson P, et al. Plasma insulin-like growth factor-I and

prostate cancer risk: a prospective study. Science

1998;279:563�/6.

[77] Stattin P, Rinaldi S, Biessy C, Stenman UH, Hallmans G,

Kaaks R. High levels of circulating insulin-like growth

factor-I increase prostate cancer risk: a prospective study in

a population-based nonscreened cohort. J Clin Oncol

2004;22:3104�/12.

[78] Finne P, Auvinen A, Koistinen H, Zhang WM, Maattanen

L, Rannikko S, et al. Insulin-like growth factor I is not a

useful marker of prostate cancer in men with elevated levels

of prostate-specific antigen. J Clin Endocrinol Metab

2000;85:2744�/7.

[79] Endrizzi J, Optenberg S, Byers R, Thompson IM Jr.

Disappearance of well-differentiated carcinoma of the

prostate: effect of transurethral resection of the prostate,

prostate-specific antigen, and prostate biopsy. Urology

2001;57:733�/6.

[80] Colao A, Marzullo P, Spiezia S, Ferone D, Giaccio A,

Cerbone G, et al. Effect of growth hormone (GH) and

insulin-like growth factor I on prostate diseases: an ultra-

sonographic and endocrine study in acromegaly, GH

deficiency, and healthy subjects. J Clin Endocrinol Metab

1999;84:1986�/91.

[81] Oliver SE, Holly J, Peters TJ, Donovan J, Persad R, Gillatt

D, et al. Measurement of insulin-like growth factor axis

does not enhance specificity of PSA-based prostate cancer

screening. Urology 2004;64:317�/22.

[82] Wu CP, Gu FL. The prostate in eunuchs. Prog Clin Biol

Res 1991;370:249�/55.

[83] Lipsett MB. Estrogen use and cancer risk. J Am Med Assoc

1977;237:1112�/5.

[84] Stattin P, Lumme S, Tenkanen L, Alfthan H, Jellum E,

Hallmans G, et al. High levels of circulating testosterone

are not associated with increased prostate cancer risk: a

pooled prospective study. Int J Cancer 2004;108:418�/24.

[85] Stattin P, Soderberg S, Hallmans G, Bylund A, Kaaks R,

Stenman UH, et al. Leptin is associated with increased

prostate cancer risk: a nested case-referent study. J Clin

Endocrinol Metab 2001;86:1341�/5.

[86] Stattin P, Kaaks R, Johansson R, Gislefoss R, Soderberg S,

Alfthan H, et al. Plasma leptin is not associated with

prostate cancer risk. Cancer Epidemiol Biomarkers Prev

2003;12:474�/5.

[87] Bok RA, Small EJ. Bloodborne biomolecular markers in

prostate cancer development and progression. Nat Rev

Cancer 2002;2:918�/26.

[88] Gutman E, Sproul E, Gutman A. Significance of increased

phosphatase activity of bone at the site of osteoblastic

metastases secondary to carcinoma of the prostate gland.

Am Cancer 1936;28:485�/95.

78 U.-H. Stenman et al.

[89] Huggins C, Hodges C. The effect of castration, of estrogen

and androgen injection on serum phosphatases in meta-

static carcinoma of the prostate. Cancer Res 1941;1:293�/7.

[90] Wymenga LF, Groenier K, Schuurman J, Boomsma JH,

Elferink RO, Mensink HJ. Pretreatment levels of urinary

deoxypyridinoline as a potential marker in patients with

prostate cancer with or without bone metastasis. BJU Int

2001;88:231�/5.

[91] Koizumi M, Yonese J, Fukui I, Ogata E. The serum level of

the amino-terminal propeptide of type I procollagen is a

sensitive marker for prostate cancer metastasis to bone.

BJU Int 2001;87:348�/51.

[92] Diaz-Martin MA, Traba ML, De La Piedra C, Guerrero R,

Mendez-Davila C, De La Pena EG. Aminoterminal pro-

peptide of type I collagen and bone alkaline phosphatase in

the study of bone metastases associated with prostatic

carcinoma. Scand J Clin Lab Invest 1999;59:125�/32.

[93] Garnero P, Buchs N, Zekri J, Rizzoli R, Coleman RE,

Delmas PD. Markers of bone turnover for the management

of patients with bone metastases from prostate cancer. Br J

Cancer 2000;82:858�/64.

[94] Kylmala T, Tammela TL, Risteli L, Risteli J, Kontturi M,

Elomaa I. Type I collagen degradation product (ICTP)

gives information about the nature of bone metastases and

has prognostic value in prostate cancer. Br J Cancer

1995;71:1061�/4.

[95] Maeda H, Koizumi M, Yoshimura K, Yamauchi T, Kawai

T, Ogata E. Correlation between bone metabolic markers

and bone scan in prostatic cancer. J Urol 1997;157:539�/

43.

[96] Noguchi M, Noda S. Pyridinoline cross-linked carboxy-

terminal telopeptide of type I collagen as a useful marker for

monitoring metastatic bone activity in men with prostate

cancer. J Urol 2001;166:1106�/10.

[97] Berruti A, Dogliotti L, Bitossi R, Fasolis G, Gorzegno G,

Bellina M, et al. Incidence of skeletal complications in

patients with bone metastatic prostate cancer and hormone

refractory disease: predictive role of bone resorption and

formation markers evaluated at baseline. J Urol

2000;164:1248�/53.

[98] Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner

P, Lacombe L, et al. A randomized, placebo-controlled trial

of zoledronic acid in patients with hormone-refractory

metastatic prostate carcinoma. J Natl Cancer Inst 2002;

94:1458�/68.

[99] Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner

P, Lacombe L, et al. Long-term efficacy of zoledronic acid

for the prevention of skeletal complications in patients with

metastatic hormone-refractory prostate cancer. J Natl

Cancer Inst 2004;96:879�/82.

[100] Brown JM, Vessella RL, Kostenuik PJ, Dunstan CR, Lange

PH, Corey E. Serum osteoprotegerin levels are increased in

patients with advanced prostate cancer. Clin Cancer Res

2001;7:2977�/83.

[101] Jung K, Lein M, Stephan C, Von Hosslin K, Semjonow A,

Sinha P, et al. Comparison of 10 serum bone turnover

markers in prostate carcinoma patients with bone meta-

static spread: diagnostic and prognostic implications. Int J

Cancer 2004;111:783�/91.

[102] Eaton CL, Wells JM, Holen I, Croucher PI, Hamdy FC.

Serum osteoprotegerin (OPG) levels are associated with

disease progression and response to androgen ablation in

patients with prostate cancer. Prostate 2004;59:304�/10.

[103] Bellows CG, Aubin JE, Heersche JN. Initiation and

progression of mineralization of bone nodules formed in

vitro: the role of alkaline phosphatase and organic phos-

phate. Bone Miner 1991;14:27�/40.

[104] Fontana A, Delmas PD. Markers of bone turnover in bone

metastases. Cancer 2000;88(12 Suppl):2952�/60.

[105] Pelger RC, Lycklama ANGB, Zwinderman AH, Hamdy

NA. Estramustine phosphate combined with orchidectomy

as first-line therapy in patients with prostate carcinoma.

Effect of age on survival. Cancer 2002;94:2596�/601.

[106] Petrioli R, Rossi S, Caniggia M, Pozzessere D, Messinese S,

Sabatino M, et al. Analysis of biochemical bone markers as

prognostic factors for survival in patients with hormone-

resistant prostate cancer and bone metastases. Urology

2004;63:321�/6.

[107] Stoch SA, Parker RA, Chen L, Bubley G, Ko YJ, Vincelette

A, et al. Bone loss in men with prostate cancer treated with

gonadotropin-releasing hormone agonists. J Clin Endocri-

nol Metab 2001;86:2787�/91.

[108] Meijer WG, van der Veer E, Willemse PH. Biochemical

parameters of bone metabolism in bone metastases of solid

tumors [review]. Oncol Rep 1998;5:5�/21.

[109] Delmas PD. Markers of bone turnover for monitoring

treatment of osteoporosis with antiresorptive drugs. Osteo-

poros Int 2000;11(Suppl 6):S66�/76.

[110] Thompson IM, Pauler DK, Goodman PJ, Tangen CM,

Lucia MS, Parnes HL, et al. Prevalence of prostate cancer

among men with a prostate-specific antigen levelB/or �/4.0

ng per milliliter. N Engl J Med 2004;350:2239�/46.

[111] Catalona WJ, Smith DS, Ratliff TL, Dodds KM, Coplen

DE, Yuan JJ, et al. Measurement of prostate-specific

antigen in serum as a screening test for prostate cancer.

N Engl J Med 1991;324:1156�/61.

[112] Maattanen L, Auvinen A, Stenman UH, Tammela T,

Rannikko S, Aro J, et al. Three-year results of the Finnish

prostate cancer screening trial. J Natl Cancer Inst 2001;

93:552�/3.

[113] Van der Cruijsen-Koeter IW, van der Kwast TH, Schroder

FH. Interval carcinomas in the European Randomized

Study of Screening for Prostate Cancer (ERSPC)*/Rot-

terdam. J Natl Cancer Inst 2003;95:1462�/6.

[114] Hakama M, Stenman UH, Aromaa A, Leinonen J, Haku-

linen T, Knekt P. Validity of the prostate specific antigen

test for prostate cancer screening: followup study with a

bank of 21,000 sera in Finland. J Urol 2001;166:2189�/91;

discussion 2191�/2.

[115] Gann PH, Hennekens CH, Stampfer MJ. A prospective

evaluation of plasma prostate-specific antigen for detection

of prostatic cancer. JAMA 1995;273:289�/94.

[116] Antenor JA, Han M, Roehl KA, Nadler RB, Catalona WJ.

Relationship between initial prostate specific antigen level

and subsequent prostate cancer detection in a longitudinal

screening study. J Urol 2004;172:90�/3.

[117] Pound CR, Partin AW, Eisenberger MA, Chan DW,

Pearson JD, Walsh PC. Natural history of progression after

PSA elevation following radical prostatectomy. JAMA

1999;281:1591�/7.

[118] Schmid HP, McNeal JE, Stamey TA. Observations on the

doubling time of prostate cancer. The use of serial prostate-

specific antigen in patients with untreated disease as a

measure of increasing cancer volume. Cancer 1993;71:

2031�/40.

[119] Freedland SJ, Dorey F, Aronson WJ. Preoperative PSA

velocity and doubling time do not predict adverse patho-

logic features or biochemical recurrence after radical

prostatectomy. Urology 2001;57:476�/80.

[120] Hudson MA, Bahnson RR, Catalona WJ. Clinical use of

prostate specific antigen in patients with prostate cancer. J

Urol 1989;142:1011�/7.

[121] Lange PH, Ercole CJ, Lightner DJ, Fraley EE, Vessella R.

The value of serum prostate specific antigen determinations

Prognostic value of serum markers for prostate cancer 79

before and after radical prostatectomy. J Urol 1989;141:

873�/9.

[122] Partin AW, Carter HB, Chan DW, Epstein JI, Oesterling

JE, Rock RC, et al. Prostate specific antigen in the staging

of localized prostate cancer: influence of tumor differentia-

tion, tumor volume and benign hyperplasia. J Urol 1990;

143:747�/52.

[123] Oesterling JE, Martin SK, Bergstralh EJ, Lowe FC. The

use of prostate-specific antigen in staging patients with

newly diagnosed prostate cancer. JAMA 1993;269:57�/60.

[124] Virtanen A, Gomari M, Kranse R, Stenman U-H. Estima-

tion of prostate cancer probability by logistic regression:

free and total prostate-specific antigen, digital rectal

examination, and heredity are significant variables. Clin

Chem 1999;45:987�/94.

[125] Hugosson J, Aus G, Bergdahl S, Fernlund P, Frosing R,

Lodding P, et al. Population-based screening for prostate

cancer by measuring free and total serum prostate-specific

antigen in Sweden. BJU Int 2003;92(Suppl 2):39�/43.

[126] Southwick PC, Catalona WJ, Partin AW, Slawin KM,

Brawer MK, Flanigan RC, et al. Prediction of post-radical

prostatectomy pathological outcome for stage T1c prostate

cancer with percent free prostate specific antigen: a

prospective multicenter clinical trial. J Urol 1999;162:

1346�/51.

[127] Raaijmakers R, Wildhagen MF, Ito K, Paez A, de Vries SH,

Roobol MJ, et al. Prostate-specific antigen change in the

European Randomized Study of Screening for Prostate

Cancer, section Rotterdam. Urology 2004;63:316�/20.

[128] Aus G, Becker C, Lilja H, Khatami A, Pihl CG, Hugosson

J. Free-to-total prostate-specific antigen ratio as a predictor

of non-organ-confined prostate cancer (stage pT3). Scand J

Urol Nephrol 2003;37:466�/70.

[129] Babaian RJ, Fritsche HA, Evans RB. Prostate-specific

antigen and prostate gland volume: correlation and clinical

application. J Clin Lab Anal 1990;4:135�/7.

[130] Benson MC, Whang IS, Pantuck A, Ring K, Kaplan SA,

Olsson CA, et al. Prostate specific antigen density: a means

of distinguishing benign prostatic hypertrophy and prostate

cancer. J Urol 1992;147(3 Pt 2):815�/6.

[131] Catalona WJ, Southwick PC, Slawin KM, Partin AW,

Brawer MK, Flanigan RC, et al. Comparison of percent

free PSA, PSA density, and age-specific PSA cutoffs for

prostate cancer detection and staging. Urology 2000;56:

255�/60.

[132] Finne P, Finne R, Auvinen A, Juusela H, Aro J, Maattanen

L, et al. Predicting the outcome of prostate biopsy in

screen-positive men by a multilayer perceptron network.

Urology 2000;56:418�/22.

[133] Kikuchi E, Nakashima J, Ishibashi M, Ohigashi T, Asakura

H, Tachibana M, et al. Prostate specific antigen adjusted

for transition zone volume: the most powerful method for

detecting prostate carcinoma. Cancer 2000;89:842�/9.

[134] Horninger W, Rogatsch H, Reissigl A, Volgger H, Klocker

H, Hobisch A, et al. Correlation between preoperative

predictors and pathologic features in radical prostatectomy

specimens in PSA-based screening. Prostate 1999;40:56�/

61.

[135] Djavan B, Remzi M, Zlotta A, Seitz C, Snow P, Marberger

M. Novel artificial neural network for early detection of

prostate cancer. J Clin Oncol 2002;20:921�/9.

[136] Finne P, Finne R, Bangma C, Hugosson J, Hakama M,

Auvinen A, et al. Algorithms based on prostate-specific

antigen (PSA), free PSA, digital rectal examination and

prostate volume reduce false-positive PSA results in pros-

tate cancer screening. Int J Cancer 2004;111:310�/5.

[137] Stephan C, Cammann H, Semjonow A, Diamandis EP,

Wymenga LF, Lein M, et al. Multicenter evaluation of an

artificial neural network to increase the prostate cancer

detection rate and reduce unnecessary biopsies. Clin Chem

2002;48:1279�/87.

[138] Vis AN, Kranse R, Roobol M, van der Kwast TH, Schroder

FH. Serendipity in detecting disease in low prostate-

specific antigen ranges. BJU Int 2002;89:384�/9.

[139] Finne P, Auvinen A, Aro J, Juusela H, Maattanen L,

Rannikko S, et al. Estimation of prostate cancer risk on the

basis of total and free prostate-specific antigen, prostate

volume and digital rectal examination. Eur Urol

2002;41:619�/27.

[140] Johansson JE, Andren O, Andersson SO, Dickman PW,

Holmberg L, Magnuson A, et al. Natural history of early,

localized prostate cancer. JAMA 2004;291:2713�/9.

[141] Fang J, Metter EJ, Landis P, Carter HB. PSA velocity for

assessing prostate cancer risk in men with PSA levels

between 2.0 and 4.0 ng/ml. Urology 2002;59:889�/93;

discussion 893�/4.

[142] Komatsu K, Wehner N, Prestigiacomo AF, Chen Z,

Stamey TA. Physiologic (intraindividual) variation of

serum prostate-specific antigen in 814 men from a screen-

ing population. Urology 1996;47:343�/6.

[143] Ross PL, Mahmud S, Stephenson AJ, Souhami L, Tanguay

S, Aprikian AG. Variations in PSA doubling time in

patients with prostate cancer on ‘‘watchful waiting’’: value

of short-term PSADT determinations. Urology 2004;64:

323�/8.

[144] Ito K, Yamamoto T, Ohi M, Takechi H, Kurokawa K,

Suzuki K, et al. Natural history of PSA increase with and

without prostate cancer. Urology 2003;62:64�/9.

[145] Ross PL, Scardino PT, Kattan MW. A catalog of prostate

cancer nomograms. J Urol 2001;165:1562�/8.

[146] Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Compet-

ing risk analysis of men aged 55 to 74 years at diagnosis

managed conservatively for clinically localized prostate

cancer. JAMA 1998;280:975�/80.

[147] D’Amico AV, Moul J, Carroll PR, Sun L, Lubeck D, Chen

MH. Cancer-specific mortality after surgery or radiation for

patients with clinically localized prostate cancer managed

during the prostate-specific antigen era. J Clin Oncol

2003;21:2163�/72.

[148] Pruthi RS, Johnstone I, Tu IP, Stamey TA. Prostate-

specific antigen doubling times in patients who have failed

radical prostatectomy: correlation with histologic charac-

teristics of the primary cancer. Urology 1997;49:737�/42.

[149] Patel A, Dorey F, Franklin J, deKernion JB. Recurrence

patterns after radical retropubic prostatectomy: clinical

usefulness of prostate specific antigen doubling times and

log slope prostate specific antigen. J Urol 1997;158:1441�/

5.

[150] Okotie OT, Aronson WJ, Wieder JA, Liao Y, Dorey F, De

KJ, et al. Predictors of metastatic disease in men with

biochemical failure following radical prostatectomy. J Urol

2004;171(6 Pt 1):2260�/4.

[151] D’Amico AV, Moul JW, Carroll PR, Cote K, Sun L,

Lubeck D, et al. Intermediate end point for prostate

cancer-specific mortality following salvage hormonal ther-

apy for prostate-specific antigen failure. J Natl Cancer Inst

2004;96:509�/15.

[152] Kelloff GJ, Coffey DS, Chabner BA, Dicker AP, Guyton

KZ, Nisen PD, et al. Prostate-specific antigen doubling

time as a surrogate marker for evaluation of oncologic drugs

to treat prostate cancer. Clin Cancer Res 2004;10:3927�/

33.

80 U.-H. Stenman et al.

[153] Stephenson AJ, Shariat SF, Zelefsky MJ, Kattan MW,

Butler EB, Teh BS, et al. Salvage radiotherapy for recurrent

prostate cancer after radical prostatectomy. JAMA

2004;291:1325�/32.

[154] Cavanaugh SX, Kupelian PA, Fuller CD, Reddy C,

Bradshaw P, Pollock BH, et al. Early prostate-specific

antigen (PSA) kinetics following prostate carcinoma radio-

therapy: prognostic value of a time-and-PSA threshold

model. Cancer 2004;101:96�/105.

[155] Bolla M, Gonzalez D, Warde P, Dubois JB, Mirimanoff

RO, Storme G, et al. Improved survival in patients with

locally advanced prostate cancer treated with radiotherapy

and goserelin. N Engl J Med 1997;337:295�/300.

[156] D’Amico AV, Manola J, Loffredo M, Renshaw AA,

DellaCroce A, Kantoff PW. 6-month androgen suppression

plus radiation therapy vs radiation therapy alone for

patients with clinically localized prostate cancer: a rando-

mized controlled trial. JAMA 2004;292:821�/7.

[157] Hansson J, Abrahamsson PA. Neuroendocrine pathogen-

esis in adenocarcinoma of the prostate. Ann Oncol

2001;12(Suppl 2):S145�/52.

[158] Cussenot O, Villette JM, Valeri A, Cariou G, Desgrand-

champs F, Cortesse A, et al. Plasma neuroendocrine

markers in patients with benign prostatic hyperplasia and

prostatic carcinoma. J Urol 1996;155:1340�/3.

[159] Angelsen A, Syversen U, Haugen OA, Stridsberg M,

Mjolnerod OK, Waldum HL. Neuroendocrine differentia-

tion in carcinomas of the prostate: do neuroendocrine

serum markers reflect immunohistochemical findings?

Prostate 1997;30:1�/6.

[160] Berruti A, Dogliotti L, Mosca A, Bellina M, Mari M, Torta

M, et al. Circulating neuroendocrine markers in patients

with prostate carcinoma. Cancer 2000;88:2590�/7.

[161] Bollito E, Berruti A, Bellina M, Mosca A, Leonardo E,

Tarabuzzi R, et al. Relationship between neuroendocrine

features and prognostic parameters in human prostate

adenocarcinoma. Ann Oncol 2001;12(Suppl 2):S159�/

64.

[162] Oefelein MG, Agarwal PK, Resnick MI. Survival of

patients with hormone refractory prostate cancer in the

prostate specific antigen era. J Urol 2004;171:1525�/8.

From the WHO International Consultation

Prediction of Patient Outcome in Prostate Cancer

Prognostic Factors 2004

Stockholm, Sweden, 9�/12 September, 2004

Prognostic value of serum markers for prostate cancer 81