Guidelines on the management of <scp>AL</scp> amyloidosis

Abstract

This guideline has been compiled by members of the Guidelines Working Group of the UK Myeloma Forum on behalf of the British Committee for Standards in Haematology (BCSH). The objective of this guideline is to provide healthcare professionals with clear guidance on the management and investigation of patients with AL amyloidosis. A Medline search for literature published between 1975 and January 2014 was performed using PubMed. The search included clinical trials in AL amyloidosis and papers or reviews where AL amyloidosis was the major focus. Abstracts from relevant meetings held between 1998 and 2012 were also included. The Cochrane database did not include any relevant information. Levels of evidence and grades of recommendation are based on the GRADE system (http://www.gradeworkinggroup.org/index.htm). The GRADE system is an established internationally recognized system for grading quality of evidence and providing strength of recommendations. The draft guideline was reviewed by the UK Myeloma Forum Executive and members of the BCSH and British Society of Haematology. The guideline was then reviewed by a sounding board of approximately 50 UK haematologists and comments incorporated where appropriate. The guidance may not be appropriate to all patients with AL amyloidosis and in all cases individual patient circumstances may dictate an alternative approach. Therapy for AL amyloidosis is based on anti-myeloma therapy that suppresses the underlying plasma cell dyscrasia along with supportive measures to manage the amyloid-related complications. All patients with symptomatic systemic AL amyloidosis with visceral organ involvement, significant soft tissue involvement, coagulopathy or neuropathic involvement should be considered for treatment. It is currently unclear whether patients with isolated carpal tunnel syndrome found incidentally in the absence of any clinical evidence of organ dysfunction would benefit from early treatment. However, such patients are at very high risk of systemic progression and need close long-term follow-up. AL amyloidosis can occur in a localized form that is most often identified in the upper respiratory, urogenital and gastrointestinal (GI) tracts, the skin and the orbit. The course of the disease is relatively benign in most patients, but severe damage to the affected organ can ultimately occur. Treatment is generally confined to local surgical intervention according to symptoms. Selected patients may benefit from local radiotherapy. Treatment in patients with AL amyloidosis is generally associated with much greater treatment-related toxicity than that seen in patients with myeloma (Moreau et al, 1998; Comenzo & Gertz, 2002; Jaccard et al, 2007). There is a conflict between the greater toxicity seen in AL amyloid patients and the need to produce a rapid and near complete response. In view of this toxicity, each patient has to be individually evaluated in terms of risk and treatment needs to be tailored accordingly, often using lower doses of agents than for multiple myeloma. A palliative approach may be appropriate for some patients with very poor risk disease. Monitoring of the clonal response after each cycle of chemotherapy is important in AL amyloidosis with a view to switching to an alternative regimen as soon as the current one is proving ineffectual. However currently there is a lack of data to guide this approach although the National Amyloid Centre currently recommends this assessment after three cycles of therapy. The initial aim of chemotherapy in AL amyloidosis is to achieve an adequate and durable haematological response as rapidly as possible whilst minimizing toxicity and treatment-related mortality (TRM). Given the rarity of the disease, there is a lack of randomized controlled trials and interpretation of phase II trial data is difficult due to the heterogeneity in patient selection between trials, particularly as regards inclusion of high risk patients. Myeloma-type chemotherapy has been used to treat AL amyloidosis for over 25 years (Kyle et al, 1997; Comenzo et al, 1998; Moreau et al, 1998; Comenzo & Gertz, 2002). A selection of the more important case series reporting chemotherapy and autologous stem cell transplantation (ASCT) are shown in Tables 1 and 2, respectively. Results with the novel agents, namely bortezomib and lenalidomide-based regimens, show great promise in both newly diagnosed and relapsed/refractory AL amyloidosis (Kastritis et al, 2007, 2010; Sanchorawala et al, 2007a; Wechalekar et al, 2008). Clinical benefit from chemotherapy typically does not occur for many months even following haematological response (Kyle et al, 1997) and patients who respond slowly often do not live long enough to derive benefit. Gradual regression of AL amyloid is often seen (Hawkins, 1994), and organ function may improve even when deposits merely stabilize rather than regress. However evidence shows that patients achieving a near complete clonal response, defined as >90% dFLC response (Pinney et al, 2011) or absolute dFLC concentration after chemotherapy of <40 mg/l (Comenzo et al, 2012) [the dFLC is defined as the difference between the involved and uninvolved free light chains (FLC), providing this is >50 mg/l prior to treatment] have significantly better outcomes than patients with poorer responses (Palladini et al, 2004; Sanchorawala et al, 2005; Dispenzieri et al, 2006; Pinney et al, 2011). Given the disease heterogeneity and increased treatment-related toxicity seen in AL amyloidosis, there is a particular need for validated risk stratification of patients that can help guide treatment. Prognostic factors and staging systems, including the Mayo staging system, are discussed in the accompanying 'Guidelines on the diagnosis and investigation of AL amyloidosis' (Gillmore et al, 2014). In particular, it is important to recognize a 'high risk' group who are likely to tolerate chemotherapy very poorly, often with a need for tailoring treatment and, in some cases, with no benefit from treatment (Goodman et al, 2004; Skinner et al, 2004). It is possible to define high risk groups based on the presence one or more of the following: poor Eastern Cooperative Oncology Group performance status (ECOG PS 3 or 4), severe cardiac disease (Mayo stage III), severe salt and water retention despite aggressive diuretic therapy, severe amyloid-related autonomic neuropathy causing marked symptomatic impairment in normal activities of daily living (excluding impotence) and liver involvement by amyloid causing bilirubin >2 times upper limit of normal. A N-terminal pro-brain natriuretic peptide (NT-proBNP) >1000 pmol/l and systolic blood pressure <100 mmHg, defined a very poor risk group of stage III patients with a median survival of 3 months and with no clear benefit from chemotherapy (Wechalekar et al, 2013). However, such a definition has not been validated in any prospective clinical trial. There is no randomized trial data to guide treatment at relapse. Patients with a good duration of response who tolerate initial treatment well may be retreated with the same initial regimen. Patients with a poor response are best treated with an alternative agent combination using agents to which the patient has not been exposed, palliation or in a clinical trial tailored to the individual patient in terms of their age, comorbidities, extent of organ involvement and the patient's wishes. Lenalidomide and pomalidomide can be considered in relapsed disease although data on durability of response are limited. Toxicity with lenalidomide is a significant issue and it is recommended to start at a dose of 15 mg daily [with further dose reduction based on glomerular filtration rate (GFR) in renal failure] in AL amyloidosis patients. High dose melphalan (HDM) and autologous stem cell transplantation (ASCT) (HDM-ASCT) may be a treatment for selected patients up to 65–70 years of age with relapsed/refractory disease or with early relapse of plasma cell dyscrasia after chemotherapy. Monitoring of the clonal response after each cycle of chemotherapy is important during treatment in AL amyloidosis, with a view to switching to an alternative regimen as soon as the current one is proving ineffectual. The disease needs to be assessed in terms of response of: Measurement of serum FLC is the most effective method for monitoring the clonal disease in the majority of AL patients (Palladini et al, 2012). Serum FLC should be measured after each cycle of chemotherapy during treatment and every 1–3 months thereafter. The use of dFLC, the difference between the involved (amyloidogenic) and uninvolved light chain concentration, has recently been recommended for disease monitoring (Dispenzieri et al, 2008; Kumar et al, 2010) and is applicable with renal impairment (Pinney et al, 2011). A difference between the involved and uninvolved FLC of 50 mg/l at diagnosis has been defined as being necessary for using changes in dFLC as a disease marker (Palladini et al, 2012) and this includes about 85% of newly diagnosed AL amyloid patients. It is important to note that 10–15% of AL patients have only minimally abnormal FLC (dFLC < 50 mg/l) and for these patients FLC cannot be used for accurate monitoring. The current serum FLC assay is well established and recognized to have some variability. Newer emerging assays for FLC analysis have not been validated and studies are needed to look at comparability and sensitivity. Monitoring the intact paraprotein is often difficult in patients with amyloidosis because of a low concentration at baseline in the majority of patients. A measurable M-protein, which has been defined as >5 g/l (Palladini et al, 2012), is useful for monitoring the haematological response for those 15% of patients with minimally abnormal FLC. One to two percent of patients with AL amyloidosis lack a measurable serum or urine marker to monitor response. Follow-up bone marrow examinations are frequently unhelpful or misleading due to the subtle nature of the plasma cell dyscrasias in most patients and inherent sampling error. The value of assessing minimal residual disease using high sensitivity flow cytometry after chemotherapy is the subject of ongoing studies and may have a role in the 1–2% of patients who have no measurable serum/urine marker for monitoring. The definitions of haematological and organ responses has recently been published as a consensus guideline (Comenzo et al, 2012) and a large international series has validated some of the response measurements (Palladini et al, 2012). The assessment of response in AL amyloid is more complicated than in multiple myeloma as it requires assessment of the haematological/plasma cell clone, assessment of the organ responses and assessment of clinical outcomes including quality of life, progression-free survival (PFS) and overall survival (OS). Although the greater the haematological response, the greater is the likelihood of organ responses, there is often discrepancy between the haematological response and organ response in a significant number of patients, with some patients showing no organ response despite a significant haematological response and others showing organ responses with less than optimal haematological responses. Haematological response is defined as follows (Comenzo et al, 2012): Partial response (PR), a 50% reduction in dFLC; very good PR (VGPR), a reduction in the dFLC to <40 mg/l; complete response (CR), normal FLC levels with a normal kappa/lambda ratio and negative serum and urine immunofixation. For the 15% of patients with AL and normal FLC or dFLC <40 mg/l at baseline, standard criteria for response used in myeloma are available if the M-protein is >5 g/l. A bone marrow examination may be required as part of a clinical trial but the consensus opinion does not support this outside of a trial in order to define a CR (Comenzo et al, 2012). Progression from CR is defined as any detectable monoclonal protein or abnormal FLC ratio (involved light chain must double). Progression from PR is defined as a 50% increase in serum M protein to >5·0 g/l or 50% increase in urine M protein to >200 mg/d (a visible peak must be present) or FLC increase of 50% to >100 mg/l. Table 3 shows the updated consensus criteria for evaluation of response of amyloid-related organ dysfunction, which are based on non-invasive testing (Comenzo et al, 2012). Multiple organ biopsies to assess amyloid deposition are of no proven value, are frequently misleading and are potentially dangerous. NT-proBNP response (>30% and >35 pmol/l decrease in patients with baseline NT-proBNP ≥ 77 pmol/l) OR NYHA class response (≥2 class decrease in subjects with baseline NYHA class 3 or 4) NT-proBNP progression (>30% and >35 pmol/l increase)a OR cTn progression (≥33% increase) OR Ejection fraction progression (≥10% increase) 50% decrease (at least 0·5 g/d) of 24-h urine protein (pre-treatment urine protein must be >0·5 g/d) Creatinine and creatinine clearance must not worsen by 25% over baseline 50% increase (at least 1 g/d) of 24-h urine protein to >1 g/d OR 25% worsening of serum creatinine or creatinine clearance 50% decrease in abnormal alkaline phosphatase value Decrease in liver size radiographically at least 2 cm Melphalan and prednisolone (MP) was shown to be beneficial in the 1990s (Bradstock et al, 1978; Schwartz et al, 1979; Kyle et al, 1982, 1997; Benson, 1986; Skinner et al, 1996), providing palliation even in patients with advanced cardiac failure (Sanchorawala et al, 2002) but haematological responses are slow, taking a median of 9–12 months (Kyle et al, 1997). In the era of novel agents leading to rapid clonal responses, MP is considered to be a sub-optimal treatment in AL. The infusional VAD regimen (vincristine, adriamycin, dexamethasone) in AL amyloidosis produced combined CR and PR rates of 65% and a median survival of 80 months among 229 patients with AL amyloidosis (Sezer et al, 1999; Gono et al, 2004) but is no longer recommended given the novel agent data, the cardiotoxicity associated with anthracyclines and the problems of infusional chemotherapy. Pulsed high-dose dexamethasone has been investigated in three series, with an overall response rate in untreated patients of approximately 34% and, subsequently, dexamethasone has been the preferred steroid in combination studies although high doses are associated with significant toxicity (Gertz et al, 1999a,b; Palladini et al, 2001; Dhodapkar et al, 2004). Palladini et al (2004) treated 46 patients with advanced AL amyloidosis with oral melphalan and high dose dexamethasone (Mel-dex) and achieved haematological combined CR and PR rates of 67% within a median of 4·5 months. The same group recently reported an impressive 4·9-year median duration of clonal remission among 9/15 patients who achieved haematological CR with Mel-dex (Palladini et al, 2007). However, two recent studies with Mel-dex using the identical regimen used by Palladini have shown median survivals of <1·5 years, reflecting that these studies had more patients with advanced cardiac disease (Lebovic et al, 2008; Dietrich et al, 2010). Mel-dex was compared with HDM-ASCT in a randomized trial of AL amyloidosis patients conducted in France (Jaccard et al, 2007). Median survival among the 50 patients in the Mel-dex group was 56 months and the regimen was well tolerated with improved outcomes compared to the transplant arm although this study was criticized for poor patient selection in the transplant arm, reflected in a high TRM (Jaccard et al, 2007). The Italian group has recently reported good long term outcomes with oral Mel-dex (Palladini et al, 2014). Use of full dose dexamethasone was associated with better responses and outcomes. Haematological response rates were 76%, with 31% complete remissions in the full-dose group, and 51% (12% CR) in patients receiving the attenuated schedule. Median survival was 7·4 years in the full-dose group and 20 months in the attenuated-dose group (Palladini et al, 2014) However, this may well be due to poorer risk patients being selected for lower dose dexamethasone. At present, there is limited data on once weekly versus pulsed 4 d dexamethasone with Mel-Dex regimen but a small study suggest non-inferiority of weekly dexamethasone although overall response rates in this study were much lower (Sanchorawala et al, 2010). Mel-dex is generally well tolerated among patients with AL amyloidosis and is often a regimen of choice in patients with neuropathic disease and in elderly patients. The variable absorption of melphalan from the GI tract led Schey et al (1998) to investigate the use of intravenous melphalan (25 mg/m2) and oral dexamethasone (IDMD) in patients with untreated multiple myeloma, with high response rates being obtained. At the National Amyloidosis Centre (NAC), 144 patients with advanced disease were treated with IDMD (Goodman et al, 2004). Haematological response rates and median OS were 54% and 40 months, respectively. However, TRM was high, at 12%, with IDMD, probably reflecting the high proportion of patients with advanced disease and poor performance status at commencement of chemotherapy. Mollee et al (2012) used a similar approach but with a dose of IV melphalan of 20 mg/m2 in patients not suitable for HDM-ASCT and found this dose was too toxic, with early deaths and excessive myelotoxicity. Dietrich et al (2010) used a dose of IV melphalan of 16 mg/m2 together with dexamethasone and had less haematological toxicity, suggesting this may be a preferable dose, but the responses were poor. Melphalan is stem cell toxic and stem cell harvesting is recommended for any patient who may be a potential candidate for later HDM and PBSCT although, practically, this may not be possible and the efficacy of HDM-based therapy in patients with relapsed or refractory disease has not been studied. A phase I/II dose-escalation study of bortezomib that specifically excluded the use of corticosteroids and excluded patients with advanced heart failure looked at two different bortezomib schedules either as a conventional biweekly or a weekly regimen, producing haematological responses in approximately 69% of patients with 37·5% CR using 1·6 mg/m2 once-weekly dosing and a rapid median time to response of 2·1 months (Reece et al, 2009). Several studies have shown rates of haematological response in excess of 80% with bortezomib in patients with relapsed or refractory disease. Haematological responses were remarkably rapid in some patients, although one-third developed grade 3 toxicity or needed to discontinue bortezomib treatment (Kastritis et al, 2007; Wechalekar et al, 2008). A collaborative European study from 33 centres of 94 patients (81% of whom had received prior therapy) showed a haematological response rate of 71%, a CR rate of 25% and a cardiac response in 29% (Kastritis et al, 2010). More recently, the combination of bortezomib-dexamethasone was given to 26 patients; received this as therapy with an overall response rate of with 31% CR and a median time to response of et al, 2011). The combination of has recently been reported by two The et al, 2012) used the following bortezomib mg/m2 IV on and to mg/m2 if well mg/m2 on and and dexamethasone 20 mg on and to 20 mg for 2 d if well patients had cardiac involvement, were treated and showed an overall haematological response rate of or an of for patients treated at and for relapsed patients and an OS of The combination of was also reported by the Mayo et al, 2012) who treated patients and with a regimen of bortezomib mg/m2 and mg/m2 15 and and dexamethasone 40 mg 15 and had symptomatic cardiac involvement, with two or more involved and responses in 16 with achieving a CR and a PR with a time to response of 2 months. is seen with bortezomib compared to IV bortezomib with clinical efficacy in multiple myeloma et al, 2012). toxicity is seen with the weekly bortezomib regimen dose reduction from the standard In a small series from the bortezomib was used in patients and to be well tolerated with grade II neuropathy in two patients et al, 2012). an particularly in view of the of responses, and may be the of choice for patients with cardiac disease although further phase studies are with lower doses a weekly may better in many patients. is in multiple myeloma when combined with either melphalan or In a small phase II study of in AL amyloidosis, haematological responses were reported in of patients complete haematological but treatment-related toxicity was (Dispenzieri et al, although this study used high dose to whilst a also from the Mayo showed that lower doses of were associated with low response rates (Dispenzieri et al, 2004). Palladini et al treated patients with advanced cardiac amyloidosis with oral and dexamethasone with early haematological responses and durable in cardiac The and dexamethasone regimen has been investigated in newly diagnosed patients with AL amyloidosis, with a clonal response in organ response in grade toxicity in of patients, TRM of and rapid responses seen in all patients achieving clonal responses by the of cycle 3 of chemotherapy (Wechalekar et al, 2007). The and efficacy of has recently been compared with Mel-dex in a study by the UK Amyloidosis Treatment (Gillmore et al, randomized patients, the toxicity and haematological response rates were between the two although was associated with more rapid haematological responses than Mel-dex (Gillmore et al, In a recent analysis of patients in a prospective study of AL amyloidosis at the grade 3 or greater toxicity in approximately of patients treated with et al, 2010). The use of and dexamethasone has also been investigated in the relapsed in patients, with 15 achieving a haematological response, complete remissions and organ responses. toxicity was with only the mg/d dose and with symptomatic being a (Palladini et al, studies show efficacy for but toxicity is an issue and, if it is to start at a dose of 50 mg daily and increase slowly if has been used in the UK as therapy over the but a proportion of patients may There is no between and agents in Lenalidomide and pomalidomide are not available for use in the UK at the time of of are seen in AL amyloidosis compared to myeloma patients and include worsening of heart increase in of cardiac damage and cardiac and it has to be used with if at in patients with stage III disease. well as efficacy and rapid of has the of being stem cell and may be useful for patients who may be for HDM and Lenalidomide is a agent that has a standard treatment in multiple myeloma. phase II studies of lenalidomide with and dexamethasone in AL amyloidosis showed response rates of (Sanchorawala et al, and 67% (Dispenzieri et al, with a median response duration and OS of and months in the Mayo study (Dispenzieri et al, 2007). were seen in these two particularly and with of patients grade 3 or in the Mayo study and a recommendation was to start patients at a dose of 15 mg daily rather than the 25 mg daily dose as used in multiple myeloma (Dispenzieri et al, 2007). Lenalidomide and dexamethasone showed significant in the relapsed with of relapsed patients, all of whom had at least two of therapy, showing a haematological response toxicity is an issue with 50% grade even at a dose of 15 mg (Palladini et al, 2014). Several studies have combined lenalidomide with an either melphalan or and dexamethasone. Moreau et al (2010) a phase I/II dose study of melphalan and dexamethasone in newly diagnosed patients, with 15 mg/d of lenalidomide achieving a complete haematological response in and an the OS and survival of and respectively. More recently, two more studies using showed much lower response rates due to significant toxicity, leading to dose in cases in one study (Sanchorawala et al, and a high proportion of cardiac deaths in the et al, Kumar et al (2012) treated patients III with and dexamethasone using the following cycles of lenalidomide given at 15 mg mg/m2 given and and dexamethasone 40 mg given 15 and The overall haematological response was with with organ responses in 29% of patients, and the and the OS were and months, respectively. A grade 3 or toxicity in 26 patients including 3 haematological toxicity in 16 patients and 3 toxicity in 25 patients et al (2012) also a similar regimen in a phase I/II but used dexamethasone 20 mg on mg oral mg/d on and lenalidomide 15 mg/d on every d in the phase II and treated patients overall in phase and 65% with of patients achieving a haematological response with a median time to progression of months. The in these studies include and GI symptoms. It is clear that lenalidomide has significant in AL amyloidosis but has to be used at a dose of 15 mg daily The role of combination of lenalidomide with is but greater toxicity and as the benefit of this approach over to be are likely to be of choice in patients with AL with neuropathic disease although there are no randomized is an with in multiple myeloma including in patients refractory to bortezomib and In a phase II Dispenzieri et al (2012) treated 33 patients with relapsed AL amyloidosis with pomalidomide and dexamethasone and showed a haematological response rate with a median overall and of and months, respectively. The grade seen were and The use of HDM-ASCT following therapy in patients with AL amyloidosis was reported in the 1990s and was by in selected patients (Comenzo et al, 1998; Moreau et al, 1998; et al, 1999; et al, Dispenzieri et al, 2001; Sanchorawala et al, 2001; et al, 2004). The mortality has been and among patients with amyloidosis than those with multiple myeloma (Gertz et al, Sanchorawala et al, The reported of include cardiac failure and GI Patients with poor renal if and patients with or symptomatic cardiac amyloid have a very high TRM et al, 1999; Comenzo & Gertz, 2002). However selected patients with cardiac involvement may as shown by the Mayo series of selected patients with cardiac involvement et al, 2012). Although of the patients received melphalan the TRM was with haematological responses and cardiac responses in and of patients with a median OS of months from In analysis of baseline only melphalan OS et al, 2012). it is to HDM-ASCT in selected patients with renal failure with doses of melphalan but