Mediators of Inflammation

Mediators of Inflammation / 2014 / Article
Special Issue

Biomarkers in Rheumatoid Arthritis

View this Special Issue

Review Article | Open Access

Volume 2014 |Article ID 545493 | 24 pages | https://doi.org/10.1155/2014/545493

Cytokines as Biomarkers in Rheumatoid Arthritis

Academic Editor: Jean Sibilia
Received28 Aug 2013
Accepted21 Jan 2014
Published09 Mar 2014

Abstract

RA is a complex disease that develops as a series of events often referred to as disease continuum. RA would benefit from novel biomarker development for diagnosis where new biomarkers are still needed (even if progresses have been made with the inclusion of ACPA into the ACR/EULAR 2010 diagnostic criteria) and for prognostic notably in at risk of evolution patients with autoantibody-positive arthralgia. Risk biomarkers for rapid evolution or cardiovascular complications are also highly desirable. Monitoring biomarkers would be useful in predicting relapse. Finally, predictive biomarkers for therapy outcome would allow tailoring therapy to the individual. Increasing numbers of cytokines have been involved in RA pathology. Many have the potential as biomarkers in RA especially as their clinical utility is already established in other diseases and could be easily transferable to rheumatology. We will review the current knowledge’s relation to cytokine used as biomarker in RA. However, given the complexity and heterogeneous nature of RA, it is unlikely that a single cytokine may provide sufficient discrimination; therefore multiple biomarker signatures may represent more realistic approach for the future of personalised medicine in RA.

1. Biomarker Research

1.1. General Features of Biomarkers

Biomarkers are defined as anatomical, physiological, biochemical, molecular parameters or imaging features that can be used to refine diagnosis, measure the progress of diseases, or predict and monitor the effects of treatment. They can also be associated with the severity of specific disease states.

Biomarkers can be detected and measured by a variety of methods including physical examination, laboratory assays, and medical imaging. Some biomarkers arepresent in particular groups of patients but not others, and as a result they are defined as qualitative biomarkers in contrast to quantitative biomarkers that are present at various degrees/levels in all patients. The accessibility of a biological biomarker, which is defined by the methods that are used to access the biomaterial necessary to measure it, is an important factor in relation to its adoption in clinical practice. If a biomarker can be obtained in a minimally invasive manner (typically from blood, saliva, or urine) or use tissue imaging as opposed to tissue sampling (biopsy), it will obviously be more attractive.

In the context of rheumatic diseases, typical biological biomarkers could encompass genetic markers, products of gene expression, autoantibodies, cytokine/growth factors, acute phase reactants, tissue abnormalities visualized by immunohistochemistry in synovial biopsy, a product of tissue degradation, or a cell subset that can be phenotyped and enumerated. The sources of these biomarkers could be the serum/plasma, urine, synovial fluid, tissue biopsy, or cells from blood, fluid, lymph node, or tissue. In contrast, a clinical biomarker (i.e., clinical surrogate) would constitute a physical variable (sign or symptom), a clinical judgment, or an outcome measurement that emerges as a sequel of the underlying disease process. In rheumatology, this variable may be not only joint counts, global assessment, pain score, duration of morning stiffness, and other clinical variables but also composite indices or functional, radiographic scores.

1.2. Specificity and Sensitivity

Sensitivity and specificity are statistical measures of the performance of biomarker using a binary classification test. This measures use used a categorical classification of patients with respect to true and false positive/negative results.

Sensitivity relates to the biomarker’s ability to identify positive results. It measures the proportion of individuals which are correctly identified by the biomarker. Sensitivity is different from positive predictive value (PPV, also called precision), representing the proportion of actual positives in the population being tested.

On the other hand, specificity relates to the ability of the test to identify negative results. It measures the proportion of people without the biomarker that are correctly not assigned to the condition. Sensitivity may be affected in case of a number of indeterminate test results. It is possible to exclude these cases from analysis or, alternatively, to treat them as false negatives (which gives the worst-case value for sensitivity but also underestimates it), but such exclusions should be stated when quoting sensitivity.

An optimal biomarker would aim to achieve 100% sensitivity (i.e., predict all people with the condition) and 100% specificity (i.e., not predict anyone from the control group). For any biomarker, there is usually a trade-off between the measures and their impact, setting acceptable limits and allowing detection of false positive (lowering specificity), but limiting false negative (increasing sensitivity).

Taking the example of anticitrullinated peptide antibodies (ACPA) in RA, sensitivity is usually reported around 68% and specificity is reported at 95% [1]. However, sensitivity is highly dependent on the group of individuals tested and values observed in established diseases that do not reflect the general RA patients’ population or early disease. Indeed, in patient with recent onset of symptoms, studies have shown that sensitivity is much lower (ranging from 35% to 50%) even if specificity remains closer to 95% [2].

Multivariate markers are as follows: the concept of biomarker algorithm or multivariate biomarkers has recently been developed based on the observation that a single biomarker is often insufficient to predict the outcome of interest, when a combination of biomarkers is better at achieving the prediction. It is usually observed that multivariate biomarkers perform better in replicate studies than univariate biomarkers.

1.3. Need for Biomarkers in Rheumatoid Arthritis (RA)

RA is a complex disease that develops as a series of events often referred to as disease continuum. Research into the preclinical and early phases of RA recently reviewed these events and categorised groups of individuals based on risk factors [3]. According to this new terminology, healthy individuals without RA are described as having potentially two main types of risks: (i) a genetic risk, for example, if they carry the shared epitope allele and (ii) an environmental risk if they smoke. They, however, do not present any laboratory evidence of symptoms or any signs of inflammatory arthritis. The first phase of RA disease progression would then be a state in which individuals develop features of systemic autoimmunity that can be measured by laboratory investigations and are known to be associated with RA (such as ACPA) [3] and more recently with carbamylated protein [4, 5]. These individuals still do not present any symptoms or signs of inflammatory arthritis. A further stage is then defined by the appearance of symptoms (such as arthralgia/morning stiffness), still with no evidence of any clinical synovitis. These individuals can come from both the genetic and environmental risk groups, from the systemic immunity group, or from the general healthy population. Finally, the last progression stage is represented by the development of clinically apparent inflammatory arthritis that may not yet fulfil the criteria for RA diagnosis [6], and hence it is being termed undifferentiated arthritis but is likely to evolve towards RA.

There are many situations in RA, which would benefit from biomarker discovery, considering that biomarkers may be broadly classified as diagnostic (detected when disease is present), prognostic (associated with disease outcome), or predictive markers (associated with drug response). Diagnosis is obviously an area where new biomarkers are still essential as RA is a condition where diagnosis relies on signs and symptoms even if recent progress has been made with the inclusion of ACPA to the recently updated criteria [6]. However, in RA diagnosis, the performance of biomarkers may greatly depend on the duration of symptoms at the time of test, the current level of inflammation, and the amount of destructive processes already undergone, as well as on the type of tissue tested. Prognostic biomarkers which predict the future course of the disease and provide information regarding the outcome irrespective of therapy would be very important in foreseeing the evolution of undifferentiated arthritis towards RA or with respect to the severity of RA which can be quite variable. Prognostic biomarker validation is therefore relatively straightforward, as it is associated with the disease and the patient and can be established (at least in theory) using data from a series of patients treated with standard treatment. The discovery of specific biomarkers for poor prognosis would, for example, enable early intervention and intensive treatment. Risk biomarkers for predicting rapid evolution or cardiovascular complications, for example, remain highly desirable. Monitoring biomarkers would be useful in predicting relapse and candidates are available using flow cytometry based cell subset phenotyping [79]. Predictive biomarkers would separate an RA patients’ population with respect to their outcome in response to a particular event taking place (i.e., particular therapy). They are therefore present/absent prior to the outcome occurring and have obvious applications with the greatest potential to affect clinical practice by targeting drugs to relevant patient subgroups. Biomarkers allowing the selection of an optimal drug for a particular patient (acknowledging that certain subset of patients respond better to certain drug than others) may represent another essential step in patients screening that would notably allow personalised medicine models to be developed, tailoring therapy to the individual, shortening time from onset to effective treatment, improving cost and risk-benefit ratios of drugs, and ultimately achieving high response rate with minimal toxicity [10]; however, in patients with long-standing RA heterogeneity in disease presentation, there remains a major obstacle even when using biomaterial as close to the disease site as synovial tissue [11].

There are several sources of tissue and body fluid that can be considered for biomarker discovery programs in RA. The suitability between the levels of invasiveness and the benefit provided by the biomarkers is however to be considered as well as the level of investigation patients would be likely to accept. Diagnostic biomarkers, considering the prevalence of the disease (1-2%), would need to use biological material which is easily accessible and a method of collection which would not impact on the progression of the disease. Blood and urine therefore appear more suitable compared to synovial tissue or fluid particularly at this early stage of the disease where mostly small joints are involved. Later in the disease continuum, tolerability for more invasive procedure such as fluid aspiration or biopsy collection would provide material reflecting the disease site more closely allowing for individual variability to be taken into account for a personalised medicine approach.

2. Cytokines as Biomarkers

2.1. Cytokine Classification

Cytokines are small proteins which play important roles in cell signalling. They are secreted by a variety of cellular sources acting either on the cell producing them (autocrine) or on the surrounding cells (paracrine). They are classified as proteins and sometimes peptides and can also be glycosylated. Cytokines usually circulate in very small amounts (picomolar 10−12 M) and, nonetheless, their concentration can increase up to 1,000-fold when required. Cytokines have originally been identified in the context of the immune system; however, it has now been shown that they are produced by and influence the behaviour of a variety of nonimmune cells. Cytokines are often referred to as “growth factors” by association with one of their most common effects, the induction of cell proliferation, despite a wide spectrum of roles in survival, apoptosis, differentiation, and functional activation (contribution to the immune response).

Over the years, cytokines have been categorized into various classes, families, or superfamilies. It has been done using either their numerical order of discovery (notably, in the interleukin family, currently up to IL-38) or a given functional activity (e.g., the larger tumour necrosis factor family). In that case, they are further divided between cytokines which enhance cellular immune responses (type 1) as opposed to those which favour antibody responses (type 2). This subclassification is performed using their function (early or late, innate or adaptive, pro- or anti-inflammatory, mitogenic, regulatory, survival functions) or, sometimes, using their primary cell of origin (monokine, lymphokine). More recently, classification has been achieved using structural homologies shared between related molecules. Nevertheless, despite sharing sequence homology and some promiscuity between their receptor systems, cytokines demonstrate specificity in their function and even opposing functions within members of the same family (best illustrated in the TNF superfamily).

Methods of detection for cytokines also vary considerably. Enzyme-linked immunosorbent assays (ELISAs) have long been considered the “gold standard,” but, nowadays, the development of multiplexing technology has allowed biomarker programs to investigate whole cytokine networks as opposed to individual candidates notably enabling large data sets to be generated from small body fluid volumes. Several multiplexing technologies are now available, including the bead-based immunoassay (often referred to as Luminex assay), membrane-based ELISAs, and Mosaic ELISAs, as well as cytometric bead arrays (CBAs). Concerns have been raised related to the sensitivity of some multiplex solid-phase assays [12] as well as interference from heterophilic antibodies [1219]. This is of particular relevance in autoimmune disease where rheumatoid factor (RF), a heterophilic autoantibody directed against the Fc portion of IgG is present notably in RA [12, 2025].

2.2. Variability and Limitations of Cytokine Measurements
2.2.1. Patient Related Variability

There are a number of features and conditions that can influence cytokine production which are related to donor variability in both health and disease. Some of these characteristics are unlikely to change during treatment (genetic/ethnic background, gender, and age); however, others may greatly limit the ability to use cytokines as biomarkers in everyday practice. These factors such as diurnal rhythmicity and sample handling factors (collection methods, storage, and plasma versus serum) may influence the measurement of cytokines and are also likely to change with not only therapy but also stress and cachexia. Such factors are likely to contribute considerably to the disparities seen among similar types of clinical studies [5355].

(1) Age and Gender Effects. Comprehensive analysis of 30 different biomarkers in 400 healthy donors, ranging in age from 40 to 80 years, showed an increase in serum interferon-inducible chemokines (MIG and IP-10), eotaxin, and soluble TNFR-II with advancing age [56]. Multiple studies discussed differences in cytokine production associated with donor age, and several reports have demonstrated that chronic, low-grade inflammation is linked with the aging process [5759]. An age-related increase in IL-6 concentration has been reported in serum, plasma, and supernatants of mononuclear cell cultures obtained from elderly subjects [60, 61]. Some studies demonstrated that plasma levels of tumor necrosis factor (TNF) are elevated in elderly populations [59, 6264]. Conversely, other cytokines regulating T cell functions, such as IL-2, may be decreased with aging. The suppressed production of IL-2 leads to a small clonal expansion of T cells thus decreasing the ability to develop specific immune responses [61]. Modifications of the immune system are globally evaluated as a form of deterioration called immunosenescence. However, ageing is also accompanied by a chronic low-grade inflammation state, showed by a 2 to 4-fold increase in serum levels of inflammatory mediators which act as predictors of mortality independently of preexisting morbidity. This proinflammatory status underlies biological mechanisms responsible for decline in physical function, and inflammatory age-related diseases are initiated or worsened by systemic inflammation [65]. The term “inflammaging” has been coined to explain the underlying changes common to the most age-associated conditions [66, 67].

Longitudinal cytokine production in paediatric and adult patients identified multiple differences in terms of proinflammatory cytokines such as IL-6, IL-8, IL-1alpha, IL-1beta, MCP-1, MIP-1alpha, IL-15, IL-5, IL-17, IL-18, and IP-10 and of anti-inflammatory cytokines such as IL-10, G-CSF, IL-13, IFN-gamma, and IL-4 between the two groups [68]. Altogether, the age of onset in RA patients is to be taken into consideration as it may reflect the cytokine production profile. Men and women also present with gender related differences in the way their immune system responds to challenge [69]. Females demonstrate better B cell-mediated immunity than age-matched males (with higher immunoglobulin levels, stronger antibody responses, and increased resistance to certain infections). Gender also influences T cell immunity, females having greater resistance to induced tolerance, an increased risk to reject grafts, and higher levels of IL-1, IL-4, and IFN-gamma in contrast to men who produce more IL-2, -4, and -13 and whose monocytes secrete more IL-1beta and TNF-alpha [70]. Differences in cytokine production profile have also been suggested to play an important role in the gender bias with regards to the ratio of relapsing remitting and secondary progressive multiple sclerosis [71] as well as susceptibility to urinary infection [72]. Aging has also been associated with alterations of the musculoskeletal system and a decline in sex hormone levels, which have a central role in the regulation of bone turnover. The effect of age combined with gender on cytokines and markers of bone metabolism production showed an increased proportion of T cells producing IFN-gamma and IL-2, IL-4, IL-10, and IL-13 particularly in elderly women after menopause [73].

(2) Circadian Rhythm. Cytokines present a circadian pattern. For example, IFN-gamma, TNF-alpha, IL-1, and IL-12 production exhibits distinct diurnal rhythms that peak in the early morning [74] and are related to the rhythm of plasma cortisol and melatonin [7577]. Taking IL-6 as an example, notably with respect to RA, IL-6 demonstrates important variation in serum or plasma levels in healthy subjects over a day period with a particular biphasic rhythm [78] altogether amounting up to a CV >23%. After correction for analytical variation, a rise in serum IL-6 in the late evening and the early morning has been reported in RA [7882] as well as high variations between and within days not necessarily indicating rhythmicity [54]. Therefore, only IL-6 changes over twice the biological variation (>50% difference) should be considered significant [78]; however, in order to obtain comparable and meaningful results, the time of sample collection should be synchronized, with a morning sample collection time being ideal. This does not affect all cytokines but is not particularly well described for many and should be considered if/when validating a biomarker for clinical use.

(3) Food Intake. Long-term food intake patterns (i.e., obesity or weight loss) have been shown to affect circulating cytokine levels, notably TNF-alpha [83]. Postprandial cytokine levels are also affected by feeding; notably circulating IL-6 levels are increased, while TNF-alpha levels are decreased [8487]. Food supplements (in particular, antioxidants such as glutathione and vitamins E and C) can attenuate the feeding-induced rise in plasma cytokines [88, 89]. Hence, patients should be instructed to maintain normal dietary habits and avoid food supplements prior to sample collection if the cytokine of interest is sensitive to such regulation [90, 91].

(4) Exercise. Physical exercise can affect cytokine levels in the circulation [54, 92]. While plasma cytokines are produced by many cell types, muscle cells are a major source of secreted cytokines during exercise [93, 94]. However, these particular responses are highly specific to the exercise protocol and physiological strain (duration, nature of the exercise, and intensity) [95, 96]. Several studies reported elevation of plasma IL-6 in healthy subjects, which peaked at the end of exercise. The magnitude of the IL-6 response was related to the duration and intensity of the muscle work, the mass of muscle recruited, and the subject’s endurance capacity [78, 9799]. In patients with RA, no changes in serum IL-6 were found after cycling. This could be due to the less strenuous exercise performed by the RA patients because of their widespread joint pain [78, 100]. In contrast, evidence suggests that the prophylactic effect of prolonged, endurance type exercise protocols may be mediated via the induction of an anti-inflammatory environment (increases in circulating levels of IL-1RA and IL-10) [101]; however, how/whether both are linked remains poorly defined. There is nevertheless consensus that exercise training protects against some types of cancers by enhancing antitumour immunity and reducing inflammatory mediators. Altogether, any unconventional strenuous activity prior to blood collection for cytokine measurements should be avoided.

(5) Stress. Stress and emotional problems were also shown to influence cytokines levels; however, studies yielded contradictory data with decrease, increase, and no change in proinflammatory cytokine production being reported [102104]. Nevertheless, lower self-rated health was associated with higher levels of inflammatory cytokines IL-1 and TNF-alpha (controlling for age, education, and physical health) [104].

2.2.2. Preanalytical Related Variability

There are several specific problems posed by sampling conditions (i.e., preanalytical issues) in addition to those described above. Cytokines act either in a paracrine or an autocrine manner as they are released and consumed locally, close to the site where the immune reaction occurs. Therefore, they are rarely detectable in peripheral blood and then only at low levels [105]. Blood may thus only partly reflect pathologies, including RA, and therefore not be the material of choice. The half-life of many cytokines is also measured in minutes; hence, the time lapse between the collection and processing of the samples may be a significant factor limiting the use of cytokines as biomarkers.

Data reproducibility can be affected by normal human variability, which is relatively easy to control in model systems (i.e., in cell culture or even animal models) but is much harder to control in real subjects. Designing and testing the sample collection (i.e., anticoagulants, stabilizing agents) and handling (temperature, elapsed time from collection to initial processing, and endogenous degrading properties of the analyte) and processing protocol/method will represent key elements in the successful development of any biomarkers [106].

(1) Serum or Plasma? In body fluids, cytokines can exist under multiple molecular forms related to posttranslational modifications (i.e., glycosylation), monomers/polymers, precursors, and degradation products or complexed with other proteins [107]. Such molecular forms can behave differently in assays used to determine their levels; therefore, choice of different analytical techniques will be determinant in selecting blood preparation. Serum and plasma are not interchangeable, and the use of one or the other will determine which technique should be used for analyte quantification (see Table 1). Therefore, a lack of consensus exists with respect to the optimal type of specimen to measure cytokines, and the question remains open as to whether plasma or serum should be used. It is important to determine if the method used to collect and prepare the sample may introduce alterations to the cytokine to be tested (i.e., cytokines, either individually or on all proteins in the sample) or whether certain preparation methods are desirable or not for certain cytokines [108].


Serum or plasmaDelays in separation
(whole blood pending processing)
Storage condition (after separation)Sensitivity to freeze-thawing (F/T) cycles

IL-1
(alpha and beta)
(i) Both are used [26]
(ii) Higher heparin plasma concentrations compared to serum [27]
(iii) Higher levels in EDTA plasma than in heparin plasma [28]
(i) Increased levels with delays in processing when kept at RT
(ii) No significant change for up to 4 days of delayed processing in samples from healthy people; however, there is significant decrease in samples from trauma patients [29]
(iii) Significant increase in serum, after delay of 48 h at 4°C, with RA patients, but in plasma there is an increase only if kept at RT [30]
(iv) Prolonged delays before separation result in increased endotoxin-induced cytokine release in contaminated tubes [31, 32]
(i) Storage at 4°C results in an
increase
(ii) Heparin plasma showed time-dependent increases in concentration [31]
No significant change in stability in plasma/serum for up to 6 F/T cycles [26]

IL-2(i) Heparin plasma concentrations are higher than in serum [27]
(ii) Comparable or higher levels in EDTA plasma compared to heparin [28]
No significant change for up to 4 days of delayed processing in samples from healthy people; however, there is a significant decrease when processing samples for trauma patients [29]

IL-4(i) Heparin plasma concentrations are higher than in serum [27]
(ii) Higher levels of IL-4 in EDTA plasma than in heparin [28] and higher concentration in serum than in heparin plasma [33]
(i) No significant change for up to 4 days of delayed processing [29]
(ii) No significant change for serum or EDTA plasma stored before centrifugation at 4°C, RT, and 35°C [30]

IL-5Slightly higher levels in EDTA plasma than in serum [30] (i) No significant change for up to 4 days of delayed processing [29]
(ii) Plasma levels significantly increased if separation delayed by 4 h stored at 4°C. Further increase if stored at RT [30]
(iii) Serum levels increased with delayed processing for 24 h at 4°C or 4 h at RT [30]

IL-6(i) Serum and EDTA plasma samples are comparable while levels in heparin and citrate plasma are lower [34]
(ii) Serum levels are higher than EDTA plasma [30]
(iii) Serum and EDTA, citrate, or heparin plasma gave comparable results [28, 35]
(iv) Heparin plasma levels are higher than those of serum [27] and this anticoagulant is not recommended due to ex vivo Il-6 release prior to assay [31]
(v) Endotoxin contamination (LPS) triggers release in heparin compared to EDTA plasma [32, 36]
(i) Reduced levels when samples are left unseparated for 24 h at 4°C or RT [26] or 4 h at RT [34]
(ii) Significant reduction in stability and recovery with time at RT [26]
(iii) Increased levels with delays in processing when left at RT
(iv) No change in samples stored at 4°C for 24 h before centrifugation [35]
(v) No change when left at 4°C or 20°C for up to 4 days before centrifugation [37]
(vi) No significant change for up to 4 days of delayed processing in samples from healthy people; however, there is a significant decrease when processing samples from trauma patients [29]
(vii) Plasma levels unchanged when stored for up to 3 h at 37°C but afterwards, an increase is observed [31]
(viii) Increased endotoxin-induced cytokine release in contaminated tubes with delays in processing [32]
No change in levels in serum stored at 4°C, −20°C, and −30°C [37] (i) No significant change for up to 6 F/T cycles
(ii) No significanceobserved after 2, 3, and 4 times of repeated F/T cycles [37]
(iii) No significant effect for up to 3 F/T cycles in EDTA plasma and serum but inconsistent stability in heparin plasma [26, 34]

IL-7(i) No significant difference between plasma and serum IL-7 levels [38]
(ii) Serum levels are significantly higher than in plasma [33]
(iii) Heparin plasma concentrations are higher than in serum [27]
(i) 2 to 4 hours of delayed processing decrease IL-7 plasma levels [38]
(ii) With 2 to 4 hours of delayed processing, serum levels are stable [38]
(iii) No significant change for up to 4 days of delayed processing [29]
Stable for up to 3 F/T cycles

IL-8(i) Comparable levels in heparin plasma and in serum [27]
(ii) Higher serum levels than in heparin plasma [33]
(iii) Lower levels in EDTA plasma than in heparin [28]
(iv) LPS induced release in whole blood is up to 100 times higher in heparin versus EDTA plasma [36]
(i) Increased levels with delays in processing if left at RT [29]
(ii) Stable levels if stored at 4°C

IL-9No significant change for up to 4 days of delayed processing [29]

IL-10(i) Higher levels in serum than in plasma [39]
(ii) Lower levels in EDTA plasma than in heparin [28]
(i) Increased levels with delays in processing if left at RT
(ii) The longer the delay, the less stable the levels
(iii) No significant change for up to 4 days of delayed processing in samples from healthy people; however, there was a significant decrease in samples from trauma patients [29]
Storage temperature affects stability: the higher the temperature, the faster the decline [37] No significant decline in levels observed after 2, 3, or 4 times of repeated F/T cycles [37]

IL-12
(p70 & p40)
Heparin and EDTA plasma levels are higher than in serum [27, 28, 30, 33] (i) Levels decrease with delayed processing [29]
(ii) No significant change for up to 4 days of delayed processing [29]
(iii) Increase in serum after 48 h of delayed processing at 4°C and 4 h at RT [30]
(iv) Stable in plasma for over 48 h at 4°C and for up to 48 h at RT [30]

IL-13(i) Heparin plasma levels are higher than those of serum [27]
(ii) Comparable levels in EDTA and heparin plasma [28]
No significant change for up to 4 days of delayed processing [29]

IL-15No significant change for up to 4 days of delayed processing [29]

IL-16Decrease after the 5th F/T cycle [40]

IL-17(i) Lower levels in EDTA plasma than in heparin [28, 33]
(ii) Higher levels in serum than in any plasma (EDTA, citrate, and heparin) [33]
(iii) Higher levels in EDTA plasma than in serum [30]
(i) No significant change for up to 4 days of delayed processing in samples from healthy people; however, there was a significant decrease in samples from trauma patients [29]
(ii) Plasma levels increased if separation delayed by 4 h at 4°C with further increase with time (up to 24 h) [30]

IL-18Similar levels in serum and EDTA plasma [30] No changes in EDTA levels over 48 h at 4°C, and significant increase after 24 h at RT [30]

TNF-alpha(i) Comparable results in serum and EDTA plasma [39]
(ii) Lower levels in sodium citrate plasma
(iii) Higher heparin and EDTA plasma levels than in serum [26, 27, 30]
(iv) LPS induced release of TNF-alpha 20 times higher when in heparin compared to EDTA plasma [36]
(v) Endotoxin induces high release
[32, 41]
Contradictory data:
(i) Reduced levels with delays in processing when kept at 4°C and RT [26, 42]
(ii) Increased levels with delays in processing if left at RT [34, 43]
(iii) No significant change for up to 4 days of delayed processing [29]
(iv) Time-dependent increases in levels with delays at 37°C in heparin plasma [31]
(i) Reduction in samples kept at RT for 20 days
(ii) Relatively stable in samples stored at 4°C [39]
(iii) Stable at −70°Cfor over 9 months [42]
Contradicting data:
(i) Levels increased with successive F/T cycles [26, 34]
(ii) No differences reported in plasma and serum for up to 10 F/T cycles [39]

TGF-beta 1(i) Higher levels in serum than plasma (citrate, EDTA) due to platelet degranulation during the clotting process [30, 4446]
(ii) EDTA plasma is not recommended because of the extreme interindividual variation of PLT activation and concurrent in  vitro GF release [44]
(iii) Sodium citrate can be used but is not as effective or reliable [44]
(iv) CTAD (citrate theophylline dipyridamole adenosine) is recommended as it blocks the in vitro release of growth factors from PLTs
(v) Plasma concentrations should be corrected by simultaneous measurement of markers of platelet degranulation [47]
(i) Increased levels with delay when plasma is left at RT or 37°C [48] due to platelet degranulation and release [45]
(ii) Lower level in serum if left at 4°C than at RT [49]
(iii) Speed of centrifugation affects recovery in plasma (2,500 ×g for 30 min yields lower levels than 1,200 ×g for 10 min) [49]
<5% deviation from baseline value in serum upon successive F/T cycles (for up to 100 F/T cycles) [48]

sCD40-ligand(i) Use of platelet poor/free plasma is recommended as it is [50] higher in serum than in plasma (EDTA, citrate, and heparin) due to clot retraction and sCD40L shedding from the platelet surface [33, 50]
(ii) EDTA anticoagulated plasma samples are not appropriate for sCD40L measurements [51]
(i) Increased levels after 3 h of delay in processing [50]
(ii) Serum levels increase with time in delayed processing [50]
(iii) No significant changes in serum or plasma levels detected after storage at 4°C for up to 48 h
(iv) Significant loss observed in serum and plasma, left at RT [52]
(v) Decreased levels with increasing centrifugation values (200–13 000 g), which gradually deplete plasma of platelets [52]
(i) Loss in serum and plasma kept for over 4 h at RT [40]
(ii) No change while stored at 4°C
(iii) Significant decrease after 24 h at 37°C [40]
(i) Stable for up to 3 F/T cycles [52]
(ii) Increased after 5 or 10 F/T cycles [40]

IFN-gamma(i) Collection in sterile pyrogen free tubes is very essential
(ii) Serum levels are higher than in plasma (EDTA, citrate, and heparin) [33]
(iii) Heparin plasma levels are higher than in serum [27]
(iv) Levels in heparin plasma are higher than in EDTA plasma [33]
(v) Levels in EDTA plasma are higher than in heparin plasma [28]
(i) Significant reduction with time at both 4°C and RT in serum and EDTA tubes [26]
(ii) IFN-gamma decreases if processing is delayed [29]
Stable for up to six F/T cycles [26]

Serum represents the soluble fraction of clotted blood. Serum preparation involves the removal of fibrinogen, platelets, and other circulating proteins. Clotting takes a minimum of 30 minutes but no longer than 60 minutes. Blood should then be centrifuged for 10 minutes and serum should be separated from the clot. Blood cells may get activated during the clot formation and cytokines may be released as a result (such as IL-1, IL-6, and CXCL8) [27, 90, 109, 110]. Rapid sample processing is therefore essential to accurately measure cytokines due to platelet release (i.e., IL-1, IL-6, sCD40L, and others) [21]. For this reason, in order to have correct estimates of specific cytokine levels, it may be preferable to measure them in plasma rather than in serum [34, 111]. This notably raised issues when comparing serum and plasma levels for TGF, IL-1, IL-6, IL-7, and so forth [38].

Plasma is the soluble fraction of anticoagulated blood. To obtain plasma, various anticoagulants can be used (ethylenediaminetetraacetic acid (EDTA), lithium/sodium heparin, and sodium citrate). Cytokine measurements were shown to be affected by the anticoagulant used [78] and, notably, lithium heparin and sodium citrate were shown to affect levels of IL-6 and TNF-alpha compared to EDTA plasma [35, 112, 113]. Citrate plasma collection also results in the reduction of total protein concentration due to the volume of citrate anticoagulant diluting the blood, in addition to an osmotic withdrawal of water from blood cells [114]. Endotoxin present in lithium heparin tubes when sterility is broken [113] can also induce cytokine release from cells, whereas EDTA inhibits endotoxin [26, 31]. Variation in cytokine levels could be attributed to anticoagulant-induced release of cytokines by blood cells notably in heparin plasma but not in EDTA plasma, [115]. Altogether, plasma collection with use of EDTA seems to bring the most consistent results [34, 35, 116] and more closely resembles data obtained in serum [31, 35, 39, 78, 90, 117]. Cytokine stability also appears increased in EDTA plasma [26, 118] perhaps through EDTA’s role as a protease inhibitor. Further mechanisms can explain differences in stability such as change in degradation rate or modification of cytokine’s structure due to the differential presence of other proteins in EDTA plasma compared to citrate plasma or serum (i.e., soluble forms of receptors) leading to a lack of recognition of the antibodies used in the ELISA. The limitation in using plasma remains the need for rapid separation after collection with changes occurring as soon as 30 minutes after sample collection [34].

Over the recent years, improvements in the collection tubes have been made, notably with the use of serum separator tubes, which include a gel that serves as a barrier between serum and the clot [106], or the substitution of plastic for glass allowing direct centrifugation [119].

Altogether, no single type of sample is optimal for every analyte; therefore, the development of assays for individual cytokines should require optimisation on a case-by-case basis, although it would be recommended to collect both serum and EDTA plasma.

(2) Time to Processing. Time is an important factor that needs to be accounted for when measuring circulating cytokines which have a relatively short half-life and an important risk of degradation notably when comparing them to other proteins such as antibodies [26, 34, 120]. Changes in the amount of cytokine detected depend on the delay and duration of sample processing and are likely due to altered production by cells after blood collection [31, 54, 120], or their binding by other proteins (i.e., soluble receptors or cells surface receptor) [42, 120, 121], or, finally, due to enzymatic activities (proteases) leading to cytokine digestion (see also Table 1). Rapid processing of samples is therefore essential, notably as samples obtained from patients often present with higher concentrations or increased activity of proteases or other factors which render specimens even more unstable than those obtained from healthy controls [111]. Ideally, samples destined for cytokine detection should be collected in sterile (endotoxin-free) tubes and processed quickly with a minimum of 30 minutes of clotting time but no longer than 60 minutes after blood draw, independently of the type of tube used (plasma or serum). Processed plasma or serum should be frozen at −80°C as soon as possible in small aliquots to avoid repeated freeze-thaw cycles [107, 122]. Some reports proposed to keep samples refrigerated at 4–8°C (but not on ice) after clotting for the duration of processing as room temperature favours proinflammatory cytokine degradation such as IL-6 but conversely stabilises TNF-alpha [26, 34, 120, 123, 124]. Most cytokines are relatively stable with the well-known exception of TNF-alpha and IL-6 [42, 125, 126]; therefore, the interval between blood draw and separation should not exceed 3–24 hours, even when the tubes are stored at 4–8°C and only when EDTA tubes are used (TNF-alpha however cannot be reliably measured any longer), although many cytokines have not been sufficiently tested [26, 35, 37, 78].

The effects of centrifugation speed are more difficult to evaluate. Gradual increase in g values (from 200 to 13,000 g) is necessary to achieve graded depletion of platelets and leucocytes from plasma; however, it reduces the levels of certain cytokines (i.e., sCD40L) [52]. Of note, the use of blood tubes with gel separator imposes a certain centrifugation speed to allow separation of serum and cells but does not allow tubes to be chilled before or during centrifugation [127].

(3) Storage Temperature and Freeze-Thaw Cycles. By and large, most cytokines and soluble markers are quite stable if frozen (see also Table 1). Storage conditions, however, vary with a choice of temperatures from short-term storage at room temperature (RT) or 4–8°C (days) to medium term (a few months) more often between −20/−30°C and long term (years) at −70°C. Direct comparison of several cytokines in plasma stored for 20 days at RT, 4°C or −70°C, showed remarkably stable levels (IL-10) except for TNF-alpha particularly at room temperature [128]. In contrast, a more recent study of reliability and reproducibility of cytokine measurements in healthy donors [122] showed that, while most cytokine measurements are stable for up to 2 or 3 years when stored at −80°C (see details in Table 1), they do not all remain stable after repeated freeze-thaw cycles. After 4 years, most cytokines were degraded. Importantly in RA, levels of certain cytokines such as TNF-alpha increase with each successive freeze-thaw cycle [54, 90, 122]. Therefore, it remained difficult to compare studies from different centres even when using the same assay for cytokine measurements (i.e., commercial kit) [39]. Altogether, the consensus would recommend storing specimens at −80°C in as many small aliquots as possible to limit freeze-thaw cycles [129].

2.2.3. Analytical Variability

(1) Assay Type. Numerous immunoassays exist to measure cytokines both in their protein form: ELISA, nitrocellulose, or other solid phase assays, immunohistochemistry, and bead-based flow cytometry multiplex immunoassays, and in their molecular form: reverse transcriptase PCR, microarrays, and in situ hybridisation (Table 2). Immunoassays use antibody to immobilise cytokines on a solid surface and then identify them with different methods for quantification using colorimetric enzymatic reactions, fluorescence, luminescence, or even, in the past, radioactivity. There are two types of assays using either one or two antibodies: one being for cytokine capture adding more specificity compared to total protein plastic binding and the second one being for detection. The major benefit to using antibodies is that assays are more specific and reproducible. Several platforms for the detection and quantification of cytokines exist. There is no universal best method for cytokine measurements; however, the oldest technique (ELISA) is often used as gold standard despite the fact that direct comparison between many commercially available kits has not been performed. Cytokines show complex protein structures (monomers/polymers, precursors, various degrees of glycosylation, and degradation products) and their activity often depends on the integrity of such structure. Minor changes that may not be detected by physicochemical measurements, immunoassays, or biophysical methods may have dramatic effects on biological activity (e.g., cytokines may lose most of their biological activity but will remain detectable if measured as mass) [130]. The presence of soluble forms of the cytokine receptors (i.e., sIL-2R, sIL-7R, and sTNF-R) in biological samples and the existence of autoantibodies to cytokines (i.e., anti-TNF-alpha, IL-6, and IL-1) [131] may or may not interfere with the recognition of cytokines by either capture or detection of antibodies [39, 132134]. Each method has advantages and limitations and should be carefully selected with respect to the research purpose. To date, most cytokine measurements in large studies essentially used ELISA, which is widely accepted as the “gold standard” method. The main limitation of ELISA remains that it allows the characterization of a single cytokine at a time, hence the development of multiplex technologies. One of the most commonly used methods for this is the multiple target based assay [135], which can measure up to 100 different analytes per sample from a small volume of body fluid [136], or more recently the cytometry bead assay (CBA) which relies on bead as solid phase and uses flow cytometry to discriminate between analytes [137]. Multiplex measurement of inflammatory cytokines in human serum by electrochemiluminescence assay was recently developed [138]. These multiplex assays are in concept close to ELISAs and dependent upon the careful choice of the capture/detection antibody pairs and proper buffering to minimize differences in assay performances [135].


Cytokine assay techniqueDescriptionCharacteristics

BioassaysBioassays (commonly used shorthand for biological assays) are typically assays by which the potency or the nature of a substance is estimated by studying its effects on living organisms
They can be conducted to measure the concentration/effects of a cytokine on a living cell
Example: IL-2 bioassay using an IL-2 dependent cell line that will undergo apoptosis in the absence of IL-2 in a dose dependent manner
They require tissue culture facility
Low specificity
Semiquantitative detection
Highly sensitive with detection limit < 1 pg/mL
Narrow analytical range
Time consuming (24–96 h)
Low precision (CV = 20–100%)
Drug interference
Laborious protocol with high staff cost

ELISAQuantitative detection of a molecule (bioactive and inactive) based on its capture by an antibody followed by its detection by another antibody coupled with a detection (commonly named ELISA)
It requires specialised equipment
Less sensitive than bioassays <10 pg/mL
Relatively large sample volume
Wide analytical range
High reagent cost
Excellent precision (CV = 5–10%)
No drug interference
Simple and relative rapid protocol

Solid phase assay (Luminex)Technology based on the detection of dyed microbeads capturing a cytokine with a first antibody and quantifying it with a second antibody coupled with fluorescence and lasers detection
It allows multiplex detection
Small sample volume
Lower sensitivity than ELISA
Large range of analytes
Sensitive to interferences from heterophilic antibodies (i.e., naturally occurring anti-antibodies), anti-cytokine antibodies, and presence of soluble receptors

Other solid phase assays Mosaic ELISA
ELISA like technology allowing multiple detection of cytokines in a 96-well plate format by spotting capture antibodies
Small sample volume
Lower sensitivity than ELISA
Only 8 analytes per test

Molecular techniquesAll techniques allowing mRNA quantification
Earlier detection of cytokines at transcriptional level however may not represent cytokine production and release
They require specialised equipment
Highly specific
Highly sensitive as they can detect changes at the single-cell level
Complete analytical range (from single cytokine to as many as needed)
Excellent precision
No drug interference
Simple and relative rapid protocol
Relatively high cost

Several studies have compared cytokine levels determined by ELISA and multiplex immunoassays with results showing either good or poor correlations between the methods. Therefore, it is not surprising that discrepancies in data comparing measurements of cytokines were observed when different commercial/manufacturers’ kits were used, even if preanalytical conditions of samples collection, separation, and storage were identical [85, 136, 139]. The use of different antibody clones to capture and detect cytokines is also likely to affect results and change the level of sensitivity of such assays. Furthermore, some monoclonal antibodies recognise different molecular complexes (monomers/polymers, precursors, glycosylation, degradation products, or total bioactive or inactive forms) [140]. In summary, comparison of the same samples (eliminating preanalytical bias) using several commercial ELISAs demonstrated that variability was mostly attributable to each assay (measuring TNF-alpha, IL-1 alpha and IL-1beta, IL-6, IL-2, IFN-gamma, and the soluble receptors of IL-2 and TNF) but yielded comparable results when the same ELISA was used at different centres [85, 139]. The nature of the different pairs of monoclonal antibodies employed in each ELISA is most likely the major source of variability, but these findings also highlight the necessity of establishing international standards for all immunoassays as ranges are also widely variable between these commercial assays. If cytokines are to be employed as clinical biomarkers for diagnosis, prognosis, and prediction, accurate and reproducible assays need to be adopted internationally.

(2) Interferences. Interferences within immunoassays are numerous, complex, and usually difficult to resolve. Proteins can show an altered expression pattern in more than one disease. The presence of lipids, complement factors, and other complex molecules in the blood was also shown to interfere with a number of assays. Human anti-animal antibodies present in biological samples (especially human anti-mouse antibodies) may cause problems; however, these may be blocked by the use of multiple species serums as blocking agents [141]. Haemolysis interference occurs rarely; however, it can affect some analytes. Lipaemia interferences were confined when using immunonephelometric and immunoturbidimetric assays, and, ideally, grossly lipaemic samples should be cleared (using ultracentrifugation of lipaemic samples with correction for volume displacement errors) or discarded. Antigen excess may, in some cases, result in false low values [142]. Complement factors and paraproteins are capable of binding to assay antibodies (capture and detection) causing interferences [142]. In addition, biological fluids may also contain naturally occurring antibodies to a variety of proteins, including cytokines themselves. Such antibodies, although at variable levels notably between normal donor and patient populations, can interfere with assays particularly if they share the same epitope on the cytokine [143]. The existence of autoantibodies against cytokines has been documented for TNF, IL-1 (alpha and beta), IL-2, IL-6, IL-8, IL-10, and IL-18 [144148]. Autoantibodies against IL-1 are the best studied. Their prevalence is high with an affinity which can reach up to 10−11 M that is very similar to the affinity of antibodies developed for immunoassays [140]. However, the main issue remains heterophilic antibodies. These antibodies are naturally produced polyclonal autoantibodies with low specificity directed against multiple poorly defined antigenic immunogens. Most often, they are present in individuals exposed to foreign proteins (e.g., domestic animals and household pets). The occurrence of false positives in immunoassays [1316] is often the result of heterophilic antibodies nonspecifically bridging the assay antibodies [18, 19]. As a result, studies have often overestimated cytokine levels notably when using the Luminex technology [12].

Blood samples from patients with autoimmune diseases, such as RA, may be problematic due to the presence of additional disease related autoantibodies [149]. RF is an autoantibody directed against the Fc portion of IgG and is found in 75% of patients presenting with RA as well as other diseases such as Sjögren’s syndrome, infective endocarditis, systemic sclerosis, and systemic lupus erythematous (SLE) [24]. RF was shown to exhibit most of the heterophilic antibody properties with several antigen cross-reactions [25] and hence immunoassay in RA is particularly sensitive to this issue and needs careful evaluation for RF interference [12, 150153]. Heterophilic immunoglobulin may further develop as a result of treatment with drugs attached to mouse (or humanised) monoclonal antibodies.

Several methods for removing heterophilic antibody (notably RF) from patients sera have been developed [21, 154156]: (i) initial serial dilutions may be recommended, particularly when results demonstrate nonlinearity suggesting the presence of heterophilic antibodies, (ii) the use of blocking reagents such as nonimmune serum from the same species as the assay antibodies, species-specific polyclonal IgG, and multispecies mixture (20% normal mouse serum, 10% goat serum, and 10% rabbit serum), as well as commercial reagents such as HeteroBlock [155], and (iii) the specific removal of immunoglobulin G using sepharose-L or polyethylene glycol precipitation (PEG 6000) has also been used. These methods act by physical removal of the immunocomplexes [155], which are then separated by centrifugation. Several reports have been published investigating interference by heterophilic antibodies in RA sera using solid phase multiplexing technology including Luminex [23, 155, 157, 158], a glass chip/chemiluminescence platform, or a multiplex sandwich ELISA. They showed clear interference (i.e., false positive) in RF-positive sera but not in negative samples [157]. In our lab, all methods were efficient at blocking/removing relatively low RF quantities in serum samples from RA patients [12]; however, none of these methods were effective when high levels of RF were present (>100 U/L) and residual RF still generated false positive results particularly when using certain types of assays (Luminex) but not others (ELISA, membrane-based ELISA, Mosaic ELISA, or CBA).

(3) Standardisation and Quality Control. Commercially available immunoassays in the form of “kits” are now extensively used. Considerable variability can arise from the use of these assays. Differences in measured levels of cytokines in identical samples using different standards ranged from 10- to 100-fold [130, 159161]. Some issues are related to the use of different epitope specificity of the antibodies, while others arise due to the use of various reference preparations (standards) for calibrating the assays [55]. Comparison of cytokine levels requires unit definition by a standard that is assay independent, which, once defined, should be used by any laboratory, thus providing a means of ensuring uniformity worldwide [130]. Variations as a result of differences in standards account for as much variability as sample collection, processing, or storage issues [31, 42, 125, 159168]. All cytokine assays should therefore be calibrated against such standards, regardless of assurances provided by the kit manufacturers. Notably, results of cytokine assays should be reported in picograms or nanograms per milliliter instead of arbitrary units. Major international efforts to organise standardisation of cytokine measurements have been conducted by the World Health Organisation, (see details at http://www.nibsc.ac.uk/products/biological_reference_materials.aspx), The National Institute for Biological Standards and Control (NIBSC), and the Biologics Evaluation and Research (The National Institutes of Health (NIH), Bethesda, MD 20205, USA) (http://www.who.int/biologicals/) [130, 131, 169]. Nonetheless, baseline values for a lot of cytokines have not yet been reliably established in healthy controls (despite a range suggested by most manufacturers), making it difficult to interpret the biological significance of minor variations in cytokine levels in patients [170]. Furthermore, some cytokine assays are sensitive at relatively high concentrations that may not always cover the physiological range even in diseases [12]. Quality control (QC) measure is also an essential step of biomarker development. Therefore, during the analytical phase, QC should be considered to document analytical performance during any studies to determine the acceptance or rejection of an analytical run during postanalytical sample analysis [136, 171]. QC samples could be prepared to evaluate the lower, middle, and upper performance limits of an assay. A number of validation samples (at least five different concentrations) should also be used to estimate intra- and interrun accuracy/precision and stability [136, 172, 173].

3. Cytokines Network in RA

Over the years, increasing numbers of cytokines have been involved in RA pathology, further to those used as target of cytokine-blocking therapies which emerged from the hypothesis that the most abundant cytokines present in the joint were more likely to be pathogenic. A large number of cytokines are detected at the disease site (through both mRNA and protein quantification) in both synovial tissue and fluid, where they have a role in perpetuating inflammation, cartilage destruction, and bone remodelling associated with RA. Several methods of detection (ELISA, immunohistochemistry) identified TNF-alpha and IL-1 as major players in the network of cytokines, notably directly expressed at the disease site in joint tissue or fluid. IL-6 and IFN-gamma are also present as well as GM-CSF and LIF. More recently, other cytokines were added to this list (IL-7, IL-15, IL-17, IL-18, IL-21, and MIP-1 notably) together with cytokines with activities targeted towards fibroblasts (TGF-betas notably) and finally several growth factors (PDGF, EGF, and VEGF) [174] and chemokines (IL-8, SDF-1, RANTES, and MCP-1). Cytokines favouring survival of infiltrating cells have also been detected (such as the pairs between IL-7 and T cell or BAFF and B cells). However, if proinflammatory cytokines (TNF-alpha, IL-1, and IL-6) are abundant in all patients, cytokines classically defined as anti-inflammatory and regulatory (IL-4, IL-10, IL-13, and TGF) [175, 176] as well as antagonist receptors (IL-1RA, or soluble IL-2R, or TNF-R) are also present. Most of these cytokines have dual roles with anti- and proinflammatory aspects depending on the context and the network they form; hence, studying their roles and actual effects is particularly complex. The redundancy and synergy between the effects of all cytokines in such an intricate network may further explain the inadequate response to single blockade therapy notably in established disease [175].

The interplay between cytokines, where excess of one may result in suppressed production of another, further complicated by interactions with soluble receptors for some of these cytokines, renders data interpretation challenging (notably for TNF-alpha and IL-1) [88, 89]. The relationship between blood and tissue is often complex and translating findings often proves difficult if not conflicting. Data on cytokine levels in humans in relation to disease activity is still limited. Increased levels of cytokines such as IL-l, IL-6, and TNF have been interpreted as an indicator of the inflammatory state. It is unlikely that these cytokines could serve as “biomarkers” in inflammatory disease, as they are linked to the disease biological processes, hence not specifically associated with a particular disease. Additionally, lack of correlation is often observed between cytokine levels (in serum/plasma) and clinical endpoints.

On the other hand, the absence of a cytokine in disease is particularly difficult to interpret. As indicated above, there may be multiple reasons for the inability to detect a cytokine when actually it is expected to be found. Even in the absence of specific or nonspecific inhibitors, excessive consumption of a cytokine versus lack of its synthesis is hard to dissociate. As an example, IL-7 levels were reported to be low in RA serum [177179]; however, they are high in synovial fluid and tissue. The presence of high levels of sIL-7R in serum [180] may explain this discrepancy and the associated loss of biological activity [177, 181].

Despite these limitations, there are some cytokine biomarkers, which appear to be relevant in RA. IL-6, despite not being disease specific [78, 92, 182], was shown to be more sensitive than CRP (despite being directly correlated with it) for the prediction of therapeutic response of RA patients to rituximab [183]. Similarly, IL-7 was shown to have some value as diagnostic biomarker associated with potential for more erosive disease [179].

3.1. Differential Cytokine Expression between Diseases

Over the years, many studies provided evidence of differential expression of cytokines between healthy control (HC) and diseases such as RA, osteoarthritis (OA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), reactive arthritis (ReA), systemic lupus erythematosus (SLE), or gout. These initially used functional assays measuring the production of cytokines in variable cell subsets using intracellular expression of cytokines (in CD4+ or CD8+, T cells or B cells, or monocytes), ELISA, ELISOPT, or mRNA quantification. Several important observations were derived from these experiments and the tables below summarise all this data as well as tissue sources and technology/experiment.

In vitro assays removed the microenvironment context; however, they reflect good the capabilities acquired through exposure to the priming effect that such microenvironment may exert (i.e., Th1/Th2 polarization, transition from naïve to memory). Altogether, they demonstrate the dysregulated expression of certain cytokines in T cells subsets notably and increased expression by monocytes in RA patients. Importantly, all cytokines tested were shown to be increased, with the exception of IL-2 and IL-4. Interestingly, RA patients’ T cells showed hyporesponsiveness to stimulation of the T cell receptor (TCR) pathways and hardly produced any cytokines despite evidence of previous activation (memory phenotype) [184]. This deficit was attributed to chronic exposure to TNF-alpha [185] and/or abnormal RAP1 signalling [186188]. The classic model of T cell naïve/memory differentiation is perturbed in RA. T cells despite being naïve with respect to antigen stimulation [189] express chemokine receptors which facilitate trafficking to sites of inflammation [7, 177]. This phenomenon was hypothesized to result from cytokine activation notably of naïve T cells (by IL-6 and TNF-alpha) bypassing the need for an antigen to achieve activation [190, 191]. Similar cells were found in RA joint (but not OA) [192] where they enable TNF-alpha production by monocytes in an antigen-independent manner. These properties of cytokine activated T cells were further extended to chemokine production and were confirmed in vivo using a cytokine cocktail containing IL-2, IL-6, and TNF-alpha [193]. Such increased ability to produce all types of cytokines reflects the chronic stage of the disease but nevertheless gives insight into potential candidates for further biomarker program.

3.2. Differential Cytokine Levels in RA Sera or SF

There are several studies comparing circulating levels of cytokine, they often show discrepancy in their results, and most do not use the appropriate biomarker development strategy. IL-1beta and TNF-alpha are increased in RA [194] and such profile is accentuated in active diseases compared to clinical remission [195]. In contrast, low levels of IL-2 and IL-7 were reported [177, 179, 194, 196]; however, those may be due to high levels of soluble sIL-2R and sIL-7R which are also present. IL-6 could not be detected in HC serum, while serum IL-6 levels are substantially increased in RA with significant circadian variations corresponding to the circadian rhythm of symptoms in RA [79]. High IL-7 [197] and IL-16 [198] were detected in sera and SF of RA patients compared to OA and are also confirmed in synovial tissues by mRNA levels. Certain cytokine levels were related to disease parameters such as IL-1RA and the number of tender and swollen joints [199], IL-18 (both sera and SF) and disease activity [200, 201], and IL-7 in the tissue (both mRNA and protein) with local levels of inflammation measured during arthroscopy [196]. IL-21 is highly produced in the synovial fluid of RA patients compared to paired serum specimens as well as healthy control sera. The increased levels of IL-21 correlate with those of IL-17 [202] and an association between levels of IL-21 and Th17 cells responses in the RA synovium was shown [202].

Similar increased serum levels of many cytokines were indeed found in other rheumatic diseases: notably PSA [203205], SLE [206, 207], AS [208210], and scleroderma [211, 212] (IL-1, IL-6, IL-7, IL-8, IL-10, IL-12, IL-16, IL-17, IL-18, and IFN-gamma, TGF-beta, or TNF-alpha, as well as IL-1RA and sIL-2R or leptin) suggesting that such rises may reflect inflammation rather than being disease specific. Therefore, the biomarker value of either one of the cytokines, or a combination of them, will likely depend on whether their disease specificity can be verified.

3.3. Cytokines as Diagnostic Biomarkers for RA

The early diagnosis of RA is critical, as it has been demonstrated that a therapeutic window of opportunity is available very early in the development of RA, when disease can be stopped efficiently, preventing structural and functional damage and leading to remission if treated. In face of such a need, clinical diagnosis remains difficult. At the (very) early stage, inflammatory arthritis often has an atypical presentation with progression towards RA that can vary in speed. Autoantibodies (RF and ACPA) are useful in RA diagnosis as recently recognised by their inclusion in the new diagnostic EULAR 2010 criteria. However, they both lack sensitivity in early disease (<50%) [213] even if ACPA specificity is quite high (over 95%) [214].

The ideal RA diagnostic biomarker should therefore be characterised by high specificity and sensitivity, both close to 100%. An ideal biomarker should also detect the presence of RA at early stages. Few, if any, biomarker testing systems achieve these levels of sensitivity and specificity although this can be approached by improvement of the assays. In advanced disease (i.e., fully developed RA), biological differences between healthy and disease states are easily detected. In contrast, in early disease, the biological distinctions between healthy and disease states or alternative diagnosis are often more subtle, and clear differentiation even for a gold standard becomes more challenging. Therefore, the evaluation of a candidate diagnostic biomarker requires an infallible diagnosis to be established which in RA remains difficult [215].

Cytokines and other soluble factors are prime candidates for diagnostic biomarkers. Several studies investigated their expression using variable methods (ELISA, multiplex assays, or gene expression) and material (tissue and body fluids). However, few studies actually compared very early inflammatory arthritis with differential outcome and still use healthy individuals or established disease patients as controls. Cytokines detected in joints were not different in 12- month disease duration compared to more advanced RA [216]; however, these findings remain to be established in very early disease. Even if right and left RA knee showed similar profiles (IL-6, IL-8, IL-10, and IFN-gamma, high expression of IL-1beta, TNF-alpha, and TGF-beta, low levels of IL-2 and GM-CSF, and no detectable IL-4 or IL-5) [217], the same pattern was observed in other diseases such as seronegative spondyloarthropathy or OA with different levels of expression.

Using Luminex technology with the blocking of heterophilic antibody, increased levels of TNF-alpha, IL-1beta, IL-6, IL-12P40, IL-13, and several chemokines (CXCL10, CCL11, CCL2, and IL-8) were observed in sera from RA patients with <6-month symptom duration compared to HC [23]. The profile was specific to RA and not reproduced in established AS or SpA but was not investigated in patients with early inflammatory symptoms who did not progress towards RA. The profile was also restricted to ACPA-positive patients suggesting increased inflammation associated with autoreactivity. In addition, ACPA was closely related to RF in this study (titres were directly correlated), questioning the efficiency of the RF-blocking methodology used as most cytokine levels were also related to ACPA levels.

In a similar study [158] comparing already diagnosed RA patients of less than 6-month symptom duration with established AS and PsA, a multiplex biomarker platform (combining cytokines, bone turnover markers, metalloproteinases, inflammatory markers, and several citrullinated epitopes) established a signature again including cytokines such as TNF-alpha, IL-1alpha and beta, IL-6, IL-12p40, IL-15, IL-17, GM-CSF, and eotaxin. However, most were also present in AS and PsA (TNF-alpha, IL-1beta, IL-6, IL-17, and eotaxin) and others were associated with autoantibody positive disease (IL-1alpha, IL-12p70, and IL-15).

Studies truly investigating early diseases and the value of cytokines as diagnostic biomarkers in a predictive manner are few. SF from early inflammatory arthritis patients before diagnosis established that patients with persistent symptoms on development of RA showed increase in Th2 cytokines (IL-4 and IL-13) but not Th1 (IFN-gamma) [218]. IL-17 was also increased however only in established RA [218]. In individuals who donated serum samples and later developed RA, a multiplex study showed significant increased levels of cytokines related to T cell activation (IL-2, IL-6), inflammation (IL-1beta, IL-1RA, and TNF-alpha), Th1 (IL-12 and IFN-gamma), Th2 (IL-4, IL-13, and eotaxin), and immune regulation (IL-10), while chemokines, stromal cell-derived cytokines, and angiogenic-related markers were elevated in patients after the development of RA rather than in individuals before the onset of RA [219]. Levels were particularly increased in ACPA-positive and RF-positive individuals. However, in all three studies, every cytokine and chemokine tested were increased (even if not significantly) and again particularly in ACPA/RF-positive patients, whereas other studies demonstrated reduction (i.e., IL-2 and IL-7). Therefore, technical issues related to heterophilic antibody interference may have to be considered when interpreting these data. A similar preclinical RA study [220] showed no detectable cytokine more than 5 years before RA onset, but during the 5-year interval before diagnosis, increased levels were associated with an increased likelihood of the risk of developing RA (IL-1 alpha, IL-1beta, IL-1RA, IL-4, IL-10, TNF-alpha, and soluble TNF-RI).

In established RA as well as in patients with less than 24-month symptom duration, reduced levels of circulating IL-7 have been reported [177, 196]. IL-7 is a pleiotropic cytokine regulating peripheral T cell homeostasis, notably in RA [177, 221, 222]. However, IL-7 is highly expressed in the joints of RA patients [196, 197, 223, 224], and such discrepancies between low systemic levels and high expression at disease site have also been reported in systemic sclerosis [225] and recently in ulcerative colitis and Crohn’s disease [226, 227]. A cohort of 250 sera from patients with very early symptoms suggesting a possible evolution towards RA (less than 6-month duration and 5-year follow-up) designed to discover diagnostic biomarkers demonstrated the potential of IL-7 as a biomarker [2].

3.4. Cytokines as Markers for Treatment Selection and Response to Therapy

Biological therapies (cytokine blockade or receptor antagonism) nowadays appear very effective in chronic inflammatory conditions such as RA, however, in a limited number of patients, with up to 40% nonresponse. Considering the cost of such therapies, biomarker prediction response and allowing for selection of the most appropriate biological treatment would have considerable impact. Most authorities recommend starting therapy with biologics after the failure to respond to at least one disease-modifying agent in RA. However, due to the limited number of studies, there is little guidance about which biological agent to select although anti-TNF remains the most commonly used.

RA patients not responding to anti-TNF showed higher synovial fluid IL-6 at baseline amongst elevated levels of IL-1beta, IL-1RA, IL-2, IL-4, IL-8, IL-10, IL-17, IFN-gamma, G-CSF, GM-CSF, and TNF-alpha. In contrast, responders had elevated IL-2 and G-CSF. In plasma, however, levels were not significantly predicting response, and IL-6 levels decreased posttreatment. In this study, SF cytokine clustering revealed 6 groups of patients with possibly underlying different cellular pathologies, and IL-6, IL-2, and G-CSF in SF may be useful in predicting response to anti-TNF [228]. Recently, we also showed that serum IL-6 was significantly higher at baseline in rituximab nonresponders and that a significant reduction followed treatment in responders only despite adequate B cell depletion in nonresponders [229]. Multivariate logistic regression analysis of synovial cytokine expression showed that TNF at baseline could only explain ~10–15% of the variance in response to TNF blockade [230], suggesting that TNF expression itself would have a limited role in relation to personalised health care. Synovial tissue analysis associated absence of sign of improvement with increased TNF and MMP-3 expression [231, 232]. In contrast, another study showed response to be associated with higher TNF bioactivity in the blood [233], which is more convenient for personalised medicine.

To date, several studies using blood have used gene expression rather than ELISA. CCL4, IL-8, and IL-1beta discriminated between responders and nonresponders to anti-TNF [234]. Several gene signatures have been published so far (some including IL-8, IL-2R) [235238] with a sensitivity of 90% and a specificity of 70% [237] and 94.4% sensitivity and 85.7% specificity for the response to anti-TNF treatment [238]. Response to anti-TNF (etanercept) was associated with reduced levels of IL-6 and increased IL-23 and IL-32 posttreatment while there was no change in nonresponders; however, no baseline level had predictive value [239].

Recently, several interferon signalling related signatures have emerged as potential biomarkers of response to biological therapies [240242] as well as for the progression of “at risk” individuals to symptomatic arthritis [243]. Such signatures are interesting as they most likely reflect an immunological status that is favourable to responding or not to therapy, although they are not really linked to the presence/absence of interferon. Indeed, these signatures combined different sets of intracellular signalling factors and transcriptional regulators (between 8 and 15 markers) and are measured through gene expression (using mostly qPCR).

4. Conclusion

Assays measuring known diagnostic biomarkers are commonly used in clinical practice. In fact, it has been reported that about 70% of the decisions made by physicians are based on the results provided by those tests [244]. However, the implementation of novel biomarkers into clinical practice proves to be a long and challenging process, which includes convincing physicians. The assessment of the impact of using the biomarker on general health is an essential step to guarantee the uptake of the biomarker into clinical practice and to further optimise its use. This area of research is likely to become increasingly important as more biomarkers enter clinical practice [245]. Given the complexity and heterogeneous nature of RA, it is unlikely that a single cytokine may provide sufficient discrimination. Many reliable cytokine assays are nowadays available with multiplex formats taking the lead (although this may not be an appropriate solution in RA due to RF interferences). These have established clinical utility for other diseases and purposes and should be easily (technically) transferable to rheumatology, although the exact performance characterization and quality assurance for the specific cytokines of interest in RA may need to be established. At present, limitation in RA lies more in the disease related complexity of networks, the elucidation of the respective role, and the redundant effect that one cytokine may have with another.

Finally, multiple biomarker signatures potentially using genetic as well as proteomic markers may represent a more realistic approach for the future of personalised medicine in RA. Such multifactorial analysis may potentially reveal patterns rather than individual biomarkers. As such, it is interesting that IL-7 alone was able to predict diagnostic at very early disease stage, whereas more complex combination of markers may be needed to predict response to therapy and define subsets of patients with more advanced and heterogonous disease.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgment

This work has been supported by the IMI Funded Project BeTheCure no. 115142-2.

References

  1. J. Avouac, L. Gossec, and M. Dougados, “Diagnostic and predictive value of anti-cyclic citrullinated protein antibodies in Rheumatoid Arthritis: a systematic literature review,” Annals of the Rheumatic Diseases, vol. 65, no. 7, pp. 845–851, 2006. View at: Publisher Site | Google Scholar
  2. V. Goëb, P. Aegerter, R. Parmar et al., “Progression to Rheumatoid Arthritis in early inflammatory arthritis is associated with low IL-7 serum levels,” Annals of the Rheumatic Diseases, vol. 72, no. 6, pp. 1032–1036, 2013. View at: Publisher Site | Google Scholar
  3. D. M. Gerlag, K. Raza, L. G. M. van Baarsen et al., “EULAR recommendations for terminology and research in individuals at risk of Rheumatoid Arthritis: report from the Study Group for Risk Factors for Rheumatoid Arthritis,” Annals of the Rheumatic Diseases, vol. 71, no. 5, pp. 638–641, 2012. View at: Publisher Site | Google Scholar
  4. J. Shi, R. Knevel, P. Suwannalai et al., “Autoantibodies recognizing carbamylated proteins are present in sera of patients with Rheumatoid Arthritis and predict joint damage,” Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 42, pp. 17372–17377, 2011. View at: Publisher Site | Google Scholar
  5. J. Shi, A. Willemze, G. M. C. Janssen et al., “Recognition of citrullinated and carbamylated proteins by human antibodies: specificity, cross-reactivity and the “AMC-Senshu” method,” Annals of the Rheumatic Diseases, vol. 72, no. 1, pp. 148–150, 2013. View at: Publisher Site | Google Scholar
  6. D. Aletaha, T. Neogi, A. J. Silman et al., “Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative,” Arthritis & Rheumatism, vol. 62, no. 9, pp. 2569–2581, 2010. View at: Publisher Site | Google Scholar
  7. C. Burgoyne, E. Vital, S. Dass, D. Corscadden, P. Emery, and F. Ponchel, “Alteration of B cell phenotype following depletion with rituximab in Rheumatoid Arthritis,” in Proceedings of the 28th European Workshop for Rheumatology Research (EWRR '08), Toulouse, France, 2008. View at: Google Scholar
  8. B. Saleem, H. Keen, and V. Goeb, “Patients with RA in remission on TNF blockers: when and in whom can TNF blocker therapy be stopped?” Annals of the Rheumatic Diseases, vol. 69, no. 9, pp. 1636–1642, 2010. View at: Publisher Site | Google Scholar
  9. F. Ponchel, V. Goëb, R. Parmar et al., “An immunological biomarker to predict MTX response in early RA,” Annals of the Rheumatic Diseases, 2013. View at: Publisher Site | Google Scholar
  10. S. Ramiro, P. Machado, J. A. Singh, R. B. Landewé, and J. A. P. da Silva, “Applying science in practice: the optimization of biological therapy in Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 12, no. 6, article 220, 2010. View at: Publisher Site | Google Scholar
  11. C. Fueldner, A. Mittag, J. Knauer et al., “Identification and evaluation of novel synovial tissue biomarkers in Rheumatoid Arthritis by laser scanning cytometry,” Arthritis Research & Therapy, vol. 14, no. 1, article R8, 2012. View at: Publisher Site | Google Scholar
  12. S. M. Churchman, J. Geiler, R. Parmar et al., “Multiplexing immunoassays for cytokine detection in the serum of patients with Rheumatoid Arthritis: lack of sensitivity and interference by rheumatoid factor,” Clinical and Experimental Rheumatology, vol. 30, no. 4, pp. 534–542, 2012. View at: Google Scholar
  13. A. M. Prince, B. Brotman, D. Jass, and H. Ikram, “Specificity of the direct solid-phase radioimmunoassay for detection of hepatitis-B antigen,” The Lancet, vol. 1, no. 7816, pp. 1346–1350, 1973. View at: Google Scholar
  14. W. M. Hunter and P. S. Budd, “Circulating antibodies to ovine and bovine immunoglobulin in healthy subjects: a hazard for immunoassays,” The Lancet, vol. 2, no. 8204, p. 1136, 1980. View at: Publisher Site | Google Scholar
  15. P. J. Howanitz, J. H. Howanitz, H. V. Lamberson, and K. M. Ennis, “Incidence and mechanism of spurious increases in serum thyrotropin,” Clinical Chemistry, vol. 28, no. 3, pp. 427–431, 1982. View at: Google Scholar
  16. M. H. Zweig, G. Csako, C. C. Benson, B. D. Weintraub, and B. B. Kahn, “Interference by anti-immunoglobulin G antibodies in immunoradiometric assays of thyrotropin involving mouse monoclonal antibodies,” Clinical Chemistry, vol. 33, no. 6, pp. 840–844, 1987. View at: Google Scholar
  17. J.-F. Benoist, D. Orbach, and D. Biou, “False increase in C-reactive protein attributable to heterophilic antibodies in two renal transplant patients treated with rabbit antilymphocyte globulin,” Clinical Chemistry, vol. 44, no. 9, pp. 1980–1985, 1998. View at: Google Scholar
  18. L. M. Boscato and M. C. Stuart, “Heterophilic antibodies: a problem for all immunoassays,” Clinical Chemistry, vol. 34, no. 1, pp. 27–33, 1988. View at: Google Scholar
  19. M. H. Nahm and J. W. Hoffmann, “Heteroantibody: phantom of the immunoassay,” Clinical Chemistry, vol. 36, no. 6, p. 829, 1990. View at: Google Scholar
  20. T. B. Martins, “Development of internal controls for the Luminex instrument as part of a multiplex seven-analyte viral respiratory antibody profile,” Clinical and Diagnostic Laboratory Immunology, vol. 9, no. 1, pp. 41–45, 2002. View at: Publisher Site | Google Scholar
  21. W. de Jager, B. J. Prakken, J. W. J. Bijlsma, W. Kuis, and G. T. Rijkers, “Improved multiplex immunoassay performance in human plasma and synovial fluid following removal of interfering heterophilic antibodies,” Journal of Immunological Methods, vol. 300, no. 1-2, pp. 124–135, 2005. View at: Publisher Site | Google Scholar
  22. W. de Jager and G. T. Rijkers, “Solid-phase and bead-based cytokine immunoassay: a comparison,” Methods, vol. 38, no. 4, pp. 294–303, 2006. View at: Publisher Site | Google Scholar
  23. W. Hueber, B. H. Tomooka, X. Zhao et al., “Proteomic analysis of secreted proteins in early Rheumatoid Arthritis: anti-citrulline autoreactivity is associated with up regulation of proinflammatory cytokines,” Annals of the Rheumatic Diseases, vol. 66, no. 6, pp. 712–719, 2007. View at: Publisher Site | Google Scholar
  24. M. Longmore, I. Wilkinson, E. Davidson, A. Foulkes, and A. Mafi, Oxford Handbook of Clinical Medicine, Oxford University Press, 2010.
  25. V. Agnello, A. Arbetter, and G. I. de Kasep, “Evidence for a subset of rheumatoid factors that cross-react with DNA-histone and have a distinct cross-idiotype,” Journal of Experimental Medicine, vol. 151, no. 6, pp. 1514–1527, 1980. View at: Google Scholar
  26. P. W. Thavasu, S. Longhurst, S. P. Joel, M. L. Slevin, and F. R. Balkwill, “Measuring cytokine levels in blood. Importance of anticoagulants, processing, and storage conditions,” Journal of Immunological Methods, vol. 153, no. 1-2, pp. 115–124, 1992. View at: Google Scholar
  27. H.-L. Wong, R. M. Pfeiffer, T. R. Fears, R. Vermeulen, S. Ji, and C. S. Rabkin, “Reproducibility and correlations of multiplex cytokine levels in asymptomatic persons,” Cancer Epidemiology Biomarkers & Prevention, vol. 17, no. 12, pp. 3450–3456, 2008. View at: Publisher Site | Google Scholar
  28. R. Patil, S. Shukre, R. Paranjape, and M. Thakar, “Heparin and EDTA anticoagulants differentially affect the plasma cytokine levels in humans,” Scandinavian Journal of Clinical & Laboratory Investigation, vol. 73, no. 5, pp. 452–455, 2013. View at: Publisher Site | Google Scholar
  29. R. P. Jackman, G. H. Utter, J. W. Heitman et al., “Effects of blood sample age at time of separation on measured cytokine concentrations in human plasma,” Clinical and Vaccine Immunology, vol. 18, no. 2, pp. 318–326, 2011. View at: Publisher Site | Google Scholar
  30. K. Skogstrand, C. K. Ekelund, P. Thorsen et al., “Effects of blood sample handling procedures on measurable inflammatory markers in plasma, serum and dried blood spot samples,” Journal of Immunological Methods, vol. 336, no. 1, pp. 78–84, 2008. View at: Publisher Site | Google Scholar
  31. P. Riches, R. Gooding, B. C. Millar, and A. W. Rowbottom, “Influence of collection and separation of blood samples on plasma IL-1, IL-6 and TNF-α concentrations,” Journal of Immunological Methods, vol. 153, no. 1-2, pp. 125–131, 1992. View at: Google Scholar
  32. G. Leroux-Roels, J. Philippe, F. Offner, and A. Vermeulen, “In-vitro production of cytokines in blood,” The Lancet, vol. 336, no. 8724, p. 1197, 1990. View at: Publisher Site | Google Scholar
  33. A. Biancotto, X. Feng, M. Langweiler, N. S. Young, and J. P. McCoy, “Effect of anticoagulants on multiplexed measurement of /chemokines in healthy subjects,” Cytokine, vol. 60, no. 2, pp. 438–446, 2012. View at: Publisher Site | Google Scholar
  34. L. Flower, R. H. Ahuja, S. E. Humphries, and V. Mohamed-Ali, “Effects of sample handling on the stability of interleukin 6, tumour necrosis factor-α and leptin,” Cytokine, vol. 12, no. 11, pp. 1712–1716, 2000. View at: Publisher Site | Google Scholar
  35. R. de Jongh, J. Vranken, G. Vundelinckx, E. Bosmans, M. Maes, and R. Heylen, “The effects of anticoagulation and processing on assays of IL-6, sIL-6R, sIL-2R and soluble transferrin receptor,” Cytokine, vol. 9, no. 9, pp. 696–701, 1997. View at: Publisher Site | Google Scholar
  36. J. B. Eggesbø, I. Hjermann, A. T. Høstmark, and P. Kierulf, “LPS induced release of IL-1β, IL-6, IL-8 and TNF-α in EDTA or heparin anticoagulated whole blood from persons with high or low levels of serum HDL,” Cytokine, vol. 8, no. 2, pp. 152–160, 1996. View at: Publisher Site | Google Scholar
  37. G. Kenis, C. Teunissen, R. de Jongh, E. Bosmans, H. Steinbusch, and M. Maes, “Stability of interleukin 6, soluble interleukin 6 receptor, interleukin 10 and CC16 in human serum,” Cytokine, vol. 19, no. 5, pp. 228–235, 2002. View at: Publisher Site | Google Scholar
  38. S. W. Read, A. Rupert, R. Stevens, A. O'Shea, and I. Sereti, “Delayed sample processing leads to marked decreases in measured plasma IL-7 levels,” Journal of Acquired Immune Deficiency Syndromes, vol. 42, no. 4, pp. 511–512, 2006. View at: Publisher Site | Google Scholar
  39. N. Aziz, P. Nishanian, R. Mitsuyasu, R. Detels, and J. L. Fahey, “Variables that affect assays for plasma cytokines and soluble activation markers,” Clinical and Diagnostic Laboratory Immunology, vol. 6, no. 1, pp. 89–95, 1999. View at: Google Scholar
  40. J. Lengellé, E. Panopoulos, and F. Betsou, “Soluble CD40 ligand as a biomarker for storage-related preanalytic variations of human serum,” Cytokine, vol. 44, no. 2, pp. 275–282, 2008. View at: Publisher Site | Google Scholar
  41. H. Redl, S. Bahrami, G. Leichtfried, and G. Schlag, “Special collection and storage tubes for blood endotoxin and cytokine measurements,” Clinical Chemistry, vol. 38, no. 5, pp. 764–765, 1992. View at: Google Scholar
  42. A. R. Exley and J. Cohen, “Optimal collection of blood samples for the measurement of tumor necrosis factor α,” Cytokine, vol. 2, no. 5, pp. 353–356, 1990. View at: Google Scholar
  43. G. Leroux-Roels, F. Offner, J. Philippe, and A. Vermeulen, “Influence of blood-collecting systems on concentrations of tumor necrosis factor in serum and plasma,” Clinical Chemistry, vol. 34, no. 11, pp. 2373–2374, 1988. View at: Google Scholar
  44. R. Zimmermann, J. Koenig, J. Zingsem et al., “Effect of specimen anticoagulation on the measurement of circulating platelet-derived growth factors,” Clinical Chemistry, vol. 51, no. 12, pp. 2365–2368, 2005. View at: Publisher Site | Google Scholar
  45. D. J. Grainger, D. E. Mosedale, and J. C. Metcalfe, “TGF-β in blood: a complex problem,” Cytokine & Growth Factor Reviews, vol. 11, no. 1-2, pp. 133–145, 2000. View at: Publisher Site | Google Scholar
  46. J. Kropf, J. O. Schurek, A. Wollner, and A. M. Gressner, “Immunological measurement of transforming growth factor-beta I (TGF- β1) in blood; assay development and comparison,” Clinical Chemistry, vol. 43, no. 10, pp. 1965–1974, 1997. View at: Google Scholar
  47. L. M. Wakefield, J. J. Letterio, T. Chen et al., “Transforming growth factor-β1 circulates in normal human plasma and is unchanged in advanced metastatic breast cancer,” Clinical Cancer Research, vol. 1, no. 1, pp. 129–136, 1995. View at: Google Scholar
  48. C. Chaigneau, T. Cabioch, K. Beaumont, and F. Betsou, “Serum biobank certification and the establishment of quality controls for biological fluids: examples of serum biomarker stability after temperature variation,” Clinical Chemistry and Laboratory Medicine, vol. 45, no. 10, pp. 1390–1395, 2007. View at: Publisher Site | Google Scholar
  49. L. Zhao, L. Wang, W. Ji, M. Lei, W. Yang, and F.-M. Kong, “The influence of the blood handling process on the measurement of circulating TGF-β1,” European Cytokine Network, vol. 23, no. 1, pp. 1–6, 2012. View at: Publisher Site | Google Scholar
  50. A. M. Halldórsdóttir, J. Stoker, R. Porche-Sorbet, and C. S. Eby, “Soluble CD40 ligand measurement inaccuracies attributable to specimen type, processing time, and ELISA method,” Clinical Chemistry, vol. 51, no. 6, pp. 1054–1057, 2005. View at: Publisher Site | Google Scholar
  51. D. Bereczki, E. Nagy, A. Pál et al., “Should soluble CD40 ligand be measured from serum or plasma samples?” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 23, no. 6, pp. 1129–1131, 2003. View at: Publisher Site | Google Scholar
  52. N. Varo, R. Nuzzo, C. Natal, P. Libby, and U. Schönbeck, “Influence of pre-analytical and analytical factors on soluble CD40L measurements,” Clinical Science, vol. 111, no. 5, pp. 341–347, 2006. View at: Publisher Site | Google Scholar
  53. D. Malone, L. M. Napolitano, T. Genuit, G. V. Bochicchio, K. Kole, and T. M. Scalea, “Total cytokine immunoassay: a more accurate method of cytokine measurement?” The Journal of Trauma, vol. 50, no. 5, pp. 821–825, 2001. View at: Google Scholar
  54. R. E. Banks, “Measurement of cytokines in clinical samples using immunoassays: problems and pitfalls,” Critical Reviews in Clinical Laboratory Sciences, vol. 37, no. 2, pp. 131–182, 2000. View at: Google Scholar
  55. M. Wadhwa and R. Thorpe, “Cytokine immunoassays: recommendations for standardisation, calibration and validation,” Journal of Immunological Methods, vol. 219, no. 1-2, pp. 1–5, 1998. View at: Publisher Site | Google Scholar
  56. G. V. Shurin, Z. R. Yurkovetsky, G. S. Chatta, I. L. Tourkova, M. R. Shurin, and A. E. Lokshin, “Dynamic alteration of soluble serum biomarkers in healthy aging,” Cytokine, vol. 39, no. 2, pp. 123–129, 2007. View at: Publisher Site | Google Scholar
  57. H. Bruunsgaard and B. K. Pedersen, “Age-related inflammatory cytokines and disease,” Immunology and Allergy Clinics of North America, vol. 23, no. 1, pp. 15–39, 2003. View at: Google Scholar
  58. H. Bruunsgaard, M. Pedersen, and B. K. Pedersen, “Aging and proinflammatory cytokines,” Current Opinion in Hematology, vol. 8, no. 3, pp. 131–136, 2001. View at: Publisher Site | Google Scholar
  59. R. J. Forsey, J. M. Thompson, J. Ernerudh et al., “Plasma cytokine profiles in elderly humans,” Mechanisms of Ageing and Development, vol. 124, no. 4, pp. 487–493, 2003. View at: Publisher Site | Google Scholar
  60. K. S. Krabbe, M. Pedersen, and H. Bruunsgaard, “Inflammatory mediators in the elderly,” Experimental Gerontology, vol. 39, no. 5, pp. 687–699, 2004. View at: Publisher Site | Google Scholar
  61. T. Fulop, A. Larbi, N. Douziech, I. Levesque, A. Varin, and G. Herbein, “Cytokine receptor signalling and aging,” Mechanisms of Ageing and Development, vol. 127, no. 6, pp. 526–537, 2006. View at: Publisher Site | Google Scholar
  62. R. Gerli, D. Monti, O. Bistoni et al., “Chemokines, sTNF-Rs and sCD30 serum levels in healthy aged people and centenarians,” Mechanisms of Ageing and Development, vol. 121, no. 1–3, pp. 37–46, 2001. View at: Publisher Site | Google Scholar
  63. J. Myśliwska, E. Bryl, J. Foerster, and A. Myśliwski, “The upregulation of TNFα production is not a generalised phenomenon in the elderly between their sixth and seventh decades of life,” Mechanisms of Ageing and Development, vol. 107, no. 1, pp. 1–14, 1999. View at: Publisher Site | Google Scholar
  64. H. Bruunsgaard, K. Andersen-Ranberg, J. V. B. Hjelmborg, B. K. Pedersen, and B. Jeune, “Elevated levels of tumor necrosis factor alpha and mortality in centenarians,” American Journal of Medicine, vol. 115, no. 4, pp. 278–283, 2003. View at: Publisher Site | Google Scholar
  65. F. Licastro, G. Candore, D. Lio et al., “Innate immunity and inflammation in ageing: a key for understanding age-related diseases,” Immunity and Ageing, vol. 2, article 8, 2005. View at: Publisher Site | Google Scholar
  66. S. Vasto, G. Candore, C. R. Balistreri et al., “Inflammatory networks in ageing, age-related diseases and longevity,” Mechanisms of Ageing and Development, vol. 128, no. 1, pp. 83–91, 2007. View at: Publisher Site | Google Scholar
  67. G. Pawelec, A. Larbi, and E. Derhovanessian, “Senescence of the human immune system,” Journal of Comparative Pathology, vol. 142, no. 1, pp. S39–S44, 2010. View at: Publisher Site | Google Scholar
  68. C. C. Finnerty, M. G. Jeschke, D. N. Herndon et al., “Temporal cytokine profiles in severely burned patients: a comparison of adults and children,” Molecular Medicine, vol. 14, no. 9-10, pp. 553–560, 2008. View at: Publisher Site | Google Scholar
  69. S. A. Ahmed, B. D. Hissong, D. Verthelyi, K. Donner, K. Becker, and E. Karpuzoglu-Sahin, “Gender and risk of autoimmune diseases: possible role of estrogenic compounds,” Environmental Health Perspectives, vol. 107, supplement 5, pp. 681–686, 1999. View at: Google Scholar
  70. A. B. Hahn, J. C. Kasten-Jolly, D. M. Constantino et al., “TNF-α, IL-6, IFN-γ, and IL-10 gene expression polymorphisms and the IL-4 receptor α-chain variant Q576R: effects on renal allograft outcome,” Transplantation, vol. 72, no. 4, pp. 660–665, 2001. View at: Google Scholar
  71. M. J. Eikelenboom, J. Killestein, J. J. Kragt, B. M. J. Uitdehaag, and C. H. Polman, “Gender differences in multiple sclerosis: cytokines and vitamin D,” Journal of the Neurological Sciences, vol. 286, no. 1-2, pp. 40–42, 2009. View at: Publisher Site | Google Scholar
  72. M. Sadeghi, V. Daniel, C. Naujokat, M. Wiesel, O. Hergesell, and G. Opelz, “Strong inflammatory cytokine response in male and strong anti-inflammatory response in female kidney transplant recipients with urinary tract infection,” Transplant International, vol. 18, no. 2, pp. 177–185, 2005. View at: Publisher Site | Google Scholar
  73. P. Pietschmann, E. Gollob, S. Brosch et al., “The effect of age and gender on cytokine production by human peripheral blood mononuclear cells and markers of bone metabolism,” Experimental Gerontology, vol. 38, no. 10, pp. 1119–1127, 2003. View at: Publisher Site | Google Scholar
  74. N. Petrovsky, P. McNair, and L. C. Harrison, “Diurnal rhythms of pro-inflammatory cytokines: regulation by plasma cortisol and therapeutic implications,” Cytokine, vol. 10, no. 4, pp. 307–312, 1998. View at: Publisher Site | Google Scholar
  75. M. Cutolo, B. Seriolo, C. Craviotto, C. Pizzorni, and A. Sulli, “Circadian rhythms in RA,” Annals of the Rheumatic Diseases, vol. 62, no. 7, pp. 593–596, 2003. View at: Publisher Site | Google Scholar
  76. M. Cutolo and R. H. Straub, “Circadian rhythms in arthritis: hormonal effects on the immune/inflammatory reaction,” Autoimmunity Reviews, vol. 7, no. 3, pp. 223–228, 2008. View at: Publisher Site | Google Scholar
  77. P. Lissoni, F. Rovelli, F. Brivio, O. Brivio, and L. Fumagalli, “Circadian secretions of IL-2, IL-12, IL-6 and IL-10 in relation to the light/dark rhythm of the pineal hormone melatonin in healthy humans,” Natural Immunity, vol. 16, no. 1, pp. 1–5, 1998. View at: Publisher Site | Google Scholar
  78. L. S. Knudsen, I. J. Christensen, T. Lottenburger et al., “Pre-analytical and biological variability in circulating interleukin 6 in healthy subjects and patients with Rheumatoid Arthritis,” Biomarkers, vol. 13, no. 1, pp. 59–78, 2008. View at: Publisher Site | Google Scholar
  79. N. G. Arvidson, B. Gudbjornsson, L. Elfman, A.-C. Ryden, T. H. Totterman, and R. Hallgren, “Circadian rhythm of serum interleukin-6 in Rheumatoid Arthritis,” Annals of the Rheumatic Diseases, vol. 53, no. 8, pp. 521–524, 1994. View at: Google Scholar
  80. R. H. Straub and M. Cutolo, “Circadian rhythms in Rheumatoid Arthritis: implications for pathophysiology and therapeutic management,” Arthritis & Rheumatism, vol. 56, no. 2, pp. 399–408, 2007. View at: Publisher Site | Google Scholar
  81. M. Cutolo, B. Villaggio, K. Otsa, O. Aakre, A. Sulli, and B. Seriolo, “Altered circadian rhythms in Rheumatoid Arthritis patients play a role in the disease's symptoms,” Autoimmunity Reviews, vol. 4, no. 8, pp. 497–502, 2005. View at: Publisher Site | Google Scholar
  82. L. J. Crofford, K. T. Kalogeras, G. Mastorakos et al., “Circadian relationships between interleukin (IL)-6 and hypothalamic- pituitary-adrenal axis hormones: failure of IL-6 to cause sustained hypercortisolism in patients with early untreated Rheumatoid Arthritis,” Journal of Clinical Endocrinology & Metabolism, vol. 82, no. 4, pp. 1279–1283, 1997. View at: Publisher Site | Google Scholar
  83. P. Dandona, R. Weinstock, K. Thusu, E. Abdel-Rahman, A. Aljada, and T. Wadden, “Tumor necrosis factor-α in sera of obese patients: fall with weight loss,” Journal of Clinical Endocrinology & Metabolism, vol. 83, no. 8, pp. 2907–2910, 1998. View at: Publisher Site | Google Scholar
  84. P. Blackburn, J.-P. Després, B. Lamarche et al., “Postprandial variations of plasma inflammatory markers in abdominally obese men,” Obesity, vol. 14, no. 10, pp. 1747–1754, 2006. View at: Publisher Site | Google Scholar
  85. C. Payette, P. Blackburn, B. Lamarche et al., “Sex differences in postprandial plasma tumor necrosis factor-α, interleukin-6, and C-reactive protein concentrations,” Metabolism, vol. 58, no. 11, pp. 1593–1601, 2009. View at: Publisher Site | Google Scholar
  86. A. J. Saah and D. R. Hoover, “‘Sensitivity” and “specificity” reconsidered: the meaning of these terms in analytical and diagnostic settings,” Annals of Internal Medicine, vol. 126, no. 1, pp. 91–94, 1997. View at: Google Scholar
  87. N. W. Tietz, C. A. Burtis, and E. R. Ashwood, Tietz Textbook of Clinical Chemistry, WB Saunders, 1994.
  88. K. Esposito, F. Nappo, R. Marfella et al., “Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress,” Circulation, vol. 106, no. 16, pp. 2067–2072, 2002. View at: Publisher Site | Google Scholar
  89. M. F. Carroll and D. S. Schade, “Timing of antioxidant vitamin ingestion alters postprandial proatherogenic serum markers,” Circulation, vol. 108, no. 1, pp. 24–31, 2003. View at: Publisher Site | Google Scholar
  90. X. Zhou, M. S. Fragala, J. E. McElhaney, and G. A. Kuchel, “Conceptual and methodological issues relevant to cytokine and inflammatory marker measurements in clinical research,” Current Opinion in Clinical Nutrition and Metabolic Care, vol. 13, no. 5, pp. 541–547, 2010. View at: Publisher Site | Google Scholar
  91. R. F. Grimble, “Nutritional modulation of cytokine biology,” Nutrition, vol. 14, no. 7-8, pp. 634–640, 1998. View at: Publisher Site | Google Scholar
  92. D. G. Rowbottom and K. J. Green, “Acute exercise effects on the immune system,” Medicine and Science in Sports & Exercise, vol. 32, no. 7, pp. S396–S405, 2000. View at: Google Scholar
  93. K. Ostrowski, T. Rohde, M. Zacho, S. Asp, and B. K. Pedersen, “Evidence that interleukin-6 is produced in human skeletal muscle during prolonged running,” Journal of Physiology, vol. 508, no. 3, pp. 949–953, 1998. View at: Publisher Site | Google Scholar
  94. A. Steensberg, C. Keller, R. L. Starkie, T. Osada, M. A. Febbraio, and B. K. Pedersen, “IL-6 and TNF-α expression in, and release from, contracting human skeletal muscle,” American Journal of Physiology, vol. 283, no. 6, pp. E1272–E1278, 2002. View at: Google Scholar
  95. H. Gabriel and W. Kindermann, “The acute immune response to exercise: what does it mean?” International Journal of Sports Medicine, vol. 18, no. 1, pp. S28–S45, 1997. View at: Google Scholar
  96. D. C. Nieman and B. K. Pedersen, “Exercise and immune function. Recent developments,” Sports Medicine, vol. 27, no. 2, pp. 73–80, 1999. View at: Publisher Site | Google Scholar
  97. B. K. Pedersen, A. Steensberg, C. Fischer et al., “The metabolic role of IL-6 produced during exercise: is IL-6 an exercise factor?” Proceedings of the Nutrition Society, vol. 63, no. 2, pp. 263–267, 2004. View at: Publisher Site | Google Scholar
  98. M. A. Febbraio and B. K. Pedersen, “Muscle-derived interleukin-6: mechanisms for activation and possible biological roles,” The FASEB Journal, vol. 16, no. 11, pp. 1335–1347, 2002. View at: Publisher Site | Google Scholar
  99. M. Penkowa, C. Keller, P. Keller, S. Jauffred, and B. K. Pedersen, “Immunohistochemical detection of interleukin-6 in human skeletal muscle fibers following exercise,” The FASEB Journal, vol. 17, no. 14, pp. 2166–2168, 2003. View at: Google Scholar
  100. B. Baslund, K. Lyngberg, V. Andersen et al., “Effect of 8 wk of bicycle training on the immune system of patients with Rheumatoid Arthritis,” Journal of Applied Physiology, vol. 75, no. 4, pp. 1691–1695, 1993. View at: Google Scholar
  101. N. P. Walsh, M. Gleeson, R. J. Shephard et al., “Position statement. Part one: immune function and exercise,” Exercise Immunology Review, vol. 17, no. 1, pp. 6–63, 2011. View at: Google Scholar
  102. A. N. Vgontzas, E. Zoumakis, E. O. Bixler et al., “Adverse effects of modest sleep restriction on sleepiness, performance, and inflammatory cytokines,” Journal of Clinical Endocrinology & Metabolism, vol. 89, no. 5, pp. 2119–2126, 2004. View at: Publisher Site | Google Scholar
  103. R. A. Barratt, S. L. Bowens, S. K. McCune, J. N. Johannessen, and S. Y. Buckman, “The critical path initiative: leveraging collaborations to enhance regulatory science,” Clinical Pharmacology and Therapeutics, vol. 91, no. 3, pp. 380–383, 2012. View at: Publisher Site | Google Scholar
  104. M. Lekander, S. Elofsson, I.-M. Neve, L.-O. Hansson, and A.-L. Unden, “Self-rated health is related to levels of circulating cytokines,” Psychosomatic Medicine, vol. 66, no. 4, pp. 559–563, 2004. View at: Publisher Site | Google Scholar
  105. D. de Groote, P. F. Zangerle, Y. Gevaert et al., “Direct stimulation of cytokines (IL-1β, TNF-α, IL-6, IL-2, IFN-γ and GM-CSF) in whole blood. I. Comparison with isolated PBMC stimulation,” Cytokine, vol. 4, no. 3, pp. 239–248, 1992. View at: Google Scholar
  106. T. C. Peakman and P. Elliott, “The UK Biobank sample handling and storage validation studies,” International Journal of Epidemiology, vol. 37, supplement 1, pp. i2–i6, 2008. View at: Publisher Site | Google Scholar
  107. J. A. D. Bienvenu, G. Monneret, M. C. Gutowski, and N. Fabien, “Cytokine assays in human sera and tissues,” Toxicology, vol. 129, no. 1, pp. 55–61, 1998. View at: Publisher Site | Google Scholar
  108. S. Barelli, D. Crettaz, L. Thadikkaran, O. Rubin, and J.-D. Tissot, “Plasma/serum proteomics: pre-analytical issues,” Expert Review of Proteomics, vol. 4, no. 3, pp. 363–370, 2007. View at: Publisher Site | Google Scholar
  109. F. S. Hosnijeh, E. J. M. Krop, L. Portengen et al., “Stability and reproducibility of simultaneously detected plasma and serum cytokine levels in asymptomatic subjects,” Biomarkers, vol. 15, no. 2, pp. 140–148, 2010. View at: Publisher Site | Google Scholar
  110. J. G. Cannon, J. W. M. van der Meer, D. Kwiatkowski et al., “Interleukin-1β in human plasma: optimization of blood collection, plasma extraction, and radioimmunoassay methods,” Lymphokine Research, vol. 7, no. 4, pp. 457–467, 1988. View at: Google Scholar
  111. T. L. Whiteside, “Cytokines and cytokine measurements in a clinical laboratory,” Clinical and Diagnostic Laboratory Immunology, vol. 1, no. 3, pp. 257–260, 1994. View at: Google Scholar
  112. J. N. Hoffmann, W. H. Hartl, E. Faist, M. Jochum, and D. Inthorn, “Tumor necrosis factor measurement and use of different anticoagulants: possible interference in plasma samples and supernatants from endotoxin-stimulated monocytes,” Inflammation Research, vol. 46, no. 9, pp. 342–347, 1997. View at: Publisher Site | Google Scholar
  113. N. Aziz, M. R. Irwin, S. S. Dickerson, and A. W. Butch, “Spurious tumor necrosis factor-α and interleukin-6 production by human monocytes from blood collected in endotoxin-contaminated vacutainer blood collection tubes,” Clinical Chemistry, vol. 50, no. 11, pp. 2215–2216, 2004. View at: Publisher Site | Google Scholar
  114. J. J. Deeks, “Systematic reviews in health care: systematic reviews of evaluations of diagnostic and screening tests,” British Medical Journal, vol. 323, no. 7305, pp. 157–162, 2001. View at: Google Scholar
  115. S. Gilbertson-White, B. E. Aouizerat, and C. Miaskowski, “Methodologic issues in the measurement of cytokines to elucidate the biological basis for cancer symptoms,” Biological Research for Nursing, vol. 13, no. 1, pp. 15–24, 2011. View at: Publisher Site | Google Scholar
  116. S. S. Tworoger and S. E. Hankinson, “Collection, processing, and storage of biological samples in epidemiologic studies: sex hormones, carotenoids, inflammatory markers, and proteomics as examples,” Cancer Epidemiology Biomarkers & Prevention, vol. 15, no. 9, pp. 1578–1581, 2006. View at: Publisher Site | Google Scholar
  117. A. Czlonkowska, A. Ciesielska, G. Gromadzka, and I. Kurkowska-Jastrzebska, “Estrogen and cytokines production—the possible cause of gender differences in neurological diseases,” Current Pharmaceutical Design, vol. 11, no. 8, pp. 1017–1030, 2005. View at: Publisher Site | Google Scholar
  118. B. B. Haab, B. H. Geierstanger, G. Michailidis et al., “Immunoassay and antibody microarray analysis of the HUPO Plasma Proteome Project reference specimens: systematic variation between sample types and calibration of mass spectrometry data,” Proteomics, vol. 5, no. 13, pp. 3278–3291, 2005. View at: Publisher Site | Google Scholar
  119. R. A. R. Bowen, Y. Chan, J. Cohen et al., “Effect of blood collection tubes on total triiodothyronine and other laboratory assays,” Clinical Chemistry, vol. 51, no. 2, pp. 424–433, 2005. View at: Publisher Site | Google Scholar
  120. X. Zhao, F. Qureshi, P. S. Eastman et al., “Pre-analytical effects of blood sampling and handling in quantitative immunoassays for Rheumatoid Arthritis,” Journal of Immunological Methods, vol. 378, no. 1-2, pp. 72–80, 2012. View at: Publisher Site | Google Scholar
  121. G. Panicker, K. S. Meadows, D. R. Lee, R. Nisenbaum, and E. R. Unger, “Effect of storage temperatures on the stability of cytokines in cervical mucous,” Cytokine, vol. 37, no. 2, pp. 176–179, 2007. View at: Publisher Site | Google Scholar
  122. W. de Jager, K. Bourcier, G. T. Rijkers, B. J. Prakken, and V. Seyfert-Margolis, “Prerequisites for cytokine measurements in clinical trials with multiplex immunoassays,” BMC Immunology, vol. 10, article 52, 2009. View at: Publisher Site | Google Scholar
  123. M. W. van der Linden, T. W. J. Huizinga, D.-J. Stoeken, A. Sturk, and R. G. J. Westendorp, “Determination of tumour necrosis factor-α and interleukin-10 production in a whole blood stimulation system: assessment of laboratory error and individual variation,” Journal of Immunological Methods, vol. 218, no. 1-2, pp. 63–71, 1998. View at: Publisher Site | Google Scholar
  124. C. A. Ray, C. Dumaual, M. Willey et al., “Optimization of analytical and pre-analytical variables associated with an ex vivo cytokine secretion assay,” Journal of Pharmaceutical and Biomedical Analysis, vol. 41, no. 1, pp. 189–195, 2006. View at: Publisher Site | Google Scholar
  125. G. Leroux-Roels, F. Offner, J. Philippe, and A. Vermeulen, “Influence of blood-collecting systems on concentrations of tumor necrosis factor in serum and plasma,” Clinical Chemistry, vol. 34, no. 11, pp. 2373–2374, 1988. View at: Google Scholar
  126. R. Thorpe, M. Wadhwa, C. R. Bird, and A. R. Mire-Sluis, “Detection and measurement of cytokines,” Blood Reviews, vol. 6, no. 3, pp. 133–148, 1992. View at: Publisher Site | Google Scholar
  127. F.-C. Lin, R. Cohen, R. Losada, and V. Bush, “Cellular sedimentation and barrier formation under centrifugal force in blood collection tubes,” Laboratory Medicine, vol. 32, no. 10, pp. 588–594, 2001. View at: Publisher Site | Google Scholar
  128. N. Aziz, P. Nishanian, J. M. G. Taylor et al., “Stability of plasma levels of cytokines and soluble activation markers in patients with human immunodeficiency virus infection,” Journal of Infectious Diseases, vol. 179, no. 4, pp. 843–848, 1999. View at: Publisher Site | Google Scholar
  129. A. J. Rai and F. Vitzthum, “Effects of preanalytical variables on peptide and protein measurements in human serum and plasma: implications for clinical proteomics,” Expert Review of Proteomics, vol. 3, no. 4, pp. 409–426, 2006. View at: Publisher Site | Google Scholar
  130. A. R. Mire-Sluis, A. Padilla, and R. G. Das, “Biological standardization of cytokines and growth factors,” Developments in Biological Standardization, vol. 97, pp. 171–176, 1999. View at: Google Scholar
  131. N. Sachdeva and D. Asthana, “Cytokine quantitation: technologies and applications,” Frontiers in Bioscience, vol. 12, no. 12, pp. 4682–4695, 2007. View at: Publisher Site | Google Scholar
  132. P. Aukrust, F. Müller, E. Lien et al., “Tumor Necrosis Factor (TNF) system levels in human immunodeficiency virus-infected patients during highly active antiretroviral therapy: persistent TNF activation is associated with virologic and immunologic treatment failure,” Journal of Infectious Diseases, vol. 179, no. 1, pp. 74–82, 1999. View at: Publisher Site | Google Scholar
  133. P. Aukrust, S. S. Frøland, N.-B. Liabakk et al., “Release of cytokines, soluble cytokine receptors, and interleukin-1 receptor antagonist after intravenous immunoglobulin administration in vivo,” Blood, vol. 84, no. 7, pp. 2136–2143, 1994. View at: Google Scholar
  134. I. Engelberts, S. Stephens, G. J. M. Francot, C. J. van der Linden, and W. A. Buurman, “Evidence for different effects of soluble TNF-receptors on various TNF measurements in human biological fluids,” The Lancet, vol. 338, no. 8765, pp. 515–516, 1991. View at: Google Scholar
  135. G. Toedter, K. Hayden, C. Wagner, and C. Brodmerkel, “Simultaneous detection of eight analytes in human serum by two commercially available platforms for multiplex cytokine analysis,” Clinical and Vaccine Immunology, vol. 15, no. 1, pp. 42–48, 2008. View at: Publisher Site | Google Scholar
  136. C. H. Chau, O. Rixe, H. McLeod, and W. D. Figg, “Validation of analytic methods for biomarkers used in drug development,” Clinical Cancer Research, vol. 14, no. 19, pp. 5967–5976, 2008. View at: Publisher Site | Google Scholar
  137. E. Morgan, R. Varro, H. Sepulveda et al., “Cytometric bead array: a multiplexed assay platform with applications in various areas of biology,” Clinical Immunology, vol. 110, no. 3, pp. 252–266, 2004. View at: Publisher Site | Google Scholar
  138. D. Dabitao, J. B. Margolick, J. Lopez, and J. H. Bream, “Multiplex measurement of proinflammatory cytokines in human serum: comparison of the Meso Scale Discovery electrochemiluminescence assay and the Cytometric Bead Array,” Journal of Immunological Methods, vol. 372, no. 1-2, pp. 71–77, 2011. View at: Publisher Site | Google Scholar
  139. P. Roux-Lombard, G. Steiner, J.-M. Dayer et al., “Preliminary report on cytokine determination in human synovial fluids: a consensus study of the European Workshop for Rheumatology Research,” Clinical and Experimental Rheumatology, vol. 10, no. 5, pp. 515–520, 1992. View at: Google Scholar
  140. J. Bienvenu, G. Monneret, N. Fabien, and J. P. Revillard, “The clinical usefulness of the measurement of cytokines,” Clinical Chemistry and Laboratory Medicine, vol. 38, no. 4, pp. 267–285, 2000. View at: Google Scholar
  141. N. Madry, B. Auerbach, and C. Schelp, “Measures to overcome HAMA interferences in immunoassays,” Anticancer Research, vol. 17, no. 4, pp. 2883–2886, 1997. View at: Google Scholar
  142. G. Dimeski, “Interference testing,” The Clinical Biochemist Reviews, vol. 29, supplement 1, pp. S43–S48, 2008. View at: Google Scholar
  143. M. Wadhwa, M. J. Seghatchian, P. Dilger et al., “Cytokines in WBC-reduced apheresis pcs during storage: a comparison of two WBC-reduction methods,” Transfusion, vol. 40, no. 9, pp. 1118–1126, 2000. View at: Publisher Site | Google Scholar
  144. M. Svenson, M. B. Hansen, L. Kayser, Å. K. Rasmussen, C. M. Reimert, and K. Bendtzen, “Effects of human anti-IL-1α autoantibodies on receptor binding and biological activities of IL-1,” Cytokine, vol. 4, no. 2, pp. 125–133, 1992. View at: Google Scholar
  145. M. B. Hansen, M. Svenson, K. Abell et al., “Sex- and age-dependency of IgG auto-antibodies against IL-1α in healthy humans,” European Journal of Clinical Investigation, vol. 24, no. 3, pp. 212–218, 1994. View at: Google Scholar
  146. K. Bendtzen, M. B. Hansen, M. Diamant, C. Ross, and M. Svenson, “Naturally occurring autoantibodies to interleukin-1α, interleukin-6, interleukin-10, and interferon-α,” Journal of Interferon Research, vol. 14, no. 4, pp. 157–158, 1994. View at: Google Scholar
  147. R. P. Revoltella, “Natural and therapeutically-induced antibodies to cytokines,” Biotherapy, vol. 10, no. 4, pp. 321–331, 1998. View at: Google Scholar
  148. M. B. Hansen, V. Andersen, K. Rohde et al., “Cytokine autoantibodies in Rheumatoid Arthritis,” Scandinavian Journal of Rheumatology, vol. 24, no. 4, pp. 197–203, 1995. View at: Google Scholar
  149. K. Chapman, “The ProteinChip Biomarker System from Ciphergen Biosystems: a novel proteomics platform for rapid biomarker discovery and validation,” Biochemical Society Transactions, vol. 30, no. 2, pp. 82–87, 2002. View at: Google Scholar
  150. C. M. Preissner, L. A. Dodge, D. J. O'Kane, R. J. Singh, and S. K. G. Grebe, “Prevalence of heterophilic antibody interference in eight automated tumor marker immunoassays,” Clinical Chemistry, vol. 51, no. 1, pp. 208–210, 2005. View at: Publisher Site | Google Scholar
  151. A. Bonetti, C. Monica, C. Bonaguri et al., “Interference by heterophilic antibodies in immunoassays: wrong increase of myoglobin values,” Acta Biomedica de l'Ateneo Parmense, vol. 79, no. 2, pp. 140–143, 2008. View at: Google Scholar
  152. S. S. Levinson and J. J. Miller, “Towards a better understanding of heterophile (and the like) antibody interference with modern immunoassays,” Clinica Chimica Acta, vol. 325, no. 1-2, pp. 1–15, 2002. View at: Publisher Site | Google Scholar
  153. N. Bolstad, D. J. Warren, J. Bjerner et al., “Heterophilic antibody interference in commercial immunoassays; a screening study using paired native and pre-blocked sera,” Clinical Chemistry and Laboratory Medicine, vol. 49, no. 12, pp. 2001–2006, 2012. View at: Publisher Site | Google Scholar
  154. W. Muller, R. Mierau, and D. Wohltmann, “Interference of IgM rheumatoid factor with nephelometric C-reactive protein determinations,” Journal of Immunological Methods, vol. 80, no. 1, pp. 77–90, 1985. View at: Publisher Site | Google Scholar
  155. K. Raza, F. Falciani, S. J. Curnow et al., “Early Rheumatoid Arthritis is characterized by a distinct and transient synovial fluid cytokine profile of T cell and stromal cell origin,” Arthritis Research & Therapy, vol. 7, no. 4, pp. R784–R795, 2005. View at: Google Scholar
  156. H. C. Vaidya and B. G. Beatty, “Eliminating interference from heterophilic antibodies in a two-site immunoassay for creatine kinase MB by using F(ab′)2 conjugate and polyclonal mouse IgG,” Clinical Chemistry, vol. 38, no. 9, pp. 1737–1742, 1992. View at: Google Scholar
  157. D. J. Todd, N. Knowlton, M. Amato et al., “Erroneous augmentation of multiplex assay measurements in patients with Rheumatoid Arthritis due to heterophilic binding by serum rheumatoid factor,” Arthritis & Rheumatism, vol. 63, no. 4, pp. 894–903, 2011. View at: Publisher Site | Google Scholar
  158. P. E. Chandra, J. Sokolove, B. G. Hipp et al., “Novel multiplex technology for diagnostic characterization of Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 13, no. 3, article R102, 2011. View at: Publisher Site | Google Scholar
  159. A. Ledur, C. Fitting, B. David, C. Hamberger, and J.-M. Cavaillon, “Variable estimates of cytokine levels produced by commercial ELISA kits: results using international cytokine standards,” Journal of Immunological Methods, vol. 186, no. 2, pp. 171–179, 1995. View at: Publisher Site | Google Scholar
  160. A. R. Mire-Sluis, R. G. Das, and R. Thorpe, “The international standard for Granulocyte Colony Stimulating Factor (G-CSF). Evaluation in an international collaborative study,” Journal of Immunological Methods, vol. 179, no. 1, pp. 117–126, 1995. View at: Publisher Site | Google Scholar
  161. A. R. Mire-Sluis, R. G. Das, and R. Thorpe, “The international standard for Macrophage Colony Stimulating Factor (M-CSF). Evaluation in an international collaborative study,” Journal of Immunological Methods, vol. 179, no. 2, pp. 141–151, 1995. View at: Publisher Site | Google Scholar
  162. R. E. G. Das and S. Poole, “The international standard for interleukin-6. Evaluation in an international collaborative study,” Journal of Immunological Methods, vol. 160, no. 2, pp. 147–153, 1993. View at: Publisher Site | Google Scholar
  163. A. R. Mire-Sluis, R. G. Das, and R. Thorpe, “Implications for the assay and biological activity of interleukin-4: results of a WHO international collaborative study,” Journal of Immunological Methods, vol. 194, no. 1, pp. 13–25, 1996. View at: Publisher Site | Google Scholar
  164. A. R. Mire-Sluis, R. G. Das, and R. Thorpe, “Implications for the assay and biological activity of interleukin-8: results of a WHO international collaborative study,” Journal of Immunological Methods, vol. 200, no. 1-2, pp. 1–16, 1997. View at: Publisher Site | Google Scholar
  165. A. R. Mire-Sluis, R. Gaines-Das, and R. Thorpe, “Immunoassays for detecting cytokines: what are they really measuring?” Journal of Immunological Methods, vol. 186, no. 2, pp. 157–160, 1995. View at: Publisher Site | Google Scholar
  166. S. Poole and R. E. G. Das, “The international standards for interleukin-1α and interleukin-1β. Evaluation in an international collaborative study,” Journal of Immunological Methods, vol. 142, no. 1, pp. 1–13, 1991. View at: Publisher Site | Google Scholar
  167. S. Romagnani, G. del Prete, R. Manetti et al., “Role of TH1/TH2 cytokines in HIV infection,” Immunological Reviews, no. 140, pp. 73–92, 1994. View at: Google Scholar
  168. N. Aziz, P. Nishanian, and J. L. Fahey, “Levels of cytokines and immune activation markers in plasma in human immunodeficiency virus infection: quality control procedures,” Clinical and Diagnostic Laboratory Immunology, vol. 5, no. 6, pp. 755–761, 1998. View at: Google Scholar
  169. A. Meager, “Measurement of cytokines by bioassays: theory and application,” Methods, vol. 38, no. 4, pp. 237–252, 2006. View at: Publisher Site | Google Scholar
  170. R. V. House, “Cytokine measurement techniques for assessing hypersensitivity,” Toxicology, vol. 158, no. 1-2, pp. 51–58, 2001. View at: Publisher Site | Google Scholar
  171. K. J. Miller, R. R. Bowsher, A. Celniker et al., “Workshop on bioanalytical methods validation for macromolecules: summary report,” Pharmaceutical Research, vol. 18, no. 9, pp. 1373–1383, 2001. View at: Publisher Site | Google Scholar
  172. J. W. Lee, “Method validation and application of protein biomarkers: basic similarities and differences from biotherapeutics,” Bioanalysis, vol. 1, no. 8, pp. 1461–1474, 2009. View at: Google Scholar
  173. J. Smolec, B. DeSilva, W. Smith et al., “Bioanalytical method validation for macromolecules in support of pharmacokinetic studies,” Pharmaceutical Research, vol. 22, no. 9, pp. 1425–1431, 2005. View at: Publisher Site | Google Scholar
  174. M. L. Hetland, I. J. Christensen, T. Lottenburger et al., “Circulating VEGF as a biological marker in patients with Rheumatoid Arthritis? Preanalytical and biological variability in healthy persons and in patients,” Disease Markers, vol. 24, no. 1, pp. 1–10, 2008. View at: Google Scholar
  175. M. Chabaud, F. Fossiez, J.-L. Taupin, and P. Miossec, “Enhancing effect of IL-17 on IL-1-induced IL-6 and leukemia inhibitory factor production by Rheumatoid Arthritis synoviocytes and its regulation by Th2 cytokines,” Journal of Immunology, vol. 161, no. 1, pp. 409–414, 1998. View at: Google Scholar
  176. P. Miossec, “Anti-inflammatory properties of interleukin-4,” Revue du Rhumatisme, vol. 60, no. 2, pp. 87–91, 1993. View at: Google Scholar
  177. F. Ponchel, R. J. Verburg, S. J. Bingham et al., “IL-7 deficiency and therapy-induced lymphopenia in Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 7, no. 1, pp. R82–R92, 2005. View at: Publisher Site | Google Scholar
  178. S. M. Churchman and F. Ponchel, “Interleukin-7 in Rheumatoid Arthritis,” Rheumatology, vol. 47, no. 6, pp. 753–759, 2008. View at: Publisher Site | Google Scholar
  179. V. Goëb, P. Aegerter, R. Parmar et al., “Progression to Rheumatoid Arthritis in early inflammatory arthritis is associated with low IL-7 serum levels,” Annals of the Rheumatic Diseases, vol. 72, no. 6, pp. 1032–1036, 2013. View at: Publisher Site | Google Scholar
  180. S. Faucher, A. M. Crawley, W. Decker et al., “Development of a quantitative bead capture assay for soluble IL-7 receptor alpha in human plasma,” PLoS ONE, vol. 4, no. 8, Article ID e6690, 2009. View at: Publisher Site | Google Scholar
  181. V. Badot, P. Durez, B. J. van den Eynde, A. Nzeusseu-Toukap, F. A. Houssiau, and B. R. Lauwerys, “Rheumatoid Arthritis synovial fibroblasts produce a soluble form of the interleukin-7 receptor in response to pro-inflammatory cytokines,” Journal of Cellular and Molecular Medicine, vol. 15, no. 11, pp. 2335–2342, 2011. View at: Publisher Site | Google Scholar
  182. J. Scheller and S. Rose-John, “Interleukin-6 and its receptor: from bench to bedside,” Medical Microbiology and Immunology, vol. 195, no. 4, pp. 173–183, 2006. View at: Publisher Site | Google Scholar
  183. S. Fabre, C. Guisset, L. Tatem et al., “Protein biochip array technology to monitor rituximab in Rheumatoid Arthritis,” Clinical and Experimental Immunology, vol. 155, no. 3, pp. 395–402, 2009. View at: Publisher Site | Google Scholar
  184. M. M. Maurice, H. Nakamura, E. A. M. van der Voort et al., “Evidence for the role of an altered redox state in hyporesponsiveness of synovial T cells in Rheumatoid Arthritis,” Journal of Immunology, vol. 158, no. 3, pp. 1458–1465, 1997. View at: Google Scholar
  185. A. P. Cope, R. S. Liblau, X.-D. Yang et al., “Chronic tumor necrosis factor alters T cell responses by attenuating T cell receptor signaling,” Journal of Experimental Medicine, vol. 185, no. 9, pp. 1573–1584, 1997. View at: Publisher Site | Google Scholar
  186. P. H. J. Remans, S. I. Gringhuis, J. M. van Laar et al., “Rap1 signaling is required for suppression of Ras-generated reactive oxygen species and protection against oxidative stress in T lymphocytes,” Journal of Immunology, vol. 173, no. 2, pp. 920–931, 2004. View at: Google Scholar
  187. P. H. J. Remans, C. A. Wijbrandts, M. E. Sanders et al., “CTLA-4Ig suppresses reactive oxygen species by preventing synovial adherent cell-induced inactivation of Rap1, a Ras family GTPase mediator of oxidative stress in Rheumatoid Arthritis T cells,” Arthritis & Rheumatism, vol. 54, no. 10, pp. 3135–3143, 2006. View at: Publisher Site | Google Scholar
  188. J. R. F. Abreu, S. Krausz, W. Dontje et al., “Sustained T cell Rap1 signaling is protective in the collagen-induced arthritis model of Rheumatoid Arthritis,” Arthritis & Rheumatism, vol. 62, no. 11, pp. 3289–3299, 2010. View at: Publisher Site | Google Scholar
  189. F. Ponchel, A. W. Morgan, S. J. Bingham et al., “Dysregulated lymphocyte proliferation and differentiation in patients with Rheumatoid Arthritis,” Blood, vol. 100, no. 13, pp. 4550–4556, 2002. View at: Publisher Site | Google Scholar
  190. D. Unutmaz, P. Pileri, and S. Abrignani, “Antigen-independent activation of naive and memory resting T cells by a cytokine combination,” Journal of Experimental Medicine, vol. 180, no. 3, pp. 1159–1164, 1994. View at: Publisher Site | Google Scholar
  191. D. Unutmaz, F. Baldoni, and S. Abrignani, “Human naive T cells activated by cytokines differentiate into a split phenotype with functional features intermediate between naive and memory T cells,” International Immunology, vol. 7, no. 9, pp. 1417–1424, 1995. View at: Google Scholar
  192. F. M. Brennan, A. L. Hayes, C. J. Ciesielski, P. Green, B. M. Foxwell, and M. Feldmann, “Evidence that Rheumatoid Arthritis synovial T cells are similar to cytokine-activated T cells—involvement of phosphatidylinositol 3-kinase and nuclear factor kappa B pathways in tumor necrosis factor a production in Rheumatoid Arthritis,” Arthritis & Rheumatism, vol. 46, no. 1, pp. 31–41, 2002. View at: Google Scholar
  193. J. T. Beech, E. Andreakos, C. J. Ciesielski, P. Green, B. M. J. Foxwell, and F. M. Brennan, “T-cell contact-dependent regulation of CC and CXC chemokine production in monocytes through differential involvement of NFκB: implications for Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 8, no. 6, article R168, 2006. View at: Publisher Site | Google Scholar
  194. L. Altomonte, A. Zoli, L. Mirone, P. Scolieri, and M. Magaro, “Serum levels of interleukin-1b, tumour necrosis factor-a and interleukin-2 in Rheumatoid Arthritis. Correlation with disease activity,” Clinical Rheumatology, vol. 11, no. 2, pp. 202–205, 1992. View at: Publisher Site | Google Scholar
  195. C. H. Burgoyne, S. L. Field, A. K. Brown et al., “Abnormal T-cell differentiation persists in Rheumatoid Arthritis patients in clinical remission and predicts relapse,” Annals of the Rheumatic Diseases, vol. 67, no. 6, pp. 750–757, 2008. View at: Publisher Site | Google Scholar
  196. S. M. Churchman and F. Ponchel, “Interleukin-7 in Rheumatoid Arthritis,” Rheumatology, vol. 47, no. 6, pp. 753–759, 2008. View at: Publisher Site | Google Scholar
  197. S. Harada, M. Yamamura, H. Okamoto et al., “Production of interleukin-7 and interleukin-15 by fibroblast- like synoviocytes from patients with Rheumatoid Arthritis,” Arthritis & Rheumatism, vol. 42, no. 7, pp. 1508–1516, 1999. View at: Google Scholar
  198. S. Blaschke, H. Schulz, G. Schwarz, V. Blaschke, G. A. Müller, and M. Reuss-Borst, “Interleukin 16 expression in relation to disease activity in Rheumatoid Arthritis,” Journal of Rheumatology, vol. 28, no. 1, pp. 12–21, 2001. View at: Google Scholar
  199. H. Amital, V. Barak, R. E. Winkler, and A. Rubinow, “Impact of treatment with infliximab on serum cytokine profile of patients with rheumatoid and psoriatic arthritis,” Annals of the New York Academy of Sciences, vol. 1110, no. 1, pp. 649–660, 2007. View at: Publisher Site | Google Scholar
  200. L. A. B. Joosten, T. R. D. Radstake, E. Lubberts et al., “Association of interleukin-18 expression with enhanced levels of both interleukin-1β and tumor necrosis factor α in knee synovial tissue of patients with Rheumatoid Arthritis,” Arthritis & Rheumatism, vol. 48, no. 2, pp. 339–347, 2003. View at: Publisher Site | Google Scholar
  201. E. A. Gouda, A. A. Aboulata, A. S. Elharoun et al., “Interleukin-18 expression in rheumatoid artheritis synovial tissue and its relation to disease activity,” The Egyptian Journal of Immunology, vol. 14, no. 2, pp. 1–10, 2007. View at: Google Scholar
  202. X. Niu, D. He, X. Zhang et al., “IL-21 regulates Th17 cells in Rheumatoid Arthritis,” Human Immunology, vol. 71, no. 4, pp. 334–341, 2010. View at: Publisher Site | Google Scholar
  203. O. Elkayam, I. Yaron, I. Shirazi, M. Yaron, and D. Caspi, “Serum levels of IL-10, IL-6, IL-1ra, and sIL-2R in patients with psoriatic arthritis,” Rheumatology International, vol. 19, no. 3, pp. 101–105, 2000. View at: Google Scholar
  204. S. E. Jacob, M. Nassiri, F. A. Kerdel, and V. Vincek, “Simultaneous measurement of multiple Th1 and Th2 serum cytokines in psoriasis and correlation with disease severity,” Mediators of Inflammation, vol. 12, no. 5, pp. 309–313, 2003. View at: Publisher Site | Google Scholar
  205. O. Arican, M. Aral, S. Sasmaz, and P. Ciragil, “Serum levels of TNF-α, IFN-γ, IL-6, IL-8, IL-12, IL-17, and IL-18 in patients with active psoriasis and correlation with disease severity,” Mediators of Inflammation, vol. 2005, no. 5, pp. 273–279, 2005. View at: Publisher Site | Google Scholar
  206. E. V. Lourenço and A. la Cava, “Cytokines in systemic lupus erythematosus,” Current Molecular Medicine, vol. 9, no. 3, pp. 242–254, 2009. View at: Publisher Site | Google Scholar
  207. K. N. Lai and D. Y. H. Yap, “Cytokines and their roles in the pathogenesis of systemic lupus erythematosus: from basics to recent advances,” Journal of Biomedicine and Biotechnology, vol. 2010, Article ID 365083, 10 pages, 2010. View at: Publisher Site | Google Scholar
  208. M. Vazquez-del Mercado, A. Garcia-Gonzalez, J. F. Muñoz-Valle et al., “Interleukin 1β (IL-Lβ), IL-10, tumor necrosis factor-α, and cellular proliferation index in peripheral blood mononuclear cells in patients with ankylosing spondylitis,” Journal of Rheumatology, vol. 29, no. 3, pp. 522–526, 2002. View at: Google Scholar
  209. J. Gratacos, A. Collado, X. Filella et al., “Serum cytokines (IL-6, TNF-α, IL-β and IFN-γ) in ankylosing spondylitis: a close correlation between serum IL-6 and disease activity and severity,” British Journal of Rheumatology, vol. 33, no. 10, pp. 927–931, 1994. View at: Google Scholar
  210. A. Bal, E. Unlu, G. Bahar, E. Aydog, E. Eksioglu, and R. Yorgancioglu, “Comparison of serum IL-1β, sIL-2R, IL-6, and TNF-α levels with disease activity parameters in ankylosing spondylitis,” Clinical Rheumatology, vol. 26, no. 2, pp. 211–215, 2007. View at: Publisher Site | Google Scholar
  211. B. W. Needleman, F. M. Wigley, and R. W. Stair, “Interleukin-1, interleukin-2, interleukin-4, interleukin-6, tumor necrosis factor α, and interferon-γ levels in sera from patients with scleroderma,” Arthritis & Rheumatism, vol. 35, no. 1, pp. 67–72, 1992. View at: Google Scholar
  212. M. Hasegawa and K. Takehara, “Potential immunologic targets for treating fibrosis in systemic sclerosis: a review focused on leukocytes and cytokines,” Seminars in Arthritis and Rheumatism, vol. 42, no. 3, pp. 281–296, 2012. View at: Publisher Site | Google Scholar
  213. A. L. M. A. Jansen, I. E. van der Horst-Bruinsma, D. van Schaardenburg, R. J. van de Stadt, M. H. M. T. de Koning, and B. A. C. Dijkmans, “Rheumatoid factor and antibodies to cyclic citrullinated peptide differentiate Rheumatoid Arthritis from undifferentiated polyarthritis in patients with early arthritis,” Journal of Rheumatology, vol. 29, no. 10, pp. 2074–2076, 2002. View at: Google Scholar
  214. V. Goëb, F. Jouen, D. Gilbert, X. le Loët, F. Tron, and O. Vittecoq, “Diagnostic and prognostic usefulness of antibodies to citrullinated peptides,” Joint Bone Spine, vol. 76, no. 4, pp. 343–349, 2009. View at: Publisher Site | Google Scholar
  215. P. E. Barker, “Cancer biomarker validation: standards and process—roles for the National Institute of Standards and Technology (NIST),” Annals of the New York Academy of Sciences, vol. 983, pp. 142–150, 2003. View at: Google Scholar
  216. J. D. Cañete, J. Llena, A. Collado et al., “Comparative cytokine gene expression in synovial tissue of early Rheumatoid Arthritis and seronegative spondyloarthropathies,” British Journal of Rheumatology, vol. 36, no. 1, pp. 38–42, 1997. View at: Google Scholar
  217. M. D. S. Hoy, J. L. Williams, and B. W. Kirkham, “Symmetrical synovial fluid cell cytokine messenger RNA expression in Rheumatoid Arthritis: analysis by reverse transcription/polymerase chain reaction,” British Journal of Rheumatology, vol. 36, no. 2, pp. 170–173, 1997. View at: Google Scholar
  218. K. Raza, F. Falciani, S. J. Curnow et al., “Early Rheumatoid Arthritis is characterized by a distinct and transient synovial fluid cytokine profile of T cell and stromal cell origin,” Arthritis Research & Therapy, vol. 7, no. 4, pp. R784–R795, 2005. View at: Google Scholar
  219. H. Kokkonen, I. Söderström, J. Rocklöv, G. Hallmans, K. Lejon, and S. R. Dahlqvist, “Up-regulation of cytokines and chemokines predates the onset of Rheumatoid Arthritis,” Arthritis & Rheumatism, vol. 62, no. 2, pp. 383–391, 2010. View at: Publisher Site | Google Scholar
  220. K. T. Jørgensen, A. Wiik, M. Pedersen et al., “Cytokines, autoantibodies and viral antibodies in premorbid and postdiagnostic sera from patients with Rheumatoid Arthritis: case-control study nested in a cohort of Norwegian blood donors,” Annals of the Rheumatic Diseases, vol. 67, no. 6, pp. 860–866, 2008. View at: Publisher Site | Google Scholar
  221. T. J. Fry and C. L. Mackall, “Interleukin-7: master regulator of peripheral T-cell homeostasis?” Trends in Immunology, vol. 22, no. 10, pp. 564–571, 2001. View at: Publisher Site | Google Scholar
  222. F. Ponchel, R. J. Cuthbert, and V. Goëb, “IL-7 and lymphopenia,” Clinica Chimica Acta, vol. 412, no. 1-2, pp. 7–16, 2011. View at: Publisher Site | Google Scholar
  223. M. Natsumeda, K. Nishiya, and Z. Ota, “Stimulation by interleukin-7 of mononuclear cells in peripheral blood, synovial fluid and synovial tissue from patients with Rheumatoid Arthritis,” Acta Medica Okayama, vol. 47, no. 6, pp. 391–397, 1993. View at: Google Scholar
  224. J. A. G. van Roon, M. C. Verweij, M. W.-V. Wijk, K. M. G. Jacobs, J. W. J. Bijlsma, and F. P. J. G. Lafeber, “Increased intraarticular interleukin-7 in Rheumatoid Arthritis patients stimulates cell contact-depedent activation of CD4+ T cells and macrophages,” Arthritis & Rheumatism, vol. 52, no. 6, pp. 1700–1710, 2005. View at: Publisher Site | Google Scholar
  225. T. Makino, S. Fukushima, S. Wakasugi, and H. Ihn, “Decreased serum IL-7 levels in patients with systemic sclerosis,” Clinical and Experimental Rheumatology, vol. 27, no. 3, pp. S68–S69, 2009. View at: Google Scholar
  226. J. C. Andreu-Ballester, J. Pérez-Griera, C. Garcia-Ballesteros et al., “Deficit of interleukin-7 in serum of patients with Crohn's disease,” Inflammatory Bowel Diseases, vol. 19, no. 2, pp. E30–E31, 2013. View at: Publisher Site | Google Scholar
  227. H. A. Kader, V. T. Tchernev, E. Satyaraj et al., “Protein microarray analysis of disease activity in pediatric inflammatory bowel disease demonstrates elevated serum PLGF, IL-7, TGF-β1 and IL-12p40 levels in Crohn's disease and ulcerative colitis patients in remission versus active disease,” The American Journal of Gastroenterology, vol. 100, no. 2, pp. 414–423, 2005. View at: Publisher Site | Google Scholar
  228. H. L. Wright, R. C. Bucknall, R. J. Moots, and S. W. Edwards, “Analysis of SF and plasma cytokines provides insights into the mechanisms of inflammatory arthritis and may predict response to therapy,” Rheumatology, vol. 51, no. 3, Article ID ker338, pp. 451–459, 2012. View at: Publisher Site | Google Scholar
  229. S. Das, E. M. Vital, S. Horton et al., “Abatacept or tocilizumab after rituximab in Rheumatoid Arthritis? An exploratory study suggests non-response to rituximab is associated with persistently high IL-6 and better clinical response to IL-6 blocking therapy,” Annals of the Rheumatic Diseases, 2014. View at: Publisher Site | Google Scholar
  230. C. A. Wijbrandts, M. G. W. Dijkgraaf, M. C. Kraan et al., “The clinical response to infliximab in Rheumatoid Arthritis is in part dependent on pretreatment tumour necrosis factor α expression in the synovium,” Annals of the Rheumatic Diseases, vol. 67, no. 8, pp. 1139–1144, 2008. View at: Publisher Site | Google Scholar
  231. J. Lindberg, E. af Klint, A. I. Catrina et al., “Effect of infliximab on mRNA expression profiles in synovial tissue of Rheumatoid Arthritis patients,” Arthritis Research & Therapy, vol. 8, article R179, 2006. View at: Publisher Site | Google Scholar
  232. T. C. M. van der Pouw Kraan, C. A. Wijbrandts, L. G. van Baarsen et al., “Responsiveness to anti-tumour necrosis factor α therapy is related to pre-treatment tissue inflammation levels in Rheumatoid Arthritis patients,” Annals of the Rheumatic Diseases, vol. 67, no. 4, pp. 563–566, 2008. View at: Publisher Site | Google Scholar
  233. H. Marotte, W. Maslinski, and P. Miossec, “Circulating tumour necrosis factor-alpha bioactivity in Rheumatoid Arthritis patients treated with infliximab: link to clinical response,” Arthritis Research & Therapy, vol. 7, no. 1, pp. R149–R155, 2005. View at: Google Scholar
  234. D. Koczan, S. Drynda, M. Hecker et al., “Molecular discrimination of responders and nonresponders to anti-TNFalpha therapy in Rheumatoid Arthritis by etanercept,” Arthritis Research & Therapy, vol. 10, no. 3, article R50, 2008. View at: Publisher Site | Google Scholar
  235. T. Häupl, B. Stuhlmüller, A. Grützkau, A. Radbruch, and G.-R. Burmester, “Does gene expression analysis inform us in Rheumatoid Arthritis?” Annals of the Rheumatic Diseases, vol. 69, supplement 1, pp. i37–i42, 2010. View at: Publisher Site | Google Scholar
  236. B. Stuhlmüller, T. Häupl, N. Tandon et al., “Microarray analysis for molecular characterization of disease activity and measuring outcomes of anti-tumour necrosis factor therapy in Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 7, article P159, 2005. View at: Publisher Site | Google Scholar
  237. T. Lequerré, A.-C. Gauthier-Jauneau, C. Bansard et al., “Gene profiling in white blood cells predicts infliximab responsiveness in Rheumatoid Arthritis,” Arthritis Research & Therapy, vol. 8, no. 4, article R105, 2006. View at: Publisher Site | Google Scholar
  238. A. Julià, A. Erra, C. Palacio et al., “An eight-gene blood expression profile predicts the response to infliximab in Rheumatoid Arthritis,” PLoS ONE, vol. 4, no. 10, Article ID e7556, 2009. View at: Publisher Site | Google Scholar
  239. S. M. Zivojinovic, N. N. Pejnovic, M. N. Sefik-Bukilica, L. V. Kovacevic, I. I. Soldatovic, and N. S. Damjanov, “Tumor necrosis factor blockade differentially affects innate inflammatory and Th17 cytokines in Rheumatoid Arthritis,” Journal of Rheumatology, vol. 39, no. 1, pp. 18–21, 2012. View at: Publisher Site | Google Scholar
  240. R. M. Thurlings, M. Boumans, J. Tekstra et al., “Relationship between the type I interferon signature and the response to rituximab in Rheumatoid Arthritis patients,” Arthritis & Rheumatism, vol. 62, no. 12, pp. 3607–3614, 2010. View at: Publisher Site | Google Scholar
  241. S. Vosslamber, H. G. Raterman, T. C. T. M. van der Pouw Kraan et al., “Pharmacological induction of interferon type i activity following treatment with rituximab determines clinical response in Rheumatoid Arthritis,” Annals of the Rheumatic Diseases, vol. 70, no. 6, pp. 1153–1159, 2011. View at: Publisher Site | Google Scholar
  242. L. G. M. van Baarsen, C. A. Wijbrandts, F. Rustenburg et al., “Regulation of IFN response gene activity during infliximab treatment in Rheumatoid Arthritis is associated with clinical response to treatment,” Arthritis Research & Therapy, vol. 12, no. 1, article R11, 2010. View at: Publisher Site | Google Scholar
  243. J. Lübbers, M. Brink, L. A. van de Stadt et al., “The type I IFN signature as a biomarker of preclinical Rheumatoid Arthritis,” Annals of the Rheumatic Diseases, vol. 72, no. 5, pp. 776–780, 2013. View at: Publisher Site | Google Scholar
  244. G. Tesch, S. Amur, J. T. Schousboe, J. N. Siegel, L. J. Lesko, and J. P. F. Bai, “Successes achieved and challenges ahead in translating biomarkers into clinical applications,” The AAPS Journal, vol. 12, no. 3, pp. 243–253, 2010. View at: Publisher Site | Google Scholar
  245. M. Pirmohamed, “Acceptance of biomarker-based tests for application in clinical practice: criteria and obstacles,” Clinical Pharmacology and Therapeutics, vol. 88, no. 6, pp. 862–866, 2010. View at: Publisher Site | Google Scholar

Copyright © 2014 Agata Burska et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

9923 Views | 2871 Downloads | 82 Citations
 PDF  Download Citation  Citation
 Download other formatsMore
 Order printed copiesOrder
 Sign up for content alertsSign up