Beyond Dose

While retrospective estimation of dose, dose quality, dose rate, and dose distribution is critical and is a central theme of the work of this center, it is by no means the end of the story. An important issue with regard to biodosimetry is that of inter-individual variability in radiation sensitivity.

It is clear that different individuals can respond differently to exactly the same dose exposure. Consequently, after a large-scale radiological event such as an IND, in which tens of thousands of individuals would be exposed to high doses, it would be highly advantageous also to be able to triage individuals based on their potential individual sensitivity to acute or delayed radiation effects, as well as based on their dose.

We previously investigated transcriptomic and metabolomic signatures associated with the individual outcome of acute radiation syndrome, and the impact of defects in the major DNA repair pathways on our biodosimeters. We now plan to take these studies further in two related areas:

• To investigate correlations between our high-throughput biomarkers and individual outcome of pneumonitis, the delayed radiation effect most likely to result in death of individuals who survived the acute phase of radiation syndrome.
• To probe the mechanisms associated with individual sensitivity by investigating the impact of altered immune / inflammatory responses on our biodosimetry markers.

Both of these approaches build on – and would not be possible without – the high-throughput biodosimetry work that we have accomplished to date. Our motivation is to investigate the feasibility of providing a high-throughput methodology to identify individuals who are particularly sensitive to acute radiation syndromes, within a large irradiated population.

Apart from known factors such as age, weight, and diabetic status, variations in individual acute sensitivity to radiation exposure can come from two basic sources:

1. genetic differences
2. stochastic differences.

We are investigating sensitivity using:

• Mouse models

Predicting Late Death From Radiation

In inbred mouse strains, the role of genetic differences as an important determinant of acute radiosensitivity has long been established.

The role of stochastic effects in inter-individual radiation sensitivity variations, while long established, has generally been less emphasized compared with genetic factors. What is meant by stochastic factors can be illustrated by the following example:

suppose we irradiate a series of genetically-identical same-aged mice, all to the same dose where, say, 25% of the animals die within a week. It is not the case that all these apparently identical mice will respond identically. Some will die within a few days, and some will survive for much longer periods.

What causes this variability? While even within inbred mice there may still be some subtle genetic and epigenetic differences, our hypothesis is that this variability of response, when the dose, genetics, and environment are essentially held fixed, is due to stochastic variations in the mouse damage response systems, which may already be present at the time of exposure, or may develop post exposure.

Thus our goal is to investigate whether our high-throughput biomarkers are reflective of these stochastic variations and to probe the mechanisms associated with such associations. By correlating breathing rate measurements, microCT imaging, and histology with our endpoints, we will also develop biomarkers of the progression and resolution of pneumonitis.

Impact of Inflammatory Disease

As 5-7% of the population have inflammatory diseases that could potentially confound our biodosimetry approaches, we will also examine the impact of inflammation on transcriptomic and metabolomic endpoints using a series of mutant mice with either attenuated or enhanced inflammatory signaling. We will compare the radiation responses in wild-type mice with those in IL10-/- and Gadd45a-/- mice, as models of active inflammatory disease. Similarly, we will also compare wild-type radiation responses to those in p38+/DN and p38Y323F mice, as models of attenuated inflammatory responses.

• Human Models

Any study of individual radiation sensitivity in humans will, of course, potentially be looking at variability caused both by genetics and by stochastics. There are many issues associated with a biomarker study of individual sensitivity to radiation effects in man. Not least is the difficulty in finding a suitable model. We use acute (early onset) erythema, following radiotherapy as the most appropriate and practical model to study interpersonal variations in sensitivity to early radiation injury.

Two complementary randomized studies are ongoing with the Department of Radiation Oncology at CUIMC. In both studies, patients receiving radiotherapy after breast-conserving surgery provide pre-exposure blood samples, which are irradiated to a standard dose and analyzed using our various endpoints. These studies are facilitated by the high throughput of the RABiT system, which allows rapid measurement of DNA damage endpoints after automated ex-vivo irradiation of a fingerstick blood sample. At the end of the study, the results will be compared to erythema, quantitated as a change in skin redness.