DOE Low Dose Radiation Research Program Workshop I Abstracts November 10-12, 1999, Washington, D.C.

2. The Risk of Cancer Induction Due to Routine Mammographic Screening

David J. Brenner, Steve Marino, and Charles Geard
Center for Radiological Research, Columbia University, 630 West 168th Street, New York, NY 10032
djb3@columbia.edu

Summary: To obtain realistic and credible risk estimates for breast-cancer mortality due to clinical mammographic imaging examinations.

Abstract: The aim of this work is to obtain realistic and credible risk estimates for breast-cancer mortality due to clinical mammographic imaging examinations. Given the increasing emphasis on clinical mammographic screening for breast cancer, it is of societal importance to provide realistic risk estimates -- with realistic confidence bounds -- for breast cancer induction from routine mammographic X rays. Concern about this risk is a major factor (both from the perspective of the patient and the family physician) in the disturbingly large numbers of women who chose not to undergo routine mammographic screening.

Direct studies of mammographic screening programs have insufficient power to quantitate risks of mammographic screening, so estimates of the risks must be based on extrapolations from other cohorts (A-bomb survivors, medical irradiations). These other cohorts, however, were exposed to far higher-energy X or g rays than the very low-energy X rays used in mammograms; dose-for-dose, low-energy X rays are considerably more radiobiologically damaging than high-energy X or g rays. So it is likely that this risk estimation process under-estimates risks of screening mammograms -- preliminary estimates are by a factor of 1.5-2.

We are using mechanistically-based techniques to extrapolate from those situations where radiation-induced breast-cancer risk estimates are better known (A-bomb survivors, medical irradiations), to risks from screening mammograms.

1. Experimental microdosimetric measurements are being used to characterize the energy-deposition characteristics of low-energy (5-28 keV) monoenergetic x rays. These studies are taking place at the X23A2 beam line of the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory. Based on these data, the biological effectiveness of realistic mammographic beams can be estimated relative to those for which breast-cancer risks have been estimated (e.g. A-bomb survivors, TB patients).

2. Direct radiobiological experiments are also underway at the NSLS of the RBE of these low-energy x rays, for the endpoints of chromosomal aberration formation and in-vitro oncogenic transformation, in mammalian cells.

The "bottom line" of these studies will be realistic estimates of the risk from routine screening mammograms, together with credible confidence limits that may reasonably be associated with these risk estimates.