CLIN.
CHEM.
26/13,
1821-1824
(1980)
Radioimmunoassayof Creatine KinaseB-lsoenzymesin Serum of Patients
with Azotemia, ObstructiveUropathy,or Carcinomaof the Prostateor
Bladder
Henry
A. Homburger, Sheldon A. Miller, and Greg L. Jacob
We measured the concentrations
of creatine kinase Bisoenzymes by radioimmunoassay
in 271 serum specimens from patients with azotemia, benign prostatic hyperplasia, adenocarcinoma of the prostate, and transitional
cell carcinoma of the bladder. There was no correlation
between the concentrations
of B-isoenzymes and creatinine in the sera of azotemic patients. Above-normal
tantly with azotemia,
prostatitis,
urinary tract infection,
or
carcinoma
of the bladder. Accordingly,
it is important
in assessing the specificity
of CK-BB as a possible
marker
for
carcinoma
of the prostate
to determine
whether
concentrations of B-isoenzymes
are increased
in association
with these
diseases.
concentrations of B-isoenzymes were found in sera from
three patients with acute renal failure, but in only two of
28 specimens from patients with chronic renal failure.
Materials and Methods
Above-normal concentrations of B-isoenzymes also were
found in sera from three of 18 patients with untreated
carcinoma of the prostate, 10 of 25 patients with treated
carcinoma, 20 of 135 patients with benign prostatic hy-
The concentrations
of B-isoenzymes
in sera were measured
by a double-antibody
radioimmunoassay
that we have described previously
(10). Briefly, we added ‘I-labeled
BB
standard,
100 pg in 100 L of tris(hydroxymethyl)methyl.
amine (Tris)-glycine
buffer; unlabeled
BB standard
in 100
iL of Tris-glycine
buffer or serum; 100 zL of a 1000-fold
dilution of rabbit antibody to the B polypeptide; and 300 L
of Tris-glycmne buffer to 10 X 75mm disposable test tubes. We
incubated
the tubes overnight
at 4 #{176}C,
then added burro
anti-rabbit IgG (300 tL) and normal rabbit serum (100 iL of
perplasia, and 10 of 33 patients with transitional cell carcinoma of the bladder. An above-normal concentration of
B-isoenzymes in serum had a low predictive value for
adenocarcinoma of the prostate, was not a sensitive indicator of the presence of carcinoma, and was noted
paradoxically in six patients with treated carcinoma who
had normal
acid phosphatase
activities
in serum.
We
conclude that routine measurement of B-isoenzymes is
not useful to establish the diagnosis of adenocarcinoma
of the prostate.
AddItIonal Keyphrases:
cut-off values
cancer
urinary-tract disease
The BB isoenzyme (CK-BB) of creatine kinase (EC 2.7.3.2)
is not detectable
in the sera of healthy individuals
except by
radioimmunoassay
(1, 2). It is of interest,
therefore,
that
several
amounts
investigators
have
(or above-normal
described
concentrations)
finding
detectable
of this isoenzyme
in the sera of patients with various diseases of the genitourinary tract (3-7).
Of particular
interest are those preliminary
reports that indicate that CK-BB may be a useful marker for
adenocarcinoma
of the prostate
(3, 8). It is also a subject
of
controversy
whether above-normal
concentrations
of CK-BB
are present in the sera of patients with chronic renal failure.
Some authors
have claimed that the apparent
isoenzyme
in
these sera is an artifact
observed
when the isoenzymes
are
measured
by fluorescence
(9).
We have developed
a radioimmunoassay
specific for the
isoenzymes
of CK that contain
the B polypeptide
subunit
(B-isoenzymes)
(10); here, we report
on the occurrence
of
above-normal
concentrations
of B-isoenzymes
in the sera of
patients
with azotemia,
obstruction
of the urinary tract due
to benign
prostatic
hyperplasia,
adenocarcinoma
of the
prostate,
and transitional
cell carcinoma
of the bladder. We
thought it necessary
to study patients with each of these diseases because carcinoma
of the prostate
may occur concomi-
Department
of Laboratory
Medicine, Section of Clinical Chemistry,
Mayo Clinic, 200 First St. Southwest,
Rochester,
MN 55901.
Received June 16, 1980; accepted
Aug. 22, 1980.
Analytical
Methods
a 20-fold dilution) and incubated
for 4 h at room
We separated
standard
Bound
the bound,
counts
labeled
were measured
temperature.
by centrifugation.
with an automatic
gamma
counting
system and the concentrations
of B-isoenzymes
calculated
from a logit-log transformation
in terms of B/B0
with a programmable
desk-top
calculator.
All tests were
performed
in duplicate.
The activities
of total and L(+)-tartrate-inhibitable
acid
phosphatase
(EC 3.1.3.2) in test sera were measured
by the
method of Kind and King (11) at pH 5.0 and 37.5 #{176}C
with
phenyl
disodium
phosphate
substrate
and with a triplechannel AutoAnalyzer
(Technicon
Instruments,
Tarrytown,
NY, 10591). Creatinine
in test sera was measured by the alkaline picrate method of Raabo and Wallhoe-Hansen
(12)
with a Vickers
Multi-channel
300 analyzer (Vickers American
Medical, Whitehouse
Station,
NJ 08889).
Patient Selection
and Analysis of Results
We measured
the concentrations
of B-isoenzymes
in serum
specimens
from 60 patients
with azotemia,
defined as a concentration
of creatinine
in serum exceeding
15 mg/L, 135
patients
with urinary tract obstruction
secondary
to benign
prostate
hyperplasia,
43 patients
with histologically
proven
adenocarcinoma
of the prostate,
and 33 patients
with histologically proven transitional
cell carcinoma
of the bladder.
Serum specimens
were stored at 4 #{176}C
for not more than 24 h
and thereafter
at -20 #{176}C
until analysis.
We reviewed the medical histories of patients with azotemia
to determine
the cause of renal failure and the duration
of
azotemia. Azotemia was defined as acute if of less than four
weeks’ duration. We reviewed the medical histories and reof microscopic
examinations
of resected prostatic
tissue
from patients with benign prostatic
hyperplasia
for evidence
of other diseases,
including
infection
of the urinary tract or
prostatitis.
The urinary tract was considered
to be infected
if urine culture demonstrated
the presence of at least i0 organisms per milliliter,
or if antibiotic
treatment
was prescribed
ports
CLINICAL CHEMISTRY, Vol. 26, No. 13, 1980
1821
Table 1. Laboratory Data Base
No. patIents wIth
B-Iso.nzym.s
>15
B-Isoenzymes, g/L
Rang.; x (SD)
PatIent group, n
Azotemic
patients,
60
(rang.,
6.0-54.8;
ag/L)
15
13.7 (6.5)
(15. 1-54.8)
Benign prostatic
hyperplasia,
135
without
concomitant
2.5-21.3;
with concomitant
diseases, 30
4.8-64.3;
(15.8-64.3)
15.6 (10.3)
3
(21.4-51.9)
10
(15.2-139.0)
4.6-51.9;
12.8(11.3)
5.0-139.0;
18.4 (26.2)
Transitional cell carcinoma of bladder, 335
Stage I,14
Stage II,12
3
13.3 (6.4)
(15.5-26.6)
5.9-29.2
13.7 (7.4)
(15.4-29.2)
3
4
(15.1-22.0)
14.5(5.5)
a
See text for definition of Stages.
16.5
Septicemia;
16
22.3
Interstitial
16
16.3
17
20.5
Aortic stenosis; syncope
Rectal prolapse; resection
17
16.5
colon
Fungal endocarditis;
amphotericin B
18
16.6
Carcinomatosis,ovarian primary
18
21.3
Metastatic liposarcoma
19
20.5
Carcinomatosls,breast primary
20
15.1 Recent cardiac surgery; urinary tract
28
16.7 Carcinomatosis,
36
38
16.5
54.8
55
17.3 Acute renal failure; urinary tract infection
16.3 Acute renal failure; Wegener’s
clinically. We reviewed the medical
records of patients
with adenocarcinoma
of the prostate
to
determine
the stage of disease (A, B, C, or D), as determined
clinically
and by histopathological
examination
of resected
tissues (13); the presence of concomitant
diseases; and prior
treatment
with radiation,
orchiectomy,
hormonal
therapy, or
cytotoxic
chemotherapy.
We classified
patients
with transitional cell carcinoma
according to the extent of tumor spread
as in situ (I), infiltrative
carcinoma
with local extension
(II),
or infiltrative
carcinoma
with local extension
and distant
metastases
(III). We recorded all other medical and surgical
diagnoses
in all cases.
To assess the usefulness
of CK-BB as a possible tumor
146
marker,
17.6
we calculated
renal primary
Chronic
renal failure; diabetes mellitus
Renal allograft; acute tubular necrosis
granulomatosis
Chronic renal failure; diabetes mellitus
the diagnostic
sensitivity,
value of an above-normal
specificity,
concentration
of
B-isoenzymes
for carcinoma
of the prostate
by the formulas
of Vecchio (14). The diseased
group comprised
all patients
with carcinoma
of the prostate.
The healthy group was all
patients with benign prostatic
hyperplasia;
results from this
group were evaluated
both before and after excluding
those
patients
with concomitant
diseases. We used 15 Lg/L as the
cutoff concentration
of B-isoenzymes
for classification
of
above-normal results. This cutoff concentration was based on
the results of concurrent
measurements
on 45 ostensibly
healthy adults. The concentrations
of B-isoenzymes in this
group were distributed
normally,
viation, and range of 10.2, 1.6, and
We used 7 U/L, with at least 26%
as the cutoff concentration
for an
phosphatase
activity in serum. In
with a mean, standard
de7.6-14.4 ig/L, respectively.
tartrate-inhibitable
activity,
above-normal
prostatic acid
a prior study of 438 healthy
G.A., and Elvebach,
L., unpublished
data) 8
2.5%
Results
24
-J
Table 1 shows the data base of all patients,
segregated
by
groups. Figure 1 shows the results of measurements
by radioimmunoassay
of B-isoenzymes
in the sera of patients with
N
Oi
18
#{149}
- #{149}
:
.
azotemia.
Fifteen
of 60 patients
in this group had abovenormal concentrations
of B-isoenzymes,
but no relationship
between
the concentrations
of these two analytes
was demonstrable by rank order correlation analysis (r = -0.131, p
#{149}
s#{149}
12
.
e
= not significant).
Table 2 lists the clinical diagnoses in patients with azotemia and above-normal
concentrations
of
-
0
0
40
80
Creatinine,
120
160
200
mg/L
Fig. 1. Serum concentrations of B-isoenzymes in patients with
azotemia
Dashed lines indicate the upper limits of the normalranges
1822
of sigmoid
UIL was the highest acid phosphat.ase
activity observed;
of adults studied had activities
of at least 7 U/L.
30
I
cellulitis
Fanconi syndrome
infection
men (Fleisher,
I,
streptococcal
nephritis;
treatment
and predictive
for an infection diagnosed
DIagnosis
16
66
1.9-26.6;
6.0-22.0;
Stage III, 7
B-leoenzymes,
pg/L
13
Adenocarcinoma of
prostate, 43
no prior treatment,
18
prior treatment, 25
Serum
creatInine,
mg/L
7
(15.6-2 1.3)
10(3.3)
diseases, 105
Table 2. Clinical Diagnoses in 15 Patients with
Azotemia and Above-Normal Concentrations of
B-lsoenzymes
CLINICAL CHEMISTRY, Vol. 26, No. 13, 1980
B-isoenzymes.
Above-normal
concentrations
were identified
in only two of 28 sera from patients with chronic renal failure
and creatinine
concentrations
greater than 30 mg/L, but were
observed
in three specimens
(one each) from patients
with
acute renal failure secondary to ischemia, inflammation,
or
urinary tract infection.
In eight of the 10 remaining
cases,
32
24
$
16
8
0
-
t
-
-
-
i
-
-
.1-
I
-.s-
-
-
-
‘5
S
ARCO
0
Benign Prostatic
without
Concomitant
Disease
Hyperplasia
with
Concomitant
Disease
III
ARCD
Carceoma
of Prostate
No Prior
Treatment
PiOf
Treatment
Transitional Cel
Carcenma
ot Bladder
Fig. 2. Serum concentrations
of B-isoenzymes in patients with
obstructive uropathy or carcinoma of the prostate or bladder
Opencircles indicate concomitantdiseases;open squares indicate specimens
with above-normalactivity of tartrate-inhibitable acid phosphatase
dictive value of an above-normal
concentration
of B-isoenzymes for carcinoma
was only 39% (Table 3). When patients
with concomitant
diseases were eliminated,
the predictive
value for carcinoma
of a positive test result increased
to 64%.
Conversely,
knowledge
that the concentration
of B-isoenzymes in serum was less than 15 itg/L did not exclude the diagnosis of carcinoma
of the prostate.
The predictive
value of
a negative result was 79%, and not significantly
different from
the prevalence
of benign
hyperplasia
in the combined
groups.
Above-normal
concentrations
of B-isoenzymes
also were
detected
in some serum specimens
from patients
with transitional cell carcinoma
of the bladder (Table 1 and Figure 2).
Positive test results were more common in patients
with infiltrative
carcinomas:
seven of 19 cases, compared
with three
of 14 cases with in situ disease, One patient with in situ carcinoma and a concentration
of B-isoenzymes
of 25 g/L had
a penetrating
duodenal ulcer; otherwise, no concomitant
diseases were identified
in the individuals
with above-normal
concentrations
of B-isoenzymes.
Discussion
concomitant
diseases
of other organs were identified
that
might have accounted
for the increased
concentrations
of
B-isoenzymes
(Table 2).
Above-normal
concentrations
of B-isoenzymes
also were
noted in sera from some patients
with benign prostatic
hyperplasia
(Table 1 and Figure 2). In 13 of 20 such cases, concomitant
diseases such as chronic prostatitis
(n = 4), acute
urinary tract infection
with renal calculi or hydroureter
(n =
4), or carcinomatosis
(n = 2) were identified.
When patients
with concomitant
diseases
were eliminated
from this group,
only seven of the 105 remaining
had above-normal
concentrations of B-isoenzymes
in serum; the highest concentration
noted was 21.3 ig/L.
The results of tests done in patients
with carcinoma
of the
prostate
also are presented
in Figure 2 and Table 1. Three of
18 sera from patients
without
prior treatment
and 10 of 25
from treated
patients
had above-normal
concentrations
of
B-isoenzymes.
In the group of untreated
patients,
only two
of six with Stage D disease and increased activities of prostatic
acid phosphatase
in serum had above-normal
concentrations
of B-isoenzymes.
In the group that had received prior treatment, several patients with normal activities of prostatic acid
phosphatase
in serum had slightly increased
concentrations
of B-isoenzymes
(Figure 2).
Given these results in patients
with carcinoma
and benign
hyperplasia
of the prostate,
it is not surprising
that the pre-
Table 3. Predictive Values of B-lsoenzyme
Measurements
Senel.
tlvlty
SpecIfIcIty
Allpatientswith
ADCA and
BPH
30%
(13/43)
85%
(115/ 135)
ADCAandBPH,
32%
93%
patients with
concomitant
(13/41)
some patients with diseases of the genitourinary
tract. However, it remains to be determined
whether concentrations
of
this analyte change with sufficient
frequency
in particular
diseases to make measurement
of it clinically
useful for diagnosis or management.
Our data indicate that the concentration of B-isoenzymes
in serum bears no relationship
to renal
function
in patients
with chronic azotemia.
We believe that
the published
reports that indicate
such a relationship
are
mistaken
in identifying
the predominant
fluorescent
species
in the sera of chronically
azotemic patients
as the BB isoenzyme. On the other hand, we have identified some examples
of above-normal
concentrations
Pv(+)
39%
PV79%
64%
78%
()
(98/105)
omitted
Calculations are for B-isoenzyme results in thediagnosis
ofcarcinomaof the
prostate (ADCA). The cutoff concentration for an above-normal concentration
of B-isoenzymes was 15 cg/L. The prevalences of ADCA in the groups comprising all patients with ADCA and benign prostatic hyperplasts (BPH) were 24%
(43/178) and 28% (41/146) before and after eliminating all patients with con-
of B-isoenzymes
in the sera
of patients
with acute renal failure secondary
to ischemia or
inflammation.
This observation
was expected
from the
knowledge
that renal tubular
epithelia,
particularly
cells of
the distal tubules, contain large amounts of the BB isoenzyme
(15). Our data are insufficient
at this time to indicate whether
measurements
of B-isoenzymes
are useful to assess the severity of renal ischemia or the presence of acute tubular necrosis. These applications
of B-isoenzyme
measurements
require further investigation.
With respect to the BB isoenzyme
as a possible marker for
adenocarcinoma
of the prostate,
several points require emphasis. For any analyte to be useful as a diagnostic
tumor
marker,
it must be present
in body fluids in above-normal
concentrations
in a high percentage
of patients with the tumor
and in a low percentage
diseases
comitant diseases.
There is an emerging consensus that the BB isoenzyme of
CK is detectable by several analytical methods in the sera of
of those with other diseases.
Alter-
natively, for measurements
to be useful in the management
of patients with cancer, the concentration
of an analyte must
reflect the extent of spread of the tumor or the rate of tumor
growth. Our data indicate that with respect to carcinoma
of
the prostate
the BB isoenzyme
does not have the attributes
of a clinically useful tumor marker. The concentration
of Bisoenzymes
carcinoma
in serum
often
is not above-normal
enough
to make
in patients
its measurement
useful
with
in
screening.
Furthermore,
prostatitis,
infection
of the urinary
tract, and many other diseases are associated
with abovenormal concentrations
of B-isoenzymes,
complicating
the
interpretation
of positive test results and limiting
the usefulness of this test in both the diagnosis
and management
of
carcinoma
of the prostate.
Our results also indicate that the concentration
of B-isoenzymes in serum often does not parallel the activity of tar-
trate-inhibitable
acid phosphatase
mented carcinoma
of the prostate.
CLINICAL CHEMISTRY,
in patients
Other
with docu-
published
Vol. 26. No. 13, 1980
data
1823
support
(16, 17). Above-normal
concentrations
some sera with normal acid phosphatase
may have been caused by the release of BB isoenzyme
this finding
of B-isoenzymes
in
activity
from other tissues, or may indicate
more widespread
dissemination
of the tumor.
Benign and malignant
transitional
epithelial
cells also
contain the BB isoenzyme (15). Once again, however, there
is little reason from our data to recommend
measuring
the
concentration of B-isoenzymes
in serum in patients suspected
of having carcinoma of the urinary bladder. Additional data
from longitudinal studies are needed to determine whether
an above-normal
concentration
tumor dissemination
in patients
prostate
or the bladder.
of B-isoenzymes
indicates
with carcinoma
of either the
This study was supported by a grant from the Mayo Clinic Department of Laboratory Medicine. We thank Cynthia Bahler for
performing radioimmunoassays
and Rodney Forsman for measuring
acid phosphatase activity.
References
I. Zweig, M. H., Van Steirteghem, A. C., and Schechter, A. N., Radioimmunoassay
of creatine kinase isoenzymes in human serum:
Isoenzyme BB. Clin. Chem. 24, 422-428 (1978).
2. Roberts, R., Sobel, B. E., and Parker, C. W., Radioimmunoassay
for creatine kinase isoenzymes. Science 194, 855-857 (1976).
3. Feld, R. D., and Witte, D. L., Presence of creatine kinase BB isoenzyme in some patients with prostatic carcinoma. Clin. Chem. 23,
1930-1932 (1977).
4. Galen, R. S., Creatine kinase isoenzyme BB in serum of renaldisease patients. Clin. Chem. 22, 120 (1977).
5. Jung, K., Scholz, D., and Precht, K., Creatine kinase isoenzyme
1824
CLINICAL CHEMISTRY, Vol. 28, No, 13, 1980
BB in serum of patients undergoing chronic hemodialysis and with
kidney transplant. Enzyme
24, 169-172 (1979).
6. Chuga, D. J., and Bachner, P., Creatine kinase isoenzyme BB in
the serum of renal-disease patients, distinct from an albumin-like
artifact. GUn. Chem. 24, 1286 (1978).
7. Weseley, S. A., Byrnes, A., Alter, S., et al., Presence of creatine
phosphokinase
brain band in the serum of chronic renal disease patients. Clin. Nephrol.
8, 345-348 (1977).
8. Silverman,
L. M., Dermer, G. B., Zweig, M. H., et al., Creatine
Idnase BB: A new tumor-associated
marker. Clin. Chem. 25,1432-1435
(1979).
R. B., Herbstman,
R., and Hermann, P., Spurious brain
creatine kinase in serum from patients with renal disease. Clin. Chem.
24, 1636-1638 (1978).
10. Homburger,
H. A., and Jacob, G. L., Creatine kinase radioimmunoassay
and isoenzyme electrophoresis compared in the diagnosis
of acute myocardial infarction. Clin. Chem. 26,861-866
(1980).
11. Kind, P. R. N., and King, E. J., Estimation
of plasma phosphatase
by determination
of hydrolyzed phenol with amino-antipyrine.
J.
9. Coolen,
Clin. Pathol.
7,322-326
(1954).
12. Raabo, E., and Wallhoe-Hansen,
P., A routine method for determining creatine avoiding deproteinization.
Scand. J. Clin. Lab.
Invest. 29, 297-301 (1972).
13. Klein, L. A., Prostatic carcinoma. N. Engi. J. Med. 300, 824-833
(1979).
14. Vecchio, T. J., Predictive value of a single diagnostic test in unselected populations.
N. Engi. J. Med. 274, 1171-1173 (1966).
15. Wold, L. E., Li, C. Y., and Homburger, H. A., Localization of the
B and M polypeptide subunits of creatine kinase in normal and neoplastic human tissues by an immunoperoxidase
technique.
Am. J.
Clin. Pathol. (in press).
16. Forman, D. T., The significance of creatine kinase (CKBB) in
metastatic
cancer of the prostate. Ann. GUn. Lab. Sci. 9, 333-337
(1979).
17. Sj#{246}vall,
K., Rubin, S., and MUntzing, J., Creatine phosphokinase
in prostatic tissue. Scand. J. Urol. Nephrot.
9, 181-184 (1975).