Doc

  1

Best Practice & Research Clinical Rheumatology Vol. 18, No. 3, pp. 249–269, 2004 doi:10.1016/j.berh.2004.03.007

available online at http://www.sciencedirect.com

  Laboratory testing in autoimmune rheumatic diseases

Joanna Sheldon* FIMLS, PhD, MRCPath

Director of the Protein reference unit, consultant clinical scientist and Honorary senior lectures Protein Reference Unit, St George’s Hospital Medical School, Cranmer Terrace, London SW15 0RE, UK

There are a number of pathological conditions in which tissue damage occurs in association with immune activation directed against components of normal tissue. The initial damaging events usually involve cells of the immune system, the T-cells, but the cell damage releases antigens that become targets for an antibody response. The detection and quantification of autoantibodies has become an important component in the diagnosis and management of autoimmune rheumatic diseases such as rheumatoid arthritis, systemic lupus erythematosus, the systemic vasculitides and systemic sclerosis.

Each of these diseases is associated with a particular autoantibody or group of autoantibodies. They are usually detected by their reaction against tissue components using subjective methods such as indirect immunofluorescence. Any positive samples are further analysed using more specific and quantitative methods for the ‘quantification’ of the specific autoantibody concentration.

It is important that these autoantibodies are not considered to be ‘gold standard’ tests: they are no more than markers of the disease with significant limitations. They are best used as part of a diagnostic panel rather than as a marker indicating one particular disease. Techniques are gradually improving, giving numerical results rather than titres, but a lack of standardization makes these results extremely variable. Many of the markers show no correlation with disease activity. Their use should be restricted to the initial investigation and not repeated every time the patient is followed up. Other markers do, however, correlate with disease activity and can be used to monitor disease.

When investigating patients who have symptoms associated with autoimmune rheumatic diseases, analytes such as immunoglobulins, complement components and C-reactive protein may all be measured.

Key words: autoantibodies; autoimmune diseases; detection; quantification.

The immune system has evolved to recognize and destroy invading pathogens with minimal damage to the host. We do, in fact, make immune responses to host antigens but usually eliminate these self-reactive cells, giving us tolerance to the large variety of host antigens.

*Tel.: þ44-208-725-5752; Fax: þ44-208-725-0025. E-mail address: jsheldon@sghms.ac.uk (J. Sheldon).

    1521-6942/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.

250 J. Sheldon

The components of the immune system include cells (lympocytes, monocytes, macrophages, neutrophils, dendritic cells), soluble components (antibodies, comp- lement, cytokines, acute-phase proteins) and of course the lymphoid tissue, where many of these components are made and where the initial stages of an immune response occur. Briefly, an antigen, usually encountered across mucosal surfaces, is picked up by antigen-presenting cells (APCs). These APCs degrade the antigen into short sequences and insert them into the human lymphocyte antigens on their surface. The APCs travel around the body showing the antigens to the T- and B-lymphocytes in the lymphoid tissue. Activation of the T- or B-cell occurs if their receptors can ligate with the antigen. The production of cytokines is an integral part of immunological activation and function. T-helper cells can be generated that will enhance B-cell function and the production of specific antibodies. The binding of antibody to antigen will activate the complement system, which will release inflammatory mediators to recruit neutrophils and cause an acute-phase response. Cytotoxic T-cells, generated particularly in response to viral antigens, can directly destroy targeted cells.

There are many suggestions for how this response, normally to a foreign antigen, results in a response to host tissue and possibly an autoimmune disease. This breakdown of tolerance can be due to the following:

† Molecular mimicry: pathogen-derived antigens, when presented to the immune system, initiate an immune response that is cross-reactive with host tissue.

† Release of hidden antigens: tissue injury may release antigens that have previously been ‘hidden’ from the immune system, and an autoimmune response occurs.

† T-cell bypass: autoreactive T-cells are usually deleted, but if an antigen becomes altered, for example by infection, drugs or ultraviolet light, new autoreactive T-cell clones may be activated.

† Cytokine imbalances and defective immune regulation: cytokines are the mediators of immune responses. There is a complex interaction between various groups of cytokines that give an individual and immunological reactivity. Inappropriate cytokine production is associated with a variety of immunological abnormalities.

Tissue damage may be caused by cellular responses, by antibody activity or by both. T-helper cells and cytotoxic T-cells are usually found in inflammatory tissue lesions and contribute to the tissue destruction and perpetuation of the immunological and inflammatory reactions. Antibodies can react directly with tissue-specific cell surface antigens or with circulating antigens that deposit in tissue with the antibodies as immune complexes. These immune complexes activate the complement cascade, causing tissue damage, the recruitment and activation of neutrophils and inflammation.

Self-reactive T-cells are the likely instigators of most autoimmune reactions, but detecting these cells is not a practical possibility. Autoantibodies, however, are much more readily detected in easily collected blood samples and are therefore likely to remain an important part of the diagnosis and monitoring of autoimmune diseases for years to come.

LIMITATIONS TO THE USE OF AUTOANTIBODY TESTING

Autoantibodies should be considered to be no more than markers of disease. They are commonly found in normal individuals in the absence of any definable disease and with increasing prevalence in an ageing population. This lack of specificity makes autoantibody testing at best only a part of a diagnostic panel.

Laboratory testing in autoimmune rheumatic diseases 251

A multitude of kits is now available for the detection and quantification of autoantibodies. Unfortunately, there are few reliable national or international standards, and there is a huge variation between reagent producers in the preparation and source of antigens and the methods. Results are often reported in arbitrary units, and every method will have different cut-off values, reference ranges and measuring range. Overall, this makes comparing methods, interpretating published data and carrying out multicentre studies difficult. Laboratories offering autoimmune serology tests should be accredited (in the UK, with clinical pathology accreditation (CPA) and participate, with satisfactory performance, in all relevant external quality assurance schemes.

    Practice points

† none of the tests is perfect—most show a lack of clinical specificity and sensitivity

† get to know your laboratory

† do not assume that results will be interchangeable between laboratories

† if the results are completely inconsistent with your clinical findings, ask

(politely) for the test to be repeated, possibly using another method

† if there is still doubt about a result, recheck it on a fresh sample

† makesurethatyouareaskingaspecificquestionwhenyourequestlaboratory

tests...and that the tests you request are capable of answering the question

     Research agenda

† to establish and introduce international reference preparations that all diagnostic companies will use

† to re-establish the clinical use of assays based on current clinical disease classifications

† to evaluate current markers with respect to new therapies

† toidentifyandvalidatenewornovelmarkerstodiagnoseandmonitordisease

 RHEUMATOID FACTOR

Rheumatoid factors (RFs) were initially described in 19401; they are antibodies with activity directed against the Fc portion of immunoglobulin (Ig) G.2 The most commonly measured and clinically useful class of RF is IgM. It can be detected by sensitive immunoassay in approximately 90% of patients with rheumatoid arthritis (RA). IgG, IgA and IgE RFs are also reported and found by similar sensitive methods in approximately 65% of patients with RA. IgA RF is reported to be associated with bone erosions and IgG RF with localized or monoarticular disease.3

Measurement of rheumatoid factor

RF can be measured by agglutination methods, but quantitative, analyser-based immuno-nephelometric or immuno-turbidimetric assays are increasingly being used.

252 J. Sheldon

    Table 1. Table of the frequency of IgM rheumatoid factors (RF) in autoimmune rheumatic diseases and healthy individuals.

 Disease

Rheumatoid arthritis

Systemic lupus erythematosus Sjo ̈gren’s syndrome Cryoglobulinaemia

Systemic sclerosis Polymyositis/dermatomyositis Healthy individuals

% RF

50–90 15–35 20–30 40 – 100 20–30

5 – 10 2–10

   Most methods available measure predominantly IgM RF, although many will also detect RFs of the other Ig classes. There is a reference preparation (WHO 1066), and the concentration should be reported in IU/ml. Most methods are said to be calibrated against this standard, but external quality assurance schemes still show marked methodological variations.

Clinical significance of rheumatoid factor

Positive or negative RF results neither confirm nor exclude RA. Table 1 shows the frequency of IgM RF in autoimmune rheumatic diseases and in healthy individuals.4 Positive RF is also found in a number of other conditions, including infections such as tuberculosis and osteomyelitis, associated with monoclonal proteins in B-cell malignancy and with increasing frequency in an ageing population.

Rheumatoid arthritis

A raised concentration of IgM RF forms part of the classification of RA. There is a correlation between higher RF concentrations and more severe disease and poorer long-term prognosis. Concentrations of IgA and IgG RFs are reported to correlate with clinical parameters in RA patients.5 Reliable methods for detection and quantification of class-specific RFs are not routinely available.

There is a correlation between higher RF concentrations and systemic symptoms, vasculitis and more severe forms of the disease6, but the use of RF in monitoring disease activity or response to treatment remains controversial. It is unlikely that RF concentrations alone would be used for patient monitoring: C-reactive protein (CRP), erythrocyte sedimentation rate and clinical parameters are more useful.

Juvenile idiopathic arthritis

Pauci-articular-onset juvenile idiopathic arthritis is seen in approximately 50% of patients, who are generally RF negative but often anti-nuclear antibody (ANA) positive.7 Approximately 25% of children with polyarticular disease do show positive RF.

Laboratory testing in autoimmune rheumatic diseases 253

ANTI-CYCLIC CITRULLINATED PEPTIDE ANTIBODIES

The measurement of antibodies to cyclic citrullinated peptide (CCP) is gradually being incorporated into the repertoire of many immunology laboratories. These antibodies are directed against citrulline residues formed in post-translational modifications of arginine8 and are reported to be highly specific (98%) and moderately sensitive (68%) for RA.9 The concentration of these antibodies may be of value, either alone or with RF measurements as markers of prognosis or of disease severity.9 – 11 These antibodies have been incorporated into a newly proposed diagnostic criteria for RA.12

    Practice points

† CRP is more reliable than RF for monitoring RA

† do not keep on requesting RF

† encourage your laboratory to use quantitative RF methods

     Practice points

† if you are using anti-CCP antibodies, do not keep requesting RF measurements

† if you do not have access to anti-CCP antibodies, keep an eye on the literature and see how their use develops

† there is likely to be a laboratory somewhere that can run an anti-CCP antibody so if you are desperate, ask your laboratory

† in a patient with back pain, do not forget to send serum and urine for paraprotein studies to exclude myeloma

 ANTI-NUCLEAR ANTIBODIES

The term ‘anti-nuclear antibodies’ refers to a diverse group of autoantibodies directed against antigens in the nucleus and cytoplasm. The antigens are common to all nucleated cells and have functions in transcription or translation, in the cell cycle or as structural proteins. A sketch of a cell showing the location of the important nuclear antigens is shown in Figure 1. The nomenclature of ANAs is complex. The antigens are named by:

† chemical structure (e.g. double-stranded [ds] DNA);

† disease association (e.g. SS-A and SS-B in Sjo ̈gren’s syndrome);

† the individual in whom they were first described (e.g. Ro, La, Sm);

† their cytological location (e.g. nucleolar, centromere);

† the particle where the antigen is found (e.g. U1 sn RNP).

This nomenclature is further complicated by some antigens having more than one name (e.g. Ro and La are also known as SS-A and SS-B, respectively), and many of the antibodies are known by their staining pattern in indirect immunofluoresence (IIF, e.g. homogeneous or speckled staining patterns).

254 J. Sheldon

dsDNA (and ssDNA): homogeneously distributed across the nucleus in non-dividing cells. Homogeneous staining seen with antibodies

to ssDNA and

(with chromosomal staining) with antibodies to dsDNA.

Extractable Nuclear Antigens: clusters of

the many ENAs arranged in ranging sized 'speckles' across the nucleus

Chromosomes: organised from chromatin for cell division. Staining seen with antibodies to dsDNA

Centromere:

40_60 discrete fine speckles seen with centromere antibodies

Nucleolar : 2-4 prominent nucleoli seen in the p2 cells

    Figure 1. Sketch of a cell showing the location of the important nuclear antigens. Detection of anti-nuclear antibodies

Screening tests for ANAs are most commonly carried out by IIF using either frozen sections of animal tissue or, more appropriately, cultured cell lines (e.g. human epithelial HEp2 cells). Neither substrate is perfect for ANA screening, some antigens being poorly expressed in rodent tissue (e.g. Scl70 and centromeric proteins), and others (e.g. Ro/SS-A) not being conserved in their native form in cell lines or in rodent tissue. Laboratories should therefore be aware of the possibility of false-negative results in the ANA screen and be prepared to run more specific follow-up tests if clinically indicated.

There are many methods available for the detection or quantification of specific antibodies to nuclear components; these include enzyme-linked immunosorbent assay (ELISA), Ouchterloney double immunodiffusion, Western blotting and countercurrent immunoelectrophoresis. The different sensitivities and specificities of these methods is likely to influence the interpretation and clinical utility of many of the tests.13

Commercially produced ELISA-based ‘ANA screens’ are increasingly becoming available. The antigens used in these reagents range from crude cell preparations to mixtures of recombinant proteins, with a consequent high variability between different manufacturers. At present, there is considerable interest in these techniques but usually for reasons of laboratory organization rather than scientifically driven improvement.

Clinical significance of anti-nuclear antibodies

The detection of a positive staining pattern should lead to a more specific assay for the relevant antigens. The standardization of these assays remains a problem for the following reasons:

† variation in the antigen (used as a ‘substrate’ for the antibodies in the patients’ serum) preparation;

† variations in antibody affinity between patients and between standards and patient samples;

† difficulties and/or lack of standardization;

† methodological differences, for example, the detection of IgG antibodies to an

antigen in some methods whereas others detect both IgG and IgM antibodies.

Laboratory testing in autoimmune rheumatic diseases 255

Because of these limitations, specific antibody results must always be interpreted with some caution, particularly when laboratories have changed methods.

Anti-nuclear antibody patterns

The staining pattern seen on IIF gives some indication of the specificity of the antibodies in the sample. Detecting subtle staining patterns has become rather an ‘art form’ in some laboratories; in reality, there are only a handful of well-defined staining patterns that everyone testing for ANAs should be able to recognize. Typical ANA patterns are shown in Figure 2. Other specificities rarely have clinical significance: the antibodies are present in low concentrations and can be found non-specifically in many infections and inflammatory conditions. Table 2 shows the major staining patterns seen on IIF on HEp2 cells, the related antigen and disease association.

Homogeneous staining patterns Antibodies to single-stranded DNA

Antibodies to DNA can be divided into three major groups:

† antibodies to the conformational epitopes inherent in the double helix structure of the native molecule;

† antibodies to the epitopes expressed on the deoxyribose phosphate backbone; † antibodies to epitopes expressed on the purine and pyrimidine bases.

Figure 2. Indirect immunofluorescence staining patterns of anti-nuclear antibodies seen on HEp2 cells. (a) Homogeneous staining pattern with chromosomal staining. (b) Speckled staining pattern. (c) Centromere staining pattern. (d) Nucleolar staining pattern.

 

256 J. Sheldon

    Table 2. Major staining patterns seen on indirect immunofluorescence on HEp2 cells, the related antigen and disease association.

 Pattern

Homogeneous Diffuse

Peripheral (rim) Speckled-coarse

Speckled-fine

Centromere (46 dots) Nucleolar-homogeneous

Nucleolar-speckled Cytoplasmic

Antigen

DNA Histone

Topoisomerase-1 (Scl70)a dsDNA

Sm

RNP

SS-A/Ro

SS-B/La

CENP PM-SC 1

Nucleolar RNA

Histidyl-t-RNA synthetase (Jo-1)

Disease association (%)

SLE (60%)

Drug-induced lupus erythematosus (95%) SLE (60%)

Progressive systemic sclerosis (15–70%) SLE

SLE (20%)

SLE-Sjo ̈gren’s overlap (100%)

SLE (25%)

Sjo ̈gren’s syndrome (60%)

SLE (35%)

Sjo ̈gren’s syndrome (40%)

SLE (15%)

Limited scleroderma (CREST) (7–21%) Polymyositis (8%)

Myositis–scleroderma overlap (50%) Scleroderma (5–43%)

Polymyositis (2%)

  SLE, systemic lupus erythematosus; see text for antigens. a May also appear as very finely speckled pattern.

 The antibodies to the purine and pyrimidine bases can only be made when the DNA is denatured to single-stranded DNA. These antibodies to single-stranded DNA are the most frequently encountered antibodies in systemic lupus erythematosus (SLE) but are also seen in most types of autoimmune rheumatic disease, in patients receiving cytotoxic therapy and in viral infections. Their lack of specificity makes them of minimal value in the investigation of patients with autoimmune rheumatic diseases.

Antibodies to double-stranded DNA

Antibodies to native or dsDNA are strongly associated with SLE, raised concentrations of anti-dsDNA being reported in approximately 60% of patients with SLE.14 The Farr assay, based on radioimmunoassay, detects high-avidity antibodies, and has the highest specificity for SLE and the best correlation with disease activity. Most laboratories have moved away from the Farr assay owing to the safety concerns associated with the use of radiolabelled reagents. ELISA assays are now the most commonly used method for quantifying antibodies to dsDNA, although these assays may detect low-affinity antibodies that are irrelevant to disease. IIF using Crithidia lucilae as the substrate shows high specificity for dsDNA antibodies but relatively low sensitivity; we use this method to confirm the presence of anti-dsDNA antibodies when the IIF and ELISA results are discrepant. The concentration of anti-dsDNA antibodies is useful for both prognosis and monitoring patients with SLE15 so serial measurements, for example, every 8–12 weeks, are justified.

Speckled staining patterns

Laboratory testing in autoimmune rheumatic diseases 257

Antibodies to extractable nuclear antigens

The extractable nuclear antigens (ENAs) are a group of antigens that leach from the cell when extracted with saline. There are many antigens within the nucleus, but only a small number have any clinical utility. The original disease associations of the ENAs were defined with the relatively insensitive gel-phase assays, double immunodiffusion and countercurrent immunoelectrophoresis. Nowadays, more sensitive ELISA-based methods are more commonly used, with a consequent decrease in diseased specificity. Furthermore, these assays detect low-affinity antibodies in addition to the more clinically significant high-affinity antibodies. In general, there is little association between the concentration of antibody to the ENAs and severity of disease, and little fluctuation in antibody concentration with time or disease severity.

Anti SS-A (Ro) and Anti SS-B (La)

The SS-A antigen consists of 52 and 60 kDa proteins (called Ro52 and Ro60, respectively) complexed with Y1–Y5 RNAs. Antibodies (mainly of the IgG isotype) to the Ro52 antigen almost always appear in association with antibodies to the Ro60 antigen, and few techniques can reliably distinguish between antibodies to the two different antigens. Antibodies to Ro52 are more often seen in patients with primary Sjo ̈gren’s syndrome, whereas antibodies to Ro60 are found more often in patients with SLE.

Testing by countercurrent immunoelectrophoresis shows antibodies to SS-A (usually in association with antibodies to SS-B) in 60% of patients with primary Sjo ̈gren’s syndrome and rarely in healthy individuals. There have been marked changes in the methodologies used for detecting these antibodies, and the more sensitive ELISA methods may increase the ‘pick-up’ rate but with a consequent increase in the number of false-positive results (reducing specificity). Anti-SS-A antibodies are also found in a number of other autoimmune diseases, including RA, SLE and polymyositis.16

Anti-SS-A antibodies (Ro52 and Ro60) are seen in 35% of patients with SLE and can be associated with neonatal lupus and with complete or partial fetal heart block. IgG crosses the placenta in the last trimester of pregnancy; if anti-SS-A antibodies are present in the maternal serum, these antibodies will also cross the placenta and may interact with the conducting fibres of the heart and induce fetal heart block in utero.17 The development of heart block and neonatal lupus are, however, relatively uncommon events. Of mothers who are known to be SS-A antibody positive, only 1 in 20 will give birth to a baby with heart block, and 1 in 15 will have a baby who develops neonatal lupus.

The SS-B protein is thought to participate in the termination of transcription of RNA polymerase III.18 Antibodies to SS-B are seen in approximately 40% of patients sjogren’s syndrome with when tested by countercurrent immunoelectrophoresis. As with testing for anti-SS-A antibodies, new techniques have resulted in an increased detection rate but reduced disease specificity. Antibodies to SS-B are also seen in SLE, RA and polymyositis.19

The role of the anti-SS-A and anti-SS-B antibodies in disease pathogenesis remains controversial. Anti-SS-A antibodies are reported to show a close association with the development of vasculitis, nephritis, lymphadenopathy and leucopenia.20 Anti-SS- B antibodies are reported to increase during disease flares in Sjo ̈gren’s syndrome.21 Anti-SS-A antibodies are often seen in patients with systemic sclerosis and occur

258 J. Sheldon

alongside the disease-specific antibodies. The presence of anti-SS-A antibodies (especially with anti-SS-B antibodies) may mark a subset of 5–10% of cases of systemic sclerosis with limited disease and a high prevalence of renal and lung disease involvement.22

The detection of anti-SS-A and anti-SS-B antibodies is an important component of the investigation of patients with suspected Sjo ̈gren’s syndrome and SLE. Specific tests should be used as a confirmatory investigation in samples showing a speckled ANA pattern on HEp2 cells. There is, however, limited value in using these tests for monitoring disease progression.

Laboratories are occasionally criticized for not running the specific tests for antibodies to the ENAs; in my laboratory, we run anti-ENAs no more than once per year on known patients unless there are good clinical indications. We will run the specific tests, even on ANA-negative (on HEp2 immunoflourescence) samples, if there is strong clinical suspicion.

Anti-Sm

The Sm antigen is a complex group of four proteins (29 kDa protein B, 28 kDa protein B, 16 kDa protein D, 13 kDa protein E) complexed with U1, U2, U4–U6 and U5 snRNAs.23 Over 80% of serum samples showing antibodies to Sm will also be positive for antibodies to ribonuclear protein (RNP), with which they share antigenic determinants. Antibodies to the U1–D1–3 snRNP complex, which forms part of the Sm complex, are highly specific for SLE although not very sensitive and are associated with renal involvement and a poor prognosis.24

Anti-ribonuclear protein

The RNP antigen is highly pleiomorphic, with three proteins (70, 33, and 22 kDa) complexed with U1 snRNA.23 The nomenclature of this complex is variable, U1-RNP, nRNP, snRNP, U-snRNP and U1-snRNP all being names given to the same antigen. The antibodies show high specificity for mixed connective tissue disease, in which they occur as the sole antibody.25 They may also be seen without antibodies to the Sm antigen in patients with SLE, a group of patients showing a lower frequency of antibodies to dsDNA and clinically apparent renal disease.26

Antibodies associated with other staining patterns

Anti-Scl-70

Antibodies to Scl-70 are directed againsts DNA topoisomerase 127, and are seen in 20 – 25% of patients with systemic sclerosis and rarely in other autoimmune diseases. Patients with anti-topoisomerase antibodies tend to be slightly younger than patients with anti-centromere antibodies, and they are particularly affected by lung fibrosis.22 The anti-topoisomerase antibodies can give a variable staining pattern depending upon the substrate; nucleolar staining with fine speckling is often seen. There are ELISA methods available for quantifying anti-Scl-70 antibodies

Nucleolar antibodies

A nucleolar staining pattern (2 – 4 prominent nucleoli usually seen in HEp2 cells) is seen with antibodies to a number of antigens. Fine speckled staining with nucleolar staining is seen with antibodies to RNA polymerase III and I. These are associated with diffuse cutaneous systemic sclerosis and with a high incidence of lung fibrosis and renal disease.28 Even, diffuse staining of the nucleoli is associated with antibodies to PM-Scl

Laboratory testing in autoimmune rheumatic diseases 259

and is seen with polymyositis–scleroderma overlap. This antibody is associated with myositis and pulmonary fibrosis and a generally good prognosis.29 A clumpy nucleolar staining pattern is seen with antibodies to the fibrillarin subunit of U3 snRNP.30 These are associated with a poor prognosis, an increased risk of heart and kidney involvement and life-threatening pulmonary hypertension.

Centromere antibodies

Antibodies to the centromeric proteins were described in 1980 and are directed against epitopes found in the kinetochore domain of the chromosome. There are three main centromeric antigens—CENP-A, -B and -C—and the minor centromeric proteins CENP-D, -E and -F. Antibodies to the three main centromeric proteins often occur in the same sera, but antibodies to CENP-B are found in almost all patients with anti- centromere antibodies. Anti-centromere antibodies have a characteristic staining pattern on IIF using HEp2 cells as a substate, and laboratories should report this pattern if it is seen. A highly sensitive and specific ELISA assay for CENP-B is also available31, although the quantification of the antibody has little clinical application.

Anti-centromere antibodies are seen in systemic sclerosis in its limited cutaneous form and in the CREST variant. The antibody can also be found (at a low frequency) in other connective tissue diseases, for example RA, SLE and primary Sjo ̈ gren’s syndrome. In patients with Raynaud’s phenomenon, the presence of anti-centromere antibodies suggests an increased risk of developing rheumatic disease. Anti-centromere antibodies are also found in a sub-group (approximately 20%) of patients with primary biliary cirrhosis, in whom the liver disease may predate the manifestations of systemic sclerosis.32

Anti-Jo-1 and antibodies to aminoacyl-tRNA synthetases

Anti-Jo-1 antibodies are part of a group of antibodies to a family of aminoacyl-tRNA synthetases33, enzymes that complex amino acids with the cognate tRNA. The enzymes within this family that are associated with antibodies are shown in Table 3.34 Antibodies to these enzymes, associated with polymyositis and dermatomyositis, are typically detected by IIF on HEp2 cells when a fine granular cytoplasmic staining pattern is seen. There are, however, a significant number of samples with antibodies to Jo-1 that do not show any staining in HEp2 cells; therefore if there are clinical indications, the antibody should be detected and quantified by specific (ELISA-based) technology. These are clearly not ANAs, but the fact that they are seen in the screening test for ANAs often means that they are considered to be part of the ‘ANA group’.

    Table 3. The family of aminoacyl-tRNA synthetase. weight of Incidence

 Molecular

Name antigen (kDa) (%)

Jo-1 50 PL-7 80 PL-12 110 EJ 75 OJ 145

Enzyme

18 Histidyl tRNA synthetase 5 Threonyl tRNA synthetase 3 Alanyl tRNA synthetase

1 Glycyl tRNA synthetase

1 Isoleucyl tRNA synthetase

Disease association

Interstitial lung disease, arthritis Raynaud’s phenomenon

   

260 J. Sheldon

Jo-1 antibodies are found in 20 – 40% of patients with aggressive polymyositis, usually in association with interstitial lung disease and arthralgia. The concentration of anti-Jo-1 antibodies is reported to correlate with disease activity so sequential measurements may be of value in monitoring patients.35

A number of antibodies to other tRNA synthetases also show an association with myositis. These antibodies are found with the highest frequency early in the disease but will persist almost regardless of disease activity or treatment, so their measurement remains in the realms of specialized laboratories.

    Practice points

† ask your laboratory whether you can have a look at how they carry out these tests—you will be amazed at how much care they take and how hard some of the patterns can be to read

† a few laboratories have enormous experience of ANA testing and a variety of techniques at their disposal for identifying odd patterns; good laboratories often send samples on for confirmation or a second opinion

† laboratoriesshouldbeabletorunanANAtestwithin1workingdayifthereis good clinical justification; ask if necessary

† ask for an ANA rather than asking specifically for antibodies to dsDNA or ENAs; this way, you will lessen the risk of missing something important

† use the concentration of anti-dsDNA antibodies to monitor but do not keep requesting antibodies to ENAs

 ANTI-NEUTROPHIL CYTOPLASMIC ANTIBODIES

The function of neutrophils is to ingest and destroy antigens; the neutrophils are therefore full of potent, damaging enzymes such as proteinase III (PR3), myeloperox- idase and elastase. Antibodies against these neutrophil enzymes have been described in association with primary systemic vasculitides. These heterogeneous disorders are characterized by widespread inflammation of the vessel walls, and their classification is based on clinical and histological findings (Table 4).36 The small-vessel vasculitides of Wegener’s granulamotosis, Churg-Strauss syndrome and microscopic polyangiitis are

     Large-vessel vasculitis Medium-sized vessel vasculitis Small-vessel disease

Table 4. The primary vasculitides.

Giant cell (temporal) arteritis

Takayasu’s arteritis

Polyarteritis nodosa Kawasaki disease

Wegener’s granulamotosisa Churg-Strauss syndromea Microscopic polyangiitisa Henoch-Scho ̈nlein purpura

Essential cryoglobulinaemic vasculitis Cutaneous leukocytoclastic angiitis

 a Associated with anti-neutrophil cytoplasmic antibodies.

 

Laboratory testing in autoimmune rheumatic diseases 261

strongly associated with the presence of anti-neutrophil cytoplasmic antibodies (ANCA), in particular antibodies to PR3 and to myeloperoxidase.

Detection and quantification of anti-neutrophil cytoplasmic antibodies

In 1988, the First International Workshop on ANCA agreed to harmonize methods for the detection of these antibodies. It was agreed that the basis of detection would be IIF using washed human buffy coat leukocytes smeared or cytospun onto slides and then fixed with ethanol or acetone and fluorescein isothiocyanate-conjugated anti-human IgG as a detection antibody.37 This protocol was reiterated by the International Consensus Statement38, which states that all new patients should be tested for IgG antibodies by IIF on ethanol-fixed human neutrophils and only positive results should be investigated with antigen-specific assays. Automated, ELISA-based screening assays for ANCA are increasingly becoming available. Many laboratories are adopting these techniques in an attempt to ‘streamline’ their services. In the absence of international standardization, however, results can be reported only in arbitrary units, the sensitivity of many of the techniques is questionable, and the diagnostic efficiency of these assays is unknown. Screening samples for ANCA using ELISA-based methods is not recommended.

The IIF of ethanol-fixed neutrophils shows two major staining patterns: a cytoplasmic granular pattern and a perinuclear pattern. These are now conventionally known as the c-ANCA pattern and the p-ANCA pattern, respectively. Typical c- and p- staining patterns are shown in Figure 3.

The c-ANCA pattern shows granular cytoplasmic fluorescence with central interlobular accentuation in the ethanol-fixed human neutrophils. This pattern is seen in association with Wegener’s granulomatosis.39,40 This pattern is usually caused by antibodies to the 29 kDa serine protease 3 (PR3) but can also be seen in association with other neutrophil cytoplasmic enzymes.

The p-ANCA pattern shows staining localized just around the nucleus. The ethanol fixation causes by redistribution of the positively charged neutrophil granule proteins towards the negatively charged DNA in the nucleus. The p-ANCA staining pattern is associated with antibodies to myeloperoxidase but is also seen with antibodies to other neutrophil enzymes and with ANAs. Samples that show p-ANCA staining without

Figure 3. Indirect immunofluorescence staining patterns of anti-neutrophil cytoplasmic antibodies (ANCA) seen on ethanol-fixed neutrophils. (a) Cytoplasmic ANCA. (b) Perinuclear ANCA.

 

262 J. Sheldon

    Table 5. Disease associations of anti-proteinase III and myeloperoxidase antibodies. Sensitivity of

  Disease entity

Wegener’s granulomatosis

Microscopic polyangiitis

Idiopathic cresentic glomerulonephritis Churg-Strauss syndrome

Polyarteritis nodosa

Anti-proteinase III (%)

85 45 25 10

5

Anti-myeloperoxidase (%)

10 45 65 60 15

   specificity for myeloperoxidase is associated with diseases such as ulcerative colitis, sclerosing cholangitis, autoimmune hepatitis and Felty’s syndrome.36

Other staining patterns can be seen and should be termed atypical ANCA. These patterns are most likely caused by antibody binding to multiple antigenic targets in the neutrophils.38 In my laboratory, we always check samples showing atypical ANCA staining with the specific assays for antibodies to PR3 and myeloperoxidase.

Clinical significance of anti-proteinase III antibodies

Antibodies to PR3 are predominantly seen in patients with Wegener’s granulomatosis (Table 5), although they are often seen in other vasculitic diseases.41 A c-ANCA staining pattern combined with positive PR3 ANCA on enzyme immunoassay is 99% specific for small-vessel vasculitis.42 Patients with PR3 ANCA relapse more frequently than patients with myeloperoxidase ANCA-associated vasculitis43 – 45 and show more widespread organ involvement, granuloma formation and active renal lesions. Studies suggest that these antibodies are involved in the pathogenesis of the vasculitis46,47, although there are patients with active disease who are ANCA negative and some patients in remission with high ANCA titres. Nevertheless, a relationship between ANCA titre and disease activity has been reported48, and the regular measurement of the titre or concentration of ANCA antibodies is an important component of monitoring of the disease. Rising titres of ANCA or rising concentrations of PR3 antibodies are frequently followed by relapses of Wegener’s granulomatosis49,50, and the persistence of ANCA after induction of remission is reported to be a risk factor for relapse.51

Clinical significance of anti-myeloperoxidase antibodies

Antibodies to myeloperoxidase are predominantly seen in patients with microscopic polyangiitis, idiopathic crescentic glomerulonephritis and Churg-Strauss syndrome (Table 5). The clinical presentations of anti-myeloperoxidase-associated ANCA vasculitis are more diverse than those seen with anti-PR3-associated ANCA vasculitis. A p-ANCA staining pattern combined with positive myeloperoxidase ANCA on enzyme immunoassay is 99% specific for small-vessel vasculitis.52 There is evidence that the serial measurement of anti-myeloperoxidase antibodies has a role in monitoring disease53, although this approach does not have such a well-established role as the measurement of anti-PR3 antibodies.

Laboratory testing in autoimmune rheumatic diseases 263

    Practice points

† laboratoriesshouldbeabletorunanANCAtestwithin1workingdayifthere is good clinical justification; ask if necessary

† ifyouhaveapatientwithdeterioratingrenalfunction,donotforgettoaskfor antibodies to glomerular basement membrane and ANAs, and send serum and urine samples for paraprotein studies

† specific anti-PR3I or anti-myeloperoxidase antibody concentrations may be useful in monitoring

† check the CRP concentration

 ANTI-CARDIOLIPIN ANTIBODIES AND ANTIBODIES TO b2 GLYCOPROTEIN-1

The term ‘anti-phospholipid antibodies’ is used to encompass a variety of different phospholipid antigens, including the negatively charged cardiolipin and phosphatidyl- serine and the neutrally charged phosphatidylcholine and phosphatidylethanolamine. Beta-2 glycoprotein-1 (b2GP1) is the major phospholipid-binding protein, and ‘anti- cardiolipin antibodies’ may bind to epitopes on this protein rather than to cardiolipin epitopes.54

The detection and quantification of antibodies to cardiolipin and to b2GP1 are an important component of the diagnosis of anti-phospholipid syndrome or Hughes’ syndrome.55 The investigation of patients with recurrent venous or arterial thrombosis, recurrent fetal loss and thrombocytopenia should include a full blood count, clotting screen and lupus anticoagulant, assays generally carried out in haematology departments.

Measurement of antibodies to cardiolipin and to b2 glycoprotein-1

Anti-cardiolipin antibodies are typically measured by ELISA-based techniques, but the methods are prone to significant problems and variability. This is because cardiolipin antibodies represent a small subset of low-affinity, highly heterogeneous antibodies whose in vitro binding may be influenced by many factors, for example, the nature of the antibody, antibody isotype, avidity and cross-reactivity. The nature of the reference preparation, antigen preparation and assay system further compound the variability.56 Methods for the quantification of IgG anticardiolipin antibodies are more robust than those for IgM anticardiolipin antibodies, and some laboratories no longer offer IgM class antibodies.

Clinical significance of antibodies to cardiolipin and to b2 glycoprotein-1

Anti-phospholipid antibodies can be seen in infections and in chronic diseases; it is well known that these antibodies give false-positive results in the Wasserman reaction serological test for syphillis. False-positive results can also be seen with hepatitis C, leprosy, lyme disease, mycoplasma, tuberculosis, HIV, Legionnaire’s disease, Q fever and varicella zoster infections. The anti-b2GP1 antibodies are usually negative in these non- specific reactions.

In 1999, preliminary classification criteria for the diagnosis of anti-phospholipid syndrome were proposed.57 The laboratory criteria are of:

264

† †

J. Sheldon

anti-cardiolipin antibodies of the IgG and/or IgM isotype in the blood, present at a medium or high titre on two or more occasions at least 6 weeks apart, measured by a standardized ELISA assay for b2GP1-dependent cardiolipin antibodies lupusanticoagulantpresentintheplasmaontwoormoreoccasionsatleast6weeks apart, detected according to the guidelines of the International Society on Thrombosis and Haemostasis.

These laboratory criteria emphasize the difference between lupus anti-coagulant and anti-cardiolipin antibodies, and the fact that they are neither synonomous nor necessarily co-existent. The diagnosis of anti-phospholipid syndrome can be made when at least one clinical and one laboratory criterion are met.

There is some correlation between anti-cardiolipin antibody concentration and disease severity, so in addition to monitoring patients’ clotting profiles, a measurement of anti-cardiolipin antibodies may be useful.

IMMUNOGLOBULINS

The clinical indications for the measurement of serum IgG, IgA and IgM concentrations are very limited. It is essential that they are measured if primary or secondary Ig deficiency or B-cell malignancy is being considered. Serum Ig quantifications should always be interpreted and reported with reference to the serum protein electrophoresis. To exclude B-cell malignancy, the urine must always be checked for Bence Jones protein. Ig concentrations may also be useful in the investigation of some tropical diseases and occasionally some liver diseases.

The concentrations of IgG, IgA and IgM show a markedly raised concentration in many autoimmune diseases. The magnitude of the increase in concentration is not directly related to the severity of disease, and there is little relationship between concentration and disease progression; repeated measurements are therefore a waste of time and money. A sensible compromise is to exclude antibody deficiency at the initial investigation and check the Ig concentrations no more than once per year and preferably less frequently.

The antibodies will be directed against both organ- and non-organ-specific targets, and well as there being high amounts of circulating immune complexes and cryoglobulins.58

    Practice points

† there is an enormous variability in the quantification of these antibodies

† itisimportanttochecktheantibodyconcentrationsonatleasttwooccasions

to make a secure diagnosis

     Practice points

† check the Ig concentrations in patients in whom immune deficiency or B-cell malignancy needs to be considered

† do not keep repeating Ig measurement in patients with RA, SLE or systemic sclerosis; they may well be abnormal but add little to your diagnosis or monitoring

 

CRYOGLOBULINS

Laboratory testing in autoimmune rheumatic diseases 265

     Type

Monoclonal Rheumatoid factor activity present Clinical syndrome Disease association

Table 6. Classification of cryoglobulins. I II

Yes Yes

No Yes (IgM)

Hyperviscosity Vasculitis

Malignancy Hepatitis C, Rheumatic disease,

Malignancy, Idiopathic

III

No Yes

Vasculitis

Hepatitis C, Rheumatic disease, Idiopathic

   Cryoglobulins are antibodies that precipitate in the cold and re-solubilize on warming; they are classified into three groups, shown in Table 6.

In general, for symptoms to be brought on by an exposure to cold, cryoproteins should exhibit their activity above 21 8C—the lowest physiological skin temperature. There are many proteins that aggregate below 16 8C that never produce symptoms but do produce laboratory artefacts. The important exceptions are the mixed cryoglobulins, which are detected after slow precipitation at 4 8C, precipitation sometimes taking up to 3 days to occur. There is often little correlation between exposure to the cold and amount of cryoprecipitate and patients’ symptoms, although correlation can occur within an individual.

Type I cryoglobulins are monoclonal proteins that typically do not have RF activity and are not effective activators of the complement cascade. This group of proteins is more likely to cause hyperviscosity and symptoms related to the precipitation of the protein, particularly in the extremities, at low temperatures. The mixed cryoglobulins (types II and III) activate complement efficiently and are likely to cause vasculitis. The symptoms associated with these types of protein are related to their immune complex behaviour rather than their precipitation in the cold. Their behaviour as cryoprecipitating proteins is essentially a laboratory artefact.

Types I and II cryoglobulins are associated with B-cell malignancies such as myeloma, Waldenstro ̈m’s macroglobulinaemia and chronic lymphocytic leukaemia. Type II cryoglobulins also occurs with some chronic infections (especially hepatitis C) and rheumatic diseases (Sjo ̈gren’s syndrome, RA, SLE). Type III cryoglobulins occur with rheumatic diseases or chronic infections but not with B-cell malignancies.

Common symptoms associated with cryoglobulins include purpura, ulceration, Raynaud’s phenomenon, arthralgia, proteinuria, renal failure and vasculitis. It is vital that samples are collected, transported and separated at 37 8C to ensure that small cryoproteins are not ‘lost’ into the cell pellet. Laboratories should have pre-warmed blood collection tubes and a warmed thermos flask; these should be brought to the patient just before the blood is collected. The blood collected into the warmed tubes and immediately put into the warm flask for immediate transport back to the laboratory.

    Practice points

† do not forget cryoglobulins—many of their symptoms overlap with those of the autoimmune rheumatic diseases

† if you are requesting cryoproteins, call the laboratory for proper instructions, tubes, etc.

 

266 J. Sheldon

C-REACTIVE PROTEIN

CRP is an acute-phase protein made in the liver under the control of cytokines such as interleukins 6 and 1, and tumour necrosis factor-alpha. In acute inlammation, the plasma CRP concentration increases within hours of the onset of the inflammatory response and falls quickly on appropriate treatment. The magnitude of inflammation is related to the magnitude of the CRP concentration.59 A raised CRP concentration is unequivocal evidence of an inflammatory process and may therefore be useful in distinguishing simple mechanical damage from more serious organic disease.

CRP measurements are useful in the autoimmune rheumatic diseases. Patients with RA usually have a raised CRP concentration that falls with appropriate treatment and correlates better with radiological evidence of progressive joint inflammation than other clinical of serological measurements.60

SLE, scleroderma, Sjo ̈gren’s syndrome and polymyositis can all be clinically active without a significant CRP response.61 Patients with these diseases occasionally have raised CRP concentrations, and patients are capable of mounting a CRP response to intercurrent bacterial infections.62 CRP concentrations are raised in vasculitis and can provide an objective marker of disease activity and response to treatment.63

COMPLEMENT

The complement proteins are a cascade of proteins that can be activated by a variety of agents including immune or antigen-antibody complexes. Serum concentrations of C3 and C4 are most commonly measured and low concentrations of C4 are associated with active immune complex disease and, in SLE, particularly with renal involvement.64 Patients with genetic deficiencies of complement (particularly C2 and C4) show an increased risk of immune complex disease65, and if you are considering complement deficiency, you should contact your laboratory for sample requirements for the total haemolytic complement (often called CH50) assay. Complement activity in vivo can be detected by the marker C3d. This remains a useful test but lies very much in the realms of the specialist laboratories.

    Practice points

† CRP is a non-specific marker of inflammation

† a raised CRP concentration is unequivocal evidence of an inflammatory

response

† the measurement of CRP is useful in monitoring RA and in distinguishing

between infection and disease ‘flare’ in SLE, etc.

     Practice points

† measureC3andC4concentrationsinpatientswithactivedisease—afallingC4 concentration is associated with an increased risk of renal complications

† if you are considering complement deficiency, check the total haemolytic complement level

 

SUMMARY

Laboratory testing in autoimmune rheumatic diseases 267

A huge range of laboratory tests is available to help in the diagnosis and management of patients with autoimmune rheumatic diseases. Not one of the tests is, however, perfect, and most tests have significant limitations. Nevertheless, when used properly, they provide a valuable minimally invasive way of diagnosing and monitoring disease.

REFERENCES

1. Waaler E. On the occurrence of a factor in human serum activating the specific agglutination of sheep corpuscles. Acta Pathologica et Microbiologica Scandinavica 1940; 1: 172–188.

2. Natvig JB, Gaarder PI & Turner MW. IgG antigens of the C gamma 2 and C gamma 3 homology regions interacting with rheumatoid factors. Clinical and Experimental Immunology 1972; 12: 177–184.

3. Jonsson T & Valdimarsson H. Is measurement of rheumatoid factor isotypes clinically useful? Annals of the Rheumatic Diseases 1993; 52: 161–164.

4. Shmerling RH & Delbanco TL. The rheumatoid factor: an analysis of clinical utility. American Journal of Medicine 1991; 91: 528–534.

5. Withrington RH, Teitsson I, Valdimarsson H & Seifert MH. Prospective study of early rheumatoid arthritis. II. Association of rheumatoid factor isotypes with fluctuations in disease activity. Annals of the Rheumatic Diseases 1984; 43: 679–685.

6. Moore TJ. Rheumatoid factors. Clinical Immunology Newsletter 1988; 18: 89–96.

7. Laxer R & Schneider R. Systemic-onset juvenile chronic arthritis. In Maddison P, Eisenberg D, Woo P & Glass DN (eds) Oxford Textbook of Rheumatology, second edn. Oxford: Oxford University Press, 1998, pp

1114 – 1131.

8. Schellekens GA, de Jong BA, van den Hoogen FH, et al. Citrulline is an essential constituent of antigenic

determinants recognized by rheumatoid arthritis specific antibodies. Journal of Clinical Investigation 1998;

101: 273–281.

9. Schellekens GA, Visser H, de Jong BA, et al. The diagnostic properties of rheumatoid arthritis antibodies

recognising a cyclic citrullinated peptide. Arthritis and Rheumatism 2000; 42: 155–163.

10. vanJaarsveldCHM,terBorgEJ,JacobsJWG,etal.Theprognosticvalueoftheantiperinuclearfactor,anti- citrullinated peptide antibodies and rheumatoid factor in early rheumatoid arthritis. Clinical and

Experimental Rheumatology 1999; 17: 689–697.

11. Kroot EJA, de Jong BW, van Leeuwen MA, et al. The prognostic value of anti-cyclic sitrullinated peptide

antibody in patients with recent-onset rheumatoid arthritis. Arthritis and Rheumatism 2000; 43: 1831 – 1835.

12. Visser H, le Cessie S, Vos K, et al. How to diagnose rheumatoid arthritis early. A prediction model for

persistent (erosive) arthritis. Arthritis and Rheumatism 2002; 46: 357–365.

13. Lock RJ & Unsworth DJ. Antibodies to extractable nuclear antigens. Has technological drift affected

clinical interpretation? Journal of Clinical Pathology 2001; 54: 187–190.

14. Swaak T & Smeenk RJ. Detection of anti-dsDNA as a diagnosis tool: a prospective study in 441 non-

systemic lupus erythematosus patients with anti-dsNDA antibody (anti-dsDNA). Annals of the Rheumatic

Diseases 1985; 44: 245–251.

15. terBorgEJ,HorstG,HummellEJ,etal.Measurementofincreasesinanti-double-strandedDNAantibody

levels as a predictor of disease exacerbation in systemic lupus erythematosus. A long term, prospective

study. Arthritis and Rheumatism 1990; 33: 634–643.

16. Manoussakis MN, Kistis KG, Liu X, et al. Detection of anti Ro(SSA) antibodies in autoimmune diseases:

comparison of five methods. British Journal of Rheumatology 1993; 32: 449–455.

17. Behan WMH, Behan PO, Reid JM, et al. Family studies in congenital heart block associated with Ro

antibody. British Heart Journal 1989; 62: 320–324.

18. Gottlieb E & Steitz JA. Function of the mammalian La protein: evidence for its activity in transcription

termination by RNA polymerase III. European Molecular Biology Organization Journal 1989; 8: 851–861.

19. Manoussakis MN, Kistis KG, Liu X, et al. Detection of human-specific anti La(SSB) antibodies patients

with rheumatoid arthritis. Journal of Autoimmunity 1995; 8: 959–969.

20. JamesJA,DickeyDJD,FujisakaA,etal.AntigenicityofarecombinantRo(SS-A)fusionprotein.Arthritisand

Rheumatism 1990; 33: 102–106.

21. Venables PJW, Charles PJ, Buchannan RC, et al. Quantitation and detection of isotypes of anti SS-B

antibodies by ELISA and Farr assays using affinity purified antigens. Arthritis and Rheumatism 1982; 26: 146 – 155.

268 J. Sheldon

22. Bunn CC, Denton CP, Shi-Wen X, et al. Anti RNA polymerase and other autoantibody specificities in systemic sclerosis. British Journal of Rheumatology 1998; 37: 15–20.

23. Habets WJ, Berden JHM, Hoch SO & van Venrooji WJ. Further characterisation and subcellular localisation of Sm and U1-ribonuclearprotein antigens. European Journal of Immunology 1985; 15: 992 – 997.

24. Holyst MM & Hoffman RW. U-small nuclear ribonucleoprotein (RNP) reactive autoantibodies: diagnostic testing and clinical interpretation. Clinical Immunology Newsletter 1998; 18: 53–57.

25. Panush RS, Edward NG, Longley S & Webster E. ‘Rhupus’ syndrome. Archives of Internal Medicine 1988; 148: 1633–1636.

26. ter Borg EJ, Groen H, Horst G, et al. Clinical association of antiribonuclearprotein antibodies in patients with systemic lupus erythematosus. Seminars in Arthritis and Rheumatism 1990; 20: 164–173.

27. Shero JH, Bordwell B, Rothfield N, et al. High titers of autoantibodies to topoisomerase I (Scl-70) in sera from scleroderma patients. Science 1986; 231: 737–740.

28. Harvey GR, Butts S, Rands AL, et al. Clinical and serological associations with anti-RNA polymerase antibodies in systemic sclerosis. Clinical and Experimental Immunology 1999; 117: 395–402.

29. Marguerie CM, Bunn CC, Copier J, et al. The clinical and immunogenic features of patients with autoantibodies to the nuclear antigen PM-Scl. Medicine 1992; 71: 327–336.

30. Bernstein RM, Steigerwald JC & Tan EM. Association of antinuclear and antinucleolar antibodies in progressive systemic sclerosis. Clinical and Experimental Immunology 1982; 48: 43–51.

31. Earnshaw W, Bordwell B, Marino C & Rothfield N. Three human chromosomal autoantigens are recognised by sera from patients with anticentromere antibodies. Journal of Clinical Investigation 1986; 77: 426 – 430.

32. McHugh NJ, James IE, Kairburn K, et al. Autoantibodies to mitochondrial and centromere antigens in primary biliary cirrhosis and systemic sclerosis. Clinical and Experimental Immunology 1990; 81: 244–249.

33. Bernstein RM. Autoantibodies in myositis. Baillie`re’s Clinical Neurology 1993; 2: 599–616.

34. Targoff IN. Humoural immunity in polymyositis/dermatomyositis. Journal of Investigative Dermatology 1995;

100: 116s–123s.

35. Polz PH, Dalakas M, Leff RL, et al. Current concepts in the idiopathic inflammatory myopathies:

polymyositis, dermatomyositis and related disorders. Annals of Internal Medicine 1989; 11: 143–157.

36. Jennette JC, Falk RJ, Andrassy K, et al. Nomenclature of systemic vasculitides—proposal of an

international consensus conference. Arthritis and Rheumatism 1994; 37: 187–192.

37. Wiik A. Delineation of a standard procedure for indirect immunofluorescence detection of ANCA. Acta

Pathologica, Microbiologica, et Immunologica Scandinavica 1989; 97(supplement 6): 12–13.

38. Savige J, Gillis D, Benson E, et al. International consensus statement on testing and reporting of antineutrophil cytoplasmic antibodies (ANCA). American Journal of Clinical Pathology 1999; 111: 507–513.

39. Ludeman J, Utect B & Gross WL. Anti-neutrophil cytoplasm antibodies in Wegener’s granulomatosis

recognize an elastinolytic enzyme. Journal of Experimental Medicine 1990; 171: 357–362.

40. Wong RCW, Silvestrini RA, Savige JA, et al. Diagnostic value of classical and atypical antineutrophil cytoplasmic antibody (ANCA) immunofluorescence patterns. Journal of Clinical Pathology 1999; 52:

124 – 128.

41. Kallenberg CGM, Brouwer E, Weening JJ & Cohen Tervaert JW. Anti neutrophil cytoplasmic antibodies:

current diagnostic and pathophysiological potential. Kidney International 1994; 46: 1–15.

42. Hagen EC, Daha MR, Heremans J, et al. Diagnostic value of standardized assays for anti-neutrophil

cytoplasmic antibodies in idiopathic systemic vasculitis. Kidney International 1998; 53: 743–753.

43. Franssen CFM, Gans ROB, Kallenberg CGM, et al. Disease spectrum of patients with antineutrophil cytoplasmic autoantibodies of defined specificity: distinct differences between patients with anti- proteinase 3 and anti myeloperoxidase autoantibodies. Journal of Internal Medicine 1998; 244:

209 – 216.

44. Geffriaud-Ricouard C, Noel LH, Chaveau D, et al. Clinical spectrum associated with ANCA of defined

antigen specificities in 98 selected patients. Clinical Nephrology 1993; 39: 125–136.

45. Westman KW, Bygren PG, Olsson H, et al. Relapse rate, renal survival and cancer morbidity in patients with Wegener’s granulomatosis or microscopic polyangiitis with renal involvement. Journal of the American

Society of Nephrology 1998; 9: 842–852.

46. Heeringa P, Brouwer E, Cohen Tervaert JW, et al. Animal models of anti-neutrophil cytoplasmic antibody

associated vasculitis. Kidney International 1998; 53: 253–263.

47. Muller Kobold AC, van der Geld YM, Limburg PC, et al. Pathophysiology of ANCA-associated

glomerulonephritis. Nephrolology, Dialysis and Transplantation 1999; 14: 1366–1375.

48. vanderWoudeFJ,RasmussenN,LobattoS,etal.Autoantibodiesagainstneutrophilsandmonocytes:tool

for diagnosis and marker of disease activity in Wegener’s granulomatosis. Lancet 1985; ii: 425–429.

49. Jayne DR, Gaskin G, Pusey CD & Lockwood CM. ANCA and predicting relapse in systemic vasculitis.

Quarterly Journal of Medicine 1995; 88: 127–133.

Laboratory testing in autoimmune rheumatic diseases 269

50. Egner W & Chapel HM. Titration of antibodies against neutrophil cytoplasmic antigens is useful in monitoring disease activity in systemic vasculitides. Clinical and Experimental Immunology 1990; 82: 244 – 249.

51. De’Oliveira J, Gaskin G, Dash A, et al. Relationship between disease activity and anti-neutrophil cytoplasmic antibody concentration in long-term management of systemic vasculitis. American Journal of Kidney Disease 1995; 25: 380–389.

52. Hagen EC, Daha MR, Heremans J, et al. Diagnostic value of standardized assays for anti-neutrophil cytoplasmic antibodies in idiopathic systemic vasculitis. Kidney International 1998; 53: 743–753.

53. Han WK, Choi WK, Roth RM, et al. Serial ANCA titres: useful tool for prevention of relapses in ANCA associated vasculitis. Kidney International 2003; 63: 1079–1085.

54. Amengual O, Atsumi T, Khamashta MA, et al. Specificity of ELISA for antibody to beta 2 glycoprotein 1 in patients with antiphospholipid syndrome. British Journal of Rheumatology 1996; 35: 1239–1243.

55. Hughes GRV. Thrombosis, abortion, cerebral disease and lupus anticoagulant. British Medical Journal 1983; 187: 1088–1089.

56. Milford Ward A & White PAE. Measuring cardiolipin antibodies. British Medical Journal 1995; 310: 1472 – 1473.

57. Wilson WA, Gharavi AE, Koike T, et al. Interntional consensus on preliminary classification criteria for definite antiphospholipid syndrome. Report of an international workshop. Arthritis and Rheumatism 1999; 42: 1309–1311.

58. HarleyJB.AutoantibodiesinSjo ̈gren’ssyndrome.InTalalN,MoutsopolousHM&KassanSS(eds)Sjo ̈gren’s Syndrome: Clinical and Immunological Aspects. Berlin: Springer, 1987, pp 218–234.

59. Colley CM, Fleck A, Goode AW, et al. Early time course of the acute phase protein response in man. Journal of Clinical Pathology 1983; 36: 203–207.

60. McConkey B, Crockson RA & Crockson AP. The assessment of rheumatoid arthritis: a study based on measurement of the serum acute phase reactants. Quarterley Journal of Medicine 1972; 41: 115–125.

61. Pepys MB. Serum C-reactive protein, serum amyloid P component and serum amyloid A protein in

autoimmune disease. Clinics in Immunology and Allergy 1981; 1: 77–101.

62. BeckerGJ,WaldburgerM,HughesGRV&PepysMB.ValueofserumC-reactiveproteinmeasurementsin

the investigation of fever in systemic lupus erythematosus. Annals of the Rheumatic Diseases 1980; 39:

50–52.

63. Hind C, Winearls CG & Pepys MB. Correlation of disease activity in systemic vasculitis with serum C-

reactive protein measurement. European Journal of Clinical Investigation 1985; 15: 89–94.

64. RuddyS,GigliI&AustenKF.Thecomplementsystemofman.I–IV.NewEnglandJournalofMedicine1972;

287: 642–646.

65. Frank MM. Complement in the pathophysiology of human disease. New England Journal of Medicine 1987;

316: 1525–1530.