Page 334 - 2014 Printable Abstract Book
P. 334
discrimination of medically relevant radiation dose levels in human and NHP samples. We have further
developed the gene predictor into a Chemical Ligation-dependent Probe Amplification Radiation Exposure
Test (CLPA-RET) capable of rapid analysis of stabilized blood samples without requiring isolation of the
RNA. The performance of CLPA-RET is being evaluated in pre-clinical and blinded studies involving
irradiated NHPs (up to 7 days post-exposure), human TBI patients, and more than 1000 non-irradiated
human subjects with potentially confounding diseases and mixed ethnicities. Results of these studies will
be presented.
(PS6-17) Application of Multivariate Modeling for Radiation Dose Assessment: a Pilot Study. David L.
Bolduc; Vilmar Villa; David J. Sandgren; G. David Ledney; William F. Blakely#; and Rolf Rolf Bünger#
#
AFRRI, Bethesda, MD. Equal contribution as the senior authors. Scientific Research Department, Armed
Forces Radiobiology Research, Institute (AFRRI), Uniformed Services University of the Health Sciences,
Bethesda, MD 20889-5603, USA.
Multivariate Radiation Dose Estimation (RDE) algorithms were formulated for estimating a 6.5 Gy
radiation dose based on changes in blood cell counts and serum biochemical parameters. An archived
blood dataset was examined from a radiation study involving 24 nonhuman primates (NHP) (Macaca
-1
60
mulatta) that received a 6.5-Gy Co γ-ray (0.4 Gy min ) total-body-irradiation (TBI) dose. Blood
biosampling was performed prior to irradiation (d 0) and on d 7, 10, 14, 21, and 25 post-irradiation. We
demonstrate how changes in complete blood counts (CBC) and serum chemistries were identified for
multivariate modeling. A correlation matrix aided with computational statistical software was then
formulated with the radiation dose (0 vs. 6.5 Gy) designated as the “dependent variable.” Independent
variables were identified based on their radio-responsiveness and relatively low multicollinearity using
stepwise-linear regression analyses. Final candidate independent variables included CBC counts (absolute
number of neutrophils, lymphocytes, and platelets) in formulating the “CBC” diagnostic Radiation Dose
Estimation (RDE) algorithm. Additionally, the formulation of a diagnostic CBC and serum chemistry “CBC-
SCHEM” RDE algorithm expanded upon the CBC algorithm with the addition of hema-tocrit and the serum
enzyme concentrations of aspartate aminotransferase, creatine kinase, and
lactate dehydrogenase. Both algorithms estimated a 6.5 Gy total body irradiation dose spanning 7 to 25
days post-irradiation with over 90 % predictive power (CBC: 91 % ±1.01, P = 0.00001, n = 92; CBC-SCHEM:
93 % ±0.88, P = 0.00001, n = 92). Only the CBC-SCHEM algorithm however, met the three assumptions of
linear-least-squares and demonstrated greater accuracy for radiation dose estimation.
Financial support for this Pilot study came from an AFRRI intramural grant (RAB32165) the views
expressed do not necessarily represent AFRRI, USUHS, or the DoD.
332 | P a g e
developed the gene predictor into a Chemical Ligation-dependent Probe Amplification Radiation Exposure
Test (CLPA-RET) capable of rapid analysis of stabilized blood samples without requiring isolation of the
RNA. The performance of CLPA-RET is being evaluated in pre-clinical and blinded studies involving
irradiated NHPs (up to 7 days post-exposure), human TBI patients, and more than 1000 non-irradiated
human subjects with potentially confounding diseases and mixed ethnicities. Results of these studies will
be presented.
(PS6-17) Application of Multivariate Modeling for Radiation Dose Assessment: a Pilot Study. David L.
Bolduc; Vilmar Villa; David J. Sandgren; G. David Ledney; William F. Blakely#; and Rolf Rolf Bünger#
#
AFRRI, Bethesda, MD. Equal contribution as the senior authors. Scientific Research Department, Armed
Forces Radiobiology Research, Institute (AFRRI), Uniformed Services University of the Health Sciences,
Bethesda, MD 20889-5603, USA.
Multivariate Radiation Dose Estimation (RDE) algorithms were formulated for estimating a 6.5 Gy
radiation dose based on changes in blood cell counts and serum biochemical parameters. An archived
blood dataset was examined from a radiation study involving 24 nonhuman primates (NHP) (Macaca
-1
60
mulatta) that received a 6.5-Gy Co γ-ray (0.4 Gy min ) total-body-irradiation (TBI) dose. Blood
biosampling was performed prior to irradiation (d 0) and on d 7, 10, 14, 21, and 25 post-irradiation. We
demonstrate how changes in complete blood counts (CBC) and serum chemistries were identified for
multivariate modeling. A correlation matrix aided with computational statistical software was then
formulated with the radiation dose (0 vs. 6.5 Gy) designated as the “dependent variable.” Independent
variables were identified based on their radio-responsiveness and relatively low multicollinearity using
stepwise-linear regression analyses. Final candidate independent variables included CBC counts (absolute
number of neutrophils, lymphocytes, and platelets) in formulating the “CBC” diagnostic Radiation Dose
Estimation (RDE) algorithm. Additionally, the formulation of a diagnostic CBC and serum chemistry “CBC-
SCHEM” RDE algorithm expanded upon the CBC algorithm with the addition of hema-tocrit and the serum
enzyme concentrations of aspartate aminotransferase, creatine kinase, and
lactate dehydrogenase. Both algorithms estimated a 6.5 Gy total body irradiation dose spanning 7 to 25
days post-irradiation with over 90 % predictive power (CBC: 91 % ±1.01, P = 0.00001, n = 92; CBC-SCHEM:
93 % ±0.88, P = 0.00001, n = 92). Only the CBC-SCHEM algorithm however, met the three assumptions of
linear-least-squares and demonstrated greater accuracy for radiation dose estimation.
Financial support for this Pilot study came from an AFRRI intramural grant (RAB32165) the views
expressed do not necessarily represent AFRRI, USUHS, or the DoD.
332 | P a g e
