PB2135 VALIDATION OF THE BOSTON UNIVERSITY STAGING SYSTEM IN AL AMYLOIDOSIS

Abstract

Background:The Mayo 2004 cardiac risk score (Blood 2004; 104(6): 1881) which incorporates NT‐proBNP and cardiac troponins (TnI/TnT) is central to the modern management of AL amyloidosis but presents a problem for institutions where NT‐BNP is not available. To address this issue the Boston University (BU) cardiac risk has been proposed which incorporates BNP instead of NT‐ProBNP (Blood. 2019 Jan 17;133(3):215).Aims:We aimed to provide external validation of the BU staging system on a cohort of patients with AL amyloidosis attending the Princess Alexandra Hospital Amyloidosis Centre.Methods44 patients with newly diagnosed AL amyloidosis had both BNP and NT‐proBNP performed on the same serum sample collected before the initiation of any therapy. Both TnI and BNP assays were performed on Beckman DxI800 immunoassay analyser (Beckman Coulter, Brea, CA, USA) using Quidel Triage BNP immunoassay and AccuTnI two‐site immunoenzymatic assay. NT‐proBNP was analysed on an Cobas E411 analyser (Roche Diagnostics, Basel, Switzerland) via an Elecsys proBNP II Cobas electrochemiluminescence immunoassay. We applied the published BNP cut‐off of 81 ng/L for the BU score, NT‐proBNP cut‐off of 332 ng/L for the Mayo 2004 score along with a TnI cut‐off of 0.1ug/mL for both staging systems. Patients were allocated to stage I if both were below the threshold, stage II if only one was above the threshold, and stage III if both biomarkers were above the threshold. Survival was measured from the time of blood sample collection until the date of death or last follow‐up and calculated by the Kaplan‐Meier method. The log‐rank test was used to assess the impact of cardiac biomarker score on survival. Concordance between Mayo 2004 staging and BU staging scores was determined using linear weighted kappa statistics.Results:Median age of the 44 patients was 67 years and 27% were female. Cardiac involvement was present in 61% and renal involvement in 54%. Initial treatment regimens varied across the cohort, including proteasome inhibitors (n = 15), immunomodulatory agents (n = 8), alkylator based regimens (n = 6), autologous stem cell transplantation (n = 7), no therapy (n = 7) and rituximab‐based therapy (n = 1). Three patients with advanced cardiac disease died before treatment commenced. Median duration of follow‐up of survivors was 7.3 years.Using the BU BNP‐based staging system we identified 12/44 (27%) of patients as stage I, 18/44 (41%) of patients as stage II and 14/44 (31%) of patients as stage III. This correlated closely with stratification via the Mayo score (Table 1), with only 1 patient miscategorised (97.7% agreement, k = 0.98). BU stage was highly predictive of outcome. Median overall survival for BU stage I was not reached, stage II was 40 months and stage III was 5 months (log rank p = 0.0012). Mayo 2004 median overall survival was identical to the BU system for stage I, II and III.Summary/Conclusion:We confirmed that the BU staging system, incorporating a BNP threshold of 81 ng/L obtained at initial consultation, stratified patients in concordance with the Mayo 2004 staging system which uses a NT‐proBNP threshold of 332 ng/L. Furthermore, we also demonstrate that the BU score displays similar prognostic information.image