Anti-GBM disease is generally the most rapidly progressive of all the glomerulonephritides, and it is often accompanied by alveolar haemorrhage. A substantial part of the work to understand it was done in the UK, particularly at Hammersmith Hospital from the 1970s.

Background

Anti-GBM has appeared to be the best understood of glomerular diseases since the pathogenicity of its autoantibodies was shown in 1967. Hundreds or thousands of scientific papers have modelled the disease in animals, or studied the property of anti-GBM antibodies. Through these we have learned much about mechanisms by which antibodies can (or do) recruit complement and cell-mediated mechanisms to cause glomerular damage. These in turn have led to advances in treatment, many of which were extended to other inflammatory forms of glomerulonephritis. Improved management of lupus and vasculitis, in particular, has followed experiences in this much rarer condition.

Timeline of international discoveries in anti-GBM disease (UK contributions in red) 1827  Bright associated oedematous disease with proteinuria and kidney disease 1837  Rayer identified acute haematuric variants of nephritis  1914  Volhard and Fahr described crescent formation  1919  Goodpasture reports crescentic lesions in a patient with lung haemorrhage  1942  Ellis introduced concept of RPGN (rapidly progressive glomerulonephritis)  1958  Characteristic immunofluorescence, Stanton and Tange  1967  Pathogenicity of autoantibodies proved, Lerner  1974  Effectiveness of nitrogen mustard reported by Couser  1976  Lockwood describes definitive combined therapy  1992  Cloning of human autoantigen  2023  Imlifidase shown to be effective.

A stark diagnosis

Ernest Goodpasture was working as a US naval doctor when he described a 19 year old man with fatal crescentic nephritis and lung haemorrhage  during the influenza pandemic of 1919. In retrospect we cannot know whether the patient had anti-GBM disease, but when immunofluorescence became available the Goodpasture label was applied to anti-GBM disease by Stanton and Tange (1967).

The presentation of anti-GBM disease was dramatic and rapidly fatal from irreversible renal failure, Type l respiratory failure caused by alveolar haemorrhage, or both.  The pathology was distinctive and the linear staining of the GBM with an IgG antibody pointed to a Classic Type ll  hypersensitivity reaction. This antibody caused disease in animal transfer experiments (Lerner, Glassock and Dixon 1967). Fulminant disease had such a poor and short outlook that effective treatments might quickly prove their worth.

Treatment

Bill Couser (Seattle) first described the logical treatment of nitrogen mustard (known to kill antibody-producing cells),  but on its own it was not effective.  Martin Lockwood, at Hammersmith Hospital, developed the concept of bringing together multiple treatments, including removing preformed antibodies by plasma exchange, preventing their resynthesis with cyclophosphamide (also azathioprine initially), and giving corticosteroids for their anti-lymphocyte and anti-inflammatory effects. Their first case report describing success with this approach was published in 1975 (Lockwood CM et al Br Med J 2:252-4). The patient’s biopsy showed 20% crescents (biopsy was performed by Dr DB Evans who had to come from Cambridge to perform it) but he progressed to needing dialysis. However the control of lung haemorrhage was a clear success. This success in a rare disease led to patients being referred from far and wide for years to come. Lockwood developed an in-house radioimmunoassay for anti-GBM antibodies that enabled antibody levels to be monitored.

Learning from experience

Experience of over fifty patients (ref) provided large amounts of clinical information, and a number of observations were followed up in a long series of laboratory research projects:

• Prognosis was related to the severity of the renal injury. Anuric patients with 100% crescents did not recover renal function but lung haemorrhage could be controlled. This did not stop some futile attempts.
• Smoking could provoke lung haemorrhage. This followed a dramatic Sunday afternoon when a young man with only modest kidney injury but subacute lung haemorrhage that had caused iron deficiency anaemia decided, after his morning plasma exchange session to go out into the car park and smoke 20 cigarettes to relieve his ennui. Catastrophic lung haemorrhage ensued and he was on a ventilator by teatime. Andy Rees reviewed the case records of the Hammersmith cases and found the relationship. (Donaghy and Rees Lancet 1983 2: 90-93).
• After a few months of treatment, the disease seldom relapsed.
  
• In patients presenting late without lung haemorrhage, antibody levels could remain asymptomatically high for two years or more without treatment.
• A strong class II HLA association was shown with HLA-DR2 (DR15) (Rees et al).
  
• Later review of larger numbers of patients strengthened predictive observations (Levy et al).

Molecular and immunological studies

• A monoclonal antibody to the autoantigen failed to bind to the GBM of many patients with severe Alport Syndrome (a phenomenon shown previously for anti-GBM sera from patients) (Cashman).
• The human autoantigen was shown to be a homologous, tissue-specific type IV collagen chain, alpha 3, encoded by a COL4A3 gene on chromosome 2 (Turner et al 1992).  This discovery came shortly after the discovery of the gene for X-linked Alport syndrome, COL4A5 (Trygvasson et al).
  
• In the rare circumstances of post-transplant anti-GBM disease in patients with X-linked Alport syndrome, the target of antibodies was likely to the the alpha 5 chain of type IV collagen (Brainwood et al). However in spontaneous anti-GBM disease the target was always the alpha 3 chain. 
• Animal experiments at Hammersmith and elsewhere showed the importance of cell-mediated immunity, including T lymphocytes, in the pathogenesis of anti-GBM disease (and in other forms of crescentic nephritis).
  
• Potential T cell epitopes were identified in animal models (Pusey at al).
• Analysis of antigen processing revealed that the peptides of the antigen that appeared to be most commonly generated were not the target of T cell sensitisation (Phelps et al, in work continued in Aberdeen).
  

Case vignette: an anuric Conservative MP

In the early days of anti-GBM care at Hammersmith, Sir Michael McNair Wilson (MP for Newbury South) presented with 100% crescents, was treated and had a torrid time. Two senior clinicians (Prof DK Peters and Dr Martin Lockwood) and the ward registrar (Dr David Oliveira) had advised him to have the potentially curative treatment, while the senior registrar (CGW) felt differently. So the patient’s bedside poll was 3-1 in favour of treatment.

In an interview on Radio 4 the patient was highly critical of the SR who had voted against treatment, and had gone on to recommend that the MP be referred to Oxford for home haemodialysis until the antibody disappeared of its own accord. In addition a septic shunt had  been treated with a penicillin to which he was allergic, a fact not declared to the ward registrar. The MP wanted a law making it mandatory that hospital doctors consult the patient’s GP before prescribing.

McNW survived and continued to serve as an MP while on home HD in Oxford. He was a keen advocate of the Patient’s Charter. The SR was subsequently appointed to a consultant post in Oxford, and was mischievously given responsibility for the patient’s care. McNair Wilson had a successful renal transplant some years later, later dying after an MI.

Authorship

First posted 1 Aug 2024

Last Updated on January 2, 2025 by neilturn