Journal of Radiation Research, Vol. 59, No. 1, 2018, pp. 1–9 doi: 10.1093/jrr/rrx047 Advance Access Publication: 21 September 2017 Cancer mortality in residents of the terrain-shielded area exposed to fallout from the Nagasaki atomic bombing 1, 1 1 2 Kenichi Yokota , Mariko Mine , Hisayoshi Kondo , Naoki Matsuda , 3 4 Yoshisada Shibata and Noboru Takamura Biostatistics Section, Division of Scientiﬁc Data Registry, Atomic Bomb Disease Institute, Nagasaki University, 12-4 Sakamoto 1-chome, Nagasaki 852-8523, Nagasaki, Japan Department of Radiation Biology and Protection, Atomic Bomb Disease Institute, Nagasaki University, 12-4 Sakamoto 1-chome, Nagasaki 852-8523, Japan Department of Radiation Medical Sciences, Atomic Bomb Disease Institute, Nagasaki University, 12-4 Sakamoto 1-chome, Nagasaki 852-8523, Japan Department of Global Health, Medicine and Welfare, Atomic Bomb Disease Institute, Nagasaki University, 12-4 Sakamoto 1-chome, Nagasaki 852-8523, Japan *Corresponding author. Biostatistics Section, Division of Scientiﬁc Data Registry, Atomic Bomb Disease Institute, Nagasaki University, 12-4 Sakamoto 1-chome, Nagasaki 852-8523, Japan. Tel: +81-95-819-7127; Fax: +81-95-819-7131; Email: email@example.com Received May 21, 2017; Revised July 26, 2017; Editorial Decision August 4, 2017 ABSTRACT The health effects of radiation exposure from the atomic bomb fallout remain unclear. The objective of the present study is to elucidate the association between low-dose radiation exposure from the atomic bomb fallout and cancer mortality among Nagasaki atomic bomb survivors. Of 77 884 members in the Nagasaki University Atomic Bomb Survivors Cohort, 610 residents in the terrain-shielded area with fallout were selected for this analysis; 1443 residents in the terrain-shielded area without fallout were selected as a control group; and 3194 residents in the direct expos- ure area were also selected for study. Fifty-two deaths due to cancer in the terrain-shielded fallout area were observed during the follow-up period from 1 January 1970 to 31 December 2012. The hazard ratio for cancer mor- tality in the terrain-shielded fallout area was 0.90 (95% conﬁdence interval: 0.65–1.24). No increase in the risk of cancer mortality was observed, probably because the dose of the radiation exposure was low for residents in the terrain-shielded fallout areas of the Nagasaki atomic bomb, and also because the number of study subjects was small. KEYWORDS: atomic bomb survivors, cancer mortality, fallout, terrain shielding, epidemiology INTRODUCTION ground (Table 1). These measurements indicate that a large portion On 9 August 1945, an atomic bomb was detonated 503 m above of the Nishiyama region was contaminated with radionuclides Nagasaki city . After the explosion, radioactive clouds containing whose radioactivity was higher than 0.8 mR/h . However, in dust and ashes formed and passed the east area at 3 m/s. The October 1948, 3 years after the bombing, no residual radioactivity Nishiyama region, located east of Nagasaki city, was showered by was detected around the Nishiyama reservoir , probably because yellow-brown droplets and was contaminated more highly than the of radioactive decay and rainfall. In 1959, Shono estimated from unshowered areas (Fig. 1). Mt Kompira (elevation: 366 m), measurements of residual radiation that the average cumulative dose which is located 2 km east of the hypocenter, shielded the until 1959 of external exposure in the Nishiyama region was 68 R Nishiyama region from direct radiation exposure. In September– . Thus, although various measurements of the residual radiation November 1945, several survey groups measured the residual radio- were performed, the dose was not reliably estimated. active intensity in the Nishiyama region [2–5]. They observed read- Regarding the somatic effects, temporary leukocytosis was ings as high as 1.08 and 1.8 mR/h using Geiger-Müller counters and observed in residents of the Nishiyama region; the mean leucocyte 3.9 mR/h using Lauritsen electroscopes at various heights above count in 25 residents increased during the period 70–80 days after © The Author 2017. Published by Oxford University Press on behalf of The Japan Radiation Research Society and Japanese Society for Radiation Oncology. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re- use, please contact firstname.lastname@example.org � 1 Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 2 � K. Yokota et al. Fig. 1. Study areas, including the terrain-shielded area. The area shielded from the detonation point of the atomic bomb 503 m above the hypocenter is shown in shades of gray. The Nishiyama region (NY) and control area (CT) were terrain shielded. Fallout was observed in the Nishiyama region, but not in the control area. The direct exposure areas on the near side (R1) and the far side (R2) of the hypocenter are also shown. Table 1. Early surveys of residual radiation in the Nishiyama region Investigator Instrument Maximum Probe height Equivalent Date measured Survey period exposure rate above the dose rate (mR/h) ground (cm) (μSv/h) First Technical Group of Geiger-Müller 1.8 5 15.8 27 Sept 1945 21 Sept to 4 Oct the Manhattan counter 1945 Engineering District Captain Warren’s Navy Geiger-Müller 1.08 100 9.5 Unknown (18 Oct 18 Oct to 17 Nov Group counter to 17 Nov 1945) 1945 Japanese scientists Lauritsen 3.9 15 34.1 1 Oct 1945 1 Oct 1945 to 21 electroscope Oct 1948 a b c See reference . See reference . See reference . the bombing, and returned to normal levels 100 days after the survivors, respectively; the prevalence of solid thyroid nodules was bombing . Regarding internal exposure, Okajima  reported signiﬁcantly (P < 0.01) higher in Nishiyama residents than in con- that Nishiyama residents had signiﬁcantly higher concentrations of trols. In 2014, Sakata et al.  reported the effects of the rain Cs by whole-body counting compared with sex- and age-matched exposure in Hiroshima and Nagasaki based on a questionnaire about controls who were not in Nagasaki city at the time of the bombing; rain exposure completed shortly after the bombing among the Life he inferred that this was likely due to the consumption of contami- Span Study (LSS) cohort. Of 733 subjects who reported rain expos- nated crops. He estimated the average internal doses in males and ure in Nagasaki, 394 deaths were observed during 1950–2005. In females in the Nishiyama region to be 2.9 μSv/y and 1.9 μSv/y, this group only, marginal association (excess relative risk = 0.08, respectively, and those in controls to be 1.9 μSv/y and 1.1 μSv/y, 95% conﬁdence interval = 0.00006–0.17, P = 0.05) was observed, respectively. However, no abnormalities were detected in chromo- but they concluded the ﬁndings may be spurious. The numbers of some studies, and no goiter, hypothyroidism, or thyroid cancers solid cancer deaths and leukemia deaths in the low-dose (<5 mGy) were detected among the residents examined . About 15 years subjects and in the reference group during 1962–2005 were 43 and after the publication of the study by Okajima , Nagataki et al.  1572, respectively. No association between rain exposure and cancer conducted a study on thyroid disease comparing 184 Nishiyama death was noted among subjects exposed to low-dose radiation. residents and 368 controls comprising atomic bomb survivors who The aim of the present study was to elucidate the effects of exposure were exposed to <0.1 mSv of atomic bomb radiation, and observed to low-dose radiation on cancer mortality in the residents who were liv- solid thyroid nodules in nine (4.9%) and three (0.8%) of these ing or staying in the fallout area from the Nagasaki atomic bomb. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Cancer mortality in the fallout area � 3 ~55 years old and after, effects of radiation exposure on cancer inci- MATERIALS AND METHODS dence in them will be attenuated and more difﬁcult to detect. It The present study was reviewed and approved by the institutional would also be difﬁcult to detect an increase in mortality in elder ethical committee of the Nagasaki University Graduate School of subjects, because the observable person-years in them should be Biomedical Sciences (No. 16012980). All of the data were obtained from the Nagasaki city government on the basis of the documented small. Furthermore, younger people at the time of the bombing have a higher cancer risk than older people . We, therefore, agreement between Nagasaki Unversity and the Nagasaki city govern- restricted study subjects to younger survivors. Thus, of the 77 884 ment. Using the data was also approved in that agreement for survivors in the study base population, 44 325 aged under 30 years research concerning the late effects of radiation exposure. We per- of age ATB were selected. Among these survivors, 1443, 610, 2180 formed analysis with anonymized data and announced the aims and and 1014 were living in the control area, the Nishiyama region, the procedure to the public (http://www-sdc.med.nagasaki-u.ac.jp/abcen- near-side direct exposure area and the far-side direct exposure area, ter/sdr/biostatistics_e.html). All methods were performed in accord- respectively. Finally, these 5247 subjects were included in the ana- ance with the relevant guidelines and regulations. lysis for the present study (Fig. 2). We also investigated cancer deaths coded as ICD-9: 140–208 or ICD-10: C00-C97 during the Nagasaki University Atomic Bomb Survivors Cohort 43 years between 1 January 1970 and 31 December 2012. Nagasaki University Atomic Bomb Survivors Cohort, which was The characteristics of the study subjects are shown in Table 2. established in 1978 and has continuously been updated, includes a The number of males aged 20–29 years at the time of the bombing large portion of atomic bomb survivors who were living in Nagasaki was smaller than that in the other age groups. Because most of the city as at 1 January 1970 . About 60 percent of this cohort people in this generation were soldiers, they were not in Nagasaki members are different from the Nagasaki members of the LSS at that time and thus not exposed to the atomic bombing. About cohort of the Radiation Effects Research Foundation. 55% of the study subjects were within 2.5–2.9 km of the hypocenter, They either currently possess, or once possessed, the Atomic and 45% were within 4.0–4.9 km; the Nishiyama region is located Bomb Handbook as a certiﬁcate of disaster issued by the Nagasaki 2.5–5 km from the hypocenter, and the number of subjects in that city government. For each cohort member, information related to area exposed to the bombing was the smallest (n = 610) among all the atomic bombing including the location of exposure, the distance the study areas. The proportion of subjects in direct exposure areas from the hypocenter, and the shielding conditions were recorded. who were 0–9 years of age ATB was smaller than that in the The individual radiation dose was estimated in those cohort mem- shielded areas. The proportion of subjects who were exposed out- bers who were within ~2 km of the hypocenter at the time of the side without being shielded was relatively high in the Nishiyama bombing . Information on their death, which has been available region, whereas this proportion was nearly identical in other areas. since 2 January 1970, unless they moved out of Nagasaki city, and the date they moved out of Nagasaki city were also recorded. The underlying cause of death was selected and coded by experienced Statistical analysis staff members of Atomic Bomb Disease Institute according to the First, we checked death rates based on observed person-years to International Statistical Classiﬁcation of Diseases and Related obtain an overall picture of cancer deaths. For the main analysis, Health Problems (ICD). We used the 9th edition (ICD-9) and the 10th edition (ICD-10) for deaths occurring from 1 January 1970 to 31 December 1994 and from 1 January 1995 onward, respectively. Study subjects To elucidate the health effects associated with the Nagasaki atomic bomb fallout, we ﬁrst identiﬁed, the areas that were shielded from direct radiation exposure by Mt Kompira (366 m high) and a 260 m ridge to the east of the hypocenter using Geographic Information System and elevation data [13, 14]; it should be noted that the atomic bomb exploded 503 m above Nagasaki city . We selected two areas from the identiﬁed areas, the Nishiyama region and the control area. According to historical records , no fallout was recorded in the control area adjoining the Nishiyama region. We also selected two areas from the direct exposure area: one was 2.0–2.4 km from the hypocenter (near-side direct exposure area) and the other was 2.5 km or more from the hypocenter (far-side dir- ect exposure area) (Fig. 1). We selected subjects who were younger than 30 years of age at the time of the bombing (age ATB); survivors who were 30 years old at the time of the bombing were 55 years old at the beginning Fig. 2. Selection of the study subjects. of the follow-up in 1970. Since cancer incidence usually increases at Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 4 � K. Yokota et al. Table 2. Characteristics of the study subjects a b c d Characteristic CT NY R1 R2 Terrain-shielded area Terrain-shielded area Direct exposure area, near Direct exposure area, far without fallout with fallout side side (n = 1443) (n = 610) (n = 2180) (n = 1014) Male Female Male Female Male Female Male Female Age ATB (years) 0–9 313 (54.9) 304 (34.8) 160 (63.2) 144 (40.3) 385 (42.6) 374 (29.3) 135 (36.7) 145 (22.4) 10–19 201 (35.3) 294 (33.7) 83 (32.8) 115 (32.2) 413 (45.7) 469 (36.7) 178 (48.4) 242 (37.5) 20–29 56 (9.8) 275 (31.5) 10 (4.0) 98 (27.5) 105 (11.6) 434 (34.0) 55 (14.9) 259 (40.1) Distance from hypocenter (km) 1.5–1.9 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 5 (0.6) 2 (0.2) 0 (0.0) 0 (0.0) 2.0–2.4 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 898 (99.4) 1275 (99.8) 0 (0.0) 0 (0.0) 2.5–2.9 570 (100.0) 873 (100.0) 140 (55.3) 214 (59.9) 0 (0.0) 0 (0.0) 336 (91.3) 548 (84.8) 3.0–3.9 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 32 (8.7) 98 (15.2) 4.0–4.9 0 (0.0) 0 (0.0) 113 (44.7) 143 (40.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Entering near hypocenter Entered 129 (22.6) 193 (22.1) 40 (15.8) 55 (15.4) 232 (25.7) 335 (26.2) 115 (31.3) 166 (25.7) Did not enter 441 (77.4) 680 (77.9) 213 (84.2) 302 (84.6) 671 (74.3) 942 (73.8) 253 (68.8) 480 (74.3) Shield conditions Outside, not shielded 110 (19.3) 113 (12.9) 89 (35.2) 114 (31.9) 158 (17.5) 143 (11.2) 63 (17.1) 65 (10.1) Outside, shielded 68 (11.9) 64 (7.3) 24 (9.5) 32 (9.0) 146 (16.2) 145 (11.4) 33 (9.0) 44 (6.8) Inside house 358 (62.8) 645 (73.9) 127 (50.2) 197 (55.2) 492 (54.5) 844 (66.1) 243 (66.0) 499 (77.2) Unknown 34 (6.0) 51 (5.8) 13 (5.1) 14 (3.9) 107 (11.8) 145 (11.4) 29 (7.9) 38 (5.9) Subtotal 570 (100.0) 873 (100.0) 253 (100.0) 357 (100.0) 903 (100.0) 1277 (100.0) 368 (100.0) 646 (100.0) a b c d e Control area. Nishiyama region. Direct exposure area on the near side of the hypocenter. Direct exposure area on the far side of the hypocenter. Age at the time of bombing. we performed Cox proportional hazard regression with adjustment where t denotes the time since commencement of follow-up; λ(t) for related factors to determine the mortality hazard ratio during and λ (t) denote the hazard rate and baseline hazard rate at t, the study period. We treated non-cancer disease death, survival to respectively; S = 1(male) or 0(female) denotes sex; A denotes age the end of the study period, and those who had moved out of ATB; G denotes area with G = 1 if the subject was exposed in the i 1 Nagasaki city as censoring. We evaluated the fallout effects on can- Nishiyama region and G = 0 otherwise; G = 1 if exposed in the 1 2 cer mortality by area adjusted by sex, age ATB, shielding condi- direct exposure area on the near side of the hypocenter and G = 0 tions (inside the house, outside shielded, or outside unshielded), otherwise; G = 1 if exposed in the direct exposure area on the far and entering into the hypocenter areas (~2 km around the hypo- side of the hypocenter and G = 0 otherwise, according to the sub- center within 3 days after the bombing) on the basis of the follow- ject’s location at the time of the bombing; P denotes shielding con- ing equation: ditions, with P = 1 (outside) or P = 0 (inside house), P = 1 1 1 2 (shielded) or P = 0 (not shielded) and P = 1 (unknown shielding 2 3 logλλ ()tt = log ( ) +βS +βA +βG +βG +βG 0 1 2 3 condition) or P = 0 (known shielding condition), according to the 12 3 4 5 ++ββP PPC +β +β , subject’s shielding from the explosion; C = 1 if the subject entered the 1 2 3 6 7 8 9 hypocenter area within the 3 days following the bombing and C = 0 Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Cancer mortality in the fallout area � 5 otherwise; and βs are unknown parameters to be estimated. We used hand, higher death rates were observed in the near side of the direct the PHREG procedure in the SAS system  for the calculations. exposure area compared with the control area. The RRs (95% CIs) of those 0–9, 10–19 and 20–29 years of age ATB were 1.20 (0.66–2.17), 1.41 (0.91–2.18) and 0.78 (0.42–1.46) in males, and 1.29 (0.55–2.97), 2.30 (1.33–3.95) and 1.07 (0.74–1.56) in females, respectively. RESULTS Among the 5247 subjects, a total of 549 cancer deaths were However, this tendency weakened in the far side of the direct exposure area. The RRs of those 0–9, 10–19 and 20–29 years observed between 1 January 1970 and 31 December 2012. Table 3 presents the site classiﬁcation of the observed cancer deaths among of age ATB were 1.04 (0.47–2.32), 1.35 (0.82–2.23) and 0.89 (0.45–1.76) in males, and 1.00 (0.31–3.23), 1.43 (0.75–2.73) the subjects. Deaths from cancers of the lung, stomach and colon were frequently observed in both males and females, but no deaths and 0.58 (0.36–0.95) in females, respectively. Death rates did not show consistent results in males. We performed Cox propor- were observed from cancers involving the thyroid, liver, melanoma, skin, ovary, prostate or bladder. To understand the overall picture tional regression analysis for cancer mortality to evaluate the effects of areas with adjustment factors, i.e. sex, age ATB, shield- of cancer deaths during the observation period, we calculated the cancer death rate by age ATB group (Table 4). In females, the can- ing condition and entering the hypocenter areas within the 3 days following the bombing. cer death rate per 100 000 population in the Nishiyama region was lower than that in the control area in every age ATB group, i.e. 43.1 Table 5 presents the hazard ratios (HRs) for cancer mortality in the Nishiyama region and the near and far sides of the direct expos- vs 103.8 (rate ratio [RR] = 0.41, 95% conﬁdence interval [95% CI] = 0.09–1.92) in those 0–9 years of age ATB, 127.4 vs 171.5 (RR = 0.74, ure area compared with in the control area. The HR for cancer mor- tality in the Nishiyama region compared with that in the control 95% CI = 0.27–2.01) in those 10–19 years of age ATB and 530.0 vs 540.4 (RR = 0.98, 95% CI = 0.55–1.74) in those 20–29 years of age area was 0.90 with a 95% conﬁdence interval (CI) of 0.65–1.24. No statistically signiﬁcant difference was observed in cancer mortality ATB. However, in males, the death rates in the Nishiyama region vs the control area in those 0–9, 10–19, and 20–29 years of age ATB between the Nishiyama region and the control area (P = 0.51). As for covariates, the hazard of cancer mortality for males was signiﬁ- were 371.1 vs 239.0 (RR = 1.55, 95% CI = 0.80–3.01), 385.8 vs 513.2 (RR = 0.75, 95% CI = 0.36–1.55) and 2105.3 vs 1256.5 (RR = 1.68, cantly higher (2.66-fold; 95% CI: 2.23–3.17) than that for females. The hazard of cancer mortality increased 1.08-fold (95% CI: 95% CI = 0.66–4.25), respectively. In males, the death rates in those 0–9and 20–29 years of age ATB in the Nishiyama region were higher 1.07–1.10) for each year of age ATB. This result reﬂects simple age- speciﬁc mortality in the Cox proportional hazard model; however, than those in the control area. Person-years in the Nishiyama area were fairly small (285) in males 20–29 years of age ATB. On the other those at a younger age ATB had a higher mortality risk . The Table 3. Site classiﬁcation of observed cancer deaths from 1970 to 2012 a b c d Sites CT NY R1 R2 Total (%) (ICD-10) Terrain-shielded Terrain-shielded Direct exposure Direct exposure area without area with fallout area, near side area, far side fallout Male Female Male Female Male Female Male Female Male Female Stomach (C16) 11 9 4 1 20 17 11 7 46 (16.6) 34 (12.5) Colon (C18) 4 12 1 2 7 13 5 5 17 (6.1) 32 (11.8) Rectal (C19–C20) 4 4 2 3 6 7 1 3 13 (4.7) 17 (6.3) Lung (C34) 5 5 6 1 25 20 19 7 55 (19.9) 33 (12.1) Breast (C50) 0 3 0 0 0 9 0 2 0 (0.0) 14 (5.1) Uterus (C53–C55) 2 0 6 2 10 (3.7) Myeloma (C90.0) 1 0 1 0 2 2 0 0 4 (1.4) 2 (0.7) Leukemia (C91–C93) 2 3 1 2 3 7 0 1 6 (2.2) 13 (4.8) Others 35 32 18 14 69 64 24 22 146 (52.7) 132 (48.5) Total 59 67 31 21 127 136 60 48 277 (100.0) 272 (100.0) Numbers of thyroid (C73), liver (C22), melanoma (C43), skin (C44), ovary (C56), prostate (C61) and bladder (C67) cancers were zero (0) in ‘Others’. Original data a b c were coded according to the ICD-9 or ICD-10; the ICD-10 codes are shown in this table. Control area. Nishiyama region. Direct exposure area on the near side of the hypocenter. Direct exposure area on the far side of the hypocenter. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 6 � K. Yokota et al. Table 4. Area-speciﬁc and age ATB–speciﬁc cancer death rates per 100 000 from 1970 to 2012 e a b c d Age ATB CT NY R1 R2 (years) Terrain-shielded area Terrain-shielded area with Direct exposure area, near side Direct exposure area, far without fallout fallout side (n = 1443) (n = 610) (n = 2180) (n = 1014) Male Female Male Female Male Female Male Female 0–9 239.0 103.8 371.1 43.1 285.8 133.5 248.8 103.3 (18/7530) (9/8673) (17/4581) (2/4645) (27/9447) (14/10 483) (9/3617) (4/3872) 10–19 513.2 171.5 385.8 127.4 723.7 393.8 691.2 245.2 (27/5261) (17/9913) (10/2592) (5/3924) (81/11 193) (56/14 222) (35/5064) (20/8157) 20–29 1256.5 540.4 2105.3 530.0 985.6 580.8 1118.9 315.0 (17/1353) (44/8142) (6/285) (16/3019) (24/2435) (75/12 914) (16/1430) (25/7937) Total 438.3 261.9 442.5 198.5 572.0 385.4 593.4 245.4 (62/14 144) (70/26 728) (33/7458) (23/11 588) (132/23 075) (145/37 619) (60/10 111) (49/19 966) a b c Data in parentheses are the number of deaths and observed person years. Control area. Nishiyama region. Direct exposure area on the near side of the hypocenter. d e Direct exposure area on the far side of the hypocenter. Age at the time of the bombing. Table 5. Hazard ratios for cancer mortality Areas compared Hazard ratio 95% Conﬁdence Interval P-value Nishiyama region vs control area 0.90 0.65–1.24 0.51 Direct exposure area on the near side of the hypocenter vs control area 1.28 1.04–1.58 0.02 Direct exposure area on the far side of the hypocenter vs control area 0.96 0.74–1.23 0.73 Cancer mortality was evaluated by area adjusted for sex, age at the time of bombing, the shielding condition, and entering into the hypocenter areas within the 3 days following the bombing. hazard of cancer mortality for those entering the hypocenter area direct exposure area were irradiated at a distance of 2.0–2.4 km from was 1.11-fold (95% CI: 0.91–1.36) higher than that for those not the hypocenter, with an estimated radiation dose of 3.8–273.1 mSv. On entering, but this difference was not signiﬁcant. On the other hand, the far side of the direct exposure area, subjects were irradiated at the hazard of cancer mortality in R1 was signiﬁcantly (P = 0.02) 2.5–3.9 km from the hypocenter, with an estimated dose of 1.9–20.9 mSv higher than that in the control area (HR = 1.28; 95% CI = 1.04– . The signiﬁcant increase in cancer mortality on the near side of the 1.58). The hazard of cancer mortality in R2, however, was not sig- direct exposure area seen in the present study is consistent with the niﬁcantly different from that in the control area (HR = 0.96; 95% results of a previous study . In contrast, no increase in cancer mortal- CI = 0.74–1.23). ity was observed on the far side of the direct exposure area, probably because of the comparatively lower dose of radiation in that area. Similarly, no increase in cancer mortality was observed in the Nishiyama DISCUSSION region (HR = 0.90; 95% CI = 0.65–1.24). No evidence of increased cancer mortality in the terrain-shielded Although a higher prevalence of solid thyroid nodules was area with radioactive fallout from the atomic bomb was found in the observed in Nishiyama residents in a previous study , no deaths present study. To conﬁrm the validity of the analysis, we simultan- caused by thyroid cancer were observed in the present study, prob- eously evaluated the effects of radiation exposure on cancer mortal- ably because thyroid cancer is not typically fatal. Immediately after ity in the direct radiation area without terrain shielding. A the bombing, a widespread residual radioactivity of ~7 μSv/h signiﬁcant increase in cancer mortality was observed in the direct (0.8 mR/h) or higher was reported in the Nishiyama region, and exposure area on the near side of the hypocenter (HR = 1.28; 95% radioactive hot spots of ~10–30 μSv/h around a reservoir were con- CI = 1.04–1.58), but no increase was observed on the far side (HR = ﬁrmed in previous surveys (Table 1); however, the radioactivity in 0.96; 95% CI = 0.74–1.23). Subjects who were on the near side of the the Nishiyama region decreased in the months after the bombing. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Cancer mortality in the fallout area � 7 The maximum cumulative external dose was estimated to be 68 R their children to keep them safe during the air raids. On the (~600 mSv)  in several surveys, while the majority of the estimates other hand, the shielded areas consisted of suburbs that had not in the Nishiyama region was in the range of 20–40 R (~170–350 been damaged in previous air raids. Fewer children aged 0–9 mSv). We note, however, that the estimates were based on the years old were thus in the direct exposure areas than in the shielded areas. Furthermore, children could not easily approach assumption that people were exposed to radiation outside for a life- time. The amount of radiation they actually received would be con- the hypocenter area immediatelyafter thebombing.These fac- siderably smaller than the estimates. tors, together with aspects related to the geography of the town, Although we restricted the study subjects to those aged <30 years meant that the proportion of people entering the hypocenter at the time of the bombing (reasons provided in ‘Study subjects’ in area in the Nishiyama region and the control area was lower than Materials and Methods), we conducted a similar analysis for those in the direct exposure areas. aged 30 years or over at the time of the bombing. The results pre- The accident at the Chernobyl nuclear power plant in 1986 sented in Tables 6 and 7 indicate no effects of radiation, even between released a huge amount of radioactive nuclides into the environ- the direct exposure area near the hypocenter and the control area (i.e. ment. Although the general population was exposed to the radi- the terrain-shielded area without fallout). We consider these results ation, there has been no persuasive evidence of any somatic effects support our decision to restrict the study subjects to atomic bomb sur- due to radiation exposure, except for signiﬁcant increases in thyroid vivors aged <30 years at the time of the atomic bombing. The propor- cancer or solid thyroid nodules among those who took milk con- tion of children in the direct exposure area was smaller than that in the taminated with I. Among more than 6000 children and adoles- Nishiyama region or the control area. Since the direct exposure areas cents diagnosed with thyroid cancers, only 15 deaths were observed were downtown and adjacent to the dockyards, which had previously by 2005 , and no increases in other cancer deaths have been been bombed in air raids, parents living downtown may have evacuated observed in the areas affected by the Chernobyl accident. Table 6. Area-speciﬁc and age ATB-speciﬁc cancer death rates per 100 000 from 1970 to 2012 among subjects who were 30 years and over of age ATB e a b c d Age ATB CT NY R1 R2 (years) Terrain-shielded area Terrain-shielded area Direct exposure area, near Direct exposure area, far without fallout with fallout side side (n = 685) (n = 243) (n = 1061) (n = 643) Male Female Male Female Male Female Male Female 30–39 1043.3 724.5 772.2 578.0 1296.5 930.3 1718.8 946.2 (13/1246) (41/5659) (2/259) (12/2076) (39/3008) (53/5697) (28/1629) (35/3699) 40–49 1826.0 1171.1 2427.2 1018.9 1473.4 1277.4 1584.2 831.2 (17/931) (31/2647) (5/206) (7/687) (31/2104) (42/3288) (24/1515) (13/1564) 50– 3973.5 1466.3 1156.1 1162.8 1025.6 463.7 1333.3 992.1 (6/151) (10/682) (2/173) (3/258) (4/390) (3/647) (4/300) (5/504) Total 1546.4 912.3 1410.7 728.2 1345.0 1017.4 1626.0 919.0 (36/2328) (82/8988) (9/638) (22/3021) (74/5502) (98/9632) (56/3444) (53/5767) a b c Data in parentheses are the number of deaths and observed person years. Control area. Nishiyama region. Direct exposure area on the near side of the hypocenter. d e Direct exposure area on the far side of the hypocenter. Age at the time of the bombing. Table 7. Hazard ratios for cancer mortality among subjects who were 30 years old and over of age at the time of bombing Areas compared Hazard ratio 95% Conﬁdence Interval P-value Nishiyama region vs control area 0.82 0.55–1.23 0.34 Direct exposure area on the near side of the hypocenter vs control area 1.07 0.84–1.36 0.58 Direct exposure area on the far side of the hypocenter vs control area 1.08 0.85–1.41 0.57 Cancer mortality was evaluated by area adjusted for sex, age at the time of bombing, the shielding condition, and entering into the hypocenter areas within 3 days after the bombing. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 8 � K. Yokota et al. Therefore, the results of the present study, suggesting no differ- CONFLICT OF INTEREST ence in the risk of cancer deaths between those exposed to radiation The authors declare that there are no conﬂicts of interest associated in the Nishiyama region and those in the control region would with this manuscript. likely not be caused by the small number of study subjects; the quantity of the radioactive materials generated by the atomic bomb FUNDING explosion was several hundred times smaller than that released to This work was supported by routine budget of the Atomic Bomb the environment due to the Chernobyl accident. Disease Institute, Nagasaki University. Following the accident at the Fukushima Daiichi Nuclear Power Station on 11 March 2011, the spatial radiation dose rate on 30 April 2011 at a point ~32 km north-west from the Fukushima Daiichi REFERENCES Nuclear Power Station was 11.57 μSv/h; this decreased to 5.69 μSv/h 1. Kerr GD, Solomon DL. The Epicenter of the Nagasaki Weapon: by 30 April 2012 . On 22 April 2012, the Japanese government a Reanalysis of Available Data with Recommended Values. Oak decided to establish an evacuation area for those areas where the Ridge, TN: Oak Ridge National Lab, 1976. ORNL-TM-5139. ambient dose rates still exceeded 3.80 μSv/h, which is equivalent to 2. Pace N, Smith RE. Measurement of the residual radiation inten- 20 mSv/y. Areas that had similar spatial radiation dose rates of 7 μSv/ sity at the Hiroshima and Nagasaki Atomic bomb sites. h or over in the Nishiyama region after the atomic bombing were Technical Report. Atomic Bomb Casualty Commission (ABCC) included in the evacuation area. Many residents of Fukushima were TR 26–59A, 1959. uneasy about possible future health effects resulting from exposure to 3. Manhattan Engineer District. Final Report of Findings of the the widespread radioactivity caused by the disaster. We hope that the Manhattan District Atomic Bomb Investigating Groups at Hiroshima results of the present study help to decrease anxiety among both and Nagasaki. Oak Ridge, TN: Manhattan Engineer District, 1946. Fukushima residents and atomic bomb survivors regarding the health 4. Arakawa ET. Radiation dosimetry in Hiroshima–Nagasaki effects associated with low-dose and temporary radiation exposure. atomic bomb survivors. New Eng J Med 1960;263:488–93. 5. Shinohara K, Morita Y, Kora K et al. Radiation of Ground in Nagasaki and Vicinity. II. Radioactivity near Nishiyama Reservoir. Limitations Collection of the Reports on the Investigation of the Atomic Bomb The present study had two limitations. First, since the follow-up of Casualties, Vol. 1. Tokyo: Japan Society for the Promotion of Nagasaki University Atomic Bomb Survivors Cohort members Science. 1953, 45–53 (in Japanese). started on 1 January 1970 , information on deaths was not 6. Shono N. Physical effects of the A-Bomb in Hiroshima and available for those who had died or had moved out of Nagasaki city. Nagasaki: amount of radiation received by A-bomb victims. Although the present analysis was restricted to those who were J Hiroshima Med Assoc 1959;12:1041–51 (in Japanese). younger than 30 years of age ATB, the results may have been biased 7. Okajima S. Dose estimation from residual and fallout radioactiv- if the contribution to cancer mortality of such individuals was not ity. Fallout in the Nagasaki–Nishiyama district. J Radiat Res negligible. 1975;16:35–41. Second, our study was ecologic due to that no individual dose esti- 8. Okajima S, Takeshita K, Antoku S et al. Radioactive fallout effects mate was available for Nishiyama residents; the dose rates presented in of the Nagasaki atomic bomb. Health Phys 1978;34:621–33. Table 1 are unique information available for us. The Dosimetry System 9. Nagataki S, Hirayu H, Izumi M et al. High prevalence of thyroid 2002 (DS02)  provided us with the direct radiation doses under nodule in area of radioactive fallout. Lancet 1989;8659:385–6. the terrain-shielding conditions; the control area described in the pre- 10. Sakata R, Grant EJ, Furukawa K et al. Long-term effects of the sent study as being shielded by a 288 m ridge was estimated to have rain exposure shortly after the atomic bombings in Hiroshima received 3–20 mSv (Gy). Since the height of the ridge that shielded and Nagasaki. Radiat Res 2014;182:599–606. the Nishiyama region from direct exposure to the atomic bomb radi- 11. Okajima S, Mine M, Nakamura T. Mortality of registered A- ation is higher (366 m) than the ridge that shielded the control area, bomb survivors in Nagasaki, Japan, 1970–1984. Radiat Res direct radiation doses in the Nishiyama region would have been lower 1985;103:419–31. than those in the control area. Further studies to estimate the individ- 12. Honda S, Mine M, Okumura Y et al. Estimation of radiation ual dose in Nishiyama residents using the available information such as doses for Nagasaki atomic bomb survivors based on ABS93D. the location of their residence at the time of the bombing and the mea- Nagasaki Med J 1997;72:30–46 (in Japanese). surements of residual radiation presented in Table 1 are necessary. 13. ESRI. ArcGIS Desktop: Release 10.2.2. Redlands, CA: Environmental Systems Research Institute; 2014. http://www. ACKNOWLEDGEMENTS esri.com/software/arcgis/arcgis-for-desktop/ (28 April 2017, Administrative and medical data on atomic bomb survivors were date last accessed). obtained with the cooperation of the Nagasaki city and Nagasaki 14. GSI. The Base Map Information, 10 m Mesh Numerical Altitude Prefectural governments. We thank the technical staff members of Model. Tokyo: Geographical Survey Institute, 2009. http://fgd. the Division of Scientiﬁc Data Registry, Atomic Bomb Disease gsi.go.jp/download/ (28 April 2017, date last accessed). Institute, Nagasaki University, for daily maintenance of the database 15. Nagasaki City. The Nagasaki Atomic Bomb Damage Records, General and the data preparation for this study. Analysis Version Vol. 1. Nagasaki: Nagasaki city, 2016, 126–31. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Cancer mortality in the fallout area � 9 16. Ozasa K, Shimizu Y, Suyama A et al. Studies of the mortality of Chernobyl accident. Report to the general assembly scientiﬁc Atomic Bomb Survivors, Report 14, 1950–2003: an overview of annex D. UNSCEAR 2008 Report Volume II. UNSCEAR, 2011. cancer and noncancer diseases. Radiat Res 2012;177:229–43. 20. Fukushima Prefecture. Fukushima Prefecture Radioactivity 17. SAS Institute Inc. SAS User’s Guide: Statistics Version 9, 4th edn. Measurement Map. http://fukushima-radioactivity.jp/ (28 April Cary, NC: SAS Institute Inc, 2014. 2017, date last accessed). 18. Okajima S, Fujita S, Harley JH. Radiation doses from residual radio- 21. Stephen DE, Dean CK, James AR et al. Survivor shielding activity. In: Roesh WC (ed). US–Japan Joint Reassessment of Atomic Part C. Improvements in terrain shielding. In: Young RW, Bomb Radiation Dosimetry in Hiroshima and Nagasaki,Vol.1. Kerr GD (eds). Reassessment of the Atomic Bomb Radiation Hiroshima: Radiation Effects Research Foundation, 1987, 205–26. Dosimetry for Hiroshima and Nagasaki-Dosimetry System 2002. 19. United Nations Scientiﬁc Committee on the Effects of Atomic Vol. 2. Hiroshima: Radiation Effects Research Foundation, Radiation (UNSCEAR). Health effects due to radiation from the 2005, 821. Downloaded from https://academic.oup.com/jrr/article-abstract/59/1/1/4209301 by Ed 'DeepDyve' Gillespie user on 16 March 2018
Journal of Radiation Research – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera