Perspectives in treatment of AL amyloidosis

Authors

  • Ashutosh D. Wechalekar,

    1. National Amyloidosis Centre, Centre for Amyloidosis & Acute Phase Proteins, Department of Medicine, Royal Free and University College Medical School, London, UK
    Search for more papers by this author
  • Philip N. Hawkins,

    1. National Amyloidosis Centre, Centre for Amyloidosis & Acute Phase Proteins, Department of Medicine, Royal Free and University College Medical School, London, UK
    Search for more papers by this author
  • Julian D. Gillmore

    1. National Amyloidosis Centre, Centre for Amyloidosis & Acute Phase Proteins, Department of Medicine, Royal Free and University College Medical School, London, UK
    Search for more papers by this author

Dr Ashutosh Wechalekar, National Amyloidosis Centre, Department of Medicine, Royal Free and University College Medical School, Rowland Hill St, London NW3 2PF, UK.
E-mail: a.wechalekar@medsch.ucl.ac.uk

Summary

Light chain (AL) amyloidosis is the most frequently diagnosed form of systemic amyloid in the western world. The historically poor prognosis of AL amyloidosis appears to be improving with currently reported median survival of c. 40 months compared to 13 months in the early 1990s when low-dose oral melphalan was the mainstay of treatment. Autologous stem cell transplantation (ASCT) achieves the highest rates of complete clonal response but is confounded by substantial treatment-related mortality in AL amyloidosis unless it is restricted to highly selected patients. Newer chemotherapy regimens appear to have a balance of better safety and respectable efficacy with overall outcomes nearly similar to ASCT, but which may be used more widely. There are few data comparing durability, depth of clonal response, rate of organ response and overall survival following ASCT or chemotherapy, but a recent small, randomized trial did not suggest superiority of ASCT to oral melphalan and dexamethasone. There is a compelling need for further and larger randomized trials in this context. At the same time, various new specific anti-amyloid drugs have shown (in early phase studies or animal models) some very promising results. This review attempts to highlight the challenges, controversies and progress in AL amyloidosis.

Amyloidosis is a generic term for a group of diseases caused by misfolding and extracellular accumulation of certain proteins as fibrillar deposits that stain with Congo red and produce pathognomonic red-green birefringence when viewed by microscopy under crossed polarized light. Over 25 different proteins with amyloidogenic potential have been identified. Systemic AL amyloidosis is the most serious and most commonly diagnosed type with an age-adjusted incidence of 5·1–12·8 per million patient-year in the US (Kyle et al, 1992). Several hundred people are thought to die from AL amyloidosis in the UK each year. Historically, AL amyloidosis had a very poor prognosis with a median survival of about 13 months (Kyle et al, 1999), but recent data suggest improving outcomes in an era marked by the adoption of autologous stem cell transplantation (ASCT) and effective combination chemotherapy. The Italian amyloidosis group recently reported a median survival of 46 months in 705 patients attending their centre (Merlini & Stone, 2006). Among 600 consecutive patients with AL amyloidosis evaluated at the National Amyloidosis Centre (NAC), UK, between 1990 and 2001, the median survival increased from 1·9 years for the cohort diagnosed between 1990 and 1995 to 3·3 years for the 1996–2001 cohort (P < 0·0001). The relative impacts of earlier diagnosis, improved supportive care and more effective chemotherapy are unclear. This review attempts to highlight emerging data and conflicting opinions on various aspects of pathogenesis and management of AL amyloidosis.

Pathogenesis of amyloidosis

Amyloidogenesis involves extensive misfolding of the native structures of the various fibril precursor proteins (Kisilevsky, 2000a). A common theme is that amyloid-forming proteins are relatively unstable and have a propensity to populate partly unfolded states that can auto-aggregate in a highly ordered manner as fibrils with predominant β-sheet structure (Pepys, 2006). Additional or alternative processes may be involved in the formation of light chain amyloid deposits from which both intact light chains and light chain proteolytic products have been isolated. For example, deposition of light chains as amyloid might involve antigen-antibody type binding to nascent collagen (Harris et al, 2000). A proposed model for the amyloid light chain (AL) fibril is based on a pseudohexagonal spiral structure with a rise of approximately the width of two dimers per 360 degree turn (Schormann et al, 1995). All amyloid deposits also contain non-fibrillar elements, which may contribute to their resistance to proteolysis, including serum amyloid P (SAP) component (Pepys et al, 2002), glycosaminoglycans (GAG) (Nelson et al, 1991) and proteoglycans (Kisilevsky, 2000b). Drugs under development include those that enhance stability of fibril precursors, disrupt β-sheet structure, deplete SAP, and interfere with GAG binding, along with immunotherapy directed against amyloid fibrils, discussed further below.

AL amyloidosis and underlying plasma cell dyscrasias

Patients with AL amyloidosis have an underlying clonal dyscrasia of plasma cells (c. 94%) or of B-lymphoid/lymphoplasmacytoid cells (Kyle & Gertz, 1995; Wechalekar et al, 2005a). The clonal cell burden in AL amyloidosis is usually small and the plasma cell proliferation fraction similar to monoclonal gammopathy of undetermined significance (MGUS) (Witzig et al, 1999). About 10–20% of patients who are diagnosed with AL amyloidosis meet the criteria for myeloma (Perfetti et al, 1999) and conversely, minor amyloid deposits, frequently of no clinical significance, have been reported in up to 31% of myeloma patients (Desikan et al, 1997). Progression of the underlying monoclonal gammopathy to overt myeloma is rare in systemic AL amyloidosis (Rajkumar et al, 1998), which, no doubt in part, reflects patients’ relatively short overall survival.

The t(11;14) immunoglobulin heavy chain gene (IGH) translocation occurs more frequently in AL amyloidosis than in MGUS or myeloma (Hayman et al, 2001; Harrison et al, 2002; Wechalekar et al, 2004; Stewart & Fonseca, 2005). Other chromosomal abnormalities reported in AL amyloidosis include deletion 13q (Harrison et al, 2002), t(4;14) translocation (Perfetti et al, 2001) and ploidy changes (Fonseca et al, 1998). Certain light chain variable region (VL) genes are overrepresented in AL amyloidosis, and there is an association between VL genotypes, organ tropism (Comenzo et al, 2001) and overall outcome (Abraham et al, 2003). Little is known about the impact of chromosomal abnormalities on the phenotype of AL amyloidosis or outcome, though preliminary studies suggest AL patients with any 14q split, including t(11;14), may have shorter than average relapse free outcome (Wechalekar et al, 2004), which contrasts with myeloma in which t(11,14) is associated with neutral or better outcome (Gertz et al, 2005a). Recent paradigms have been introduced in myeloma, in which treatment is stratified according to genetic risk profile (Stewart et al, 2007), but more studies are required to determine the significance of cytogenetic abnormalities in AL amyloidosis.

Principles of treatment

Potential approaches to the treatment of AL amyloidosis are shown in Fig 1, though clinical management of amyloidosis is presently focused on reducing amyloid formation by suppressing production of the respective fibril precursor protein. Thus, treatment of AL amyloidosis comprises chemotherapy that targets the underlying clonal B-cell dyscrasia with the aim of reducing production of amyloidogenic light chains, coupled with appropriate supportive measures. Treatment regimens for AL amyloidosis have essentially been adapted from those developed in multiple myeloma, though most patients with AL amyloidosis have a low-grade plasma cell dyscrasia and small clonal burden (Gertz et al, 1991). The dose-intensive and consolidated prolonged treatment approach widely used in myeloma is evidently not required in some patients with AL amyloidosis who may achieve complete haematological remissions after just two or three cycles of chemotherapy (Wechalekar et al, 2007) or ASCT (Sanchorawala et al, 2007a). This is a crucial issue, since toxicity of chemotherapy in AL amyloidosis may be substantially greater than in myeloma due to the reduced functional reserve of amyloidotic organs and poor performance status. The challenge in AL amyloidosis therefore, is to achieve adequate haematological responses as rapidly as possible whilst minimizing morbidity and treatment-related mortality (TRM). There are very few robust prospective data comparing combination chemotherapy to ASCT, though open series suggest that overall survival may be broadly similar. Between 1994 and 2004, 394 of 701 (45%) consecutive patients with AL amyloidosis attending the Boston Amyloid Centre, USA, underwent ASCT (Skinner et al, 2004) as first line therapy, with median overall survival of 4·6 years. During a similar period in the UK, only 57 of c. 800 (7%) patients evaluated at the NAC, UK (Goodman et al, 2006) underwent ASCT, but among 220 patients who received first-line treatment with vincristine, Adriamycin and dexamethasone (VAD), median survival was 6·6 years (Goodman et al, 2005). Although open series such as these must be compared with due caution, they do highlight different approaches to the treatment of AL amyloidosis which might result in similar overall survival.

Figure 1.

 Potential therapeutic strategies in AL amyloidosis. GAG, glycosaminoglycans; SAP, serum amyloid P component; CPHPC, R-1-[6-[R-2-carboxy-pyrrolidin-1-yl]-6-oxo-hexanoyl]pyrrolidine-2-carboxylic acid.

Defining a response

Consensus response criteria in AL amyloidosis have recently been published (Gertz et al, 2005b). Responses are divided into haematological responses and organ responses, i.e. changes in clonal disease burden/monoclonal immunoglobulin production and changes in amyloidotic organ function. Improvement in organ function occurs slowly and is dependent upon haematological response to treatment, the latter being a strong predictor of survival in AL amyloidosis (Lachmann et al, 2003). Until very recently, meaningful quantification of the haematological response to therapy was not usually possible, as most patients with AL amyloidosis have very low-level paraproteins, which cannot be measured accurately, or their clonal immunoglobulin was detectable only by qualitative immunofixation electrophoresis (IFE) of serum and/or urine. Introduction of the FreeliteTM serum free light chain (FLC) assay represents a landmark advance in the management of AL amyloidosis, as it demonstrates a serially quantifiable monoclonal excess of kappa or lambda light chains in more than 95% of AL patients (Lachmann et al, 2003). Despite some criticism about its method of standardization, the FLC assay is now widely available and provides the best assessment of clonal response to treatment in most AL patients. We have reported markedly improved survival among patients whose aberrant FLC was suppressed >50% after treatment (Lachmann et al, 2003), regardless of the type of chemotherapy received.

Recent studies of ASCT in AL amyloidosis suggest that survival benefit associated with a complete FLC response is greater than that associated with a partial response (PR) (Sanchorawala et al, 2005; Dispenzieri et al, 2006), and this may be especially true for patients with dominant cardiac amyloid. However in many patients with AL amyloidosis, a partial clonal response can suffice, and identifying such patients by serial monitoring of organ function, and SAP scintigraphy where available (see below), is critical. This may be a key distinction between the management of myeloma and amyloidosis: in myeloma there is much emphasis on durability of clonal response, which is associated with complete response (CR), whereas in AL amyloidosis, a partial clonal response may halt amyloid deposition or even lead to its regression in some patients in whom attempts to achieve a complete clonal response may cause severe toxicity or death. Few robust data are yet available in AL on the relative rates of relapse among patients with partial or complete clonal responses.

We developed whole body radio-labelled SAP scintigraphy (Hawkins et al, 1988) for diagnosis and quantitative monitoring of amyloid deposits, and it is now routinely used at the NAC (Hawkins et al, 1990). SAP scans show that amyloid deposits exist in a state of dynamic turnover that varies greatly between patients, although the processes by which amyloid deposits are removed, presumably involving tissue enzymes and macrophages, are not understood. However, organ function and/or biomarkers of organ function can improve following successful haematological treatment (Palladini et al, 2006) even when amyloid deposits do not measurably regress, suggesting either that amyloidogenic precursor proteins and/or prefibrillar aggregates may themselves be harmful, or that merely halting progressive accumulation of amyloid can be beneficial.

Chemotherapy for AL amyloidosis

Conventional non-transplant chemotherapy has been used in AL amyloidosis for over 25 years and some of the more recent data are summarized in Table I.

Table I.   Selected chemotherapy studies in AL amyloidosis.
Regimen (reference)Clonal response (%)Overall survival (months)TRM (%)Toxicity (≥grade 3); (%)
  1. ns, not specified; MP, Oral Melphalan and prednisone or Melphalan, prednisone, colchicine; MPC, Oral Melphalan, prednisone and colchicine; VBMCP, Vincristine, carmustine, melphalan, cyclophosphamide, and prednisone; HDD, high dose dexamethasone; IFN, interferon; IDM - IV intermediate dose melphalan 25 mg/m2; VAD, Vincristine, Adriamycin, Dexamethasone; CTD, Cyclophosphamide, thalidomide and dexamethasone.

MP or MPC (Kyle et al, 1985)ns25·2ns 
MPC (Skinner et al, 1996)ns10·6NILns
MP or MPC (Kyle et al, 1997)2818nsns
VBMCP (Gertz et al, 1999) 3129nsns
Modified – HDD (Palladini et al, 2001)35209ns
Thalidomide (standard dose) (Seldin et al, 2003)25nsns50
Low dose melphalan (Sanchorawala et al, 2002)335·7ns (57% early death)80% hospitalized (grade not specified)
HDD followed by dexamethsone and IFN (Dhodapkar et al, 2004)5331751
Melphalan dexamethasone (Palladini et al, 2004)67Median not reachedNIL11
Thalidomide (low dose) (Dispenzieri et al, 2004a)NilnsNILns
Intermediate dose melphalan (IDM) (Goodman et al, 2004)544412ns
Thalidomide dexamethasone (Palladini et al, 2005)48nsns65
VAD (Goodman et al, 2005)65804ns
Bortezomib (Wechalekar et al, 2006)7722NILns
CTD (Wechalekar et al, 2007)74Median not reached432
Lenalidomide ± dexamethasone (Sanchorawala et al, 2007b)67nsNIL35
Lenalidomide ± dexamethasone (Dispenzieri et al, 2007)75nsNIL73
Bortezomib (Reece et al, 2007)45nsNIL42

Oral alkylators

Several early studies suggested benefit from oral alkylators in AL amyloidosis (Buxbaum et al, 1979; Kyle et al, 1982, 1985; Benson, 1986; Gertz & Kyle, 1986; Cohen et al, 1987), culminating in Kyle and colleagues’ seminal large randomized trial that confirmed modest superiority of melphalan and prednisone (MP) over colchicine, which effectively acted as a placebo (Kyle et al, 1997). However, median survival for MP-treated patients was only 18 months, and was no better in a subsequent trial of a more complex alkylator combination (Gertz et al, 1999). Another randomized trial of MP produced similar results (Skinner et al, 1996). Low dose continuous oral melphalan achieved similar overall results, though importantly demonstrated that patients who were able to receive prolonged treatment and a cumulative dose of >300 mg had a 50% chance of clonal response (Sanchorawala et al, 2002); the fact that 57% patients in this study died of progressive amyloidosis within 4 months compellingly highlights the need for a more rapid response. Haematological responses to MP, among the one-third of AL patients who do respond, take a median of 9 (Offer et al, 2005) to 12 months (Kyle et al, 1997) using FLC and conventional paraprotein criteria respectively. Significant progression of amyloidosis during such time was highlighted in one study in which many patients who received oral MP prior to planned ASCT (Sanchorawala et al, 2004) became ineligible for this due to deteriorating amyloidotic organ function.

Intermediate dose chemotherapy

Various recent studies suggest improved clonal response rates and better outcomes following ‘intermediate’ dose chemotherapy regimens when compared with ‘low dose’ oral alkylators. Pulsed high-dose dexamethasone produces high rates of response including complete haematological responses in nearly a quarter of patients, which occur much more rapidly than MP, but with much greater toxicity (Dhodapkar et al, 2004). Lower doses of dexamethasone are better tolerated but are associated with a much lower 35% overall response rate (Palladini et al, 2001). Studies in myeloma of lower doses of pulsed dexamethasone in combination with other agents showed less toxicity and mortality than high-dose dexamethasone, which may be of relevance to the treatment of amyloidosis (Rajkumar et al, 2006), especially if response rates are not compromised. Oral melphalan and dexamethasone (Mel–dex) was reported by Palladini et al (2004) who treated 46 AL patients and achieved haematological combined complete and partial response rates of 67%, and which occurred in a median of 4·5 months . A recent randomized trial yielded similar results (Jaccard et al, 2005). The regimen was well tolerated and an impressive 4·9-year median duration of clonal remission was recently reported among 9/15 patients who achieved complete responses (Palladini et al, 2007). The UK amyloid treatment guidelines (Guidelines Working Group of UK Myeloma Forum, British Committee for Standards in Haematology, British Society for Haematology, 2004) advocate combination regimens such as VAD and intermediate dose intravenous melphalan (25mg/m2) and dexamethasone (IDMD). Among patients evaluated at the NAC, haematological combined CR and PR rates and median survival were 65% and 80 months, respectively, for 229 cases treated with VAD (Goodman et al, 2005), and 54% and 40 months, respectively, for 144 patients with more advanced disease who were treated with IDMD (Goodman et al, 2004). TRM was 4% with VAD and 12% with IDMD, the latter associated with a greater proportion of patients with advanced disease and poor performance status. Experience that has been acquired since the UK amyloid treatment guidelines were published (Guidelines Working Group of UK Myeloma Forum, British Committee for Standards in Haematology., British Society for Haematology, 2004) has lately led to a substantial change in practice in the UK, comprising recommendation of cyclophosphamide, thalidomide and dexamethasone (CTD) – see below – or oral Mel–Dex as first-line treatment in lieu of VAD or IDMD in most patients. A randomized trial of CTD versus Mel–Dex is in progress at the NAC.

Thalidomide, lenalidomide and bortezomib

The advent of immunomodulatory drugs has ushered in a new era in the treatment of plasma cell disorders. Standard myeloma doses of thalidomide are tolerated poorly in AL amyloidosis, frequently necessitating early discontinuation; haematological responses occur in less than a third of patients (Dispenzieri et al, 2003; Seldin et al, 2003). Lower doses of thalidomide are better tolerated but have very limited efficacy as monotherapy (Dispenzieri et al, 2004a). Thalidomide and dexamethasone in combination is more effective but toxicity remains problematic (Palladini et al, 2005) with >60% developing ≥grade 3 toxicity. We recently reported the use of risk adapted CTD in AL amyloidosis (Wechalekar et al, 2007). This regimen achieved combined haematological CR and PR rates of 74%, which is amongst the highest of any reported non-ASCT regimen; moreover, all clonal responses occurred within 3 months and resulted in organ responses in 31% of cases (Wechalekar et al, 2007). Three-year survival among complete responders was 100% and median survival of the whole cohort was 41 months. Toxicity ≥grade 3 occurred in 32% patients, necessitating cessation of therapy in 8%; TRM was 4%. The safety and efficacy of CTD is currently being investigated as part of a prospective study at the NAC.

Bortezomib shows early promise in patients with AL amyloidosis who have relapsed or refractory clonal disease. Among 20 such patients attending our centre, 77% achieved CR or PR, although the median progression-free interval of their clonal disease was only 6 months (Wechalekar et al, 2006) and one-third developed grade 3 toxicity or needed to discontinue bortezomib treatment. The preliminary results of an ongoing dose escalating phase I/II study of bortezomib in AL amyloidosis in Europe and North America reports a lower response rate and more manageable toxicity, possibly reflecting the lower drug dosages used for the early patients (Reece et al, 2007); further results are eagerly awaited to determine the preferred starting dose of bortezomib in AL patients. In myeloma, bortezomib combination chemotherapy (Mateos et al, 2006; Manochakian et al, 2007; Palumbo et al, 2007) appears to be highly effective, raising the possibility of this approach in AL amyloidosis. Phase II results for lenalidomide with dexamethasone in AL amyloidosis are encouraging with 67% overall clonal response rates (Sanchorawala et al, 2007b), in apparent contrast to lenalidomide as a single agent, where the response rates were disappointing (Dispenzieri et al, 2007). In keeping with the general theme in AL amyloidosis, toxicity from lenalidomide is more frequent than in myeloma, although seemingly manageable and lower doses (10–15 mg) appear to be better tolerated (Dispenzieri et al, 2007; Sanchorawala et al, 2007b). Severe fluid retention and other problems that can complicate treatment of cardiac amyloidosis with thalidomide (Palladini et al, 2005), seem not to be common with lenalidomide. Skin toxicity appears to be much more frequent in patients with AL amyloidosis receiving this agent for reasons as yet unclear (Sviggum et al, 2006) but is usually not an indication for discontinuing therapy.

Stem cell transplantation for AL amyloidosis

Autologous stem cell transplantation

Efficacy of high-dose melphalan with ASCT in myeloma prompted open studies in AL amyloidosis and Table II summarizes some of the results of such studies. Anecdotal reports (Majolino et al, 1993a,b) were followed by small studies (Comenzo et al, 1996) reporting efficacy of ASCT in AL amyloidosis. A major problem in patients with amyloidosis is limited eligibility due to the risk of procedural morbidity and mortality with probably only c. 20% of cases being genuinely ‘low risk’ for TRM with ASCT, as per stringent criteria (Goodman et al, 2006). Though earlier studies reported a TRM of 20–40% (Moreau et al, 1998; Vesole et al, 2003; Mollee et al, 2004), substantially higher than in myeloma, it has lately levelled out to 10–12% in the experienced North American transplant centres (Skinner et al, 2004; Gertz et al, 2005c). Use of a risk-adapted approach further reduces the TRM markedly to less than 5% (Cohen et al, 2005) but the reduction in melphalan dose probably compromises its efficacy (Gertz et al, 2004) and needs prospective study to find an optimal dose. More stringent patient selection (Table III) is likely to achieve lower TRM but would also limit ASCT to an even smaller proportion of patients presenting with AL amyloidosis.

Table II.   Selected studies of autologous stem cell transplantation for AL amyloidosis.
ReferenceNumber of patientsHaematological response (%)Overall survivalTRM (%)
  1. CR, complete response; ns, not specified; ITT, on intention to treat analysis.

  2. *Risk adapted melphalan with adjuvant thalidomide dexamethasone.

  3. †Double autograft trial – 22 patients not in CR after first ASCT received a second ASCT.

Comenzo et al (1998)2562 (CR)68% at 24 monthsns
Moreau et al (1998)213057·5% actuarial 4-year survival43
Mollee et al (2004)205660 months35
Skinner et al (2004)35640 (CR)55 months13
Gertz et al (2005c)17168>6 years for respondersns
(Cohen et al (2005)*457976% at 18 months4·4
Jaccard et al (2005)256448 months24
Schonland et al (2005)4150 (CR)89% at 2 years7
Seldin et al (2006) (age >65 years)6532 (CR)48 months10·3
Perfetti et al (2006)2255 (36% CR)68 months14
Goodman et al (2006)916663 months23
Sanchorawala et al (2007c)6267 CR (56% ITT)Not reached8
Table III.   Criteria for eligibility for autologous Stem Cell transplantation in the current randomized UK amyloidosis treatment trial (UKATT)*.
  1. ECOG, Eastern Cooperative Oncology Group; NYHA, New York Heart Association.

  2. *Patients must satisfy all the above criteria to be eligible for a stem cell transplant. When analysed retrospectively, patients satisfying all the criteria had no TRM in the recently published UK series (Goodman et al, 2006).

ECOG performance status of 0 or 1
No greater than NYHA class I or II heart failure
No more than 2 organs involved by amyloid by consensus guidelines (Gertz et al, 2005a)
Age ≤65 years
Creatinine clearance ≥0·8 ml/s (50ml/min)
Bilirubin ≤1·5 times and alkaline phosphatase ≤2× upper limit of normal
Inter ventricular and left ventricular posterior wall thicknesses of ≤15 mm by echocardiography
Absence of clinically important amyloid related autonomic neuropathy
Absence of clinically important amyloid related gastro intestinal haemorrhage

The ASCT achieves the highest rates of complete clonal response among the current treatments for AL amyloidosis. Forty-one per cent of patients achieved a haematological CR by conventional assessments in the largest series of 394 patients from Boston (Skinner et al, 2004). Greater than 90% suppression of aberrant FLC was reported in 56% of cases who underwent ASCT at Boston amyloid centre (Sanchorawala et al, 2005). Patients with myeloma who do not achieve a complete or very good partial response after ASCT may benefit from a second ASCT (Attal et al, 2003), and using a similar rationale, this tandem approach achieved a remarkable 67% complete response rate in a study in AL amyloidosis (Sanchorawala et al, 2007c). These results are very impressive but only a few AL patients are fit enough to undergo tandem transplantation, and the long-term outcome in this group of patients is not yet known. Among AL patients who have undergone ASCT in the UK, 35% achieved haematological CR and a further 31% PR using stringent combined FLC and conventional criteria (Goodman et al, 2006). These various studies all suggest a CR rate that is much superior to the c. 20% rate achieved with most intermediate dose regimens (Palladini et al, 2004; Wechalekar et al, 2007) and the negligible rates with traditional MP (Kyle et al, 1997; Sanchorawala et al, 2002; Offer et al, 2005). Median overall survival among the 45% of eligible patients who initiated ASCT in Boston was 4·6 years (Skinner et al, 2004) and, more impressively, the median survival has not yet been reached after 10 years in the cohort who were in complete haematological remission and alive at year 1 after ASCT (Sanchorawala et al, 2007a). In the UK, median survival of patients who survived beyond day +100 after ASCT was 8·5 years (Goodman et al, 2006). Although responses to ASCT have undoubtedly been very durable in some patients, the same is true for others who have received intermediate dose regimens like oral Mel–dex (Palladini et al, 2007), and there is presently a paucity of clear comparative data on relapse-free survival in the literature.

Allogeneic stem cell transplantation in AL amyloidosis

Isolated case reports indicate that allogeneic transplantation has been performed rarely in AL amyloidosis, the first successful case being reported in 1998. (Gillmore et al, 1998). In a recent European Group for Blood and Marrow Transplantation (EBMT) report, overall and progression-free survival at 1 year among 19 cases were 60% and 53%, respectively (Schonland et al, 2006); a number of these patients had advanced disease and seven received full intensity conditioning. The overall TRM was substantial at 40%, and even higher at 50% among those who received total body radiation. Reduced-intensity allogeneic (RIC) transplantation is more appealing in AL amyloidosis as early morbidity and TRM are markedly lower than the traditional full intensity method. However, the efficacy of the proposed graft-versus-plasma cell effect on low grade AL plasma cell dyscrasias remains unknown, and there are of course concerns about morbidity caused by graft-versus-host disease and accompanying immunosuppressive treatment. There are a few case reports of RIC allografts in AL amyloidosis (Kawai et al, 2004; Imamura et al, 2006), but systematic data of any kind is lacking. At present, allogeneic stem cell transplantation is best restricted to clinical trials and highly selected patients in experienced centres.

Comparison between stem cell transplantation and chemotherapy

This is a matter of unresolved and, in all fairness, poorly informed debate. A study in France is the only prospective randomized controlled trial that has been reported, and was relatively small. The Mayo group reported superior outcome of their transplant cohort compared with historical chemotherapy controls (Dispenzieri et al, 2004b), but the latter had not received any of the promising ‘intermediate’ dose regimens described above. Type, extent and severity of organ involvement is a critical factor in determining outcome in AL amyloidosis, which is a very heterogeneous disease, and in one study patients who were eligible for ASCT but had actually been treated with chemotherapy had median overall survival of 42 months (Dispenzieri et al, 2001). NAC data shows that a median survival of VAD-treated patients who would have been eligible for ASCT [age <70 years, Eastern Cooperative Oncology Group (ECOG) performance status ≤2, number of organs ≤2 and creatinine <176·8 μmol/l; n = 105] was 8·0 years compared to 29 months among ASCT-ineligible cases. Comparisons between series of selected patients receiving different treatments in different centres at different times must be considered with much caution.

A recent trial in France (Jaccard et al, 2005), the only prospective randomized comparison between chemotherapy and ASCT, cast some doubt on the role of ASCT as first line therapy in AL amyloidosis. ASCT failed to demonstrate superiority over oral Mel–Dex chemotherapy either in terms of patient survival (48 months vs. 56 months respectively) or clonal response rates (64% vs. 65% respectively). TRM was 24% in the ASCT group (comparable with older multicentre series of ASCT but substantially higher than more recent single centre reports) versus 2% among patients receiving Mel–Dex. It is, however, likely that many of the patients would not have met the current eligibility criteria for ASCT in the UK and USA. Interestingly however, there was no difference in survival between the two groups on a landmark analysis from day +100 (i.e. excluding the effect of TRM from the analysis). A key limitation of the study was the small number of patients with only 50 cases in each arm. A larger study with adequate sample size is required to investigate this further. Data on durability of haematological response to various transplant and non-transplant chemotherapies in AL amyloidosis are extremely scanty and the hypothesized benefit of ASCT in this regard has not yet been proven. There is an improvement in the quality of life for patients with AL amyloidosis who achieve a good response after an ASCT (Seldin et al, 2004) but comparative quality of life studies have not been performed and are plainly vital in future prospective trials evaluating different approaches to treatment of this extremely unpleasant and as yet incurable disease.

Organ transplantation in AL amyloidosis

Solid organ transplantation has a role in selected patients with AL amyloidosis, particularly in those with single organ failure. However, most patients have either multisystem disease, poor performance status or are too old to be considered suitable transplant candidates. Solid organ transplantation needs to be accompanied by measures to suppress the underlying clone (such as ASCT or one of the more effective intermediate dose regimens) in order to prevent recurrence of amyloid in the graft. Cardiac transplantation holds promise for younger (<60 years) patients with predominant (and advanced) cardiac disease, no evidence of myeloma and good extra-cardiac organ function. We recently reported good outcomes for five such patients with AL amyloidosis who underwent sequential cardiac followed by autologous stem cell transplantation (Gillmore et al, 2006). Similarly, median graft survival among 16 patients with AL amyloidosis who received a renal transplant was 7·4 years despite poor clonal responses in some such cases (Wechalekar et al, 2005b). In contrast, outcome among the few patients with AL amyloidosis who have undergone orthotopic liver transplantation has generally been poor, reflecting the presence of advanced amyloid deposits in other organ systems in each case. Thus, there appears to be promising role for cardiac and renal transplantation in AL amyloidosis in carefully selected cases.

Novel approaches to amyloid treatment

Inhibiting fibrillogenesis

Eprodisate (Kiacta®; Neurochem Inc, Quebec, Canada) is a negatively charged, sulphonated molecule of low-molecular weight that has structural similarities to heparan sulphate (Kisilevsky & Szarek, 2002). Eprodisate is thought to inhibit the formation of AA amyloid fibrils by inhibiting their interaction with glycosaminoglycans (Kisilevsky et al, 2007), and in a landmark recent phase II/III placebo controlled trial in AA amyloidosis (Dember, et al 2007), treatment with this novel agent was associated with reduction by 54% of the risk of doubling serum creatinine (= 0·027), and a halving of the risk of a 50% reduction in creatinine clearance (= 0·011). Glycosaminoglycans are a universal constituent of amyloid deposits and inhibiting the interaction between GAGs and amyloid fibrils remains a promising therapeutic approach in other types of amyloidosis. Future studies with eprodisate in AL amyloidosis would be of much interest.

Enhancing regression – targeting serum amyloid P component 

Serum amyloid P binds to amyloid fibrils in vitro and protects them from degradation by phagocytic cells and proteolytic enzymes (Tennent et al, 1995). R-1-[6-[R-2-carboxy-pyrrolidin-1-yl]-6-oxo-hexanoyl]pyrrolidine-2-carboxylic acid (CPHPC) is a drug developed in our unit in collaboration with Roche (Basel, Switzerland), which cross-links pairs of SAP molecules in vivo (Pepys et al, 2002) and triggers their prompt and virtually complete clearance from the blood. Phase I/II clinical studies indicate that CPHPC is extremely well tolerated and safe, and that it results in sustained depletion of circulating SAP and substantially depletes SAP from amyloid deposits. Larger randomized controlled trials will be necessary to determine efficacy.

Immunotherapy

All amyloid fibrils share common structural motifs and an attractive strategy under investigation is the development of therapeutic antibodies to enhance their clearance. This approach has proved successful in a mouse model of AL amyloidosis, and a potentially therapeutic chimaeric antibody will shortly enter the first phase of clinical study (O’Nuallain et al, 2006). Plasma cell antigens, such as CD32B, may also be promising targets for immunotherapy in AL amyloidosis, and phase I trials of a humanized monoclonal antibody directed against this antigen are due to begin shortly (Boruchov et al, 2007).

Stabilizing amyloid fibril precursor proteins

The relative instability of fibril precursor proteins is a key property that potentiates amyloid fibrillogenesis. In vitro studies support the hypothesis that amyloid fibril precursor proteins can be stabilized by drugs that bind to them, thereby inhibiting amyloid fibril formation. Whilst this concept is yet to be developed in AL amyloidosis, two randomized, placebo-controlled trials using agents to inhibit tranthyretin amyloid fibril formation in familial amyloid polyneuropathy are in process, one using diflunisal, a non-steroidal anti-inflammatory drug (Adamski-Werner et al, 2004) and the other Fx-1006A under the sponsorship of the US company FoldRx Pharmaceuticals Inc (http://www.foldrx.com).

Current clinical trials in AL amyloidosis

A number of clinical trials in AL amyloidosis are presently attempting to address some of the questions outlined above. A comparative trial of lower doses of Mel (with dexamethasone) versus ASCT is ongoing at the Mayo Clinic in the United States. We are recruiting patients to a multicentre trial in the UK (the UK amyloidosis treatment trial – UKATT), which is a two arm randomized comparison between oral CTD versus ASCT for fitter patients (criteria provided in Table III) and oral Mel-dex versus CTD with dose attenuation as indicated for patients not satisfying these criteria. The commercially sponsored phase I/II study of bortezomib in AL amyloidosis is still recruiting and the final results are awaited with interest. The Boston amyloidosis group is studying the role of a second ASCT in patients who have relapsed or are refractory to an earlier ASCT. Use of drugs like amifostine, which can reduce chemotherapy-induced toxicity (Phillips et al, 2004), may permit higher doses of melphalan to be given during ASCT, and this is being investigated in AL amyloidosis in an ECOG study. A trial of lenalidomide, cyclophosphamide and dexamethasone in AL amyloidosis is due to start in the near future in the United States.

In summary

An approach to treatment

Selecting the optimal initial treatment in a patient with newly diagnosed AL amyloidosis remains challenging. Although few data are available from prospective controlled studies, there is increasing consensus that the critical requirement for good outcome is achievement of a substantial and rapid haematological response. At the same time it is crucial to minimize toxicity and TRM. Our own strategy at the present time has evolved to recommend one of the ‘newer’ intermediate dose regimens (risk adapted CTD or oral mel–dex) as first-line treatment in the majority of cases and with ASCT considered as an alternative option in patients who are at exceptionally low risk of TRM (as defined by criteria in Table III). Stem cell collection should be considered prior to administration of potentially stem cell toxic treatment in any patient who might subsequently undergo ASCT. Amyloidotic organ function can improve markedly following chemotherapy and patients who may not have been good candidates for ASCT at diagnosis may become eligible at a later time. There is no evidence at present that it is beneficial to continue chemotherapy far beyond the achievement of a clonal plateau or complete haematological response. It seems reasonable to consider an alternative treatment in patents that fail to show any clonal response after 3–4 cycles of intermediate dose chemotherapy, and it is plainly necessary to accept a higher risk ASCT when it is employed as second- or third-line therapy in refractory or relapsed cases. Oral MP, low-dose alkylators, single agent thalidomide and possibly single agent lenalidomide, probably have very limited roles in the treatment of AL amyloidosis. Supportive care is critical whilst awaiting chemotherapeutic clonal responses to translate into organ responses, which typically takes many months. A multidisciplinary approach with involvement of cardiologists, renal physicians and other specialists (depending on the organ systems affected), in addition to the treating haematologists, is crucial to providing optimal individualized supportive treatment.

Amyloidosis – a changing field

Just over a decade ago, AL amyloidosis was widely regarded as an untreatable and rapidly fatal disease with the majority of patients being either unfit for treatment or receiving ineffective treatment. Consequently, the median survival from diagnosis was less than 1 year. However, the field is rapidly changing. Most specialist centres now report a median survival >4 years from diagnosis and far better survival among selected patients who are diagnosed early or do not have critical organ dysfunction at the time of presentation. A number of newer chemotherapy regimens, adapted from the expanding myeloma armamentarium, have resulted in marked improvement in clonal response rates among patients with AL amyloidosis. Open labelled studies report similar survival with intermediate dose chemotherapy regimens or ASCT but large prospective trials with prolonged follow up and assessment of quality-of-life issues are urgently required to determine the preferred initial therapy. Due to the rarity and clinical heterogeneity of this disease, international collaboration will undoubtedly be crucial to achieving this goal. Novel agents that can directly target amyloid fibrils, such as CPHPC, antibody-based approaches or eprodisate appear to show great promise and may help to speed functional organ improvement. The field of amyloidosis has moved into a very exciting era with a real prospect of curing this disease on the horizon.

Acknowledgements

This study was supported by MRC Programme Grant G97900510 (P.N.H.), Wolfson Foundation, UCL Amyloidosis Research Fund and NHS Research and Development Funds.

Author contribution

ADW, JDG and PNH contributed to reviewing the data and writing the article.

Ancillary