209elevated cell proliferation. Expression profi ling of these genes can provide useful
information about cancer and planning its personalized treatment.
Microarray methods have revealed unexpected subgroups within the diagnostic
categories of the hematologic cancers that are based on morphology and have dem-
onstrated that the response to therapy is dictated by multiple independent biologic
features of a tumor. Some examples of applications of this approach are:
- These expression signatures can be combined to form a multivariate predictor of
survival after chemotherapy for diffuse large-B-cell lymphoma. - Gene-expression profi ling has been used as an alternative approach to mapping
chromosomal translocations in leukemias. Gene-expression signatures can be
combined with the use of statistical algorithms to predict chromosomal abnor-
malities with a high degree of accuracy. - In B-cell acute lymphoblastic leukemia, gene-expression profi ling at the time of
diagnosis provides information that could predict which patients would relapse
and which would remain in continuous complete remission. - ZAP-70 gene expression identifi es a chronic lymphocytic leukemia (CLL) sub-
type with unmutated immunoglobulin genes, inferior clinical outcome, and dis-
tinct gene expression profi le. RT-PCR and immunohistochemical assays for
ZAP-70 expression can be applied clinically and would yield important prognos-
tic information for CLL patients.
An important goal is to develop a platform for routine clinical diagnosis that can
quantitatively measure the expression of a few hundred genes. Such a diagnostic
platform would enable a quick determination of important molecular subgroups
within each hematologic cancer. As new clinical trials designed, one must include
genomic-scale gene-expression profi ling in order to identify the genes that infl uence
the response to the agents under investigation. Thus the molecular diagnosis of the
hematologic cancers can be refi ned on the basis of new advances in treatment and
facilitate the development of tailored therapies for molecularly defi ned diseases.
Gene expression profi ling has been done of prostate tumors using IHC on tissue
microarrays. Positive staining for MUC1, a gene highly expressed in the subgroups
with aggressive clinicopathological features, is associated with an elevated risk of
recurrence, whereas strong staining for AZGP1, a gene highly expressed in the
other subgroup, is associated with a decreased risk of recurrence. In multivariate
analysis, MUC1 and AZGP1 staining are strong predictors of tumor recurrence
independent of tumor grade, stage, and preoperative prostate-specifi c antigen (PSA)
levels. These fi ndings suggest that prostate tumors can be usefully classifi ed accord-
ing to their gene expression patterns, and these tumor subtypes may provide a basis
for improved stratifi cation for prognosis and treatment.
Gene-expression profi ling has been used to improve the design of cancer drugs
that have shown some promise in clinical trials. Some of the cancer signatures can
predict clinical response in individuals treated with anticancer drugs. Notably, sig-
natures developed to predict response to individual agents, when combined, could
also predict response to multidrug regimens. Finally, integration of chemotherapy
response signatures with signatures of oncogenic pathway deregulation may help to