| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
The crux of the criticism of Law et al is that their own evaluations of risks of mammography screening are based on risk figures (from the NRPB, now HPA Radiation Protection Division) for radiation induced breast cancer, which were derived from epidemiology studies of cohorts exposed to medical X-rays. The inference in their argument is that the reported elevated values of relative biological effectiveness (RBE) for in vitro transformation are restricted to comparisons between mammography X-rays and high energy photons, or simulated A-bomb spectra as in our own work, and are therefore irrelevant to risk considerations in mammography. In vitro transformation studies using human cells in fact show elevated values of RBE for mammography X-rays compared with 200–220 kVp X-rays, Co-60 gamma rays and simulated A-bomb spectra.
For in vitro neoplastic transformation in human cells, the data of Frankenberg et al [1] can be used to derive limiting values (i.e. values at low doses) of RBE of 4.67±3.93 for mammography X-rays compared with 200 kVp X-rays. Gögglemann et al [2], using the same endpoint, provide data from which a limiting RBE can be derived of 3.58±1.77 for a mammography source compared with 220 kVp X-rays. The mean energy of the reference X-rays is quoted as being between 60 keV and 100 keV by Frankenberg et al [1] and 96 keV by Gögglemann et al [2], much lower than A-bomb spectra [3]. Despite these large differences in energy, no significant difference in the limiting RBE is observed between our data [4] and that of Frankenberg et al [1] and Gögglemann et al [2]. When pooled with our own results for this cell line and endpoint (limiting RBE of 4.42±2.02 for mammography X-rays compared with the atomic bomb spectrum simulated source) the combined data provide a best estimate for limiting RBE of 4.0±0.7 [4] for mammography X-rays compared with reference X-rays/photons with energies above 60 keV to 100 keV. These reference energies are comparable with the X-ray energies suggested by Law et al as being those used by themin the epidemiological studies of radiation-induced breast cancer. The reason why there is an appreciable change in the effectiveness of X-rays in the range below about 90 keV may be related to the preponderance of electrons with energies in the range 5–12 keV produced by mammography X-rays compared with higher energy X-rays (these electrons have ranges of 1–3.5 µm, comparable with chromosome dimensions) and to the generation of densely ionizing Auger electrons [5].
We accept that in vitro transformation data provide only indicative interim guidance on the RBE of mammography X-rays for breast cancer induction in women. In our paper [6] we used a range of values of RBE from 1 to 6 to encompass uncertainties in measured RBE values and to explore the sensitivity of the ratio of cancer detection to induction (DIR) on this parameter. The issue of whether/how risk estimates from in vitro measurements and epidemiological studies made at dose levels of a few tens of cGy can be extrapolated down to levels comparable with natural radiation background levels or single mammography exposures remains a source of contention (see for example the correspondence and response to Redpath and Mitchel in this issue), and indeed highlights the need for further research in this field.
Law et al describe the basis of the breast cancer risk estimates which they have used in the past in a range of publications. As they explain – there is in fact no definitive description of the origin of the particular values which they have used which is accessible to us. The derivation of radiation risks for breast cancer from pooled data is far from simple. Eight epidemiology studies which are available for breast cancer risk estimation have been reviewed in detail in a pooled analysis by Preston et al [7] and an overview has recently been provided by the National Academy of Sciences the Committee on the Biological Effects of Ionizing Radiation (BEIR) VII report [8]. Amongst these studies are the New York acute post-partum mastitis cohort (APM) and the Massachusetts tuberculosis fluoroscopy cohort (TB, two separate cohorts in fact) to which Law et al refer. As Law et al explain – the APM cohort was exposed to higher energy X-rays (up to 250 kVp) than the TB cohorts (up to 80 kVp). Preston et al [7] derive somewhat lower values of excess relative risk (per Gray) and higher values of excess absolute risk (per person year per Gray) for the APM cohort compared with the TB cohorts. The APM and TB cohorts appear to have similar statistical power, but we are not clear as to how the risks were combined to give the values used by Law et al. The risks derived from these cohorts will be at least in part influenced by exposures similar in photon energy to those used for the in vitro transformation studies we have described. The remaining five studies considered by Preston et al for evaluation of breast cancer risk are for orthovoltage therapeutic X-ray exposures or high energy gamma radiations [7]. The in vitro transformation data for mammography X-rays, 200–220 kVp X-rays and higher energy gamma radiations may thus be more pertinent than Law et al indicate to past and future evaluations of breast cancer risk in mammography.
We understand that breast cancer risks are currently being reviewed by the Advisory Group on Ionising Radiation (AGIR) of the Health Protection Agency (HPA) as part of a review of solid cancer risks (http://www.hpa.org.uk/radiation/advisory_groups/agir/index.htm). It is to be hoped that these complex considerations will be published in detail so that the bases of derived risk figures are patent. This is particularly important in the case of radiation induced breast cancer where factors such as the dose and dose rate effectiveness factor (DDREF), RBE, and the dependence of RBE on photon energy are so intimately related [9–11].
Yours etc.,
1 Medical Physics, University Hospital Birmingham NHS Foundation Trust Edgbaston, Birmingham, B15 2TH, 2 Radiation Biophysisics Group, School of Physics and Astronomy, The University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
References
Related articles in BJR:
This article has been cited by other articles:
 |
|
 |
|
  BJR review of the year - 2006 Br. J. Radiol., March 1, 2007; 80(951): 147 - 151. [Full Text] [PDF]
|
|
|
|
|
|
B M Moores Relative biological effectiveness and exposure of the female breast Br. J. Radiol., February 1, 2007; 80(950): 141 - 142. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
