Cyclophosphamide, Bortezomib and Methylprednisolone (CyBorMe) for the Treatment of AL Amyloidosis: Initial Experience From a Single Center

Abstract

The use of cyclophosphamide, bortezomib and dexamethasone (CyBorD) is widely accepted in the treatment of AL amyloidosis (AL). Recently, the substitution of dexamethasone by methylprednisolone (CyBorMe) appeared to improve response rates and survival outcomes. All consecutive newly diagnosed AL amyloidosis treated with CyBorMe from 01/19 to 08/20 were evaluated. A historic cohort of patients treated with CyBorD was used for comparison (01/13–08/20). Methylprednisolone was given IV at 500 mg weekly for 4 weeks in the CyBorMe group. 43 patients were treated with CyBorD and 14 with CyBorMe. After a median of 4 cycles of CyBorD and 3 cycles of CyBorMe, Hematological Response was seen in 90.6% and 92.8% of cases, including CR in 28.5% and 35.7%, VGPR in 33.3% and 35.7% and PR in 30.9% and 21.4% for CyBorD and CyBorMe, respectively. Time to first response was faster in the CyBorMe group (4 vs. 6 weeks) and cardiac response was observed in 44% and 31% of patients treated with CyBorMe and CyBorD, respectively. CyBorMe appeared to be efficacious and well tolerated in patients with AL amyloidosis. Prospective studies with CyBorMe in the stage III/IV group are warranted aiming to minimize toxicity.

Introduction

Immunoglobulin light chain amyloidosis (AL) is a rare systemic disease characterized by a misfolded protein able to deposit in organs and compromise their function [1]. Recent clinical data suggests that AL amyloidosis clonal plasma cells do have an intrinsic sensitivity to bortezomib [2]. In 2012, two small retrospective studies reported on the use of a combination with cyclophosphamide, bortezomib and dexamethasone (CyBorD) for the treatment of AL amyloidosis [3, 4]. Palladini et al. also reported their experience in 230 AL amyloidosis patients treated with upfront CyBorD in two major amyloid referral centers in Europe [5]. This collaborative effort reported an overall hematological response of 60%, including complete response in 23% of patients. Additionally, a recent study reported on the use of bortezomib in 915 newly-diagnosed AL amyloidosis patients [6]. Overall response rate (intent-to-treat) was 65%, with 49% complete response (CR)/very good partial response/low dFLC response. Due to the potential effects of advanced cardiac disease on outcomes, recently the UK group has explored the use of methylprednisolone in combination with bortezomib and cyclophosphamide for the treatment of AL amyloidosis [7]. In the present study, we aimed to assess the effect of the substitution of dexamethasone by methylprednisolone on hematological response rates, and rapidity of response for patients with AL amyloidosis treated with bortezomib and cyclophosphamide at our amyloid referral center.

Methods

All consecutive newly diagnosed symptomatic patients with AL amyloidosis, treated with cyclophosphamide, bortezomib and methylprednisolome (CyBorMe) or CyBorD at Tom Baker Cancer Center (TBCC) were identified from our institutional database. The primary objective of the study was to assess degree of hematological (HR), organ (OR) and rapidity of response after treatment with CyBorMe and to compare with a historic cohort of cases treated with CyBorD. Approval for the review of these records was obtained from the TBCC Institutional Review Board.

Treatment Regimens

Patients were planned to receive 6 cycles of therapy. CyBorMe was administered in a 28-day cycle by using oral cyclophosphamide (300 mg/m²) weekly for 4 weeks, bortezomib 1.3–1.5 mg/m² weekly for 4 weeks, and methylprednisolone 500 mg IV weekly for 4 weeks. Dose modifications were made as per physician discretion. CyBorD was administered as reported previously by our group [8, 9]. Most patients received subcutaneous administration of bortezomib (88% and 92% for CyBorD and CyBorMe, respectively).

Response Assessment

The diagnosis of AL amyloidosis and assessment of HR and OR was performed based on the consensus criteria published in 2005 and modified in 2012 [10, 11]. Time to first response was defined as the time from treatment to PR or better. Response assessment was made every 2–4 weeks on the first cycle and monthly thereafter. Best response instead was defined as the maximal hematological response censored with a given treatment. Minimal Residual Disease (MRD) negativity was assessed by multiparameter flow cytometry as reported by our group [9].

Statistics

Two-sided Fisher exact test was used to test for differences between categorical variables. A P value of < 0.05 was considered significant. All statistical analyses were performed by using the SPSS 24.0 software.

Results

Fourteen-patients with AL amyloidosis treated with CyBorMe were identified. A cohort of 43 cases treated with CyBorD was used for comparison. Four patients treated with CyBorD died before 1 cycle due to cardiac complications and were not included for response analysis. Clinical characteristics are shown in Table 1. After a median of 3 cycles (1–6), HR was seen in 13 of the CyBorMe cases (92.8%) including CR in 5 (35.7%) and VGPR in 5 (35.7%). MRD negativity was reported in 10.2% of cases treated with CyBorD and 28.5% of the CyBorMe patients (p = 0.2) (Table 2). MRD negativity was achieved at a median of 3 months for CyBorMe and 4 months for CyBorD (p = 0.3) Median time to first and best response was shorter in the CyBorMe group (4 versus 6 weeks, p = 0.002 and 4 versus 12 weeks, p = 0.007). In addition, median dFLC at 1 month was 31 and 25 for the CyBorD and CyBorMe groups, respectively (p = 0.3). All organ response (AOR) was documented in 34% and 35.7% of patients treated with CyBorD and CyBorMe, respectively (p = 0.9). Further, cardiac response was observed in 4/9 (44%) evaluable patients treated with CyBorMe and 9/29 (31%) of those who received CyBorD (p = 0.4). Median time of cardiac response in the CyBorMe group was 8 weeks (range 4–24) as compared to 24 weeks in the CyBorD group (8–24 weeks, p = 0.001).

Table 1.

Clinical characteristics of patients with AL amyloidosis treated with CyBorMe and CyBorD at our institution

Characteristic	N = 43 (CyBorD)	N = 14 (CyBorMe)	P value
Age (median)	64	69	0.1
Gender
Male	24 (56%)	9 (64%)	0.5
Female	19 (44%)	5 (36%)
Hb (g/L)	126	116	0.2
Creatinine (µmol/L)	92	137	0.3
B2microglobulin (µmol/L)	2.81	3.0	0.3
Albumin (g/L)	29	29	0.6
Stage I	6 (13.9%)	4 (28%)	0.5
Stage II	10 (23.2%)	2 (14%)
Stage III	10 (23.2%)	2 (14%)
Stage IV	15 (34.8%)	6 (42%)
Unknown	2 (4.6%)	0
LDH (IU/L)	208	242	0.3
BMPC (%)	7.5%	11%	0.2
NTproBNP ng/L	1810	1539	0.4
Light chain:
Kappa	6 (14%)	6(42%)	0.06
Lambda	36(86%)	8 (57%)
Organ involvement
Cardiac involvement	72%	71%	0.9
Kidney involvement	72%	78%	0.6
Liver involvement	9.3%	21%	0.2
Nerve involvement	25.5%	14%	0.3
GI involvement	25%	7%	0.1
Lung involvement	6.9%	0%	0.3

BMPC Bone marrow plasma cells; FLC Free-light chains only

Table 2.

Hematological and organ response for patients with AL amyloidosis treated with CyBorD and CyBorMe at the Tom Baker cancer center

Characteristic	CyBorD group	CyBorMe group	P value
Overall response rate (ORR)	90.6%	92.8%	0.9
Complete response (CR)	28.5%	35.7%
Very good partial response (VGPR)	33.3%	35.7%
Partial response (PR)	30.9%	21.4%
**MRD neg CR by flow cytometry	10.2%	28.5%	0.2
Time to first response (weeks, median)	6	4	0.002
dFLC at 1 month (median)	31	25	0.3
Time to best response (weeks)	12 weeks	4 weeks	0.007
Median number of cycles	4	3	0.8
Discontinuation	25%	7%	0.1
Organ response	34%	35.7%	0.9
Cardiac response	31% (9/29)	44% (4/9)	0.4

**Minimal Residual Disease (MRD) assessment by flow cytometry

At the time of analysis, 25 and 3 patients in the CyBorD and CyBorMe groups have died and/or progressed with a median follow up of 25 and 7 months, respectively. Early death (within 3 months) was seen in 6.9% of patients treated with CyBorD and 7% of those treated with CyBorMe mainly associated with cardiac complications (p = 0.9).

Toxicity

CyBorD and CyBorMe were generally well-tolerated. Methylprednisolone was reduced in 57% of patients compared to 69.7% for dexamethasone in the CyBorD group (p = 0.3). At presentation, 5 patients in the CyBorD and 1 in the CyBorMe group required some form of heart device insertion. Two patients (14.2%) in the CyBorMe group had worsening heart failure and required methylprednisolone adjustment as well as diuretic modifications as compared to 8 (18%) in the CyBorD group. The rate of infections was similar between CyBorMe and CyBorD patients (35.7% vs. 37.2%, p = 0.5). Patients with CyBorD had a 13.9% rate of thromboembolic events compared to 7.1% in CyBorMe (p = 0.3). The rate of hospitalizations associated to infections was similar in both groups (13.9 vs. 14%, p = 0.5).

Discussion

AL amyloidosis is a clonal plasma cell disorder characterized by the production of misfolded immunoglobulin light chain that deposits in tissues of various organs [12]. A recent meta-analysis demonstrated that bortezomib treatment in AL amyloidosis improves ORR, CR, cardiac response rate and survival rate compared to controls without bortezomib therapy [13]. A very important factor to consider is the need for rapid response and minimal toxicity for patients with the so-called high-risk AL amyloidosis (NTproBNP > 8500). The median survival is < 6 months when AL patients with heart failure are untreated for the underlying plasma cell dyscrasia [14]. Initially, the European collaborative study reported lower haematological response rates in patients with stage IIIb disease [15]. Further, Manwani et al., reported on the importance of rapid responses in patients with stage IIIb [16]. Based on these reports, recently Mahmood et al., explored the possible benefit of the substitution of dexamethasone by methylprednisolone aiming to decrease toxicity in a very preliminary study [7]. Methylprednisolone is an intermediate-acting steroid with a half-life of 18–36 h compared to 36–54 h for dexamethasone [17]. These methylprednisolone properties could potentially induce rapid responses with more limited toxicity. Our study suggests that CyBorMe is a safe and feasible treatment strategy for AL patients. Response rates appear similar between the two groups. However, patients on the CyBorMe group had a faster time to first and best response compared to CyBorD. In addition, a trend towards lower dFLC after one month and higher cardiac response rate was noted. The rate of steroid reduction in the CyBorD group could potentially limit rapidity and depth of response. Our study has some inherent limitations, given the small patient sample size and limited follow-up, it is not possible to draw strong conclusions. However, it clearly paves the path for further assessment of this strategy on patients with advanced stage disease requiring fast responses with the emphasis on minimal toxicity.