Use of Proteomic and Hematology Biomarkers for Prediction of Hematopoietic Acute Radiation Syndrome Severity in Baboon Radiation Models

Health Phys. 2018 Jul;115(1):29-36. doi: 10.1097/HP.0000000000000819.

Abstract

Use of plasma proteomic and hematological biomarkers represents a promising approach to provide useful diagnostic information for assessment of the severity of hematopoietic acute radiation syndrome. Eighteen baboons were evaluated in a radiation model that underwent total-body and partial-body irradiations at doses of Co gamma rays from 2.5 to 15 Gy at dose rates of 6.25 cGy min and 32 cGy min. Hematopoietic acute radiation syndrome severity levels determined by an analysis of blood count changes measured up to 60 d after irradiation were used to gauge overall hematopoietic acute radiation syndrome severity classifications. A panel of protein biomarkers was measured on plasma samples collected at 0 to 28 d after exposure using electrochemiluminescence-detection technology. The database was split into two distinct groups (i.e., "calibration," n = 11; "validation," n = 7). The calibration database was used in an initial stepwise regression multivariate model-fitting approach followed by down selection of biomarkers for identification of subpanels of hematopoietic acute radiation syndrome-responsive biomarkers for three time windows (i.e., 0-2 d, 2-7 d, 7-28 d). Model 1 (0-2 d) includes log C-reactive protein (p < 0.0001), log interleukin-13 (p < 0.0054), and procalcitonin (p < 0.0316) biomarkers; model 2 (2-7 d) includes log CD27 (p < 0.0001), log FMS-related tyrosine kinase 3 ligand (p < 0.0001), log serum amyloid A (p < 0.0007), and log interleukin-6 (p < 0.0002); and model 3 (7-28 d) includes log CD27 (p < 0.0012), log serum amyloid A (p < 0.0002), log erythropoietin (p < 0.0001), and log CD177 (p < 0.0001). The predicted risk of radiation injury categorization values, representing the hematopoietic acute radiation syndrome severity outcome for the three models, produced least squares multiple regression fit confidences of R = 0.73, 0.82, and 0.75, respectively. The resultant algorithms support the proof of concept that plasma proteomic biomarkers can supplement clinical signs and symptoms to assess hematopoietic acute radiation syndrome risk severity.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Acute Radiation Syndrome / blood*
  • Acute Radiation Syndrome / diagnosis*
  • Acute Radiation Syndrome / etiology
  • Algorithms
  • Animals
  • Biomarkers / blood*
  • Dose-Response Relationship, Radiation
  • Gamma Rays / adverse effects*
  • Hematology*
  • Male
  • Papio
  • Proteome / radiation effects*
  • Proteomics / methods
  • Severity of Illness Index*
  • Whole-Body Irradiation

Substances

  • Biomarkers
  • Proteome