Biomedical Engineering Reference
In-Depth Information
developed to facilitate communication among research and clinical data handlers to
resolve QA problems in clinical data.
10.3.3 Data Centralization
WRI opted to develop a hybrid data warehouse for centralization of the internally
generated and the public data needed in data analysis [37]. The details of this DW
are discussed in Chapter 8. Currently, WRI is furthering the development of the
clinical component of the data warehouse with a modular data model that enables
multiple clinical projects to be readily supported.
10.3.4 Genomic and Proteomic Studies
10.3.4.1 Allelic Imbalance to Characterize Genomic Instability
Allelic imbalance (AI) studies examine patterns of changes to the physical structure
of the chromosomes. This technique has been used at WRI to characterize genomic
alterations in breast disease and cancer.
AI in Breast Disease
Progression of breast cancer has long been viewed as a nonobligatory sequence of
changes in tissue histology from normal epithelium to a malignant tumor. During
the process of breast cancer development, genetic changes are thought to accumu-
late in a random fashion, but little is known about the timing of when these critical
mutations occur and how such genetic changes can influence progression of the
disease.
To investigate the timing of critical genetic changes in human breast disease,
patterns of AI have been examined in tissue samples that represent various stages
along the continuum of breast cancer development to study the evolution of
genomic instability. The samples were first subjected to laser microdissection to sep-
arate the diseased cells from all of the surrounding normal tissues. The DNA from
these cells was then isolated and 52 genetic markers known as microsatellites,
which represent 26 chromosomal regions throughout the genome, were assayed to
assess patterns of AI.
This study showed that preinvasive ductal carcinoma in situ (DCIS) lesions con-
tain levels of genomic instability that are very similar to that of advanced invasive
tumors, suggesting that the biology of a developing cancer may already be predeter-
mined by the in situ stage. Conversely, levels of AI in atypical ductal hyperplasia
(ADH) lesions are similar to those in disease-free tissues, which may be why patients
with ADH who receive prompt treatment usually have a good prognosis [38].
AI in Breast Cancer
An important problem in treating patients with breast cancer is trying to determine
which patients will respond to treatment and which patients will not respond well.
One way is to stratify breast cancer patients into two groups: those who are
expected to have a favorable outcome and those who likely will have a less favor-
able outcome (such as disease recurrence or premature death). The structural char-
acteristics of the cancer cells that can be seen by a pathologist can be used for patient
Search WWH ::
Custom Search