Outcome of Patients with Immunoglobulin Light Chain Amyloidosis with Lung, Liver, Gastrointestinal, Neurologic and Soft Tissue Involvement after Autologous Hematopoietic Stem Cell Transplantation

Abstract

Background

There is limited information on the outcome when organs other than heart or kidneys are involved by immunoglobulin light chain amyloidosis (AL).

Methods

We report the outcome of 53 patients with AL with gastrointestinal (GI), peripheral nerve (PN), liver, lung or soft tissue involvement, who underwent high-dose chemotherapy and autologous hematopoietic stem cell transplantation (auto- HCT) at our institution between 1997 and 2013.

Results

Median age at auto-HCT was 56 years (range, 35-74). One, two, or three or four organs were involved in 43%, 22%, 28% and 4% of patients, respectively. Concurrent cardiac, renal or both were involved in 24 (45%) patients. Forty-six patients received induction therapy before auto-HCT. The 100-day and 1-year treatment-related mortality (TRM) were 3.8% (n=2), and 7.5% (n=4), respectively. Forty-one (80%) patients achieved a HR. Organ response at 1 year after auto-HCT was seen in 23 (57%) of the 40 evaluable patients. With a median follow-up of 24 months, the median progression free survival (PFS) and overall survival (OS) were 36 and 73 months, respectively.

Conclusion

Auto-HCT was associated with a low TRM, durable organ responses and a median OS of > 6years in selected patients with AL and GI, PN, liver, lung or soft tissue involvement.

Introduction

Immunoglobulin light chain systemic amyloidosis (AL) is a monoclonal plasma cell disorder characterized by the deposition of misfolded immunoglobulin free light chains as amyloid fibrils [1]. These free light chains are secreted by the clonal plasma cells and accumulate in multiple organs, causing organ dysfunction. The aim of therapy in AL is to reduce the production of misfolded light chains and preserve the function of involved organs. The treatment of AL is patterned after the treatment of multiple myeloma, another clonal plasma cell disorder. The treatment include the use of conventional cytotoxic chemotherapy agents such as melphalan and cyclophosphamide in standard doses [2, 3]; immunomodulatory agens (IMiD) including thalidomide [4] and lenalidomide [5]; the proteasome inhibitor (PI) bortezomib [6] in combination with corticosteroids; and high- dose melphalan followed by autologous hematopoietic stem cell transplantation (auto-HCT). Auto-HCT has been associated with longer survival in selected patients when compared with conventional chemotherapy agents [7-10], and thus considered an effective treatment strategy that has been associated with hematologic response, an improvement in organ function [8], quality of life and survival [11]. However, in the only phase 3, randomized trial, published by Jaccard et al, the outcome with high-dose melphalan plus auto-HCTwas not superior to standard-dose melphalan plus dexamethasone[12]. This trial was criticized because of the high treatment-related mortality rate (24%) in the high-dose melphalan group, a rate that is more than twice the rate in centers performing auto-HCT for A. That study was also notable for the inclusion of patients with more than 3 involved organs, a suboptimal dose of high-dose melphalan in 10 of 37 patients and the inclusion of transplant centers that perform one or fewer auto-HCT for AL in a year. Previous reports on the outcome of high-dose therapy and auto-HCT in AL have primarily focused on patients with cardiac or renal involvement [13, 14]. Limited data are available on the role of auto-HCT when organs other than heart and kidneys are involved by AL. In this study, we evaluated the role of high-dose melphalan followed by auto-HCT in patients with gastrointestinal tract (GI), peripheral nerves (PN), liver, lungs, or soft tissues involvement by AL.

Materials and Methods

Patients and Diagnosis

Diagnosis of AL was established by Congo red staining for amyloid fibrils with concomitant demonstration of plasma cell clonality by serum free light chain (FLC) studies, serum immunofixation electrophoresis (IFE) and bone marrow biopsy and immunohistochemistry (CD138, kappa or lambda FLC predominance) according to the criteria established at the 10th International Symposium on Amyloid and Amyloidosis [15, 16], In selected cases, free light chain origin of amyloid fibrils was confirmed by mass spectrometry or electron miscroscopy.

Involvement of individual organs (GI, PN, liver, lung or soft tissue) by AL was also established according to the criteria established at the 10th International Symposium on Amyloid and Amyloidosis [15]. Patients with GI involvement had endoscopic evaluation and biopsies. Patients with liver involvement were evaluated by liver function tests, imaging studies included CT scans and ultrasound, and liver biopsy when feasible. Patients with lung and soft tissue involvement were evaluated by imaging studies, including CT scan and biopsy of the involved organ when feasible. Patients with PN involvement were evaluated by electromyography (EMG), nerve conduction studies (NCS) and sural nerve biopsy (2 patients) to confirm the diagnosis. All patients underwent auto- HCT between 1997 and 2013 at the University of Texas MD Anderson Cancer Center.

Auto-HCT

The choice of preparative regimen was based on the patient's performance status and the treating physician's preference. Patients received high-dose melphalan at 200mg/m² or 140 mg/m² at the treating physician’s discretion.

Hematologic and Organ Response assessment

All patients underwent clinical, laboratory and radiological evaluation at baseline and at 100 days and 1 year post auto-HCT. Hematologic response (HR) was defined according to the IMWG criteria [17]. Organ involvement and responses were defined by the criteria established at the 10th International Symposium on Amyloid and Amyloidosis [15].

Statistical analysis

The primary endpoint of this study was to assess the hematologic and organ response after auto-HCT. Treatment related mortality (TRM) was defined as death due to any cause other than relapse of AL within 100 days or 1 year after auto-HCT. Overall survival (OS) was defined as time from auto-HCT to death or last follow-up. Progression-free survival (PFS) was defined as time from auto-HCT to disease progression, death or the last follow-up. Survival was estimated by using the Kaplan–Meier method. Fisher’s exact test and Chi-square test were used for analyzing the differences between categorical variables. The associations of multiple factors on OS were evaluated using the Cox proportional hazards model. All P-values were two-tailed, and we accepted P ≤ 0.05 as significant. All statistical analyses were performed using STATA (StataCorp. 2011. Stata Statistical Software: Release 12. College Station, TX: StataCorp LP.)

Results

Patients

We identified 53 patients with AL who met the inclusion criteria for these analyses. Patient characteristics are summarized in Table 1. The median follow up among surviving patients was 25 months (range, 3-144).

Table1.

Clinical characteristic

Clinical characteristic	Group	Numbers (%)
Age (years)	<56 ≥56	25 (27) 28 (53)
Sex	Male Female	32 (60) 21 (40)
Light chain subtype	Kappa lambda unknown	19 (36) 32 (60) 2 (4)
dFLC( mg/dl) pre-transplant	<18 ≥18 unknown	21 (40) 21 (40) 11 (20)
Bone marrow	<10% ≥10%	21 (40) 32 (60)
Bone marrow amyloid	Absence Presence	36 (68) 17 (32)
β2 microglobulin (mcg/ml)	<3.5 ≥3.5 unknown	36 (68) 16 (30) 1 (2)
Serum creatinine (mg/dl)	<2 ≥2	47 (89) 6 (11)
Urine protein 24 hours (g/d)	≤1 >1 unknown	35 (66) 17 (32) 1 (2)
Albumin serum	<3.5 ≥3.5	15 (28) 38 (72)
Alkaline phosphatase	≤126 >126	40 (75.5) 13 (24.5)
Induction type	Novel* conventional no induction	33 (62) 13 (25) 7 (13)
Melphalan dose (mg/m2)	<200 200	7 (13) 46 (87)
Number of involved organ (including heart and Kidney)	1 2 3 4	23 (43) 12 (23) 15 (28) 3 (6)

^*Novel therapy defined as IMid or PI based therapy alone or in combinations.

Hematologic Involvement at Diagnosis

Thirty-two (60%) patients had λ FLC subtype while 19 (36%) patients had κ FLC subtype of AL. Serum FLC were available in 42 (79%) patients before auto-HCT. The median difference in involved and uninvolved FLC (dFLC) was 59 (range: 0 – 1237) pre-transplant. As shown in Table 1, 21 (39%) patients had a dFLC of <18, while 21 (39%) patients had a dFLC of >18. Median plasma cell % in the bone marrow at diagnosis was 13% (range 1-100).

Organ Involvement at Diagnosis

Organ involvement was defined according to the ISA criteria [15]. Clinical features of organ involvement are summarized in Table 2. Patients with GI involvement presented with abdominal mass, bleeding, diarrhea, motility disturbance, nausea or weight loss. Upper or lower GI endoscopy was performed to confirm the diagnosis of AL. A limited number of patients also underwent Barium studies to evaluate the GI tract. Patients with neurologic involvement had symmetrical peripheral neuropathy or autonomic neuropathy. In selected patients, electromyography (EMG) or nerve conduction studies (NCS) were also performed. Sural nerve biopsies were also performed in 2 of 17 (12%) patients with peripheral neuropathy. Patients with liver involvement presented with hepatomegaly or abnormal alkaline phosphatase (at least 1.5 times the upper limit of normal). Ultrasound or CT scans were performed to confirm hepatomegaly. Liver biopsy was done if amyloid deposition was not confirmed elsewhere. Patients with lung involvement presented with either dyspnea or hemoptysis. They had diffuse interstitial infiltrates, lung nodules or mass lesions on imaging studies. Lung biopsy was performed to confirm the diagnosis. Patients with soft tissue involvement presented with either tongue enlargement, para-spinal muscle mass, muscle weakness or pain, diffuse skin nodules, or lymph node enlargement. Organ involvement was further confirmed by imaging studies including CT scans, and biopsy of the involved organ if a diagnosis of AL had not been established elsewhere.

Table 2.

Clinical Presentation Of The Involved Organs

Evaluation of the involved organs	Clinical	Laboratory	Imaging	Other
Kidney	Generalized edema, ascites, dypnea	Proteinuria Biopsy	NA	AL deposition on renal biopsy
Heart	Dyspnea, pedal edema, fatigue	Elevated NT-Pro BNP; Troponin T or I	Echo	Endomyocar dial biopsy
Peripheral Nerve	Pain, tigling, numbness, parestheis	Nerve biopsy	NA	EMG, NCS
GI	Diarrhea, nausea, weight loss, constipation		Barium studies	Upper or lower GI endoscopy
Soft tissue	Tongue enlargement, Lymph node enlargement, Muscle pain or weakness, Carpal tunnel syndrome	NA	CT scan of the involved area	Biopsy of the involved organ
Liver	Hepatomegaly	Abnormal liver function tests	CT scan, Ultrasound	Liver biopsy
Lung	Shortness of breath, hempotyiss	NA	CXR, CT scan	PFT, lung biopsy

Patients with cardiac involvement most commonly presented with symptoms of congestive heart failure, and the diagnosis was established by an echocardiogram, cardiac biomarkers or endomyocardial biopsy in selected patients. Patients with renal involvement most commonly presented with generalized edema, with proteinuria >2 g/24 hours and renal biopsy was performed in selected patients to confirm AL deposition.

Twenty-nine (55%) patients did not have heart or kidney involvement at diagnosis. Twenty-three (43%) patients had a single organ involved which was not heart or lung. That includes peripheral nerves (10), soft tissues including lymphadenopathy, macroglossia or skeletal muscles (5), GI tract (4), liver (3) and lungs (1). Twelve (22%) patients had 2, 15 patients (28%) had 3, and 3 patients (5%) had 4 organs involved. There were 24 (45%) patients with concurrent renal or cardiac involvement. Nine (17%) patients had concurrent heart and kidney, 9 (17%) patients had heart and 6 (11%) had kidney involvement in addition to the organs described in this paper.

Induction Therapy

Overall, 33 (62%) patients received pretransplant induction therapy with either an IMiD or a PI, 13 (25%) received conventional chemotherapy and 7 (13%) received no induction prior to auto-HCT. In 33 patients who received induction with either a PI or IMiD, 14 (26%) patients received a bortezomib-based regimen, 9 (17%) patients received bortezomib in combination with lenalidomide or thalidomide, 10 (18%) received an IMiD-based regimen (lenalidomide 8, thalidomide 2) withoutout bortezomib. The median duration of induction therapy prior to auto-HCT was 3.6 months.

TRM

The 100-day TRM was 3.8% (n=2). One patient died of sepsis and multiple organ failure and the other due to thrombotic thrombocytopenic purpura (TTP). The 1 year TRM was 7.5% (n=4). Two more patients died between 100-day and 1-year evaluations. One patient died of sepsis, while the cause of death was unclear for the second patient.

Hematologic response (HR)

The HR was evaluated 100 days after auto-HCT. Two patients died before 100-day evaluation. According to IMWG criteria [17], of the 51 evaluable patients, 41 (80%) achieved an HR (≥PR). Ten (20%) patients achieved a complete response (CR), 11 (21%) a very good partial response (VGPR), and 20 (39%) achieved a partial response (PR). Serum free light chain (FLC) assays were available in 42 (79%) patients both before and after auto-HCT. The median difference in involved and uninvolved FLC (dFLC) was 59 (range: 0 – 1237) before and 30.5 (range: 0.2 – 553) after the auto-HCT.

Organ response

Organ response was evaluated at 100-day and 1-year after auto-HCT. The response was assessed by clinical, laboratory imaging and other diagnostic tests to meet the response criteria established by the 10th International Symposium on Amyloid and Amyloidosis [15]. Organ response in at least 1 organ was seen in 30 (59%) of the 51 evaluable patients at 100-day, and 23 (57%) of the 40 evaluable patients at 1-year evaluation after auto-HCT. Only 40 patients had organ response evaluation at one year because 4 patients died before the one year landmark, and 9 patients were unable to return to our institution for detailed and reliable response evaluation. Organ response at 100-day evaluation was seen in 26 of 41 (63 %) patients with an HR, and in 4 of 10 (40%) patients (4/10) without a HR (P=0.2). Details of organ response are summarized below. There was resolution of gastrointestinal symptoms (nausea, vomiting, malabsorption, diarrhea) in 11 (52%) patients with GI involvement. There was an improvement in paresthesias and orthostatic hypotension in 8 (50%) patients with PN involvement, with improvement in EMG and NCS in selected patients. In patients with liver involvement, there was a 4 cm decrease in liver span in one patient and at least 50% decrease in alkaline phosphatase value in 4 patients. In patients with lung involvement, 4 (80%) had an organ response, with resolution or significant improvement of pulmonary infiltrates and nodules on CT scans, and improvement in symptoms and pulmonary function tests In patients with soft tissues involvement 7 (47%) had an organ response with improvement in or resolution of tongue swelling, dysphagia, lymphadenopathy, skin nodules and muscle pain or swelling (Table 3).

Table 3.

Organ response to auto-HCT

Organ Involved	Number (%)	Numbers Evaluable (%)	Organ Response (%)	Stable disease (%)	Progressed (%)
Gastrointestinal	24 (45)	21 (87.5)	11 (52)	9 (43)	1(5)
Peripheral nerve	17 (32)	16 (94)	8 (50)	8 (50)	0
Liver	8 (15)	6 (75)	5 (83)	0	1(17)
Lungs	5 (9)	5 (100)	4 (80)	1 (20)	0
Soft tissue	17 (32)	15 (88)	7 (47)	7(47)	1 (6)

Survival

The median overall survival (OS) after auto-HCT was 73 months (Figure 1). The median progression free survival (PFS) was 36 months (figure 2). In landmark analysis starting at 100 days after auto-HCT, the median OS for 41 patients with a HR at day 100 was 90 months vs. 34 months in 10 patients without a HR (P =0.4). The median OS was 94 months for 30 patients with an organ response at day 100 vs. 64 months in 21 patients without an organ response (P= 0.05). In landmark analysis starting at 1 year after auto-HCT, the median OS was 81 months for 23 patients with an organ response at 1-year vs. 55 months for 17 patients without an organ response (P=0.04).

Figure 1.

Median overall survival (OS) from auto-HCT

Figure 2.

Median progression free survival (PFS) from auto-HCT

Factors predicting survival before auto-HCT

On univariate analysis for 2-year OS, we evaluated the following variables: number of organs involved, percentage of plasma cell in bone marrow, free light chain subtype (kappa, lambda), 24-hour urine total protein value, β2 microglobulin, and serum albumin value at diagnosis and dFLC value pre auto-HCT. None of the factors analyzed had a statistically significant impact on the outcome (Table 4).

Table 4.

Univariate analysis for 2-year OS

Variable	Group	Hazard ratio	95% CI	P-value
Number of organ involvement	1-2 3-4	Ref. 1.9	0.5-7.3	0.3
Bone marrow PC%	<10 ≥10	1.4 Ref.	0.4-5.4	0.6
Free light chain subtype	kappa lambda	0.3 Ref.	0.03-2.2	0.2
DFLC ( mg/dl) pre- transplant	<18 ≥18	0.4 Ref.	0.1-2	0.25
Urine protein 24 hours (g/d)	<1 ≥1	0.4 Ref.	0.1-1.4	0.1
β2 microglobulin (mcg/ml)	<3.5 ≥3.5	0.3 Ref.	0.1-1.2	0.09
Albumin serum (g/dl)	<3.5 ≥3.5	2.3 Ref.	0.6-8.6	0.2

Discussion

We report the clinical characteristic and outcomes of 53 patients with AL with involvement of at least one of these 5 organs: GI, PN, liver, lungs, and soft tissues. Our results demonstrate that selected patients receiving auto-HCT can achieve durable hematologic and organ responses, with a long OS and relatively low TRM.

HR was seen in 80% patients [17], which is comparable to previous studies in similar patients [18-20]. Gertz et al. reported an overall hematologic and organ response of 64% in a multicenter trial that enrolled 30 patients [21]. In another study reported from the Mayo Clinic, 76% patients achieved HR, including 38.7% achieving CR [19]. Although overall HR in our study was comparable to other groups, the CR rate was somewhat lower (24%) than reported by others. This relatively lower CR rate may be due to several factors, such as response evaluation soon after auto-HCT underestimating the actual response, higher disease burden, and the use of lower melphalan dose in 13% of patients before auto-HCT. Organ response in at least one organ was seen in 59% of patients, which also compares favorably with other reports [19, 22]. Vesole et al. reported a functional improvement in at least one organ in 36% (28 of 77) patients at 100-day post auto-HCT [22]. Gertz et al reported outcomes of 434 patients with AL transplanted since 1996, and the organ response rate was 46.8% [19]. Some studies have shown a direct correlation between HR, organ response and OS. [11, 23]. Although our result showed a direct correlation between organ response and OS, there was no significant association between HR and organ response.

We previously showed durable HR and noticeable OS in Al patients with cardiac or renal involvement who received auto-HCT [24, 25]. Organ response in patients with renal involvement was 29% in 1 year after auto-HCT. One-year TRM in patients with cardiac involvement was less than 4%. In this study, we focused on the characteristics and outcomes of organs affected by AL that are relatively difficult for response assessment. Using the standard published criteria, encouraging responses were seen in the affected organs that were durable in most patients, with a low TRM. In general, AL is associated with high TRM after an auto-HCT [26, 27]. In a meta-analysis conducted by Comenzo et al. [28], 100-day TRM was 21% and 39% in single and multicenter clinical trials, respectively. However, a risk-adapted approach to high-dose melphalan significantly reduced the risk of TRM after auto-HCT [27]. The low TRM (3.4%) in our study may be attributed to the inclusion of patients with non-cardiac AL, improvement in supportive care, and melphalan dose reduction in patients with older age or significant comorbidities.

Our study had the usual limitations of a retrospective chart review, including a heterogeneous population, potential selection bias, and perhaps inclusion of relatively good risk patients that were able to undergo high-dose melphalan and less than one-third had cardiac involvement.

In conclusion, our results support the use of high dose melphalan followed by auto-HCT in selected patients with AL. We report the durable HR, organ responses, a median OS of > 6 years with low TRM.

Highlights.

• This report describes the outcome of 53 patients with AL with gastrointestinal (GI), peripheral nerve (PN), liver, lung or soft tissue involvement after auto- HCT

• Approximately 60% of evaluable patients achieved an organ response within a year

• Median PFS and OS were 36 and 73 months, respectively

• Auto-HCT was associated with a low TRM, durable organ responses and a median OS of > 6years in selected patients with AL and GI, PN, liver, lung or soft tissue involvement.