Cancer risks after radiation exposure in middle age

Abstract

Background: Epidemiological data show that radiation exposure during childhood is associated with larger cancer risks compared with exposure at older ages. For exposures in adulthood, however, the relative risks of radiation-induced cancer in Japanese atomic bomb survivors generally do not decrease monotonically with increasing age of adult exposure. These observations are inconsistent with most standard models of radiation-induced cancer, which predict that relative risks decrease monotonically with increasing age at exposure, at all ages.

Methods: We analyzed observed cancer risk patterns as a function of age at exposure in Japanese atomic bomb survivors by using a biologically based quantitative model of radiation carcinogenesis that incorporates both radiation induction of premalignant cells (initiation) and radiation-induced promotion of premalignant damage. This approach emphasizes the kinetics of radiation-induced initiation and promotion, and tracks the yields of premalignant cells before, during, shortly after, and long after radiation exposure.

Results: Radiation risks after exposure in younger individuals are dominated by initiation processes, whereas radiation risks after exposure at later ages are more influenced by promotion of preexisting premalignant cells. Thus, the cancer site-dependent balance between initiation and promotion determines the dependence of cancer risk on age at radiation exposure. For example, in terms of radiation induction of premalignant cells, a quantitative measure of the relative contribution of initiation vs promotion is 10-fold larger for breast cancer than for lung cancer. Reflecting this difference, radiation-induced breast cancer risks decrease with age at exposure at all ages, whereas radiation-induced lung cancer risks do not.

Conclusion: For radiation exposure in middle age, most radiation-induced cancer risks do not, as often assumed, decrease with increasing age at exposure. This observation suggests that promotional processes in radiation carcinogenesis become increasingly important as the age at exposure increases. Radiation-induced cancer risks after exposure in middle age may be up to twice as high as previously estimated, which could have implications for occupational exposure and radiological imaging.

Figure 1

Excess relative risks (ERRs) per Gy for cancer incidence in Japanese atomic bomb survivors as a function of age at radiation exposure. ERRs were estimated at an attained age of 80 years and sex averaged, except for female breast cancer. Solid cancers refers to all primary malignant tumors excluding hematopoietic cancers. The data points are derived from Walsh (5) and Little (6) and the error bars represent 90% confidence intervals. The curves represent fits to the ERR data using the quantitative mechanistic model (23) described in the text and the parameters detailed in Table 1.

Figure 2

Schematic illustrating the dominant factors determining the variation in radiation-induced cancer risk with age at exposure. Jagged arrows indicate different times of radiation exposure, and the solid circles represent risks at a given attained age (eg, 80 years). A) Excess risk per year due to radiation initiation; for an exposure at a younger age, initiated cells have longer to exploit their growth advantage over normal cells. B) Excess risk per year due to radiation promotion; people irradiated at older ages, when there are more premalignant cells for promotion to act upon, are expected to have larger promotion-driven risks. C) Excess lifetime risks due to radiation-induced initiation and promotion. Initiation and promotion result in very different variations in cancer risk as a function of age at exposure; the downturn in excess lifetime risk shown in (C) for very old ages at exposure is due mainly to competing risks.

Figure 3

Estimates of absolute lifetime radiation-induced cancer risks (per 0.1 Gy per 100 000 persons), as a function of age at exposure. The smooth curves, results of this analysis, are predicted absolute radiation-induced lifetime cancer risks in a US population as a function of age of exposure; the shaded band represents the 95% confidence intervals, estimated with Monte Carlo simulations described in the text. The stepwise lines represent estimates of the same absolute lifetime radiation-induced cancer risks taken from table 12D-1 of the Biological Effects of Ionizing Radiation-VII report (14); these latter estimates were based on analysis of essentially the same datasets as the current analysis, but with the constraint that the excess relative risk was not permitted to increase with increasing age at exposure.