Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
Biomarkers in Rare Diseases
A. Ferlini a C. Scotton a G. Novelli b, c
a
Section of Microbiology and Medical Genetics, Department of Medical Sciences, University of Ferrara, Ferrara,
and b Genetics Section, Department of Biomedicine and Prevention, University of Rome ‘Tor Vergata’, Rome, Italy;
c
Pharmacogenomics Working Party, European Medicines Agency (PGWP EMA), London, UK
Key Words
Biomarkers · Diagnostics · Genetics · Genomics · Mendelian ·
Omics · Proteomics · Rare diseases · Therapies ·
Transcriptomics
Abstract
Background: Nowadays 7,000 rare diseases (RDs) have been
identified with a prevalence less than 5/10,000. Despite of
the enormous effort the European Union (EU) has already
invested in this field, still 4,000 RDs remain orphan of genet-
ic diagnosis and causative gene identification. The genetic
definition of RDs represents a prerequisite for being diag-
nosed, for having a robust prevention, for entering in a spe-
cific standard of care, and ultimately, for being included in
clinical trials, often via personalized medicine. It is well estab-
lished that biomarkers can offer a way to speed up research
by understanding the pathophysiological mechanisms of
diseases. In particular, biomarkers will offer an invaluable
tool for monitoring disease progression, prognosis and re-
sponse to drug treatment. Methods: In this review, we sum-
marize the different types of biomarkers and their impor-
tance as well as their translational applications in RDs. We
have reviewed the current knowledge on biomarkers state-
of-the-art via literature data, specific websites and EU sourc-
es regarding past, pending and current projects. Results:
Here we provide a comprehensive scenario of biomarkers
research, its applications in clinical practice, with special em-
phasis on translational research applicable to diagnostic and
clinical trials. The experience of the EU project BIO-NMD is
© 2013 S. Karger AG, Basel
1662–4246/13/0166–0313$38.00/0
E-Mail karger@karger.com
www.karger.com/phg
Published online: February 3, 2014
also mentioned. Conclusion: Biomarkers represent key fea-
tures in both diagnostics and research on rare diseases and
will encounter wide exploitation in translational and person-
alized medicine. © 2013 S. Karger AG, Basel
Rare Diseases and Biomarkers: State-of-the-Art
In the European Union (EU) countries, a rare disease
(RD) is any disease affecting fewer than 5 people in 10,000,
according to the definitions adopted by the European
Parliament and the Council on Orphan Medicinal Prod-
ucts in the Orphan Drugs Regulation (European Com-
mission) No. 141/2000 and in the Commission Commu-
nication COM (2008) 679/2 on Rare diseases: Europe’s
challenges. It translates into approximately 425,000 peo-
ple throughout the EU’s 27 Member States [1]. Most pa-
tients suffer from even rarer diseases affecting 1 person in
100,000 or more. It is estimated that today in the EU,
6,000–8,000 distinct rare diseases (RDs) affect 6–8% of
the population – between 27 and 36 million people (for
details see www.eucerd.eu and Orphanet database). More
than 80% are caused by genetic defects. Screening for ear-
ly diagnosis, followed by suitable care, can improve the
quality of life and life expectancy. Due to the limited in-
terest of pharmaceutical industry in developing and mar-
keting products, the EU and national governments have
prioritized law and funding to promote the development
of ‘orphan drugs’ for patients with RDs. Of the 7,000 RDs
157.82.153.40 - 5/17/2015 3:11:15 PM
Alessandra Ferlini, MD, PhD
Section of Microbiology and Medical Genetics, Department of Medical Sciences
University of Tokyo
University of Ferrara
Downloaded by:
Via Fossato di Mortara 74, IT–44121 Ferrara (Italy)
E-Mail fla @ unife.it
collected in the OMIM database (including all genet-
ic diseases; http://www.ncbi.nlm.nih.gov/omim), only
3,000 genes are known, despite that the recent history of
medical genetics has clearly shown that a prerequisite to
approaching diagnosis and cure for a RD is to identify the
causative mutated gene. Therefore, the identification of
virtually all RD causing genes is imperative.
Although the relevance of the single RD is significant-
ly lower than 1/2,000, the amount of RDs is collectively
7,000, suggesting that one out of 20 patients will be af-
fected by an orphan disease [2]. This implies that RDs do
have a heavy socioeconomic impact. Joint effort will max-
imize the success and reduce costs, and it is, therefore,
mandatory to cooperate within a national and suprana-
tional context, as recently stated by the novel cooperation
initiative by the EU, Canada and the US: The Interna-
tional Rare Diseases Research Consortium (IRDiRC,
http://ec.europa.eu/research/health/medical-research/
rare-diseases/irdirc_en.html).
Indeed, many RDs remain without genetic diagnosis
and/or therapy, and IRDiRC announced the goals to dis-
cover 200 new therapies and to diagnose most of them by
2020.
RDs research is considered pivotal for other common
diseases. Indeed, many RDs show some pathophysiologi-
cal mechanisms that can be applied and translated to oth-
er disorders [2]. On the contrary, within the same disease,
the signs and the symptoms may be very different: for
example, whilst the RD spinal muscular atrophy (SMA,
OMIM 253300) affects the anterior horn cells, the Char-
cot Marie Tooth (CMT, OMIM 118200) peripheral neu-
ropathy affects the peripheral nerve. The neuromuscular
junction is involved in congenital myasthenic syndromes
and, within this disease category, different parts of that
junction are affected, meaning that a drug may benefit
one form of congenital myastenic syndromes, but not all.
Meanwhile, problems with the muscle cell itself cause
muscular dystrophies such as Duchenne muscular dys-
trophy (DMD, OMIM 310200). In addition to these con-
ditions, which have a genetic basis and are caused by de
novo or inherited mutations, other diseases affecting the
same system can be acquired. For example, renal cell car-
cinoma is a disease of the cell metabolism with deregula-
tion of metabolic pathways involved in oxygen, iron and
nutrient sensing, but it also occurs as a specific manifesta-
tion in rare syndromes such as von Hippel-Lindau (VHL,
OMIM 193300), Birt-Hogg-Dubé (BHD, OMIM 135150)
syndromes and hereditary leiomyomatosis and renal cell
carcinoma (HLRCC, OMIM 150800) [3]. In order to un-
derstand these contradictions, the worldwide research
314 Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
has focused on identification of characteristic determi-
nants or signs for a group of diseases or therapeutic re-
sponse: the so-called biomarkers.
In 2001, the National Institutes of Health defined bio-
markers as ‘a characteristic that is objectively measured
and evaluated as an indicator of normal biological pro-
cesses, pathogenic processes, or pharmacologic responses
to a therapeutic intervention or other healthcare inter-
vention’ [4]. In other words, biomarkers include tools
and technologies that are able to discern, predict and
monitor diseases and/or treatment response. Measuring
can be performed using body fluids or tissue (cells, skin,
etc.) evaluation or alternatively may be done indirectly by
using imaging technologies.
Biomarkers can be identified at different levels (DNA,
RNA, protein); the advent of high-throughput technolo-
gies such as genomics, transcriptomics, proteomics, me-
tabolomics, and modern imaging have allowed the dis-
covery of more markers than before. Recent studies show
the benefit of high-throughput platform interaction for
biomarkers discovery in spinal muscular atrophy [5].
In this review, we present an overview of biomarker
characteristics, types, applications in RDs, and future per-
spectives.
Ideal Biomarker
A biomarker is ‘a measurable biological characteristic
that is an indicator of normal biologic processes, patho-
genic processes, and/or response to therapeutic or other
interventions’ [6]. It may be, for example, a particular
gene mutation, a transcription factor, a protein found in
the blood or urine, or the result of a scan. It can always
be tied to a feature of the disease – for example, levels of
a protein biomarker in the blood may increase as a dis-
ease progresses, even before the patient or the clinician
have noted a worsening in symptoms. Where clinical
measurements may not be reliably correlated to an aspect
of a disease or may not be sensitive enough to detect small
changes, especially over a short period of time, biomark-
ers can offer an objective measurement of something
which is directly related to an aspect of the condition.
They may allow the detection of small changes, make
predictions about the benefits of a therapy or suggest a
disease’s likely course. An appropriate biomarker should,
therefore, document disease progression or disease activ-
ity and should assess any effects of therapeutic interven-
tion.
Following the rules of regulatory authorities such as
the European Medicines Agency (EMA), an ideal bio-
marker should have the following characteristics:
157.82.153.40 - 5/17/2015 3:11:15 PM
Ferlini /Scotton /Novelli
University of Tokyo
Downloaded by:
• Analytical validity: It should be measuring a specific
parameter, both accurately and within a detection
range clearly distinguishing between diseased and
normal status. It can be assessed using specific tests
that have to be accurate, reliable, and reproducible.
The limits of the detection systems should be clearly
defined. Analytical performance characteristics also
depend on the methodology used for biomarker detec-
tion, sample preparation as well as the expertise of the
testing laboratory.
• Clinical validity: It should finely measure changes in
disease characteristics with a bidirectional correlation
(the measured change in the biomarker should always
indicate a change in the correlated disease feature and
a change in the disease features should always be mir-
rored in the biomarker). The biomarkers should be
stable, which means not influenced by other environ-
mental factors such as diet, exercise, stress, age, sex, or
genetic determinants.
• Clinical utility: It is the likelihood that the biomarker
will lead to improved outcome with a given interven-
tion. For example, it is the value that the biomarker
provides in predicting drug response or prognostic
evaluation of a certain disease of a marker defined
population over and above other standard clinical fea-
tures.
• Feasibility: This is a technical aspect related to the bio-
marker detection. A biomarker should be easy to iden-
tify and measure, and not variable, depending on the
type of sample collection, methods or platform used
for its identification.
• Time- and cost-effectiveness: The cost/effect relation-
ship is important especially if a biomarker can be wide-
ly used in large populations and for monitoring fre-
quent diseases (as diabetes or cancer, for example).
• Non-invasiveness: Ideally, a biomarker should be ear-
ly, indicative after active damage, sensitive, correlate
with severity, accessible in peripheral tissue, analyti-
cally stable in tissue, bridge across species, and finally
localize damage. The low invasiveness also implies
easy optainability biomarkers, as in body fluids or by
scanning techniques (e.g. MRI or PET, etc.).
The use of biomarkers fitting this profile in clinical tri-
als would mean that they can be conducted over a shorter
time and with a smaller, more specifically selected, pa-
tient cohort.
Personalized or Stratified Medicine
Personalized medicine is an important concept of
health in which we can administer ‘the right drug to the
Biomarkers and Genetics
right patient at the right time’ [7]. It is intuitive that hav-
ing such a deal realized, there will be an enormous im-
provement in health and life quality together with very
relevant economic repercussions. This is also true for RDs
for which both the small number of patients and the short
running period of novel trials make personalized medi-
cine crucial for the therapy success. In this view, the most
important aims of the use of biomarkers for drug devel-
opment identified by the EMA are the improved design
of trials and enhanced cohort selection [8]. The pharma-
cogenetic biomarkers can predict the patient’s response
to a particular therapy. In the future, having a pharmaco-
genetic biomarker predicting drug response, may theo-
retically allow selecting an appropriate cohort before
treatment starts. This is also the case with side effects that
may be very severe. Patients who may be at high risk of
presenting severe adverse effects due to the drug reaction
may be identified by a biomarker profile and excluded
from a specific drug, which can be negatively health effec-
tive [8].
Similarly, they will be able to provide stratified patient
cohorts for clinical trials, for example, by ‘identifying pa-
tients with a particular disease sub-type or disease sever-
ity as a target’ [8] allowing smaller numbers of partici-
pants, more meaningful results and reduced costs.
In the optimized scenario, a biomarker can inform a
patient on treatment choices – predicting which drug can
work best for which patient, on which dose regimen and
an appropriate and efficacious administration route. Bio-
markers are, therefore, very important (for some EMA/
Food and Drug Administration approved markers) and
mandatory in personalized medicine, minimizing side ef-
fects and maximizing efficacy.
Biomarker Discovery
Identifying biomarkers for a specific disease/condition
is strenuous work, regarding the intensive experimental
research necessary (fig. 1). Biomarkers discovery research
can be carried out following different possible approach-
es and are summarized below:
(1) Candidate approach: Using this strategy, a pre-
liminary in silico selection of candidate biomarkers
based on meta-analysis of data (from literature or from
the public domain) is done. Candidates can be consid-
ered, e.g. proteins with proven functional relationship
with the disease or some aspects of it, transcripts that are
deregulated or other biological markers that have been
associated with specific biological and/or clinical as-
pects of the disease. This approach is less demanding but
not effective, at least until the full functions and the en-
157.82.153.40 - 5/17/2015 3:11:15 PM
Public Health Genomics 2013;16:313–321 315
DOI: 10.1159/000355938
University of Tokyo
Downloaded by:
Fig. 1. Schematic representation of biomarker
discovery. Different technologies, both genomics
and proteomics, are able to identify potential bio-
markers; subsequently data interpretation, con-
nection between the 2 approaches and validation
will confirm the biomarker identification.
tire knowledge of the human genome and proteome,
and, at the end, the human profilomics will be fully de-
fined.
(2) High-throughput approach: This strategy takes ad-
vantage of the novel techniques (e.g. parallel sequencing,
mass spectrometry (MS), array expression profiling to-
wards whole exome sequencing, and similar omic ap-
proaches) to pick up and identify a small cohort of bio-
markers. This is a very cost-intensive approach, still in
development, but with a high potential (e.g. see the EU
BIO-NMD project).
(3) High-throughput/candidate-combined approach:
This strategy uses the high-throughput techniques using
in silico methods (software) that prioritize the candidates
to be further selected. This is a very useful approach, since
the large amount of data the omic methods generate can
make the discovery of biomarkers very difficult. For ex-
ample, the output of whole exome sequencing for identi-
fying SNPs that are related to a specific phenotype (dis-
ease severity, organ involvement) can be filtered using
software that prioritize the data on the basis of the func-
tional interactome map. In this way, the immense amount
of data normally resulting from omic analysis can be fil-
tered using the interactome and then have a better valida-
tion of and applicability on the disease.
(4) Parallel-omics: This novel approach is complex,
high demanding and is based on a parallel omic analysis
(genomic, proteomic, metabolomics, etc.) often per-
formed on the same individual/patient, at different times
and/or disease stages. This is a recently described method
to identify biomarkers for monitoring dynamic changes
in the personal omic profiles, an approach that is tightly
linked to the personalized medicine concept [9]. Al-
316 Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
though still at an experimental stage, it can be widely used
when omic techniques will become less expensive and
better interpretable in terms of bioinformatics filters ap-
plied.
It deserves to be mentioned that a gene mutation caus-
ing a Mendelian disease has to be considered a genetic
biomarker (specific marker for genetic diagnosis), ac-
cording to the IRDiRC statement (http://www.irdirc.eu),
and, thus, all the gene discovery methods and tools can be
considered as specific strategies for diagnostic biomarker
identification.
Biomarkers Development and Regulatory Aspects
In order to translate newly identified exploratory bio-
markers into clinical practice and to verify their clinical
utility (see EMA definitions below), they need to first pass
through a regulatory procedure.
This procedure as first step goes through qualification,
which is a conclusion that, ‘within the stated context of
use, the results of assessment with a biomarker can be re-
lied upon to adequately reflect a biological process, re-
sponse or event, and support use of the biomarker during
drug or biotechnology product development, ranging
from discovery through post-approval’ [10].
This process leads either to a Committee for Medicinal
Products for Human Use (CHMP) opinion or to a Scien-
tific Advice on novel drug development tools. Recently
EMA approved briefing meetings in order to quickly pro-
pose novel biomarkers even in a preliminary phase. EMA
guidelines are generally applicable to a variety of bio-
marker categories with the aim of harmonizing qualifica-
tion of biomarkers across regions considering the differ-
ent international regulatory authorities.
157.82.153.40 - 5/17/2015 3:11:15 PM
Ferlini /Scotton /Novelli
University of Tokyo
Downloaded by:
Fig. 2. Biomarker flowchart, from the discovery to the development, based on the emerging paradigm for improving translational sci-
ence. The figure shows the mechanistic flowchart from basic science to clinical utility, going through the various validation steps.
Fig. 3. Schematic representation of bio-
marker application. Biomarkers play a role
in disease screening and progression (dis-
ease biomarker) and in monitoring differ-
ent clinical outcomes of a particular thera-
py or treatment (therapeutic biomarker in
clinical trials).
The 3 most important characteristics biomarkers
should have, as EMA has stated in a recent published doc-
ument [11], are analytical validity, clinical validity, and
clinical utility (listed and detailed above).
Figure 2 summarizes the mechanistic flowchart for
biomarker discovery and development.
Ethical Considerations
When biomarkers are used in clinical practice, they do
present ethical issues, including privacy rules on genetic
information communication, recruiting patients or vol-
unteers for clinical trial or research purposes, collecting
human samples for sharing data, patients’ databases and
registries, and furthermore, all the issues related to the
data ownership, exploitation and dissemination. In gen-
eral, the RDs field needs academia, industry and parent
advocacy working closer together by establishing com-
Biomarkers and Genetics
mon guidelines that are to be followed, according to the
existent normative [12].
Types of Biomarkers: Applications
Diagnostic/Prognostic Biomarker
A diagnostic or prognostic biomarker classically iden-
tifies patients that can be correlated with clinical out-
comes, such as disease screening, clinical measures mon-
itoring and disease progression, and disease risk predic-
tion (fig. 3).
The diagnostic biomarkers are signatures that sepa-
rate the healthy population from the disease phenotype
or discriminate against groups based on phenotype se-
verity.
For example, the amyotrophic lateral sclerosis (ALS,
OMIM 105400) is a neurodegenerative disease character-
ized by lack of specific diagnostic test. Hwang et al. [13]
157.82.153.40 - 5/17/2015 3:11:15 PM
Public Health Genomics 2013;16:313–321 317
DOI: 10.1159/000355938
University of Tokyo
Downloaded by:
have correlated the increased level of HMGB1 auto-Ab in
serum samples with the onset of amyotrophic lateral scle-
rosis. This higher HMGB1 auto-Ab level also correlates
with the disease severity.
A study on Marfan syndrome (MFS, OMIM 154700),
a systemic disease of the connective tissue characterized
by a combination of cardiovascular, muscle-skeletal and
pulmonary manifestations, represents an example of a
prognostic biomarker. Patients with Marfan syndrome
suffer from an increased risk of cardiovascular manifesta-
tions that may cause premature death; this study shows
that high serum levels of the transforming growth factor-β
(TGF-β) correlate with larger aortic root diameters and
faster aortic root growth predicting cardiovascular events,
being, therefore, of prognostic value [14].
A recent study on DMD confers a specific role to ma-
trix metalloproteinase-9 (MMP-9) and tissue inhibitors
of metalloproteinase-1 (TIMP-1) proteins as diagnostic
or prognostic biomarkers; indeed MMP-9 and TIMP-1
serum levels were significantly higher in DMD patients
compared to healthy controls. In addition, MMP-9 levels
were also significantly higher in older, nonambulant pa-
tients, compared to ambulant patients [15].
As mentioned before, all genes causative for RDs are
diagnostic biomarkers. This implies that dystrophin and
the cystic fibrosis transmembrane conductance regulator
(CFTR, OMIM 602421) have to be considered as bio-
markers that allow a precise genetic diagnosis of DMD
and cystic fibrosis (CF, OMIM 219700).
Predictive/Therapeutic Biomarker
A predictive biomarker highlights the different out-
comes of a particular therapy or treatment, discriminat-
ing the patients in 2 categories: responders and nonre-
sponders, or high and low responders (fig. 3).
In the majority of cases, therapeutic biomarkers are
SNPs, as for typical pharmacogenetic biomarkers.
In RDs, there are a few examples of therapeutic bio-
markers. A discrimination between groups depending on
the pharmacological treatment response has been report-
ed in a recent study on idiopathic nephrotic syndrome, a
rare disease affecting the kidneys, where the authors
found a significant difference in serum proteomes be-
tween steroid-sensitive nephrotic syndrome (SSNS) and
steroid-resistant nephrotic syndrome (SRNS, OMIM
256370) patients [16].
The predictive biomarkers in clinical research and
practice are important since evaluating the treatment/in-
tervention efficacy/efficiency plays a crucial role in mak-
ing decisions on treatment choices and therapy outcomes.
318 Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
Types of Biomarkers: Features
Genomic Biomarkers
While mutations in genes in Mendelian disorders are
classically deterministic, polymorphisms or allelic varia-
tions could be related to disease susceptibility, disease
progression or therapeutic response.
Accordingly, with the Food and Drug Administration/
EMA definition, genomic biomarkers include both DNA
and RNA determinants. Since the publication of the hu-
man genome sequence, increasing knowledge in the fields
of biology and informatics led to the development of
high-throughput technologies such as microarray and
next-generation sequencing platforms.
Microarray technologies are able to quantitatively ex-
amine the DNA/RNA levels/configurations of many
genes in diseased and healthy samples simultaneously
and at different time points (e.g. pre- and post-treat-
ment). Similarly, the new next-generation sequencing
platforms allow determining genomic nucleotide chang-
es in multiple samples simultaneously, reducing time and
costs compared to a standard Sanger sequencing technol-
ogy. In addition, RNA-sequencing (RNAseq) analyzes
the whole transcriptome both quantitatively and qualita-
tively providing information about allelic-specific ex-
pression and alternative spliced isoforms.
These techniques will facilitate the study of human ge-
nome variability increasing the biomarker discovery, as
detailed above.
miRNAs represent a field of intense investigation in
the field of biomarkers. Recent studies found that the lev-
els of muscle specific miRNAs (myomir) are increased in
the serum of animal models [17] as well as patients with
neuromuscular disorders [18].
Significantly, myomirs increased in more severe pheno-
types, possibly being biomarkers of disease severity. A spe-
cific miRNA signature has also been identified by Eisen-
berg et al. [18] to distinguish between different categories
of neuromuscular disorders, for example, Duchenne/
Becker or specific limb girdle muscular dystrophies.
The miRNAs studied have extended beyond other
RDs, for example, Megiorni et al. [19] observed significantly
upregulated levels of 2 miRNAs in nasal epithelial tissues
from cystic fibrosis (OMIM 219700) patients versus
healthy controls. The author suggested a role of miRNAs
as biomarkers in disease severity by modulating CFTR
gene levels as well as the expression of important mole-
cules involved in the inflammatory response [19].
Genomic biomarkers also include the epigenomic
modification such as DNA methylation. Recent studies
demonstrated a correlation between DNA methylation
157.82.153.40 - 5/17/2015 3:11:15 PM
Ferlini /Scotton /Novelli
University of Tokyo
Downloaded by:
and the clinical outcome in Friedreich ataxia (FRDA,
OMIM 229300). In particular, the authors evaluated the
methylation status immediately upstream and down-
stream of the GAA triplet expansion that determines the
pathological phenotype. The hypermethylation of the up-
stream region correlates with the clinical severity, while
the hypomethylation of the downstream region correlates
with the age at onset [20]. These results suggest an impor-
tant role of methylation and more in general epigenomics
as diagnostic and prognostic biomarkers.
Proteomic Biomarkers
Proteomic studies have the potentiality to identify bio-
markers more closely related to biological function/dys-
function.
Protein biomarkers have the remarkable advantage to
be determined in accessible body fluids (blood, urine) of
patients being, therefore, usable as early indicators of the
disease condition, repeatedly monitored markers for
drug response or adverse effects in clinical trials. It is,
however, known that a limit of proteomic biomarkers
may be related to the poor mirroring they may have of the
original disease in the target tissue/organ, making them
often low specific. The most used method to measure pro-
teins is the immunoassay, based on the properties of an-
tibodies to capture and detect specific protein domains.
Immunoassay can pick up a single protein using mono-
clonal antibodies and is widely considered the privileged
tool to validate biomarkers because of its high sensitivity.
In the last years, many efforts have been done to develop
multi-analyte readouts using multiplexed immunoas-
says. These assays are able to investigate several proteins
simultaneously using very low amounts of samples. Dif-
ferent commercial multiplexes are now available, and
these panels are able to screen up to a few hundred pro-
teins in the same sample. For example, in idiopathic pul-
monary fibrosis (IPF, OMIM 178500) a multiplexed assay
allowed evaluating 92 proteins in plasma samples from
more than 200 patients, identifying 3 biomarkers associ-
ated with disease progression [21]. Other studies used
ELISA immunoassay to associate high serum levels of an
extracellular matrix glycoprotein, Tenascin-C (TN-C), to
an increased risk of developing dilated cardyomyopathy
in patients affected by Emery-Dreifuss muscular dystro-
phy (EMD, OMIM 310300) [22].
MS is another important protein assay, even if, prior
to analysis, the samples have to be extensively processed.
The most common method to detect proteins and bio-
markers by MS involves the cleavage of large molecules
into peptides. After this step, peptides can be analyzed by
Biomarkers and Genetics
various MS technologies, such as gas chromatography-
MS or liquid chromatography-MS. The innovative pro-
teomics approach, such as 2-DE/MALDI-MS and LC/
ESI-MS adopted by Sedic et al. [23], allowed the identifi-
cation of carbonic anhydrase3, creatine kinase and throm-
bospondin4 proteins as plasma diagnostic biomarkers in
S-Adenosylhomocysteine hydrolase deficiency (AHCY,
OMIM 613752), a rare congenital disorders characterized
by psychomotor delay and severe myopathy.
This technology was also used to identify diagnostic or
therapeutic biomarkers in other biological fluids such as
urine; indeed, Kistler et al. [24] examined urine pro-
teomics by capillary electrophoresis coupled to MS to
pick up both diagnostic and therapeutic markers. Indeed,
they could define a biomarker model usable either for di-
agnosis or for therapy monitoring in female Fabry pa-
tients (FD, OMIM 301500). Proteomic biomarkers are
also very appealing as screening tests (e.g. neonatal test-
ing) for their affordable costs and low invasiveness.
Other Biomarkers
Identifying biomarkers also include all diagnostic
tests, imaging technologies and other approaches that are
able to measure the disease status of patients.
For example, MRI is a safe and noninvasive approach
without ionizing radiation, able to resolve muscular and
connective tissue, fat and bone. In muscle diseases, MRI
scans may be used to correlate the clinical phenotype
with the muscle biopsy morphology and immunocyto-
chemistry. Kinali et al. [25] have demonstrated that MRI
scanning can be a reliable tool to inform the choice of
muscles to be assessed during clinical trials. In this paper,
reporting data on a clinical trial based on antisense oli-
gonucleotide in DMD, the results of the MRI scan serve
as biomarkers of disease progression, in particular mus-
cles, by identifying which ones are sufficiently preserved.
Generally, imaging biomarkers are considered very ap-
pealing, and recently, there has been intensive research
on these biomarkers, as demonstrated by a specific EU
call.
The BIO-NMD EU Project
As we have summarized above, biomarkers are ex-
tremely useful tools for mirroring and/or surrogate end-
points for establishing efficacy in clinical trials, for an-
ticipating the severity of disease course (disease stratifica-
tion and prognosis) for use as diagnostic tools, for
defining both pharmacokinetic and pharmacodynamic
157.82.153.40 - 5/17/2015 3:11:15 PM
Public Health Genomics 2013;16:313–321 319
DOI: 10.1159/000355938
University of Tokyo
Downloaded by:
action of drugs, to define mechanisms of drug response
(patient responders vs. nonresponders), and for assessing
drug safety in trials.
The BIO-NMD project is a research network funded
by the EU from 2009 to 2012 and aimed to discover and
validate biomarkers in genetic, rare and neuromuscular
diseases. The main goal of the project was designing ap-
propriate, harmonized and efficacious therapies for
NMDs by bridging basic research and clinical research
(http://www.bio-nmd.eu).
Two pivotal diseases were chosen: dystrophinopathies
and COL6-related myopathies.
BIO-NMD was based both on a targeted biomarkers
discovery (proceeded by candidates) and on omic ap-
proaches (genomic, transcriptomics, proteomics). The
first approach (targeted search for biomarkers) is mainly
based on the selection of candidate genes/proteins using
bioinformatic tools. This prioritization consists of the
definition of functional pathways (interactome), which
disclose candidate biomarkers to be further studied via
other strategies. The second (omics) meant to analyze the
whole genome, transcriptome or proteome sets using in-
novative technologies, such as whole exome sequencing,
sequencing of the full transcriptome (RNAseq) and pro-
teome profiling, by custom designed multiplexes immu-
noassays or using novel high-throughput techniques on
the whole proteome. In addition, BIO-NMD adopted si-
multaneously all omic approaches (genomics, transcrip-
tomics and proteomics) on different human tissues/cells/
fluids in order to translate omic data into noninvasive
human biomaterials or low-invasive tissues, to be used for
monitoring disease progression, prognosis and drug re-
sponse, therefore, optimizing the choice of appropriate
and often personalized therapies. An example of the BIO-
NMD project approach is reported in Kotelnikova et al.
[26].
The International Rare Diseases Research
Consortium: The Goal
As joint effort will maximize the success and reduce
the cost of developing therapies for RDs, it is, therefore,
crucial to cooperate within a national and supranational
context; this has been recognized via the formation of a
novel, cooperative initiative by the EU, Canada and the
USA, and the International Rare Diseases Research Con-
sortium (IRDiRC) (http://ec.europa.eu/research/health/
medical-research/rare-diseases/irdirc_en.html). The
IRDiRC states that diagnosis for most patients affected by
320 Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
a RD remains challenging and that most RDs are lacking
dedicated therapies. The objective of the IRDiRC is to de-
liver diagnostic tests for most RDs and 200 new therapies
for RD patients by 2020.
Conclusion and Future Vision
The importance of biomarkers in RDs for clinical di-
agnosis or treatment response is increasing. Their use,
from monitoring disease progression to predicting re-
sponse to treatment, drives the research towards a per-
sonalized medicine. In this context, biomarker discovery,
verification, validation, and qualification with the regula-
tory authorities are crucial steps towards the selection of
the right patients for the right therapy. For RDs, however,
the bottleneck we face is the small number of patients of-
ten disseminated around the world, with consequent dif-
ferences in the standard of care, diagnosis and therapy.
This is why a worldwide cooperation is required in RDs
for data sharing, harmonizing common patient’s stratifi-
cation procedures, having wide biobank and sample re-
pository, and, last but not least, increasing critical mass
and experience.
Accordingly to the IRDiRC guidelines and hopes, we
should intensify the effort to work in very large coopera-
tion actions, with shared visions and strategies. This will
maximize success in RD research and will benefit patients
improving their quality of life.
Acknowledgement
The EU BIO-NMD project (grant 241665) to A.F. and G.N. is
kindly acknowledged.
References 1 Beck M: ‘Rare’ and ‘Ultra Rare Diseases’. Out-
look Develop Drugs 2012;1:e107.
2 Roubertoux PL, de Vries PJ: From molecules
to behavior: lessons from the study of rare ge-
netic disorders. Behav Genet 2011; 41: 341–
348.
3 Pavlovich CP, Schmidt LS: Searching for the
hereditary causes of renal-cell carcinoma. Nat
Rev Cancer 2004;4:381–393.
4 Biomarkers Definitions Working Group: Bio-
markers and surrogate endpoints: preferred
definitions and conceptual framework. Clin
Pharmacol Ther 2001;69:89–95.
157.82.153.40 - 5/17/2015 3:11:15 PM
Ferlini /Scotton /Novelli
University of Tokyo
Downloaded by:
 5 Finkel RS, Crawford TO, Swoboda KJ,
Kaufmann P, Juhasz P, Li X, Guo Y, Li RH,
Trachtenberg F, Forrest SJ, Kobayashi DT,
Chen KS, Joyce CL, Plasterer T; Pilot Study of
Biomarkers for Spinal Muscular Atrophy Tri-
al Group: Candidate proteins, metabolites
and transcripts in the Biomarkers for Spinal
Muscular Atrophy (BforSMA) clinical study.
PLoS One 2012;7:e35462.
6 Mitrpant C, Fletcher S, Wilton SD: Person-
alised genetic intervention for Duchenne
muscular dystrophy: antisense oligomers and
exon skipping. Curr Mol Pharmacol 2009; 2:
110–121.
7 Ma Q, Lu AY: Pharmacogenetics, pharma-
cogenomics, and individualized medicine.
Pharmacol Rev 2011;63:437–459.
8 European Medicines Agency: Qualification
opinion of Alzheimer’s disease novel method-
ologies/biomarkers for PET amyloid imaging
(positive/negative) as a biomarker for enrich-
ment for use – in predementia AD clinical tri-
als. 2011. http://www.ema.europa.eu/docs/
en_GB/document_library/Regulatory_and_
procedural_guideline/2011/12/
WC500118364.pdf.
9 Chen R, Mias GI, Li-Pook-Than J, Jiang L,
Lam HY, Chen R, Miriami E, Karczewski KJ,
Hariharan M, Dewey FE, Cheng Y, Clark MJ,
Im H, Habegger L, Balasubramanian S,
O’Huallachain M, Dudley JT, Hillenmeyer S,
Haraksingh R, Sharon D, Euskirchen G, La-
croute P, Bettinger K, Boyle AP, Kasowski M,
Grubert F, Seki S, Garcia M, Whirl-Carrillo
M, Gallardo M, Blasco MA, Greenberg PL,
Snyder P, Klein TE, Altman RB, Butte AJ,
Ashley EA, Gerstein M, Nadeau KC, Tang H,
Snyder M: Personal omics profiling reveals
dynamic molecular and medical phenotypes.
Cell 2012;148:1293–1307.
10 EMEA: Definitions for genomic biomarkers,
pharmacogenomics, pharmacogenetics, ge-
nomic data and sample coding categories.
2006. http://www.ema.europa.eu/docs/en_
GB/document_library/Scientific_guide-
line/2009/09/WC500002880.pdf.
11 European Medicines Agency: ICH guideline
E16 genomic biomarkers related to drug re-
sponse: context, structure and format of qual-
ification submissions. 2009. http://www.ema.
europa.eu/docs/en_GB/document_library/
Scientific_guideline/2010/09/WC500097060.
pdf.
Biomarkers and Genetics
12 McCormack P, Woods S, Aartsma-Rus A,
Hagger L, Herczegfalvi A, Heslop E, Irwin J,
Kirschner J, Moeschen P, Muntoni F, Ouil-
lade MC, Rahbek J, Rehmann-Sutter C,
Rouault F, Sejersen T, Vroom E, Straub V,
Bushby K, Ferlini A: Guidance in social and
ethical issues related to clinical, diagnostic
care and novel therapies for hereditary neuro-
muscular rare diseases: ‘translating’ the trans-
lational. PLoS Curr 2013;10:5.
13 Hwang CS, Liu GT, Chang MD, Liao IL,
Chang HT: Elevated serum autoantibody
against high mobility group box 1 as a potent
surrogate biomarker for amyotrophic lateral
sclerosis. Neurobiol Dis 2013;58:13–18.
14 Franken R, den Hartog AW, de Waard V,
Engele L, Radonic T, Lutter R, Timmermans
J, Scholte AJ, van den Berg MP, Zwinderman
AH, Groenink M, Mulder BJ: Circulating
transforming growth factor-β as a prognostic
biomarker in Marfan syndrome. Int J Cardiol
2013;168:2441–2446.
15 Nadarajah VD, van Putten M, Chaouch A,
Garrood P, Straub V, Lochmüller H, Ginjaar
HB, Aartsma-Rus AM, van Ommen GJ, den
Dunnen JT, ‘t Hoen PA: Serum matrix metal-
loproteinase-9 (MMP-9) as a biomarker for
monitoring disease progression in Duchenne
muscular dystrophy (DMD). Neuromuscul
Disord 2011;21:569–578.
16 Wen Q, Huang LT, Luo N, Wang YT, Li XY,
Mao HP, Zhang L, Dong XQ, Yu XQ: Pro-
teomic profiling identifies haptoglobin as a
potential serum biomarker for steroid-resis-
tant nephrotic syndrome. Am J Nephrol 2012;
36:105–113.
17 Mizuno H, Nakamura A, Aoki Y, Ito N, Kishi
S, Yamamoto K, Sekiguchi M, Takeda S,
Hashido K: Identification of muscle-specific
microRNAs in serum of muscular dystrophy
animal models: promising novel blood-based
markers for muscular dystrophy. PLoS One
2011;6:e18388.
18 Eisenberg I, Alexander MS, Kunkel LM:
miRNAS in normal and diseased skeletal
muscle. J Cell Mol Med 2009;13:2–11.
Public Health Genomics 2013;16:313–321
DOI: 10.1159/000355938
19 Megiorni F, Cialfi S, Cimino G, De Biase RV,
Dominici C, Quattrucci S, Pizzuti A: Elevated
levels of miR-145 correlate with SMAD3
down-regulation in Cystic Fibrosis patients. J
Cyst Fibros 2013;pii:S1569–1993(13)00043-
X.
20 Evans-Galea MV, Carrodus N, Rowley SM,
Corben LA, Tai G, Saffery R, Galati JC, Wong
NC, Craig JM, Lynch DR, Regner SR, Brocht
AF, Perlman SL, Bushara KO, Gomez CM,
Wilmot GR, Li L, Varley E, Delatycki MB, Sar-
sero JP: FXN methylation predicts expression
and clinical outcome in Friedreich ataxia.
Ann Neurol 2012;71:487–497.
21 Richards TJ, Kaminski N, Baribaud F, Flavin
S, Brodmerkel C, Horowitz D, Li K, Choi J,
Vuga LJ, Lindell KO, Klesen M, Zhang Y, Gib-
son KF: Peripheral blood proteins predict
mortality in idiopathic pulmonary fibrosis.
Am J Respir Crit Care Med 2012;185:67–76.
22 Niebroj-Dobosz I, Madej-Pilarczyk A,
Marchel M, Sokołowska B, Hausmanowa-Pe-
trusewicz I: Circulating tenascin-C levels in
patients with dilated cardiomyopathy in the
course of Emery-Dreifuss muscular dystro-
phy. Clin Chim Acta 2011;412:1533–1538.
23 Sedic M, Kraljevic Pavelic S, Cindric M, Vis-
sers JP, Peronja M, Josic D, Cuk M, Fumic K,
Pavelic K, Baric I: Plasma biomarker identifi-
cation in S-adenosylhomocysteine hydrolase
deficiency. Electrophoresis 2011; 32: 1970–
1975.
24 Kistler AD, Siwy J, Breunig F, Jeevaratnam P,
Scherl A, Mullen W, Warnock DG, Wanner
C, Hughes DA, Mischak H, Wüthrich RP,
Serra AL: A distinct urinary biomarker pat-
tern characteristic of female Fabry patients
that mirrors response to enzyme replacement
therapy. PLoS One 2011;6:e20534.
25 Kinali M, Arechavala-Gomeza V, Cirak S,
Glover A, Guglieri M, Feng L, Hollingsworth
KG, Hunt D, Jungbluth H, Roper HP, Quinli-
van RM, Gosalakkal JA, Jayawant S, Nadeau
A, Hughes-Carre L, Manzur AY, Mercuri E,
Morgan JE, Straub V, Bushby K, Sewry C,
Rutherford M, Muntoni F: Muscle histology
vs MRI in Duchenne muscular dystrophy.
Neurology 2011;25:76:346–353.
26 Kotelnikova E, Shkrob MA, Pyatnitskiy MA,
Ferlini A, Daraselia N: Novel approach to me-
ta-analysis of microarray datasets reveals
muscle remodeling-related drug targets and
biomarkers in Duchenne muscular dystro-
phy. PLoS Comput Biol 2012;8:e1002365.
157.82.153.40 - 5/17/2015 3:11:15 PM
321
University of Tokyo
Downloaded by: