Hematology Reports 2011; volume 3(s3):e2

The use of molecular profiling though it is not clear at this time whether all
cancer cells have this capacity or only a mi- Correspondence: Dr Miguel Angel Piris, Molecular
for diagnosis and research nority of them, or even whether this is just Pathology Programme, Spanish National Cancer
in non-Hodgkin’s lymphoma transiently acquired. These cells develop into Research Centre (CNIO), Av. Melchor Fernandez
a tumor when they get the capacity for local Almagro, 3, 28029 Madrid, Spain.
Miguel Angel Piris infiltration, probably through development of Tel. +34.91.732.8000 (ext 3600) - Fax: +34.91.224 6923.
a signature identified as epithelial-mes- E-mail: mapiris@cnio.es
Centro Nacional de Investigaciones
enchymal transition. Eventually, these cells
Oncologica, Madrid, Spain Key words: non-Hodgkin’s lymphoma, molecular
gain the capacity for distant metastasis, re-
diagnosis, genetic abnormality, treatment.
quiring drug treatment. Finally, during
therapy, these tumors can develop genomic Conflict of interest: the author reports no con-
instability, which allows them to escape the flicts of interest.
Abstract therapy. The steps for this sequence are still a
matter of investigation, and new concepts and This work is licensed under a Creative Commons
Molecular profiling facilitates the under- Attribution NonCommercial 3.0 License (CC BY-
data may dramatically change our current un-
standing of the genetic processes underlying NC 3.0).
derstanding of cancer.
the development of cancer, and makes it pos-
©Copyright M.A. Piris, 2011
sible to use specific signatures to prognosti- Licensee PAGEPress, Italy
cate clinical outcome and to predict response Genetic sequencing of cancer cells Hematology Reports 2011; 3(s3):e2
to specific treatments. There has been a great A key advance in our study of cancer has doi:10.4081/hr.2011.s3.e2
increase in the availability of tools for ex- come from the development of technology and
ploring genetic abnormalities in cancer cells, reduction of the cost of gene sequencing,
which have allowed a more comprehensive which has produced a great increase in the
characterization of the mutations, transloca- amount of sequencing data that are starting to challenge for researchers, therefore, is now to
tions, and copy-number variations that may af- be available.1 This has led to the formation of understand the data that are available on ge-
fect the development of cancer or therapy re- the International Cancer Genome Consor- netic changes, and move on from that to dis-
sponse. An improved understanding of the mo- tium. The goal of this consortium is to coordi- cover the implications for targeted therapy of
lecular basis of cancer is helping also in the nate the generation of comprehensive cata- the patients.
identification of new molecular targets for logs of genomic abnormalities (somatic muta-
therapy. tions) in tumors from 50 different cancer
types and/or subtypes that are of clinical, sci- Therapy driven by molecular
entific, or social importance across the globe.2 diagnosis
For each of these 50 cancer types/subtypes, The concept we are now trying to sustain
Introduction the objective of the consortium is to sequence and develop is phrased as therapy driven by
at least 500 cases. This project is ongoing, for molecular diagnosis. Research strategies aim
A key aim in the management of cancer is example in breast cancer, lymphoma, and to match the individual genetic variability of
to associate the diagnosis to the treatment, so other common types of cancer – some of the tumor samples and patients with therapies ad-
tumor types have still to be defined. It is the justed to these variables. This will be achieved
that treatments are more effective and the
most ambitious project that has ever been de- through the combination of: early diagnosis for
probability of curing patients is improved. For
veloped in the field of biomedicine or cancer screening of patients at risk; the definition of
this, the concept of molecular cancer tax-
research, as it encompasses the sequencing of prognostic markers that would allow us to as-
onomy is essential – a system that allows cli-
50,000 genomes. sign individual treatments and stratify pa-
nicians to assign treatments to patients,
tients by their individual risk; and the develop-
based on the morphology/immunopheno- Multiple mutations in cancer cells ment of therapies for the newly identified ther-
type/molecular features of the disease. Tax- An important finding discovered in recent apeutic targets.
onomy requires the precise definition of clini- years is that most cancer samples have mul-
copathologic entities and molecular markers tiple somatic mutations, rather than the small Tools for molecular diagnosis
that are specific to each condition. This classi- number that was originally expected. In that Analysis of the genomes of cancer patients
fication system leads to the identification of context, it is not surprising that targeted ther- now allows the characterization of several
underlying molecular alterations, targets for apies that are directed at single mutations are types of genetic changes. These include point
therapy and predictive and prognostic markers not fully effective. A study of tumor cells in mutations, insertion or deletion of sequences,
for patient stratification. pancreatic cancer found that these tumors had inter- and intrachromosomal translocations,
an average of 63 genetic alterations, the ma- and copy-number changes.4 These somatic
jority of which were point mutations, and some mutations can be examined using DNA mi-
The cancer sequence cancers had more than 100 genetic alterations croarrays, to investigate single-nucleotide
Cancer molecular profiling has greatly fa- (Figure 1).3 The frequency of mutational polymorphisms (SNPs), copy-number varia-
cilitated a more comprehensive under- changes is nevertheless changing among dif- tions, and gene-expression signatures.
standing of the processes underlying the de- ferent tumor types, with higher frequency in Changes in gene expression and protein prod-
velopment of cancer cells. An early stage is tumors associated with smoking or exposure ucts can also be measured using tissue im-
the presence of inherited plus acquired ge- to ultraviolet light, such as lung cancer and munohistochemical microarrays. In addition,
netic mutations and epigenetic changes in melanoma. The mutations characterized in microRNA (miRNA) arrays are starting to be
the cells, which at this stage are not yet neo- pancreatic cancer are not distributed randomly used, and techniques for analyzing miRNAs
plastic cells, but acquire a selective survival across the genome, but affect particular sig- have been applied using formalin-fixed,
advantage. Cancer cells then acquire the ca- naling pathways and processes in the cells, paraffin-embedded tissue studied using quan-
pacity for self-renewal, so-called stemness, al- forming a complex system of changes. The titative polymerase chain reaction (QT-PCR)

[page 4] [Hematology Reports 2011; 3(s3):e2]

on samples. Identification of SNPs allows the Identifying treatment targets leukemia revealed variability in the expression
analysis of gene polymorphisms in patients Lamb et al. have reported their development of two gene clusters associated with B-cell-re-
and copy-number variation in tumors. How- of a reference database of gene-expression ceptor signaling and mitogen-activated protein
ever, interpretation of these results is not profiles from cell lines. These can be mined for kinase activation. Variations in the expression
simple – millions of SNPs can be identified in connections with gene signatures of specific tu- of these two clusters identified three groups of
an individual, but most of these cannot be mors or experimental conditions, to form a patients who had different risks of treatment-
linked to specific risk for a disease, or to re- connectivity map that relates tumor signatures free survival.9 Using such techniques, there-
sponse to therapy. In most cases, the impact of with response to specific treatments.8 Multiple fore, gene expression may be used to predict
an SNP is not individually important, although practical applications of this approach have al- patients’ treatment needs at early stages of the
in some cases it may produce a different gene ready been generated. For instance, an expres- disease. These potential drug sensitivities can
product or alter the regulation of a gene with sion profile of genes thought to play a role in be tested in vitro using measurable pharmaco-
significant pathogenic results. the pathogenesis of chronic lymphocytic dynamic markers. In this way, the introduction
The use of tissue microarrays has led to the
development of antibodies that can be used to
identify specific cancer phenotypes. For ex-
ample, germinal center B-cell-expressed tran-
script 1 (GCET1) has been used to charac-
terize B-cell lymphomas,5 and the expression
of B-lymphocyte-induced maturation protein-1
(BLIMP-1) has been found to constitute the
best marker of plasma cell differentiation
among normal and neoplastic B-cells.6 Im-
munohistochemical staining in tissue mi-
croarrays can also be used to quantify the ex-
pression of key regulatory proteins (e.g. Bcl-2,
Lyn, Syk). These may be highly important in
the future because, by using such microarrays,
expression of multiple markers can be visual-
ized at the same time in a tumor sample. More-
over, as tumor cells are unstable and their ge-
netic composition changes with time, this
technique should allow us to monitor differ-
ences both between tumors and within the Figure 1. The number of genetic alterations detected through sequencing and copy-number
same tumor over time. analyses in each of 24 pancreatic cancers.3 From Jones S et al. Core signaling pathways in
human pancreatic cancers revealed by global genomic analyses. Science 2008;321:1801-6.
Identifying molecular risk Reprinted with permission from AAAS.
Tumor specimens can be analyzed to iden-
tify specific molecular risk scores. For ex-
ample, in samples from patients with advanced
Hodgkin’s lymphoma, quantitative reverse-
transcription PCR was applied to 30 genes to
develop a molecular risk score.7 There are
likely to be hundreds of markers that affect pa-
tients’ molecular risk in this condition, but it is
not feasible to work with large numbers of
markers, so they were reduced to a reasonable
number, to focus on the best predictors. These
were integrated into an 11-gene model, incor-
porating genes from four functional pathways:
the cell cycle, apoptosis, macrophage activa-
tion, and interferon regulatory factor 4. This
produced a single molecular risk score for indi-
vidual patients, which predicted response to
treatment and 5-year failure-free survival (de-
fined as the time interval between treatment
initiation and treatment failure or last follow-
up). Importantly, in this and other studies it
has been found that the molecular risk was in-
dependent of clinical risk. Thus, molecular and
clinical risk could be combined, and in this way
it was possible to identify about 25% of pa- Figure 2. Integration of the molecular risk score and clinical risk (stage IV disease) identified
tients who were at very high risk of not re- a quartile of patients with particularly poor prognosis.7 Refer to main text for definition of
failure-free survival. This research was originally published in Blood. Sanchez-Espiridion B
sponding to therapy, carrying a very unfavor- et al. A molecular risk score based on 4 functional pathways for advanced classical Hodgkin
able prognosis (Figure 2).7 lymphoma. Blood 2010;116:e12-17.© The American Society of Hematology.

[Hematology Reports 2011; 3(s3):e2] [page 5]

of new therapies for hard-to-treat cancers may marker for a subset of germinal center-de-
be accelerated. Conclusions rived lymphomas. Blood 2008;111:351-8.
A second tool for identifying molecular 6. Garcia JF, Roncador G, Sanz AI, et al.
treatment targets is gene set enrichment It may be possible to stratify patients for tar- PRDM1/BLIMP-1 expression in multiple B
analysis (GSEA) – a computational method geted therapy using molecular diagnosis. This and T-cell lymphoma. Haematologica
that determines whether an a priori-defined approach requires clear, specific markers that 2006;91:467-74.
set of genes shows statistically significant can be applied at diagnosis, and pharmacody- 7. Sanchez-Espiridion B, Montalban C, Lopez
differences between two phenotypes namic markers to assess responses to treat- A, et al. A molecular risk score based on 4
(www.broad.mit.edu/gsea). This approach ment. New drugs and combinations may also functional pathways for advanced classical
has been used to screen a database of drug- be assayed in vitro using these molecular tech- Hodgkin lymphoma. Blood 2010;116:e12-17.
associated gene-expression profiles for mole- niques. Rapid and flexible clinical trials are 8. Lamb J, Crawford ED, Peck D, et al. The Con-
cules whose profile overlapped with the needed to move ahead with the introduction of nectivity Map: using gene-expression signa-
gene-expression signature of glucocorticoid new drugs developed in this way.
tures to connect small molecules, genes, and
sensitivity in acute lymphoblastic leukemia.
disease. Science 2006;313:1929-35.
This indicated that the gene-expression pro-
9. Rodriguez A, Villuendas R, Yanez L, et al.
file of the mTOR inhibitor rapamycin
Molecular heterogeneity in chronic lym-
matched the signature of glucocorticoid sen- References
phocytic leukemia is dependent on BCR
sitivity. Further testing demonstrated that ra-
signaling: clinical correlation. Leukemia
pamycin could induce glucocorticoid sensi- 1. Human genome at ten: the sequence ex-
tivity in neoplastic lymphoid cells.10 This 2007;21:1984-91.
plosion. Nature 2010;464:670-1.
demonstrates how GSEA can be used to iden- 10. Wei G, Twomey D, Lamb J, et al. Gene ex-
2. Hudson TJ, Anderson W, Artez A, et al. In-
tify promising new or combination therapies. ternational network of cancer genome pression-based chemical genomics identi-
Likewise, in cutaneous T-cell lymphoma, this projects. Nature 2010;464:993-8. fies rapamycin as a modulator of MCL1
approach demonstrated that the histone 3. Jones S, Zhang X, Parsons DW, et al. Core and glucocorticoid resistance. Cancer Cell
deacetylase inhibitor vorinostat interfered signaling pathways in human pancreatic 2006;10:331-42.
with the signal-transduction pathway of the cancers revealed by global genomic 11. Wozniak MB, Villuendas R, Bischoff JR, et
T-cell receptor, and synergized with phospho- analyses. Science 2008;321:1801-6. al. Vorinostat interferes with the signaling
inositide-3 kinase inhibitors.11 Thus, the de- 4. Stratton MR, Campbell PJ, Futreal PA. The transduction pathway of T-cell receptor
velopment of new drugs and combination cancer genome. Nature 2009;458:719-24. and synergizes with phosphoinositide-3
therapies is being accelerated through the 5. Montes-Moreno S, Roncador G, Maestre L, kinase inhibitors in cutaneous T-cell lym-
use of molecular diagnostic techniques. et al. Gcet1 (centerin), a highly restricted phoma. Haematologica 2010;95:613-21.

[page 6] [Hematology Reports 2011; 3(s3):e2]