Radiation

HPA response to Review of risks from tritium

HPA response statement on Review of risks from tritium by the independent Advisory Group on Ionising Radiation (AGIR)

The HPA welcomes the AGIR Review of risks from tritium1 as an authoritative review of tritium dose estimation and risks. The review considers the properties and sources of tritium, its relative biological effectiveness (RBE), epidemiological studies of tritium exposed populations, and the biokinetics of tritiated water (HTO) and forms of organically bound tritium (OBT). The report also deals with the specific issue of doses to female germ cells.

Sources of exposure to tritium by workers and members of the public include nuclear reactors, nuclear fuel reprocessing plants, nuclear weapon manufacture and the preparation of medical and research products. In addition, the possible future operation of fusion reactors is a potential source of exposures.

Important properties of tritium as a radioactive isotope of hydrogen are the low energy of beta particles (average 5.7 keV, maximum 18.6 keV) emitted during decay (half-life of 12.3 years) to helium, and its incorporation into organic molecules within the body (OBT). Low-energy beta particles are more effective per unit absorbed dose (Gy) in causing cancer-related cellular damage than are, for example, gamma rays from cobalt-60. OBT may be retained for long periods in the body and the possibility of heterogenous distribution within cells is of potential importance because of the short track length of tritium beta particles (average 0.56 µm, maximum 6 µm).

The AGIR report provides a thorough review and analysis of experimental data on the RBE of tritium beta emissions in comparison with x-rays and gamma rays. Particular attention was paid to in vivo studies relating to cancer induction but in vitro data and developmental and reproductive end-points were also considered. It is concluded that there is good experimental evidence, supported by theoretical considerations, for RBE values for tritium of 1–2 compared to x-rays and 2–3 compared to gamma rays. Given the dependence of RBE values on the reference radiation, it is recommended that a high energy gamma-ray source such as cobalt-60 is to be preferred as a reference; consideration should also be given to the effect of dose rate. It is noted that animal carcinogenesis studies have resulted in low estimates of RBE, close to 1 (range 1–2.5). However, each of these studies has limitations and they do not provide information on RBE at low doses. Commenting on the possible effects of transmutation to helium and accumulation of tritium in the hydration shell of DNA, the conclusion reached is that their contribution to observed RBE values is likely to be small.

The overall judgement of the AGIR subgroup concerning tritium RBE compared to cobalt-60 gamma rays is that the available experimental data and biophysical considerations support the use of a value of 2–3 for induction of stochastic effects in humans exposed to tritium at low doses and dose rates. A rounded RBE value of 2 is recommended for use in epidemiological studies and retrospective dose assessments. Further, the report suggests that the International Commission on Radiological Protection (ICRP) should consider adopting a specific value of 2 for the radiation weighting factor (wR) for tritium beta particles.

The AGIR report provides a comprehensive account of, and commentary on, relevant epidemiological studies. Studies relating to occupational and environmental exposures that include tritium have been undertaken in the UK, USA, Canada and Germany. None of these studies provide quantitative information on risks from tritium. The subgroup concluded that the best prospects for more informative analyses were from follow-up of nuclear worker studies in Canada, the Savanna River site in the USA, and UK nuclear workforces at Sellafield, Chapelcross, Capenhurst, and the AWE and the UKAEA. The AGIR subgroup recommends that consideration should be given to an international collaborative epidemiological study of populations exposed to tritium.

On the basis of a review of biokinetic data from studies on humans and experimental animals, the AGIR subgroup concludes that the ICRP models used for the calculation of dose coefficients for HTO and OBT make appropriate use of the available information. It is noted that while the current ICRP model for HTO takes account of incorporation into OBT as a single component of retention, a future model will include a further smaller but longer-term component of OBT retention. It is also noted that the dose coefficient given for OBT ingestion is a generic value for tritium incorporated into a range of food materials (carbohydrates, proteins, etc). Different biokinetic assumptions might apply to specific forms of OBT, for example, to OBT in fish from the Cardiff Bay area present as a result of discharges from the radiochemical production laboratories of GE Healthcare.

The AGIR subgroup considered the special case of exposures to DNA precursors and conclude that, since exposures can be expected to be low and the compounds will be largely catabolised to HTO, risks to workers and the general population are low.

Concerns have been expressed concerning possible tritium incorporation into female germ cells and the risk of heritable effects that this may pose. In principle, tritium incorporated into female germ cells during fetal development could remain present for decades until fertilisation. Such incorporated tritium could therefore cause significant genetic damage. Specific calculations were undertaken based on OBT concentrations measured in fish caught in the Cardiff Bay area and consumption rates applied to the local critical group. These calculations indicate that the risk of hereditary effects due to such exposures is several orders of magnitude less than the 3-4% spontaneous incidence of hereditary effects. Thus the report concludes that no specific protection of females from tritium is required with the caveat that existing evidence does not enable account to be taken of any consequences there might be resulting from bystander effects or genomic instability during in utero development.

The AGIR report conclusions and recommendations are accepted by HPA, specifically that:

On RBE, it is noted that the AGIR subgroup considers that the values obtained in animal carcinogenesis studies might be less than the maximum values that are applicable to cancer induction at low doses and dose rates. However, the point is also made that there are theoretical reasons for expecting values in animals (and humans) to differ from those obtained using cells in culture. Given the uncertainties associated with the current data, the HPA considers that further studies with suitable animal models of carcinogenesis would be valuable.

While the HPA accepts the conclusion that tritiated DNA precursors do not present a large risk to workers or the public, the risks remain uncertain. Further work to quantify doses and risks, including RBE values, for specific forms of OBT would be welcome.

While the HPA endorses the AGIR subgroup conclusions concerning RBE for tritium, the suggestion that the ICRP should consider the use of a wR of 2 relates to the wider issue of the definition and use of the ICRP protection quantities, equivalent and effective dose. The ICRP has been made aware of this recommendation and copies of the report will be sent to the ICRP and the relevant ICRP committees by the HPA. The ICRP position is that radiation weighting for all radiation types was reconsidered in the development of the 2007 recommendations2 (currently in press). The HPA accepts the judgements made by the ICRP in the calculation and use of its protection quantities and will provide further comments on this subject in the HPA response to the new ICRP recommendations.

The HPA welcomes the AGIR subgroup recommendation that further epidemiological studies should be undertaken. In general, the HPA recognises the importance of maximising the information that can be obtained from such studies, particularly concerning risks from chronic radiation exposures, including exposures to internal emitters.

References

1. HPA (2007). Review of risks from tritium. Report of the independent Advisory Group on Ionising Radiation. Docs. HPA, RCE-4.

2. ICRP (2007). 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann. ICRP, 37(2-4) in press.


Last reviewed: 18 April 2008