Content uploaded by Eduardo Martínez-Morillo
Author content
All content in this area was uploaded by Eduardo Martínez-Morillo on Dec 15, 2014
Content may be subject to copyright.
Peptidomics of Urine and Other Biofluids for
Cancer Diagnostics
Josep Miquel Bauc¸a,
1
Eduardo Martı´nez-Morillo,
2
and Eleftherios P. Diamandis
3,4,5*
BACKGROUND:Cancer is a leading cause of death world-
wide. The low diagnostic sensitivity and specificity of
most current cancer biomarkers make early cancer di-
agnosis a challenging task. The comprehensive study of
peptides and small proteins in a living system, known
as “peptidomics,” represents an alternative technolog-
ical approach to the discovery of potential biomarkers
for the assessment of a wide variety of pathologies. This
review examines the current status of peptidomics for
several body fluids, with a focus on urine, for cancer
diagnostics applications.
CONTENT:Several studies have used high-throughput
technologies to characterize the peptide content of dif-
ferent body fluids. Because of its noninvasive collection
and high stability, urine is a valuable source of candi-
date cancer biomarkers. A wide variety of preanalytical
issues concerning patient selection and sample han-
dling need to be considered, because not doing so can
affect the quality of the results by introducing bias and
artifacts. Optimization of both the analytical strategies
and the processing of bioinformatics data is also essen-
tial to minimize the false-discovery rate.
SUMMARY:Peptidomics-based studies of urine and
other body fluids have yielded a number of biomol-
ecules and peptide panels with potential for diagnosing
different types of cancer, especially of the ovary, pros-
tate, and bladder. Large-scale studies are needed to val-
idate these molecules as cancer biomarkers.
© 2013 American Association for Clinical Chemistry
Cancer is a major clinical problem worldwide. Ac-
counting for approximately 1 in every 4 deaths, cancer
represents the second leading cause of death in devel-
oped countries, after cardiovascular diseases. It is esti-
mated that more than 1.6 million new cancer cases are
diagnosed every year in the US (1 ). Highly heteroge-
neous and with only a few effective therapeutic strate-
gies, cancer represents a challenging medical condition
for both healthcare professionals and governments.
The possibility of detecting cancer at early stages,
before it spreads to anatomically distant tissues, has
long interested physicians and scientists, because early
diagnosis is a key factor for successful treatment out-
comes. For instance, the overall 5-year survival rate for
ovarian cancer is ⬍40%, whereas the rate increases to
90% if it is detected in its early stages (2 ). Conse-
quently, Finding new tools, such as endogenous
biomolecules, that could help identify patients with
early-stage disease is highly desirable. As stated by the
WHO in 1968, the ideal biomarker for a disease should
be measurable via a simple, reliable, and affordable
method and have a high diagnostic sensitivity and
specificity. The biomarker should be present in higher-
than-normal concentrations during early disease
stages, and its concentration should reflect the extent
or severity of the disease. Defining the target popula-
tion for whom the test would be applied is also of major
concern.
Unfortunately, only a few cancer biomarkers have
entered routine use. Even fewer have been approved for
population screening or diagnosis (3 ). One of the most
frequently used cancer biomarkers is prostate-specific
antigen (PSA).
6
Despite its widespread measurement,
many issues relating to overdiagnosis and overtreat-
ment have arisen because serum PSA also increases in
benign prostatic hyperplasia and other nonmalignant
diseases (4 ). Similarly and despite being considered the
best biochemical marker for breast cancer, carbohy-
drate antigen 15.3 is also increased in other tumors,
such as pancreatic and colorectal cancers, as well as in a
number of benign pathologies. The lack of diagnostic
sensitivity for this antigen, the serum concentrations of
1
Servei d’Ana` lisis Clı´niques, Hospital Universitari Son Espases, Palma de Mal-
lorca, Spain;
2
Samuel Lunenfeld Research Institute, Joseph and Wolf Lebovic
Health Centre, Mount Sinai Hospital, Toronto, Ontario, Canada;
3
Department of
Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario,
Canada;
4
Department of Clinical Biochemistry, Toronto General Hospital, Uni-
versity Health Network, Toronto, Ontario, Canada;
5
Department of Pathology
and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, Canada.
* Address correspondence to this author at: Department of Pathology and
Laboratory Medicine, Mount Sinai Hospital, Suite 6-201, Box 32, 60 Murray St.,
Toronto, Ontario, M5T 3L9 Canada. Fax 416-619-5521; e-mail ediamandis@
mtsinai.on.ca.
Received June 21, 2013; accepted October 22, 2013.
Previously published online at DOI: 10.1373/clinchem.2013.211714
6
Nonstandard abbreviations: PSA, prostate-specific antigen; CSF, cerebrospinal
fluid; LC-MS/MS, liquid chromatography–tandem mass spectrometry; CA125,
carbohydrate antigen 125.
Clinical Chemistry 60:7
000 – 000 (2014) Reviews
1
http://hwmaint.clinchem.org/cgi/doi/10.1373/clinchem.2013.211714The latest version is at
Papers in Press. Published November 8, 2013 as doi:10.1373/clinchem.2013.211714
Copyright (C) 2013 by The American Association for Clinical Chemistry
which barely increase in early-stage malignancy, is also
an important limitation (5 ). Clearly, there is an urgent
need to discover and validate new biomarkers with bet-
ter performance characteristics.
High-throughput technologies that generate mas-
sive quantities of data have become known as “omics,”
a suffix derived from “genomics,” the comprehensive
study of genes and other DNA sequences (i.e., the
genome)— hence transcriptomics, proteomics, metabo-
lomics, epigenomics, and peptidomics. Three steps are
essential in the process of developing a biomarker (3 ):
(a) the discovery of candidate molecules in defined pa-
tient groups, (b) validation of the biomarkers for their
capacity to assist in disease assessment, and (c) imple-
mentation in the clinical setting.
This review focuses on the current situation of
cancer biomarker discovery through peptidomics. The
emphasis is on urine, but this review also covers inves-
tigations of other body fluids. Technological aspects of
peptidomics and their applications to different types of
cancer are also reviewed.
Proteomics and Peptidomics
Since the dawn of the genomics and transcriptomics
era, numerous efforts have been directed toward dis-
covering biomarkers that could help in the diagnosis,
prognosis, or monitoring of different diseases. The
main limitation of nucleic acid– based approaches is
that recognition of an inherited predisposition to dis-
ease is usually not sufficient to identify the biological
processes and mechanisms by which they operate (6 ).
This limitation can be partially alleviated with pro-
teomics. A complete analysis of the protein content of a
cell, tissue, or organism comprises all of the layers of
information gathered from the genome and the tran-
scriptome, plus posttranslational modifications (e.g.,
phosphorylation, glycosylation). Proteomics is the
large-scale study of the full complement of proteins in a
living system—their structures, their physicochemical
properties, and their functions. Proteomic technolo-
gies have the potential to detect dynamic changes in the
production of proteins via these technologies’ integra-
tion of the proteome’s genetic and epigenetic features
(7 ). The proteome is hence much more complex than
the genome or the transcriptome, and it appears to re-
flect actual cellular processes more accurately than the
genome or the transcriptome. Proteins are the effectors
of biochemical actions (Fig. 1). From a pathophysio-
logical point of view, genetic analyses can predict the
risk of developing a disease, whereas proteomic ap-
proaches have a capacity both to show when the risks
become evident as a disease and to facilitate monitor-
ing of the therapeutic response. Both the concentra-
tions of proteins and their posttranslational modifica-
tions may be altered during disease progression (8 ).
The peptidome constitutes the low molecular
weight proteome. The term “peptidomics,” a term
coined in 1996, is the systematic and comprehensive
analysis of the small proteins and endogenous peptides
of biological samples at a defined time. Peptidomics
typically encompasses polypeptides ⱕ20 kDa, al-
though no clear limit has been established. It is inter-
esting that results obtained with the first proteomic
methodologies appeared to indicate that the peptide
content samples was too simple and easily cleared by
the kidney to carry useful information. That turned out
not to be the case, for research efforts with peptidomics
have already yielded positive results.
Most peptides in biological systems are not syn-
thesized as such, but rather are derived from precursor
proteins via proteolytic cleavage by endogenous pepti-
dases in a specific or nonspecific way (Fig. 2), e.g., the
activation of some zymogens of the coagulation cas-
cade or the maturation of insulin. Other peptides are
generated in situ and then traverse the endothelial vas-
culature, if they are sufficiently small to enter the blood
passively or the wall becomes permeable owing to dis-
ease conditions (9, 10 ).
Proteolytic processing has been theorized to be
necessary to facilitate metabolic variation so that indi-
Fig. 1. Application of “omics” technologies for dis-
covering novel biomarkers.
Peptidomics may encompass additional sources of informa-
tion owing to proteolytic processing of proteins by active
enzymes.
Reviews
2Clinical Chemistry 60:7 (2014)
viduals and species may better adapt to exogenous
stimuli (11 ). Actually, peptides in body fluids are be-
lieved to be due to an imbalance between the activities
of proteases on the one hand and the actions of pro-
tease inhibitors on the other. In this way, endogenous
proteases are differentially regulated in the contexts of
many physiological and pathologic phenomena (12 ).
Therefore, it is reasonable to hypothesize that studying
protease activity and regulation can lead to improved
detection and a deeper understanding of the molecular
mechanisms of some diseases.
Peptides also play central roles in healthy physio-
logical processes (13 ). That is the case for many cyto-
kines, growth factors, and some neuropeptides for
which proteome mapping studies have revealed no
precursor protein (14 ), suggesting that they are synthe-
sized as such in the central nervous system and are not
breakdown products of precursor proteins. Even pep-
tides processed from large proteins usually show bio-
logical functions and activities different from their par-
ent molecules (7 ). Given that a high percentage of
proteins undergo proteolytic cleavage, identifying and
characterizing their breakdown products might be of
interest, because they could be even more informative
than the precursor protein (10 ). The study of differen-
tial protease activities could be an inviting field for the
application of peptidomics to medicine. For example,
neoplastic processes are involved in the transformation
and proliferation of certain cell types and thus alter the
concentrations and activities of specific proteins and
enzymes, such as proteases. Therefore, not only do pro-
teins in the system (proteomics) become altered, but
their metabolic products (peptides), which should be
regarded as an extension of the proteome, also change.
Peptidomics of Body Fluids
The peptidome constitutes a still mostly unexplored
source of biological information, and it might provide
useful biomarkers for disease assessment. Peptides in
body fluids are proxies for protein synthesis, process-
ing, and degradation. Worth mentioning is that some
peptide biomarkers have already entered the clinic, al-
though none of them were discovered with contempo-
rary peptidomic methods (15 ) (Table 1). The peptides
most widely known are the aminoterminal propeptide
of brain natriuretic peptide (which is measured in se-
rum for assessing heart failure) and C-peptide (for
monitoring endogenous insulin production in diabetes
patients). Collagen N-terminal telopeptides are mea-
sured in measured in urine as biomarkers of bone turn-
over (16 ). Thus, body fluids represent attractive
sources to mine for informative proteins and peptides
(Table 2).
PEPTIDOMICS OF BLOOD
Blood fluid (serum or plasma) is regarded as the most
valuable specimen for biomarker elucidation (17 ), be-
cause blood is the transportation medium for most
tissue-derived molecules in the organism. Therefore,
Fig. 2. Generation of a series of peptides from a core
peptide produced via endopeptidase activity.
Amino- and carboxypeptidases can further cleave the core
peptide, generating numerous fragments, which can then
be identified by mass spectrometry
(12, 64)
.
Table 1. Examples of current peptide biomarkers
used in clinical diagnosis.
Biomarker Fluid Disease
NT–pro-BNP
a
Blood (serum) Heart failure, ventricular
dysfunction
Pro-GRP Blood (serum) Neuroendocrine tumors,
small cell lung cancer

-CTX Blood (serum) Bone turnover
PINP Blood (serum) Bone turnover
Pancreatic
polypeptide
Blood (serum) Neuroendocrine tumors
Osteocalcin Blood (serum) Osteoporosis

2
-Microglobulin Blood (serum) Renal disease and
inflammation
Calcitonin Blood (serum) Medullary thyroid
carcinoma
Cystatin C Blood (serum) Renal failure
C-peptide Blood (serum),
urine
Diabetes mellitus
VIP Blood (plasma) Pancreatic tumor
ANF Blood (plasma) Heart failure
NTX Urine Bone turnover

-Amyloid (1–42) CSF Alzheimer disease
a
NT–pro-BNP, N-terminal end of the pro–brain natriuretic peptide; pro-
GRP, pro–gastrin-releasing peptide;

-CTX, cross-linked collagen type I
C-terminal telopeptide; PINP, procollagen type I N-terminal propeptide;
VIP, vasoactive intestinal peptide; ANF, atrial natriuretic factor; NTX,
collagen type 1 N-terminal telopeptide.
Peptidomics in Cancer Reviews
Clinical Chemistry 60:7 (2014) 3
this biofluid can reveal the pathophysiological states of
a broad spectrum of tissues and organs. Compared
with healthy cells, disease-affected cells within tissues
could differentially harbor peptides and proteins even-
tually released into the interstitial fluid and later into
the bloodstream. The high protein content of serum
makes it an attractive fluid for peptidomics; however,
proteins and peptides are present in serum over a wide
and dynamic range of concentrations—⬎10 orders of
magnitude (7 ). This fact represents a considerable an-
alytical challenge, because a few high-abundance pro-
teins (albumin, immunoglobulins, transferrin,
␣
1
-
antitrypsin, haptoglobin) hamper the identification of
low-abundance molecules, which are more likely to be
biomarker candidates (18 ). Given that the composi-
tion of blood reflects the metabolic state of the entire
body as it transports molecules released from virtually
any tissue or organ, pathophysiological changes in any
single organ could easily be missed. Because the clot-
ting time has a substantial effect on the polypeptide
content of serum, plasma is often used instead (19 ).
Nevertheless, a few studies have demonstrated the
applicability of serum peptidomic profiling to a range
of medical conditions, although most of these studies
have not been validated for biases and artifacts. Shen et
al. (20 ) explored the plasma peptidome (the entire col-
lection of protein breakdown products) in the quest for
breast cancer biomarkers and detected increased con-
centrations of cancer-relevant protein products, in-
cluding extracellular matrix components, innate
immune system molecules, proteases, and protease in-
hibitors. In another study, Villanueva et al. (13 ) com-
pared the peptidomes of patients with metastatic thy-
roid carcinoma with those of age- and sex-matched
controls and obtained a 12-peptide signature for iden-
tifying malignancy that had a 95% diagnostic sensitiv-
ity and 95% specificity. These and similar approaches
have been challenged for bias and artifacts (http://
www.jci.org/eletters/view/26022).
PEPTIDOMICS OF CEREBROSPINAL FLUID
Cerebrospinal fluid (CSF) is considered an outstanding
source of biomarkers for neurologic diseases. A color-
less fluid produced in the choroid plexus in the brain,
CSF provides mechanical protection, nutrient supply,
waste product removal, and metabolite transportation.
Via continual interactions, CSF contains molecules
that can reflect many of the processes of the central
nervous system (14 ). Proteomic efforts have been
aimed at characterizing the CSF peptidome to discover
potential biomarkers for neurodegenerative conditions,
neuropsychiatric disorders, traumatic brain injury, brain
tumors, and aging-related conditions (21, 22 ). Wijte et al.
used an enhanced mass spectrometry– based approach
for both free peptides and peptides bound to proteins to
evaluate postmortem CSF from patients with Alzheimer
disease and identified a series of candidate peptides for its
diagnosis, such as VGF, nerve growth factor, and C4 com-
plement precursor (23 ). Additional verification studies
remain to be done.
An analysis by Zougman et al. (14 ) revealed 391
peptides derived from 91 different proteins, and a more
recent study yielded 626 unique peptide sequences of
⬍5 kDa that were derived from 104 proteins (24 ). The
enrichment of CSF with small peptides might be due to
a higher rate of filtration from plasma compared with
components of higher molecular weight.
One of the major disadvantages of CSF studies is the
invasive nature of the collection procedure, making it in-
appropriate for general screening of presumably healthy
individuals or all patients with neuropathologies.
PEPTIDOMICS OF SALIVA
Saliva is a multifunctional body fluid secreted by the
major salivary glands (parotid, submandibular, and
sublingual glands) and other glands distributed in the
oral cavity. It lubricates the oral cavity and participates
in digestion and preventing infections (25 ). Saliva con-
tains bacteria, cellular debris, crevicular fluid, and se-
rum components, and as an alternative sample for
noninvasive collection, saliva has promising, unique
features (26 ). Proteolytic degradation occurs as soon as
proteins enter the oral cavity and continues after a sa-
liva sample is collected. This process leads to great vari-
ation in the peptide profile and thus limits the repro-
ducibility of peptidomic analyses. Other preanalytical
variables, such as sex, age, diet, and circadian rhythms,
can also play important roles in the peptide composi-
tion of saliva (27 ). Despite these shortcomings, pep-
tidomic analyses of saliva have been used to assess a
variety of pathologies, including Sjögren syndrome, xe-
rostomia, and diabetes (28 ). Studies of oral cancer
Table 2. Characteristics of body fluids
for peptidomics.
Body fluid
Invasiveness of
sample-collection
procedure
Relative
peptide
stability
Serum Minimally invasive Unstable
Plasma Minimally invasive Unstable
Urine Noninvasive Stable
CSF Invasive Not known
Saliva Noninvasive Unstable
Ascitic fluid Invasive Unstable
Pleural fluid Invasive Unstable
Tears Noninvasive Not known
Reviews
4Clinical Chemistry 60:7 (2014)
(29, 30 ) have identified a number of peptides that are
overproduced in patients with squamous cell carci-
noma. Hu et al. (29 ) reported a combination of 5 pro-
teins that could be used to detect oral cancer with 90%
diagnostic sensitivity and 83% specificity.
PEPTIDOMICS OF TEARS
Tears are a complex extracellular fluid that can be as-
sessed noninvasively. As with plasma, the concentra-
tions of proteins and peptides in tears span several
orders of magnitude (31 ), with low interday but re-
markably pronounced interindividual variation (32 ).
de Souza et al. (33 ) identified 491 proteins in tears, and
a more recent study yielded 1543 proteins (31 ). The
most comprehensive peptidomics study of tears to date
characterized 30 endogenous peptides, most of which
were derived from proline-rich protein 4, a protein of
unknown function produced at high concentrations in
lacrimal acinar cells (34 ). As a body fluid, tears have a
composition that reflects the pathophysiological state
of the underlying tissues and organs and has proved
useful for assessing both ocular and systemic patholo-
gies, such as dry eye, meibomian gland dysfunction,
and Sjögren syndrome (35, 36 ).
Urine as a Source of Biomarkers
Urine is formed in the kidney by ultrafiltration of the
plasma for the elimination of metabolic waste prod-
ucts. Because urine is stored in the bladder for several
hours before elimination, proteolytic degradation by
endogenous proteases has largely been considered to
have been completed by the time of voiding (37 ). Re-
cent findings, however, have demonstrated the pres-
ence of a wide spectrum of proteases in urine (38 ) that
might generate various combinations of different en-
dogenous peptides, depending on how long the urine
resides in the bladder. Proteolytic degradation of a
urine sample can be minimized if second morning
urine samples are used, because the time between the
first and second voiding can be easily monitored and
standardized. Despite such variation, urine is regarded
as stable body fluid, especially compared with blood, in
which proteases are known to be activated during and
after blood drawing and thereby generate a consider-
able number of breakdown products that may skew
proteomic and peptidomic approaches (37, 39 ).
Urine has long been an attractive body fluid for
study, not only in biomarker research but also in clin-
ical diagnosis, mainly because it can be obtained
noninvasively in large amounts. Currently, several
routinely analyzed biomolecules serve as highly dis-
criminating markers. For example, measurements of
urine catecholamines and their metabolites help in as-
sessing pheochromocytoma and in evaluating albu-
minuria as an index of glomerular function.
Some factors require consideration when plan-
ning experiments with urine. Urine shows high daily
biological variation, reflecting the effects that many en-
dogenous and exogenous factors have on its produc-
tion and composition. Diet, exercise, and water intake
are 3 major factors influencing the quality and the
comparability of samples. Urine contains components
released not only from the kidneys and the bladder but
also from many other organs, and biological processes
can have profound impacts on its fluctuating content
(40 ). Even cardiovascular, autoimmune, and infec-
tious diseases affect the presence and concentrations of
some protein molecules (41 ).
URINARY PEPTIDOME
Despite containing very small amounts of proteins,
urine samples from healthy and diseased individuals
are attractive for exploring proteomic disease. In 1997,
Heine et al. (42 ) reported the presence of 13 proteins in
human urine. Since then, many other investigators
have assessed the human uroproteome and have drawn
different, methodology-dependent conclusions. The
initial approaches with 2-dimensional electrophoresis
identified 1400 spots (43 ), a number that increased
when liquid chromatography was introduced. Adachi
et al. (44 ) stated that urine from healthy donors con-
tains at least 1543 different proteins, mostly extracellu-
lar and membrane bound. This finding led the authors
to suggest the possibility of specific transport pathways
for lysosomal and plasma membrane proteins for
reaching the urine. The lower protein content of urine
compared with plasma reduces the possibility for high-
abundance proteins to mask potential biomarkers.
Studies have also demonstrated urine to be highly en-
riched for small peptides (45 ); healthy individuals and
patients with Fanconi syndrome contain a ⬎100-fold
enrichment of molecules ⬍10 kDa, compared with
higher molecular weight polypeptides, perhaps be-
cause the former pass freely through the glomerulus
(46 ). On the other hand, low-abundance and low-mass
peptides can become bound to large carrier proteins
that act as harvesters in the circulation (47 ). The major
constituents of the urinary peptidome appear to be col-
lagen fragments, especially from the collagen
␣
1 chain,
which probably reflect the physiological turnover of
tissue extracellular matrix (39 ).
Technological Aspects of Peptidomics
The discovery of novel biomarkers depends not only
on the concentration of the biomarker candidate in the
sample and the complexity of the matrix but also on
the analytical sensitivity of the detection method and
Peptidomics in Cancer Reviews
Clinical Chemistry 60:7 (2014) 5
the sample-preparation steps. The design of study
strategies and analyses of bioinformatics data is crucial
for reproducible and unbiased results (48 ). Any pep-
tidomic analysis requires a robust and comprehensive
procedure.
SAMPLE PREPARATION
To enrich the low molecular weight components in a
sample for peptidomics analyses requires sample-
preparation steps different from those required for
proteomic analyses (17, 49 ). The preanalytical phase is
the most challenging. A wide range of variables, both
exogenous and endogenous, can affect the results
(50, 51 ). The considerable stability of the proteome’s
composition and concentrations in urine allows sam-
ples to be stored for6hatroom temperature with little
change and for years at ⫺20 °C (37, 52 ). Fiedler et al.
investigated the influence of many variables on final
peptidomics results (53 ). Significant differences were
observed not only between first and second morning
urine samples but also between first-stream and mid-
stream urine samples. Bacteriuria and hematuria had a
great effect, even at low concentrations, on the peptide
profile. Freeze–thaw cycles can influence the final re-
sults when assessing exogenous variables; thus, repro-
ducibility is improved with once-frozen urine samples.
To minimize such potentially confounding factors
and preanalytical variations requires that samples be
collected and handled in a standardized manner. The
Human Kidney and Urine Proteome Project (http://
www.hkupp.org) is an international initiative of the
Human Proteome Organization to establish collection
and manipulation procedures for proteomics. In Eu-
rope, the European Kidney and Urine Proteomics or-
ganization (http://www.eurokup.org) promotes inter-
actions between scientists in the field, with the goal of
improving the understanding and assessment of kid-
ney disease through urine proteomics. Both associa-
tions have proposed recommendations for standard-
ized urine-processing steps (with special emphasis on
sample collection, centrifugation, and thawing), which
should minimize biases among studies.
USE OF MASS SPECTROMETRY IN PEPTIDOMICS
Traditionally, hypotheses for biomarker discovery
have been derived from an understanding of disease
biology (54 ). Over the past few decades, however,
many researchers have turned to mass spectrometry to
discover candidate molecules that could serve as bio-
markers. The advantages of mass spectrometry for
identifying and quantifying peptides in complex bio-
logical samples have facilitated the development of
novel biochemical approaches for diagnosis, not only
of cancer but of other diseases as well. Studies of pro-
teins and peptides have used different methodologies.
Two-dimensional gel electrophoresis has been
used extensively, but it is a time-consuming technique
with poor interassay reproducibility, especially at low
molecular weights because it cannot separate and thus
distinguish molecules of ⬍10 kDa. Capillary electro-
phoresis–mass spectrometry yields robust and highly
reproducible analyses of low molecular weight peptides
and is compatible with many volatile buffers and ana-
lytes (19, 22 ); however, the long processing times make
this technique challenging to use for large-scale studies.
One of the most suitable platforms for urine peptide
profiling is SELDI and MALDI followed by mass spec-
trometry identification with a TOF detector. This ap-
proach focuses on peptides in the range of 1–20 kDa.
Immobilization, the key step in the entire process,
reduces sample complexity, but at the expense of a
great loss of information (50 ). Finally, liquid chro-
matography followed by tandem mass spectrometry
(LC-MS/MS) is capable of providing large amounts
of information with high reproducibility. Only cap-
illary electrophoresis and liquid chromatography are
able to interface directly with tandem mass spectrom-
etry instruments for peptidomics studies with the re-
quired depth of analysis, dynamic range, and enhanced
accuracy of quantification (51 ). In addition, analytical
methodologies that increase analytical sensitivity have
been developed. One example is selected reaction mon-
itoring, which uses a nonscanning mode of operation
on an LC-MS/MS instrument (55 ). It increases the de-
tection capability by 2 to 3 orders of magnitude com-
pared with conventional scanning modes.
Mass spectrometry facilitates both biomarker dis-
covery and verification/validation. Mass spectrometers
help in characterizing proteins and peptides and their
modifications. One of the clearest advantages over
other platforms is its capacity to qualitatively screen
and analyze thousands of molecules without previous
knowledge of their existence or relevance to particular
pathophysiological conditions. Quantification of pre-
viously discovered candidates is essential for evaluating
their diagnostic capabilities.
As with proteomics, both absolute and relative
quantification of peptides generally require the use of
stable isotope–labeled molecules as internal standards.
Their use can overcome the problems of matrix effects,
variation in sample preparation, and instrument fluc-
tuations. As outlined elsewhere, the isotopic label
should be introduced into the work flow as early as
possible to increase the number of steps being con-
trolled and thereby decrease imprecision (56 ). This
technique is fairly time-consuming and expensive,
however. Label-free strategies for relative quantifica-
tion are based on comparing signal intensities pro-
duced by identical peptides in different analyses and
rely on the accuracy of the hypothesis that identical
Reviews
6Clinical Chemistry 60:7 (2014)
peptides will behave similarly across different experi-
ments and therefore permit direct comparisons (57 ).
With urine samples, even absolute quantification
is usually uninformative, so analyte concentration is
commonly corrected for creatinine or protein excre-
tion, or it is based on a 24-h urine collection, thus re-
ducing the dietary and exercise effects on variation in
results. In contrast to transcriptomics and proteomics,
no “housekeeping” peptides have been successfully
identified to date (51 ).
DATA PROCESSING AND BIOINFORMATICS
Peptidomics and proteomics require considerable
computing power to obtain statistically significant and
reproducible data. Peptide identification is one of the
most challenging aspects (58 ). Online databases con-
tain peptide sequences for a variety of body fluids and a
myriad of disease conditions, and they serve as univer-
sal platforms for aiding in defining and verifying can-
didate biomarkers (40 ). A substantial proportion of
the human urinary proteome database is derived from
studies that assessed transplantation or renal disease,
whereas the data derived from studies of prostate, re-
nal, and bladder cancers, as well as pheochromocyto-
mas, are relatively few.
The most demanding issue with peptidomics is re-
lated to the nonspecificity of the peptide ends. As
Hölttä et al. have stated (24 ), no restrictions regarding
enzyme-cleavage specificity can be applied during
analyses of bioinformatics data. The consequence is a
huge increase (up to 1000-fold) in the number of se-
quences to consider. This situation contrasts with that
of proteomics, in which protease digestion (usually
with trypsin) ensures specific endings for each peptide
molecule. For this reason, peptidomics suffers from
higher false-positive rates and less accurate results.
Urine Peptidomics for Disease Diagnostics
Most of the literature on urine peptidomics addresses
impairment of kidney function and outcomes of kid-
ney transplantation (41, 59 ). The few studies that have
searched for cancer biomarkers have focused on blad-
der, ovarian, and prostate cancers. Genitourinary ma-
lignancies are responsible for 1 of every 6 cancer deaths
in men and 1 of every 10 in women (1 ).
OVARIAN CANCER
Ovarian cancer is the deadliest gynecologic malig-
nancy. Current diagnostic strategies are based on mea-
suring carbohydrate antigen 125 (CA125) in serum
(60 ) in combination with vaginal ultrasonography.
Measurement of CA125 lacks diagnostic sensitivity and
specificity for early diagnosis, however, and many ef-
forts have focused on finding protein and peptide mol-
ecules that could be useful as diagnostic biomarkers.
Some proteins, such as human epididymal secretory pro-
tein 4 (61 ) and osteopontin (62 ), have shown utility, al-
though none has surpassed CA125. Using serum, one of
the first peptidomics-based studies combined peaks of
unknown identity, presumably representing low molecu-
lar weight polypeptides, and claimed to distinguish be-
tween individuals with no malignancy and patients with
ovarian cancer (stages I–IV) with 100% sensitivity and
95% specificity (63 ). The results described in this report
have now been invalidated because of preanalytical, ana-
lytical, and bioinformatics artifacts (64 ).
PROSTATE CANCER
Prostate cancer, the most prevalent malignancy in
men, ranks second in lethality (1 ). Novel noninvasive
markers with higher diagnostic sensitivity and specific-
ity are needed. Although PSA-derived forms and the
ribonucleic acid marker PCA3 (prostate cancer antigen
3) seem to add some degree of diagnostic specificity,
they have not met expectations (65 ). Other protein
candidates that have been suggested require large-scale
validation. The first comparison of the urine proteome
used 2-dimensional gel electrophoresis followed by
MALDI-TOF mass spectrometry fingerprinting of
voided urine samples after prostatic massage to evalu-
ate age-matched men with benign prostatic hyperpla-
sia (66 ). Calgranulin B/MRP-8 was highlighted for ver-
ification and validation. Subsequent research with
urine samples has yielded additional candidate mole-
cules, including the matrix metalloproteinases (67 )
and engrailed-2 (68 ), although none have yet been val-
idated with large cohorts. Hypothesizing that first-void
urine contains prostatic fluid, Theodorescu et al. (69 )
used a filter with a 20-kDa cutoff followed by capillary
electrophoresis–mass spectrometry analysis and ob-
tained a biomarker panel of 12 urinary peptides based
on the results. They proposed that this peptide panel,
used in combination with age, free PSA, and total PSA,
could improve current diagnosis by increasing the area
under the ROC curve from 0.77 (based on the free-PSA
percentage and patient age) up to 0.82. Nevertheless,
this peptide panel also remains to be validated.
BLADDER CANCER
Bladder cancer is the fifth most common cancer in
Western societies. Current diagnostic strategies are
based on cytoscopy and urine cytology, but these meth-
ods have high interobserver imprecision and low re-
producibility. Given that the bladder is in intimate
contact with urine after its production in the kidney,
this body fluid has been mined heavily for both protein
and peptide biomarkers that might help, not only in
detecting bladder cancer, but also in distinguishing
muscle-invasive from noninvasive malignancy (70, 71 ).
Peptidomics in Cancer Reviews
Clinical Chemistry 60:7 (2014) 7
A large number of peptides with different concentrations
in urine samples from patients with invasive bladder can-
cer, compared with patients with noninvasive cancer and
with controls, have been found, but most of these peptides
appear to be fragments of abundant proteins. In fact,
Theodorescu et al. (52 ) proposed a proteomic pattern of
22 polypeptides with high diagnostic sensitivity and spec-
ificity for urothelial cancer and highlighted fibrinopeptide
A as a potential diagnostic biomolecule. Bryan et al. (70 )
identified 8 peptides with significantly different concen-
trations in patients with and without muscle-invasive
urothelial carcinoma. Such peptides were identified as
derived from albumin, fibrinogen, hemoglobin, and
prealbumin—all high-abundance proteins.
OTHER CANCERS
Anatomically distant sites can influence urine compo-
sition. Studies of the urine peptidome have used this
rationale to pursue possible markers of lung cancer
(72 ) and gastrointestinal cancer (73 ). Using SELDI,
Husi et al. (74 ) found that a nonnegligible number of
the candidate proteins belonged to the family of small
calcium-binding proteins, S100, which have been re-
lated to the growth of tumors of the upper gastrointes-
tinal tract. None of the identified candidates fulfilled
the requirements for a single marker, so a protein–
peptide pattern served for screening and prediction of
outcome. A diagnostic sensitivity of up to 98% was
reported, but most of the candidates had also been de-
scribed for other malignancies, compromising the pat-
tern’s specificity.
If one steps back and considers these results as a
whole, one sees that most of the peptide panels have
not been validated properly. This lack of validation
studies represents one of the major shortcomings of
peptidomics for reaching the clinical setting and places
the usefulness of peptidomics for cancer diagnostics
under a critical eye.
Translation to the Clinic
Despite intensive efforts, no molecule described in any
proteomics or peptidomics study has entered the clinic.
For retrospective and prospective validation studies
of candidate molecules and to avoid artifacts and
methodology-related false-positive results, other meth-
odologies (e.g., immunoassays) are preferred (3, 75 ).
Sometimes initial studies based on small populations
show a statistical significance that, because of bias in
patient selection or other confounders, becomes lost in
subsequent studies. The lessons from this experience
could help in improving the planning of future strate-
gies. Large-scale population studies are rare and carry a
large financial burden. Finally, reaching statistical sig-
nificance is not sufficient for candidate biomarkers. As
with novel drugs, biomarkers have to show some clin-
ical improvement over those currently in use; other-
wise, they will not be adopted.
Given the complexity of any biological process, a
single biomarker has been widely viewed to be unlikely
to discriminate a pathologic process with sufficient
sensitivity and specificity. Therefore, the incorporation
of combinations of multiple, independent biomarkers
into a diagnostic or predictive panel may be more likely
to be useful. Nevertheless, each of the individual bio-
markers used in any panel must be independently ver-
ified and validated to ensure clinical utility. This re-
quirement makes the design of large-scale validation
studies even more difficult.
Recently, a new perspective that transcends classic
proteomics and peptidomics suggests that the study of
individual or global protease activities might also yield
indicators or predictors of disease (76 –78 ). This new
approach has been termed “functional peptidomics.” It
relies on the fact that tumor progression and invasive-
ness may lead to the differential production and secre-
tion of exoproteases; thus, the study of their functions
might not only reflect the true biological/pathologic
state of an organism but also overcome reproducibility
problems related to preanalytical variables. This ap-
proach is still in its beginning stages, however, and con-
clusions about its applicability cannot yet be drawn.
Future Challenges
The impressive growth in high-throughput biology has
dominated science during the last decade, mainly ow-
ing to the leap in the development of new technologies.
Substantial efforts in proteomics have focused on the
discovery and validation of sensitive and specific diag-
nostic biomarkers for many human pathologies. Deep
biochemical and pathophysiological knowledge is crit-
ical for solving clinical questions, and every step in the
procedure must be planned and executed meticu-
lously. Standardized handling procedures are expected
to aid tremendously in the generation of clinically use-
ful and reproducible data.
Peptidomics is a relatively new field, and few studies
of explorations and characterization of the peptidome
have yet been published. There is no strong evidence that
peptidomics will yield better results than proteomics, but
biological and chemical reasoning supports work in that
direction. Proteomics is undoubtedly the dominant tech-
nology in the postgenomics era, and peptidomics repre-
sents a largely unexplored step forward.
Author Contributions: All authors confirmed they have contributed to
the intellectual content of this paper and have met the following 3
Reviews
8Clinical Chemistry 60:7 (2014)
requirements: (a) significant contributions to the conception and design,
acquisition of data, or analysis and interpretation of data; (b) drafting
or revising the article for intellectual content; and (c) final approval of
the published article.
Authors’ Disclosures or Potential Conflicts of Interest: Upon man-
uscript submission, all authors completed the author disclosure form.
Disclosures and/or potential conflicts of interest:
Employment or Leadership: E.P. Diamandis, Clinical Chemistry,
AACC.
Consultant or Advisory Role: None declared.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: None declared.
Expert Testimony: None declared.
Patents: None declared.
References
1. Siegel R, Naishadham D, Jemal A. Cancer statis-
tics, 2013. CA Cancer J Clin 2013;63:11–30.
2. Lowe KA, Chia VM, Taylor A, O’Malley C, Kelsh
M, Mohamed M, et al. An international assess-
ment of ovarian cancer incidence and mortality.
Gynecol Oncol 2013;130:107–14.
3. Pavlou MP, Diamandis EP, Blasutig IM. The long
journey of cancer biomarkers from the bench to
the clinic. Clin Chem 2013;59:147–57.
4. Hoffman RM. Clinical practice. Screening for pros-
tate cancer. N Engl J Med 2011;365:2013–9.
5. Duffy MJ, Evoy D, McDermott EW. CA 15–3: uses
and limitation as a biomarker for breast cancer.
Clin Chim Acta 2010;411:1869–74.
6. Mischak H, Apweiler R, Banks RE, Conaway M,
Coon J, Dominiczak A, et al. Clinical proteomics:
a need to define the field and to begin to set
adequate standards. Proteomics Clin Appl 2007;
1:148–56.
7. Schulte I, Tammen H, Selle H, Schulz-Knappe P.
Peptides in body fluids and tissues as markers of
disease. Expert Rev Mol Diagn 2005;5:145–57.
8. Candiano G, Musante L, Bruschi M, Petretto A,
Santucci L, Del Boccio P, et al. Repetitive frag-
mentation products of albumin and
␣
1-
antitrypsin in glomerular diseases associated with
nephrotic syndrome. J Am Soc Nephrol 2006;17:
3139–48.
9. Liotta LA, Ferrari M, Petricoin E. Clinical
proteomics: written in blood. Nature 2003;425:
905.
10. Petricoin EF, Belluco C, Araujo RP, Liotta LA. The
blood peptidome: a higher dimension of informa-
tion content for cancer biomarker discovery. Nat
Rev Cancer 2006;6:961–7.
11. Doucet A, Butler GS, Rodriguez D, Prudova A,
Overall CM. Metadegradomics: toward in vivo
quantitative degradomics of proteolytic post-
translational modifications of the cancer pro-
teome. Mol Cell Proteomics 2008;7:1925–51.
12. Villanueva J, Shaffer DR, Philip J, Chaparro CA,
Erdjument-Bromage H, Olshen AB, et al. Differ-
ential exoprotease activities confer tumor-specific
serum peptidome patterns. J Clin Invest 2006;
116:271–84.
13. Villanueva J, Martorella AJ, Lawlor K, Philip J,
Fleisher M, Robbins RJ, Tempst P. Serum pep-
tidome patterns that distinguish metastatic thy-
roid carcinoma from cancer-free controls are un-
biased by gender and age. Mol Cell Proteomics
2006;5:1840–52.
14. Zougman A, Pilch B, Podtelejnikov A, Kiehntopf
M, Schnabel C, Kumar C, Mann M. Integrated
analysis of the cerebrospinal fluid peptidome and
proteome. J Proteome Res 2008;7:386–99.
15. Norden AG, Rodriguez-Cutillas P, Unwin RJ. Clin-
ical urinary peptidomics: learning to walk before
we can run. Clin Chem 2007;53:375–6.
16. Worsfold M, Powell DE, Jones TJ, Davie MW.
Assessment of urinary bone markers for monitor-
ing treatment of osteoporosis. Clin Chem 2004;
50:2263–70.
17. Tirumalai RS, Chan KC, Prieto DA, Issaq HJ, Con-
rads TP, Veenstra TD. Characterization of the low
molecular weight human serum proteome. Mol
Cell Proteomics 2003;2:1096–103.
18. Echan LA, Tang HY, Ali-Khan N, Lee K, Speicher
DW. Depletion of multiple high-abundance pro-
teins improves protein profiling capacities of hu-
man serum and plasma. Proteomics 2005;5:
3292–303.
19. Kolch W, Neususs C, Pelzing M, Mischak H. Cap-
illary electrophoresis-mass spectrometry as a
powerful tool in clinical diagnosis and biomarker
discovery. Mass Spectrom Rev 2005;24:959–77.
20. Shen Y, Tolic N, Liu T, Zhao R, Petritis BO,
Gritsenko MA, et al. Blood peptidome-
degradome profile of breast cancer. PLoS One
2010;5:e13133.
21. Craft GE, Chen A, Nairn AC. Recent advances in
quantitative neuroproteomics. Methods 2013;61:
186–218.
22. Schiffer E, Mischak H, Novak J. High resolution
proteome/peptidome analysis of body fluids by
capillary electrophoresis coupled with MS. Pro-
teomics 2006;6:5615–27.
23. Wijte D, McDonnell LA, Balog CI, Bossers K,
Deelder AM, Swaab DF, et al. A novel peptidom-
ics approach to detect markers of Alzheimer’s
disease in cerebrospinal fluid. Methods 2012;56:
500–7.
24. Hölttä M, Zetterberg H, Mirgorodskaya E,
Mattsson N, Blennow K, Gobom J. Peptidome
analysis of cerebrospinal fluid by LC-MALDI MS.
PLoS One 2012;7:e42555.
25. Helmerhorst EJ, Oppenheim FG. Saliva: a dynamic
proteome. J Dent Res 2007;86:680–93.
26. Amado F, Lobo MJ, Domingues P, Duarte JA,
Vitorino R. Salivary peptidomics. Expert Rev Pro-
teomics 2010;7:709–21.
27. Cabras T, Pisano E, Boi R, Olianas A, Manconi B,
Inzitari R, et al. Age-dependent modifications of
the human salivary secretory protein complex. J
Proteome Res 2009;8:4126–34.
28. Lee YH, Wong DT. Saliva: an emerging biofluid
for early detection of diseases. Am J Dent 2009;
22:241–8.
29. Hu S, Arellano M, Boontheung P, Wang J, Zhou H,
Jiang J, et al. Salivary proteomics for oral cancer
biomarker discovery. Clin Cancer Res 2008;14:
6246–52.
30. Xie H, Onsongo G, Popko J, de Jong EP, Cao J,
Carlis JV, et al. Proteomics analysis of cells in
whole saliva from oral cancer patients via value-
added three-dimensional peptide fractionation
and tandem mass spectrometry. Mol Cell Pro-
teomics 2008;7:486–98.
31. Zhou L, Zhao SZ, Koh SK, Chen L, Vaz C, Tanavde
V, et al. In-depth analysis of the human tear
proteome. J Proteomics 2012;75:3877–85.
32. Gonzalez N, Iloro I, Duran JA, Elortza F, Suarez T.
Evaluation of inter-day and inter-individual vari-
ability of tear peptide/protein profiles by MALDI-
TOF MS analyses. Mol Vis 2012;18:1572–82.
33. de Souza GA, Godoy LM, Mann M. Identification
of 491 proteins in the tear fluid proteome reveals
a large number of proteases and protease inhib-
itors. Genome Biol 2006;7:R72.
34. Hayakawa E, Landuyt B, Baggerman G, Cuyvers
R, Lavigne R, Luyten W, Schoofs L. Peptidomic
analysis of human reflex tear fluid. Peptides
2012;42C:63–9.
35. Soria J, Dura´ n JA, Etxebarria J, Merayo J,
Gonza´ lez N, Reigada R, et al. Tear proteome and
protein network analyses reveal a novel penta-
marker panel for tear film characterization in dry
eye and meibomian gland dysfunction. J Pro-
teomics 2013;78:94–112.
36. Tomosugi N, Kitagawa K, Takahashi N, Sugai S,
Ishikawa I. Diagnostic potential of tear proteomic
patterns in Sjögren’s syndrome. J Proteome Res
2005;4:820–5.
37. Schaub S, Wilkins J, Weiler T, Sangster K, Rush D,
Nickerson P. Urine protein profiling with surface-
enhanced laser-desorption/ionization time-of-
flight mass spectrometry. Kidney Int 2004;65:
323–32.
38. Marimuthu A, O’Meally RN, Chaerkady R, Sub-
bannayya Y, Nanjappa V, Kumar P, et al. A
comprehensive map of the human urinary pro-
teome. J Proteome Res 2011;10:2734–43.
39. Good DM, Zurbig P, Argiles A, Bauer HW, Beh-
rens G, Coon JJ, et al. Naturally occurring human
urinary peptides for use in diagnosis of chronic
kidney disease. Mol Cell Proteomics 2010;9:
2424–37.
40. Siwy J, Mullen W, Golovko I, Franke J, Zurbig P.
Human urinary peptide database for multiple dis-
ease biomarker discovery. Proteomics Clin Appl
2011;5:367–74.
41. Albalat A, Mischak H, Mullen W. Clinical appli-
cation of urinary proteomics/peptidomics. Expert
Rev Proteomics 2011;8:615–29.
42. Heine G, Raida M, Forssmann WG. Mapping of
peptides and protein fragments in human urine
using liquid chromatography-mass spectrometry.
J Chromatogr A 1997;776:117–24.
43. Pieper R, Gatlin CL, McGrath AM, Makusky AJ,
Mondal M, Seonarain M, et al. Characterization
of the human urinary proteome: a method for
high-resolution display of urinary proteins on
two-dimensional electrophoresis gels with a yield
of nearly 1400 distinct protein spots. Proteomics
2004;4:1159–74.
Peptidomics in Cancer Reviews
Clinical Chemistry 60:7 (2014) 9
44. Adachi J, Kumar C, Zhang Y, Olsen JV, Mann M.
The human urinary proteome contains more than
1500 proteins, including a large proportion of
membrane proteins. Genome Biol 2006;7:R80.
45. Norden AG, Sharratt P, Cutillas PR, Cramer R,
Gardner SC, Unwin RJ. Quantitative amino acid
and proteomic analysis: very low excretion of
polypeptides ⬎750 Da in normal urine. Kidney Int
2004;66:1994–2003.
46. Konvalinka A, Scholey JW, Diamandis EP. Search-
ing for new biomarkers of renal diseases through
proteomics. Clin Chem 2012;58:353–65.
47. Geho DH, Liotta LA, Petricoin EF, Zhao W, Araujo
RP. The amplified peptidome: the new treasure
chest of candidate biomarkers. Curr Opin Chem
Biol 2006;10:50–5.
48. Diamandis EP. Cancer biomarkers: Can we turn
recent failures into success? J Natl Cancer Inst
2010;102:1462–7.
49. Finoulst I, Pinkse M, Van Dongen W, Verhaert P.
Sample preparation techniques for the untar-
geted LC-MS-based discovery of peptides in com-
plex biological matrices. J Biomed Biotechnol
2011;2011:245291.
50. Diamandis EP. Mass spectrometry as a diagnostic
and a cancer biomarker discovery tool: opportu-
nities and potential limitations. Mol Cell Pro-
teomics 2004;3:367–78.
51. Ling XB, Mellins ED, Sylvester KG, Cohen HJ.
Urine peptidomics for clinical biomarker discov-
ery. Adv Clin Chem 2010;51:181–213.
52. Theodorescu D, Wittke S, Ross MM, Walden M,
Conaway M, Just I, et al. Discovery and validation
of new protein biomarkers for urothelial cancer: a
prospective analysis. Lancet Oncol 2006;7:230–
40.
53. Fiedler GM, Baumann S, Leichtle A, Oltmann A,
Kase J, Thiery J, Ceglarek U. Standardized pep-
tidome profiling of human urine by magnetic
bead separation and matrix-assisted laser desorp-
tion/ionization time-of-flight mass spectrometry.
Clin Chem 2007;53:421–8.
54. Pisitkun T, Johnstone R, Knepper MA. Discovery
of urinary biomarkers. Mol Cell Proteomics 2006;
5:1760–71.
55. Picotti P, Aebersold R. Selected reaction
monitoring-based proteomics: Workflows, poten-
tial, pitfalls and future directions. Nat Methods
2012;9:555–66.
56. Ong SE, Mann M. Mass spectrometry-based pro-
teomics turns quantitative. Nat Chem Biol 2005;
1:252–62.
57. Neilson KA, Ali NA, Muralidharan S, Mirzaei M,
Mariani M, Assadourian G, et al. Less label, more
free: approaches in label-free quantitative mass
spectrometry. Proteomics 2011;11:535–53.
58. Menschaert G, Vandekerckhove TT, Baggerman
G, Schoofs L, Luyten W, Van Criekinge W. Pep-
tidomics coming of age: a review of contributions
from a bioinformatics angle. J Proteome Res
2010;9:2051–61.
59. Ling XB, Sigdel TK, Lau K, Ying L, Lau I, Schilling
J, Sarwal MM. Integrative urinary peptidomics in
renal transplantation identifies biomarkers for
acute rejection. J Am Soc Nephrol 2010;21:646–
53.
60. Buys SS, Partridge E, Black A, Johnson CC, Lam-
erato L, Isaacs C, et al. Effect of screening on
ovarian cancer mortality: the Prostate, Lung,
Colorectal and Ovarian (PLCO) Cancer Screening
Randomized Controlled Trial. JAMA 2011;305:
2295–303.
61. Hellstrom I, Raycraft J, Hayden-Ledbetter M, Led-
better JA, Schummer M, McIntosh M, et al. The
HE4 (WFDC2) protein is a biomarker for ovarian
carcinoma. Cancer Res 2003;63:3695–700.
62. Kim JH, Skates SJ, Uede T, Wong KK, Schorge JO,
Feltmate CM, et al. Osteopontin as a potential
diagnostic biomarker for ovarian cancer. JAMA
2002;287:1671–9.
63. Petricoin EF, Ardekani AM, Hitt BA, Levine PJ,
Fusaro VA, Steinberg SM, et al. Use of proteomic
patterns in serum to identify ovarian cancer. Lan-
cet 2002;359:572–7.
64. Diamandis EP. Point: proteomic patterns in bio-
logical fluids: Do they represent the future of
cancer diagnostics? Clin Chem 2003;49:1272–5.
65. Stephan C, Jung K, Semjonow A, Schulze-Forster
K, Cammann H, Hu X, et al. Comparative assess-
ment of urinary prostate cancer antigen 3 and
TMPRSS2:ERG gene fusion with the serum
[⫺2]proprostate-specific antigen–based prostate
health index for detection of prostate cancer. Clin
Chem 2013;59:280–8.
66. Rehman I, Azzouzi AR, Catto JW, Allen S, Cross
SS, Feeley K, et al. Proteomic analysis of voided
urine after prostatic massage from patients with
prostate cancer: a pilot study. Urology 2004;64:
1238–43.
67. Roy R, Louis G, Loughlin KR, Wiederschain D,
Kilroy SM, Lamb CC, et al. Tumor-specific urinary
matrix metalloproteinase fingerprinting: identifi-
cation of high molecular weight urinary matrix
metalloproteinase species. Clin Cancer Res 2008;
14:6610–7.
68. Morgan R, Boxall A, Bhatt A, Bailey M, Hindley R,
Langley S, et al. Engrailed-2 (EN2): a tumor spe-
cific urinary biomarker for the early diagnosis of
prostate cancer. Clin Cancer Res 2011;17:
1090–8.
69. Theodorescu D, Schiffer E, Bauer HW, Douwes F,
Eichhorn F, Polley R, et al. Discovery and valida-
tion of urinary biomarkers for prostate cancer.
Proteomics Clin Appl 2008;2:556–70.
70. Bryan RT, Wei W, Shimwell NJ, Collins SI, Hussain
SA, Billingham LJ, et al. Assessment of high-
throughput high-resolution MALDI-TOF-MS of
urinary peptides for the detection of muscle-
invasive bladder cancer. Proteomics Clin Appl
2011;5:493–503.
71. Munro NP, Cairns DA, Clarke P, Rogers M, Stan-
ley AJ, Barrett JH, et al. Urinary biomarker pro-
filing in transitional cell carcinoma. Int J Cancer
2006;119:2642–50.
72. Tantipaiboonwong P, Sinchaikul S, Sriyam S,
Phutrakul S, Chen ST. Different techniques for
urinary protein analysis of normal and lung can-
cer patients. Proteomics 2005;5:1140–9.
73. Hong CS, Cui J, Ni Z, Su Y, Puett D, Li F, Xu Y. A
computational method for prediction of excretory
proteins and application to identification of gas-
tric cancer markers in urine. PLoS One 2011;6:
e16875.
74. Husi H, Stephens N, Cronshaw A, MacDonald A,
Gallagher I, Greig C, et al. Proteomic analysis of
urinary upper gastrointestinal cancer markers.
Proteomics Clin Appl 2011;5:289–99.
75. Makawita S, Diamandis EP. The bottleneck in the
cancer biomarker pipeline and protein quantifica-
tion through mass spectrometry– based approaches:
current strategies for candidate verification. Clin
Chem 2010;56:212–22.
76. Villanueva J, Philip J, Chaparro CA, Li Y, Toledo-
Crow R, DeNoyer L, et al. Correcting common
errors in identifying cancer-specific serum peptide
signatures. J Proteome Res 2005;4:1060–72.
77. Villanueva J, Nazarian A, Lawlor K, Yi SS, Robbins
RJ, Tempst P. A sequence-specific exopeptidase
activity test (SSEAT) for “functional” biomarker
discovery. Mol Cell Proteomics 2008;7:509–18.
78. Findeisen P, Neumaier M. Functional protease
profiling for diagnosis of malignant disease. Pro-
teomics Clin Appl 2012;6:60–78.
Reviews
10 Clinical Chemistry 60:7 (2014)
