Somalogic st2 and ntprobnp assays predict heart failure mortality as effectively as the elisa assay
Debbs J, Bryson TD, Zeld N, Aurora L, Gui H, Luzum JA, Peterson E, She R, Williams LK, and Lanfear DE. Somalogic st2 and ntprobnp assays predict heart failure mortality as effectively as the elisa assay. Journal of the American College of Cardiology 2020; 75(11):1091.
J Am Coll Cardiol
Background Biomarkers are critical for modern heart failure (HF) care. There are several established prognostic markers such as N-terminal pro-b-type natriuretic peptide (NTproBNP) and soluble suppressor of tumorgenicity 2 (ST2). Recent advances in multiplexing/multi-marker platforms offer faster and broader data generation. However, how these newer methods compare to FDA-approved ELISA-based assays remains unclear. The SOMALogic® SOMAscan assay is an aptamer-based technology that quantifies thousands of proteins simultaneously, including NTproBNP and ST2. The purpose of this study is to compare the test results and performance in predicting mortality using ELISA vs. SOMA for each marker. Methods Patients age ≥18 years and meeting Framingham definition for HF were enrolled in a prospective registry (Oct 2007 - March 2015) at Henry Ford Hospital. Only patients with an ejection fraction < 50 % were used for the analysis (N= 687 for ST2 and N= 902 for NTproBNP). We tested the correlation of SOMA vs ELISA for each marker and report spearman correlation coefficient. We then tested each marker in Cox models adjusted for clinical risk score (MAGGIC) and compared the HR and model improvement (using calculated area under the curve [AUC]) for ELISA vs SOMA versions. Results First, we calculated the correlation between SOMA and ELISA values for both ST2 and NTproBNP. The correlation for ST2 was 0.74 (p<0.001) and NTproBNP was 0.88 (p<0.001), respectively. Next, we used a Cox proportional hazards model to predict death for the four variables corrected for the clinical score MAGGIC. Both versions of both markers were significantly associated with survival time. The hazard ratio for ELISA-ST2 was 1.12 (95% CI 1.09-1.16, p < 0.001) and for SOMA ST2 was 1.11 (95% CI 1.08-1.15, p<0.001). The hazard ratios for ELISA NTproBNP and SOMA NTproBNP were identical (1.12 95% CI 1.09-1.15, p<0.001). Uno's area under the ROC curve analysis showed there was no difference between marker versions in mortality prediction for ST2 nor for NTproBNP. Conclusion These results indicate that the SOMAscan assay results for ST2 and NTproBNP are strongly correlated to the standard ELISA versions and have equivalent prognostic information.