Abstract 1782

Poster Board I-808

The heart is involved in approximately two thirds of patients with AL amyloidosis and the prognosis of the disease is largely dependent on the severity of cardiac damage. Heart dysfunction can be assessed by measuring the serum concentrations of N terminal natriuretic peptide type B (NT-proBNP) and of cardiac troponins (cTn). The combination of these biomarkers allows an accurate prognostic stratification of patients with AL amyloidosis. Recently, new generation of more sensitive assays for cTn, characterized by low limits of detection and low imprecision, have been developed, aiming at identifying minimal cardiac damage. We report the impact of a highly sensitive (hs) cTnT assay on detection of heart involvement, prognosis and response to therapy in 109 consecutive newly diagnosed patients with AL amyloidosis.

The hs-cTnT was measured on the Modular E instrument with a precommercial immunoassay from Roche Diagnostics on frozen sera collected at the time of diagnosis and stored at -80°C. The 99th centile of hs-cTnT concentration in sera from 546 healthy volunteers is 14 ng/L (95%CI 12.4-24.9 ng/L). NT-proBNP and cTnI were measured with commercially available assays. The upper reference limit of NT-proBNP is 332 ng/L and for cTnI is 40 ng/L. Heart involvement was defined according to the International Society of Amyloidosis consensus criteria as a mean left ventricular wall thickness (mLVW) >12 mm in the absence of other causes.

Sixty-nine of the 109 patients (63%) fulfilled the echocardiographic criteria of heart involvement. Among these patients, 74% had elevated cTnI, 96% had high hs-cTnT and 100% had elevated NT-proBNP. Fifteen patients who did not have echocardiographic heart involvement at presentation reached a mLVW >12 mm within 6 months from diagnosis. Among them, cTnI was elevated at presentation in 33% of cases, hs-cTnT in 80% (p=0.01 compared to cTnI) and NT-proBNP in 73%, indicating that hs-cTnT and NT-proBNP can detect amyloid cardiac involvement when it is still unapparent by conventional echocardiographic criteria. Thirty-seven (34%) patients died and the median follow-up of living patients was 38 months (range 14-67 months). Survival was reduced in patients with hs-cTnT >14 ng/L (58% vs. 84% surviving at 3 years, p=0.01) and NT-proBNP >332 ng/L (58% vs. 86% surviving at 3 years, p=0.02) and cTnI >40 ng/L (52% vs. 75% surviving at 3 years, p=0.007). At multivariate analysis ln(hs-cTnT) was the only variable significantly associated with survival (HR 1.576, 95%CI 1.048-2.371, p=0.03). All the patients were treated with melphalan-dexamethasone and 66 (60%) reached hematologic response. The hs-cTnT cutoff best predicting survival after treatment was 68 ng/L (45% vs. 80% surviving at 3 years, p=0.0003). This cutoff, in combination with hematologic response to therapy, could differentiate the patient population into 3 groups with significantly different survival. Estimated survival at 3 years was 30% in non responders with hs-cTnT >68 ng/L, 51% in those who obtained either hematologic response or had hs-cTnT <68 ng/L (p=0.05) and 95% in responders with hs-cTnT <68 ng/L (p<0.0001).

Cardiac troponin measured with a high-sensitivity assay significantly improved the sensitivity, compared with commercially available troponin assay, for cardiac damage caused by AL amyloidosis and represents now the most powerful prognostic determinant. Patients who obtain hematologic response and a hs-cTnT concentration below the threshold of 68 ng/L enjoy prolonged survival.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution