-
Research Article
A Novel Quantitative Multiplex TissueImmunoblotting for
Biomarkers Predicts aProstate Cancer Aggressive PhenotypeGuangjing
Zhu1, Zhi Liu1, Jonathan I. Epstein2, Christine Davis1,Christhunesa
S. Christudass1, H. Ballentine Carter1, Patricia Landis1, Hui
Zhang2,Joon-Yong Chung3, Stephen M. Hewitt3, M. Craig Miller4, and
Robert W.Veltri1
Abstract
Background: Early prediction of disease progression in menwith
very low-risk (VLR) prostate cancer who selected activesurveillance
(AS) rather than immediate treatment could reducemorbidity
associated with overtreatment.
Methods: We evaluated the association of six
biomarkers[Periostin, (�5, �7) proPSA, CACNA1D, HER2/neu, EZH2,
andKi-67] with different Gleason scores and biochemical
recurrence(BCR) on prostate cancer TMAs of 80 radical prostatectomy
(RP)cases. Multiplex tissue immunoblotting (MTI) was used to
assessthese biomarkers in cancer and adjacent benign areas of 5
mmsections. Multivariate logistic regression (MLR) was applied
tomodel our results.
Results: In the RP cases, CACNA1D, HER2/neu, and Peri-ostin
expression were significantly correlated with aggressivephenotype
in cancer areas. An MLR model in the cancerarea yielded a ROC-AUC ¼
0.98, whereas in cancer-adjacent
benign areas, yielded a ROC-AUC ¼ 0.94. CACNA1D andHER2/neu
expression combined with Gleason score in a MLRmodel yielded a
ROC-AUC ¼ 0.79 for BCR prediction. In thesmall biopsies from an AS
cohort of 61 VLR cases, anMLR model for prediction of progressors
at diagnosis retained(�5, �7) proPSA and CACNA1D, yielding a
ROC-AUC of0.78, which was improved to 0.82 after adding tPSA into
themodel.
Conclusions: Themolecular profile of biomarkers is capable
ofaccurately predicting aggressive prostate cancer on
retrospectiveRP cases and identifying potential aggressive prostate
cancerrequiring immediate treatment on the AS diagnostic biopsy
butlimited in BCR prediction.
Impact: Comprehensive profiling of biomarkers using MTIpredicts
prostate cancer aggressive phenotype in RP and AS bio-psies. Cancer
Epidemiol Biomarkers Prev; 24(12); 1864–72. �2015 AACR.
IntroductionProstate cancer is the most common cancer in men in
the
United States and is the second most common cause of cancerdeath
in men. Prostate cancer usually occurs after age 50 and
theincidence increases with age. There are 233,000 estimated
newcases and 29,480 estimated prostate cancer deaths in 2014 in
theUnited States (1). Radical prostatectomy (RP) is an
effectivetreatment for patients with organ-confined disease and has
beendemonstrated to reduce the risk of death fromprostate cancer
(2).
Nearly 40% of prostate cancer patients who choose
definitivetherapy will undergo RP. In 38% to 52% of cases,
advanceddisease with potentially bad prognosis are found in
surgicalspecimens (3). Each category of extraprostatic disease is
associatedwith significantly increased risk of cancer recurrence
and progres-sion, measured at the earliest time with a detectable
prostate-specific antigen (PSA; >0.20 ng/mL), or biochemical
recurrence(BCR; ref. 4). The natural history of prostate cancer
progressionafter BCR following surgery can be highly variable
(i.e., 3–13years); however, at least two thirds of BCR patients
developdisease spread if left untreated and many will die because
ofdistant progression (5).
In most cases, BCR is used as a surrogate measure for
moreclinically meaningful critical endpoints such as distant
progres-sion or cancer-specific mortality. Inaccurate risk
classificationscould result in inappropriate or unnecessary
treatments. Unfor-tunately, existing tools (e.g., nomograms like
Shariat, ref. 6;Swanson, ref. 7; and Capra, ref. 8) that rely
solely on clinicalvariables such as PSA velocity, grade, and stage
are unable topredict which men will go on to metastasis and
ultimatelyprostate cancer-specific death. These tools may be unable
toidentify those men that have already demonstrated BCR.
There-fore, a better method to predict whether a prostate cancer
patienthas a more aggressive phenotype at surgery will help to
standard-ize treatments of the more aggressive cancer patients
early andefficiently and spare patients of less aggressive cancer
from themorbidity associated with adjuvant therapy.
1The James Buchanan Brady Urological Institute, The Johns
HopkinsUniversity School of Medicine, Baltimore, Maryland.
2Department ofPathology, The Johns Hopkins Hospital, Baltimore,
Maryland. 3Exper-imental Pathology Laboratory, Laboratory of
Pathology, Center forCancer Research, National Cancer Institute,
NIH, Bethesda, Maryland.4Statistical Consultant, Lindale,
Texas.
Note: Supplementary data for this article are available at
Cancer Epidemiology,Biomarkers & Prevention Online
(http://cebp.aacrjournals.org/).
Current address for C.S. Christudass: Department of Neurological
Sciences,Christian Medical College, Vellore, Tamil Nadu, India.
Corresponding Author: Robert W. Veltri, The Brady Urological
Institute, TheJohns Hopkins University School of Medicine, 600
North Wolfe Street, Balti-more, MD 21287. Phone: 410-955-6380; Fax:
410-502-7711; E-mail:[email protected]
doi: 10.1158/1055-9965.EPI-15-0496
�2015 American Association for Cancer Research.
CancerEpidemiology,Biomarkers& Prevention
Cancer Epidemiol Biomarkers Prev; 24(12) December 20151864
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
PSA screening and prostate biopsies have resulted in
overdi-agnosis and overtreatment of prostate cancer (9–11). In
fact, low-grade, low-stage prostate cancer of up to approximately
56%menremained undetected during their lifetime (9–12). In the
presentera, 30 men require treatment for prostate cancer in order
to saveoneman (13). Such overtreatment can always create a chance
of adecreased quality of life once sexual function andurinary
functionare compromised (14). The accurate identification of men
withprostate cancer that are destined to progress to life
threateningdisease and who will benefit from curative
interventionmust be ahigh research priority, with the goal of
reducing unnecessarytreatments.
In 1995, Dr. Ballentine Carter started an active surveillance
(AS)program with the delayed surgical intervention as a
treatmentoption in Johns Hopkins Medical Institutions (JHMI,
Baltimore,MD). Patients are enrolled if theymeet Epstein inclusion
very low-risk (VLR) criteria (15–17). These criteria include PSA
density(PSAD) 6 on follow-upbiopsy/biopsies or retropubic radical
prostatectomy criteriawere met during AS. For each patient,
continuous 5 mm tissuesections were cut and used either for H&E
staining and path-ologic assessment or for profiling biomarkers
using MTI. TheH&E slides were first reviewed by a pathologist
and the cancerarea were marked for reference of data analysis later
on. Onlyslides with cancer and adjacent to the corresponding read
H&Eslides were used for MTI study.
All cases used for TMAs and AS biopsies were consented underan
Institutional Review Board-approved protocol at Johns Hop-kins
University School of Medicine.
Biomarkers Predict a Prostate Cancer Aggressive Phenotype
www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 24(12)
December 2015 1865
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
Multiplex tissue immunoblottingMTI is a method that can be used
to detect multiple molecular
targets in FFPE tissues, while retaining both quantitative
andhistomorphologic diagnostic and prognostic features
(24–26).Starting with a 5-mm FFPE tissue section on a standard
glass slide,proteinswere transferred from the tissue sections
(TMAor biopsy)onto a series of overlapped thin membranes (P-Film,
20/20GeneSystems, Inc.). Each membrane was probed with one of
thesix biomarkers [CACNA1D, Periostin,HER2/neu, EZH2,Ki67 and(�5,
�7 proPSA)] followed by incubation with FITC-conjugatedsecondary
antibodies. Total proteins collected on the blottedmembranes were
biotinylated, and followed by incubation withstreptavidin-linked
Cy5. The fluorescence signals of biomarkersand total proteins were
acquired using a Typhoon 9410 imager(GE Healthcare) and quantified
with ImageQuant5.2 software(GE Healthcare). The quantified signals
of biomarkers weredivided by that of total protein for
normalization. The log-transformed ratio was used for downstream
data analysis. Sup-plementary Table S1 provides the detailed
information about theprimary antibodies and dilutions used in the
study. The quality ofthe antibodieswas checkedusingWestern blot
analysis of prostatecancer cell lines (data not shown). This MTI
method helps topreserve valuable tissue and while multiplexing
prognostic bio-markers for prostate cancer.
Statistical analysisGraphPad Prism 6 software was used to
generate scatterplots
of the biomarker expressions within the different Gleason
scoregroups (3þ3, 3þ4, 4þ3, and �8) as well as to generate
thepredictive probability plots. One-way ANOVA analysis fol-lowed
by Dunnet multiple comparisons test was used to
evaluate the biomarker expressions in the four Gleason
scoregroups. STATA 13.0 (STATA, StataCorp LP) was used for
allstatistical modeling. Statistical significance was defined as a
P <0.05. Statistical modeling included standard and/or
backwardsstepwise multivariate logistic regression (MLR) to
discriminateless aggressive from more aggressive prostate cancer.
Decisioncurve analysis (DCA; refs. 27–29) was used to evaluate
differentMLR models for the prediction of more aggressive
prostatecancer, patient that experienced BCR, and progressors. DCA
is amethod for evaluating and comparing different predictionmodels
(27–29), which gives an expected net benefit perpatient relative to
the assumption that all patients are treatedor not treated. For
these evaluations, treatment was defined asthe
significant/interested changes in the patients (aggres-siveness,
BCR, progressors, etc.). The interpretation of netbenefit is made
by comparing a model curve to a baselinecurve where all patients
are considered as interested changes,and if the model curve is
above the standard curve at a certainprobability, then the model
would be considered a betterpredictor of the outcome.
ResultsDifferential expression of CACNA1D, HER2/neu, and
Periostinin prostate cancer with different Gleason scores
We optimized all MTI primary antibodies for six of our
inter-ested biomarkers on a test TMA (TMA 475) containing
normalcontrol and prostate cancer tissue samples. Our results
showedthat proteins on a single 5 mm FFPE tissue slide could be
success-fully transferred onto a series of six membranes and probed
withsix biomarkers simultaneously (Fig. 1).
Table 1. Prostate cancer patients demographics in TMA681 and 682
(N ¼ 80)Recurrence
Variable Subgroups No Yes Total
Age 59.04 � 6.57 58.82 � 5.80Gleason score 6 7 1 8
7 (3þ4) 14 3 177 (4þ3) 15 3 18� 8 11 15 26
TNM stage T2 26 2 28T3A 16 15 31T3B 3 5 8
Gland weight (g) 51.94 � 14.90 56.19 � 21.38PSA (ng/mL) 8.29 �
5.29 13.17 � 10.15PSAD (ng/mL/g) 0.17 � 0.11 0.25 � 0.20Exposure
(y) 5.06 � 4.50 3.82 � 2.36Race A 6 3 9
H 1 0 1O 3 1 4C 37 18 55
Surgical margin Negative 41 16 57Positive 6 6 12
Seminal vesicle status Negative 46 16 62Positive 6 6 12
Lymph node status Negative 46 22 68Positive 1 0 1
Capsular penetration (fcp) Negative 38 16 54Positive 8 6 14
Capsular penetration (ecp) Negative 34 8 42Positive 13 14 27
Organ confinement status Nonconfined 21 20 41Confined 26 2
28
NOTE: Age, gland weight, PSA, PSAD, exposure data were shown as
mean� SD. For race data, A, African American; H, Hispanic; O,
Others; C, Caucasian. PSAD (ng/mL/g) ¼ PSA (ng/mL)/gland weight
(g).
Zhu et al.
Cancer Epidemiol Biomarkers Prev; 24(12) December 2015 Cancer
Epidemiology, Biomarkers & Prevention1866
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
After optimization of the six biomarkers, we performed MTIwith
these six biomarkers on our study, TMAs sections containingcancer
and caner-adjacent benign tissue from80different prostatecancer
patients and determined the relative expression of thebiomarkers.
These sections include four groups with differentGleason scores
(3þ3, 3þ4, 4þ3, and � 8). Among the six
biomarkers, CACNA1D, HER2/neu, and Periostin were
differen-tially regulated among the four Gleason score groups in
cancerarea and cancer-adjacent benign areas. In the cancer areas,
CAC-NA1D and HER2/neu relative expression were significantly
lowerin the Gleason score groups of 4þ3 and � 8 compared with
theGleason score group of 3þ3 (P < 0.05). There was a trend
of
Figure 1.Test of antibodies on MTI using TMA475. Representative
results of MTIusing the six biomarkers. Signalintensity:
white–red–yellow–green–blue–black from maximum tominimum.
0.6
0.4
0.2
0.0
-0.2
-0.4
0.6
0.4
0.2
0.0
-0.2
-0.4
0.4
0.2
0.0
-0.2
-0.4
0.4
0.2
0.0
-0.2
-0.4
-0.6
3+3 3+4 4+3 8 and above
3+3 3+4 4+3 8 and above
3+3 3+4 4+3 8 and above 3+3 3+4 4+3 8 and above
3+3 3+4 4+3 8 and above
3+3 3+4 4+3 8 and above
Gleason score
Gleason score
Gleason score Gleason score
Gleason score
Gleason score
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
2
1
0
-1
Exp
ress
ion
Exp
ress
ion
Exp
ress
ion
Exp
ress
ion
Exp
ress
ion
Exp
ress
ion
Cancer area Cancer adjacent benign area
CACNA1D CACNA1D
HER2/neu HER2/neu
Periostin Periostin
3+33+44+38 and above
3+33+44+38 and above
3+33+44+38 and above
3+33+44+38 and above
3+33+44+38 and above
3+33+44+38 and above
A
B
C F
E
D
Figure 2.Expression of biomarkers in cancer areas and
cancer-adjacent benign areas. A–C, prostate cancer areas. D–F,
cancer-adjacent benign areas. Data are shown asmean� SD. � and �� ,
statistical significance (P < 0.05) between Gleason score groups
of 4þ3 and �8 versus the Gleason score group of 3þ3,
respectively.
Biomarkers Predict a Prostate Cancer Aggressive Phenotype
www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 24(12)
December 2015 1867
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
gradually decreased relative expression of both markers
withincreased Gleason scores (Fig. 2A and 2B). Pearson
correlationanalysis showed that CACNA1Dhad a strong negative
correlationwith Gleason score (correlation coefficient ¼ �0.73, P
< 0.001),whereas HER2/neu showed amoderate negative correlation
(cor-relation coefficient¼�0.39, P¼ 0.001). Interestingly,
CACNA1Dand HER2/neu expression behaved similarly in the
cancer-adja-cent benign areas (Fig. 2A and 2B vs. 2D and 2E),
suggesting apossible field effect for these markers. Periostin
expression wassignificantly higher in prostate cancer
caseswithGleason score�8compared with that of Gleason score 3þ3 in
cancer-adjacentbenign areas (P < 0.05, Fig. 2F). Although
Periostin expressionwas not statistically significantly different
in cancer areas, therewas a trend for the expression to increase
with increasing Gleasonscore (Fig. 2C).
Biomarker expression in TMAs distinguishes less aggressivefrom
more aggressive prostate cancer
Although the intermediate risk group of clinically
localizedprostate has a Gleason score of 7, it includes both
Gleason 3þ4and Gleason 4þ3, patients with the two pathologic
statuses havesignificantly different treatment: most Gleason 3þ4,
if it is organ-confined, could bemanaged conservativelywhile all
Gleason4þ3necessitates immediate interventionwith surgery and/or
adjuvanttherapy. Therefore, we grouped Gleason score 3þ3 and 3þ4
casesas a less aggressive phenotype and Gleason score 4þ3 and
�8cases as a more aggressive phenotype. Using MLR, our resultsshow
that a model retaining CACNA1D, HER2/neu, and Ki67
expression distinguishes less aggressive and more
aggressiveprostate cancer in cancer areas (Fig. 3A and
Supplementary TablesS2 and S3). Using this model, the predictive
probability was closeto 100% formost cases in the aggressive groups
(Gleason 4þ3 and� 8; Fig. 3B). The ORs of Ki-67, CACNA1D, HER2/neu
were4.41e7,
-
respectively. DCA curve showed only a limited power for
predic-tion of BCRwithGleason score alonewithin the probability
rangeof 0.10 to 0.50, while combination of Gleason score
withmolecular biomarkers showed improved prediction
capabilitywithin a wider probability range (0.10–0.80; Fig. 4B).
The pre-dictive probability of BCR using Gleason score alone or
combi-nation of Gleason score with MBs is shown in Fig. 4C and
4D,respectively.
Biomarkers and tPSA separate progressors and nonprogressorsin
the active surveillance cohort
Using the biomarkers evaluated in the TMAs, we also evaluatedthe
predictive ability of the biomarkermolecular profiles in a totalof
61 AS biopsy cases (32 nonprogressors and 29 progressors). Inthe AS
program, progressors are defined as men that met all theabove VLR
criteria at entry (tumor increased in volume and/orGleason score
> 6); however, duringmonitoring, they were foundto have a more
aggressive prostate cancer than entry criteria onbiopsy follow-up
(i.e., AS event; or clinically aggressive prostatecancer). An MLR
model that retained (�5, �7) proPSA andCACNA1D was able to
distinguish progressors from nonprogres-sors in the AS cohort (Fig.
5A). This model had an accuracy of72.13% in the prediction of
progressors. The ORs for (�5, �7)proPSA and CACNA1D are 0.20 and
7.93, respectively. Thepredictive probability of this model is
shown in Fig. 5D.
We further tried to separate progressors and nonprogressors
inthe AS cohort by using prebiopsy tPSA alone or adding the
tPSAresult to the biomarker model. We found that tPSA alone
haslimited prediction capability with an ROC-AUC of 0.73,
whereascombining molecular biomarkers and tPSA improved the ROC-AUC
of the model to 0.83, but without significant differencecompared
with the biomarker-only model (P ¼ 0.26). The com-binationmodel
(tPSA-MB) had a sensitivity of 79.31%, specificityof 78.13%, with
the tPSA had an OR of 1.24 in this model. The
combined model could correctly classify the two groups with
anaccuracy of 78.69% (Fig. 5A and Supplementary Tables S4 andS5). A
plot of the predictive probabilities is shown in Fig. 5C and5E. DCA
curve showed that the tPSA only and the molecularbiomarker-only
model improved the net benefit within the prob-ability range of
0.35 to 0.80 and 0 to 0.80, respectively, while thecombined tPSA-MB
model improved the net benefit within theprobability range of 0 to
0.75. DCA curve comparing the threemodels to each other showed that
the combined model wassuperior to the biomarker-only model within
the probabilityrange of 0.30 to 0.65 (Fig. 5B). This suggests that
a subset of thebiomarkers evaluated have the potential to
differentiate progres-sors from nonprogressor patients that can
remain annual mon-itoring in our preliminary data of 61 biopsy
cases with highspecificity. Adding tPSA improves the model,
especially withinthe probability range of 0.30 to 0.65.
DiscussionThe development and validation of sensitive andmore
accurate
methods for early detection are pivotal in the management
ofprostate cancer, the overdiagnosis and overtreatment of which is
asignificant healthcare problem (13). AS has become the
preferredmethod for the management of VLR and many LR
patients.Currently, the new NCCN guidelines have included VLR and
LRgroup of prostate cancer for AS (23). For VLR or LR
cases,abnormal morphologic changes are minimal, and the futuretumor
progress could not be effectively predicted by pathologicgrading.
Meanwhile, molecular level changes do occur and maysignal the
potential disease progression. Characterization of pros-tate cancer
necessitates a comprehensive panel of biomarkersconsidering the
heterogeneity of prostate cancer biology (30).Therefore, we reason
that a comprehensive panel of biomarkersinvolving the glandular
acini and the stromal areas couldbeuseful
0.00 -1
00.
10.
20.
3
0.25
0.50
1.00
0.75
0.00 0.25 0.50 0.75 1.00
Sen
siti
vity
Net
ben
efit
1 – Specificity Threshold probability
0 0.2 0.4 0.6 0.8 1
AllGS
GS
NoneGS & MB
GS & MBReference
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
No recurrence Recurrence No recurrence Recurrence
GS GS & MB
Pre
dic
tive
pro
bab
ility
A
C D
B
Figure 4.Prediction of BCR using Gleason scorealone and Gleason
score combinedwith biomarkers. Prediction BCR withGleason score
only (A) versus Gleasonscore and biomarkers (B). DCA of
BCRprediction using the GS or GS and MBmodel in cancer areas (C and
D).Predictive probability of BCR usingthe GS or GS and MB model in
cancerareas.
Biomarkers Predict a Prostate Cancer Aggressive Phenotype
www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 24(12)
December 2015 1869
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
in the prediction of VLR and LR prostate cancer progression in
anAS monitoring program.
To make full use of the valuable biopsy samples with limitedsize
of cancer area, we adopted MTI as a method to profile theexpression
of six biomarkers on a single 5-mm section in TMAs of80 RP cases.
Using a backward stepwise MLR model, we foundCACNA1D, HER2/neu, and
Periostin expression could signifi-cantly distinguish less
aggressive from more aggressive prostatecancer with high
sensitivity and specificity in the cancer area.Interestingly, in
the cancer-adjacent benign area, a different MLRmodel retains
CACNA1DandPeriostin could also distinguish lessaggressive from more
aggressive prostate cancer, implying theimportance of these
biomarkers in the prediction of aggres-siveness in RP cases.
Moreover, our data of the cancer-adjacentbenign area support the
concept of a possible field effect thatinvolves changes of tumor
progression at the molecular levelbefore changes in
glandularmorphology.Ourmodel also showedan improved prediction of
BCR compared with the predictionusing Gleason score alone.
Furthermore, in a total of 61 AS cases,we successfully separated
progressors and nonprogressors in theAS cohort with (�5, �7) proPSA
and CACNA1D among theseverified biomarkers, which are retained in a
MLR model formolecular biomarkers to predict the AS cases that
require defin-itive treatment.
It is worthwhile to note that our RP long-term follow-up
MLRbiomarker model differs from the cancer area and the
cancer-adjacent benign area suggests that biology of molecular
altera-tions in progression and changes in the field may differ in
thecancer areas. This is supported by a systemic genomic analysis
ofprostate cancer andmorphologically normal tissue (30).
Notably,the changes of biomarkers in the cancer-adjacent benign
areashow a slightly different pattern compared with the cancer
area,with changes of stromal protein Periostin expression being
moresignificant. Further characterization of the molecular features
ofprostate cancer with VLR or LR biomarker expression patterns
inthese two histologic areas could provide an additional means
tostratify the AS and RP cases to more accurately assess the
progres-sion of prostate cancer and effectively personalize the
manage-ment of treatment.
Interestingly, CACNA1D stands out as a new useful marker
inprostate cancer area, cancer-adjacent benign area, and also
biop-sies in the prediction of cancer progression (upgrading).
CAC-NA1D is an ERG target gene upregulated in TMPRSS2-ERG–positive
prostate cancer cells (31). Overexpression of CACNA1Din prostate
cancer cells was reported to regulate the ligand inde-pendent
activation of AR. ERG-induced expression of CACNA1Dwas reported to
promote entry of calcium ions into cytosol (32).Here, we found that
the expression of CACNA1D showed a trend
0.00
0.25
0.50
0.75
1.00
0.00 0.25 0.50 0.75 1.00
Sen
siti
vity
Net
ben
efit
1 – Specificity Threshold probability
0 0.2 0.4 0.6 0.8 1
00.
10.
20.
30.
40.
5
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0
1.2
0.8
0.4
0.0Pre
dic
tive
pro
bab
ility
Nonprogressor Progressor Nonprogressor Progressor Nonprogressor
Progressor
tPSA
tPSA
tPSA
MB
MB
MB
MB & tPSA
MB & tPSA
MB & tPSA
Reference
All None
A
C D E
B
Figure 5.Prediction of active surveillance progressor status
with tPSA, biomarkers alone, biomarkers and tPSA. A, ROC comparison
of models using tPSA, molecularbiomarkers alone, and biomarkers
combined with tPSA. B, DCA of the three models. Progressor
predictive probability of tPSA-only (C), biomarker-only (D),and
biomarkers and tPSA combined model (E).
Zhu et al.
Cancer Epidemiol Biomarkers Prev; 24(12) December 2015 Cancer
Epidemiology, Biomarkers & Prevention1870
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
of decrease with increasing Gleason score in RP
specimens.However, the progressor AS biopsies have relatively
higher expres-sion of CACNA1D as an early event that consists with
results inour RP model for 3þ3 and 3þ4 Gleason scores. It has
beenreported that blocking L-type calcium channel-mediated Ca2þ
influx suppressed castration-induced apoptotic cell death in
pros-tate epithelial cells (33, 34). We presume that the reduced
expres-sion of CACNA1D in more aggressive prostate cancer and
pro-gressorsmay contribute to decreasedCa2þ influx and results in
theescape of apoptosis, thus suggesting that decreased
CACNA1Dexpression could be a more aggressive phenotype in higher
gradeprostate cancer or progressors in AS.
The role of HER2/neu in prostate cancer remains controver-sial
(35, 36). HER2/neu can often be overexpressed in prostatecancer
(37, 38), but may depend on the antibody reagents usedand two
differentially expressed forms of HER2/neu transcriptswere reported
(39), which may need to be considered duringinterpretation of
results. Antibodies detecting variable epitopesof HER2/neu may be
one reason for the controversy. In thecurrent study, we showed that
in cancer area, along withCACNA1D and Gleason score, HER2/neu could
be used toefficiently predict aggressiveness and recurrence of
prostatecancer in RP or biopsy specimens.
Moreover, prostate cancer progression involves both epithe-lial
changes and stromal changes. Periostin was identified as
animportant biomarker well correlating with the aggressive
phe-notype (40–42). Both stromal and epithelial Periostin
expres-sion was reported to be significantly increased in tumor
tissues:stromal expression was significantly higher than
epithelialexpression as compared with normal tissue. Although
highstromal expression was significantly associated with
shortersurvival, a low epithelial score significantly correlated
withshorter PSA-free survival, suggesting that Periostin
apparentlyplays an opposing biologic role depending on its tissue
local-ization (43). Here, we included Periostin in our
biomarkerpanel quantification in both epithelia and stroma since
our MTImethod could not separate the expression of these two
areas.Our results suggest that along with CACNA1D, Periostin
couldpredict the malignancy of prostate cancer based on the
cancer-adjacent area (Fig. 3B).
Our MLR modeling system showed its power in the predic-tion of
prostate cancer progression in biopsies, and RP caseswith high
sensitivity and specificity. Our early data were val-idated on 80
RP cases of our TMAs and 61 AS biopsies atdiagnosis (i.e., entry to
AS). In the latter case, the AS study wascompromised by the fact
that the total cancer surface area in theAS biopsies is often
limited. Hence, due to this minimal cancerarea in VLR prostate
cancer, the expansion of the AS biopsysample size is needed for
obtaining additional results for theverification of the statistical
stability of our models. Alterna-tively, the AS program can be
altered to expand criteria for entryinto the program, which is
already being done at JHMI andother institutions (21, 44). In
addition, MTI shows its advan-tage in multiplexing up to six
biomarkers using the same 5-mmtissue section, it does have
technical disadvantages: it relies onthe relatively high expression
of protein and the use of goodquality antibodies. The detection of
lowly expressed proteinlevels is limited, such as Ki-67 in AS
biopsies. Finally, even ifsuper-thin P-film was used in MTI, the
protein transferredonto the membranes shows a gradual decrease from
the oneat the bottom closest to the tissue to the one on top.
We
normalized the expression of each biomarker to the totalprotein
to balance these differences in total protein contenton the same
membrane.
Notably, we have developed a panel of six biomarkersprofiled on
the same 5-mm biopsy tissue slide using MTI, anintegrated,
high-throughput quantitative molecular biomarker-based method for
predicting prostate cancer progression. Ourmodel involving these
biomarkers shows robust strength in theprediction of aggressiveness
in the RP cases and more impor-tantly, this method and biomarkers
could be used to predict theprogress of VLR and LR prostate cancer
at diagnosis, which willreduce the overdiagnosis and overtreatment
and greatly facil-itate the effective management of prostate cancer
patients.Future studies will significantly expand the number of
biopsycases studied and also include nuclear morphometry of
Feulgen(DNA)-stained or H&E-stained biopsies of AS patients as
avariable to more accurately predict the prostate cancer
aggres-sive phenotype of progressors. Given the new problems
ofmanaging AS cohorts, new criteria for selection and new
tech-nology for multiplexing tissue biomarkers are needed to
sup-port the advancement of this concept to improve healthcare
inthe future for prostate cancer.
Disclosure of Potential Conflicts of InterestC.S. Christudass
has ownership interest in a patent filed in the United States
(PCT/US2015/015074). No potential conflicts of interest were
disclosed by theother authors.
Authors' ContributionsConception and design: Z. Liu, C.S.
Christudass, H. Zhang, S.M. Hewitt, R.W.VeltriDevelopment of
methodology: G. Zhu, Z. Liu, C.S. Christudass, S.M. Hewitt,R.W.
VeltriAcquisition of data (provided animals, acquired and managed
patients,provided facilities, etc.): G. Zhu, Z. Liu, J.I. Epstein,
C. Davis, C.S. Christudass,H.B. Carter, P. LandisAnalysis and
interpretation of data (e.g., statistical analysis,
biostatistics,computational analysis): G. Zhu, Z. Liu, C. Davis,
M.C. Miller, R.W. VeltriWriting, review, and/or revision of
themanuscript:G. Zhu, Z. Liu, J.I. Epstein,C.S. Christudass, H.
Zhang, J.-Y. Chung, M.C. Miller, R.W. VeltriAdministrative,
technical, or material support (i.e., reporting or organizingdata,
constructing databases): G. Zhu, C. Davis, H.B. Carter, P. Landis,
J.-Y.Chung, S.M. Hewitt, M.C. Miller, R.W. VeltriStudy supervision:
G. Zhu, H.B. Carter, H. Zhang, R.W. Veltri
AcknowledgmentsThe authors appreciate the help of Department of
Pathology in Johns
Hopkins Medical Institutions and the PCBN for the construction
of TMA. TheP-Films used in MTI were kindly provided by Vladimir
Knezevic in 20/20GeneSystems, Inc. The authors also appreciate the
help of Nan Hu and ChaoyuWang in Dr. Philip Taylor's lab at NCI for
assisting in the conduction ofexperiments.
Grant SupportThis work was supported by two grants: EDRN/NCI
U01CA152813 grant (to
H. Zhang P.I.) with an administrative supplement to Dr. Robert
W. Veltri,Prostate Cancer Foundation and Patana Fund of The Brady
Urological Instituteat the Johns Hopkins University.
The costs of publication of this articlewere defrayed inpart by
the payment ofpage charges. This article must therefore be hereby
marked advertisement inaccordance with 18 U.S.C. Section 1734
solely to indicate this fact.
Received May 19, 2015; revised July 28, 2015; accepted August
28, 2015;published OnlineFirst September 24, 2015.
Biomarkers Predict a Prostate Cancer Aggressive Phenotype
www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 24(12)
December 2015 1871
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
References1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics,
2014. CA Cancer J Clin
2014;64:9–29.2. Loeb S, Schaeffer EM, Trock BJ, Epstein JI,
Humphreys EB, Walsh PC. What
are the outcomes of radical prostatectomy for high-risk prostate
cancer?Urology 2010;76:710–4.
3. Ward JF, ZinckeH, Bergstralh EJ, Slezak JM,Myers RP, BluteML.
The impactof surgical approach (nerve bundle preservation versus
wide local excision)on surgical margins and biochemical recurrence
following radical prosta-tectomy. J Urol 2004;172:1328–32.
4. Stephenson AJ, Scardino PT, Eastham JA, Bianco FJ Jr, Dotan
ZA, DiBlasioCJ, et al. Postoperative nomogram predicting the
10-year probability ofprostate cancer recurrence after radical
prostatectomy. J Clin Oncol2005;23:7005–12.
5. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD,
Walsh PC.Natural history of progression after PSA elevation
following radical pros-tatectomy. JAMA 1999;281:1591–7.
6. Shariat SF, Kattan MW, Vickers AJ, Karakiewicz PI, Scardino
PT. Criticalreview of prostate cancer predictive tools. Future
Oncol 2009;5:1555–84.
7. Swanson GP, Yu C, Kattan MW, Hermans MR. Validation of
postoperativenomograms in prostate cancer patients with long-term
follow-up. Urology2011;78:105–9.
8. Cooperberg MR, Hilton JF, Carroll PR. The CAPRA-S score: A
straightfor-ward tool for improved prediction of outcomes after
radical prostatectomy.Cancer 2011;117:5039–46.
9. DraismaG, Boer R,Otto SJ, van derCruijsen IW,Damhuis RA,
Schroder FH,et al. Lead times and overdetection due to
prostate-specific antigen screen-ing: estimates from the European
Randomized Study of Screening forProstate Cancer. J Natl Cancer
Inst 2003;95:868–78.
10. Etzioni R, Penson DF, Legler JM, di Tommaso D, Boer R, Gann
PH, et al.Overdiagnosis due to prostate-specific antigen screening:
lessons fromU.S.prostate cancer incidence trends. J Natl Cancer
Inst 2002;94:981–90.
11. Walsh PC. Overdiagnosis due to prostate-specific antigen
screening: les-sons from U.S. prostate cancer incidence trends. J
Urol 2003;170:313–4.
12. Tornblom M, Eriksson H, Franzen S, Gustafsson O, Lilja H,
Norming U,et al. Lead time associated with screening for prostate
cancer. Int J Cancer2004;108:122–9.
13. Schroder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S,
Nelen V, et al.Prostate-cancer mortality at 11 years of follow-up.
N Engl J Med 2012;366:981–90.
14. CooperbergMR, LubeckDP,MengMV,Mehta SS, Carroll PR. The
changingface of low-risk prostate cancer: trends in clinical
presentation and primarymanagement. J Clin Oncol
2004;22:2141–9.
15. Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic
and clinicalfindings topredict tumor extent of nonpalpable (stage
T1c) prostate cancer.JAMA 1994;271:368–74.
16. Bastian PJ, Mangold LA, Epstein JI, Partin AW.
Characteristics of insignif-icant clinical T1c prostate tumors. A
contemporary analysis. Cancer2004;101:2001–5.
17. Tosoian JJ, Trock BJ, Landis P, Feng Z, Epstein JI, Partin
AW, et al. Activesurveillance program for prostate cancer: an
update of the Johns Hopkinsexperience. J Clin Oncol
2011;29:2185–90.
18. 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.
19. Walsh PC. Radical prostatectomy versus watchful waiting in
early prostatecancer. J Urol 2005;174:1291–2.
20. Bill-Axelson A, Holmberg L, Ruutu M, Haggman M, Andersson
SO, BratellS, et al. Radical prostatectomy versus watchful waiting
in early prostatecancer. N Engl J Med 2005;352:1977–84.
21. Klotz L. Active surveillance for prostate cancer: a review.
Curr Urol Rep2010;11:165–71.
22. Veltri R, Isharwal S. Reply to indolent prostate cancer.
Urology 2011;77:e2.23. Mohler J, Bahnson RR, Boston B, Busby JE,
D'Amico A, Eastham JA, et al.
NCCN clinical practice guidelines in oncology: prostate cancer.
J NatlCompr Canc Netw 2010;8:162–200.
24. Chung JY, Braunschweig T, Baibakov G, Galperin M, Ramesh A,
Skacel M,et al. Transfer and multiplex immunoblotting of a paraffin
embeddedtissue. Proteomics 2006;6:767–74.
25. Chung JY, Hong SM, Choi BY, Cho H, Yu E, Hewitt SM. The
Expression ofPhospho-AKT, Phospho-mTOR, and PTEN in Extrahepatic
Cholangiocar-cinoma. Clin Cancer Res 2009;15:660–7.
26. Chung JY, Braunschweig T, Hu N, Roth M, Traicoff JL, Wang
QH, et al. Amultiplex tissue immunoblotting assay for proteomic
profiling: a pilotstudy of the normal to tumor transition of
esophageal squamous cellcarcinoma. Cancer Epidemiol Biomarkers Prev
2006;15:1403–8.
27. Vickers AJ, Cronin AM, Elkin EB, Gonen M. Extensions to
decision curveanalysis, a novel method for evaluating diagnostic
tests, predictionmodels and molecular markers. BMC Med Inform Decis
Mak 2008;8:53.
28. Vickers AJ, Elkin EB. Decision curve analysis: a novelmethod
for evaluatingprediction models. Med Decis Making
2006;26:565–74.
29. Steyerberg EW, Vickers AJ. Decision curve analysis: a
discussion. MedDecisMaking 2008;28:146–9.
30. CooperCS, Eeles R,WedgeDC,Van LooP,GundemG,Alexandrov LB, et
al.Analysis of the genetic phylogeny of multifocal prostate cancer
identifiesmultiple independent clonal expansions in neoplastic and
morphologi-cally normal prostate tissue. Nat Genet
2015;47:367–72.
31. Paulo P, Ribeiro FR, Santos J, Mesquita D, Almeida M,
Barros-Silva JD,et al. Molecular subtyping of primary prostate
cancer reveals specificand shared target genes of different ETS
rearrangements. Neoplasia2012;14:600–11.
32. Chen R, Zeng X, Zhang R, Huang J, Kuang X, Yang J, et al.
Ca1.3 channelalpha protein is overexpressed and modulates androgen
receptor transac-tivation in prostate cancers. Urol Oncol
2014;32:524–36.
33. Martikainen P, Isaacs J. Role of calcium in the programmed
death of ratprostatic glandular cells. Prostate 1990;17:175–87.
34. Connor J, Sawczuk IS, Benson MC, Tomashefsky P, O'Toole KM,
OlssonCA, et al. Calcium channel antagonists delay regression of
androgen-dependent tissues and suppress gene activity associated
with cell death.Prostate 1988;13:119–30.
35. Carles J, Lloreta J, Salido M, Font A, Suarez M, Baena V, et
al. Her-2/neuexpression in prostate cancer: a dynamic process? Clin
Cancer Res2004;10:4742–5.
36. Salomon DS, Brandt R, Ciardiello F, Normanno N. Epidermal
growthfactor-related peptides and their receptors in
humanmalignancies. Crit RevOncol Hematol 1995;19:183–232.
37. Neto AS, Tobias-Machado M, Wroclawski ML, Fonseca FL,
Teixeira GK,Amarante RD, et al. Her-2/neu expression in prostate
adenocarcinoma: asystematic review and meta-analysis. J Urol
2010;184:842–50.
38. Isharwal S, Miller MC, Epstein JI, Mangold LA, Humphreys E,
PartinAW, et al. Prognostic value of Her-2/neu and DNA index for
progres-sion, metastasis and prostate cancer-specific death in men
with long-term follow-up after radical prostatectomy. Int J Cancer
2008;123:2636–43.
39. Gang A. Differential expression of full-length and a
truncated Her-2/neuoncogene receptor in prostate cancer assessed
using relative quantitativeRT-PCR. Mol Urol 1998;2:305–10.
40. Chen J, Xi J, Tian Y, Bova GS, Zhang H. Identification,
prioritization, andevaluation of glycoproteins for aggressive
prostate cancer using quantita-tive glycoproteomics and
antibody-based assays on tissue specimens.Proteomics
2013;13:2268–77.
41. Tian Y, Bova GS, Zhang H. Quantitative glycoproteomic
analysis ofoptimal cutting temperature-embedded frozen tissues
identifying gly-coproteins associated with aggressive prostate
cancer. Anal Chem2011;83:7013–9.
42. Tian Y, Choi CH, Li QK, Rahmatpanah FB, Chen X, Kim SR, et
al. Over-expression of periostin in stroma positively associated
with aggressiveprostate cancer. PLoS ONE 2015;10:e0121502.
43. Nuzzo PV, Rubagotti A, Zinoli L, Ricci F, Salvi S, Boccardo
S, et al.Prognostic value of stromal and epithelial periostin
expression in humanprostate cancer: correlation with clinical
pathological features and the riskof biochemical relapse or death.
BMC Cancer 2012;12:625.
44. Suardi N, Briganti A, Gallina A, Salonia A, Karakiewicz PI,
Capitanio U,et al. Testing themost stringent criteria for selection
of candidates for activesurveillance in patients with low-risk
prostate cancer. BJU Int 2010;105:1548–52.
Cancer Epidemiol Biomarkers Prev; 24(12) December 2015 Cancer
Epidemiology, Biomarkers & Prevention1872
Zhu et al.
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/
-
2015;24:1864-1872. Published OnlineFirst September 24,
2015.Cancer Epidemiol Biomarkers Prev Guangjing Zhu, Zhi Liu,
Jonathan I. Epstein, et al. Biomarkers Predicts a Prostate Cancer
Aggressive PhenotypeA Novel Quantitative Multiplex Tissue
Immunoblotting for
Updated version
10.1158/1055-9965.EPI-15-0496doi:
Access the most recent version of this article at:
Material
Supplementary
http://cebp.aacrjournals.org/content/suppl/2015/09/24/1055-9965.EPI-15-0496.DC1
Access the most recent supplemental material at:
Cited articles
http://cebp.aacrjournals.org/content/24/12/1864.full#ref-list-1
This article cites 44 articles, 7 of which you can access for
free at:
E-mail alerts related to this article or journal.Sign up to
receive free email-alerts
Subscriptions
Reprints and
[email protected]
To order reprints of this article or to subscribe to the
journal, contact the AACR Publications Department
Permissions
Rightslink site. Click on "Request Permissions" which will take
you to the Copyright Clearance Center's (CCC)
.http://cebp.aacrjournals.org/content/24/12/1864To request
permission to re-use all or part of this article, use this link
on July 7, 2021. © 2015 American Association for Cancer
Research. cebp.aacrjournals.org Downloaded from
Published OnlineFirst September 24, 2015; DOI:
10.1158/1055-9965.EPI-15-0496
http://cebp.aacrjournals.org/lookup/doi/10.1158/1055-9965.EPI-15-0496http://cebp.aacrjournals.org/content/suppl/2015/09/24/1055-9965.EPI-15-0496.DC1http://cebp.aacrjournals.org/content/24/12/1864.full#ref-list-1http://cebp.aacrjournals.org/cgi/alertsmailto:[email protected]://cebp.aacrjournals.org/content/24/12/1864http://cebp.aacrjournals.org/
/ColorImageDict > /JPEG2000ColorACSImageDict >
/JPEG2000ColorImageDict > /AntiAliasGrayImages false
/CropGrayImages false /GrayImageMinResolution 200
/GrayImageMinResolutionPolicy /Warning /DownsampleGrayImages true
/GrayImageDownsampleType /Bicubic /GrayImageResolution 300
/GrayImageDepth -1 /GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true
/GrayImageFilter /DCTEncode /AutoFilterGrayImages true
/GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict >
/GrayImageDict > /JPEG2000GrayACSImageDict >
/JPEG2000GrayImageDict > /AntiAliasMonoImages false
/CropMonoImages false /MonoImageMinResolution 600
/MonoImageMinResolutionPolicy /Warning /DownsampleMonoImages true
/MonoImageDownsampleType /Bicubic /MonoImageResolution 900
/MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode
/MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None
] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false
/PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000
0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true
/PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ]
/PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier ()
/PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped
/False
/CreateJDFFile false /Description > /Namespace [ (Adobe)
(Common) (1.0) ] /OtherNamespaces [ > /FormElements false
/GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks
false /IncludeInteractive false /IncludeLayers false
/IncludeProfiles false /MarksOffset 18 /MarksWeight 0.250000
/MultimediaHandling /UseObjectSettings /Namespace [ (Adobe)
(CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /NA
/PageMarksFile /RomanDefault /PreserveEditing true
/UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling
/LeaveUntagged /UseDocumentBleed false >> > ]>>
setdistillerparams> setpagedevice