2019 Volume 83 Issue 4 Pages 775-782
Background: Cardiac involvement occurs in more than half of the patients with light-chain amyloidosis (AL), but the characteristics, treatment and prognosis of cardiac AL (CAL) are not fully described.
Methods and Results: A total of 227 patients with CAL diagnosis between January 2009 and March 2017 at Peking Union Medical College Hospital were included. Patients with Mayo stages I, II and III AL accounted for 0.9%, 49.8% and 49.3%, respectively. Autologous stem cell transplantation, bortezomib combinations, non-bortezomib regimens and palliative treatment were given as first line therapy in 3.1%, 44.1%, 30.8% and 22.0% of patients, respectively. Overall hematological response and cardiac response were achieved in 60.6% and 37.2% of evaluable patients, respectively. The median overall survival (OS) was 17 months in all patients, and 10 months in those with Mayo stage III. In patients with Mayo stage III disease who survived for >1 month, the bortezomib group survived significantly longer than the non-bortezomib group (median OS, not reached vs. 12 months, P=0.019). Three independent prognostic factors for survival were identified: N-terminal fragment of B-type natriuretic peptide (NT-proBNP) ≥5,000 pg/mL, bone marrow plasma cells ≥10%, and systolic blood pressure <100 mmHg.
Conclusions: CAL patients had poor prognosis, but those treated with bortezomib combinations had a better outcome than the non-bortezomib group.
Light-chain amyloidosis (AL) is a rare systemic disorder caused by the deposition of misfolded amyloid fibrils derived from the monoclonal immunoglobulin light-chain in organs.1 Cardiac involvement occurs in more than half of patients at diagnosis of AL and is a major determinant of the treatment options and prognosis.2,3 Cardiac AL (CAL) patients have a poor prognosis and the median survival without treatment is approximately 6 months.4 In recent years, overall survival (OS) has been improved by high-dose melphalan followed by autologous stem cell transplantation (HDM/ASCT) for selected patients and/or combination chemotherapy with novel antiplasma cell agents such as bortezomib;5–7 however, the median survival of patients with Mayo stage III CAL remains as short as 7 months.8 Therefore, early recognition and appropriate treatment are necessary to improve long-term survival. In this study, we describe the clinical features, treatment, and outcomes of a large cohort of Chinese CAL patients, and explore the risk factors associated with survival.
Medical records were reviewed to identify patients who had a discharge diagnosis of AL between January 2009 and March 2017 at Peking Union Medical College Hospital. The definition of AL and assessment of organ involvement were based on the consensus criteria published in 20059 and modified in 2012.10 AL was biopsy-proven by Congo red staining, and the deposits were characterized as AL type by immunohistochemistry, immunofluorescence, or laser microdissection with mass-spectrometry-based proteomic analysis.11 Myeloma was carefully ruled out by special organ damage (anemia, hypercalcemia, renal failure, and bone lesion) caused by myeloma, especially in patients with ≥10% plasma cells in the bone marrow. Cardiac involvement was defined as mean left ventricular wall thickness on echocardiography >12 mm in the absence of hypertension or other potential causes of left ventricular hypertrophy, or N-terminal fragment of B-type natriuretic peptide (NT-proBNP) >332 ng/L in the absence of renal failure or atrial fibrillation. Only patients with CAL were included in this study and compared with AL patients without cardiac involvement.
Baseline Evaluation and Follow-upPatients’ work-up included clinical evaluation and a complete laboratory assessment (full blood count, basic biochemistry, protein analysis with serum and urine immunofixation and electrophoresis, and serum immunoglobulin free light-chains). Bone marrow aspirate and biopsy data were also obtained.
Cardiac evaluation at baseline and during follow-up included a complete physical examination, 12-lead ECG, echocardiography, magnetic resonance imaging (MRI) and assessment of cardiac biomarkers. NT-proBNP and cardiac troponin I (cTnI) levels were measured with standard commercially available assays. Cardiac function was evaluated using the New York Heart Association (NYHA) classification.
Disease StageThe Mayo stage criteria8,12–14 were used for stratification. Patients were classified as having stage I, II or III disease based on whether they had none, one or both of the following findings: NT-proBNP ≥332 ng/L and cTnI ≥0.1 ng/mL.12 Furthermore, stage III patients were divided into stage IIIa or stage IIIb based on whether NT-proBNP was below or above 8,500 ng/L.8
Treatment and ResponseRegimens used for treatment included HDM/ASCT, bortezomib-containing regimens, regimens without bortezomib (including melphalan-dexamethasone, alkylating agent-thalidomide-dexamethasone combination, and lenalidomide combinations), and palliative treatment. Patients who met the criteria were candidates for HDM/ASCT:15 physiologic age ≤70 years; blood pressure ≥90 mmHg; estimated glomerular filtration rate (eGFR) ≥30 mL/min/1.73 m2; ECOG performance status ≤2; Mayo stage I or II; NYHA functional status Class I or II; ejection fraction >55%; no severe pleural effusions. Therapeutic decisions were made based on the physican’s advice and the patient’s wishes.
Hematological and organ responses were assessed according to the Consensus Opinion from the 10th International Symposium on Amyloid and Amyloidosis.9 A hematological very good partial response (VGPR) was defined as the difference between involved and uninvolved serum free light-chain concentration (dFLC) <40 mg/L.10
Survival was calculated from the date of diagnosis until the date of last follow-up, or death. Last follow-up time was October 2017. Early death was defined as death within 3 months after diagnosis. The primary outcome measure was OS and the effect of a hematological response to treatment on survival.
Ethical StatementThis study was approved by the Clinical Research Ethics Committee of Peking Union Medical Center Hospital, Beijing, China. Because of the study’s retrospective nature, written informed consent by patients was waived by the Committee.
Statistical AnalysisStatistical analysis was undertaken using SPSS version 22.0 software (IBM, Armonk, NY, USA). Categorical data are expressed as percentages and were compared using the chi-square or Fisher’s exact test. Mean and standard deviation are presented for normally distributed variables and median and range for non-normally distributed ones. All P values were two-sided with a significance level of 0.05. Receiver-operating characteristic (ROC) analysis with death at 1 year was used to identify the threshold for NT-proBNP, cTnI, left ventricular ejection fraction (LVEF), systolic blood pressure (SBP), and dFLC, which were then analyzed as dichotomous variables. Survival was assessed by the Kaplan-Meier method and compared by log-rank test. Cox multivariable regression models were fitted, including non-collinear predictive variables. Patients who died prior to response assessment were classified as non-responders.
During the study period, 332 patients were newly diagnosed with AL; of them, 227 (68.4%) had cardiac AL and 105 did not. The demographic and clinical characteristics of both groups are compared in Table 1. CAL patients had a median age of 57.0 years, and the ratio of male to female was 2.0:1. Compared with non-CAL patients, CAL patients had a higher median serum dFLC concentration, and more patients had bone marrow plasma cells ≥10%. CAL patients had more gastrointestinal and peripheral neuropathic involvement but less renal involvement. More than one-third of CAL patients had ≥3 organs involved, which was more than in the non-CAL patients. When biomarkers were evaluated, CAL patients had a lower median level of 24-h urine protein, but a higher median level of serum albumin.
| Characteristic | CAL (n=227) |
Non-CAL (n=105) |
P value |
|---|---|---|---|
| Age, years | 57.0 (37–81) | 58.0 (20–84) | 0.851 |
| Male sex | 151 (66.5%) | 64 (61.0%) | 0.323 |
| Symptoms to diagnosis, months | 12.0 (1–104) | 8.0 (1–432) | 0.077 |
| M protein | |||
| κ-type | 59 (26.0%) | 38 (36.2%) | 0.057 |
| dFLC, mg/L (n=182) | 200.6 (1.6–4,263.0) | 51.0 (0.6–1,575.3) | <0.001 |
| Bone marrow ≥10% plasma cells | 33/198 (16.7%) | 6/94 (6.4%) | 0.016 |
| Organ involvement | |||
| Renal involvement | 153 (67.4%) | 83 (79.0%) | 0.030 |
| PN | 52 (22.9%) | 7 (6.7%) | <0.001 |
| Liver | 49 (21.6%) | 25 (23.8%) | 0.651 |
| Gastrointestinal | 29 (12.8%) | 5 (4.8%) | 0.025 |
| Involved organs ≥3 | 85 (37.4%) | 6 (5.7%) | <0.001 |
| Biomarkers | |||
| 24-h urine protein, g | 1.5 (0.0–25.0) | 4.6 (0.06–34.0) | <0.001 |
| eGFR, mL/min/1.73 m2 | 83.0 (6.0–136.0) | 88.1 (9.3–177.0) | 0.304 |
| Serum albumin, g/L | 33.0 (13.0–53.0) | 30.0 (11.0–50.0) | 0.005 |
| ALP, g/L | 89.0 (30.0–919.0) | 76.0 (25.0–1,546.0) | 0.264 |
AL, light-chain amyloidosis; ALP, alkaline phosphatase; CAL, AL with cardiac involvement; dFLC, difference between involved and uninvolved serum free light-chain concentration; eGFR, estimated glomerular filtration rate; PN, peripheral neuropathy.
Cardiac parameters in CAL patients are listed in Table 2. About half of them were in NYHA Class III or IV and had SBP <100 mmHg. The median value of the different cardiac biomarkers was 4,566 ng/L (range, 222–103,277) for NT-proBNP and 0.10 ng/mL (range, 0–16.5) for cTnI; 103 patients (47.2%) and 66 patients (30.3%) had NT-proBNP ≥5,000 ng/L and 8,500 ng/L, respectively. Among all the CAL patients, 49.8% had Mayo stage II disease, 49.3% had stage III disease, and only 2 patients had stage I disease. Of the patients with stage III disease, the number with stage IIIa disease was about 1.5-fold the number with stage IIIb disease. Echocardiography was performed in 217 patients (95.6%). The mean LV wall thickness was 14.2±3.8 mm, and the median EF was 57% (range, 21–88%). Cardiac MRI was done in 74 patients (32.6%) and late gadolinium enhancement was positive in 69 patients (93.2%). Endomyocardial biopsy was performed in 59 patients (26.0%); Congo red stain was positive in 98.3% of endomyocardial biopsy specimens, and the only patient with negative Congo red stain was proved to have CAL by laser microdissection with mass spectrometry-based proteomic analysis. Therefore, the sensitivity of endomyocardial biopsy was as high as 100%.
| Parameter | Median (range) or no. of patients (%) | P value | |
|---|---|---|---|
| Cardiac AL (n=227) | Non-CAL (n=105) | ||
| NYHA grade III–IV | 107 (48.9%) | 0 | <0.001 |
| SBP, mmHg | 100 (55–160) | 118 (78–170) | <0.001 |
| <100 | 94 (42.7%) | 19 (18%) | <0.001 |
| cTnI, ng/mL | 0.10 (0–16.5) | 0.03 (0–0.07) | <0.001 |
| ≥0.1 | 107 (50.7%) | 0 | <0.001 |
| NT-proBNP, ng/L | 4,566 (222–103,277) | 151 (7–890) | <0.001 |
| ≥5,000 | 103 (47.2%) | 0 | <0.001 |
| ≥8,500 | 66 (30.3%) | 0 | <0.001 |
| BNP, ng/L | 691 (53–5,000) | 42 (5–116) | <0.001 |
| Pleural effusion | 87 (37.4%) | 2 (1.9%) | <0.001 |
| Mayo stage | n=217 | n=100 | <0.001 |
| I | 2 (0.9%) | 90 (90%) | |
| II | 108 (49.8%) | 10 (10%) | |
| III | 107 (49.3%) | 0 | |
| IIIb | 44 (20.3%) | 0 | |
| Echocardiography | n=217 | n=95 | |
| IVS, mm | 14.2±3.8 | 9±1.5 | <0.001 |
| LVPW, mm | 12.8±2.8 | 8±1.3 | <0.001 |
| LVEF, % | 57 (21–88) | 66.5 (52–84) | <0.001 |
| LVEF <48% | 63 (29%) | 0 | <0.001 |
| E/A* | 1.6±0.8 | 0.8±0.2 | 0.07 |
| E/E’* | 17 (8–42) | NA | |
| CMR | n=74 | NA | |
| LGE | 69 (93.2%) | ||
| LVEDV index | 63.7 (32.9–90.3) | ||
| RVEDV index | 63 (21–107) | ||
| LVEF, % | 56 (30–80) | ||
| RVEF, % | 52.5 (25.5–74.2) | ||
| Endomyocardial biopsy | n=59 | ||
| Congo red stain | 58 (98.3%) | NA | |
AL, light-chain amyloidosis; CAL, AL with cardiac involvement; BNP, B-type natriuretic peptide; cTnI, cardiac troponin I; E/A, ratio of E and A; E/E’, ratio of E and E’; IVS, interventricular septal thickness; LGE, late gadolinium enhancement; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVPW, left ventricular posterior wall thickness; CMR, cardiac magnetic resonance imagining; NA, not acquired; NT-proBNP, N-terminal fragment of B-type natriuretic peptide; NYHA, New York Heart Association; RVEDV, right ventricular end-diastolic volume; RVEF, right ventricular ejection fraction; SBP, systolic blood pressure.
First-Line Treatment A total of 50 patients (22.0%) received palliative treatment. Of the remaining 177 patients, 7 patients (3.1%) received ASCT, 100 (44.1%) received bortezomib combinations, and 70 (30.8%) received non-bortezomib regimens (melphalan-dexamethasone, n=43; thalidomide-dexamethasone/thalidomide-cyclophosphamide-dexamethasone, n=17; and lenalidomide combinations, n=10).
Hematological Response Of the 177 patients who received first-line treatment, 40 were excluded for the following reasons: 16 stopped after 1 or 2 cycles of chemotherapy because of toxicity or economic problems; 24 did not attend the first follow-up visit within 3 months after diagnosis. Therefore, 137 patients were evaluated for response. The hematological responses are listed in Table 3. An overall hematological response was achieved in 6 ASCT patients (85.7%). The bortezomib group had a higher overall hematological response than the non-bortezomib group (66.2% vs. 46.8%, P=0.030). The median time from treatment to hematological response was 4, 1, and 4 months in the ASCT, bortezomib and non-bortezomib groups, respectively.
| All evaluable patients (n=137) |
ASCT (n=7) |
Bortezomib group (n=83) |
Non-bortezomib regimen (n=47) |
|
|---|---|---|---|---|
| HemR | ||||
| CR | 49 (35.7%) | 5 (71.4%) | 30 (36.1%) | 14 (29.8%) |
| VGPR | 20 (14.6%) | 1 (14.3%) | 15 (18.1%) | 4 (8.5%) |
| PR | 14 (10.2%) | 0 | 10 (12.0%) | 4 (8.5%) |
| NR | 54 (39.4%) | 1 (14.3%) | 28 (33.7%) | 25 (53.2%) |
| ORR | 60.5% | 85.7% | 66.2% | 46.8% |
| CarR | 37.2% | 3 (42.9%) | 37 (44.6%) | 11 (23.4%) |
ASCT, autologous stem cell transplantation; CarR, cardiac response; CR, complete hematological response; HemR, hematological response; NR, no response; ORR, overall response rate; PR, partial response; VGPR, very good partial response.
Cardiac Response (CarR) In the 137 evaluable patients, organ response was achieved in 56 patients (40.9%) and CarR in 51 (37.2%) (Table 3). The median time from treatment to CarR was 5 months (1–21 months). CarR was more frequently associated with hematological complete response (36 of 49 patients, 73.5%) and VGPR (11 of 20 patients, 55%) than with PR (4 of 14 patients, 28.5%) (P=0.019). CarR was achieved in 44.6% of the 83 patients in the bortezomib group, which was statistically higher than the non-bortezomib group (23.4%, P=0.002).
All patients with CarR showed a decrease of ≥30% in NT-proBNP, which declined from a median value of 3,389 ng/L (range, 349–16,496) to 740 ng/L (range, 64–6,000), and a decreased rate of 71% (range, 32–97%) was noted. A total of 41 of these patients (80.4%) achieved an improvement of at least 1 grade in NYHA Class; 10 patients with a hematological response ≥VGPR underwent repeated cardiac MRI and of them, 5 (50%) had a reduction of ≥2 mm in septal thickness, 2 (20%) showed an improvement of ≥10% in LVEF and 5 (50%) showed an improvement of ≥10% in right ventricular EF. Of the 5 patients with a hematological response ≥VGPR underwent repeated echocardiography, 4 (80%) had a reduction of ≥2 mm in septal thickness and 2 (40%) showed an improvement of ≥10% in LVEF.
Male patients and those with kappa-type, gastrointestinal involvement, ≥3 involved organs, SBP <100 mmHg, non-bortezomib treatment, or advanced cardiac amyloid had worse hematological responses with significant difference (Supplementary Table).
SurvivalMedian follow-up time was 28 months (2–123 months) for all patients. At the time of last follow-up, 15 patients (6.6%) had been lost and 116 (51.1%) had died. The median OS was 17 months [95% confidence interval (CI) 10.7–23.3], and the 3, 6, 12, and 24-month OS rates were 72.6%, 66.3%, 53.0%, and 44.6%, respectively (Figure 1A). The median OS for patients with Mayo stage II, IIIa or IIIb disease was 22, 19, and 3 months, respectively (P<0.001) (Figure 1B).
(A) Overall survival (OS) for cardiac light-chain amyloidosis patients. (B) Survival for cardiac light-chain amyloidosis at different Mayo stages.
With regard to treatment options, all ASCT patients remained alive, and palliative patients had a median OS of only 3 months (95% CI 1.9–4.1). The median OS in patients treated with chemotherapy was 29 months (95% CI 17.5–40.5), which was longer than that in patients with palliative treatment (P<0.001). The median OS was not reached in patients treated with bortezomib combinations, compared with 18 months in the non-bortezomib group (P=0.205). In patients with Mayo stage III disease, the bortezomib group survived longer than the non-bortezomib group (median OS, 30 vs. 10 months, P=0.051) (Figure 2). In patients with Mayo stage III disease who survived for >1 month, the bortezomib group survived significantly longer than the non-bortezomib group (median OS, not reached vs. 12 months, P=0.019).
(A) Survival for cardiac light-chain amyloidosis patients treated with bortezomib combinations or non-bortezomib regimen. (B) Survival for cardiac light-chain amyloidosis at Mayo stage III in the bortezomib and non-bortezomib groups. (C) Survival by 1-month landmark analysis for cardiac light-chain amyloidosis at Mayo stage III in the bortezomib and non-bortezomib group.
Survival was significantly associated with hematological response (Figure 3A). Survival rates at 12 and 24 months for patients evaluable for a hematological response were: for patients with complete response, both 96%; for patients with VGPR or PR, 83.8% and 77.8%; for non-responders, 20.7% and 14.8% (P<0.001). The OS for hematological responders with CarR was longer than that for patients without CarR (P<0.001) (Figure 3B).
(A) Survival by hematological response: CR (complete hematological response); VGPR-PR (very good partial response and partial response); NR (no response). (B) Survival by cardiac response in patients with hematological response: COR (cardiac organ response); non-COR (non-cardiac organ response).
Risk factors associated with OS on univariate and multivariate analyses are given in Table 4. Using ROC analysis, the NT-proBNP cutoff identified for death at 1 year was 5,000 pg/mL [area under the curve (AUC), 0.730; P<0.001], the SBP cutoff was 100 mmHg (AUC, 0.677; P<0.001), the cTnI cutoff was 0.1 ng/mL (AUC, 0.634; P=0.001), the LVEF cutoff was 48% (AUC, 0.635; P=0.001) and the dFLC cutoff was 500 mg/L (AUC, 0.615; P=0.009). Univariate analysis revealed the following risk factors: male sex, symptom-to-diagnosis interval >6 months, dFLC ≥500 mg/L, bone marrow plasma cells ≥10%, ≥3 involved organs, peripheral neuropathy, liver or gastrointestinal involvement, NYHA class ≥3, SBP <100 mmHg, cTnI ≥0.1 ng/mL, NT-proBNP ≥5,000 pg/mL, and LVEF <48%. When multivariate analysis was performed, only 3 independent determinants were identified: NT-proBNP ≥5,000 pg/mL [hazard ratio (HR) 2.101, 95% CI 1.251–3.529], bone marrow plasma cells ≥10% (HR=1.884, 95% CI 1.045–3.394), and SBP <100 mmHg (HR=1.673, 95% CI 1.029–2.719). Patients with 0, 1, 2 or all 3 factors had a median OS of 78, 17, 4, and 1 month, respectively (P<0.001).
| Factor | Univariate analysis | Multivariate analysis (n=131) | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Age ≥65 years | 1.139 | 0.735–1.765 | 0.561 | – | – | – |
| Male | 1.752 | 1.149–2.672 | 0.009 | |||
| Interval ≥6 months | 2.123 | 1.268–3.555 | 0.004 | |||
| κ-type | 1.478 | 0.993–2.200 | 0.054 | |||
| dFLC ≥180 mg/L | 1.398 | 0.915–2.137 | 0.122 | |||
| dFLC ≥500 mg/L | 2.187 | 1.407–3.401 | 0.001 | |||
| Bone marrow ≥10% plasma cells | 1.640 | 1.023–2.629 | 0.040 | 1.884 | 1.045–3.394 | 0.035 |
| Involved organs ≥3 | 1.309 | 1.090–1.572 | 0.004 | |||
| PN involvement | 1.776 | 1.199–2.631 | 0.004 | |||
| Liver involvement | 1.552 | 1.017–2.370 | 0.042 | |||
| Gastrointestinal involvement | 1.659 | 1.033–2.666 | 0.036 | |||
| eGFR <30 mL/min/1.73 m2 | 1.738 | 0.932–3.239 | 0.082 | |||
| NYHA grade III or IV | 2.311 | 1.574–3.393 | 0.000 | |||
| SBP <100 mmHg | 1.816 | 1.253–2.632 | 0.002 | 1.673 | 1.029–2.719 | 0.038 |
| cTnI ≥0.1 ng/mL | 1.656 | 1.124–2.440 | 0.011 | |||
| NT-proBNP ≥5,000 pg/mL | 2.362 | 1.610–3.465 | 0.000 | 2.101 | 1.251–3.529 | 0.005 |
| Pleural effusion | 1.264 | 0.873–1.830 | 0.216 | – | – | – |
| IVS ≥14 mm | 1.007 | 0.680–1.493 | 0.971 | – | – | – |
| LVEF <48% | 1.839 | 1.227–2.754 | 0.003 | |||
CI, confidence interval; HR, hazard ratio. Other abbreviations as in Tables 1,2.
In this retrospective cohort study, we analyzed 227 consecutive patients with CAL in a single center in China. Our study had 2 novel findings. First, we emphasized that bortezomib combinations given as first-line treatment were superior to non-bortezomib regimens, especially in patients with Mayo stage III disease. Second, using Cox regression analysis, we disclosed some baseline risk factors for the outcome that differed from those reported in the literature.
Although not statistically significant, we showed that CAL patients had a longer duration from symptom onset to diagnosis. Furthermore, in the univariate analysis, we revealed that symptom-to-diagnosis interval >6 months was a risk factor associated with OS. Early detection allows for treatment that can halt cardiac damage, potentially reducing the risk for sudden death;16 therefore, timely examinations for early diagnosis and treatment are imperative. In patients highly suspected of CAL, if initial assessment of amyloid from less-invasive sites, such as the tongue, periumbilical fat, labial salivary gland, or bone marrow biopsy specimen is negative, cardiac biopsy should be considered, which has a sensitivity of 100% when obtaining a pathological diagnosis17 and conclusively identifies CAL. As well, Austin et al showed that late gadolinium enhancement by cardiac MRI had a sensitivity of 88% and specificity of 95% in patients with biopsy-proven amyloidosis.18 Our study also showed that late gadolinium enhancement by cardiac MRI had a high sensitivity of 93.2%. Therefore, CMR evaluation is recommended as the first-line non-invasive diagnostic tool for CAL.19 In our cohort, endomyocardial biopsy was performed in 26.0%, and CMR was done in 32.6% of all patients, which were high percentages, enhancing the credibility of the present study.
Although the OS for CAL has improved in the past 3 decades, early death in CAL remains an obstacle to improving outcomes, and half of the patients with Mayo stage IIIb disease die within 3 months, which was also demonstrated in our study. Bortezomib, the first therapeutic proteasome inhibitor, has limited cardiotoxicity, and was found in our study and reported in the literature to have a median time to first response of 1–2 months,20,21 which was shorter than in the ASCT and non-bortezomib groups. Both prospective and retrospective studies of AL patients treated with bortezomib showed high hematological response rates (68–71%),20,22 and the combination of bortezomib, dexamethasone and cyclophosphamide increases the hematological response rate to 94%.21,23 Concerning CAL, a combination of bortezomib, dexamethasone and an alkylating agent (BDex+AA) was proved to have a hematological response rate of 68–96%, and a CarR rate of 32–60%, which was higher than that in the non-bortezomib group.24,25 Furthermore, a study in 106 CAL patients with symptomatic heart failure showed that BDex+AA improved survival more than in the non-bortezomib group.26 In our cohort with more patients included, we also demonstrated that bortezomib combinations significantly improved hematological response and CarR compared with non-bortezomib regimens. We also showed that bortezomib combinations improved median OS significantly in CAL patients with Mayo stage III who survived for >1 month when the response of bortezomib could be observed. Therefore, bortezomib combinations are recommended for naïve patients with significant (Mayo stage III) CAL and those refractory to other therapies.27
Wechalekar et al explored predictors of OS in 346 CAL patients with Mayo stage III disease and demonstrated that NT-proBNP >8,500 ng/L and SBP <100 mmHg were the only factors that independently affected OS.8 Risk factors described in other studies include NYHA functional class, estimated glomerular filtration rate, response to chemotherapy, amyloid load,28 old age, elevation of cTnI, and LV systolic or diastolic dysfunction.29 We identified 3 independent risk factors for OS: NT-proBNP ≥5,000 pg/mL, SBP <100 mmHg, and bone marrow plasma cells ≥10%. The latter has already been proved to be a risk factor for survival in AL.30 Compared with Wechalekar’s model, which included only cardiac parameters, our model for predictors of OS also included an AL parameter.8 However, this model was exploratory and should be validated in larger studies.
Study LimitationsFirst, because of its retrospective nature, many patients were lost and may have been a source of bias. Second, our hospital has undertaken the FLC assay only since 2012, so serum FLC concentrations before 2012 were not available. Third, because of its complexity, we did not analyze the influence of second-line therapy on survival. Despite these limitations, we tried our best to obtain full information for 227 CAL patients, and provide some meaningful characteristics and outcomes.
In summary, using a series of CAL patients in whom a high proportion were diagnosed by cardiac biopsy and evaluated with CMR, we showed that CAL patients had a poor prognosis, especially those with Mayo stage III disease. Patients treated with bortezomib combinations had a better hematological response rate and CarR than the non-bortezomib group. Furthermore, bortezomib combinations improved median OS significantly in CAL patients with Mayo stage III who survived for >1 month. NT-proBNP ≥5,000 pg/mL, bone marrow plasma cells ≥10%, and SBP <100 mmHg were independent risk factors for OS.
This work was supported by the National Key Research and Development Program of China grants no. 2016YFC0901500 and no. 2016YFC0901502.
None.
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-18-1048
