Age at exposure and attained age variations of cancer risk in the Japanese A-bomb and radiotherapy cohorts

Authors

  • Schneider Uwe,

    1. Institute of Physics, Science Faculty, University of Zürich, Zürich 8057, Switzerland and Radiotherapy Hirslanden, Uwe Schneider Institute of Radiotherapy, Witellikerstr. 40, Zürich 8032, Switzerland
    Search for more papers by this author
    • a)

      Author to whom correspondence should be addressed. Electronic mail: uwe.schneider@uzh.ch; Telephone: +41-443872559.

  • Walsh Linda

    1. Institute of Physics, Science Faculty, University of Zürich, Zürich 8057, Switzerland and BfS - Federal Office for Radiation Protection, Radiation Protection and Health, Neuherberg 85764, Germany
    Search for more papers by this author

Abstract

Purpose:

Phenomenological risk models for radiation-induced cancer are frequently applied to estimate the risk of radiation-induced cancers at radiotherapy doses. Such models often include the effect modification, of the main risk to radiation dose response, by age at exposure and attained age. The aim of this paper is to compare the patterns in risk effect modification by age, between models obtained from the Japanese atomic-bomb (A-bomb) survivor data and models for cancer risks previously reported for radiotherapy patients. Patterns in risk effect modification by age from the epidemiological studies of radiotherapy patients were also used to refine and extend the risk effect modification by age obtained from the A-bomb survivor data, so that more universal models can be presented here.

Methods:

Simple log-linear and power functions of age for the risk effect modification applied in models of the A-bomb survivor data are compared to risks from epidemiological studies of second cancers after radiotherapy. These functions of age were also refined and fitted to radiotherapy risks. The resulting age models provide a refined and extended functional dependence of risk with age at exposure and attained age especially beyond 40 and 65 yr, respectively, and provide a better representation than the currently available simple age functions.

Results:

It was found that the A-bomb models predict risk similarly to the outcomes of testicular cancer survivors. The survivors of Hodgkin's disease show steeper variations of risk with both age at exposure and attained age. The extended models predict solid cancer risk increase as a function of age at exposure beyond 40 yr and the risk decrease as a function of attained age beyond 65 yr better than the simple models.

Conclusions:

The standard functions for risk effect modification by age, based on the A-bomb survivor data, predict second cancer risk in radiotherapy patients for ages at exposure prior to 40 yr and attained ages before 55 yr reasonably well. However, for larger ages, the refined and extended models can be applied to predict the risk as a function of age.

Ancillary