1. PSAD i s a biomarker to predict re-classification of AS patients on repeat biopsy (11, 13-14).
3. PSA derivatives (i.e. freePSA and (-2,-5,-7)ProPSA)) in serum and tissue (18-20).
4. DNA contentand nuclear morphometry of AS biopsy (21-24).
5. Ki67; Our group found that Ki--67 to be univariately significant to predict PSA recurrence in long term follow-up CaP (22).
6. p300 and Nuclear Morphometry in CaP:
Her-2/neu oncogene over-expression and DNA content: The FBBL at JHMI demonstrated that Her-
2Jneu as well as DNA content was increased significantly in biochemical progression, metastasis and CaP
specific survival (22, 24-25).
8. Calci um channel, voltage dependent, I tvpe, alpha 1d subunit (CACNA1D):
Periostin (POSTN
UNKNOWN
Prostate and Urologic Cancers Research Group
The Early Detection Research Network of the NCI is charged with the discovery, development and
validation of biomarkers for early detection and prognosis related to neoplastic disease. Our laboratory is an NCI EDRN (U01CA152813) working on "Glycoprotein biomarkers for the early detection of aggressive prostate cancer". This EDRN administratiVE! supplement is a collaboration with Robert Veltri on his project
to identify men with very low risk (indolent) prostate cancer (CaP) at the diagnostic biopsy at selection for
active surveillance (AS). We will assess biopsy tissue using quantitative nuclear histomorphometric measurements and molecular biomarkers to predict an unexpected catastrophic CaP in such men with indolent CaP. At Johns Hopkins Hospital w1e use the Epstein criteria that includes; PSA density (PSAD)
<0.15 ng/mVcm3, Gleason score SS, S2 cons involved with cancer, and ::;;SO% of any core involved with
cancer to select AS. Our approach will study 140 AS men (70 with a expected outcome and 70 with a disastrous outcome) using nuclear histomorphometry and pre-qualified biomarkers quantified by digital microscopy. Previously, our laboratory combined measurements of DNA content and (-2)pPSA in the serum and (-5,-?)pPSA in biopsy tissue to identify 7/10 men that would fail surveillance based on the primary diagnostic biopsy. We now will devHiop a clinical, morphological and biomarker 'signature' for identifying severe aggressive disease from a AS diagnostic biopsy. Our approach will combine nuclear morphometry measured by digital microscopy with a unique biopsy tissue biomarker profile (DNA content, Ki67, Her2neu, CACND1 and periostin). Fcr the molecular targets we will use a multiplex tissue blot (MTB) immunohistochemistry method. The Aims o'f our work include 1) to utilize retrospective archival biopsy
material from 70 AS cases where the outcome was unexpected and disastrous and collect an equal number of AS cases (n=140) and perform assays for morphology and biomarker targi ts proposed, 2) and predict
failure using Cox proportional hazards statistical modeling.
The Early Detection Research Network of the NCI is charged with the discovery, development and
validation of biomarkers for early detection and prognosis related to neoplastic disease. Our laboratory is an NCI EDRN (U01CA152813) working on "Glycoprotein biomarkers for the early detection of aggressive prostate cancer". This EDRN administratiVE! supplement is a collaboration with Robert Veltri on his project
to identify men with very low risk (indolent) prostate cancer (CaP) at the diagnostic biopsy at selection for
active surveillance (AS). We will assess biopsy tissue using quantitative nuclear histomorphometric measurements and molecular biomarkers to predict an unexpected catastrophic CaP in such men with indolent CaP. At Johns Hopkins Hospital w1e use the Epstein criteria that includes; PSA density (PSAD)
<0.15 ng/mVcm3, Gleason score SS, S2 cons involved with cancer, and ::;;SO% of any core involved with
cancer to select AS. Our approach will study 140 AS men (70 with a expected outcome and 70 with a disastrous outcome) using nuclear histomorphometry and pre-qualified biomarkers quantified by digital microscopy. Previously, our laboratory combined measurements of DNA content and (-2)pPSA in the serum and (-5,-?)pPSA in biopsy tissue to identify 7/10 men that would fail surveillance based on the primary diagnostic biopsy. We now will devHiop a clinical, morphological and biomarker 'signature' for identifying severe aggressive disease from a AS diagnostic biopsy. Our approach will combine nuclear morphometry measured by digital microscopy with a unique biopsy tissue biomarker profile (DNA content, Ki67, Her2neu, CACND1 and periostin). Fcr the molecular targets we will use a multiplex tissue blot (MTB) immunohistochemistry method. The Aims o'f our work include 1) to utilize retrospective archival biopsy
material from 70 AS cases where the outcome was unexpected and disastrous and collect an equal number of AS cases (n=140) and perform assays for morphology and biomarker targi ts proposed, 2) and predict
failure using Cox proportional hazards statistical modeling.
Aim #1 To select 140 retrospective (n=70
eventually had a catastrophic
outcome and another n=70 cases where the outcome is as expected, a very low risk cancer). Dr. Epstein,pathologist, will be responsible for these
tasks.
Cases available for the project are listed
in Table 1— These must be collected,reviewed and marked for cancer areas.
Aim #2 Quantitative Nuclear Morphometry
(QNM) and Molecular Biomarkers
by MTI on AS cohort
Optimization of five
biomarkers using the
Multiplex tissue
immunoblotting
(MTI)1quarter
QNM technology is
standardized and
requires 15-20 minutes
per case and we have
140 cases to run.
QNM – this will require at least 5
qurtersa to collect cells from 140 cases,
create the database and then analyze.
MTI for (-5,-7)ProPSA, Ki67,
Her2/neu, POSTN, &CACNA1D; - We
can only run about 5-6 cases per every 2
days and make one run per week for all
the markers.
A total of five quarters required..
Aim #3 Construct and validate computerized
based histologic classifier using Cox
proportional hazards analysis.
Dr. Bruce Trock will
assist in preparing the
Cox proportional
hazard models.
A proportional hazards model will be
developed using the predictors from the
best model in Aim 2. For continuous
variables with evidence of non-linearity
we will explore alternative metrics using
restricted cubic splines. We may use
several approaches to modeling. If the
number of predictors derived in Aim 1 is
not large (<1/10th the number of biopsy
progression events) we will include all
predictors and bootstrap the model.
Estimates after all data is collected and
audited: 8-10 weeks (part-time basis)
There are currently no biomarkers annotated for this protocol.
No datasets are currently associated with this protocol.
