Publications

Documents of the NRPB: Volume 11 , No. 1

Risks of Second Cancer in Therapeutically Irradiated Populations: Comparison with Cancer Risks in the Japanese Bomb Survivors and in Other Exposed Groups: Report of an Advisory Group on Ionising Radiation


Author(s):

n/a

Summary:

This report reflects understanding and evaluation of the current scientific evidence as presented and referenced in the full document.

The summary and conclusions and the recommendations of the report are given below.

Price:

£22.50
ISBN 0-85951-442-0

Order

PDF download:

Not available

     

Summary and conclusions

Please note that the paragraph numbering from the full published document has been retained in the extract below but that references cited have not been included and a number of minor editorial changes have been made.

  1. Follow-up of the Japanese atomic bomb survivors is the primary source of data for the evaluation of radiation-induced cancer risk. However, patients treated for first primary cancer will be an increasingly important secondary source of data for the evaluation of radiation-induced cancer risk. In this document the available data relating to incidence of second cancers induced by radical radiotherapy have been examined. Cancer risks in these datasets are compared with risk estimates derived from the Japanese atomic bomb survivor cancer incidence data.
  2. High dose fractionated irradiation results in high levels of cell sterilisation. This influences recovery processes by determining cell survival and makes a contribution to repopulation by surviving cells, which may take place toward the end of a radiotherapy treatment schedule. Broadly similar considerations apply to brachytherapy using radioisotopes. These features contrast with the acute exposures of the Japanese atomic bomb survivors for whom cell sterilisation, recovery and repopulation were relatively less important in determining cancer risk.
  3. In evaluating data on second cancers following radiotherapy, the greater role of cell sterilisation might be expected to alter the dose-response relationship. At doses below those which can cause appreciable cell sterilisation the dose-response relationship is approximately linear-quadratic (i.e. the risk increases with increasing dose). At higher doses, the risk may no longer increase with dose as a result of the loss of those cells previously at the highest risk because of the damage they had sustained, with possibly even a decrease in second tumour incidence at the highest dose levels due to cell sterilisation. This implies that higher cancer risks could occur in tissues receiving only moderate dose, e.g. on the edge of treatment fields. Some evidence supports this prediction but it is not clear whether it is equally applicable to all tumours.
  4. It is important to continue to acquire dose-response data in relation to second tumour position in radiotherapy patients in order to clarify this. If correct, the proposition implies that older radiotherapy regimens, which had less spatial precision, greater dose heterogeneity and greater dose scatter, would be more likely to induce second tumours than those in current use. It also suggests that reduction of radiation dose at positions outside the intended treatment volume would be an important preventive measure. However, if there are tumour types for which the dose-response relationship is increasing with dose to very high dose levels, then reduction of in-field dose or field size would be required for reduction of cancer risk. Such tumour types may also be the most obvious candidates for increased risk as a result of radiation-chemotherapy interactions.
  5. Cancer can be viewed as a disease representing the sequential acquisition of genetic abnormalities. For example, the loss of the tumour suppressor gene function or the gain of characteristics through oncogene activation can lead to a step-wise progression to a life-threatening cancer. Some of these genetic changes affect the ability of a cell to respond to stress and exert control over proliferation and cell death. However, in many specific types of tumours the sequence of such events and their roles in neoplastic progression are unknown. The consequence of these genetic events is double-edged since acquisition of genomic abnormalities or instability can also influence therapeutic responsiveness and the development of treatment resistance.
  6. Cytotoxic agents referred to as 'alkylating drugs' remain among the most widely used in cancer chemotherapy. The main action of an alkylator is to damage the genetic material of both target and normal tissues. The intensity of treatment, namely duration and total dose, with alkylators is a significant factor in cancer induction in non-target tissues. In general terms chemotherapy produces a large increase in the risk of acute leukaemia, and does not efficiently produce other types of cancer. Patients treated with combined modalities constitute a potential source of information for the estimation of radiation-associated risks if the effects of the chemotherapeutic component can be set into perspective. Although, in general, there is little strong evidence for synergy between the two types of therapy in their carcinogenic action, it is recognised that the combination of intensive alkylator chemotherapy and intensive radiotherapy yields the greatest risk for treatment-related haematological and solid malignancies. It is in the combined modality treatments of childhood cancers that long-term effects are revealed with increasing evidence of the contributing effect of chemotherapy to iatrogenic neoplasia. This produces a dilemma in which a potentially curative procedure may carry a quantifiable risk of second cancer induction. In practical terms there is a case for generating new approaches to evaluate risks and benefits of cancer therapy.
  7. In general, for most cancer sites the relative risks for the induction of second cancers are comparable with or less than the relative risks derived from the Japanese data. In many cases the difference is not statistically significant, due at least in part to the small numbers of cases that are involved when total cancers are subdivided into individual types. For lung cancer, bone cancer, ovarian cancer and leukaemia the differences are statistically significant. At least for leukaemia, the discrepancy can be explained by cell-sterilisation effects and leukaemia subtype differences. For leukaemia there are seven available second cancer studies whereas for most cancer sites (all but lung, bone, female breast, thyroid, leukaemia) only a single second cancer study is available. The fact that in general the relative risks in the second cancer studies are lower than those based on the Life Span Study (LSS) data is reassuring for the therapist in so far as the risks derived from the LSS are likely to constitute an upper bound to the risks associated with radiotherapy.
  8. Although the analyses imply reasonable compatibility of cancer risks between the Japanese atomic bomb survivors and the medically irradiated groups when the effects of cell sterilisation (and for leukaemia, subtype differences) are taken into account, the analyses do not take into account the pattern of accumulation of dose over time. To this extent, the findings of general compatibility of cancer risk between the Japanese atomic bomb survivors and the second cancer groups must be somewhat qualified.
  9. There are few indications of modification of radiation-induced second tumour relative risk among those treated with adjuvant chemotherapy, nor are there strong indications of modification of radiation-induced relative risk by heritable genetic factors. There is weak evidence that the relative excess radiogenic cancer risks are lower among those patients with cancer-prone disorders than among non-susceptible patients. However, the absolute excess radiogenic cancer risk in those persons with cancer-prone conditions may well be higher than in normal (non-susceptible) patients.
  10. The treatment of cancer using radiation or cytotoxic drugs continues to be a two-edged sword where the risk of second cancers needs to be weighed against therapeutic benefit by the clinician and the patient. There is an advance towards new forms of therapy, devised to target tumour cells in a manner which reflects or manipulates their genetic and phenotypic properties. The risk of radiation-induced second cancers in novel combined modalities will be difficult to predict from existing information, although the current study provides a baseline for therapeutic situations which have been judged against the alternative source of atomic bomb survivor data. In pursuing the primary task of the Advisory Group, to see whether the radiation risk estimates in radiotherapy patients are in line with those of the LSS of atomic bomb survivors, the members became acutely aware of the problems of abstraction of high quality data from published reports. Our recommendations address some of these problems by highlighting the need for enhanced data recording, preservation, and linkage so that full use can be made of clinical databases. We suggest that accessible databases will improve the value of risk estimate studies and provide clues to the impact of changes in therapy design.
  11. In view of the importance of collecting details of treatment for persons treated for malignant neoplasms, the Royal College of Radiologists has recommended that:
    The medical records relating to radiotherapy and chemotherapy for all children and young adults (up to the age of 20 years at the time of diagnosis), treated in the past should be kept not only for the duration of the patient's lifetime, but as a permanent record... An important improvement for the future would be the preparation of a detailed record of chemotherapy and a summary of that treatment as a separate identifiable document, irrespective of whether this document is stored together with the other medical records or not... Wherever practicable, prospective records for both adult and child patients should be kept on a computer database and should include sufficient details of radiotherapy and chemotherapy to enable reconstruction of these treatments at a later date... Ideally, adult records should be kept according to the same guidelines as those of children and young adults, as a permanent record on a computer database. Where this is not possible, hard-copy records and simulator films should be kept for the patient's lifetime plus five years. The records of patients treated in research investigation should be retained according to the protocol for the research. EC guidance is that patient records used in connection with clinical trials should be kept for at least 15 years.

Recommendations

  1. With increasing numbers of patients being successfully treated with radiotherapy (with or without adjuvant chemotherapy) and surviving longer, the available data on radiation-induced second cancers should increase. It is recommended that the assessment of cancer incidence in all these and other groups exposed to high doses of ionising radiation be maintained and enhanced.
  2. The newly available linkage systems of regional and national cancer registries should be utilised to flag second and subsequent cancers so that details of treatment-induced malignancies can be assessed. It is recommended that pilot national studies of second malignancies using the cancer registration scheme be undertaken to ascertain their value for future epidemiological and biological studies. The results of the pilot studies, particularly with respect to the detail and accuracy of ascertainment of past therapy details, should be used to further improve the system of cancer registration in this particular area.
  3. Anecdotal observations constitute a further source of information, not considered in this report. They could form a rich source of clues to unknown determinants of second cancer formation and may highlight areas for future studies. There would be value in the establishment of an adverse event reporting system for second cancers similar to other such systems.
  4. The Royal College of Radiologists has recommended that:
    The medical records relating to radiotherapy and chemotherapy... should be kept... for at least 15 years.
    This particularly highlights the need for the records of children and young adults treated for malignant disease to be kept permanently. We support these recommendations and further recommend that such record keeping should be standardised so that retrospective dose calculations can be used for second cancer risk evaluation. A summary of the chemotherapy given should be kept in the notes with the radiotherapy details. Compliance with this should be audited.
  5. Studies should be conducted into the cost-effectiveness of methods of maintaining databases of clinical/treatment records for long-term archival purposes, and the feasibility of applying such methods in multicentre studies. These databases should allow efficient record linkage and high quality access for research purposes.
  6. Clinical trial design should continue to be improved such that data accrued meet the minimal quality hurdles for evaluation of second cancer risk in multi-trial analyses. There should be assessment of second cancer risks in gene-therapy based procedures, or in modalities involving biological response modifying agents.
  7. Studies of sensitive subgroups and patients who demonstrate high sensitivity to radiotherapy or chemotherapy in treatment for their first cancer should be encouraged. This is particularly important for paediatric groups given their long-term survival potential. Radiosensitive subgroups or other occult high risk groups of patients effectively reduce the maximum radiation and drug exposures considered tolerable for the general population. The risk of second tumours arising in more aggressively treated normal individuals will become a pressing issue if subgroups can be identified.
  8. There should be further experimental and epidemiological investigation of possible interactions between radiotherapy and chemotherapy as modifiers of cancer risk and risk benefit.
  9. Further experimental investigation of the dynamics of repopulation of normal and pre-malignant cells during and after irradiation would help to determine their influence on the dose response for the incidence of second cancer following radiotherapy.
  10. There should be further analyses of the Japanese atomic bomb survivors and the second cancer datasets using well-established cytogenetic and molecular endpoints as potential correlates of the radiation lesions relevant to the particular cancer types. Such analyses may enable better account to be taken of the biological effect of different patterns of delivery of dose, and thereby more reliably determine the extent of compatibility of cancer risks in these groups.

Acknowledgements

  1. The Advisory Group is very grateful for the helpful comments of Professor A Barrett. This document makes use of data obtained from the Radiation Effects Research Foundation (RERF) in Hiroshima, Japan. RERF is a private foundation funded equally by the Japanese Ministry of Health and Welfare and the US Department of Energy through the US National Academy of Sciences. The conclusions in this document are those of the Advisory Group and do not necessarily reflect the scientific judgement of RERF or its funding agencies. The International Radiation Study of Cervical Cancer patients (IRSCC) was coordinated by the International Agency for Research on Cancer. The Advisory Group gratefully acknowledges the permission to use these two original sets of data (RERF and IRSCC). Some of this work was funded by the European Commission under contract FI4P-CT95-0009.



Last reviewed: 29 August 2008