Stem-cell Mobilization and Autologous Transplant for Immunoglobulin Light Chain Amyloidosis

Background

Stem cell transplantation has been an established modality for the treatment of multiple myeloma for over 30 years. This technique is applied globally and is considered a standard of care. Patients who achieve a complete response following stem cell transplantation with amyloidosis have a longer progression-free survival than patients who achieve a complete response following autologous stem cell transplant for multiple myeloma. This is understandable given the fact that patients with amyloidosis have a low tumor mass at diagnosis (average 10% plasma cells), have far less frequent unfavorable genetics (5% versus 25%), and have a nonproliferative clone of plasma cells unlike multiple myeloma. Conceptually, transplanting amyloidosis is like transplanting MGUS, and one would expect more durable responses and longer times to next treatment.

The differences between transplanting multiple myeloma and transplanting amyloidosis are significant, however. In multiple myeloma, the bone marrow is compromised, but organ function, particularly the heart, the liver, and the kidney, are functioning well. The converse is the case in light chain amyloidosis. The bone marrow is minimally involved, there is a very low tumor mass, but these patients typically have advanced cardiac or renal or hepatic dysfunction. Organ compromise leads to significant complications with the application of myeloablative chemotherapy. These issues of organ dysfunction and late diagnosis of amyloidosis translate into a proportion of patients with light chain amyloid eligible for stem cell transplant at no more than 20 to 25%.

Treatment related mortality and eligibility criteria

The primary consideration for proceeding with stem cell transplantation for amyloidosis is safety. Currently, treatment-related mortality from all causes by day 100 in myeloma is <1%. Given the availability of many new drugs for amyloidosis, mortality rates in excess of 3% should be considered unacceptable. In the early days of transplantation, there were few alternatives for therapy and included melphalan-dexamethasone and vincristine-doxorubicin-dexamethasone. With those regimens, deep responses were uncommon, and patients often had little alternative but to undergo stem cell transplantation. With the availability of new agents, including proteasome inhibitors, immunomodulatory drugs (IMIDs) and monoclonal antibodies, subjecting a patient to a stem cell transplant if the expected morbidity and mortality is high is considered unacceptable.(1) However, with careful patient selection, autologous stem cell transplantation is an effective therapy for light chain amyloidosis. In publications over the last five years where patient selection has been refined, complete hematological and organ responses can be seen in 50 to 60% of patients.(2) Patients that had brain natriuretic peptide levels <300 pg/mL and troponin I levels ≤0.07 ng/mL had better outcomes.(3) Due to refinements in patient selection and the availability of alternative, highly-effective chemotherapies for patients with amyloidosis, the early mortality over time has fallen substantially.(4) When the Mayo program began, therapy-related mortality was as high as 14.5%. It is currently 2.4%. We believe that safety of transplantation is the paramount consideration for eligibility.(4)

Stem cell transplant in patients with renal failure

Although refined patient selection is required for good outcomes, it is not to say that excellent outcomes cannot be achieved in subsets of patients historically deemed unsuitable for stem cell transplantation. Accounts of patients on steady-state dialysis with successful stem cell transplantation for light chain amyloidosis have been reported,(5) with a complete hematologic response in 70% of evaluable patients one year after stem cell transplant and a median overall survival of 5.8 years. This has a positive impact on the survival of patients with AL amyloidosis. In one trial, significant differences in overall survival were seen, depending on when acute renal failure occurs. Patients transplanted on steady-state dialysis can do remarkably well. Patients who develop acute renal insufficiency and require dialysis within 30 days of autologous stem cell transplant have unacceptably high rates of treatment-related mortality. If it is expected that the patient will sustain acute renal insufficiency, either associated with very low serum albumin levels or markedly elevated urinary protein levels, the patient should not be considered for transplantation.(6) The Mayo group has transplanted a cohort of patients with an EGFR of <45 mL/minute. Dialysis was required in 6.7% within 100 days of autologous stem cell transplant, and dialysis was required in 22% in patients with renal stage 3 disease. In this cohort, 100-day mortality was 14%. Impaired renal function does predict for a higher rate of hospitalization, progression to dialysis, and early mortality.(7) A powerful predictor of acute renal injury during stem cell transplant for renal amyloidosis is severe hypoalbuminemia, serum albumin <2 g/dL. Evidence of acute renal failure requiring hemodialysis was 25% when the albumin was this low. Extreme caution should be used in conditioning patients for transplant who have a serum albumin <2 g/dL, and melphalan dose modification should be strongly considered.(8) With careful patient selection, the overall response rate was 75%, the complete response rate was 25%, and the progression-free survival was 40 months.

Stem cell transplant in elderly patients

In patients above the age of 70 the treatment-related mortality was 3%, and transplant can be performed safely. We have transplanted patients with three or more organs involved. We have found this can be done safely as long as the severity of cardiac involvement is mild to moderate. An NT-proBNP level >2000 was a powerful predictor of overall survival and was far more important than the number of organs involved. Number of organs should not be considered a contraindication to safe transplantation if cardiac function is preserved.(9)

Patients who have lung, liver, gastrointestinal, neurologic, and soft tissue involvement have also been safely transplanted.(10) In one trial, 46 patients received induction therapy before transplant. The day-100 treatment-related mortality was 3.8%. Median progression-free and overall survivals were 36 and 73 months, respectively. Stem cell transplantation in selected AL patients with gastrointestinal, hepatic, pulmonary, and soft tissue involvement can be considered safe for transplantation. Mayo has also transplanted patients with factor × deficiency and a bleeding diathesis. Post-stem cell transplant factor × levels increased in all patients evaluated, but the degree of improvement in factor × levels were correlated with an improvement in markers of renal involvement by amyloid. Patients with factor × deficiency with appropriate support can be safely transplanted.(11) However, splenic rupture has been reported in a patient with factor × deficiency undergoing autologous stem cell transplant, so caution is required.(12)

As in all patients receiving myeloablative chemotherapy, bacterial infections are common. They have been reported in 24% of patients. And when they occur, treatment-related mortality was 10%. Infections are gram-positive in 51%, anaerobic in 16%, gram-negative in 13%, and fungal in 9%. Serum creatinine >2 mg/dL was a risk factor for infection.(13) Among 377 AL patients acute kidney injury during leukocyte engraftment was very common and attributed to the engraftment syndrome. Patients who have a serum creatinine rise at the time of leukocyte engraftment should be considered to have engraftment syndrome and be considered for corticosteroid therapy.(14)

Induction therapy

It is controversial whether patients with light chain amyloidosis and a small plasma cell burden require induction chemotherapy prior to autologous stem cell transplantation. The HOVON reported extended follow-up with vincristine-doxorubicin-dexamethasone induction followed by high-dose melphalan in a phase 2 trial. Four-year overall survival rate was 62% and, after transplantation, was 78%. Median survival of all patients was 96 months. The authors concluded that vincristine-doxorubicin-dexamethasone should not be applied as induction therapy for amyloidosis but did recommend a two-step approach consisting of induction therapy followed by high-dose melphalan.(15) The most common induction regimen is bortezomib-containing. Induction therapy with bortezomib and dexamethasone was reported in 56 patients. The complete remission rate in the bortezomib high-dose melphalan arm was 68%. In this randomized study, those patients who did not have induction therapy but transplant only had a complete response rate of only 36% (p=0.03); moreover, a survival advantage was reported for those transplanted patients who received bortezomib induction, suggesting that bortezomib-dexamethasone induction with transplant was superior to transplant alone.(16) In a non-randomized trial where half of the patients received bortezomib pre-transplant and the other half went directly to transplant, the overall response rate in the bortezomib-treated group was 92% versus 69% in the group that received no treatment. Time to best response was also shorter in the bortezomib-treated group (3 versus 14 months).(17) A nonrandomized trial of novel-agent induction was reported, consisting of a proteasome inhibitor in 13, an IMID in 16, a proteasome inhibitor and cyclophosphamide in 3, and an IMID and a proteasome inhibitor in 3. A progression-free survival advantage at 3 years in patients receiving novel-agent induction was 79% and was 53% for those who went directly to transplant without induction. This suggests that a novel-agent induction regimen prior to stem cell transplant could improve outcomes.(18) The Mayo group reported on 415 AL patients that had induction prior to stem cell transplant. Hematologic complete and very good partial response rates were significantly higher in those patients who had <10% plasma cells in the bone marrow. Multivariate analysis risk factors for inferior survival included failure to give induction therapy, advanced amyloid stage, and bone marrow >10% plasma cells. Induction therapy pretransplant may improve outcomes among AL patients due to a rapid reduction of toxic light chains.(19) At MD Anderson, IMID or proteasome-inhibitor-based induction was associated with a longer overall survival compared with patients who receive no induction before autologous stem cell transplant. They reported a two-year overall survival of 87% with induction therapy compared to 73% without induction therapy. Response depth was also greater in the induction therapy cohort.(20) In the long-term follow-up from Boston University, the incorporation of bortezomib into induction yielded durable hematologic responses with corresponding organ responses and prolonged survival. The overall response rate was 100%, and the median overall and progression-free survival had not been reached with a median follow-up of 77 months.(21)

In the HOVON 104 trial, bortezomib-based induction followed by stem cell transplant was performed on 50 patients. Treatment-related mortality was 0%. Hematologic responses at six months after transplant were 86%, with 46% complete and 26% very good partial response, confirming the high efficacy of bortezomib as an induction regimen for patients with AL amyloidosis.

The preponderance of evidence suggests that induction therapy with bortezomib-based regimens should be strongly considered in all patients with transplant-eligible light chain amyloidosis.

Mobilization

Stem cell mobilization (SCM) can be performed with different strategies that combine chemo-mobilization and plerixafor administration. Main studies reporting results on SCM in AL amyloidosis are reported in the Table 1. Stem cell mobilization is associated with unusual morbidity in AL amyloidosis compared to multiple myeloma, even when performed without chemotherapy. In a large monocentric series, 11% of patients SCM with G-CSF did not proceed to stem cell transplantation due to complications and 4% died during SCM. Hypoxia and hypotension are rare life-threatening event during SCM with G-CSF and may be the result of a capillary leak syndrome triggered by the growth factor. Capillary leak syndrome may also be responsible of worsening of fluid retention. This is particularly relevant in patients with cardiac and renal AL amyloidosiswho present with heart failure and nephrotic syndrome. In another study, 10% had worsening of fluid retention during G-CSF SCM that resulted in a delay in stem cell transplantation. In chemo-mobilization, G-CSF dosage is half than in SCM with G-CSF alone. The Heidelberg group reported a series of patients in which SCM was performed with G-CSF alone or with chemotherapy that was cyclophosphamide, doxorubicin and dexamethasone (CAD) in most cases. CAD mobilization was highly effective, with a median of 8 × 10⁶ CD34+/Kg body weight (range, 0 – 46) collected cells with a median of one leukapheresis procedure. Main adverse events during CAD mobilization were cardiovascular events, gastrointestinal toxicity and infections. When compared with G-CSF alone SCM, no differences in rates of grade III or IV adverse events were seen (20% vs. 10% and 4% vs. 0%). In order to improve chemo-mobilization tolerability, a SCM with only cyclophosphamide and dexamethasone was proposed. In poor mobilizers, as in patients previously exposed to melphalan, the addition of plerixafor is effective to reach the requisite number of CD34+ cells. This drug has been proven effective and safe in several series of patients with AL amyloidosis. Particularly, a study comparing SCM with G-CSF alone and G-CSF and plerixafor, showed that adding plerixafor resulted in higher numbers of collected CD34+ cells with no obvious differences in toxicity. Mobilization strategy, either with or without chemotherapy, should be aimed to obtain the requested CD34+ cell collection in the shortest time, in order to proceed rapidly to stem cell transplant. Mobilization related toxicity may also result in a delay in high dose treatment and should be carefully considered. Recently, a study reported that low serum albumin and high NT-proBNP are associated with higher peri-mobilization morbidity/mortality. Thus, patients’ selection remains of utmost importance. Finally, SCM should provide an adequate number of CD34+ cells to have a long-term recovery of bone marrow hemopoietic function. The Heidelberg group showed that infusion of > 6.5 × 10⁶ CD34+ cells/Kg resulted in shorter duration of leukocyte reconstitution and higher platelet count at 12 months form high dose treatment. Reconstitution of good hemopoietic function after stem cell transplant might result in a better hematologic tolerability of a second line treatment and be associated with a reduction of secondary hematologic neoplasia.

Table 1.

Studies on PBSC mobilization in patients with AL amyloidosis

Study	N° of patients	Mobilization approach	Observed AEs	Median collected CD34+ cells	Main remarks
Skinner, et al.*	312	G-CSF	4% mortality	>2.0 × 106/Kg	11 % of patients did not proceed to HSCT due perimobilization morbidity/mortality
Perotti, et al.	42	G-CSF alone G-CSF and CTX	4.7% grade ≥2 AEs	8.2 × 106/Kg 8.9 × 106/Kg	Mobilization effective in 95% of cases, with a median number of collections was 1.8
Gertz, et al.*	434	G-CSF	10% marked fluid retention	7.16 × 106/Kg	Patients that had perimobilization fluid retention required 1 month of recovery before HSCT
Lee, et al.	5	Plerixafor and G-CSF	No mobilization related AEs	5.9 × 106/Kg	Plerixafor is effective and well tolerated in cardiac AL amyloidosis
Kaul, et al.	12	Plerixafor and G-CSF	25% grade 1 AEs	13.8 × 106/Kg	Plerixafor allows collection of sufficient CD34+ cells for HSCT and cryopreservation
Dhakal, et al.	49	G-CSF alone Plerixafor and G-CSF	28% hospitalization 37% hospitalization	6.3 × 106/Kg 12.8 × 106/Kg	Combination of plerixafor and G-CSF results in a more effective PBSC collection
Lisenko, et al.	110	G-CSF alone CAD and G-CSF Ifosfamide and G-CSF	10% grade 3 AEs 26% grade 3 AEs 43% grade 3 AEs	4.9 × 106/Kg 8.8 × 106/Kg 19.1 × 106/Kg	CAD mobilization is an effective approach and tolerability may be implemented suspending doxorubicin. G-CSF alone is also effective and safe
Yeh, et al.	101	G-CSF (and plerixafor if needed)	14% hospitalization 4% mortality	-	Low serum albumin, high NT-proBNP and increased IVS thickness were associated with higher morbidity/mortality
Badar, et al.	53	Plerixafor and G-CSF	2% grade 2 AEs	12.4 × 106/Kg	Mobilization with plerixafor and G-CSF is effective and safe in AL amyloidosis

CAD, cyclophosphamide, doxorubicin and dexamethasone; CTX, cyclophosphamide

^*These two papers reported a large series of patients with AL amyloidosis treated with HSCT by Boston and Mayo Clinic groups. Only information about mobilization are reported in the table. In Skinner’s paper, a requirement for HSCT was the collection of at least 2.0 × 10⁶ CD34+ cells/Kg (median value not reported).

Conditioning

Proper dosing of melphalan, the standard conditioning agent for amyloidosis, is important to optimize outcomes. Southwest Oncology Group investigated two sequential cycles of modified high-dose melphalan at 100 mg/m² and reported a treatment-related mortality of 12% and grade 3 and higher non-hematologic adverse events in 81% of patients. Reported outcomes with this approach do not appear to be superior to single induction with melphalan 200 mg/m².(22) Risk adjustment of the melphalan dose is used at some centers. One center reported success by using melphalan 200 mg/m² in patients who had a performance status of zero or one, two or less organs involved, a creatinine of 1.5 mg/dL or less, an ejection fraction >50%, and a BNP of 200 pg/mL or less. All other patients received 140 mg/m². In this schema, there was no treatment-related mortality.(23) In the Mayo Clinic experience, full-dose conditioning at melphalan 200 is associated with a higher rate of very good partial response or better, complete response, and organ response compared to reduced-dose conditioning. We also found progression-free survival to be superior in the full-intensity group at four years, 55 versus 31% in the reduced-intensity group, as well as a higher overall survival rate at four years. Conditioning dose remained an independent factor for progression-free and overall survival. As a consequence at Mayo and Heidelberg, we generally will not transplant a patient if we believe they will not tolerate full-dose melphalan 200, as we have been unable to demonstrate that melphalan 140 is better than current novel agent-based chemotherapy.(24) Boston University does use a risk-adapted modified conditioning dose of melphalan for patients with poor baseline functional status, advanced age, renal compromise, and cardiac involvement. In this very large cohort median overall survival was 6.1 years. Median event-free survival was 4.3 years and was driven by hematologic complete response. The treatment-related mortality was 3%. This group continues to use modified doses of melphalan for conditioning.(25) The Boston group has also reported on a second course of high-dose melphalan for the treatment of relapsed disease. Eleven patients achieved a hematologic complete response at one year. Three patients died of progressive disease. Only one-third of patients who relapsed after a first stem cell transplant achieve hematologic complete response with a second stem cell transplant.(26)

Consolidation treatment

We have explored the possibility of using consolidation chemotherapy following autologous stem cell transplant to deepen response and maintain the response. We identified 72 patients receiving consolidation. In patients who achieved a very good partial response or better at day 100, consolidation therapy post-autologous transplant did not improve progression-free or overall survival. However, for patients who achieve less than a very good partial response at day +100, consolidation improves progression-free survival for patients and should be strongly considered for patients who, at day 100, have not achieved a very good partial response. In one trial patients with persistent clonal plasma cell disease 3 months post-SCT received 9 months of adjuvant thalidomide/dexamethasone. 42% achieved an improvement in hematological response. By intention-to-treat, overall hematological response rate was 71% (36% complete response), with 44% having organ responses.(27) In a risk adapted trial of stem cell transplantation in 40 patients those patients with less than complete hematological response (CR) received bortezomib dexamethasone consolidation. Twenty-three patients received consolidation and in 86% response improved demonstrating the value of consolidation therapy in patients whose responses felt not to be sufficiently deep to result in an organ response.(28) this trial was subsequently updated to include 143 patients receiving risk adapted transplantation. The CR rate was particularly high (62%) in patients offered bortezomib consolidation. Median event-free survival with risk adapted stem cell transplantation was 4.04 years and median OS was 10.4 years.(29)

Prognosis

A number of attempts have been made to assess prognosis following autologous stem cell transplantation. In a case series of 82 patients undergoing stem cell transplantation, two-dimensional global longitudinal strain echocardiography was highly predictive of survival in patients with AL amyloidosis. A global longitudinal strain of - 17% was associated with improved survival.(30) Predictors of early treatment failure following initial therapy of AL amyloidosis has also been assessed. Patients with early treatment failure were older, had a higher prevalence of cardiac and multi-organ involvement, and had a higher proportion of patients with the t(11;14) genetic abnormality. In a multivariate analysis, the presence of t(11;14) and failure to proceed with autologous stem cell transplant were significant predictors of early treatment failure, suggesting that stem cell transplant is a very important component of management.(31)

The most important endpoint following stem cell transplant in amyloidosis is normalization of the serum-free light chain and, in fact, the achievement of MRD negativity. In amyloidosis, since the light chain is the precursor of the amyloid protein, virtual complete eradication of the amyloidogenic light chain is required for optimal response. In one Mayo Clinic study of 313 patients, higher values of the free light chain at diagnosis predicted overall response, and an early overall response predicted an improved overall survival.(32) A stringent complete response is the optimal endpoint after stem cell transplantation in immunoglobulin light chain amyloidosis. We assessed a total of 540 patients, of whom 220 achieved a complete response. Progression-free survival was significantly shorter in patients failing to achieve a stringent complete response. A bone marrow examination post-transplant is important because it can identify patients who fail to achieve a stringent complete response and thereby predicts earlier progression.(33) The endpoint of treatment with stem cell transplantation should be stringent complete response with normalization of the light chain ratio.

Outcomes

Long-term outcomes have been well-reported for autologous stem cell transplant. Mayo Clinic report of ten-year survival after stem cell transplant, 43% of patients were ten-year survivors.(30) We recently updated this to our 15-year overall survival rate.(31) A higher proportion of patients with 15-year survivorship received full-dose melphalan conditioning and achieved a complete response. The median overall survival among those patients who achieved a complete response was 19.3 years. Fifteen-year overall survival was observed in 30% of patients. A report from the Bone Marrow Transplant Registry has shown improved outcomes after autologous stem cell transplant for light chain amyloidosis.(32) Post-transplantation survival in AL has improved with a dramatic reduction in early post-transplant mortality and excellent five-year survival. From 2007 to 2012, therapy-related mortality at 100 days post-transplant was 5%. In addition, five-year overall survival improved to 77%.(32) We have identified that a bone marrow plasma cell percentage of 20% or greater has an important impact on response and survival. These patients have a poor outcome independent of their cardiac risk factors and stem cell transplant eligibility.(33)

There are special subsets of patients with amyloidosis. One is IgM amyloidosis present in 7% of light chain amyloidosis patients. These patients have similar survival but very different presentations, with a higher incidence of neurologic and pulmonary involvement. When these patients undergo stem cell transplant, it is effective, and therapy is safe. The median progression-free survival is 93 months for patients achieving a response.(34)

The National Amyloidosis Center of Britain reported 264 patients transplanted. The median overall survival post-transplant was 87%, and a hematological response was achieved in 94.8%. The reported treatment-related mortality from 2013 to 2018 was 1.1%.(35) Stem cell transplantation was compared directly with melphalan plus dexamethasone in the treatment of light chain amyloidosis. In a matched control series, patients undergoing stem cell transplant had a three-year overall survival rate of 83.6% and was far superior to melphalan and dexamethasone, with an overall mortality hazard ratio of 2.56.(36) Another comparative trial of outcomes using stem cell transplant versus standard chemotherapy was reported.(37) The majority of patients in the chemotherapy cohort received bortezomib-based treatment. The median progression-free and overall survivals were superior in the transplant cohort not reached versus nine months, and overall survival was 74 months versus 8 months. A multivariable analysis demonstrated an improved progression-free survival and overall survival were associated with stem cell transplant compared with chemotherapy. The first phase 3 study comparing chemotherapy with transplantation in 100 patients was inconclusive because of very high therapy-related mortality associated with stem cell transplant (9 deaths among 37patients) and that many patients assigned to stem cell transplant were never transplanted (26%).

Summary

Autologous stem cell transplant is a highly effective technique for the treatment of amyloidosis. Patients can be selected to reduce therapy-related mortality below 3%. The goal of therapy is normalization of the involved immunoglobulin free light chain and its ratio.

Key points:

1. Stem Cell Transplantation results in high hematologic and organ response rates in amyloidosis

2. Virtually all eligible patients can undergo safe stem cell transplantation

3. With proper patient selection therapy related mortality is now <3%

Synopsis.

Stem cell transplantation was one of the first proven effective regimens for the management of immunoglobulin light chain amyloidosis. Criteria for patient selection and the mobilization regimen become important features in ensuring a safe outcome. The technique of stem cell transplantation has evolved considerably in parallel with the development of new chemotherapeutic agents for the management of amyloidosis. Optimal outcomes require both the use of effective novel agent induction and appropriate application of high-dose chemotherapy with subsequent stem cell reconstitution.

Clinic care points.

1. Stem Cell transplantation is an important and effective therapy for the management of amyloidosis

2. The procedure should only be undertaken if the estimated therapy related mortality is less than 3%

3. Stem cell mobilization is successful in over 95% of patients and can be done using cyclophosphamide with filgrastim or filgrastim and plerixafor