Comments to ICRP on the ICRP Draft 2005 Recommendations 1. General I found the document well structured and readable and my comments are restricted to points requiring clarification. I have raised a series of questions for which I would appreciate answers. 1.1. Index It is difficult to check whether some points have been covered without and index that will be included later? 1.2. Framework documents Will time be given for consideration and comment on the new Framework Documents? 1.3. BEIR 7 There has been much more epidemiological data collected since 1990 with continued follow-up on the A-bomb survivors; medically exposed groups; NRRW data on occupational exposure; radon in homes data; and environmental exposure following the Chernobyl accident. This data was reviewed in the UNSCEAR 2000 report and this is presumably included but various versions of BEIR 7 are out for review and should be published in mid 2005. Will there findings be considered by ICRP for inclusion in the recommendations? 1.4. Lung cancer anomaly Data from other studies are reasonably compatible with the A- bomb data taking into account uncertainties in the data, however, there are still differences of x 2-3 in lung cancer risks compared with lung cancer in radon-exposed miners. Can this anomaly be resolved before publication of the recommendations? 1.5. Extrapolation of cancer risks to low doses This will be the topic considered in a forthcoming ICRP Task Group report. Although it seems likely that ICRP will stick with the linear, no-threshold hypothesis it would be good to see the arguments justifying this approach. 2. Tissue reactions Page 16, para. 47: ¡®Deterministic¡¯ has now been replaced by ¡®tissue reactions¡¯ which I believe is an improvement. However, the radiation weighting factor based upon RBE needs to be included in the document. I note that new low dose data for cataracts are currently being reviewed and needs to be included. This could apply to future cataract cases. 3. Tissue weighting factors Pages 24-5, para. 80-1: The definition of detriment has changed and the weighting factors for individual tissues based upon their detriment contributions are still being finalised and will be part of a new Foundation Document which I hope will be available for comment before the recommendations are finalised? 4. Non-cancer diseases after radiation exposure Page 33, para. 123-5: There is increasing evidence of raised risks of heart disease, stroke, respiratory and digestive diseases among Japanese A-bomb survivors (and radiotherapy patients for heart disease) that are consistent both with a linear dose-response and with no raised risk below 0.5 Sv. It is unclear what forms of cellular or tissue mechanisms might underlie such a wide range of disorders. You state that ICRP judges that the data available on non-cancer disease does not allow for their inclusion in estimating radiation detriment for doses up to a few tens of mSv. However, the potential importance of these data needs to be recognized and if it is linear then the excess of absolute risk for non-cancer mortality might be similar to that for cancer mortality. I understand that UNSCEAR will be publishing data in the future and this should be included in the ICRP recommendations if they are not to be out-of-date very soon. 5. Radiation Weighted Dose and Effective Dose 5.1. Heterogeneity in organs Page 18, para. 50: Heterogeneity is recognized for internal radiations (alpha, low-energy beta and Auger radiations) but why not for external radiations including low-energy X radiation and neutron recoils? 5.2. Radiation weighted dose Page 18, para. 51: I welcome the replacement of ¡®Equivalent dose¡¯ by ¡®radiation weighted dose¡¯ but I am against the use of different units other than the sievert because we have argued that the use of a single unit for different quantities is fully justified in other fields and is now well established in Protection. 5.3. Use of risk estimates Page 19, para. 54: I accept that the risk estimates are subject to various sources of uncertainty and have been developed for protection purposes. However, the ICRP provide an excellent and authoritive assessment of the risk data that will continue to be used for retrospective studies, taking into account the specific organ at risk. 6. Radiation and Tissue Weighting Factors It is accepted that wR and wT only relate to stochastic effects but will be required for tissue reactions. There is recognition that wR is for the external field and the internal field in an organ may be very different. wR for alpha particles remains at 20 but could be influenced by further studies on radon in miners. I welcome the reduction in wR for protons from 5 to 2. 6.1. Neutron Radiation Weighting Factors ICRP published ICRP 92 as a report to put the radiation weighting factors wR and quality factors into the context of the linear energy transfer (LET) based upon the determined relative biological effectiveness for appropriate end points. wR was defined in an arbitrary way in ICRP 60 and caused problems in comparing radiological and operational quantities. In ICRP 60, the neutron wR was defined as a step function with a smooth function as an approximation that was used in most operational calculations. wR = 5.0 + 17.0 (exp[-{ln (2 En}2 / 6]) The recognition that neutrons below 1 MeV produce mainly gamma radiation in humans is welcomed. A smooth function, related to LET is accepted as more appropriate for all purposes. The neutron weighting factor is based upon LET or the effective quality factor in the human body qE is good. It is argued in ICRP 92 that, because the current wR values have been incorporated into national and European legislation, there should not be a reduction in wR for 1 MeV neutrons. Thus instead of being directly proportional to qE, the radiation weighting factor is: wR = 1.6 (qE ¨C 1) + 1 = 1.6 qE ¨C 0.6 (1) that preserves a value of wR of about 20 at incident neutron energies near 1 MeV and gives a value of 1.0 for low LET radiations. This is not a scientific reason for not using qE directly. The neutron weighting factors will be reduced by a factor of 2 below 1 MeV and the function proposed is: wR = 2.5 + 18.2(exp[-{ln En}2 / 6]) For energies ¡Ý 1 MeV, the formula from ICRP 60 is used, viz: wR = 5.0 + 17.0 (exp[-{ln (2 En}2 / 6]) It is recognised by ICRP that the functions in the proposed recommendations differ from those proposed in ICRP 92 but this is not justified. Kellerer et al (2004) have endorsed the ICRP 92 approach using a wR related to LET without necessarily supporting the formula in equation (1). They propose two options: Option 1 which does not include the artificial factor of 1.6 and so would be more appropriate for Scheme cases since it reflects the actual risk: wR = 2 + 10(exp [-{ln En}2 / 4] + 2.5 (exp [-{ln En/20}2 / 12] Option 2 includes the artificial factor of 1.6 to preserve the current factors in legislation: wR = 2.6 + 16(exp [-{ln En}2 / 4] + 4 (exp [-{ln En/20}2 / 12]. Option 2 gives a wR of about 1.6 times that for option 1 and gives similar results to those from ICRP 92 and the new recommendations. Option 1 would surely be justified scientifically as a basis for wR. The fluence to effective dose conversion coefficients have been published by Pelliccioni (2004) for isotropic neutron and incidence and form the basis for considering the effects of the changes in wR for dose equivalent measurements. The ratios of the radiation weighting factors for neutrons for the various references to ICRP 60 are given in Table 1 where ICRP Rec. refers to the new draft recommendations. Table 1. Radiation Weighting Factors for Neutrons Neutron Ratios of Radiation Weighting Factors Fluence Ratio(3) Energy MeV ICRP 92 ICRP 60 ICRP Rec. ICRP 60 RPD(1) ICRP 60 RPD(2) ICRP 60 RPD(1) RPD(2) ICRP 92 ICRP 60 1.00E-06 0.500 0.500 0.400 0.520 0.769 0.499 1.00E-05 0.500 0.500 0.400 0.520 0.769 0.500 1.00E-04 0.500 0.500 0.400 0.520 0.769 0.499 1.00E-03 0.499 0.499 0.398 0.517 0.769 0.497 1.00E-02 0.422 0.479 0.327 0.429 0.763 0.416 1.00E-01 0.456 0.625 0.305 0.451 0.677 0.434 2.00E-01 0.593 0.724 0.387 0.589 0.657 0.557 5.00E-01 0.817 0.877 0.531 0.822 0.646 0.800 1.00E+00 0.965 1.000 0.637 0.990 0.643 0.966 2.00E+00 1.059 1.000 0.719 1.116 0.644 1.062 5.00E+00 1.110 1.000 0.779 1.196 0.651 1.088 1) RPD(1) is option 1 in Kellerer et al (2004) 2) RPD(2) is option 2 in Kellerer et al (2004) 3) Pelliccioni (2004) It can be seen that both the draft 2005 recommendations and those in ICRP 92 will have weighting factors between 50% and 100% of those in ICRP 60 and these will lead to fluence ratios that will be similarly reduced depending upon the hardness of the neutron spectrum. The use of option 1 in Kellerer et al (2004) could lead to a further reduction of 0.7, on average, and would be appropriate for the new recommendations. There are differences in the formulae between the new recommendations and ICRP 92 and these will need to be resolved. 6.2. Tissue weighting factors I understand that these weighting factors could change in the final recommendations and could have significant influence in Medicine during radiotherapy and in assessing the risk of radio-diagnosis. 7. Conclusions I welcome the intention of ICRP is to revise rather than make wholesale changes as much of the previous recommendations have been incorporated into regulations throughout the World, including Europe and the UK. 8. References ICRP 60 1990 Recommendations of the International Commission of Radiological Protection. Annals of the ICRP 21 (1-3) (1991). ICRP 92 Relative Biological Effectiveness (RBE), Quality Factor, and Radiation Weighting Factor (wR) Annals of the ICRP (2003). ICRP Rec. 2005 Recommendations of the International Commission on Radiological Protection. Downloaded from the ICRP web site (2004). Kellerer, A. M., Leuthold, G., Mares, V. and Schraube, H. Options for the modified radiation weighting factor of neutrons. Radiat. Prot. Dosim. 109 181-188 (2004). Pelliccioni, M. The impact of ICRP publication 92 on the conversion coefficients in use for cosmic ray dosimetry Prot. Dosim. 109 303-309 (2004).