Diagnosis and monitoring of amyloidosis

9 Diagnosis and monitoring of amyloidosis P. N. H A W K I N S

The amyloidoses are a diverse and heterogeneous group of disorders characterized by the common finding of amyloid deposits in various tissues and organs. Diagnosis requires a high index of suspicion since clinical features of the amyloid diseases are extremely varied and non-specific. Confirmation of the diagnosis rests upon demonstration of the tissue amyloid deposits, which until recently always necessitated histopathological examination of affected tissues. Material containing amyloid gives green birefringence when it is stained with Congo red and viewed under crossed polarised light, and this characteristic universal property remains the gold standard by which other diagnostic techniques are judged in amyloidosis. Immunohistochemical staining of amyloidotic tissues remains the simplest method for identifying the amyloid fibril type. However, biopsies provide small samples of tissue which can only provide limited information on the anatomical distribution, extent and natural history of amyloid deposits. A major advance in clinical amyloidosis has been the development of radiolabelled serum amyloid P component (SAP) as an agent which specifically targets amyloid deposits in vivo. Combined scintigraphic imaging and metabolic analyses using labelled SAP have provided a wealth of new information on the natural history of many different forms of amyloid and their response to treatment (Hawkins et al, 1988a, 1990a,b, 1991a, 1993a,b, 1994a). For the first time the status of amyloid deposits throughout the body has been evaluated directly, rather than just the effects of these deposits on organ function. Most encouragingly it has been clearly demonstrated in several different forms of amyloid that reduction of the supply of the protein precursors of amyloid fibrils can be followed by major regression of the deposits. Although many anecdotal accounts of such regression in A A amyloid and a few in AL amyloid have been reported, the unavoidable limitations of biopsies meant that direct confirmation of such regression was rarely available. Use of labelled SAP to directly monitor amyloid in vivo, presages a new era in the development of treatments aimed at halting deposition of amyloid and promoting its regression. CLINICAL DIAGNOSIS OF AMYLOID Amyloid may be an unexpected histological finding on biopsy of the kidneys, liver, heart, gut, peripheral nerves, lymph nodes, skin, thyroid or bone Baillidre's Clinical Rheumatology-Vol. 8, No. 3, August 1994 ISBN 0-7020-1868-6

635 Copyright 9 1994, by Bailli~re Tindall All rights of reproduction in any form reserved

636

v.N. HAWKINS

marrow, or indeed any tissue obtained at surgery. On the other hand, certain clinical features may alert the clinician to the diagnosis of amyloidosis especially when associated conditions are present. For example, the finding of proteinuria or nephrotic syndrome in any patient with a chronic inflammatory disorder, such as rheumatoid arthritis, merits a search for A A amyloidosis, whereas similar clinical findings in a patient with monoclonal gammopathy should suggest the possibility of AL amyloid. However, with the occasional exception of hepatic or splenic enlargement patients with A A amyloid rarely have other suggestive clinical signs. By contrast in AL

Figure 1. Characteristic facial appearance of a patient with systemic AL amyloidosis showing spontaneous periorbital purpura due to amyloid infiltration of cutaneous blood vessels and connective tissue.

DIAGNOSIS AND MONITORING

637

amyloidosis there may be some highly characteristic features (Gertz and Kyle, 1989) including cutaneous purpura, especially around the eyes, and macroglossia (Figure 1). Carpal tunnel syndrome, peripheral and autonomic neuropathy, and restrictive cardiomyopathy should also strongly suggest AL amyloidosis in patients with clonal B-cell dyscrasias. Hereditary amyloid syndromes are rare, and most are characterized predominantly by peripheral and autonomic neuropathy, although varying degrees of cardiac, renal and other visceral involvement are usually present. These syndromes are caused by point mutations in the genes for various proteins including transthyretin (Nakazato et al, 1984), apolipoprotein AI (Nichols et al, 1988; Soutar et al, 1992), lysozyme (Pepys et al, 1993), and fibrinogen a-chain (Benson et al, 1993), and are transmitted in an autosomal dominant manner. Another increasingly important group of patients are those receiving longterm dialysis for end-stage renal failure, many of whom develop osteoarticular and occasionally systemic deposits of f32M amyloidosis (Gejyo et al, 1985; Gorevic et al, 1985). A high index of suspicion for amyloid should therefore be maintained in subjects with any of these predisposing clinical or genetic disorders. HISTOCHEMICAL DIAGNOSIS OF AMYLOID

Biopsy Before subjecting an individual with suspected amyloidosis to any sort of biopsy procedure, it is advisable to undertake a coagulation screen to exclude an acquired deficiency of clotting factors IX or X, which occurs occasionally in AL. It is important to also consider that there is an increased risk of bleeding in any patient with systemic amyloidosis since amyloid deposits are usually present in the walls of small blood vessels throughout the body. Biopsy of an affected major organ, for example, the kidney or heart, is likely to give a positive result in the vast majority of cases, but is invasive. It is often considered preferable to biopsy a more accessible tissue in the first place, especially in patients in whom the index of suspicion for amyloid is low. Commonly used procedures include fine-needle aspiration of abdominal subcutaneous fat (Westermark and Stenkvist, 1973) and biopsy of the rectum or labial salivary glands (Delgado and Mosqueda, 1989; Hachulla et al, 1993). In experienced hands any of these techniques may provide positive results in 80% or more patients with systemic amyloidosis (Kyle and Greipp, 1983; Libby et al, 1983; Klemi et al, 1987) whereas biopsies of other accessible sites such as the skin and gums give poorer results (Kyle and Greipp, 1983). However, a negative biopsy result by no means excludes the diagnosis of amyloid, and the positive yield from any of the above procedures may be substantially reduced in non-specialist centres. Fixation in neutral buffered formalin and wax embedding are satisfactory for routine Congo red studies but most immunohistochemical staining reactions are most intense on cryostat sections of fresh-frozen tissue.

638

P.N. HAWKINS

Histochemistry The presence of amyloid is suggested by the appearance of amorphous pink hyaline material in tissue sections that have been stained routinely with haematoxylin and eosin. Occasionally amyloid material may give green birefringence when eosin-stained sections are viewed under intense crosspolarized light. On the other hand, even quite large amounts of amyloid may be missed, particularly in connective tissues, and additional stains must always be performed to confirm the diagnosis. Congo red staining was first used to identify amyloid deposits in tissue sections in 1922 and the characteristic apple-green birefringence seen under cross-polarized filters was described in 1927. The alkaline-alcoholic Congo red method described by Puchtler et al (1962) gives very specific and consistent results and remains the pathognomonic histochemical test for amyloidosis. Congo red also stains collagen, elastic fibres and hyaline material quite strongly although in these cases polarization studies give white birefringence. The stain is unstable and must be prepared every 8 weeks or less. Unlike for most staining techniques in which tissue sections are usually cut at 2-3 ixm, the optical properties of Congo red demand a tissue thickness of 7-8 ~m for optimal results. False-negative and falsepositive results are commonly encountered, and can be minimized by the inclusion in every staining run of a positive control tissue containing modest amounts of amyloid. Interpretation of Congo red uptake in sections of myocardium, bone marrow and connective tissues can be quite difficult. Other preparations containing amyloid fibrils including synovial fluid and bone marrow aspirates, fibrils isolated from amyloidotic tissue ex vivo and even fibrils synthesized in vitro all share the same tinctorial properties with Congo red as tissue deposits. The sensitivity of Congo red histology can be optimized by using a polarization microscope equipped with strain-free objectives, well aligned high-quality Polaroid polarizing filters and a powerful quartz halide light source. Sensitivity may be enhanced further using an ultraviolet light source, under which amyloid stained with Congo red gives bright red fluorescence against a dark or greenish background, although certain slide mounting media interfere with this method. The affinity of Congo red for amyloid fibrils was previously utilized as the basis for the intravenous Congo red test; extravascular sequestration of dye into the amyloid deposits was determined by colorimetric analysis of a plasma sample obtained following injection of the substance. The technique was neither very sensitive nor particularly safe, and has now been abandoned. Other substances (Elghetany and Saleem, 1988) have been used to stain amyloid in tissues including alternative cotton dyes such as Sirius and Pagoda reds. The cationic dye, thioflavine T, and the anionic dye, thioflavine S, are fluorochromes which have been used to detect amyloid in tissue for many years, but also stain cartilage, sulphated mucosubstances, keratin, hyaline and fibrinoid material. Metachromatic stains such as crystal violet, methyl violet and toluidine blue have also been widely used but none of

DIAGNOSIS AND MONITORING

639

these methods are as reliable as Congo red for routine purposes and are not recommended. Various modifications of the Congo red method have been devised in attempts to subclassify the type of amyloid, including pretreatment of tissue sections with potassium permanganate, trypsin, alkaline guanidine and by autoclaving. These physical and chemical procedures which can differentially modulate the Congophilias of different types of amyloid are frequently misleading and have been rendered obsolete by immunohistochemical techniques. Likewise the pattern of the topographic distribution of amyloid within a particular organ is also unreliable as a guide to the type of amyloid involved (Ohyama et al, 1990; Buck and Koss, 1991).

Immunohistochemistry Antisera to most fibril precursor proteins are commercially available and immunohistochemistry always gives reliable positive results in AA and fl2M types of amyloid. In AL amyloid the deposits can be stained in only about half of all cases using commercially available antisera to whole Kor h chains, probably because the light-chain fragment from which the fibrils are formed is usually the N-terminal variable domain, and much of this is unique for each monoclonal protein. Immunohistochemical staining of transthyretin [3-protein and prion protein amyloids may require pretreatment of tissue sections with formic acid or alkaline guanidine or deglycosylation to enhance antigen availability. The demonstration of AP in all types of amyloid, using standard antibodies to SAP, is a sensitive marker for corroborating the presence of amyloid and on occasions excluding other immunoglobulin deposition diseases. It is important to appreciate that although many amyloid fibril proteins can be identified immunohistochemicaUy, the demonstration of amyloidogenic proteins in tissues does not, on its own, establish the presence of amyloid. Congo red staining must always be performed in addition. A good example of this phenomenon is the presence of amorphous non-amyloid deposits of [3-protein throughout the brains of patients with Alzheimer's disease.

Electron microscopy The appearance of amyloid fibrils in tissues under the electron microscope is not completely specific, and on occasions they cannot be convincingly identified. Although there are claims that electron microscopy is more sensitive than light microscopy, it is not sufficient by itself to confirm the diagnosis of amyloidosis. PROBLEMS OF HISTOLOGICAL DIAGNOSIS The tissue sample must be adequate, for example, submucosa and vessels should be obtained in a rectal biopsy specimen. The unavoidable problem of

640

e.N. HAWKINS

sampling error means that biopsy cannot satisfactorily reveal the extent or distribution of amyloid and failure to find amyloid does not exclude the diagnosis. Considerable experience with Congo red staining is required if clinically important false-negative and false-positive results are to be avoided. Amongst patients referred with a diagnosis of amyloid to our unit, previous rectal biopsies had yielded positive results in less than half of cases and in 5% re-examination of biopsies in our own laboratory revealed that no amyloid was actually present. Immunohistochemical staining requires positive and negative tissue and antisera controls as well as demonstration that staining is specific by absorption of positive antisera with isolated pure antigens. SAP AS A SPECIFIC TRACER IN AMYLOIDOSIS

Amyloid deposits in all different forms of the disease, both in man and in animals, contain the non-fibrillar glycoprotein amyloid P component (AP). AP is identical to and derived from the normal circulating plasma protein, SAP. As a member of the pentraxin protein family it consists of ten identical non-covalently associated subunits, each of mass 25 462 Da, which are noncovalently associated in two pentameric disc-like rings interacting face-toface. SAP is a calcium-dependent ligand binding protein, the best defined specificity of which is for the 4,6-cyclic pyruvate acetal of [3-D-galactose. It is therefore a lectin, but it also binds avidly and specifically to DNA (Pepys and Butler, 1987), to chromatin (Butler et al, 1990) and to glycosaminoglycans (GAGs) (Hamazaki, 1987), particularly heparan and dermatan sulphates, and this latter interaction may well underlie the deposition of SAP as AP in amyloid. In addition to being a plasma protein, SAP is also a normal minor constituent of certain extracellular matrix structures, and has been designated tissue AP. This protein, assumed to be derived from SAP, is covalently associated with collagen and/or other matrix components in the lamina rara interna of the human glomerular basement membrane (Dyck et al, 1980) and it is also present on the microfibrillar mantle of elastin fibres throughout the body (Breathnach et al, 1981). The normal function of SAP is not known, nor its role, in the pathogenesis of amyloidosis. However, no deficiency or evidence indicating significant polymorphism of SAP has been described and it has been conserved extremely stably in evolution. SAP is only produced by hepatocytes and its plasma concentration is tightly regulated: women, mean 24mg/litre, SD 8, range 8-55 (n=274); men, mean 32mg/litre, SD 7, range 12-50 (n = 226) (Nelson et al, 1991a). The value remains normal even during deposition of SAP into amyloid, indicating that even though SAP is not a major acute phase protein the rate of synthesis and secretion can be significantly increased (Hawkins et al, 1990b). The universal presence in amyloid deposits of AP, presumed to be derived from circulating SAP suggested the use of radioisotope-labelled SAP as a diagnostic tracer in amyloidosis. The precursor-product relationship

DIAGNOSIS AND MONITORING

641

between SAP and AP was established in the mouse model of experimentally induced A A amyloidosis (Baltz et al, 1986), and since 1987 the approach has been developed in man though it remains presently of restricted availability. Isolation and radiolabelling of SAP

For clinical studies SAP has been isolated in sterile, pure form from the plasma of single accredited donors to the British National Transfusion Service. Donors in this scheme are prospectively and rigorously screened for transmissible conditions and the plasma is heat treated before isolation of the SAP (Hawkins et al, 1988a, 1991b). The pure protein is oxidatively labelled with radioiodine using N-bromosuccinimide under conditions which preserve its function intact: 125I (half-life 13.2 h) is used for whole body scintigraphic imaging; 124I (half-life 4 days) for positron emission tomography; 125I (half-life 60 days) for metabolic studies and 131I (half-life 8 days) for whole body counting and delayed low resolution imaging studies. In each 123iodine imaging study patients receive an intravenous bolus of 100 p~g of SAP carrying about 180 MBq of radioactivity, resulting in an effective dose equivalent of less than 4 mSv after thyroid blockade. More than 800 patient studies have now been performed in our unit and no adverse effects have been encountered. Metabolism of SAP

In healthy individuals SAP is largely confined to the plasma compartment from which it is cleared with a half-time of about 24 h (Hawkins et al, 1990b). SAP is metabolized exclusively in the liver (Hutchinson et al, 1994b) and in radioiodine tracer studies the associated radioactivity is released rapidly back into the circulation, mainly in the form of iodotyrosine; there is no retention or extravascular accumulation of radioactivity, all of which can be recovered in the urine within 14 days. The metabolism of SAP is unaltered in patients undergoing an acute phase response or with other diseases. In patients with amyloidosis labelled SAP is initially cleared from the plasma more rapidly reflecting extravascular sequestration of the protein into the amyloid deposits. SAP is not significantly metabolized once it has localized to amyloid and labelled protein molecules have been shown to persist in this site unaltered for many months. The turnover studies indicate that plasma SAP is in a dynamic equilibrium with the AP within amyloid deposits; in most patients with clinically significant systemic amyloidosis the pool of SAP/AP within the amyloid is substantially larger than the normal plasma complement, by a factor of 200-fold in some cases (Figure 2). The 'localization' of up to 95% or more injected labelled SAP in patients with advanced systemic amyloidosis thus represents a dilution phenomenon: extremely small numbers of radioactive SAP molecules are distributed amongst a vast excess of unlabelled native molecules within the plasma and the amyloid deposits alike. Since the latter compartment is much larger, most of the labelled SAP becomes 'trapped' within the amyloid. The free exchange of SAP between the two compartments occurs during all phases of amyloid deposition and mobilisation, as well as during the steady state.

642

P . N . HAWKINS

Hepatic synthesis of SAP > 2 mg/hr

Circulating SAP

SAP/AP in amyloid deposits

50-100 mg

up to 20,000 mg

Catabolism

Half-life of SAP in circulation - 24 hours No significant catabolism of SAP/AP in amyloid deposits Figure 2. Schematic representation of SAP biodistribution and metabolism in a patient with extensive systemic arnyloidosis. SAP is in a free equilibrium between circulating and amyloid compartments. Whereas up to 95 % or more of the SAP in the small plasma compartment at any one moment will within 1 h have shifted into the large amyloid compartment, only a very small proportion of the SAP molecules in the latter will have moved in the opposite direction. In a labelled SAP study, the tracer is 'diluted' proportionally between both compartments and, in effect, trapped in the larger amyloid pool.

Scintigraphic studies with 123I-SAP The properties of 123I as a medium energy, short half-life, pure ~/-emitter endow this isotope with ideal characteristics as a scintigraphic imaging agent. High-resolution scans are obtained 24 h after 123I-SAP administration (Hawkins et al, 1988a, 1990a) and show that in healthy subjects and patients with diseases other than amyloidosis the tracer is confined to the circulating blood pool. Whole body images thus show the major blood vessels and organs which contain a large blood volume; degraded radioactive material can be seen in the urinary bladder. In patients with amyloidosis uptake of 123I-SAP, diagnostic of amyloid, can be seen in one or more sites in the vast majority of cases (Figure 3). Deposits can be readily identified in the kidneys, liver, spleen, adrenal glands, bone marrow and periarticular structures. The sensitivity of SAP scanning for diagnosis of systemic amyloid is 100% in A A type, 84% in AL and in excess of 95% in heredo-familial forms. The characteristic organ distribution of amyloid differs in the different forms of the disease which in some cases, for example, bone marrow deposits in patients with AL type, is completely specific (Hawkins et al, 1994a). Uptake of tracer into separate organs can be measured using computed region of interest analyses (Hawkins et al, 1993a) and whole body retention of radioactivity, an index of the total quantity of amyloid within a patient, can also be determined scintigraphically.

643

DIAGNOSIS AND MONITORING

~:, , ,

(a)

(b)

(c)

Figure 3, Whole body scintigraphs 24 h after intravenous injection of 123I-labelled human SAP. (a) Anterior view of normal control subject showing distribution of residual tracer in the blood pool and radioactive breakdown products in urine in the bladder. (b) Posterior view of patient with juvenile chronic arthritis complicated by A A amyloidosis. There is uptake of tracer in the spleen, kidneys and adrenal glands, a typical distribution of A A amyloid in which the spleen is involved in 100% of cases, kidneys in 75% and adrenals in 40%. Note the reduced blood pool and bladder signal compared to a. Interestingly this patient, whose amyloid was first diagnosed by renal biopsy 15 years ago when nephrotic syndrome developed, and who was then treated with chlorambucil, had been in complete remission for at least 10 years during which there had been no acute phase response. At the time of this scan there was no biochemical abnormality in blood or urine, despite the very appreciable amyloid deposits. This illustrates the discordance between presence of amyloid and clinical effects. (c) Anterior view of patient with monoclonal gammopathy complicated by extensive A L amyloidosis. There is uptake and retention of tracer in the liver, spleen, kidneys, bone marrow and soft tissues around the shoulder. Note the absence of blood pool or bladder signal resulting from complete uptake of the tracer dose into the substantial amyloid deposits. This pattern of amyloid distribution revealed by scintigraphy is pathognomonic for AL amyloidosis, bone marrow uptake never having been seen in any other type.

644

e.N. HAWKINS

Specificity and quantitative aspects of labelled SAP studies The binding properties of SAP to purified amyloid fibrils in vitro are characteristic of a protein-ligand interaction. Scatchard analysis suggests that the binding constant is in the order of 10 -8 litres/mol (Hawkins et al, 1991b). The mouse model of experimentally induced A A amyloid is analogous to reactive systemic amyloidosis in man, and has been used to confirm that radioiodinated pure human SAP behaves (and localizes to amyloid deposits) in vivo in a manner identical to the native unlabelled protein in fresh whole normal human serum (Hawkins et al, 1988b). Clinical scintigraphic studies in over 300 amyloid patients and 200 patients with other diseases have demonstrated that labelled SAP only localizes to amyloidotic organs and not unaffected tissues or to those involved by other disease processes. When labelled SAP scans and turnover studies are repeated in amyloid patients after a short interval, identical results are obtained. There is no evidence of any interaction between labelled SAP and normal 'tissue' SAP, and nor is there any interaction between labelled control proteins, including albumin and C reactive protein (Vigushin et al, 1993), and amyloid fibrils in vitro or in animal or clinical studies. SAP can be extracted from amyloidotic tissues in quantities which correlate with the amount of amyloid present in the tissue sample (Vigushin et al, 1994). No SAP can be extracted from normal tissue under similar conditions. Data obtained at autopsy from 15 patients with amyloidosis who, shortly beforehand, had undergone quantitative radioiodinated SAP studies, indicate that the uptake of labelled protein into tissue is proportional to both the amount of native SAP in these tissues and also to the quantity of amyloid estimated independently.

APPLICATIONS OF SAP ISOTOPE STUDIES Labelled SAP imaging and limited turnover studies can be conveniently performed together and provide diagnostic results in more than 90% of patients with systemic amyloidosis. As a non-invasive technique with radiation dosimetry well within routine limits, SAP scintigraphy is evidently suitable as a method for screening patients known to be at risk of developing amyloidosis, such as those with chronic inflammatory diseases, monoclonal gammopathies, and amyloidogenic gene mutations. Scans enable the organ distribution and quantity of amyloid to be ascertained whereas turnover studies provide a robust measurement of the whole body amyloid burden. Serial studies have provided an effective method for elucidating the natural history of the amyloidoses, and to directly monitor the effects of treatment on the amyloid deposits themselves.

AA amyloidosis SAP scintigraphy is completely specific and extremely sensitive for the diagnosis and evaluation of the tissue distribution of A A amyloid deposits

DIAGNOSIS A N D MONITORING

645

(Hawkins et al, 1993b). The spleen is always involved and is probably the first organ in which significant amyloid deposition occurs. Amyloid can be imaged in the kidneys in at least 80% of cases, the adrenal glands in up to one-half and the liver in about one-quarter (Figure 3(b)). Extensive liver and spleen uptake obscures visualization of the kidneys and adrenal glands in some patients. The scintigraphic appearance of A A amyloid deposits in patients with different underlying inflammatory conditions is similar, although in all cases there is very limited correlation between the quantity of amyloid in any organ and the resulting degree of functional impairment. For example, some 5% of patients with A A amyloidosis have extensive renal deposits but do not have significant proteinuria. Moderate impairment of splenic and adrenal function is common, although liver function is rarely disturbed until very large quantities of amyloid are present. The majority of visceral organs affected by A A amyloid are not clinically enlarged. The early and universal involvement of the spleen renders the SAP scan useful for screening patients with suspected A A amyloidosis. Subclinical deposits were identified in 5% of patients with rheumatoid arthritis attending our own rheumatology clinic. Prospective serial SAP isotope studies have shown that AA amyloid is progressive in most patients in whom the underlying inflammatory disorder remains active. The rate of accumulation, however, varies considerably, not only among different individuals but also between different organs within the same individual. In some patients with rheumatoid arthritis complicated by A A amyloidosis, the whole body load of amyloid may double over a 2-year period in the presence of a mean plasma SAA level of 100 mg/litre, whilst occasionally in others there may be no appreciable change at all. Indeed, in a few patients the quantity of amyloid may progressively increase for a period and then reach a plateau level despite there having been no change in clinical inflammatory activity or the acute phase response. This phenomenon has also been documented in one individual with subclinical amyloidosis (Figure 4). Among patients with A A amyloid in whom the underlying acute phase response had remitted, either spontaneously or following treatment, serial scans over 2-3-year period have shown an absence of progression of amyloid in about one-half of cases, and regression in the remainder. The rate of regression varied considerably from patient to patient, and some cases with very extensive deposits at presentation gave near-normal scans within 3 years. Regression of A A amyloid is accompanied by clinical benefits, for example, reduction of proteinuria, in most individuals.

AL amyloidosis In a series of 100 patients with systemic AL amyloidosis (Hawkins et al, 1994a), labelled SAP studies demonstrated that the deposits were considerably more heterogeneous with respect to their anatomical distribution, quantity, rate of progression and response to treatment than in AA type. Scans were specific, but diagnostic in only 85% of cases with histologically proven AL amyloid. Deposits were identified in the spleen in 64% of cases,

646

P.N. HAWKINS

the liver 49%, kidneys 25 %, carpal tunnel region 23 % and adrenal glands in 5%. Bone and bone marrow images were obtained in one-third of patients and have not been observed in patients with other types of amyloidosis. Disappointingly, cardiac deposits were identified only occasionally, probably as a result of motility of the heart, ventricular blood pool activity, chest wall attenuation of signal and possibly delayed passage of labelled SAP across the non-fenestrated myocardial capillary endothelium. Most AL

t

(a)

(b)

(c)

Figure 4. Screening for amyloid with 123I-SAP scintigraphy. (a) Posterior whole body scan of a 17-year-old male patient with a 10-year history of juvenile rheumatoid arthritis. The patient had a sustained major acute phase response, but no proteinuria or other features to suggest A A amyloidosis. There is abnormal uptake in the spleen, indicating early amyloid deposits in this site, (b) Scan of the same patient 18 months later. There is a substantial increase in splenic amyloid, and adrenal deposits are now clearly evident. Blood pool background is correspondingly less. There was a persistent acute phase response during the interval. (c) Scan of the same patient 30 months after presentation, There is no measurable change since the second study, despite sustained inflammatory activity. No detectable dysfunction of the spleen, adrenal glands, or kidneys had yet developed. The quantity of amyloid appears to have reached a plateau, despite the circulating SAA levels having remained steadily elevated at 250-300 mg/ litre throughout the study period.

DIAGNOSIS AND MONITORING

647

patients with negative scans had clinical evidence of amyloid limited to only one organ system, namely the heart, peripheral and autonomic nerves, lungs, muscle, skin and lymph nodes. The majority of patients with cardiac amyloid had substantial deposits in other organs which were imaged readily. There have been no studies as yet in which asymptomatic individuals with monoclonal gammopathy have been screened for AL amyloid using these techniques. Serial studies, performed at 6-monthly intervals, have been obtained in 27 patients, of whom 18 underwent cytotoxic chemotherapy. Regression of amyloid was observed in one-third of treated cases (Figure 5), and in no untreated case. An absence of progression was noted in one-third of both

Figure 5. Serial posterior whole body 123I-5AP scintigraphs of a man with AL amyloidosis presenting with hepatomegaly and massive elevation of liver alkaline phosphatase. Initially there was uptake in the spleen, liver and bone marrow, largely obscuring any renal signal. Chemotherapy comprising vincristine, adriamycin and dexamethasone was given before the second scan which shows increased spleen uptake, reduced liver uptake, and some renal uptake, but no change in total amyloid load determined by measurements of the clearance and retention of the tracer (not shown). Subsequently he suffered from recurrent splenic infarction and splenectomy was performed. Thereafter, in scan 3, there was increased tracer uptake in liver although a notably lower total amyloid load. Six months later, in scan 4, liver and kidney uptake, plasma clearance and whole body retention of tracer were all reduced, indicating regression of amyloid. This was accompanied by clinical improvement and reduction of liver alkaline phosphatase to normal levels.

648

P.N. HAWKINS

treated and untreated individuals. The rate of accumulation of amyloid in patients with progressive disease was often slower than expected given the speed with which clinical deterioration typically occurs in systemic AL amyloidosis. Preliminary observations suggest that a favourable labelled SAP study result, that is, a small whole body amyloid load, lack of progression, or regression of deposits, are indicators of a better prognosis.

Hereditary systemic amyloidosis Major visceral amyloid deposits can be imaged and quantified in most patients with familial amyloid polyneuropathy due to mutations in the gene for plasma transthyretin. Renal deposits are demonstrable by SAP scanning in the majority of cases, the spleen in two-thirds and the adrenal glands in one-third. Cardiac amyloid deposits cannot be readily identified (although they are frequently present), and the small peripheral and autonomic nerve deposits are also beyond the limits of detection of this method. A much rarer clinical syndrome is hereditary non-neuropathic systemic amyloidosis, a form of which was first described by Ostertag in 1932. This disorder is now known to be caused by various point mutations in the genes of at least three different plasma proteins, namely apolipoprotein AI, lysozyme and the a-chain of fibrinogen. Major visceral amyloid deposits, variably involving the liver, spleen, kidneys and adrenal glands have been shown by SAP scintigraphy in all affected individuals (Table 1). As in every other type of amyloidosis, the correlation between the amount of amyloid and resulting organ dysfunction in hereditary amyloidosis is very poor.

Table1. Distribution of amyloiddepositsusing 123I-SAPscintigraphy in patients with hereditary systemicamyloidosisdue to point mutationsin the genesfor apolipoproteinAI, lysozyme and the a-chain of fibrinogen. One-fifth of the cases studied had no clinical features of amyloidosis. Type of hereditary systemicamyloidosis Apolipoprotein A-I Lysozyme Fibrinogen a-chain

n 10 2 6

% positive 100 100 100

Distribution of amyloidon 123I-SAP scans

Massive spleen, liver, kidneys Massive spleen, liver, kidneys Spleen, kidneys, adrenals, liver

Extensive amyloid deposits have been demonstrated by SAP scintigraphy in many apparently healthy carriers of amyloidogenic gene mutations, some individuals having now been followed up for 5 years without clinical manifestations of disease. Longer-term studies of affected patients and gene carriers should shed considerable light on the penetrance and natural history of these disorders. Hereditary amyloidosis due to variants of transthyretin are potentially amenable to treatment by orthotopic liver transplantation since the mutant gene is expressed exclusively in this organ (Holmgren et al, 1993). So far more than 70 liver transplants have been performed for variant transthyretin-associated FAP and SAP scintigraphy has provided objective evidence of regression of visceral amyloidosis in several such cases (Figure 6).

DIAGNOSIS AND MONITORING

649

Figure 6. Serial 123I-SAP isotope studies in patient AS, with FAP due to TI'R Met-30, who was the second individual to undergo liver transplantation for this condition, in November 1990. The study in April 1990, prior to surgery, shows substantial visceral amyloid deposits. The second study, 1 year after transplantation, shows that the splenic and adrenal amyloid deposits have regressed.

152Mamyloidosis The relatively small osteoarticular deposits of [32M amyloid which occur in patients receiving long-term dialysis for end stage renal failure can be visualized with SAP imaging in most patients (Nelson et al, 1991b). Some anatomical sites are favoured, for example, the wrists, knees and feet, by being remote from the large central blood pool background signal. Asymptomatic articular deposits can be frequently identified, and systemic deposits have been seen in a few patients who were dialysed for more than 10-15 years. Retrospective and lately prospective studies (Hawkins et al, 1994b) in affected patients who have received renal transplants indicate that regression of ~2M amyloid deposits occurs in a proportion of cases.

650

P.N. HAWKINS

Other types of amyloidosis

Silent, major visceral amyloid deposits have been imaged in individuals carrying the abnormal cystatin C gene which causes hereditary cerebral haemorrhage with amyloidosis, Icelandic type (Hawkins et al, 1991a). Localized, nodular forms of AL amyloidosis are beyond the resolution of SAP scanning, and the role of 123I-SAP imaging in Alzheimer's disease and other forms of localized amyloidosis has not been extensively defined although preliminary results have been disappointing. COMPARISON OF LABELLED SAP STUDIES WITH HISTOLOGY

Biopsy histology and SAP scintigraphy are completely different and complementary techniques both of which can specifically demonstrate the presence of amyloid deposits in tissues (Vigushin et al, 1994). Labelled SAP scintigraphy surveys the whole body non-invasively and is a sensitive method for diagnosis, quantification, monitoring and, sometimes in the case of AL disease, characterizing the type of amyloidosis. Histology is more sensitive for revealing microscropic deposits and permits immunotyping of fibril proteins, but is invasive, cannot quantify the whole body amyloid load or be used for monitoring the progress or regression of deposits generally. Both techniques require specialist experience and histology performed routinely yields a significant number of false-positive and false-negative results. Although labelled SAP scans and turnover studies are still highly specialized tools of restricted availability, it is anticipated that they will become more accessible and widely used in the near future. 123I-SAP scintigraphic studies have now been performed in over 20 units worldwide, both in collaboration with our own unit and independently (Sa'ile et al, 1993), and encouragingly it has lately been possible to label SAP with 99mTc, an inexpensive and universally available isotope (see below). SINGLE PHOTON EMISSION COMPUTED TOMOGRAPHY WITH 123I-SAP

Single photon emission computed tomography (SPECT) can be performed in conjunction with any ~/-emitting isotope used in conventional planar scintigraphy. The system is relatively time consuming but enables thin transaxial, coronal and sagittal planes to be visualized. Although we have not performed SPECT studies frequently, additional and clinically useful information may be obtained in some cases. The greater sensitivity of SPECT was not, however, sufficient to image cerebrovascular deposits in hereditary cerebral haemorrhage with amyloidosis, Icelandic type. POSITRON EMISSION TOMOGRAPHY WITH 124I-LABELLED SAP

Positron emission tomography (PET) is a tomographic method in which transaxial images can be constructed with far greater resolution than with

DIAGNOSIS AND MONITORING

651

SPECT. The system is inherently superior for quantitative studies. Unfortunately, there is a paucity of positron emitting isotopes which can be used for protein labelling, and those available have relatively unfavourable properties. We have performed PET studies using 124I-labelled SAP in two patients with systemic amyloidosis (AA and AL, one case each) and cranial PET studies in six other subjects (Hawkins et al, 1991a). In systemic amyloidosis extremely high resolution images, offering fine anatomical detail, were obtained of renal A A amyloid deposits (Figure 7) and cardiac AL deposits at intervals of 24-120 h after isotope administration. Cranial PET studies were performed in two patients with familial Alzheimer's disease in whom intracerebral and cerebrovascular [3-amyloid deposits could be assumed to be present, and three patients with hereditary cerebral haemorrhage with amyloidosis, Icelandic type, a disorder characterized by the presence of cerebrovascular amyloid deposits derived from genetically variant cystatin C. In addition an apparently healthy 53-year-old man with a strong family history of Alzheimer's disease was also studied. Major constraints on isotope dosage governed by the 4-day half-life and additional undesirable high energy ~/-emissions of 1241resulted in relatively low count rates and hence poor images in these studies, but useful quantitative information was obtained in all cases. The results indicated some

Figure 7. Positron emission tomography 72 h after intravenous injection of 124I-SAP in a patient with A A amyloidosis. This abdominal transaxial image shows localization of tracer in both kidneys and very low levels of background activity.

652

P . N . HAWKINS

accumulation of the tracer, relative to the concurrently observed decline in plasma activity, in an outer cranial 'rim'. The significance of these results is unclear, particularly since the control subject gave a result similar to the cerebral amyloid cases. However, given his positive family history, subclinical cerebral amyloidosis could not be excluded; further control studies have not yet been performed. In summary, PET imaging studies with 124I-SAP offer several advantages as compared with conventional 123I-SAP scintigraphy which include inherently greater resolution, more accurate regional quantification and, because the isotope has a longer half-life, later images can be obtained with beneficial diminution of the levels of blood pool background. Disadvantages included the relatively small amount of radioactivity which could be administered given the suboptimal physical properties of 124I. Given the scarcity of 124I and expensive PET scanning equipment, these studies can presently be regarded as suitable for research purposes only. SAP SCINTIGRAPHY WITH 99mTc-LABELLED SAP The widespread application of SAP scintigraphy has been limited by the availability of clinical grade pure SAP, and the requirement for the imaging isotope 1231I. Most nuclear medicine techniques use 99mTc, a metal nuclide, which unlike 123Iis inexpensive and universally available. The chemistry for labelling proteins with metal nuclides is challenging but we have recently succeeded in labelling SAP with 99mTcvia a triamide thiolate chelating agent attached covalently to the protein (Hutchinson et al, 1994b). The99mTcSAP complex was highly resistant to competition from other chelators for the bound isotope and the capacity of the labelled SAP to undergo specific calcium-dependent ligand binding in vitro remained intact. The plasma clearance of the complex was the same as that of iodinated human SAP in normal mice, and in amyloidotic mice it showed highly specific localization to experimentally induced A A amyloid deposits. 99mTc-labelled SAP has so far been tested in three normal subjects and in three patients with systemic amyloidosis (AA, two cases; AL one case) and has shown specific uptake into amyloid deposits in the kidneys, spleen and adrenals, previously demonstrated in the same cases by 123I-SAP scans (Figure 8). Possible disadvantages may include some reduction of specific ligand binding activity in vivo, and a degree of separation of 99mTcfrom the SAP in the circulation. Unbound 99mTC is partly excreted through the biliary system and a proportion is retained in the kidneys. Further clinical studies are presently underway in our unit comparing the in vivo behaviour of 99mTc-SAP and 123I-SAP. NON-SPECIFIC CLINICAL INVESTIGATIONS IN AMYLOIDOSIS Non-specific scintigraphic studies The accumulation of bone-seeking tracers into amyloid deposits was first reported in 1975 (Van Antwerp et al, 1975) since which time there have been

DIAGNOSIS AND MONITORING

653

Figure 8. Posterior whole body scintigraph 3 h after i.v. injection of 99mTc-labelled SAP in a patient with A A amyloidosis. There is uptake of the tracer into amyloid deposits in the spleen, kidneys, adrenal glands and possibly the liver.

numerous descriptions of this property of 99mTc-labelled methylenediphosphonate and pyrophosphate compounds (Yood et al, 1981; Falk et al, 1983; Janssen et al, 1990). The rich calcium content of amyloid is thought to account for the phenomenon which has variously been described in AL, AA, FAP and haemodialysis-associated amyloidosis, although studies have shown that the calcium content of amyloidotic organs and the tracer uptake are not related in a linear manner. Soft tissue uptake of bone scanning agents has been reported into amyloid deposits in skeletal muscles, the tongue and

654

P.N. HAWKINS

heart, skin, liver, spleen, kidneys, thyroid and breast. The technique, however, is entirely non-specific, and not particularly sensitive. Furthermore it may yield completely contrasting results when repeated in the same individual at a later date (Lomena et al, 1988). The relative merits of 99mTc-MDP and 99mTc-PYP are controversial. Echocardiography has been reported to be far more sensitive for identifying cardiac amyloid than bone scanning by most groups. Another diagnostic clue to the presence of systemic amyloidosis (simply reflecting hyposplenism) is reduced splenic uptake on conventional 99mTcsulphur colloid liver-spleen scanning, especially when the spleen is of normal size or enlarged (Straub and Boyer, 1993). Another tracer which may localize non-specifically to amyloid is 67gallium citrate (Lee et al, 1986; Gertz et al, 1990). None of the above tracer studies have been generally adopted into routine clinical practice in amyloidosis, presumably because the information provided by them is too non-specific. However, these studies should suggest the possibility of amyloidosis and encourage specific clinical investigations to be pursued. Attempts have been made to develop other specific tracers for imaging amyloid deposits, the most successful of which are studies utilizing 131I-labelled 132Mas a scanning agent in anuric patients with haemodialysisassociated amyloidosis in whom the deposits are derived from ~2M (Floege et al, 1989). Labelled antibody studies have been performed with some success in experimental murine A A amyloidosis using monoclonal antibodies directed against the A A protein subunits (Marshall et al, 1986), and some preliminary clinical studies have been reported in which labelled antibodies reacting with islet amyloid polypeptide showed pancreatic uptake in patients with diabetes mellitus. Scintigraphic studies with labelled antibodies all have the major disadvantage that a host immune response to the tracer is evoked, and the proportion of labelled antibody which is taken up by the target antigen is generally less than 1% of the injected quantity. The role of scintigraphy with 9 9 m Tc-labelled pentavalent dimercaptosuccinic acid (99mTc(V)DMSA) in amyloidosis remain to be defined. This agent has been reported to localize to tumour deposits in medullary carcinoma of the thyroid, almost all of which contain localized amyloid deposits (Ohta et al, 1984; Clarke et al, 1988). Other reports suggest that this tracer may localize to AL deposits in association with plasmacytomas (Ohta et al, 1989) and in the heart (Ohta et al, 1991). In another study the tracer localized to myeloma deposits in the absence of any amyloid (Ohnishi et al, 1991), and in two of our own patients with extensive systemic A A and AL amyloidosis there was no abnormal uptake of this compound whatsoever. Other scintigraphic methods in routine use which may be useful for demonstrating the effects of amyloid, rather than imaging the deposits themselves, include gastric emptying studies in FAP and neuropathic forms of AL, and 9 9 m Tc-HMPAO cerebral blood flow studies in hereditary cerebral haemorrhage with amyloidosis, Dutch type (Haan et al, 1990). Conventional bone scans do not demonstrate the bone and bone marrow deposits of AL amyloid which are evident on aa3I-SAP scintigraphy in up to one-third of these patients.

DIAGNOSIS AND MONITORING

655

Echocardiography The various modalities of echocardiography are now firmly established as valuable, albeit non-specific, methods for evaluating cardiac amyloid. The appearance on a routine two-dimensional echocardiograph of small, stiff, concentrically hypertrophied ventricles with a sparkling 'ground glass' appearance to the ventricular wall is highly suggestive of amyloid. Suspicion of the diagnosis is supported further when these findings are associated with decreased voltages and a pseudo-infarct pattern on electrocardiography. Contemporary echocardiographic techniques, including pulsed wave Doppler studies, have shown that a spectrum of diastolic filling abnormalities are present early on in cardiac amyloidosis which precede the characteristic restrictive filling pattern (Klein et al, 1989, 1990). In advanced disease global systolic impairment may occur as well. Most echocardiographic studies have been performed in systemic AL amyloidosis. Histological evidence of AL is known to be present in more than 90% of cases, whereas significant functional or structural abnormalities can be identified using echocardiography in about 50% of patients. The mean left ventricular wall thickness is a useful guide to prognosis (CuetoGarcia et al, 1985), though may not correlate well with patients' symptoms or clinical features. Although clinically significant cardiac A A amyloidosis is very unusual, studies have noted echocardiographic abnormalities in 10% of cases (Hamer et al, 1992). Abnormalities are frequently detected in hereditary amyloidosis due to genetic variants of transthyretin (Eriksson et al, 1984). Radiology and other anatomical imaging modalities The breadth of findings reported in patients with amyloidosis studied with ultrasound, plain radiography, computed tomography and magnetic resonance imaging are beyond the scope of this article, and needless to say extensive. Although no form of anatomical imaging can show abnormalities specific for amyloid, such studies may occasionally suggest the possibility of the diagnosis, for example large hyperechoic kidneys on ultrasound or pulmonary 'fibrosis' with lymphadenopathy on chest X-ray. Clearly, however, these techniques have a more frequent role in monitoring the effects of amyloid deposition over time. SUMMARY The diagnosis of systemic amyloidosis is only occasionally suspected on clinical grounds alone and is more often considered when an associated condition, such as a chronic inflammatory disease or monoclonal gammopathy, is present. No blood test is diagnostic of amyloid although routine haematological and biochemical investigations have important roles in defining the underlying disease process in amyloidosis, and evaluating organ function. A number of non-invasive investigations including echocardiography, electrocardiography and soft tissue scintigraphy with bone-seeking

656

1,. y. HAWI~INS

tracers give characteristic results in some patients with amyloidosis, but are non-specific. The diagnosis can only be confirmed by demonstrating the presence of amyloid deposits in the tissues. Histology is the traditional method in routine clinical practice and is sensitive for revealing microscopic deposits and permits immunotyping of fibril proteins. Disadvantages are that biopsies are invasive, open to sampling error and can only give limited information on the distribution and extent of amyloid deposits in an individual. Scintigraphic and turnover studies with radioiodinated SAP are new specific methods for confirming the presence of amyloid in tissues, based on the affinity of SAP for all types of amyloid fibril. Labelled SAP scans survey the whole body macroscopically for the presence and anatomical distribution of amyloid in a quantitative manner, and SAP turnover studies provide information on the whole body amyloid load. Although the availability of SAP scintigraphy presently remains restricted, the technique has been used in over 400 patients with amyloid in prospective studies, and has already provided a number of new insights into the natural history of amyloidosis. These include the observation that there is a consistently poor correlation between the quantity of amyloid in an organ and the resulting degree of functional impairment. Amyloid deposits accumulate at rates which vary substantially between different organs in a single subject and between individuals with similar types of amyloidosis, even when the rates of amyloid fibril precursor protein supply are apparently similar. In some patients amyloid accumulation may plateau without any measurable alteration in the precursor supply. In patients with amyloidosis in whom the supply of fibril precursors is reduced, either as a result of therapy directed towards the underlying process or through a natural remission, substantial regression of amyloid frequently occurs. This has been observed in patients with A A , A L and variant TTR-associated amyloidosis, and is usually associated with clinical benefits. In some such cases, however, the function of affected organs may continue to deteriorate despite halting the accumulation of amyloid, presumably because irreversible structural damage has already occurred. These findings indicate that amyloid deposits may be mobilized from tissues and, therefore, that they may always be in a dynamic state of turnover. The natural history of most forms of amyloidosis is that deposition far exceeds the capacity for mobilization, an observation which has previously fuelled the widespread belief that amyloid deposits are inert.

REFERENCES

Baltz ML, Caspi D, Rowe IF et al (1986) Pathogeneticmechanisms and precursor product relationshipsin murineamyloidosis.In Glenner GG, OssermanEF, Benditt EP et al (eds) Amyloidosis, pp 101-108. New York: Plenum Press. BensonMD, LiepnieksJ, UemichiT et al (1993) Hereditaryrenal amyloidosisassociatedwith a mutant fibrinogenc~-chain.Nature Genetics 3: 252-255. BreathnachSM, MelroseSM, BhogalBet al (1981) AmyloidP componentis located on elastic fibre microfibrilsof normal human tissues. Nature 293: 652-654.

DIAGNOSIS AND MONITORING

657

Buck FS & Koss MN (1991) Hepatic amyloidosis: morphologic differences between systemic AL and A A types. Human Pathology 22: 904-907. Butler PJG, Tennent GA & Pepys MB (1990) Pentraxin-chromatin interactions. Serum amyloid P component specifically displaces Hl-type histones and solubilizes native long chromatin. Journal of Experimental Medicine 172: 13-18. Clarke SEM, Lazarus CR, Wraight Pet al (1988) Pentavalent [99mTc]DMSA, [131I]MIBG, and [99mTc]MDP--an evaluation of three imaging techniques in patients with medullary carcinoma of the thyroid. Journal of Nuclear Medicine 29: 33-38. Cueto-Garcia L, Reeder GS, Kyle RA et al (1985) Echocardiographic findings in systemic amyloidosis: spectrum of cardiac involvement and relation to survival. Journal of the American College of Cardiology 6: 737-743. Delgado AW & Mosqueda A (1989) A highly sensitive method for diagnosis of secondary amyloidosis by labial salivary gland biopsy. Journal of Oral Pathology and Medicine 18: 310-314. Dyck RF, Lockwood M, Kershaw M et al (1980) Amyloid P-component is a constituent of normal human glomerular basement membrane. Journal of Experimental Medicine 152: 1162-1174. Elghetany MT & Saleem A (1988) Methods for staining amyloid in tissues: a review. Stain Technology 63: 201-212. Eriksson P, Backman C, Bjerle P e t al (1984) Non-invasive assessment of the presence and severity of cardiac amyloidosis. A study in familial amyloidosis with polyneuropathy by cross sectional echocardiography and technetium-99m pyrophosphate scintigraphy. British Heart Journal 52: 321-326. Falk RH, Lee VW, Rubinow A et al (1983) Sensitivity of technetium-99m-pyrophosphate scintigraphy in diagnosing cardiac amyloidosis. American Journal of Cardiology 51: 826830. Floege J, Nonnast-Daniel B, Gielow Pet al (1989) Specific imaging of dialysis-related amyloid deposits using 131I-beta-2-microglobulin. Nephron 51: 444-447. Gejyo F, Yamada T, Odani S e t al (1985) A new form of amyloid protein associated with chronic hemodialysis was identified as [32-microglobulin. Biochemical and Biophysical Research Communications 129: 701-706. Gertz MA & Kyle RA (1989) Primary systemic amyloidosis--a diagnostic primer. Mayo Clinic Proceedings 64: 1505-1519. Gertz MA, Brown ML, Hauser MF & Kyle RA (1990) Utility of gallium imaging of the kidneys in diagnosing primary amyloid nephrotic syndrome. Journal of Nuclear Medicine 31: 292-295. Gorevic PD, Casey TT, Stone WJ et al (1985) Beta-2 microglobulin is an amyloidogenic protein in man. Journal of Clinical Investigation 76: 2425-2429. Haan J, Van Kroonenburgh MJPG, Algra PR et al (1990) Hereditary cerebral hemorrhage with amyloidosis--Dutch type. Tc-99m HM-PAO single photon emission computed tomography. Neuroradiology 32: 142-145. Hachulla E, Janin A, Marc Flipo R et al (1993) Labial salivary gland biopsy is a reliable test for the diagnosis of primary and secondary amyloidosis. A prospective clinical immunohistologic study in 59 patients. Arthritis and Rheumatism 36: 691-697. Hamazaki H (1987) Ca2---mediated association of human serum amyloid P component with heparan sulfate and dermatan sulfate. Journal of Biological Chemistry 262: 1456-1460. Hamer JPM, Janssen S, Van Rijswijk MH & Lie KI (1992) Amyloid cardiomyopathy in systemic non-hereditary amyloidosis. Clinical, echocardiographic and electrocardiographic findings in 30 patients with A A and 24 patients with AL amyloidosis. European Heart Journal 13: 623-627. Hawkins PN, Myers M J, Lavender JP & Pepys MB (1988a) Diagnostic radionuclide imaging of amyloid: biological targeting by circulating human serum amyloid P component. Lancet h 1413-1418. Hawkins PN, Myers MJ, Epenetos A A et al (1988b) Specific localization and imaging of amyloid deposits in vivo using 123I-labeled serum amyloid P component. Journal of Experimental Medicine 167: 903-913. Hawkins PN, Lavender JP & Pepys MB (1990a) Evaluation of systemic amyloidosis by scintigraphy with a23I-labeled serum amyloid P component. New England Journal of Medicine 323: 508-513.

658

P . N . HAWKTNS

Hawkins PN, Wootton R & Pepys MB (1990b) Metabolic studies of radioiodinated serum amyloid P component in normal subjects and patients with systemic amyloidosis. Journal of Clinical Investigation 86: 1862-1869. Hawkins PN, Tyrrell P, Jones T et al (1991a) Metabolic and scintigraphic studies with radiolabeled serum amyloid P component in amyloidosis: applications to cerebral deposits and Alzheimer disease with positron emission tomography. Bulletin of Clinical Neuroscience 56: 178-190. Hawkins PN, Tennent GA, Woo P & Pepys MB (1991b) Studies in vivo and in vitro of serum amyloid P component in normals and in a patient with A A amyloidosis. Clinical and Experimental Immunology 84: 308-316. Hawkins PN, Richardson S, MacSweeney JE et al (1993a) Scintigraphic quantification and serial monitoring of human visceral amyloid deposits provide evidence for turnover and regression. Quarterly Journal of Medicine 86: 365-374. Hawkins PN, Richardson S, Vigushin DM et al (1993b) Serum amyloid P component scintigraphy and turnover studies for diagnosis and quantitative monitoring of A A amyloidosis in juvenile rheumatoid arthritis. Arthritis and Rheumatism 36: 842-851. Hawkins PN, Vigushin DM, Richardson Set al (1994a) Evaluation of 100 cases of systemic AL amyloidosis by serum amyloid P component (SAP) scintigraphy. In Kisilevsky R et al (eds) Amyloid and Amyloidosis, (in press). Pearl River, NY: Parthenon Publishing. Hawkins PN, Tan SY & Pepys MB (1994b) General aspects of amyloidosis. In Van Ypersele C & Driieke T (eds) Oxford Monographs on Clinical Nephrology: Dialysis Amyloidosis (in press). Oxford: Oxford University Press. Holmgren G, Ericzon B-G, Groth C-G et al (1993) Clinical benefit and regression of amyloidosis following liver transplantation in patients with familial amyloid polyneuropathy due to transthyretin variant Met-30. Lancet 341: 1113-1116. Hutchinson WL, Noble GE, Hawkins PN & Pepys MB (1994a) Serum amyloid P component is taken up and catabolised in hepatocytes. In Kisilevsky R et al (eds) Amyloid and Amyloidosis, (in press). Pearl River, NY: Parthenon Publishing. Hutchinson WL, Mather SJ, Staltieri Met al (1994b) Scintigraphic imaging of amyloid deposits with 99mTc-labelled serum amyloid P component. In Kisilevsky R et al (eds) Amyloid and Amyloidosis, (in press). Pearl River, NY: Parthenon Publishing. Janssen S, Piers DA, Van Rijswijk MH et al (1990) Soft-tissue uptake of 99mTc-diphosphonate and 99mTc-pyrophosphate in amyloidosis. European Journal of Nuclear Medicine 16: 663-670. Klein AL, Hatle LK, Burstow DJ et al (1989) Doppler characterization of left ventricular diastolic function in cardiac amyloidosis. Journal of the American College of Cardiology 13: 1017-1026. Klein AL, Hatle LK, Burstow DJ et al (1990) Comprehensive Doppler assessment of right ventricular diastolic function in cardiac amyloidosis. Journal of the American College of Cardiology 15: 99-108. Klemi PJ, Sorsa S & Happonen RP (1987) Fine-needle aspiration biopsy from subcutaneous fat. An easy way to diagnose secondary amyloidosis. Scandinavian Journal of Rheumatology 16: 429-431. Kyle RA & Greipp PR (1983) Amyloidosis (AL): clinical and laboratory features in 229 cases. Mayo Clinic Proceedings 58: 665-683. Lee VW, Skinner M, Cohen AS et al (1986) Renal amyloidosis. Evaluation by gallium imaging. Clinical Nuclear Medicine 11: 642-646. Libby CA, Skinner M & Cohen AS (1983) Use of abdominal fat tissue aspirate in the diagnosis of systemic amyloidosis. Archives oflnternal Medicine 143: 1549-1552. Lomena F, Rosello R, Pons F et al (1988) Abnormal scintigraphic evolution in A A hepatic amyloidosis. Clinical Nuclear Medicine 13: 194-196. Marshall J, MeNally W, Muller D et al (1986) In vivo radioimmunodetection of amyloid deposits in experimental amyloidosis. In Glenner GG, Osserman EF, Benditt EP et al (eds) Amyloidosis, pp 163-174. New York: Plenum Press. Nakazato M, Kangawa K, Minamino N et al (1984) Identification of a prealbumin variant in the serum of a Japanese patient with familial amyloidotic polyneuropathy. Biochemical and Biophysical Research Communications 122: 712-718. Nelson SR, Tennent GA, Sethi D et al (1991a) Serum amyloid P component in chronic renal failure and dialysis. Clinica Chimica Acta 200: 191-200.

DIAGNOSIS AND MONITORING

659

Nelson SR, Hawkins PN, Richardson S e t al (1991b) Imaging of haemodialysis-associated amyloidosis with 131I-serum amyloid P- component. Lancet 338: 335-339. Nichols WC, Dwulet FE, Liepnieks J & Benson MD (1988) Variant apolipoprotein AI as a major constituent of a human hereditary amyloid. Biochemical and Biophysical Research Communications 156: 762-768. Ohnishi T, Noguchi S, Murakami Net al (1991) Pentavalent technetium-99m-DMSA uptake in a patient having multiple myeloma without amyloidosis. Journal of Nuclear Medicine 32: 1785-1787. Ohta H, Yamamoto K, Endo K et al (1984) A new imaging agent for medullary carcinoma of the thyroid. Journal of Nuclear Medicine 25: 323-325. Ohta H, Endo K, Kanoh T et al (1989) Technetium-99m (V) DMSA uptake in amyloidosis. Journal of Nuclear Medicine 30" 2049-2052. Ohta H, Okada T, Furukawa Y e t al (1991) Tc-99m(v) DMSA uptake in cardiac amyloidosis. Clinical Nuclear Medicine 16: 67%4575. Ohyama T, Shimokama T, Yoshikawa Y & Watanabe T (1990) Splenic amyloidosis: correlations between chemical types of amyloid protein and morphological features. Modern Pathology 3: 419-422. Ostertag B (1932) Demonstration einer eigenartigen familiaren paraamyloidose. Zentralbl. Aug. Patol. 56: 253-254. Pepys MB & Butler PJG (1987) Serum amyloid P component is the major calcium-dependent specific DNA binding protein of the serum. Biochemical and Biophysical Research Communications 148: 308-313. Pepys MB, Hawkins PN, Booth DR et al (1993) Human lysozyme gene mutations cause hereditary systemic amyloidosis. Nature 362: 553-557. Puchtler H, Sweat F & Levine M (1962) On the binding of Congo red by amyloid. Journal of Histochemistry and Cytochemistry 10: 355-364. Saile R, Deveaux M, Hachulla E et al (1993) Iodine-123-1abelled serum amyloid P component scintigraphy in amyloidosis. European Journal of Nuclear Medicine 20: 130-137. Soutar AK, Hawkins PN, Vigushin DM et al (1992) Apolipoprotein AI mutation Arg-60 causes autosomal dominant amyloidosis. Proceedings of the National Academy of Sciences of the USA 89: 7389-7393. Straub RF & Boyer TD (1993) The liver-spleen scan in systemic amyloidosis: a clue to the diagnosis. Journal of Clinical Gastroenterology 16: 340-343. Van Antwerp JD, O'Mara RE, Pitt MJ & Walsh S (1975) Technetium-99m diphosphonate accumulation in amyloid. Journal of Nuclear Medicine 16: 238-240. Vigushin DM, Pepys MB & Hawkins PN (1993) Metabolic and scintigraphic studies of radioiodinated human C-reactive protein in health and disease. Journal of ClinicaIInvestigation 91: 1351-1357. Vigushin DM, Pepys MB & Hawkins PN (1994) Comparison of histology with SAP scintigraphy for evaluation of amyloidosis. In Kisilevsky R et al (eds) Amyloid andAmyloidosis, (in press). Pearl River, NY: Parthenon Publishing. Westermark P & Stenkvist B (1973) A new method for the diagnosis of systemic amyloidosis. Archives of lnternal Medicine 132: 522-523. Yood RA, Skinner M, Cohen AS & Lee VW (1981) Soft tissue uptake of bone seeking radionuclide in amyloidosis. Journal of Rheumatology 8: 760-766.