U.S. patent application number 10/092769 was filed with the patent office on 2003-01-23 for insulin-like growth factor system and cancer.
This patent application is currently assigned to Diagnostic System Laboratories, Inc.. Invention is credited to Diamandi, Anastasia, Khosravi, M. Javad, Mistry, Jehangir.
Application Number | 20030017513 10/092769 |
Document ID | / |
Family ID | 23926118 |
Filed Date | 2003-01-23 |
United States Patent
Application |
20030017513 |
Kind Code |
A1 |
Khosravi, M. Javad ; et
al. |
January 23, 2003 |
Insulin-like growth factor system and cancer
Abstract
Method of monitoring or diagnosing disease conditions, including
disease of the prostrate, that involve measuring a combination of
tumour markers and at least one component of the IGF axis. The
invention is exemplified with prostate cancer and benign prostatic
hyperplasia, the tumor marker prostate specific antigen, and the
insulin-like growth factors and their binding proteins.
Inventors: |
Khosravi, M. Javad;
(Toronto, CA) ; Diamandi, Anastasia; (Toronto,
CA) ; Mistry, Jehangir; (League City, TX) |
Correspondence
Address: |
HAYNES AND BOONE, LLP
901 MAIN STREET, SUITE 3100
DALLAS
TX
75202
US
|
Assignee: |
Diagnostic System Laboratories,
Inc.
Webster
TX
|
Family ID: |
23926118 |
Appl. No.: |
10/092769 |
Filed: |
March 7, 2002 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10092769 |
Mar 7, 2002 |
|
|
|
09484903 |
Jan 18, 2000 |
|
|
|
Current U.S.
Class: |
435/7.23 |
Current CPC
Class: |
G01N 33/57496 20130101;
Y10S 436/813 20130101; G01N 2333/65 20130101; G01N 33/574 20130101;
G01N 2333/4745 20130101; G01N 2800/52 20130101; G01N 33/57434
20130101; G01N 33/57415 20130101; G01N 33/57423 20130101; G01N
33/57484 20130101; G01N 33/74 20130101; G01N 33/57419 20130101;
G01N 2333/96455 20130101 |
Class at
Publication: |
435/7.23 |
International
Class: |
G01N 033/574 |
Claims
We claim:
1. A diagnostic tool for discriminating between benign and
malignant disease, said tool comprising a indicator ratio selected
from the group consisting of the concentration ratios: a)
IGF/kallikrein-like protein, b) IGFBP/kallikrein-like protein, c)
IGF/IGFBP/kallikrein-like protein, d) (intact IGFBP/total
IGFBP)/kallikrein-like protein, and e) (IGF+IGFBP)/kallikrein-like
protein.
2. The diagnostic tool of claim 1, said tool comprising a indicator
ratio selected from the group consisting of: IGF-I/free PSA, b)
intact IGFBP-3/free PSA, c) (IGF-I/total IGFBP-3)/free PSA, d)
(intact IGFBP-3/total IGFBP-3)/free PSA, e) (IGF-I+intact
IGFBP-3)/free PSA, and f) intact IGFBP3.
3. The diagnostic tool of claim 1, wherein said tool is used to
distinguish between benign conditions and lung cancer, breast
cancer, colon cancer or prostate cancer.
4. The diagnostic tool of claim 1, wherein said tool is used to
distinguish between benign conditions and prostate cancer.
5. A method of predicting cancer in a patient, comprising
determining an indicator ratio selected from the group consisting
of the concentration ratios of: a) IGF/kallikrein-like protein, b)
IGFBP/kallikrein-like protein, c) IGF/IGFBP/kallikrein-like
protein, d) (intact IGFBP/total IGFBP)/kallikrein-like protein, and
e) (IGF+IGFBP)/kallikrein-like protein, whereby a indicator ratio
different from a standard normal value indicates the existence of
cancer.
6. The method of claim 5, wherein the indicator ratio is selected
from the group consisting of: a) IGF-I/free PSA, b) intact
IGFBP-3/free PSA, c) (IGF-I/total IGFBP-3)/free PSA, d) (intact
IGFBP-3/total IGFBP-3)/free PSA, e) (IGF-I+intact IGFB-P-3)/free
PSA, and f) intact IGFBP3.
7. The method of claim 5, wherein the indicator ration is used to
distinguish between benign conditions and lung cancer, breast
cancer, colon cancer or prostate cancer.
8. The method of claim 5, wherein said indicator ratio is used to
distinguish between benign conditions and prostate cancer.
9. The method of claim 8, wherein the method further includes a
measurement of free/total PSA.
10. A method of monitoring cancer in a patient, comprising
determining an indicator ratio selected from the group consisting
of concentration ratios of: a) IGF/kallikrein-like protein, b)
IGFBP/kallikrein-like protein, c) IGF/IGFBP/kallikrein-like
protein, d) (intact IGFBP/total IGFBP)/kallikrein-like protein, and
e) (IGF+IGFBP)/kallikrein-like protein, whereby a change in the
indicator ratio indicates progression of the cancer.
11. The method of claim 10, wherein the indicator ratio is selected
from the group consisting of a) IGF-I/free PSA, b) intact
IGFBP-3/free PSA, c) (IGF-I/total IGFBP-3)/free PSA, d) (intact
IGFBP-3/total IGFBP-3)/free PSA, e) (IGF-I+intact IGFBP-3)/free
PSA, and f) intact IGFBP3.
12. The method of claim 10 wherein the cancer is selected from the
group consisting of lung cancer, colon cancer, breast cancer and
prostate cancer.
13. The method of claim 11, wherein the cancer is prostate
cancer.
14. The method of claim 13, wherein the method further includes a
measurement of free/total PSA.
15. A method of predicting cancer in a patient, comprising
measuring at least two IGF axis components and a tumor marker.
16. The method of claim 15, wherein the IGF axis components are
selected fro the group consisting of IGF-I, IGF-II, IGFBP-1,
IGFBP-2, IGFBP-3, IGFBP-4, IGFBP-5, IGFBP6, IGFBP-rP-1, IGFBP-rP-2,
IGFBP-rP-3, IGFBP-rP4, IGFBP-rP-5, IGFBP-rP-6, IGFBP-rP-7,
IGFBP-rP-8, IGFBP-rP-9, IGFBP protease, GH, GHBP, GH receptor, IGF
receptor, IGF proteases, ALS, IGF receptor antagonists, and GH
receptor antagonists and wherein the tumor marker is selected from
the group consisting of PSA, kallikrein, S-100 protein, C219,
GCDFP-15/gp17, riboflavin carrier protein, vitamin carrier
proteins, human chorionic gonadotropin, alpha-fetoprotein, lactate
dehydrogenase, cytokeratin 19 fragment, CYFRA 21-1, carbohydrate
antigen 19.9, macrophage-colony stimulating factor, abnormal
prothrombin PIVKA-II, tissue polypeptide antigen, carcinoembryonic
antigen, cancer antigen 125, CA72-4, CA15-3, squamous cell antigen,
neuron specific enolase, focal adhesion kinase, soluble CD44
(sCD44), soluble CD30 (sCD30), tissue polypeptide specific antigen
(TPSA), total alkaline phosphatase (T-ALP), urinary Dpd/creatinine
(Cre) ratios, bone specific alkaline phosphatase (B-ALP),
N-acetylneuraminic (Neu5Ac), vascular endothelial growth factor
(VEGF), glutathione peroxidase, melanoma antigen (MAGE), mesothelin
and megakaryocyte potentiating factor (MPF), cyclin-dependent
kinase inhibitor p27 (Kip1), PGP9.5, proliferating cell nuclear
antigen (PCNA), Cyclin D1, epidermal Growth Factor (EGF),
transforming growth factor alpha (TGF alpha), estrogen
receptor-related protein (ERRP), multidrug resistance marker
(MDRM), protein kinase C (PKC), Gs alpha, inhibin, cathepsin D,
H19, the steroid hormones, p53, and cytokines and interleukins.
Description
BACKGROUND
[0001] 1. Field of the Invention
[0002] This invention relates to the measurement of the IGF-axis
component levels and tumor marker levels for use in assessing
cancer risk and/or progression and/or distinguishing between cancer
and other non-malignant disorders. The invention is exemplified
with prostate cancer (CaP) and benign prostatic hyperplasia (BHP),
the tumor marker prostate specific antigen (PSA), and the
insulin-like growth factors (IGF) and their binding proteins
(IGFBP). Specifically, IGF-I, intact IGFBP-3, fragment IGFBP-3,
total IGFBP-3, free PSA and total PSA were assayed and certain
permutations of these measurements were found to present improved
diagnostic indicators. The method is predicted to have general
applicability to other IGF-system related cancers.
[0003] 2. Description of the Prior Art
[0004] The insulin-like growth factor (IGF) family of high affinity
IGF binding proteins (IGFBP-1-6) (14) has recently evolved to a
superfamily status in order to accommodate a related group of newly
discovered low affinity IGFBPs called the IGFBP related proteins
(5). The conventional view of IGFBPs as the sole regulators of IGF
bioavailability and bioactivity has also evolved to include the
IGF-independent properties of IGFBPs (6, 7). IGFBPs, particularly
IGFBP-3, have been recently identified as potent apoptotic agents
(8-12), presumably mediating the effects of cellular growth
suppressing mechanisms (8, 11, 12). The emerging new concept
appears to similarly broaden the pathophysiological roles of the
IGF peptides to include their potential involvement in regulation
of the IGFBPs' bioactivity (8). In this ever-expanding maze of
reciprocal molecular interactions, post-translational modification
by selective proteolysis is rapidly gaining acceptance as the key
modulator of the IGF/IGFBP system and a major determinant of their
effects on cellular growth and metabolism (13, 14).
[0005] Insulin-like growth factors (IGF-I and -II) are mitogenic
and anti-apoptotic agents produced primarily by the liver and
locally by a wide variety of tissues. IGFs circulate mostly
complexed with IGFBP-3, which in association with the acid-labile
subunit (ALS) forms an approximately 150 kD ternary protein complex
(1-4). Under normal conditions, nearly all of the circulating IGFs
remain ternary complexed (75-80%), and smaller proportions (20-25%)
are associated with the low molecular weight IGFBPs (IGFBP-1,
IGFBP-2, IGFBP-4, IGFBP-5, and IGFBP-6) or exist in the free form
(1-4).
[0006] Dysregulation and/or over-expression of the IGF system have
been long implicated in the etiology of both benign and malignant
proliferative disorders (3,4, 15-19). Malignant cells of various
origins have been shown to express various components of the IGF
system (3, 4,11-13, 18-22), and increased IGF-I levels, as seen in
acromegaly, have been found in association with benign prostatic
hyperplasia (BPH) (23, 24) and colonic tumors (25, 26). High levels
of circulating IGF-I has been more recently identified as risk
factors for the development of prostate, breast, and lung cancers
(27-30), while over-expression of both IGF-I and IGF-II has been
linked to colorectal cancers (31). In prostate, both benign and
malignant cells have been found to express IGFs, IGFBPs and their
respective receptors (18, 23). IGF-I has been shown to promote
prostate cell growth, while prostate specific antigen (PSA) has
been identified as an IGFBP-3 protease, presumably capable of
augmenting tissue access to the IGF peptides (18, 23, 32).
[0007] In men over 50 years of age, cancer of the prostate (CaP)
and benign prostatic hyperplasia (BPH) are among the most commonly
diagnosed malignant and benign proliferative disorders,
respectively (33). However, serum levels of PSA, the most reliable
predictor of CaP available to date, is also increased in BPH,
resulting in a diagnostic "gray-zone" in the PSA range of
.about.4-10 .mu.g/L (34). In addition, PSA levels of less than 4
.mu.g/L does not necessarily indicate disease-free status because
significant numbers of men with organ-confined CaP reportedly
express normal PSA levels (35). These significant limitations of
PSA testing invariably result in a diagnostic dilemma, allowing for
loss opportunity for early cancer detection, or unnecessary
surgical approaches to a readily treatable benign disorder.
Although the ratio of free/total PSA levels in serum is
significantly reduced in CaP and its determination is now used to
heighten the diagnostic accuracy of PSA testing (36, 37), there is
still a great need to further improve our ability to discriminate
between BPH and prostate cancer (35).
SUMMARY OF THE INVENTION
Abbreviations and Definitions
[0008] ACT--alpha-1-antichymotrypsin.
[0009] ALS--Acid Labile Subunit. A protein found in the 150 kD
ternary complex wherein most of the circulating IGF-I is found. ALS
is sensitive to inactivation by acid, urea and certain
detergents.
[0010] Body fluid--Any biological fluid, including but not limited
to the following: serum, plasma, lymph fluid, synovial fluid,
follicular fluid, seminal fluid, amniotic fluid, milk, mammary
fluid, whole blood, urine, spinal fluid, saliva, sputum, tears,
perspiration, mucus, tissue culture medium, tissue extracts and
cellular extracts. Preferably, the body fluid is blood, plasma,
serum or seminal fluid.
[0011] BHP--benign prostatic hyperplasia.
[0012] CaP--cancer of the prostate.
[0013] DHT--Dihydrotestosterone.
[0014] GH--Growth hormone.
[0015] GHBP--GH binding protein.
[0016] IGF--Insulin-like Growth Factor.
[0017] IGF-axis components--Those components that modulate the
IGF/GH cascades including GH, GHBP, GH receptor, IGF I and II, IGF
receptors, IGF proteases, IGFBP-1 through -6 and the IGFBP related
proteins IGFBP-rP-1-9, IGFBP proteases, ALS, IGF and GH receptor
antagonists, and the like. Altered levels of IGF axis components
are known to be associated with a variety of malignant diseases,
including breast, ovarian, endometrial, colorectal and prostate
cancer, as discussed herein, and also with papillary thyroid
cancer, Wilms tumor and possibly other CNS tumors, choroid plexus
papilloma, meningiomas, hepatocellular carcinoma, rhabdomyosarcoma,
gastric carcinomas, liver cancer, and colon cancer, leukemias,
pituitary adenomas and tumors, and lung cancer.
[0018] IGCFBP--Any IGF binding protein, including IGFBP-1 to 6 and
the IGFBP related proteins IGFBP-rP-1 to 9.
[0019] IGFBP-3--The major circulating IGF binding protein. Intact
IGFBP-3 refers to that portion of the total IGFBP-3 which is
undegraded (and exists mainly in complexed form). Fragment IGFBP-3
refers to the fragmented forms of IGFBP-3. With the assay described
herein, amino terminal fragments of IGFBP-3 that lack the carboxyl
terminal amino acids are detected as fragment IGFBP-3. Total
IGFBP-3 refers to complexed, uncomplexed, intact and fragmented
IGFBP-3. The various forms of IGFBP-3 are referred to collectively
as "IGFBP-3 variants."
[0020] Indicator ratio--As used herein a ratio of the measured
levels of IGF axis components with or without kallikrein-like
components, such as PSA, which is useful to distinguish between
benign and cancerous conditions or useful in monitoring the
progression of a cancerous disease. The specification teaches how
to evaluate various permutations of measurements of IGF axis
components and tumor markers and test for clinically significant
associations. For example, an indicator ratio of IGF-I/free PSA
means the concentration of IGF-I divided by the concentration of
free PSA.
[0021] Kallikrein--A group of serine proteases with homology to
PSA, including at least K1, K2, and preprokallikreins.
"Kallikrein-like proteins" includes the kallikreins and various
forms of PSA.
[0022] PSA--Prostate specific antigen. Free PSA is the fraction of
PSA that is not complexed with other proteins, such as ACT. Total
PSA is free PSA and complexed PSA.
[0023] Ratio--Any ratio referred to herein expressly refers to and
includes the inverse ratio. Thus if the ratio of IGF-I/free PSA is
informative about a particular disease state, of course the ratio
of free PSA/IGF-I will be equally informative.
[0024] SHBG--Sex hormone binding globulin.
[0025] T--Testosterone.
[0026] Tumor Marker--As used herein, the term includes any marker
associated with tumors or tumor progression, including PSA and
kallikrein. Other tumor markers are known, measurements of any of
which may be combinable with measurements of IGF axis components to
provide increased discriminating power. An exemplary listing of
potential tumor markers that might be useful together with
measurement of IGF axis components includes: S-100 protein, C219,
GCDFP-15/gp17, riboflavin carrier protein (RCP) and other vitamin
carrier proteins (VCP), human chorionic gonadotropin (hCG),
alpha-fetoprotein (AFP), lactate dehydrogenase, cytokeratin 19
fragment (C,K19) or CYFRA21-1, carbohydrate antigen 19.9 (CA19.9),
macrophage-colony stimulating factor (M-CSF), abnormal prothrombin
(PIVKA-II), tissue polypeptide antigen (TPA), carcinoembryonic
antigen (CEA), cancer antigen (CA) 125, CA72-4, CA15-3, squamous
cell antigen (SCC), neuron specific enolase (NSE), focal adhesion
kinase (FAK), soluble CD44 (sCD44), soluble CD30 (sCD30), tissue
polypeptide specific antigen (TPSA), total alkaline phosphatase
(T-ALP), urinary Dpd/creatinine (Cre) ratios, bone specific
alkaline phosphatase (B-ALP), N-acetylneuraminic (Neu5Ac), vascular
endothelial growth factor (VEGF), glutathione peroxidase, melanoma
antigen (MAGE), mesothelin and megakaryocyte potentiating factor
(MPF), cyclin-dependent kinase inhibitor p27 (Kip1), PGP9.5,
proliferating cell nuclear antigen (PCNA), Cyclin D1, epidermal
Growth Factor (EGF), transforming growth factor alpha (TGF alpha),
estrogen receptor-related protein (ERRP), multidrug resistance
marker (MDRM), protein kinase C (PKC), Gs alpha, inhibin, cathepsin
D, H19, the steroid hormones, p53, and cytokines and
interleukins.
[0027] The invention in its broadest sense consists of a method of
predicting cancer in a patient measuring at least two IGF axis
components and a tumor marker. The measurements are combined in
statistically significant permutations, as described herein, to
provide an improved means of discriminating between cancerous and
non-cancerous conditions.
[0028] More particularly, the invention provides a diagnostic tool
for discriminating between benign and malignant disease. The tool
is an indicator ratio which is a concentration ratio such as
IGF/kallikrein-like protein, IGFBP/kallikrein-like protein,
IGF/IGFBP/kallikrein-like protein, (intact IGFBP/total
IGFBP)/kallikrein-like protein, and (IGF+IGFBP)/kallikrein-like
protein.
[0029] The indicator ratio may also be IGF-I/free PSA, intact
IGFBP-3/free PSA, (IGF-I/total IGFBP-3)/free PSA, (intact
IGFBP-3/total IGFBP-3)/free PSA, and (IGF-I+intact IGFBP-3)/free
PSA. It has also been discovered that intact IGFBP-3 is a valid
indicator of prostate CaP and this marker may be used alone or may
be combined with existing tumor marker measurements or ratios, such
as free PSA or free/total PSA. The diagnostic tool can distinguish
between benign conditions and lung cancer, breast cancer, colon
cancer or prostate cancer.
[0030] The tool can also be used in method of predicting cancer in
a patient, wherein the method comprises the determination of one or
more of the above indicator ratios. The indicator ratio is compared
to the ratio obtained in a normal patient population and
significant deviations from the norm indicate cancer. The method
can also be used to monitor the progression of disease.
BRIEF DESCRIPTION OF FIGURES
[0031] FIG. 1. Percentile distribution plots. Percentile
distribution of IGF-I (FIG. 1A), intact IGFBP-3 (FIG. 1B), and free
PSA (FIG. 1C) levels measured in patients with BPH (n=75) and CaP
(n=84) are shown.
[0032] FIG. 2. Relationship of IGF-I, intact, and total IGFBP-3
with free PSA. Correlation of IGF-I (FIG. 2A), total IGFBP-3 (FIG.
2B) and intact IGFBP-3 (FIG. 2C) levels in subjects with BPH vs the
corresponding free PSA levels are shown. Values are mean of
duplicate measurements.
[0033] FIG. 3. Relationship of intact, fragment and total IGFBP-3
with IGF-I. Correlation of total IGFBP-3 (FIG. 3A), intact IGFBP-3
(FIG. 3B), and fragment IGFBP-3 (FIG. 3C) levels measured in
subjects with BPH vs the corresponding IGF-I levels are shown.
Values are mean of duplicate measurements.
[0034] FIG. 4. Relationship of intact, fragment and total IGFBP-3
with IGF-I. Correlation of total IGFBP-3 (FIG. 4A), intact IGFBP-3
(FIG. 4B), and fragment IGFBP-3 (FIG. 4C) levels measured in
subjects with CaP vs the corresponding IGF-I levels are shown.
Values are mean of duplicate measurements.
[0035] FIG. 5. Distribution of various concentration ratios in BPH
and CaP patients. Box plot of IGF-I/free PSA (FIG. 5A), intact
IGFBP-3/free PSA (FIG. 5B) and (IGF-I/total BP3)/free PSA (FIG. 5C)
ratios in subjects with BPH and CaP. The median (centerline) and
the 95% limits about the median are shown. Abbreviations are
described in footnote to Table 1.
[0036] FIG. 6. Distribution of various concentration ratios in BPH
and CaP patients. Box plot of intact (IGFBP-3/total IGFBP-3)/free
PSA (FIG. 6A), (IGF-I+intact IGFBP-3)/free PSA (FIG. 6B), and free
PSA/total PSA (FIG. 6C) ratios in subjects with BPH and CaP. The
median (centerline) and the 95% limits about the median are shown.
Abbreviations are described in footnote to Table 1.
[0037] FIG. 7. Receiver operating characteristics (ROC) curves.
Comparative potential of IGF-I and IGF-I/free PSA ratio relative to
free PSA/total PSA ratio in discriminating between BPH and CaP
patients is shown. 1-Specificity (1-[true negatives/true
negatives+false positives]) versus sensitivity (true positives/true
positives+false negatives) is plotted. The corresponding area under
the curve (AUC) and confidence intervals (CI) are described in the
text. Abbreviations are described in footnote to Table 1.
[0038] FIG. 8. Receiver operating characteristics (ROC) curves.
Comparative potential of (IGF-I/total IGFBP-3)/free PSA and
IGF-I/total IGFBP-3 ratios relative to free/total PSA ratio in
discriminating between BPH and CaP patients is shown. 1-Specificity
versus sensitivity is plotted, as in FIG. 7. The corresponding area
under the curve (AUC) and confidence intervals (CI) are described
in the text. Abbreviations are described in footnote to Table
1.
[0039] FIG. 9. Receiver operating characteristics (ROC) curves.
Comparative potential of IGF-I+intact IGFBP-3 and (IGF-I+intact
IGFBP-3)/free PSA ratio relative to free PSA/total PSA ratio in
discriminating between BPH and CaP patients is shown. 1-Specificity
versus sensitivity is plotted, as in FIG. 7. The corresponding area
under the curve (AUC) and confidence intervals (CI) are described
in the text. Abbreviations are described in footnote to Table
1.
[0040] FIG. 10. Receiver operating characteristics (ROC) curves.
Comparative potential of intact IGFBP-3/free PSA, intact
IGFBP-3/total IGFBP-3, and (intact IGFBP-3/total IGFBP-3)/free PSA
ratios in discriminating between BPH and CaP patients is shown.
1-Specificity versus sensitivity is plotted, as in FIG. 7. The
corresponding area under the curve (AUC) and confidence intervals
(CI) are described in the text. Abbreviations are described in
footnote to Table 1.
DETAILED DESCRIPTION OF THE INVENTION
[0041] Generally speaking, the invention is directed to improved
diagnostic indicators of malignant disease states which involve
measurement of various permutations of IGF axis components together
with tumor markers in order to provide increased diagnostic
accuracy. The invention is exemplified with prostate cancer and the
analtyes: IGF-I, intact, fragment and total IGFBP-3, and free and
total PSA. However, the invention can be broadened to other IGF
axis components and possibly to other tumor markers as well. If
other tumor markers prove useful in the way that PSA and IGF axis
components have proved useful, then the invention will be
applicable to other cancers, and several reasonable predictions in
this regard are provided.
[0042] Specifically, the permutations of IGF-I/free PSA, intact
IGFBP-3/free PSA, (IGF-I/total IGFBP-3)/free PSA, (intact
IGFBP-3/total IGFBP-3)/free PSA, and (IGF-I+intact IGFBP-3)/free
PSA and measurement of intact IGFBP-3 alone have been shown to be
useful indicators of prostate cancer. Further, these indicator
ratios together with the ratio of free/total PSA in multivariate
analysis provides improved discriminating potential than does the
ration of free/total PSA alone.
EXAMPLE 1
[0043] In view of the growing evidence describing association of
the IGF system with cancer, we used prostate cancer as a model and
investigated differences in serum levels of IGF-I and intact,
fragment and total IGFBP-3 in a group of patients with BPH and CaP.
The age-matched patient populations were carefully selected to have
total PSA in the diagnostic gray-zone range. Because of the highly
complex nature of the IGF regulation, particularly involving
proteolysis (13, 14), we postulated that investigation of IGF-I and
IGFBP-3 variants in relation to levels of PSA in serum might help
identify innovative approaches for enhancing differential
BPH/cancer detection. As the IGF system appears to be more widely
associated with malignancy (3, 4, 11-13, 18-22), we further
speculated that the present approach could have broader application
in human cancer diagnostics and monitoring.
[0044] Patient population and samples Serum samples from 159
patients with benign prostatic hyperplasia (BPH 75 males aged
55-75; mean age.+-.SD, 65.6.+-.6.0) or prostate cancer (84 males
aged 52-75; mean age.+-.SD, 64.8.+-.6.2) were provided by Dr. E. P.
Diamandis (Mount Sinai Hospital, Toronto, Ontario). The samples
were from patients with total PSA levels between 1.75-13.5 .mu./L
and with histologically confirmed disease status at biopsy. All
specimens were residuals from routine or research test samples and
were stored frozen at -70.degree. C., with less than three
freeze/thawing cycles prior to analysis.
[0045] Analytical methods IGF-I, intact IGFBP-3, fragment IGFBP-3
and total IGFBP-3 were assayed by ACTIVE enzyme-linked
immunosorbent assay (ELISA) kits manufactured by Diagnostic Systems
Laboratories, Inc. (DSL, Inc., Webster, Tex.). These assays are
based on non-competitive ELISA involving a solid-phase capture
antibody and a soluble horseradish peroxidase (HRP)-labeled
detection antibody. The DSL IGF-I ELISA employed is a modification
of a previously described method involving acid-ethanol extraction
(38, 39) and measures total IGF-I. The assay incorporates a
101-fold sample pretreatment (acid-neutralization) dilution factor
and a total incubation time of less than 3 hours, and demonstrates
an overall variance of less than 10% (39, 40).
[0046] The intact, fragment and total IGFBP-3 ELISAs were developed
based on our knowledge of antibody binding specificity and IGFBP-3
complex epitope recognition derived by systematic evaluation of 10
different IGFBP-3 monoclonal antibodies in four different binding
experiments (41), including further performance assessment using a
polyclonal detection antibody. The IGFBP-3 ELISAs incorporate
identical components and protocols, and involve a common monoclonal
capture antibody in combination with a polyclonal (ELISA-3) or two
different monoclonal (ELISA- 1 and -2) detection antibodies in a
manner previously reported for development of total and
non-phosphorylated IGFBP-1 (42). These assays also include a
101-fold sample pre-dilution and total incubation times of about 3
hours, and their analytical specification and performance
characteristics have been recently described (43, 44).
[0047] Concentrations of free and total PSA were determined by the
Hybritech Tandem-R total and free (45) non-competitive
immunoradiometric (IRMA) methods (Hybritech Inc., San Diego,
Calif.).
[0048] Data Analysis The ELISA results were analyzed using the data
reduction packages included in the Labsystems Multiskan microplate
ELISA reader (Labsystems, Helsinki, Finland) with cubic spline
(smoothed) curve fit.
[0049] The analysis of differences between IGF-I and IGFBP-3
concentrations in the two groups of subjects was performed with
nonparametric Mann-Whitney U test. Association of IGF-I and IGFBP-3
variants in serum with the other continuous parameters was examined
using Spearman correlation. Receiver operating characteristics
(ROC) curves were plotted as 1-Specificity (1 -[true negatives/true
negatives+false positives] on the x axis) versus sensitivity (true
positives/true positives+false negatives on the y axis) and the
areas under the ROC curves (AUC) were calculated. Univariate and
multivariate unconditional logistic regression models were
developed to evaluate the ability of IGF-I and IGFBP-3 levels to
predict the presence of prostate cancer. The plots were established
by StatView (Abacus Concepts Inc, Berkeley Calif. 94704-1014). The
statistical analysis was performed by SigmaStat (Superior
Performing Software Systems Inc, Chicago Ill. 60606-9653) and SAS
(SAS Institute, Cary, N.C. 27513).
[0050] IGF system components in BPH vs CaP In serum samples from a
group of subjects with total PSA in the range of 1.75-13.5 .mu.g/L
we identified significantly higher IGF-I and intact IGFBP-3 levels
in those with CaP than BPH (p<0.001), while changes in fragment
and total IGFBP-3 were statistically insignificant (Table 1).
1TABLE 1 Descriptive statistics of measured variables BPH Subjects
CaP subjects Variable.sup.a Mean SE Range Mean SE Range P (t) PSA
5.04 0.168 2.61- 4.85 0.200 13.5- 0.173 12.1 1.75 (f) PSA 1.01
0.056 0.31- 0.757 0.049 0.15- <0.001 3.15 2.59 IGF-I 101.2 5.45
10.9- 126.6 4.89 28.0- <0.001 220 218 (i) BP-3 1.12 0.072 0.14-
1.48 0.068 0.32- <0.001 2.71 278 (f) BP-3 3.62 0.192 1.4- 3.39
0.168 1.19- 0.257 9.8 7.81 (t) BP-3 2.07 0.061 0.81- 2.22 0.061
1.17- 0.091 2.91 4.10 .sup.AValues for total (t) PSA, free (f) PSA,
IGF-I are in .mu.g/L; intact (i) IGFBP-3 (BP-3), fragment (f)
IGFBP-3 and total (t) IGFBP-3 are in mg/L SE = Standard error of
the mean; P = probability
[0051] In these samples, the mean (.+-.SE) IGF-I and intact IGFBP-3
levels were 101.2.+-.5.45 .mu.g/L (range 10.9-220) and
1.12.+-.0.072 mg/L (range 0.14-2.71) in BPK and 126.6.+-.4.89
.mu.g/L 28-218) and 1.48.+-.0.068 mg/L (range 0.32-2.78) in CaP
patients. As expected (36), the total PSA levels were relatively
similar, while free PSA showed statistically significant
differences, and were 1.01.+-.0.056 .mu.g/L (range 0.31-3.15) and
0.757.+-.0.049 .mu.g/L (range 0.15-2.59) in BPH and CaP patients,
respectively.
[0052] Differences in IGF-I and intact IGFBP-3 levels in BPH vs CaP
were further evident in percentile distribution plots, which for
the most part identified clear separation between the two groups of
patients. Only at the high end of the distribution plots, the IGF-I
as well as the intact IGFBP-3 levels in BPH and CaP appeared to
overlap. The percentile plots of the free PSA levels demonstrated a
similar but inverse pattern in relation to BPH vs CaP patients
(FIG. 1). Percentile plots for total IGFBP-3, fragment IGFBP-3 and
total PSA were non-discriminative and showed complete overlap
throughout the range of the measurements (data not shown).
[0053] In comparative correlation analysis (Spearman), neither
IGF-I nor IGFBP-3 variants correlated significantly with the
corresponding total or free PSA levels detected in CaP subjects
(Table 2). Similarly, IGF-I and IGFBP-3 variants did not
demonstrate any significant correlation in relation to the total
PSA levels in BPH (Table 2). In contrast, BPH levels of IGF-I,
intact and total IGFBP-3 showed negative correlation (Spearman
correlation) vs the corresponding free PSA levels (R=-0.261-0.351,
p=0.024-<0.001) (Table 2). Similar results were obtained when
the data were subjected to linear regression analysis by the
Least-Squares method, and correlation coefficient determined by the
Pearson method (FIGS. 2A-2C). Interestingly, in BPH and CaP
patients, IGF-I levels correlated more tightly with intact IGFBP-3
than with total IGFBP-3, while fragment IGFBP-3 was negatively
related to IGF-I in BPH and not at all in CaP (FIGS. 3A-3C and
4A-4C).
2TABLE 2 Correlation matrix for IGF-I and IGFBP-3 comparison to
PSA.sup.a BPH subjects Comparative method IGF-I (i) BP-3 (f) BP-3
(t) BP-3 (t) PSA r -0.203 -0.126 -0.109 -0.167 p 0.083 0.286 0.359
0.158 (f) PSA r -0.261 -0.308 -0.096 -0.351 p 0.025 0.008 0.418
0.0025 CaP subjects (t) PSA r 0.0192 -0.069 0.021 -0.004 p 0.864
0.540 0.857 0.974 (f) PSA r -0.058 -0.190 0.019 -0.083 p 0.604
0.0821 0.866 0.462 .sup.aSpearman correlation
[0054] Calculated parameters The inverse relation of IGF-I and
intact IGFBP-3 versus free PSA, and their apparent
disease-dependent associations, prompted evaluation of various
concentration ratios for their discriminating ability. We examined
a number of different permutations and identified the following
ratios as the most discriminative:
[0055] a) IGF-I/free PSA,
[0056] b) intact IGFBP-3/free PSA,
[0057] c) (IGF-I/total IGFBP-3)/free PSA,
[0058] d) (intact IGFBP-3/total IGFBP-3)/free PSA, and
[0059] e) (IGF-I+intact IGFBP-3)/free PSA.
[0060] As shown in FIGS. 5 and 6, the median values for the above
ratios were significantly different in BPH vs CaP subjects, and
were 94.91 and 185 (FIG. 5A), 0.85 and 2.09 (FIG. 5B), 56.26 and
82.4 (FIG. 5C), 0.489 and 0.938 (FIG. 6A), 0.978 and 2.361 (FIG.
6B). As previously reported (36, 37), the median values for
free/total PSA ratio was lower in CaP than in BPH subjects and were
0.144 and 0.202 respectively (FIG. 6C).
[0061] Although the observed differences were all highly
significant (p<0.001), the medians for the new parameters showed
an increase of 1.46 fold (for IGF-I/total IGFBP-3)/free PSA ratio)
to 2.46 fold (for intact IGFBP-3/free PSA) in CaP vs BPH subjects.
The free/total PSA ratio showed a relative change of only 1.39
fold. Thus, the above ratios demonstrated increased ability to
discriminate between BHP and CaP patients who have PSA levels in
the gray-zone.
[0062] Receiver operating characteristic curves In attempts to
better define the cancer differentiating potential of the measured
and calculated parameters, receiver operating characteristics
curves (ROC) were constructed. As shown in Table 3, ratios
demonstrated better discriminating powers than the individual
variables.
3TABLE 3 Characteristics of ROC curves for measured and calculated
parameters Function* AUC SE P 95% CI of Area (f) PSA/(t) PSA 0.689
0.0425 <0.0001 0.605 to 0.772 IGF-I 0.655 0.0438 0.0002 0.569 to
0.741 IGF-I/(t) PSA 0.646 0.0445 0.0005 0.559 to 0.733 IGF-I/(f)
PSA 0.728 0.0404 <0.0001 0.649 to 0.808 LGF-I/(t) BP-3 0.609
0.0460 0.0088 0.519 to 0.699 IGF-I/(t) BP-3/(t) PSA 0.617 0.0453
0.0049 0.528 to 0.706 IGF-I/(t) BP-3/(t) PSA 0.725 0.0405
<0.0001 0.646 to 0.805 (i) BP-3 0.663 0.0435 <0.0001 0.578 to
0.749 (f) BP-3 0.553 0.0463 0.1266 0.462 to 0.644 (t) BP-3 0.565
0.0461 0.0791 0.475 to 0.655 (i) BP-3/(f) PSA 0.737 0.0404
<0.0001 0.584 to 0.754 (i) BP-3/(t) BP-3 0.685 0.0427 <0.0001
0.601 to 0.769 (i) BP-3/(t) BP-3/(f) PSA 0.747 0.0395 <0.0001
0.670 to 0.824 IGF-I + (i) BP-3 0.670 0.0432 <0.0001 0.585 to
0.754 (IGF-I + (i) BP-3)/(t) PSA 0.733 0.0405 <0.0001 0.653 to
0.8 12 IGF-I X (i) BP-3 0.669 0.0436 <0.0001 0.584 to 0.755
IGF-I X (i) BP-3/(f) PSA 0.710 0.0419 <0.0001 0.600 to 0.768
*Abbreviation are described in the footnote to Table 1.
[0063] Ratios involving total PSA or fragment IGFBP-3 were least
discriminating, while those based on IGF-I, intact IGFBP-3 and free
PSA appeared most discriminating. Compared to the currently used
free/total PSA (AUC, 0.689; 95% CI, 0.605-0.772), several
permutations, particularly the ratios of (intact IGFBP-3/total
IGFBP-3)/free PSA, intact IGFBP-3/free PSA, (IGF-I+intact
IGFBP-3)/free PSA, IGF-I/free PSA, and (IGF-I/total IGFBP-3)/free
PSA demonstrated better discriminative potential (Table 3 and FIGS.
7-10) than the prior art methodology.
[0064] Among these, the (intact IGFBP-3/total IGFBP-3)/free PSA
(AUC, 0.747; 95% CI, 0.670-0.824) and intact IGFBP-3/free PSA (AUC,
0.737; 95% CI, 0.585-0.745) ratios, involving intact IGFBP-3, had
better differentiating power than the IGF-I/free PSA ratio (AUC,
0.728; 95% CI, 0.649-0.808). The ratios of (IGF-I+intact
IGFBP-3)/free PSA (AUC, 0.733; 95% CI, 0.653-0.812) and
(IGF-I/total IGFBP-3)/free PSA (AUC, 0.725; 95% CI, 0.646-0.805)
were also found to have potential.
[0065] We also examined other functions involving IGF-I, intact
IGFBP-3 and free PSA (logarithmic ratios, difference), but none
appeared promising. The comparative ability of IGF-I/free PSA,
intact IGFBP-3/free PSA, and free/total PSA ratio in
differentiating between BPH and CaP patients at selected cut-off
values is summarized in Table 4. The observation that the
IGF-I/free PSA or intact IGFBP-3/free PSA ratio could be used to
significantly improve the specificity of CaP detection is of
particular importance as it complement the free/total PSA testing,
which has a problem of a poor cancer detection specificity.
Obviously, the inverse relation of IGF-I/free PSA or intact
IGFBP-3/free PSA relative to free/total PSA testing in terms of
limits of sensitivity versus specificity could potentially further
enhance the diagnostic differentiation between CaP and BPH. As has
been demonstrated for this particular application, the present
findings further exemplify the significant clinical potential of
combination testing of biomarkers that respond inversely to
pathophysiological changes. The combined testing of such variables
followed by data analysis by appropriate statistical manipulation
may obviously help enhancing the diagnostic accuracy of disease
versus non-disease identification.
4TABLE 4 Comparison of sensitivity vs specificity at selective
cut-off points.sup.a Cut-off Parameter value** Sensitivity (%)
Specificity (%) Cap value (f) PSA/(t) 0.419 100 2.7 <cut-off PSA
0.28 95 17 <cut-off 0.24 90 27 <cut-off IGF-I/(f) PSA 410 100
14 >cut-off 273 95 29 >cut-off 264 90 29 >cut-off (i)
BP-3/(f) PSA 5200 100 13 >cut-off 3500 95 29 >cut-off 3100 90
34 >cut-off .sup.aData from ROC curves *Abbreviation are
described in the footnote to Table 1. **Values throughout for total
(t) PSA, free (f) PSA, IGF-I are in .mu.g/L; intact (i) IGFBP-3
(BP-3), fragment (f) IGFBP-3 and total (t) IGFBP-3 are in mg/L. All
units are changed to mg/L and the cut-off value is therefor
unitless in this table.
[0066] Univariate and multivariate analysis. We developed
univariate and multivariate logistic regression models in attempts
to further demonstrate the cancer predictive values of the new
determinants. As shown in Table 5, increased levels of IGF-I/free
PSA and intact IGFBP-3/free PSA ratios were found to be associated
with increased probability for cancer.
5TABLE 5 Univariate analysis for predicting presence of CaP using
unconditional logistic regression modeling Covariate* Crude risk
ratio 95% C.I. p-value** (t) PSA 0.93 0.77-1.13 0.48 (f) PSA 0.28
0.12-0.61 0.0016 (f) PSA/(t) PSA 0.20 0.042-0.83 0.0038 IGF-I 2.57
1.16-5.73 0.02 IGF-I/(f)PSA 6.71 3.34-13.1 <0.001 (i) BP-3 2.63
1.52-4.55 <0.001 (t) BP-3 1.68 0.91-3.05 0.093 (i) BP-3/(t) BP-3
2.19 1.25-3.80 0.003 (i) BP-3/(f) PSA 3.30 1.82-5.90 <0.001 Age
0.99 0.94-1.04 0.62 *Abbreviation are described in the footnote to
Table 1. **Test for trend
[0067] In multivariate analysis as shown in Tables 6 and 7, IGF-I
and IGFBP-3-based variables were considered separately because of
the strong correlation between these parameters as well as to show
their separate relation to PSA measurement. These regression models
were adjusted for IGF-I/free PSA or intact IGFBP-3/free PSA, as
well as for total PSA, free/total PSA, and age, all of which were
considered as continuous variables. These models identified both
IGF-I/free PSA (crude risk ratio=2.8, 95% CI=1.72-4.61, p<0.001)
and intact IGFBP-3/free PSA (crude risk ratio=1.656, 95%
CI=1.18-2.34, p<0.004) as independent factors in predicting the
presence of CaP and thus, highly useful for differentiating between
CaP and BPH patients. In both cases, free/total PSA ratio appear to
significantly improve the predictive power of the indicator ratios
in these multivariate models (Table 6 and 7).
6TABLE 6 Multivariate analysis for predicting presence of CaP using
unconditional logistic regression modeling Covariate* Crude risk
ratio 95% C.I. p-value** (t) PSA 1.05 0.84-1.32 0.66 (f) PSA/(t)
PSA 0.30 0.15-0.66 0.002 IGF-I/(f) PSA 2.80 1.72-4.61 <0.001 Age
1.0 0.97-1.10 0.28 *Abbreviations are described in the footnote to
Table 1. **Test for trend
[0068]
7TABLE 7 Multivariate analysis for predicting presence of CaP using
unconditional logistic regression modeling Covariate* Crude risk
ratio 95% C.I. p-value** (t) PSA 1.06 0.83-1.33 0.65 (f) PSA/(t)
PSA 0.39 0.16-0.82 0.016 (i) BP-3/(f) PSA 1.66 1.18-2.34 0.0035 Age
1.03 0.96-1.09 0.38 *Abbreviations are described in the footnote to
Table 1. **Test for trend
[0069] Discussion IGFs are endocrine, paracrine, and autocrine
hormone that play significant roles in cellular growth and
differentiation (3, 4). Although cumulating evidence has implicated
involvement of the IGF system in cellular carcinogenesis (3, 4,
11-13, 18-22), the proposed association has not been invariably
confirmed. In case of the most intensely investigated prostate
cancer, earlier data identified little or no difference in serum
IGF-I in CaP vs normal patients (18). In addition, the high IGF-I
levels as occur in acromegaly are reportedly associated with BPH
but not CaP (23, 24). These differences may partly be due to the
effects of population and IGF assay differences (18), and partly
because acromegaly arguably results from exposure to a relatively
balanced increase in the levels of both IGF-I and its major binding
protein IGFBP-3, with little change in IGF-I/IGFBP-3 ratio
(17,46).
[0070] In contrast, prostate carcinogenesis might be more closely
dependent on dysregulation of the IGF system, resulting in
over-expression of the effector IGF-I physiology (15, 23).
Predictably, regulation of the IGFs' action depends on the
integrated effects of the systemic/locally produced IGFs, IGFBPs,
various cell surface receptors and proteases that cleave IGFBPs
and, thus, modulate their bioactivities. The profound effects of
diseased/tissue-associated physiological changes and factors such
as nutrition, genetics and aging on the rate of IGFs' production
(3,4, 17, 47, 48, 49) further compound this complexity. As the
systemic levels of IGFs could be influenced by multiple variables,
results of serum determinations may not consistently reflect
disease status as long as the masking effects of non-disease
influences are not carefully considered. The latter may be
exemplified by the reported failure of IGF-I association with
prostate and breast cancer risk in older individuals (28, 29).
[0071] Consistent demonstration of IGFs' involvement with cancer
may be even more complex if dysregulation of the bioactive (tissue
accessible) form of the IGFs is the most important determinant of
outcome. As the bioactive form of IGFs may be difficult to
accurately quantify (50), approaches involving analysis of the
IGFs, relevant IGFBPs and/or various permutations thereof may
better represent changes in the dynamics of the IGF regulation. The
predictive power of such analytical indicator(s) could be further
strengthened by their inclusion in multivariate analysis that could
also involve determinations of tissue and/or tumor-specific
(associated) markers.
[0072] Based on the above rational, we measured serum IGF-I and
IGFBP-3 in a group of patients with BPH or CaP and who had total
PSA in the diagnostic "gray zone" range. Of major interest was the
investigation of intact, fragment, and total IGFBP-3 levels, as
indicator of cancer-associated proteolysis, which has been long
implicated in etiology of various malignancies (13, 14, 51). Among
the key findings were identification of significantly higher levels
of IGF-I and intact IGFBP-3 in CaP patients (p<0.00 1), while
changes in fragment and total IGFBP-3 were statistically
insignificant. The observed differences were further evident in
percentile distribution plots, which consistently identified higher
percentages of patients with CaP at a given IGF-I or intact IGFBP-3
level, except at the high end where the plots overlapped and,
similar to those for fragment and total IGFBP-3, became
non-discriminative. Confirming previous observations (36, 37), the
total PSA levels were similar in the two group of patients
(p=0.173), while the free PSA levels were significantly lower in
CaP (p<0.001).
[0073] In comparative correlation analysis, free PSA showed
negative correlation with levels of IGF-I and IGFBP-3 (intact and
total) in BPH but not in CaP patients. In contrast, no correlations
were found in comparisons involving total PSA. The latter may be
expected as the total PSA levels quantified include its
catalytically active form, which is reportedly inactivated upon
release by complex formation with protease inhibitors, particularly
alpha-1-antichymotrypsin (ACT) (52-54). The free form of PSA,
amounting to about 10-40% of the total levels (55), is considered
catalytically inactive (52-54), which our findings appear to
confirm, but for CaP only. The reported higher levels of PSA
binding to ACT in CaP (52-55) would theoretically protect against
potential IGFBP-3 proteolysis. However, the inverse relation of
free PSA with IGF-I or IGFBP-3 in BPH suggests circulation of at
least a proportion of free PSA in catalytically active form in the
BPH patient. This is consistent with a recent report suggesting
that internal cleavage (nick) of PSA in BPH patients may be
responsible for PSA's relatively lower binding to ACT as well as
its lower chymotrypsin-like activity in comparison to seminal
plasma PSA. Interestingly, PSA in BPH patients and in seminal
plasma from normal individuals had similar trypsin-like catalytic
activity (54).
[0074] Whether the trypsin-like activity of free PSA in BHP
patients and its inability to efficiently complex with ACT is
responsible for its inverse relation to IGF-I and IGFBP-3 as
demonstrated here remains to be investigated. Theoretically,
IGFBP-3 proteolysis would lead to IGF dissociation, facilitating
clearance and/or tissue uptake of both IGF-I and IGFBP-3 fragments
and thus, a relative reduction in IGF-I and IGFBP-3 levels with
increasing PSA. Our data suggest that blocking of a proportionally
higher levels of catalytically active PSA by ACT, or other
inhibitors, might be partly responsible for the higher systemic
IGF-I and IGFBP-3 levels in CaP. This combined with possible
increases in prostatic IGF-I and IGFBP-3 production may further
account for their elevated levels in CaP patients. The inhibition
of PSA action in cancer could reportedly involve both liver and
prostatic sources of ACT (53). Whether enhanced inhibition of
IGFBP-3 proteolysis at local prostatic levels would favor
enhancement (or inhibition) of tumor growth will have to be
clarified.
[0075] We recently reported significant association of high IGFBP-3
levels in primary breast tumor extracts with unfavorable prognostic
indicators of the disease (56), and more recently found that in
breast nipple aspirate fluid (NAF) IGFBP-3 levels were directly and
IGFBP-3 fragment levels were inversely related to breast cancer
risk (57). IGFBP-3 levels in NAF were also inversely associated
with PSA. It is also noteworthy that anti-proliferative (apoptotic)
properties of IGFBP-3, including ability of IGFBP-3 fragments to
inhibit the mitogenic effects of IGF-I have been described
(8-14).
[0076] Identification of markers with inverse relation in CaP (i.e.
IGF-I and intact IGFBP-3 vs free PSA) prompted the examination of
several concentration ratios and measurement permutations in
relation to PSA. Among the various possibilities, permutations of
total IGF-I/free PSA, intact IGFBP-3/free PSA, (IGF-I/total
IGFBP-3)/free PSA, (intact IGFBP-3/total IGFBP-3)/free PSA, and
(IGF-I+intact IGFBP-3)/free PSA appeared most promising. By ROC
analysis, determination of total IGF-I/free PSA and intact
IGFBP-3/free PSA demonstrated better discriminating potential than
the currently used free PSA/total PSA ratio. Although several other
permutations, notably the ratio of (intact IGFBP-3/total
IGFBP-3)/free PSA showed even better discriminative power, the
relative improvement may not be significant enough to warrant
inclusion of a third measurement component.
[0077] The potential of growth factor/tumor marker permutations
were further confirmed by multivariate analysis, which identified
IGF-I/free PSA and intact IGFBP-3/free PSA as independent
parameters for discriminating between BPH and CaP. As shown in
Table 4, at a cutoff value of 0.28, the free/total PSA ratio
identified 95% of cancer patients (false negative rate of 5%) with
a specificity of 17% (false positive rate of 83%), confirming
previous reported observations (36, 37). On the other hand, the
IGF-I/free PSA ratio at a cut-off value of 273, identified 29/o of
cancer patients with a specificity of 95%. Intact IGFBP-3/free PSA
at a cut-off value of 3500 also detected 29% of patients with
cancer with specificity of 95%. Because the free/total PSA ratio
significantly improves the predictive power of the above
multivariate models, IGF-I/free PSA or intact IGFBP-3/free PSA
ratios could potentially further enhance the diagnostic
differentiation between CaP and BPH. As described earlier, the
observation that the IGF-I/free PSA or intact IGFBP-3/free PSA
ratio could be potentially used to significantly improve the
specificity of CaP detection is of particular importance as it
complements the free/total PSA testing, which has a problem of poor
cancer detection specificity.
[0078] In summary, in a group of subjects with total PSA in the
diagnostic gray-zone, we identified significantly higher IGF-I and
intact IGFBP-3 levels in those patients with CaP than those with
BPH. Among several possibilities, the IGF-I/free PSA and intact
IGFBP-3/free PSA ratios demonstrated potential for significant CaP
diagnostic improvements. As the new permutations of markers appear
to compliment traditional testing (Table 4), analysis of IGF-I/free
PSA or intact IGFBP-3/free PSA may further enhance the diagnostic
discrimination between BPH and CaP. Further, the inclusion of UPS
additional IGF axis parameters, such as IGFBP-2, or studies of
increased sample size may still further improve the discriminating
power of the indicator ratios.
[0079] To our knowledge, this represents the first attempt in
establishing the potential utility of IGFs and/or IGFBPs testing in
relation with a tumor associated biomarker. We believe that the
idea has general applicability in human cancer diagnostics and
could be readily evaluated to involve other representatives of the
IGF system (e.g. IGF-I and IGF-II, and IGFBP-1-9 and IGFBP related
proteins). In light of these findings, application of such vertical
(e.g., IGF-I/intact IGFBP-3 or IGF-I/IGFBP-2) or lateral (e.g.
IGF-I/free PSA or IGFBP-2/free PSA) marker permutations may prove
valuable in differential diagnosis of human cancer in general and
prostate cancer in particular.
EXAMPLE 2
[0080] As indicated above, the ratio of IGF-I/free PSA or intact
IGFBP-3/free PSA in relation to CaP vs BPH (see Table 4) can be
used as a significantly better predictive of BPH than is the
currently used ratio of free/total PSA, which has a proven
specificity problem. Based on the above findings, we predict that
determinations of additional IGF axis components in various
permutations with PSA or other tumor markers may further enhance
discrimination between various cancer and benign conditions.
[0081] In addition to PSA, the kallikreins might represent a tumor
marker that can be used in various permutations with IGF axis
components to enhance the discriminating power of diagnostic tests.
PSA and kallikreins, such as human kallikrein 2 (hK2), are members
of a multigene family of serine proteases that share up to 80%
sequence homology. Further, the kallikreins have been shown to be
expressed in various biological fluids of normal or malignant
origins (58). For example, PSA and hK2 have been detected in serum
and/or tissue extract of patients with breast (59) and lung cancers
(60), as well as in those with prostate cancer. Thus, we further
speculate that determination of ratios of IGFs (free or total
IGF-I, or IGF-II)/kallikrein or intact IGFBP-3 or other
IGFBPs/kallikrein could have potential utility in the detection of
various cancers.
[0082] For example, because high IGF-I levels are also associated
with breast, and lung cancers, one might predict that the indicator
ratios described herein (with or without a tumor marker such as PSA
or a kallikrein) would also be useful predictors for these
conditions. In contrast, over-expression of both IGF-I and IGF-II
has been linked to colorectal cancers. Thus, it might be predicted
that and IGF-I.times.IGF-II or IGF-I+IGF-II, alone or taken
together with a measurement of a suitable tumor marker (such as
CA-19.9), would provide increased predictive power of colorectal
cancers over current methodologies.
[0083] Similarly we predict that determination of the ratios of
IGFs (IGF-I or IGF-II) or IGFBPs (IGFBP-1-6 and/or IGFBP-rP- 1-9)
in relation to PSA or kallikrein proteins (or other tumor markers)
would have increased discriminating power in the differential
diagnosis of prostate, breast and lung and possibly other cancers.
As exemplified for prostate cancer, determination of IGF/IGFBP
ratios could be of significant clinical value in other human
cancers, such as cancer of the breast, colorectal, and lung
cancers. Although a PSA-equivalent marker for these cancers has not
been described as yet, we predict that measurement of the
concentration ratio of IGF-I and/or IGF-II (free or total) in
relation to IGFBPs (IGFBP-1, IGFBP-2, IGFBP-3, IGFBP4, IGFBP-5,
IGFBP-6) could be useful in differentiating cancerous from
non-cancerous disease.
[0084] We further predict that ratios of IGF-I or IGF-II (free or
total) in relation to IGFBP related protein (IGFBP-rPs1-9) which
share significant N-terminal homology with the classical IGFBPs
(IGFBP-1-6), but bind IGFs with lower affinity, could be also of
significant clinical value. To date up to nine different
IGFBP-related binding proteins (IGFBP-rP-1-9) have been identified
(5), and additional IGFBP-rPs will most probably be discovered. For
example, in the case of the prostate cancer, ratios of
IGF-I/IGFBP-rP-1, or ratio of intact IGFBP-3/IGFBP-rP-1 might prove
to be very useful. The same may be also true for other human
cancers which may benefit from determinations of ratios of IGF-I or
IGF-II (free or total)/IGFBP-rPs. In this context, determinations
of IGFBP-2/IGF-I, or IGFBP-2/IGF-II may be of significant clinical
value in relation to colon cancer, while ratios of
IGFBP-2/IGFBP-rP-1 might have diagnostic value in breast tumor.
[0085] We further speculate that the circulating levels of the IGF
superfamily of molecules (5) might be a surrogate marker (18) of
the tissue levels (e.g. prostate, breast and colon) of these
variables. If this prediction is true, then identification of
markers with statistically significant but weak association with a
given cancer at serum level (as a result of blood volume dilution
effect) could lead to identification of markers with significantly
stronger and more distinct association with the cancer at the
tissue level. If the serum/tissue relation is correct, then
measurement of the IGF axis components and one or more tumor
markers at the tissue level (e.g. by immunohistochemical staining
and electron microscopy, or in tumor cell extracts) might be also
highly beneficial.
[0086] Of course, we realize that this general approach of
combining IGF axis component measurements together with a tumor
marker measurement would require systematic evaluation of the
effect of the various markers and their permutations as described
above, using several of the currently available mathematical
regression modeling. The latter could be readily performed in the
hand of an expert statistician and a good computer software (such
as that described above). Experiments are planned to test and
confirm the above predictions.
References
[0087] The following citations are hereby incorporated by reference
into this specification.
[0088] 1. Lamson G, Giudice L, Rosenfeld R. 1991 Insulin-like
growth factor binding proteins: Structural and molecular
relationships [review]. Growth Factors 5:19-28.
[0089] 2. Cohen P, Rosenfeld R G. 1994 Physiologic and clinical
relevance of the insulin-like growth factor binding proteins
[review]. Curr Opin Pediatr 6:462-467.
[0090] 3. Jones J I, Clemmons D R. 1995 Insulin-like growth factors
and their binding proteins: Biological Actions [review]. Endocrin
Rev 16:3-34.
[0091] 4. Rajram S, Baylink D J, Mohan S. 1997 Insulin-like growth
factor-binding proteins in serum and other biological fluids:
regulation and functions [review]. Endocrin Rev 18:801-831.
[0092] 5. Hwa V, Oh Y, Rosenfeld R G. 1999 Insulin-like growth
factor binding proteins: a proposed supperfamily. Acta Paediatr
Suppl 88:37-45.
[0093] 6. Kelley K M, Oh Y, Gargosky S E, Guecv Z, Matsumoto T, Hwa
V, Ng L, Simpson D M, Rosenfeld R G. 1996 Insulin-like growth
factor-binding proteins (IGFBPs) and their regulatory dynamics.
Intl Biochem Cell Biol 28:619-637.
[0094] 7. Ferry Jr R J, Cerri R W, Cohen P. 1999 Insulin-like
growth factor binding proteins: New proteins, new functions. Horm
Res 51:53-67
[0095] 8. Rechler M M. 1997 Editorial: Growth inhibition by
insulin-like growth factor (IGF) binding protein-3 --What's IGF got
to do with it. Endocrinology 138:2645-47.
[0096] 9. Valentinis B, Bhala A, Deangelis T, Baserga R. Cohen
P.1995 The human insulin-like growth factor (IGF) binding protein-3
inhibits the growth of fibroblasts with a targeted disruption of
the IGF-I receptor gene. Mol Endocrinol 9:361-367.
[0097] 10. Zadeh S M, Binoux M. 1997 The 16-kDa proteolytic
fragment of insulin-like growth factor (IGF) binding protein-3
inhibits the mitogenic action of fibroblasts growth factor on mouse
fibroblasts with a targeted disruption of the type-1 IGF receptor
gene. Endocrinology 138:3069-3072.
[0098] 11. Rajah R, Valentinis B, Cohen P. 1997 Insulin-like growth
factor-binding protein-3 induces apoptosis and mediates the effects
of transforming growth factor-b on programmed cell death through a
p53- and IGF-Independent mechanisms. J Biol Chem
272:12181-12188
[0099] 12. Oh Y. 1998 IGF-Independent regulation of breast cancer
growth by IGF binding proteins. Breast Cancer Res Treat
47:283-293.
[0100] 13. Rajah R, Katz L, Nunn S, Solberg P. Beers T, Cohen P.
1995 Insulin-like growth factor binding protein (IGFBP) proteases:
Functional regulations of cell growth. Prog Growth Factor Res
6:273-284.
[0101] 14. Giudice L C. 1995 Editorial: IGF binding protein-3
protease regulation: How sweet it is. J Clin Endocrinol Metab
80:2279-2281.
[0102] 15. Russell P J, Bennett S, Stricker P.1998 Growth factor
involvement in progression of prostate cancer [review]. Clin Chem
44:705-723.
[0103] 16. Holly Jeff. 1998 Insulin-like growth factor and new
opportunities for cancer prevention. Lancet 351:1373-1374.
[0104] 17. Rosen C J, Pollak M. 1999 Circulating IGF-I: New
perspectives for a new century [review]. TEM 10:136-141
[0105] 18. Cohen P. 1998 Serum insulin-like growth factor-I levels
and prostate cancer risk-Interpreting the evidence. J Natl Cancer
Inst 90:876-879
[0106] 19. Baserga R. 1995 The insulin-like growth factor I
receptor: a key to tumor growth? Cancer Res55:249-452.
[0107] 20. Li S L, Goko H, Xu Z D, Kimura G, Sun Y, Kawachi M H,
Wilson T G, Wilczynski S, Fujita-Yamaguchi Y. 1998 Expression of
insulin-like growth factor (IGF)-II in human prostate, breast,
bladder, and paraganglioma tumors. Cell Tissue Res 291:469-479.
[0108] 21. Glick R P, Lichtor T, Unterman T G. 1997 Insulin-like
growth factors in central nervous system tumors. J neurooncol
35:315-3125,
[0109] 22. Lahm H, Amstad P, Wyniger J, Yilmaz A, Fischer J R,
Schreyer M, Givel J C. 1994 Blockade of the insulin-like growth
factor-I receptor inhibits growth of human colorectal cancer cells:
evidence of a functional IGF-II-mediated autocrine loop. Int J
Cancer 58:452-459.
[0110] 23. Grimberg A, Cohen P. 1999 Growth hormone and prostate
cancer: guilty by association? J Endocrinol Invest 22:64-73.
[0111] 24. Colao A, Marzullo P, Spiezia S, Ferone D, Giaccio A,
Cerbone G, Pivonello R, Di Somma C, Lombardi G. 1999 Effects of
growth hormone (GH) and insulin-like growth factor I on prostate
diseases: an ultrasonographic and endocrine study in acromegaly, GH
deficiency, and healthy subjects. J Clin Endo Metab
84:1986-1991
[0112] 25. Cats A, Dullaart R P, Kleibeuker J H, Kuipers F, Sluiter
W J, Hardonk M J, de vires E G. 1996 Increased epithelial cell
proliferation in the colon of patients with acromegaly. Cancer Res
56:523-526.
[0113] 26. Orme S M, Mcnally R J, Cartwright R A, Belchetz P E.
1998 Mortality and cancer incidence in acromegaly: a retrospective
cohort study. United Kingdom Acromegaly Study Group. I Clin Endo
Metab 83:2730-2734.
[0114] 27. Chan J K Stamnpfer M J, Giovannucci E, Gann P H, Ma J,
Wilkinson P, Hennekens C H, Pollak M. 1998 Plasma insulin-like
growth factor-I and prostate cancer risk: A prospective study.
[0115] Science 279:563-566.
[0116] 28. Hankinson S E, Willett W C, Colditz G A, Hunter D J,
Michaud D S, Deroo B, Rosner B, Speizer F E, Pollak M. 1998
Circulating concentrations of insulin-like growth factor-I and risk
of breast cancer. The Lancet 351:1393-1396.
[0117] 29. Wolk A, Mantzoros C S, Andersson S-O, Bergstrom R,
Signorello L B, Lagiou P, Adami H-O, Trichopoulos D. 1998
Insulin-like growth factor-I and prostate cancer risk: a
population-based, case control study. J Natl Cancer Inst
90:911-915.
[0118] 30. Yu H, Spitz M R, Mistry J, Gu J, Hong W K, Wu X. 1999
Plasma levels of insulin-like growth factor-I and lung cancer risk:
a case-control analysis. J Natl Cancer Inst 91:151-156
[0119] 31. Manousos O, Souglakos J, Bosetti C, Tzonou A,
Chatzidakis V, Trichopoulos D, Adami H O, Mantzoros C. 1999 IGF-I
and IGF-IL in relation to colorectal cancer. Int J Cancer
83:15-17.
[0120] 32. Cohen P, Graves H C, Peehl D M, Kamarei M, Giudice L C,
Rosenfeld R G. 1992 Prostate-specific antigen (PSA) is an
insulin-like growth factor binding protein-3 protease found in
seminal plasma. J Clin Endocrinol Metab 75:1046-1053.
[0121] 33. Carter H B, Coffey D S. 1990 The prostate: an increasing
medical problem. Prostate 16:39-48.
[0122] 34. Diamandis E P. 1998 Prostate specific antigen- its
usefulness in clinical medicine. Trends Endocrinol Metab
9:310-316
[0123] 35. Catalona W J, Smith D S, Ornstein D K. 1997 Prostate
cancer detection in men with serum PSA concentrations of 2.6-4.0
ng/mL and benign prostate examination. Enhancement of specificity
with free PSA measurements. JAMA 277:1452-1455.
[0124] 36. Luderer A A, Chen Y T, Soriano T F, Kramp W J, Carlson
G, Cuny C, Sharp T, Smith w, et al. 1995 Measurement of the
proportion of free to total prostate-specific antigen improves
diagnostic performance of prostate-specific antigen in the
diagnostic gray zone of total prostate-specific antigen. Urology
46:187-194.
[0125] 37. Catalona W J, Partin A W, Slawin K M, Brawer M K,
Flanigan R C, Patel A, Richie J P, deKernion J B, et al. 1998 Use
of percentage of free prostate-specific antigen to enhance
differentiation of prostate cancer from benign prostate disease: a
prospective multicenter clinical trial. JAMA279:1542-1547.
[0126] 38. Khosravi M J, Diamandi A, Mistry J, Lee P D K. 1996 A
non-competitive ELISA for human serum insulin-like growth factor-I.
Clin Chem 42:1147-1154.
[0127] 39. Lee P K D, Mistry J, Hintz R L. 1996 Active insulin-like
growth factor-I (IGF-I) assays. Webster: Diagnostic Systems
Laboratories.
[0128] 40. Yu H, Mistry J, Nicar M J, Khosravi M J, Diamandis A,
Doom J V, Jull A. Insulin-like growth factors (IGF-I, Free IGF-I,
and IGF-II) and insulin-like growth factor binding proteins
(IGFBP-2, IGFBP-3, IGFBP-6, and ALS) in blood circulation. 1999 J
Clin Lab Analysis 13:166-172.
[0129] 41. Khosravi J, Diamandi A, Mistry J, Krishna R G. 1999 The
high molecular weight insulin-like growth factor binding protein
complex: Epitope mapping, immunoassay, and preliminary clinical
evaluation. J Clin Endocrinol Metab 84:2826-2833.
[0130] 42.Khosravi M J, Diamandi A, Mistry J. 1997 Immunoassay of
insulin-like growth factor binding protein-1. Clin Chem
43:523-532.
[0131] 43. Diamandi A, Mistry J, Khosravi M J. Specificity of new
immunoassays for human insulin-like growth factor binding protein-3
(IGFBP-3). Proc of the 80.sup.th Annual Meet of the Endocrine Soc.
1998; P3-360.
[0132] 44. Diamandi A, Mistry J, Krishna R G, Khosravi M J. 1999
Immunoassay of insulin-like growth factor binding protein-3: New
means to quantifying IGFBP-3 proteolysis. I Clin Endocrinol Metab
(submitted).
[0133] 45. Woodrum D L, French C M, Hill T M, Roman S J, Slatore H
L, Shaffer J L, York L G, Eure K L, Loveland K G, Gasior G H,
Southwick P C, Shamel B. 1997 Analytical performance of the
Tandem-R free PSA immunoassay measuring free prostate-specific
antigen. Clin Chem 15::, 43:1203-1208.
[0134] 46. Juul A, Main K, Blum W F, Lindholm, J, Ranke M B, and
Shakkeback N E. 1994 The ratio between serum levels of insulin-like
growth factor (IGF)-I and the IGF binding proteins (IGFBP)-1, 2,
and 3 decreases with age in healthy adults and is increased in
acromegalic patients. Clin Endocrinol 41:85-93.
[0135] 47. Thiissen J P, Ketelslegers J M, Underwood L E, 1994
Nutritional regulation of the insulin-like it growth factors.
Endocr Rev 15:80-101.
[0136] 48. Harrela M, Koistinen It Kapiro J, Lehtovirta M,
Tuomilehto J, Erikisson J, Toivanen L, Koskenvuo M, Leinonen P,
Koistinen R, Seppala M. 1996 Genetics and environmental components
of interindividual variation in circulating levels of IGF-I,
IGF-II, IGFBP-1, and IGFBP-3. J Clin Invest 98:2612-2615.
[0137] 49. Kao P C. Matheny A P, Lang C A. 1994 IGF-I comparisons
in healthy twin children. J Clin Endocrinol Metab. 78:310-312.
[0138] 50. Khosravi M J, Diamandis A, Mistry J. 1998 Factors
influencing immunoassay levels of free insulin-like growth factors
(IGFs) as determined by new ELISAs. Proc of the 80.sup.th Annual
Meet of the Endocrine Soc. 1998; P3-355.
[0139] 51 Baciuchka M, Remacle-Bonnet M, Garrouste F, Favre R,
Sastre B, Pommier G. 1998 Insulin-like growth factor (IGF)-binding
protein-3 (IGFBP-3) proteolysis in patients with colorectal cancer:
possible association with the metastatic potential of the tumor.
Int J Cancer 79:460-467.
[0140] 52. Stenman U-H, Leinonen J, Alfthan H, Rannikko S, Tukhanen
K, Alfthan 0. 1991 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
51:222-226.
[0141] 53. Cho N H, Park C, Park D S. 1997 Expression of alpha
1-antichymotrypsin in prostate carcinoma. J Korean Med Sci.
12:228-33.
[0142] 54. Chen Z, Chen H, Stamey T A. 1997 Prostate specific
antigen in benign prostatic hyperplasia: purification and
characterization. J Urol 157:2166-70.
[0143] 55. Diamandis E P. 1998 Prostate specific-antigen--its
usefulness in clinical medicine. Trends Endocrinol Metab
9:310-316.
[0144] 56. Yu K Levesque M A, Khosravi M J, Papanastasiou-Diamandi
A, Clark G M, Diamnandis E P. 1998 Insulin-like growth factor
binding protein-3 and breast cancer survival. Int I Cancer.
79:624-628.
[0145] 57. Sauter E R, Litwin S, Engstrom P F, Diamandis A,
Khosravi J, Diamandis E P. 1999 Insulin-like growth factor binding
protein-3 (IGFBP-3) is associated with breast cancer risk. Cancer
Epidemiolog Biomarkers and Prev (submitted).
[0146] 58. Youse G M, Obiezu C V, Luo L Y, Black M H Diamandis E P.
Prostase/KLK-1 is a new member of the human kallikrein gene family,
is expressed in prostate and breast tissues, and is hormonally
regulated. Cancer Res. 59:4252-4256, 1999.
[0147] 59. Borchert G H, Yu H, Tomlinson G, Giai M, Roagna R,
Ponzone R, Sgro L, Diamandis E P. Prostate specific antigen
molecular forms in breast cyst fluid and serum of women with
fibrocystic breast disease J Clin Lab Anal 13:75-81, 1999
[0148]
[0149] 60. Zarghami N, D'costa M, Tsuyuki D, Asa S L, Diamandis E
P. Expression of the prostate-specific antigen by lung tissue. Clin
Cancer Res 3:1201-6, 1997.
* * * * *