Amyloidosis

Etiology

Amyloidosis is a disease caused by protein misfolding. These misfolded proteins infiltrate, aggregate, and form insoluble fibrils that can affect the normal function of a number of vital organs.

In the amyloidosis nomenclature, a distinction is made between amyloidosis that develops from mutations in the amyloid fibril protein itself and amyloidosis associated with genetic mutation in nonamyloid proteins. The former are referred to as hereditary amyloidoses ; examples include mutations in the genes for transthyretin and apolipoprotein A, both of which are uncommon in young children. This is in contrast to amyloid A (AA) amyloidosis , which develops in patients with chronic inflammatory states. It is estimated that, worldwide, approximately 45% of all amyloid cases are AA amyloidosis. In the past, chronic infectious diseases such as tuberculosis, malaria, leprosy, and chronic osteomyelitis accounted for most cases of AA amyloidosis. With effective treatment for these infections, other causes of AA have become more common. A number of chronic inflammatory rheumatic diseases, such as rheumatoid arthritis (RA) , juvenile idiopathic arthritis (JIA) , and ankylosing spondylitis , as well as hereditary autoinflammatory diseases, have an increased risk for the development of AA amyloidosis. AA amyloidosis has also been associated with granulomatous diseases such as sarcoidosis, cystic fibrosis, Crohn disease, malignancies such as mesothelioma and Hodgkin diseases, intravenous drug abuse, and other infections, such as bronchiectasis and HIV. Approximately 6% of AA amyloidosis cases have no identified disease association. AL amyloidosis (formerly known as idiopathic amyloidosis or myeloma-associated amyloidosis ) is extremely rare in children, occurring in middle-aged or older individuals.

Epidemiology

Only AA amyloidosis affects children in appreciable numbers. The factors that determine the risk for amyloidosis as a complication of inflammation are not clear, because many individuals with long-standing inflammatory disease do not demonstrate tissue amyloid deposition, whereas some children with relatively recent onset of disease may develop amyloid. In developed countries, before the initiation of therapy with disease-modifying antirheumatic drugs (DMARDs) and biologic agents, RA was the most common inflammatory disease associated with AA amyloidosis. Patients who had a long history of poorly controlled severe disease with extraarticular manifestations were the most at risk for developing amyloidosis, and the median time from first symptoms of their rheumatic condition to the diagnosis of amyloidosis was 212 mo. The full effect of DMARD and biologic therapy in RA-associated amyloidosis has yet to be fully appreciated, but studies are showing a sustained decline in the number of new cases.

JIA is another rheumatic disease associated with development of AA amyloidosis, with the highest prevalence in patients with systemic JIA, followed by those with polyarticular disease (see Chapter 180 ). In the pre-DMARDs and prebiologics era, the prevalence of AA amyloidosis in JIA patients ranged from 1–10%. Higher prevalence was seen in Northern European patients, especially Polish patients, who had a prevalence of 10.6%; lower prevalence was observed in North America. The reasons for this discrepancy are not completely understood, although it is speculated that selection bias, genetic background, and tendency toward earlier, more aggressive therapy in North Americans may have played a role. AA amyloidosis has been observed in JIA patients as early as 1 yr after diagnosis. Similar to RA, the occurrence of new amyloid cases has significantly decreased in the past 20 yr because of the increased efficacy of treatment with DMARDs and biologics.

The hereditary autoinflammatory diseases define a group of illnesses characterized by attacks of seemingly unprovoked recurrent inflammation without significant levels of either autoantibodies or antigen-specific T cells, which are typically found in patients with autoimmune diseases (see Chapter 188 ). Although seemingly unprovoked, these attacks are often initiated by stress, immunization, or trauma, suggesting that gene-environment interactions play an important role in pathogenesis. Although there is some variability among the autoinflammatory diseases, common findings include fevers, cutaneous rashes, arthritis, serositis, and ocular involvement. The inflammatory attacks are accompanied by intense acute-phase responses (erythrocyte sedimentation rate and C-reactive protein) and high levels of serum amyloid A (SAA). Amyloidosis AA is associated with some but not all the hereditary autoinflammatory diseases.

Familial Mediterranean fever (FMF) is the most common of the mendelian autoinflammatory diseases and is seen most frequently in the Armenian, Arab, Turkish, and Sephardi Jewish populations. FMF is an autosomal recessive disease that results from mutations in the MEFV gene, which encodes the pyrin/marenostrin protein. MEFV mutations affecting the M680 and M694 amino acid residues are associated with early onset of FMF, severe disease, and an increased risk of AA amyloidosis. Patients residing in Armenia, Turkey, and Arabian countries have an increased risk of developing AA amyloidosis compared to patients with the same mutations of MEFV living in North America. While one may assume that FMF patients who have frequent, severe attacks would be at the most risk for the development of AA amyloidosis, this is not always the case. Some patients have had a history of frequent attacks and never develop amyloidosis, and others develop amyloidosis at an early age. There is also a subset of FMF patients referred to as phenotype II . These patients present with AA amyloidosis before their first FMF attack. In this group the distribution of the common MEFV mutation is similar to that found in FMF patients with typical symptoms.

Tumor necrosis factor receptor–associated periodic syndrome (TRAPS) is associated with mutations in the TNFRSF1A gene, which encodes the 55 kDa tumor necrosis factor (TNF) receptor protein (TNFR1). It is estimated that 14-25% of patients with TRAPS develop AA amyloidosis. Patients with mutations in TNFRSF1A that affect cysteine residues have the highest risk of developing AA amyloidosis. It is thought that these cysteine residues participate in assembly of disulfide bonds important for TNFR1 folding, and that disruption of these bonds affects protein folding.

Mutations in the NLRP3 gene (also known as CIAS1 , cold-induced autoinflammatory syndrome 1) cause 3 clinically distinct diseases: familial cold autoinflammatory syndrome (FCAS) , Muckle-Wells syndrome (MWS) , and neonatal-onset multisystem inflammatory disease (NOMID) , also known as chronic infantile neurologic cutaneous and articular (CINCA) syndrome . Mutations in NLRP3 are inherited in an autosomal dominant fashion or as de novo mutations in patients with the most severe disease. A smaller portion of patients have been found to carry somatic mutations in NLRP3.

FCAS is generally the least severe of the cryopyrinopathies and is rarely associated with AA amyloidosis. MWS presents with fevers, myalgias, arthralgias, urticarial-like rash, and progressive sensorineural hearing loss. AA amyloidosis is quite common in MWS, affecting up to one third of the patients. NOMID/CINCA is the most severe cryopyrinopathy. Historically, 20% of patients died before reaching adulthood, but with current therapies, many are living longer lives. Some NOMID patients develop AA amyloidosis as they get older, although not as often as MWS patients, possibly because of a shortened life span in these patients.

Hyper-IgD syndrome (HIDS) is another autoinflammatory disease that presents in early childhood with chills, high fevers, abdominal pain, lymphadenopathy, and occasional rash. HIDS is an autosomal recessive disease that involves loss-of-function mutations in the MVK gene that encodes the mevalonate kinase enzyme. Severe MVK mutations that completely abolish enzyme activity are identified in patients with mevalonic aciduria , who present with recurrent fevers, dysmorphic features, and developmental delays. HIDS-associated mutations are milder loss-of-function mutations. Inflammatory markers, including SAA, are high during attacks and may remain elevated in the intercurrent period. AA amyloidosis is rare in HIDS but has been reported.

Although seen less frequently than in the hereditary periodic fever syndromes, the risk of AA amyloidosis has been well established in patients with Crohn disease. AA amyloidosis occurs in an estimated 1% of U.S. patients and up to 3% in Northern European patients. Conversely, AA amyloidosis presenting in patients with ulcerative colitis is extremely rare, with estimated prevalence of 0.07%. The patients have a long-standing history of aggressive, poorly controlled disease, although there are reports of amyloidosis in patients with well-controlled inflammatory markers.

Transthyretin-related hereditary amyloidosis is an autosomal dominant disorder with variable penetrance and onset in the 2nd to 3rd decade of life. More than 120 single or double mutations in the TTR gene are responsible for disease. Manifestations include neuropathy (familial amyloidotic polyneuropathy: motor, sensory, autonomic), familial amyloid cardiomyopathy, nephropathy, and ocular disease.

Pathogenesis

The deposition of AA amyloid fibrils is a result of a prolonged inflammatory state that leads to misfolding of the AA amyloid protein and deposition into tissues. The precursor protein of the fibrils in AA amyloidosis is an apolipoprotein called serum amyloid A (SAA). SAA is expressed by 3 different genes that are localized on chromosome p15.1. SAA1 and SAA2 are 2 isoforms that are acute-phase reactants synthesized by the liver that can form amyloid. SAA is produced in response to proinflammatory cytokines such as interleukin (IL)-1, IL-6, and TNF-α and can increase >1,000-fold during inflammation. It has been speculated that SAA has a role as a chemoattractant and in lipid metabolism. Supporting this theory is the finding that amyloid deposition occurs initially in organs that are major sites of lipid and cholesterol metabolism, such as the kidney, liver, and spleen. Approximately 80% of secreted SAA1 and SAA2 are bound to lipoprotein.

Under normal circumstances, SAA secreted by the liver is completely degraded by macrophages. The secreted SAA protein is 104 amino acids in length and is primarily secreted in an α-helix structure. For reasons not completely understood, patients with AA amyloidosis have a flaw resulting in incomplete degradation and accumulation of intermediate SAA products. In these patients, SAA is transferred to the lysosome, where the c-terminal portion of the SAA protein is cleaved, allowing the remaining protein to fold into a β-pleated sheet configuration. Deposited amyloid contains only 66-76 amino acids, compared to the 104 in secreted SAA. These cleaved fragments polymerize and form fibrils that are deposited in the extracellular space and bind proteoglycans and other proteins such as serum amyloid P. These fibrils then become resistant to proteolysis and deposit in organ tissues.

Development of AA amyloid may be associated with a number of risk factors. The gene encoding SAA1 has polymorphisms that, when present, carry a 3-7–fold increased risk for the development of AA amyloidosis. Caucasian patients with RA, JIA, or autoinflammatory diseases who have the SAAα/α (alpha/alpha) genotype have an increased risk of amyloidosis. In that group of patients, the SSA1γ (gamma) allele is associated with a decreased risk of amyloidosis. Interestingly, the risk in Japanese patients is reversed, with SAAα/α genotype is associated with a decreased susceptibility to amyloidosis development but the SAA1γ genotype carries an increased risk.