How I treat AL amyloidosis

Systemic immunoglobulin light chain (AL) amyloidosis is caused by the conversion of light chains (LCs) from their soluble states into highly organized fibrillar aggregates that deposit in tissues, resulting in progressive organ damage and dysfunction.¹ Rapid organ deterioration imposes early diagnosis and prompt effective therapies to halt disease progression and possibly rescue organ function.² LCs, LC aggregates, or preceding intermediaries may induce proteotoxicity, leading to cell dysfunction and death, as shown in cardiac cells.^3-5 Cardiomyocytes, in response, produce high levels of natriuretic peptide type-B and its N-terminal fragment (BNP and NT-proBNP, respectively) that are pivotal for patient management.⁶ Furthermore, amyloid fibrils may cause cell damage and distortion of tissue architecture contributing to organ dysfunction.⁷ Accordingly, although experimental strategies aiming at reducing amyloid load are being developed, current therapy aims at suppressing the synthesis of the amyloid protein while adapting to patient’s frailty.

Structural features of LCs underlie their amyloidogenicity. Approximately 80% are of the λ isotype, and a limited pool of LC variable region genotypes are responsible for the production of amyloidogenic free LCs (FLC), consistent with a biased selection.^8-11 Specific LC germline use can partly explain organ tropism.^11,12 Somatic mutations, acquired during clonal selection, determine LC higher flexibility, kinetic instability, and a higher dynamic state that underlie amyloidogenicity and tissue toxicity.^13-16 Studies are ongoing to develop LC stabilizers expected to inhibit fibril formation and proteotoxicity.^17,18N-glycosylation of κ LCs is more common in patients with AL amyloidosis than in those with monoclonal gammopathy of undetermined significance (MGUS) or multiple myeloma (MM) and can be readily detected in serum by mass spectrometry.^19,20 These genetic and posttranslational characteristics may be exploited, possibly through machine learning,²¹ for detecting patients with plasma cell (PC) dyscrasias at high risk for developing AL amyloidosis, facilitating early diagnosis.²² 

Several studies have delineated the biologic characteristics of the B-cell/PC clone that is usually small and indolent.²³ Amyloidogenic PCs are vulnerable to proteasome inhibitors, accounting for improved clinical outcomes after the introduction of bortezomib.^24,25 Cytogenetic abnormalities are detected in approximately 80% of patients via fluorescent in situ hybridization (FISH). t(11;14) is the most frequent (40% to 60% of patients) and is associated with poorer hematologic response rates and overall survival with bortezomib and dexamethasone with or without cyclophosphamide.²⁶ The BCL-2 inhibitor venetoclax is very effective in patients with t(11;14),²⁷ and larger, controlled studies of this agent are warranted. Gain(1q21) is present in 15% to 20% of patients and is associated with poor response to oral melphalan.²⁸ Compared with MGUS/MM, AL amyloidosis shows less intraclonal heterogeneity²⁹ and different transcriptional programs,³⁰ suggesting that AL amyloidosis may be more amenable to eradication than MM.

A 58-year-old, very active man, with low-intermediate risk immunoglobulin G (IgG) λ MGUS diagnosed 1 year earlier, presented at the second yearly control with normal hematology and biochemistry values. Serum monoclonal spike was stable (8 g/L), as well as FLC concentration (λ, 180 mg/L; κ, 20 mg/L; ratio, 0.11), and proteinuria was 220 mg/d with urine immunofixation electrophoresis showing monoclonal λ FLC. Bone marrow biopsy showed 8% λ typic plasma cells, and FISH analysis showed t(11;14) in 35% of interphases. The patient was asymptomatic and working full time. The finding of a significant increase of NT-proBNP to 745 ng/L from the previous 114 ng/L raised the suspicion of heart involvement, and an appropriate workup was started. The abdominal fat aspirate showed amyloid deposits that were typed as AL λ by mass spectrometry. Echocardiography was normal. Cardiac magnetic resonance was consistent with early cardiac amyloidosis. Cardiac troponin I was 0.01 ng/mL, and the patient was classified as cardiac stage³¹ II. We thoroughly discussed the opportunity to start anticlone therapy, carefully considering toxicity and benefit balance. The patient decided to start therapy, and because he was eligible for autologous stem cell transplantation (ASCT), we started treatment with cyclophosphamide, bortezomib, and dexamethasone (CyBorD). During the first course, asymptomatic orthostatic hypotension (80/50 mm Hg) occurred but resolved spontaneously. After 4 courses, a very good partial response (VGPR) was achieved (λ, 48 mg/L; κ, 17 mg/L; difference between involved and uninvolved FLC [dFLC], 31 mg/L; positive serum and urine immunofixation). Hemopoietic stem cells were harvested uneventfully. However, 3 months later, NT-proBNP increased to 954 ng/L, and the patient reported unusual fatigue after strenuous physical exercise while maintaining VGPR. The patient consented to proceed to high-dose melphalan (200 mg/m²). The procedure was uneventful. Three months after ASCT, he achieved complete hematologic response (CR), with reduction of NT-proBNP to 584 ng/L, thus qualifying for cardiac response. Eighteen months after ASCT, the patient is in CR, with negative minimal residual disease (MRD) by next-generation flow cytometry, and NT-proBNP is now 98 ng/L.

This patient illustrates the possibility to diagnose AL amyloidosis at a very early asymptomatic stage and to fully exploit the therapeutic armamentarium and restore organ function. About 3% to 5% of patients with a known precursor PC disorder will progress to AL amyloidosis. However, even in referral centers, it takes a median time of almost 1 year to reach the diagnosis, based on signs or symptoms, in patients with preexisting PC dyscrasias.³² When symptoms manifest, even a minor delay translates into significant shortening of survival.³³ For this reason, we advocated the use of sensitive markers of amyloid organ involvement during the follow-up of individuals with MGUS to facilitate a presymptomatic diagnosis.³⁴ In this patient, an increase of NT-proBNP raised suspicion of amyloid cardiac involvement. NT-proBNP is a very sensitive marker, but it is not specific (being elevated in other conditions), and heart involvement can be confirmed by echocardiography or magnetic resonance. Our diagnostic approach is reported in Figure 1. Once diagnosis is established, it is necessary to assess the risk to carefully balance expected benefit and possible treatment toxicity. Cardiac involvement is the main determinant of frailty and survival. Current validated staging systems are reported in Figure 2.^31,35-39 The patient was cardiac stage II and considering the young age and perfect fit, he was eligible for high-dose melphalan and ASCT. Eligibility criteria vary for centers (Figure 3); however, only ∼20% of newly diagnosed patients are eligible for this intensive treatment. ASCT is safe, with <3% treatment-related mortality in referral centers, and highly effective, providing ≥VGPR in approximately 70% of patients and a median survival > 15 years in patients achieving CR.^40,41 Induction therapy with bortezomib-based regimens should be considered in all transplant-eligible patients.^42,43 Furthermore, the recently concluded Andromeda trial, reported an unprecedented high rate of deep hematologic responses (≥VGPR in 78.5%) for daratumumab plus CyBorD, which was recently approved by the US Food and Drug Administration and European Medicines Agency for newly diagnosed AL amyloidosis.⁴⁴ Daratumumab-CyBorD is a new standard of care and will become also the preferred induction therapy before ASCT. Importantly, daratumumab-CyBorD is also effective in patients whose amyloid clones harbor the t(11;14) who have poorer outcomes with CyBorD alone.⁴⁴ Because daratumumab was not available in Italy at the time this patient was seen, he received CyBorD achieving VGPR after 4 courses. Based on data suggesting that ASCT can be deferred to relapse after successful induction, without detrimental effects on overall survival, we elected to postpone ASCT.^45-47 This practice is common in Europe. Actually, in a survey including 3064 patients treated between 2011 and 2018 in 10 European referral centers, ASCT was used as frontline therapy in 6% of patients and as second-line therapy in 10%.⁴⁸ Although some centers prefer transplant for all eligible patients upfront, we generally prefer to defer ASCT in subjects who attain CR and/or organ response after induction. However, at our center, we proceed to ASCT in patients with MM. In this patient, the appearance of mild symptoms and increase in NT-proBNP suggested that attaining VGPR was not sufficient to halt disease progression, and the patient proceeded to ASCT. Approximately 30% of patients do not respond to first-line treatment, and they should promptly be switched to another class of drugs. Quality of organ response is strictly dependent on the depth of hematology response, and although in some patients VGPR or even PR is enough to reach it, in patients with highly toxic LCs, persistence of even MRD may be sufficient to hamper amelioration of organ function.⁴⁹ In patients who achieve CR without organ response, MRD evaluation may help decide regarding further therapy. Proposed criteria for grading organ response are now being validated.⁵⁰ The criteria for hematologic and organ response are reported in Table 1.^39,51-54 In this fit patient, stem cell collection was uneventful, although the incidence of major complications (hypotension, hypoxia, cardiac arrhythmia, and fluid retention) during stem cell mobilization and collection is approximately 15%. The patient was eligible for full-dose melphalan (200 mg/m²), which is important to optimize outcomes, and reached CR and cardiac response.

A 58-year-old woman began to experience exertional dyspnea. Three months later, she was hospitalized because worsening dyspnea and pleural effusion. Echocardiogram reported left ventricular hypertrophy with preserved ejection fraction (55%). Diuresis resolved pleural effusion, and she was discharged. Over the next 6 months, symptoms worsened, and cardiac catheterization showed normal coronaries. Two months later, the patient was seen in another cardiology center because of recurrent cardiac syncope and nonsustained ventricular tachycardia. An implantable cardioverter defibrillator (ICD) was implanted, and oral amiodarone was started. Echocardiography showed diastolic dysfunction and increased wall thickness (posterior wall, 16 mm; interventricular septum, 15 mm) with reduced ejection fraction (43%). The suspicion of cardiac amyloidosis was raised, and a scintigraphy with ^99mTc 3,3-diphosphono1,2-propanodicarboxylic acid and an endomyocardial biopsy were performed. The scintigraphy showed no cardiac uptake, and the biopsy showed amyloid deposits that were not further characterized. NT-proBNP was 10 740 ng/L, and troponin I was 0.44 ng/mL. The patient was referred to our center 2 months later, more than 1 year after the onset of symptoms. She reported fatigue, dyspnea for minor exertion, petechiae, and ecchymoses in the upper part of the body. Physical examination revealed mild macroglossia, jugular vein distention, pleural effusion, hepatomegaly, distal edema, and hypotension (86/53 mm Hg). Serum and urine electrophoresis and immunofixation revealed an IgG κ spike (28 g/L) with free κ LCs and Bence Jones proteinuria. Serum κ FLCs were 303 mg/L (ratio, 25.2; dFLC, 291 mg/L). A bone marrow biopsy speciman showed 31% monotypic κ PC with gain1q21 and amyloid deposits. No other myeloma-defining events were present. Abdominal fat aspirate was positive for amyloid that was typed as AL κ by immuno-electron microscopy. NT-proBNP was 15 585 ng/L, troponin I was 0.53 ng/mL, and echocardiography confirmed advanced amyloid involvement. The patient was stage IIIb, and because there was no sign of other organ damage, we started evaluation for cardiac transplantation. In the meantime, she started treatment with CyBorD with reduced bortezomib (1 mg/m²) and dexamethasone (20 mg) dosage. After the second cycle, she experienced worsening of heart failure that required hospitalization. Investigations revealed reduction of the serum monoclonal component to 19 g/L and of dFLC to 35 mg/L, qualifying for VGPR, whereas NT-proBNP increased to 16 834 ng/L and troponin I to 0.63 ng/L. Three days after admission, the patient collapsed and died with pulseless electrical activity.

This case illustrates the devastating effects of delayed diagnosis: in this rapidly progressing disease, “time is life,” and even few months can irreversibly compromise any chance to benefit from effective therapies. Symptoms that should trigger an appropriate workup for amyloidosis are listed in Table 2. The diagnosis of cardiac amyloidosis can be augmented by machine learning applied to electrocardiographic and echocardiographic data.⁵⁵ In the European survey, the percentage of patients presenting with very advanced cardiac damage (stage IIIb) remained the same (15%-16%), and median survival remained disappointingly low (4.5 months) in the periods 2004 to 2010 and 2011 to 2018.⁵⁶ A recent survey among US community and academic hematologists showed that low disease awareness, inadequate referral practices, and inadequate screening/testing procedures account for a 1- to 1.5-year diagnostic delay.⁵⁷ Scintigraphy with bone tracers (PYP or DPD) is used to search for cardiac transthyretin amyloidosis that usually shows strong uptake.⁵⁸ However, the presence of the monoclonal protein should be ascertained first, using appropriate methods (Figure 1), because its presence makes cardiac biopsy mandatory and scintigraphy unnecessary. In this case, the cardiologists elected to perform scintigraphy and cardiac biopsy concomitantly. Because of recurrent syncope and registered ventricular arrhythmias, an ICD was implanted. Probably, patients with very advanced cardiac amyloidosis (New York Heart Association class 4, NT-proBNP >8500 ng/L, and systolic blood pressure <90 mm Hg) are not appropriate candidates for ICD implantation.⁵⁹ The patient reported recurrent purpura, a common finding particularly in advanced stages of the disease, that is attributed to amyloid infiltration of blood vessels. Coagulation abnormalities are multifactorial, including factor x and other factor deficiencies, and occur in a significant proportion of patients.⁶⁰ These extremely fragile patients are very sensitive to treatment toxicity, and we attenuated treatment dosages. Although early and deep response is associated with improved survival in patients with stage IIIb,⁶¹ in this subject, cardiac involvement was likely already too advanced. At present, there is no treatment capable of improving the outcome of most patients with stage IIIb. However, subcutaneous daratumumab can induce rapid and deep responses, which makes it a valuable option in these patients. A multicenter phase 2 trial of single-agent daratumumab in subjects with stage IIIb is underway (NCT04131309). Moreover, removing amyloid deposits with antibodies is expected to improve outcomes and is currently being tested in clinical trials (Table 3). Young patients with stage IIIb with isolated heart involvement should be considered for cardiac transplantation that is now associated with outcomes comparable to nonamyloid cardiomyopathy.⁶² However, scarcity of donors frequently renders time on waiting list incompatible with patient’s needs.

A 70-year-old man with a 1-year history of rheumatoid arthritis treated with hydroxychloroquine and prednisone (2.5 mg/d) and of smoldering MM (IgG λ; FLC ratio, 0.08; dFLC, 141 mg/L; bone marrow PC infiltrate, 25%; t(11;14) in 25% of interphases) was referred to our center because of bilateral carpal tunnel syndrome and proteinuria of 1.2 g/d (creatinine, 0.71 mg/dL). There was no sign of amyloid cardiac involvement (normal echocardiography; NT-proBNP, 42 ng/L; troponin I, 0.03 ng/mL). Serum amyloid A apolipoprotein (SAA) level was normal. Abdominal fat aspirate was positive, and the amyloid deposits reacted with anti-λ (and not with anti-SAA and anti-transthyretin) antibodies at immuno-electron microscopy. A diagnosis of AL amyloidosis with renal and soft tissue involvement was made, and the patient was enrolled in the ANDROMEDA trial and randomized to the daratumumab-CyBorD arm. After 2 infusions, treatment was interrupted because of a respiratory infection. However, by then, a VGPR had been reached (dFLC, 13 mg/L; free light chain ratio [FLCR], 0.61 ; positive serum immunofixation) with renal response (proteinuria, 0.5 g/d). Cycle 2 was also interrupted after the second infusion because of a respiratory infection. The following 4 cycles were completed without complications, but we elected to withhold daratumumab maintenance. The patient is currently asymptomatic 2 years after treatment discontinuation, and VGPR is maintained (dFLC, 14 mg/L; FLCR, 0.81; positive serum immunofixation).

In this patient, symptoms were recognized early, and the diagnosis was made when he was still renal stage I. However, this subject had a chronic inflammation (rheumatoid arthritis) that can give rise to AA (reactive) amyloidosis. Amyloidosis reactive to chronic phlogosis is caused by deposition of SAA-derived amyloid fibrils and presents with renal damage.⁶³ Treatment is aimed at controlling the underlying inflammation. Moreover, carpal tunnel syndrome can be found both in AL and in transthyretin amyloidosis. Thus, accurate tissue typing was mandatory.

In this patient, bone marrow PC infiltrate was higher than 20%. This measure of clonal bulk is an additional negative prognostic factor independent of cardiac involvement,⁶⁴ as well as dFLC level.^37,53,54 The patient could be treated with daratumumab-CyBorD in the ANDROMEDA trial. This regimen is well tolerated, but 1 of the most common adverse events with daratumumab in AL amyloidosis is respiratory infection,⁶⁵ as occurred in this subject who was also immunocompromised by steroid treatment of rheumatoid arthritis. Nevertheless, the infection was manageable, and VGPR was reached after only 2 infusions. Prophylaxis of infections may be useful in patients with AL amyloidosis treated with daratumumab as was shown in MM.⁶⁶ Besides its high efficacy, the ability of daratumumab to induce deep responses after the first infusion^67,68 is particularly relevant in patients with AL amyloidosis who need to reach a deep reduction of the amyloid LCs as soon as possible.

The role of maintenance has not been adequately explored in patients with AL amyloidosis. In the ANDROMEDA study, treatment with daratumumab could be continued for up to 2 years if well tolerated, but there was no control arm for maintenance. In our patient, who was prone to infections, we elected to withhold maintenance. Controlled studies are warranted to ascertain whether maintenance has a role in AL amyloidosis. At our center and at the Mayo Clinic,⁶⁹ maintenance is considered in patients with large clones (eg, fulfilling the SLiM CRAB criteria) and/or high-risk cytogenetics.

A 72-year-old man developed worsening speech difficulty caused by slowly progressing macroglossia. After 18 months, foamy urine, peripheral edema, impotence, and symptomatic postural hypotension with occasional syncope ensued, and the patient sought medical advice. Proteinuria was present and the general practitioner suspected amyloidosis and referred the patient to our center. He presented with typical macroglossia and submandibular gland swelling, nephrotic syndrome (proteinuria, 5.9 g/d; normal creatinine), symptomatic orthostatic hypotension, and hepatomegaly (8 cm below costal margin). Echocardiography revealed amyloid heart involvement, NT-proBNP was 5500 ng/L, and cardiac troponin I was 0.05 ng/L. Alkaline phosphatase was elevated (480 U/L; upper reference limit, 150 U/L) with normal bilirubin and aminotransferases. Immunofixation electrophoresis showed monoclonal λ FLC in serum and urine, and circulating λ FLC concentration was 849 mg/L (ratio, 85; dFLC, 840 mg/L). Bone marrow plasma cell infiltrate was 16%, with no chromosomal abnormalities by interphase fluorescence in situ hybridization (iFISH). Abdominal fat aspirate showed amyloid deposits characterized as AL-λ type by immuno-electron microscopy. A diagnosis of AL amyloidosis with cardiac, renal, liver, soft tissue, and autonomic nervous system involvement was made, and the patient was enrolled in a clinical trial comparing melphalan and dexamethasone (MDex) with MDex plus bortezomib (BMDex). The patient was randomized to the BMDex arm. The use of fitted elastic leotards was recommended. Diuretic therapy was initiated before chemotherapy with furosemide 25 mg/d, and the patient lost 2 kg in 1 week. Higher doses of furosemide were not tolerated because of worsening hypotension. During the first BMDex course, fluid retention worsened again. By the end of cycle 1, fluid retention and hypotension resolved and did not recur during subsequent cycles. After 3 cycles, a partial response was reached, and VGPR was attained after 6 cycles and maintained at the completion of cycle 8 (positive urine immunofixation; dFLC, 33 mg/L; ratio, 0.30), with cardiac (NT-proBNP 686 ng/L), renal (proteinuria 1.4 g/d), and liver (alkaline phosphatase 150 U/L) response, and treatment was discontinued. The patient was followed every 6 months for 5 years, when an increase in dFLC was observed (141 mg/L; ratio, 0.06) with stable NT-proBNP (767 ng/L) and proteinuria (1.6 g/d). The only symptoms were mild fatigue and persistent involvement of soft tissues. Rescue treatment with ixazomib, lenalidomide, and dexamethasone was started. After 2 cycles, partial hematologic response was achieved (dFLC, 61 mg/L; ratio, 0.16) with stable symptoms, increased NT-proBNP (3835 ng/L), stable proteinuria (1.4 g/d), and an increase in serum creatinine (from 0.77 to 1.4 mg/dL). Treatment was continued for 4 more cycles with stable hematologic and organ parameters. After lenalidomide discontinuation, NT-proBNP level decreased to 480 ng/L. Eighteen months later, a new rise in dFLC was documented (169 mg/L; ratio, 0.09) that was associated with stable NT-proBNP (576 ng/L), proteinuria (1.5 g/d), and creatinine (1.5 mg/dL). Rescue treatment with pomalidomide and dexamethasone was started. A VGPR was re-established after 3 cycles (dFLC, 21 mg/L; ratio, 0.48; persistent positive urine immunofixation). A suspected pomalidomide-related increase in NT-proBNP was observed during therapy (up to 3057 ng/L) while renal involvement remained stable (proteinuria, 1.3 g/d; creatinine, 1.3 mg/dL). The patient is now 81 years old, and he is enjoying an active life despite persistent macroglossia, with resolution of symptoms of heart and renal involvement and of hypotension.

Symptoms of AL amyloidosis depend on organ involvement and can be often mistaken for manifestations of more common diseases. However, typical soft tissue involvement and combination of signs of simultaneous involvement of different organs can be a clue to diagnosis as in this patient. Yet, multiorgan involvement makes these patients frail and difficult to treat intensively. Supportive measures to manage fluid retention, hypotension, and malnutrition should be started early, possibly before initiation of chemotherapy (Table 4).⁷⁰ Dexamethasone can exacerbate fluid retention and arrhythmias and should be used at lower dosage (≤20 mg) in patients with advanced heart involvement (New York Heart Association class III or IV, cardiac stage IIIb, repetitive ventricular arrhythmias).⁷¹ At diagnosis, this patient was cardiac stage II and renal stage II, and he was not a candidate for stem cell transplantation because of age, heart involvement, and hypotension, qualifying as intermediate risk. Bortezomib is the most used agent in upfront therapy, based on retrospective uncontrolled studies.^35,72 Running controlled clinical trials has been difficult in AL amyloidosis because of the relative rarity of this condition and the easy access to drugs licensed for MM. This patient was enrolled in the AC-004-EU/EMN-03 trial comparing MDex and BMDex (NCT01277016). This academic, investigator-initiated study showed that BMDex grants higher hematologic response rates than MDex (81% vs 57%; VGPR/CR, 64% vs 39%) and was the only study to demonstrate an overall survival advantage for the experimental arm in AL amyloidosis.⁷³ BMDex is well tolerated and relatively inexpensive. Moreover, it can potentially overcome the effect of the 2 most common chromosomal abnormalities found in this disease: t(11;14) and gain 1q(21).^26,28 In this patient, BMDex induced VGPR with cardiac, renal, and liver response that lasted 5 years. It has recently been shown that restaging discriminates survival even after upfront therapy.⁷⁴ The patient had asymptomatic hematologic relapse and was rescued when he still was cardiac stage II. There are no validated criteria for hematologic progression in AL amyloidosis, and there is disagreement on when treatment should be started at relapse.^75-77 This is relevant not only for individual patient management but also for the lack of validated definition of progression-free survival in clinical trials. However, it is generally agreed that organ progression should not be awaited, because this is associated with shorter survival and time to dialysis.^39,51 Indeed, it is common practice at different referral centers to start treatment when early, relatively small increases in FLC (approximately 40% of baseline value) are detected.^78,79 Small increases in FLC should be regarded cautiously particularly in patients who presented with advanced cardiac involvement. Rescue therapy is discussed in Figure 4. This patient was rescued with a combination of ixazomib and lenalidomide,⁸⁰ and pomalidomide at first and second relapse, before progression became symptomatic. Although he never reached CR, his disease was controlled for 10 years, with a good quality of life.

The optimal goal of therapy is yet to be established and possibly varies during follow-up. In patients with this rapidly progressing disease, it is necessary to closely monitor (every 2 cycles) the outcome of therapy to promptly switch regimen and avoid further organ deterioration.^61,81 A deep reduction of FLCs should be achieved very early,⁸¹ but ideal timing and depth vary based on disease severity and treatment regimen. With daratumumab, FLCs usually drop after the first administration.⁶⁷ In patients with advanced cardiac involvement, an earlier (after 1 cycle) assessment of response is advisable to allow a rapid shift in nonresponders. This can be particularly relevant in subjects receiving daratumumab. At the end of frontline therapy, either CR or organ response should be reached, and the recently validated composite hematologic and organ response criteria can offer guidance in this assessment.⁸² However, in patients with an aggressive underlying clone (eg, with SLiM CRAB or high-risk myeloma cytogenetics) CR might be more obstinately pursued, as recently suggested by the Mayo Clinic group.⁶⁹ In the longer term, biomarkers of organ involvement should progressively decrease and normalize, indicating amyloid FLC concentration is kept at a level that does not prevent recovery of organ dysfunction.⁵⁰ In fit patients who fail to attain organ response despite CR, MRD assessment can be useful, and further therapy can be considered if MRD is detected.⁴⁹ The role of MRD as an end point in AL amyloidosis should be explored in future studies.

A 77-year-old woman was referred to a hematologist because of bilateral enlarged cervical lymph nodes. Subsequent investigations showed mediastinal and retroperitoneal lymphadenopathy (<2 cm). An IgMλ monoclonal component was detected. Blood tests were all normal. A lymph node biopsy showed amyloid deposits. No clonal lymphocytes or plasma cells were detected in her bone marrow. The patient was followed closely without therapy. After 2 years, bilateral, symmetrical, ascending, sensitive, peripheral neuropathy appeared, and the patient was referred to our center. The concentration of the monoclonal protein was 25 g/L, λ FLC level was 40 mg/L (dFLC, 24 mg/L; ratio, 0.40), hemoglobin level was 11.5 g/dL, and alkaline phosphatase level was 360 U/L (upper reference limit, 150 U/L) with normal bilirubin and aminotransferases. Testing for anti-myelin–associated glycoprotein and anti-ganglioside antibodies was negative. Nerve conduction studies showed a purely axonal pattern. There were no signs of amyloid cardiac or renal involvement. We repeated a bone marrow biopsy that showed a 7% clonal, CD20-positive lymphoplasmacytic infiltrate. The clone harbored both the MYD88^L265P and the CXCR4 mutations. The patient was treated with bendamustine, rituximab, and dexamethasone, which was well tolerated, and after 4 cycles, a low dFLC response was reached (dFLC, 4 mg/L; ratio, 0.63; monoclonal component, 10 g/L). Two more cycles were performed without improvement of hematologic response while alkaline phosphatase normalized. The patient is hematologically and clinically stable 1 year after treatment discontinuation.

IgM-related AL amyloidosis is a distinct clinical entity that accounts for 6% to 10% of all AL amyloidosis cases.^83,84 Localized amyloid deposits are more common, as well as lymph nodes, peripheral nervous system, and lung involvement, and amyloid FLC levels tend to be lower. The level of cardiac biomarkers is prognostic, but liver involvement and peripheral neuropathy also predict a poor outcome and were incorporated in a staging system specific for IgM-AL amyloidosis.⁸⁴ Peripheral neuropathy is frequently symptomatic, but it is not detected by nerve conduction studies. Having both peripheral nervous system and liver involvement, this patient classified stage III despite normal cardiac biomarkers.

In most patients with IgM-AL amyloidosis, the clone can be detected in the bone marrow at diagnosis. The underlying clone can be a pure PC (in ∼25% of cases) or a lymphoplasmacytic neoplasm (in ∼65% of cases) as in the present case, and this is relevant for the choice of treatment.⁸⁵ Lymphoplasmacytic clones harbor the MYD88 (95%-97%), CXCR4 (30%-40%), ARID1A (17%), and CD79B (8%-15%) mutations. Patients with IgM-AL amyloidosis caused by pure plasma cell clones should be treated like patients with non–IgM-AL amyloidosis, whereas patients with lymphoplasmacytic clones are generally treated with rituximab-containing regimens, based on treatments developed in Waldenström macroglobulinemia. Rituximab can be combined with dexamethasone and cyclophosphamide, bortezomib, or bendamustine.^86-88 Patients with IgM-AL amyloidosis can also successfully undergo autologous stem cell transplant.⁸⁹ Ibrutinib can be an option in patients with favorable mutational pattern, but a first report highlighted a rather poor tolerability and scarce efficacy of this drug.⁹⁰ Patients with IgM-AL amyloidosis have lower response rates and shorter survival than other patients with AL amyloidosis when adjusted to stage.⁸⁵ 

Neuropathy was the main reason to start treatment in this subject. The differential diagnosis of monoclonal gammopathies of neurologic significance is challenging.⁹¹ In this patient, age was considered a contraindication to ASCT, and we preferred a bortezomib-free combination because of neuropathy. The validated response criteria of the International Society of Amyloidosis consistently predict survival and can be used in IgM-AL amyloidosis.⁸⁴ 

After many years of empirical treatment, the therapy of patients with AL amyloidosis has now entered the realm of evidence-based medicine, with 3 international clinical trials completed in the last 2 years.^44,73,92 We should now further exploit these achievements with new controlled trials based on validated end points. However, our evidence-based knowledge only covers a small part of the management of patients with AL amyloidosis. Daratumumab-bortezomib combinations can be used as frontline therapy in almost all patients, and the key eligibility criteria to safely proceed to ASCT are now shared by referral centers. However, important questions, such as best treatment at relapse, role of maintenance, and the most cost-effective approach in patients with specific chromosomal abnormalities still need to be answered in controlled studies. For these reasons, patients should still be referred to specialized centers and treated within clinical trials whenever possible. With a growing therapeutic armamentarium, survival has more than tripled; however, no progress was made in patients with stage IIIb. Amyloid-directed therapy is an appealing opportunity to accelerate recovery of organ function, particularly in these patients. New therapies are emerging, and more than 20 interventional trials are now recruiting. Immunotherapies showing promise in MM will hopefully be offered to patients with AL amyloidosis. The anti-CD38 antibody isatuximab showed encouraging results in a recent phase 2 clinical trial in relapsed/refractory patients (NCT03499808).⁹³ Chimeric antigen receptor T cells, and bispecific T-cell engager antibodies and antibody–drug conjugates, such as belantamab mafodotin that is being evaluated in a European trial (NCT04617925), will allow targeting additional PC antigens. Finally, the continuous, stimulating progress in basic and translational research are offering new insights in the mechanisms of disease, and there is hope that new treatment targets will eventually be made available.

The authors received research support by a grant from CARIPLO “Harnessing the plasma cell secretory capacity against systemic light chain amyloidosis” (grant 2018-0257), a grant from the Italian Ministry of Health “Towards effective, patient-tailored anti-plasma cell therapies in AL amyloidosis: predicting drug response and overcoming drug resistance” (grant GR-2018-12368387), and by a grant from Cancer Research United Kingdom, Fundación Científica Asociación Española Contra el Cáncer, and Associazione Italiana per la Ricerca sul Cancro under the Accelerator Award 2017 Program “Early detection and intervention: understanding the mechanisms of transformation and hidden resistance of incurable haematological malignancies.”

Contribution: G.P. and G.M. wrote the manuscript.

Conflict-of-interest disclosure: G.P. reports honoraria for lectures from and membership on advisory boards with Janssen and honoraria for lectures from Pfizer and Siemens. G.M. declares no competing financial interests.

Correspondence: Giampaolo Merlini, Amyloidosis Research and Treatment Center, Foundation IRCCS Policlinico San Matteo, Viale Golgi, 19, 27100 Pavia, Italy; e-mail: gmerlini@unipv.it.