||5R03CA073922-02 Interpret this number
||Massachusetts General Hospital
||Optimal Screening for Prostate CA with Serial Psa Levels
DESCRIPTION (Applicant's Description) The goal of this pilot study is to
develop an approach to early detection of carcinoma of the prostate (CP)
that is optimized, i.e. that, when compared to the standard approach,
results in the greatest increase in years of life saved for a given
increment in testing. The principal investigator is currently developing
precise statistical models of long term longitudinal PSA behavior using data
from participants in the Veterans Affairs (VA) Normative Aging Study (NAS)
as part of a currently funded study. After the distributional assumptions
of these models have been tested, we will use them to develop an optimized
early detection approach that reassesses the risk that the patient has
prostate cancer after each successive PSA measurement based on all previous
measurements. The decision for the next step, either biopsy or when to have
the next PSA test, is based on the current assessment of risk. Those at
elevated risk will be measured more frequently for PSA levels, and those at
highest risk are referred for more invasive diagnostic workup such as
biopsy. Thus, more screening resources are allocated to patients who are at
increased risk. For the intermediate risk group, time until next PSA
measurement will be optimized to maximize the expected years of life saved,
using published estimates of treatment efficacy by stage and age. The
approach will be developed in three stages: (I) A statistical method for
estimating the risk of CP based on single or multiple longitudinal PSA
levels, and the duration between the PSA levels will be developed, (II) A
computer simulation of an early detection program for CP incorporating the
method for calculating risk of CP will be written, and (III) The simulation
will be used to identify the optimal scheme for assigning patients into
normal, intermediate, and high risk categories. This program will be
similar to an approach the PI has developed for early detection of ovarian
cancer with longitudinal levels of the marker CA125. The long-term
objectives, beyond the scope of this proposal, are to expand these
approaches to early detection of CP to include other tests such as "free
PSA" and digital rectal exam, to implement these approaches in software
appropriate for use in clinical settings, and to compare such optimized
screening strategies with current screening strategies in a randomized
Ccm1 is required for arterial morphogenesis: implications for the etiology of human cavernous malformations.
, Plummer N.W.
, Adams J.A.
, Marchuk D.A.
, Li D.Y.
Development (Cambridge, England), 2004 Mar; 131(6), p. 1437-48.