Humoral immunity changes in Rheumatic diseases

01Definition and overview

Humoral immunity is the arm of adaptive immunity carried by antibodies made by B cells and plasma cells, with complement as its main effector. In the autoimmune rheumatic diseases it goes wrong in three ways:

• autoantibodies: antibodies against the body's own molecules, nuclear (ANA), cytoplasmic (ANCA, anti-Jo-1), membrane-bound (antiphospholipid) or soluble (rheumatoid factor against IgG);
• immune complexes: antigen bound to antibody, which may circulate and deposit in tissues or form in them;
• complement consumption or deficiency: the cascade that clears immune complexes and kills microbes is activated by them, damaging tissue and falling in the blood.

Clinically, an autoantibody does one or more of four jobs: it supports a diagnosis (anti-dsDNA in lupus), classifies a disease into subsets with different courses (the scleroderma and myositis antibodies), tracks activity (anti-dsDNA and complement in lupus nephritis), or predicts a risk (anti-Ro and fetal heart block, antiphospholipid antibodies and thrombosis). The disease topics name their own antibodies; this topic is where they are explained.

02Etiology and pathophysiology

Loss of tolerance

The body makes self-reactive B and T cells all the time. Most are deleted or silenced in the bone marrow and thymus (central tolerance), and the rest are held in check in the periphery by regulatory T cells, the lack of T cell help and limited survival signals. Autoimmunity begins when these checks fail:

• genetic susceptibility, above all in HLA class II genes that present self peptides to T cells, and in genes of B cell signalling and interferon;
• too much self-antigen in the wrong place: nuclear debris from apoptotic cells that is cleared slowly, as in lupus; citrullinated proteins made in the smoker's lung by the enzyme peptidylarginine deiminase, as in rheumatoid arthritis;
• molecular mimicry, in which antibodies against a microbe cross-react with a self protein, as in rheumatic fever;
• excess survival signals, such as the B cell factor BLyS (BAFF), and type I interferon, which keep autoreactive B cells alive and active.

Once started, the response spreads to more epitopes on the same molecule and to other molecules in the same particle (epitope spreading), which is why anti-Ro and anti-La, or anti-Sm and anti-RNP, travel together. Antibodies switch from IgM to high-affinity IgG with T cell help, and they can appear years before the illness.

How antibodies damage tissue

The Gell and Coombs scheme sorts the damage, simplified here; in a real disease several types act at once:

Immune complexes

Whether a complex harms depends on its size and on clearance. Large complexes, made with excess antibody, are cleared by phagocytes of the liver and spleen after complement coats them and carries them on red cells. Small complexes, made with a slight excess of antigen, stay in the blood and lodge where blood is filtered under pressure: the glomerulus, the synovium, the skin and the choroid plexus. Two classic models show it:

• the Arthus reaction: local complexes after an injection into someone with high antibody levels, with swelling and necrosis at the site within hours;
• serum sickness: systemic complexes 1 to 2 weeks after a foreign protein (antitoxins, antithymocyte globulin) or some drugs, with fever, rash, arthralgia, proteinuria and low complement, which settles as the antigen is cleared.

Complement

Three pathways meet at C3:

• the classical pathway, started when C1q binds IgG or IgM in immune complexes, consuming C1, C4 and C2;
• the lectin pathway, started by mannose-binding lectin on microbial sugars, also using C4 and C2;
• the alternative pathway, which ticks over continuously and amplifies the others, and consumes C3 without C4.

C3 is cleaved into C3b, which coats targets for phagocytes and clearance, and C3a; the C5 convertase then releases C5a, a strong chemoattractant and anaphylatoxin, and assembles C5b to C9, the membrane attack complex.

Cryoglobulins

Cryoglobulins are immunoglobulins that precipitate below body temperature and dissolve on warming. They block small vessels in the cold, and mixed cryoglobulins, which are immune complexes of IgM rheumatoid factor bound to IgG, deposit and fix complement:

03Clinical features

Which antibody goes with which disease

Figures are approximate: they vary with the method, the cut-off and the population, and the table simplifies.

Complement patterns

04Diagnosis and management

Asking the right question

Autoantibody tests are ordered for a clinical question, not as a screen. The same positive result means a lot in a young woman with arthritis, a rash and proteinuria, and little in someone with tiredness alone. Panels ordered without a question produce positives that cause anxiety, referrals and sometimes treatment for a disease the patient does not have. Once an ANA, ENA, RF or anti-CCP is positive, repeating it adds nothing; it is anti-dsDNA, complement and, in some settings, ANCA that are followed over time.

ANA

Indirect immunofluorescence on HEp-2 cells, a human epithelial tumour line with large nuclei and many antigens, is the reference method. Diluted serum is laid on the cells, bound antibody is shown with a fluorescent anti-human IgG, and the result is reported as a titre, the last dilution that still stains, with a pattern.

• Titre: 1 in 80 is the entry criterion for lupus classification. Each laboratory sets its own cut-off from healthy controls; a titre of 1 in 160 or more, found in only about 1 in 20 healthy adults, carries more weight, and a titre near the cut-off is weaker evidence. The titre does not measure activity.
• Pattern: the international consensus (ICAP) names each pattern with a code, and the pattern suggests what to test next:

An ANA reported as negative does not exclude every antibody. Anti-Jo-1 and other cytoplasmic antibodies are missed if only nuclear staining is read, and anti-Ro can stain weakly on HEp-2 cells. When the picture suggests myositis, Sjögren syndrome, subacute cutaneous lupus or neonatal lupus, the specific antibodies are requested directly. Automated solid-phase ANA screens are faster but detect only the antigens they contain.

ENA and the specific antibodies

The extractable nuclear antigens (Ro, La, Sm, RNP, Scl-70, Jo-1 and others), centromere protein B, and the myositis and scleroderma antibodies are measured by solid-phase immunoassays: ELISA, chemiluminescence, multiplex beads or line blots. A weak band on a line blot is often a false positive, so a result is trusted most when it fits the HEp-2 pattern and the clinical picture. Myositis-specific antibodies are usually mutually exclusive, so two in one patient suggest a false positive.

Anti-dsDNA

Immunoassays (ELISA, chemiluminescence, beads) are sensitive and quantitative but also pick up low-avidity antibody, so they are less specific. Immunofluorescence on Crithidia luciliae, a protozoon whose kinetoplast holds pure dsDNA, and the older Farr radioimmunoassay are very specific but less sensitive, so a positive immunoassay in a patient without a typical picture is often confirmed by Crithidia. For monitoring, the same method is used each time.

Rheumatoid factor and anti-CCP

Rheumatoid factor, an IgM antibody against the Fc part of IgG, is measured by nephelometry or ELISA. Anti-CCP detects antibodies against citrullinated peptides, the ACPA. Both count in the 2010 classification of rheumatoid arthritis, high titres more. Anti-CCP is the more specific, so a positive anti-CCP with inflammatory arthritis is strong evidence; rheumatoid factor alone is weak evidence in an older patient. Rheumatoid factor can also cause false positives in other IgM immunoassays.

ANCA

ANCA are antibodies against proteins in the granules of neutrophils and monocytes. By the 2017 international consensus:

1. 1

   Immunoassay first

   In suspected granulomatosis with polyangiitis or microscopic polyangiitis, high-quality antigen-specific immunoassays for PR3-ANCA and MPO-ANCA are the first test, without a categorical need for immunofluorescence; they discriminate vasculitis from other disease better.
2. 2

   If negative

   When suspicion stays high, another immunoassay or immunofluorescence is done, and a biopsy is considered; a few patients with vasculitis are ANCA negative.
3. 3

   If positive

   The result is weighed against the clinical picture, and a biopsy is sought where possible. In a pulmonary-renal syndrome, ANCA and anti-GBM are requested as emergencies, and treatment is not delayed for the result or the biopsy.

On immunofluorescence of ethanol-fixed neutrophils, PR3-ANCA usually gives a granular cytoplasmic pattern (cANCA) and MPO-ANCA a perinuclear one (pANCA), because the charged MPO moves to the nucleus on fixation. A perinuclear pattern without PR3 or MPO (atypical ANCA) is common in ulcerative colitis, primary sclerosing cholangitis and autoimmune hepatitis and does not mean vasculitis. ANCA also appear with infections, such as endocarditis and tuberculosis, and with drugs. Rising or persisting ANCA during remission predicts relapse only weakly, so treatment is not changed on the level alone.

Antiphospholipid tests

Lupus anticoagulant, a clotting test done with two methods, and the ELISAs for anticardiolipin and anti-β2-glycoprotein I IgG and IgM. A result counts only if it persists for at least 12 weeks. Anticoagulants distort the lupus anticoagulant test: direct oral anticoagulants cause false positives, and warfarin and heparin make it hard to read, while the ELISAs are unaffected. Infections cause transient, mostly low-titre IgM antibodies, and syphilis tests can be falsely positive. The details, and the risk profile, are in antiphospholipid syndrome.

Complement

• C3 and C4 are measured by nephelometry; their levels reflect both production and consumption, and both rise in the acute phase, so a normal value can hide modest consumption. A falling trend in the same patient means more than a single value.
• CH50, the classical haemolytic complement, tests the whole classical pathway from C1 to C9 by lysis of antibody-coated sheep red cells; AP50 tests the alternative pathway. An absent CH50 means a missing component. Complement is fragile, so a sample left at room temperature gives a falsely low result.
• Further tests answer specific questions: C1-inhibitor level and function for angioedema; C1q and anti-C1q antibodies in lupus nephritis and urticarial vasculitis; C3 nephritic factor; individual components when a deficiency is suspected.

Cryoglobulins, immune complexes and immunoglobulins

• Cryoglobulins: blood is collected into warmed tubes and kept at 37 °C until the serum is separated; if it cools first the cryoglobulin is lost with the clot. The serum is then kept at 4 °C for up to 7 days, the precipitate measured as a cryocrit and typed by immunofixation. Low C4 and a positive rheumatoid factor are clues, and hepatitis C is sought in every mixed cryoglobulinaemia.
• Circulating immune complex assays are no longer useful in practice; immune complexes are shown in tissue, by immunofluorescence of a kidney or skin biopsy (the "full house" of IgG, IgA, IgM, C3 and C1q in lupus nephritis, granular IgA in IgA vasculitis).
• Immunoglobulins: a polyclonal rise in IgG is common in lupus, Sjögren and autoimmune hepatitis and raises the ESR. Serum protein electrophoresis with immunofixation and serum free light chains look for a paraprotein when the ESR is high without explanation, in type I cryoglobulinaemia and in older patients with bone pain, where myeloma must be excluded. A raised IgG4 suggests IgG4-related disease but is not specific. IgG is measured before and during B cell depletion with rituximab, which can cause hypogammaglobulinaemia, and selective IgA deficiency makes IgA-based tests falsely negative.

False positives and how to avoid them

• testing without a clinical question, above all ANA panels in tiredness or widespread pain;
• age, female sex and relatives of patients, who carry more low-titre antibodies;
• infections: Epstein–Barr virus, hepatitis B and C, HIV, endocarditis and tuberculosis can raise ANA, rheumatoid factor, antiphospholipid antibodies and ANCA;
• drugs: anti-TNF drugs (ANA and anti-dsDNA), procainamide (ANA, anti-histone), hydralazine (ANA, anti-histone, MPO-ANCA), minocycline and propylthiouracil (ANA or MPO-ANCA);
• weak bands on line blots and low-avidity anti-dsDNA on immunoassays;
• other diseases with ANA: autoimmune thyroid disease, autoimmune hepatitis, primary biliary cholangitis, some cancers.

05Epidemiology

have an ANA at 1 in 40

1 in 3healthy adultshave an ANA at 1 in 40

at 1 in 80

1 in 8healthy adultsat 1 in 80

at 1 in 160

1 in 20healthy adultsat 1 in 160

of patients with lupus, at 1 in 80

98per centof patients with lupus, at 1 in 80

In 15 reference laboratories, healthy adults were ANA positive in about 32 per cent at a dilution of 1 in 40, 13 per cent at 1 in 80, 5 per cent at 1 in 160 and 3 per cent at 1 in 320. In a national American survey about 14 per cent of people over 12 had an ANA, women about twice as often as men, rising with age. Lupus, by contrast, affects about 1 in 1000 people or fewer in Europe. So most people with a positive ANA do not have and will not develop a connective tissue disease, and the higher the titre, the stronger the signal: a meta-analysis found the ANA's specificity for lupus rose from about 67 per cent at 1 in 40 to about 97 per cent at 1 in 320, while its sensitivity fell from 98 to 86 per cent.

Other antibodies in health:

• rheumatoid factor in a few per cent of young adults and more of older people;
• anti-CCP in only about 1 to 2 per cent of healthy people in most studies, and high titres more rarely still;
• antiphospholipid antibodies in about 1 to 5 per cent, mostly low-titre and transient;
• anti-dsDNA, anti-Sm, anti-Scl-70, anti-centromere and PR3- or MPO-ANCA rarely, so they carry far more weight when found.

How often each antibody occurs in each disease is shown in the table under clinical features; frequencies differ between ethnic groups, as anti-Sm and anti-RNP are commoner in patients of African and Asian descent, and anti-MDA5 in East Asia.

06Prognosis

Autoantibodies often appear long before the disease and predict its course:

• in stored blood from people who later developed lupus, almost 9 in 10 had at least one lupus antibody before diagnosis, on average more than three years earlier (ANA, anti-Ro, anti-La and antiphospholipid first; anti-dsDNA later; anti-Sm and anti-RNP last, closer to the onset);
• ACPA and rheumatoid factor appear years before rheumatoid arthritis, and a person with joint pains and high-titre anti-CCP has a high risk of developing it;
• most healthy people with a low-titre ANA and no symptoms never develop a systemic disease; they need no follow-up beyond advice to return with new symptoms.

Within a disease, antibodies and complement predict organ risk and flares:

Antibody levels alone rarely change treatment: patients are treated for their disease, not their numbers, and the tests help decide when to look harder.