Transient transcriptional response to low-dose ionizing radiation in humans undergoing IMRT

Abstract

20018 Background: As intensity modulated radiation therapy techniques are increasingly utilized to treat cancer, the area of normal tissue exposed to ionizing radiation is increasing. The biologic risks associated with this normal tissue low dose exposure (LDIR) are fundamentally unknown and of concern to cancer survivors following therapy. Current modeling for health regulations presupposes a linear, no-threshold model of radiation effects, which estimates the effect and risk at low dose by extrapolation from measured effects at high doses. Cell culture models of ionizing radiation (RT) exposure show variable effects, not consistent with a linear dose-response relationship. We therefore undertook the first study to our knowledge of transcriptional effects of LDIR over time in vivo in solid tissue in humans. Methods: Tissue was collected at pre-RT, 3, 8, and 24 hours post-IR at sites receiving 10cGy. Transcriptional response at 3 and 8 hours were compared to the 0 and 24 hour time points. If transcripts are up regulated or down regulated at 3 and 8 hours compared with 0 and 24 hours, we have detected a transient response. The method of Rocke (2005), which was designed to detect differentially expressed gene groups using the responses of multiple probe sets corresponding to gene groups, was used to allow us to test whether there is differential expression for each patient separately, as well as for all the patients together. Results: Significant (p < 0.05) transient up regulation was shown in zinc finger proteins, keratins, BMP receptors, BAG, cyclins and BCL 6. Down regulation was detected in TNF, protein disulfide isomerase, interleukins, heat shock proteins, and S100. Nine gene groups did not show significant change; however, the number of significant gene groups (11) far exceeds the number expected by chance (2). In most cases in which a gene group was shown to be transiently altered, the tests of individual patients showed that most or all of the individuals also had differential expression of the same type. Conclusions: We have shown that it is possible to detect transient responses to LDIR in vivo in humans, and have identified eleven gene groups that demonstrate transient changes, as measured by a statistically principled analysis method. No significant financial relationships to disclose.