ICRP task group 123: classification of harmful radiation-induced effects on human health for radiological protection purposes

1. INTRODUCTION

Within the current system of radiological protection, tissue reactions are based on injury in populations of normal cells and are characterised by a threshold dose set at 1% incidence within the population and an increase in the severity of the reaction as dose increases. Tissue reactions, also still commonly termed ‘deterministic effects’, are a wide variety of different types of health effects, and they are sometimes modifiable after radiation exposure, for example, by biological response modifiers. The stochastic effects, cancer, and heritable effects nominally result from damage in a single cell. In this case, the frequency of the effect, but not the severity, increases with increasing dose, without a dose threshold. The objective of the system is based on managing and controlling exposures to ionising radiation so that harmful tissue reactions are prevented, and the risks of stochastic effects are reduced to the extent reasonably achievable. Classification of health effects into these two distinct categories directly underpins the implementation of the system of radiological protection (ICRP, 2012).

In recent years, several authors have highlighted the need to revisit the classification of effects. While cataract was still considered a tissue reaction in ICRP Publication 118 (2012), the threshold for effects was reduced from 2 to 0.5 Gy with no differentiation between acute and protracted or fractionated exposures. Limited evidence for a progressive increase in severity with dose and over time was reported, with an inverse relationship between the latency period for opacification and dose. In contrast, Hamada et al. (2014) proposed that minor opacities may not necessarily progress to be vision impairing in nature. It was also clear from the chiefly epidemiological studies that the evidence could be interpreted as supportive of no threshold, and it is also worth noting the lack of evidence in support of an increase of severity with dose. Since the ICRP Publication 118, a number of mechanistic studies have been published on radiation effects in the lens, which tend to support stochastic (mutational) mechanisms, or indeed that cataract may be classified as a tissue reaction or a stochastic effect, depending on the type of cataract (Ainsbury et al., 2022, and references therein, including Pawliczek et al., 2022).

Circulatory disease was also considered in ICRP Publication 118 (2012) as a tissue reaction with an ‘approximate’ threshold of 0.5 Gy for acute and fractionated/protracted exposures, with the evidence suggesting that doses at this level would lead to an incidence of the order of 1% of circulatory disease in exposed individuals (but also recognising the high degree of uncertainty, and with severity not explicitly considered in ICRP Publication 118). Since the publication of ICRP Publication 118, there has been increasing evidence regarding the mechanisms underpinning the various effects which are collectively termed diseases of the circulatory system and which include cardiovascular and cerebrovascular diseases, vascular remodelling, stroke, and ischaemic heart disease, amongst others. The dose response in respect of each of these effects remains unclear, with inconsistent data in the published literature, possibly indicating different responses for the different types of diseases of the circulatory system (Little et al., 2023). Kozbenko et al. (2023) recently published an adverse outcome pathway for vascular remodelling with six key events including oxidative stress, altered signalling pathways, and increased proinflammatory mediators. A recent review of omics studies on radiation-induced cardiovascular disease showed the relevance of biological data for a deeper understanding of the mechanistic links underlying cardiac toxicity following irradiation. Although the review found that there are still significant research and knowledge gaps that need to be addressed, it highlighted the potential of biological (omics) data for integration with epidemiological findings to reduce uncertainty in radiation risk assessment (Azimzadeh et al., 2022). The identification of some consistent omics data after different radiation exposure scenarios suggests common adverse consequences of radiation exposure for the cardiovascular system and challenges current assumptions of cell and tissue response at different doses, dose rates, and time points.

Effects on the nervous system were also addressed in ICRP Publication 118 (2012); however, at the time of publication, the available evidence was not able to support the estimation of a potential threshold for effects following single acute exposure. It was recognised, however, that irradiation to the developing brain of children can cause long-term cognitive and behavioural defects and that children are more susceptible to cognitive impairment following exposure to doses >0.1 Gy. Lopes et al. (2022) recently summarised epidemiological studies of low-to-moderate radiation exposure in adulthood. The authors found significant excess relative risks of cerebrovascular disease incidence and mortality and Parkinson's disease in a similar dose range as with children. In an article summarising the 2021 Symposium on ‘Radiation risks of the central nervous system’, Zabloska et al. (2022) concluded that low-dose effects are observed following exposure in utero or in early childhood, including reduced neuronal proliferation, altered neurogenesis, neuroinflammation, and various neurological and psychological consequences. More recently, Dauer et al. (2023) conducted a meta-analysis of six cohorts of the Million Person Study (MPS), where more than half a million workers were exposed to 6.9- to 47.6-mGy brain dose. Five of the six cohorts revealed a positive association with Parkinson's disease with an excess relative risk at 100 mGy of 0.17 (95% CI: 0.05–0.29).

ICRP Publication 118 also recognised that while the prudent judgement at the time of publication was to provide threshold values for both the lens and the circulatory system assuming ‘the same incidence of injury irrespective of the acute or chronic nature of the exposure’ and over a period >20 years for the lens and >10 years for the circulatory system, the authors also specifically recommended further studies in this area. In terms of wider effects, ICRP Publication 118 (2012) considered but drew few conclusions in relation to the respiratory, musculoskeletal, and endocrine systems as well as retinal effects. Recent publications may suggest that additional radiation exposure-linked effects should be considered in future ICRP publications, including systemic immune and/or metabolic effects and wider effects in the eye including glaucoma. Biological response modifiers were additionally highlighted as an area of increasing importance as more is known about their effects (ICRP, 2012).

The clear need to reconsider the classification of effects in the light of this emerging evidence is highlighted in the recent publications of Clement et al. (2021), who proposed that for protection purposes, it may be desirable to distinguish severe and other tissue reactions, or short-term and long-term health effects, but also noted that some radiation-linked effects may not fit well into either category. Also highlighted was the need to assess the impact of any newly proposed classification scheme on the system, including on tolerability and management of radiological risks, and that any new classification must change the fundamental requirement of prevention of severe tissue reactions and optimisation of protection against stochastic effects. Classification of effects also emerged as a key theme of the 2021 ICRP Workshop on the Future of Radiological Protection, as summarised by Rühm et al. (2022). Several presenters at the workshop highlighted that the current system of classification is incompatible with the new and emerging evidence in this area and suggested new ways to consider classification (e.g. Nakamura, 2023; Trott, 2023).

In summary, the case for clarification of the rationale of the current system of classification and the consideration of evolution in this area in response to the emerging evidence on various health effects is clear.

2. ESTABLISHMENT OF ICRP TASK GROUP 123

In 2022, ICRP Task Group (TG) 123 on ‘Classification of Harmful Radiation-induced Effects on Human Health for Radiological Protection Purposes’ was established as a joint TG between ICRP Committees 1 on Radiation Effects and 4 on Application of the Commission's Recommendations. The TG has the following objectives:

To clarify the rationale behind the current classification (based on a review of relevant ICRP Publications) and the primary protection objectives of the ICRP system.

A total of 12 colleagues from Committees 1 and 4, as well as experts from the wider community, contribute to the work of ICRP TG 123, which seeks to fulfil the above aims by reviewing the relevant evidence published since the ICRP Publication 118 as well as by considering the uncertainties, including but not limited to the underlying mechanisms of the various effects, the threshold doses, individual sensitivity, and variation with age. Work with the ICRP committees and other TGs is clearly necessary and mandated in the terms of reference, as is collaboration with other international bodies considering the evidence in relation to classification and/or responsible for implementation of revisions to the system, including the International Atomic Energy Agency (IAEA), International Radiation Protection Association (IRPA), Multidisciplinary European Low Dose Initiative (MELODI), National Energy Agency (NEA), and United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) (ICRP, 2023).

3. PROGRESS TO DATE

The TG has met several times online and in person and has agreed on the basis for the work, including that reclassification must not change the fundamental requirement to prevent severe tissue reactions and optimise protection at lower doses and dose rates and that the wide range of potential implications of any changes must be fully understood. The draft table of contents has been elaborated, and writing tasks have been assigned. In addition, a total of seven mentees have been appointed to assist with reviewing the literature in respect of the evolution of the current system as well as the new and emerging evidence in relation to classification of radiation-linked health effects (including through joint meetings/workshops and collaboration with other TGs, Committees 1 and 4 on the effects and the application of any proposed changes, and wider international partners, as outlined above). It is aimed to complete the reviewing tasks in 2024 and then to work together to define and consider the potential evolution of the way in which health effects are classified, likely through an options appraisal style approach. Stakeholder input to the work is also essential, and this will be ensured through at least two community workshops sharing the initial and then the final approaches proposed by the TG.

In conclusion, classification of effects is a cornerstone of the current system of radiological protection, and it is essential that this remains fit for purpose in support of the system. Any evolution of the way in which radiation-linked health effects are classified will have major implications for practical radiation protection, and as such the work of ICRP TG 123 is an essential part of the current review of the system of radiological protection.