Radiation exposure is a possible predisposing factor for monoclonal gammopathy of undetermined significance (MGUS), but the association has been uncertain. We investigated the relationship between radiation exposure and MGUS prevalence by using data from the M-protein screening for Nagasaki atomic bomb survivors between 1988 and 2004. Radiation exposure was assessed by exposure distance from the hypocenter and exposure radiation dose. We computed prevalence ratios (PRs) and the 95% confidence intervals (CIs) adjusting for exposure age and sex. A total of 1082 cases of MGUS were identified from 52 525 participants. MGUS prevalence was significantly higher in people exposed at distance within 1.5 km than beyond 3.0 km (PR, 1.4; 95% CI, 1.1-1.9) among those exposed at age 20 years or younger, but it was not found among those exposed at age 20 years or older. MGUS prevalence was also significantly higher in people exposed to more than 0.1 Gy than those exposed to less than 0.01 Gy (PR, 1.7; 95% CI, 1.0-2.8) among those exposed at age 20 years or younger. Thus, people exposed at younger age exhibited a significantly high risk of MGUS when exposed to a high radiation dose. There was no clear association between radiation exposure and the malignant progression of MGUS. Further detailed analysis is needed.