Patients and methods

We retrospectively reviewed medical records of all patients with primary systemic AL amyloidosis who had been treated with ASCT at our institution. The patients were identified by searching our ASCT Database into which data are prospectively entered as a quality assurance activity. All patients gave written and verbal consent to proceed with ASCT, and to have their outcome data entered into the database, analyzed and published. The ASCT consent form was approved by our research ethics board.

Diagnosis of primary AL amyloidosis required histopathological demonstration of amyloidosis by Congo red stain on organ biopsy together with evidence of plasma cell monoclonality as defined by the presence of a urine or serum monoclonal paraprotein, light chain restriction of bone marrow plasma cells or proof of light chain amyloid by immunohistochemical stains. Three patients did not have serum monoclonal gammopathy or urine light chains, but all three had light chain restricted bone marrow plasmacytosis between 5 and 15%, two had amyloid identified on bone marrow biopsy, two had amyloid typed by immunohistochemical stains of organ biopsy to demonstrate light chain amyloid, and one had measurable serum free light chains. Cardiac involvement was determined by clinical symptoms of congestive heart failure, pleural and/or pericardial effusions; as well as echographic evidence of concentric hypertrophy with abnormal echogenicity of the septum and/or septal enlargement >11 mm. No patients had hereditary, localized AL, or secondary amyloidosis, nor overt multiple myeloma.

A total of 15 patients with primary systemic AL amyloidosis were identified who received ASCT between October 1997 and January 2005. All patients were less than 65 years of age and had a WHO performance status 0–2. Only one patient had received a course of chemotherapy prior to stem cell mobilization. Patient characteristics and main clinical manifestations of primary amyloidosis are summarized in Table 1. Major organ sites were considered to be cardiac, renal, neurologic and gastrointestinal (intestinal and hepatic included in this category). Patients with significant weight loss, hypotension (systolic blood pressure <90 mmHg), syncope and gastrointestinal bleeding prior to transplantation; as well as those with overt organ dysfunction including congestive heart failure and liver dysfunction (hepatomegaly and alkaline phosphatase >200 IU/l) were included as transplantation candidates. Seven patients had nephrotic range proteinuria, one patient had mild azotemia, but no patient had a creatinine >200 mol/l.

Table 1 - Patient characteristics.

Full table

The patients were divided into three risk categories according to Comenzo and Gertz¹⁵ as follows: (1) good risk patients are of any age and have 1–2 organs involved, no cardiac involvement and creatinine clearance >50 ml/min, (2) intermediate risk patients are <71 years old and have 1–2 organs involved, one of which must include cardiac or renal with creatinine clearance <51 ml/min, and (3) poor risk patients have either three organs involved or advanced cardiac involvement.

A total of 12 patients received cyclophosphamide 2 g/m² day 1 with GCSF 300 g/day (<70 kg) or 480 g/day (>70 kg) s.c. days 7–12 to mobilize blood stem cells. One patient with biopsy-proven lymph node involvement with amyloidosis and some evidence of a proliferative B-cell disorder in her bone marrow received one cycle of CHOP (cyclophosphamide, adriamycin, vincristine, prednisone) plus G-CSF to mobilize stem cells. The initial patient transplanted in 1997 had previously received three cycles of melphalan and prednisone pre-transplant and was treated with DICEP (dose intense cyclophosphamide 4.5 g/m², etoposide 750 mg/m² and cisplatin 75 mg/m²) as the mobilization regimen.

HDM consisted of 200 mg/m² administered by i.v. push on day -1 for 14 patients. The final patient had severe cardiac dysfunction with elevated proBrain Natiuretic Peptide (proBNP) level of 3206 pg/ml (reference: >900 indicated dyspnea of cardiac origin), troponin T level of 0.08 g/l (normal <0.03), troponin I level of 0.38 g/l (normal <0.05), and intraventricular septum (IVS) width of 14 mm on echocardiogram. This patient received an arbitrary 15% dose reduction of melphalan to 170 mg/m² on day -1. GCSF at 300 g/day (<70 kg) or 480 g/day (>70 kg) was given from day +7 post stem cell infusion until the absolute neutrophil count (ANC) was >1.5 10⁹/l. All patients received 5–6 units of random donor platelets for a platelet count less than 20 10⁹/l and two units of packed red blood cells for hemoglobin less than 80 g/l or symptomatic anemia.

Results

Response assessment

Table 2 summarizes the sites of involvement, prognostic risk category and response to therapy for all 15 patients. Complete organ response was defined as resolution of nephrotic range proteinuria (<3 g/day), resolution of nephropathy, normalization of echocardiogram, and reduction of liver span and liver enzymes back to baseline for each organ, respectively. Patients were considered to have improved if they had symptomatic improvement of organ-related symptoms or objective improvement of post-ASCT investigations. Stable disease was defined as no worsening or improvement of organ dysfunction. Organ complete responses occurred in four patients (27%), three of them, who had involvement of only one major organ system. Clinical improvement or organ stability was seen at 1 year in an additional seven patients (52%) patients, giving an overall clinical benefit in 11 of the 14 patients evaluable 1 year post-ASCT (79%).

Table 2 - Sites of amyloid involvement and response to ASCT.

Full table

Complete hematologic response was defined as <5% bone marrow plasma cells and the absence of monoclonal paraprotein in the urine and serum by serum protein electrophoresis. Complete hematological responses sustained for 12 months were observed in nine of 14 (64%) evaluable patients.

Of the four patients in the poor risk category, two patients died of progressive disease at 10 and 30 months post-ASCT. A third patient progressed 7 months post-ASCT with worsening renal dysfunction, but remains alive on dialysis 55 months post-ASCT. The final poor prognosis patient is only 3 months post-ASCT and at this time, he already has some improvement in his symptoms of cardiac failure. His Kappa/lambda free light chain ratio has decreased from 43 to 11.7 at 3 months post transplant and his serum kappa light chain decreased by >50%. However, his intraventricular septal thickness remains unchanged (13 mm) and his serum Troponin I and proBNP remain elevated (0.3 g/l and 9043 pg/ml, respectively). All four patients with intermediate risk disease remain alive without recurrent disease from 41 to 67 months post-ASCT. One of these patients initially presented with symptomatic cardiac failure requiring diuretics and afterload reduction with an angiotensin receptor blockade. By the time of ASCT he had colonic involvement with gross hematochezia, liver dysfunction, salivary gland involvement, and a WHO performance status of 2. Despite these high-risk features, he tolerated HDM/ASCT well, and remains in complete organ and hematological remission 62 months post-ASCT. Interestingly, he had significant fatigue as well as lower extremity edema requiring diuretic therapy until 11 months post-ASCT. Since that time he has been working fulltime as a foreman for the city. A second intermediate risk patient had massive hepatomegaly, diarrhea secondary to biopsy-proven intestinal amyloid, and biopsy-proven renal amyloid with a creatinine of 117 mol/l, creatinine clearance of 72 ml/min, and non-nephrotic proteinuria. She achieved improvement in her liver disease, stable renal function, complete hematological response, and remains alive 45 months post-ASCT. Five of seven patients in the good risk category remain symptomatically well 17–90 months post-ASCT, without any evidence of progressive disease. The other two progressed at 20 and 36 months post-ASCT.

Survival

Vital status was censored on 1 May 2005 with a median post-ASCT follow-up period of 45 months. In all, 12 patients (80%) remain alive and 10 patients (67%) remain symptomatically well, not requiring other therapy. The actuarial 4-year event-free and overall survival rates were 60% (95% CI 32–89%) and 75% (95% CI 50–100%), respectively (see Figure 1).

Figure 1.

Survival following high-dose melphalan and ASCT for 15 patients with primary systemic AL amyloidosis.

Full figure and legend (14K)

Engraftment and toxicity

The patients underwent apheresis of a median of 24 l of blood (range 10.1–65.3 l) to collect stem cells. A median of 7.6 10⁶ CD34⁺ cells/kg (range 3.9–17.9 10⁶ CD34⁺ cells/kg) were infused on day 0. Time to engraftment of ANC >0.5 10⁹/l was 12 days (11–16) and to platelets >20 10⁹/l was 12 days (10–20).

During stem cell mobilization with cyclophosphamide and GCSF, one patient developed transient liver dysfunction with a peak of total bilirubin at 62 mol/l, and hematochezia secondary to raised nodules in the descending and transverse colon, biopsy-proven to be amyloid. Bleeding resolved with blood product support and resolution of pancytopenia. Two patients developed herpes simplex lesions of the skin prior to ASCT, which resolved with acyclovir. One patient developed Respiratory Syncytial Virus pneumonia prior to ASCT, which resolved with Ribavirin and intravenous immunoglobulin.

For all 15 patients treated with HDM/ASCT, there was no treatment-related mortality and no requirement for intensive care support. On day 0 of ASCT, two patients with cardiac involvement developed diuretic responsive pulmonary edema. One of these patients developed transient azotemia with peak creatinine of 226 mol/l on day 2. A patient with nephrotic syndrome also developed transient azotemia with peak creatinine of 218 mol/l on day 5. Post-ASCT, two patients developed hemorrhagic cystitis, which resolved with hematopoietic engraftment, and one patient developed corticosteroid-responsive engraftment syndrome with fevers and hypoxia. In the initial 100 days post transplant, bacterial bronchial pneumonia was observed in two patients, one patient had a genital herpes simplex outbreak and three patients developed varicella zoster skin lesions.

Discussion

There are no randomized studies comparing standard therapy with melphalan and prednisone to HDM and ASCT. Nevertheless, ASCT has become an acceptable treatment approach for patients with primary AL amyloidosis based on similarities of the disease to multiple myeloma, as well as encouraging reports of ASCT outcome from major amyloidosis referral centers. Of concern, however, is the high reported TRM of ASCT in this setting despite the significant selection bias of patients offered ASCT. Eligibility itself for transplantation has been demonstrated to be a favorable prognostic factor for survival as demonstrated in one study which included age less than 70, cardiac septal thickness <15 mm, cardiac ejection fraction >55% and serum creatinine <2 mg/dl and direct bilirubin <2.0 mg/dl.¹³ This highly selected cohort was not treated with ASCT and the median survival with chemotherapy was 42 months, much better than the previous expected 18-month survival in all patients with amyloidosis.¹³ Based on this information, one may question the advisability of nonreferral centers performing ASCT for this disease.

The largest prior study included 701 consecutively referred primary AL amyloidosis patients and reported that only 56% of patients met eligibility criteria for ASCT.¹⁶ For the 312 patients who initiated treatment in this study, the median survival was 4.6 years, and 100 day TRM rate was 13%. Higher organ response rates and overall survival were seen in the 40% of patients who achieved a complete hematological response to ASCT.¹⁶ A second pivotal retrospective study compared 63 patients who underwent HDM/ASCT with 63 matched controls who received standard chemotherapy.¹² This study suggested a survival benefit of HDM/ASCT at 1 (89 vs 71%), 2 (81 vs 55%) and 4 years (71 vs 41%).¹² HDM/ASCT conferred a survival benefit despite a reported TRM rate of 13%.¹² These recent studies suggest that HDM/ASCT is an effective treatment for primary AL amyloidosis.

Our site performs approximately 120–140 combined autologous and allogeneic hematopoietic stem cell transplants per year for a variety of conditions. Although we are not a tertiary referral center for primary AL amyloidosis, our 15 patient experience is similar to that reported in larger series. Five (36%) of our patients achieved significant or complete improvement in organ function, and only three (20%) did not at least experience stabilization of organ function. Although other centers have reported organ response rates of approximately 50% for melphalan doses 200 mg/m², they often did not treat patients who had adverse prognostic features such as cardiac disease or multi-system involvement with this high dosage, thereby biasing response rate assessment in favor of 200 mg/m² relative to those seen with 140 mg/m². The hematological complete remission rate of 64% we report may have been overestimated because immunofixation was not routinely performed post-ASCT. Perhaps measurement of free serum light chains will further refine the definition of hematological remission in the future. The 3 year survival rate of 75%, and lack of treatment-related mortality compares favorably with other series. Our patient population, response rates and lack of treatment-related mortality are similar to that of the ECOG study by Gertz et al¹⁷ which indicated that ASCT for amyloidosis could be performed at transplant centers with limited experience in managing AL amyloidosis.

Limitations of our experience include a small, heterogeneous group of patients treated over a prolonged period of time. We also are unable to compare our results to a control group of patients who did not receive ASCT. Furthermore, primary AL amyloidosis is not reported to the Alberta Cancer Registry and we, therefore, cannot identify the number of primary AL amyloidosis patients who were diagnosed during the same years and were not referred for ASCT. Another limitation of these results is that all patients did not undergo typing of the amyloid protein. It has been reported that a diagnosis of amyloid by Congo Red staining with green birefringence and/or positive electron microscopy is not sufficient to make a diagnosis of AL amyloid, even in the presence of a monoclonal gammopathy, and that typing of the amyloid should be routinely performed to rule out the possibility of hereditary amyloidosis.¹⁸

The decision as to who should receive ASCT is a difficult one. The mortality from untreated amyloidosis is high, and ASCT gives the best opportunity for long lasting hematological and organ improvement and complete response. However, the benefits are balanced by a high TRM rate. Prognostic risk categorization based on organ involvement, cardiac involvement and renal function has been attempted to better select who should be offered ASCT.¹⁵ Some studies indicate that patients with symptomatic cardiac involvement and those with greater than two organs involved with amyloid should not be offered ASCT due to a high risk of mortality.^{6, 11, 15} However, without transplantation, these patients have a life expectancy of <6 months and HDM/ASCT may be the only chance for symptomatic improvement and improved survival. Recently, Disperzieri et al¹⁹ reported that troponins (cTNT) and N-terminal proBrain Natriuretic Peptide (NT-proBNP) levels effectively predict survival following ASCT, and elevations of troponin I predict for TRM. This newly proposed staging system helps select a subgroup of patients with high cTNT and NT-proBNP (stage III-i or III-t) with a very poor median survival post-ASCT. These studies suggest that this high-risk group should be enrolled in clinical trials with new agents rather than undergo stem cell transplant. The distinction is less clear for the earlier stages and longer follow-up is needed in order to optimize this new staging system.¹⁹ These tests were measured on our most recent patient with cardiac amyloidosis, and both were elevated. For this reason, the dose of melphalan was arbitrarily decreased to 170 mg/m² and was fairly well tolerated. The optimal, or maximum tolerated dose of melphalan still needs to be properly evaluated in phase I trials for AL amyloidosis patients who have severe cardiac disease or 3–4 major organs involved.

One of our most remarkable patients presented at age 44 years with symptomatic congestive heart failure, pericardial and pleural effusions, and lower GI bleeding from extensive, biopsy-proven colonic amyloidosis. Since he had a WHO performance status of 2 on cardiac medications, no significant azotemia, and gave written informed consent despite being told the high risk of TRM, we proceeded with HDM/ASCT. He went on to have a complete organ and hematological response and remains symptom free, off cardiac medications 62 months post-ASCT.

One intermediate risk patient was a 44 year old female with predominant hepatic involvement with an elevated alkaline phosphatase, renal, and gastrointestinal involvement. Moreau et al⁷ reported that hepatomegaly associated with alkaline phosphatase level >200 IU/l was one of the poor prognostic factors associated with inferior response and survival. However, our patient with liver involvement remains relapse-free 45 months post-ASCT. Our two cases illustrate that patients with symptomatic cardiac or hepatic dysfunction with multi-organ involvement should still be considered for transplantation because it is possible that they can respond and have prolonged survival.

The most remarkable result was the lack of TRM experienced by our 15 patients following HDM and ASCT. This lack of TRM may have been due to referral bias, but we included all referred patients, even those with multi-organ involvement, syncope, hypotension, and symptomatic cardiac disease or liver dysfunction. Of note, however, is that all of our patients were <65 years of age, had WHO performance status levels 2, and had creatinine levels <200 mol/l. One study does concur that patients with WHO performance status >2 should not be transplanted.¹⁴ In addition, 14 of 15 patients had no pretreatment with melphalan and prednisone. A recent study has demonstrated that newly diagnosed patients with AL amyloidosis eligible for ASCT did not benefit from initial treatment with oral melphalan and prednisone and there was a survival disadvantage for patients with cardiac involvement if ASCT was delayed by oral melphalan.²⁰ The low toxicity and absence of TRM may have been due to ASCT expertise at our center, but this is not specifically related to amyloidosis due to the rarity of the disease.

In conclusion, it is feasible to safely administer HDM/ASCT to patients with primary AL amyloidosis at transplant centers that do not have specific expertise with AL amyloidosis. This is true even those with poor prognosis, multi-organ involvement who are less than 65 years of age and have a reasonably good performance status of level 2 or better.

References

References