Key Points
Question
Does the risk of a second primary malignant neoplasm after radioactive iodine (RAI) treatment for differentiated thyroid cancer vary by age?
Findings
This cohort study using data from the US National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program for 1975 to 2021 included 72 412 participants and found that RAI treatment was associated with elevated myeloma risk in individuals aged 15 to 44 years. Among adults aged 45 to 64 years, risks were elevated for prostate cancer, salivary gland cancer, and nodal non-Hodgkin lymphoma, while adults 65 years and older had elevated risks of stomach, esophagus, and nonepithelial skin cancers, and acute myeloid leukemia.
Meaning
The age-specific patterns in RAI-associated risk of a second primary malignant neoplasm highlight the need for tailored surveillance strategies in survivors of differentiated thyroid cancer.
Abstract
Importance
Radioactive iodine treatment for differentiated thyroid cancer has been associated with second primary malignant neoplasms, but age-specific risks remain poorly understood.
Objective
To evaluate the association of age with increased risk of second primary malignant neoplasm after radioactive iodine treatment for differentiated thyroid cancer (DTC).
Design, Setting, and Participants
This retrospective cohort study used data from 8 cancer registries in the US National Cancer Institute’s Surveillance, Epidemiology, and End Results Program for 1975 to 2021. Patients who received radioactive iodine treatment for DTC were identified; those with fewer than 2 years of follow-up, missing follow-up data, or distant metastases at diagnosis were excluded. Included patients were categorized into 3 age groups: 15 to 44, 45 to 64, and 65 years or older. Among 5-year survivors, median (IQR) follow-up time was 14.7 (8.6-24.0) years in the 15- to 44-year age group; 8.3 (6.4-11.4) years in the 45- to 64-year group; and 9.3 (6.9-12.7) years in the 65 years or older group. Data were analyzed from May to July 2024.
Exposure
Radioactive iodine vs no radioactive iodine.
Main Outcomes and Measures
Relative risks (RR) and 95% CIs for solid and hematologic second primary malignant neoplasms, calculated using multivariable Poisson regression models adjusted for age, sex, and latency.
Results
The study sample comprised 72 412 patients with nonmetastatic DTC, of whom 28 432 (39%) were age 15 to 44 years; 34 009 (47%), age 45 to 64 years; and 9971 (14%), 65 years or older. In the 15- to 44-year age group, radioactive iodine was associated with an increased risk of hematologic cancers (RR, 1.35; 95% CI, 1.02-1.80), specifically myeloma (RR, 4.22; 95% CI, 1.68-10.62). In the 45- to 64-year age group, the analyses showed increased risks of prostate cancer (RR, 1.61; 95% CI, 1.10-2.37), salivary gland cancer (RR, 10.22; 95% CI, 1.27-82.24), and nodal non-Hodgkin lymphoma (RR, 2.81; 95% CI, 1.34-5.89). The overall risk of a solid second primary malignant neoplasm was not elevated (RR, 0.94; 95% CI, 0.76-1.16), but hematologic second primary malignant neoplasm risk was elevated (RR, 1.73; 95% CI, 1.14-2.60). Patients in the group that was 65 years and older showed elevated risks of stomach (RR, 4.06; 95% CI, 1.05-15.81), esophagus (RR, 11.42; 95% CI, 1.40-93.3), nonepithelial skin cancers (RR, 10.52; 95% CI, 1.09-88.77), and acute myeloid leukemia (RR, 3.26; 95% CI, 1.15-9.24). The overall risk of a solid second primary malignant neoplasm was elevated (RR, 1.88; 95% CI, 1.59-2.21), whereas risk of hematologic malignant neoplasm was not (RR, 1.35; 95% CI, 0.97-1.87).
Conclusions and Relevance
This cohort study found that the risk of radioactive iodine−associated second primary malignant neoplasm among patients who received radioactive iodine for DTC varies significantly by age group, with distinct patterns in middle-aged compared with older adults. These findings suggest a need for age-specific surveillance strategies in radioactive iodine−treated DTC survivors. Further research is needed to establish optimal screening protocols.
This cohort study evaluates the association of age with increased risk of second primary malignant neoplasm after radioactive iodine treatment for differentiated thyroid cancer.
Introduction
The incidence of thyroid cancer has risen steadily over the past several decades, more than doubling from 5.0 to 13.6 cases per 100 000 persons between 1975 and 2021.1 Notably, most cases (63.5%) occur in adults age 45 years and older, with 40.7% of cases in those age 45 to 64 years, and 22.8% in those 65 years or older.1 This substantial burden in older adults demands age-specific understanding of treatment-related consequences. Differentiated thyroid cancer (DTC) accounts for 94% of thyroid cancers, and high-risk cases require treatment with thyroidectomy followed by radioactive iodine (RAI).2,3 Although RAI treatment has demonstrated survival benefits,4 it has been associated with several adverse effects, including gastrointestinal symptoms, xerostomia, gonadal damage, and most concerningly, an increased risk of second primary malignant neoplasms (SPMs).5,6
The relationship between RAI and adverse effects, particularly SPMs, has prompted research into its long-term safety profile. Current literature presents conflicting evidence, with some studies identifying clear associations between RAI and specific SPMs, while others report no significant association.7,8,9,10,11,12 This inconsistency poses a critical challenge for clinical decision-making, particularly for older adults. A fundamental limitation of existing research is the lack of age-stratified analyses, despite compelling evidence that both radiation sensitivity and remaining life expectancy—key factors in SPM development—vary significantly with age.13,14,15,16 The absence of age-specific risk profiles leaves a knowledge gap that potentially allows for suboptimal treatment decisions regarding adults aged 45 years and older, who constitute the majority of thyroid cancer cases.
We hypothesized that age-related differences in radiation sensitivity and remaining lifespan would produce distinct SPM risk profiles. Such findings could inform both treatment decisions and cancer screening for individuals with DTC. Therefore, our study investigated whether older age is associated with an increased risk of second primary malignant neoplasm after radioactive iodine treatment for DTC.
Methods
The Icahn School of Medicine at Mount Sinai Program for the Protection of Human Subjects deemed this study to be exempt from review because it used only deidentified and publicly available data (STUDY-23-01598); informed consent was waived for the same reason. The study followed the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
Study Design and Population
This retrospective cohort study used methods described by Pasqual et al14 who previously assessed RAI-associated SPM risks in patients younger than 45 years of age using data from the US National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program for 1975 to 2017. We applied the analyses to older adults to evaluate the association between RAI treatment and SPM risk specifically and to enable direct comparisons of SPM risk patterns across 3 age groups (15 to 44, 45 to 64, and 65 years or older).
Our data were collected from 8 SEER cancer registries, which together account for 8.3% of the US population: 3 metropolitan areas (Atlanta, Georgia; San Francisco-Oakland; California; and Seattle-Puget Sound, Washington) and 4 states (Hawaii, Iowa, New Mexico, and Utah). For the period from 1975 to 2021, we identified 28 432 individuals aged 15 to 44 years; 34 009 adults aged 45 to 64 years; and 9971 adults age 65 years and older who were diagnosed with nonmetastatic DTC. All patients were diagnosed with a primary papillary or follicular carcinoma. Race data were derived from the database records and were coded per the options in the SEER database: Black, White, or other, which included American Indian/Alaska Native and Asian/Pacific Islander.
Patients were stratified into 2 groups by RAI treatment status: received or not received/unknown. Received status included radioisotopes (1988+ [per SEER terminology]), treatment other than beam radiation (1973-1987 only), and radiation, method not otherwise specified (NOS) or source NOS. Not received/unknown status included none or unknown, refused (1988+), and recommended, unknown if administered. Patients who received beam radiation before 1988 or other radiation/radiation NOS after 1988 were excluded.
Surveillance for hematologic and solid SPMs began 2 and 5 years after DTC diagnosis, respectively, to account for the differing latency periods of radiation carcinogenesis.13,14,15 Patients with fewer than 2 years of follow-up, missing follow-up data, or distant metastases at time of diagnosis were excluded. The outcomes of interest were solid malignant neoplasms (excluding thyroid cancer) and hematologic malignant neoplasms. Surveillance ended at time of SPM diagnosis, death, last follow-up, or December 31, 2021, whichever occurred first. The variable, months since index calculated, was used to find the time of SPM diagnosis; and the variable, person time years calculated, was used to find the time of last follow-up or death. These follow-up intervals were used to calculate person-time at risk, which was incorporated into the Poisson regression model through a log offset term and used to estimate expected SPM counts based on SEER incidence rates stratified by age, sex, and calendar year.
Statistical Analyses
Descriptive statistics were performed to describe the study population. We calculated effect sizes appropriately (Cohen h, Cohen d, and Cramer V) for continuous and categorical variables. To assess whether age modified the effect of RAI on SPM risk, we formally tested an interaction between RAI and age group in our Poisson regression model using a type III Wald test. The interaction term was significant (χ22 = 15.43; P < .001), indicating that the association between RAI and SPM risk varied significantly across the 3 age groups. Given the strength of this interaction, we conducted age-stratified analyses.
From the SEER 8 data, we found the observed number of SPMs by cancer site. Using SEER cancer incidence rates, we calculated the expected number of SPMs by site in our study population, stratified by age and sex. Then, we calculated relative risks (RR) and 95% CIs of solid and hematologic SPMs according to RAI receipt status using multivariable Poisson regression models adjusted for age at DTC diagnosis, sex, and latency. The model was also adjusted for attained age by using the log of the expected number of cases—calculated from SEER incidence rates by 10-year age groups—as an offset, allowing estimation of SPM risk per unit of person-time.14
In each cohort, we calculated crude SPM incidence rates for the RAI-treated and untreated groups (observed cases divided by person-time years). The absolute excess number of SPMs attributable to RAI was then obtained by multiplying the rate difference between the 2 groups by the total person-years of the RAI cohort. 95% CIs were derived under a Poisson assumption and applied to the rate difference before scaling to counts. All analyses were performed from May to July 2024 using SEER*Stat, version 8.4.3, and SAS, version 9.4 (SAS Institute Inc).
Results
Age 15 to 44 Years Group
We identified 28 432 individuals between the ages of 15 and 44 years diagnosed with nonmetastatic DTC from 1975 to 2021. This group was composed of 23 314 female (82.0%) and 5118 male (18.0%) participants, including 1297 Black (4.6%), 23 353 White (82.1%), and 3782 individuals of other race (13.3%). Of these patients, 26 746 were 2-year survivors and 26 723 were 5-year survivors. The demographic characteristics of this group are described in eTable 1 in Supplement 1.
Second Primary Malignant Neoplasms
Among the 26 746 patients who were 2-year survivors, 178 hematologic SPMs were observed over the surveillance period. The overall risk of hematologic SPMs was significantly higher in the RAI group (RR, 1.35; 95% CI, 1.02-1.80; eFigure in Supplement 1). Specifically, the risk of myeloma increased (RR, 4.22; 95% CI, 1.68-10.62).
Among the 26 723 patients who were 5-year survivors, 1661 solid SPMs were observed over the surveillance period. The overall risk of solid SPMs was not significantly higher for RAI-treated patients (RR, 1.00; 95% CI, 0.90 to 1.10). There were no elevated risks of specific solid SPMs in this age group. We estimated that 6.59 (95% CI, −1.17 to 14.34) excess myeloma cases were attributable to RAI over the follow-up period in this cohort.
Age 45 to 64 Years Group
We identified 34 009 adults from 45 to 64 years diagnosed with nonmetastatic DTC from 1975 to 2021. This group was composed of 25 847 female (76.0%) and 8162 male (24.0%) participants, including 1449 Black (4.3%), 27 887 White (82.0%), and 4673 individuals of other race (13.7%). Of these patients, 23 799 were 2-year survivors and 19 695 were 5-year survivors. Of the 2-year survivors, 75.0% (17 889 of 23 799) were female and 43% (10 250 of 23 799) were treated with RAI (Table 1). Among 5-year survivors, RAI was used more often in male (2926 of 5910; 50%) than in female patients (7324 of 17 889; 41%) (Cramer V = 0.25; 95% CI, 0.22-0.28).
Table 1. Characteristics of 2- and 5-Year Survivors of Differentiated Thyroid Cancer Aged 45 to 64 Years, by Radioactive Iodine Treatment (RAI) Status, SEER 8.
| Characteristic | Received RAI, No. (%) | |||||
|---|---|---|---|---|---|---|
| 5-y Survivors | 2-y Survivors | |||||
| Yes | No/unknown | Effect size (95% CI) | Yes | No/unknown | Effect size (95% CI) | |
| Total | 8772 (45) | 10 923 (55) | 0.25 (0.22 to 0.28) | 10 250 (43) | 13 549 (57) | 0.28 (0.25 to 0.30) |
| Sex | ||||||
| Female | 6390 (43) | 8604 (57) | 0.25 (0.22 to 0.28) | 7324 (41) | 10 565 (59) | 0.09 (0.07 to 0.10) |
| Male | 2382 (51) | 2319 (49) | 2926 (50) | 2984 (50) | ||
| Age at diagnosis, y | ||||||
| 45 to <50 | 2761 (47) | 3138 (53) | 0.03 (0.019 to 0.047) | 3319 (45) | 4023 (55) | 0.05 (0.04 to 0.06) |
| 50 to <55 | 2483 (44) | 3111 (56) | 3007 (43) | 3926 (57) | ||
| 55 to <60 | 2010 (44) | 2594 (56) | 2296 (41) | 3275 (59) | ||
| 60 to <65 | 1518 (42) | 2080 (58) | 1628 (41) | 2325 (59) | ||
| Racea | ||||||
| Black | 398 (40) | 594 (60) | 0.07 (0.05 to 0.08) | 500 (39) | 766 (61) | 0.07 (0.05 to 0.08) |
| White | 6996 (43) | 9104 (57) | 8147 (42) | 11 165 (58) | ||
| Other | 1378 (53) | 1225 (47) | 1603 (50) | 1618 (50) | ||
| Attained age at SPM, y | ||||||
| 45 to <50 | NA | NA | 0.02 (0.010 to 0.04) | 1 (100) | 0 | 0.03 (0.02 to 0.04) |
| 50 to <55 | 47 (44) | 59 (56) | 23 (74) | 8 (26) | ||
| 55 to <60 | 184 (53) | 164 (47) | 28 (54) | 24 (46) | ||
| 60 to <65 | 246 (42) | 340 (58) | 34 (52) | 31 (48) | ||
| Time to SPM, median (IQR), mo | 102 (80 to 134) | 102 (78 to 142) | 0.00002 (−0.03 to 0.03) | 75 (45 to 122) | 79 (40 to 116) | −0.07 (−0.12 to −0.03) |
| Follow-up time, median (IQR), y | 8.34 (6.5 to 11.0) | 8.33 (6.3 to 11.7) | 0.003 (−0.03 to 0.03) | 6.08 (3.6 to 10.0) | 6.42 (3.2 to 9.5) | −0.07 (−0.12 to −0.03) |
| Follow-up time, y | ||||||
| 2 to <5 | NA | NA | 0.10 (0.08 to 0.11) | 1577 (36) | 2763 (64) | 0.13 (0.11 to 0.14) |
| 5 to <10 | 3140 (43) | 4081 (57) | 3079 (44) | 3995 (56) | ||
| 10 to <20 | 4273 (49) | 4447 (51) | 4235 (49) | 4396 (51) | ||
| ≥20 | 1359 (36) | 2395 (64) | 1359 (36) | 2395 (64) | ||
| Time period of diagnosis, y | ||||||
| 1975-2009 | 4400 (44) | 5676 (56) | 0.04 (0.02 to 0.05) | 4782 (44) | 6012 (56) | 0.10 (0.09 to 0.12) |
| 2009-2015 | 3278 (47) | 3732 (53) | 3437 (47) | 3902 (53) | ||
| 2015-2021 | 1094 (42) | 1515 (58) | 2031 (36) | 3635 (64) | ||
| Tumor histologic findings | ||||||
| Papillary | 879 (52) | 816 (48) | 0.05 (0.03 to 0.06) | 9099 (42) | 12 495 (58) | 0.07 (0.06 to 0.09) |
| Follicular | 7430 (44) | 9550 (56) | 1065 (52) | 991 (48) | ||
| Unknown | 463 (45) | 557 (55) | 86 (58) | 63 (42) | ||
| Thyroid cancer stage per TNM | ||||||
| Size <4 cm and no lymph node involvement | 3157 (37) | 5442 (63) | 0.32 (0.30 to 0.33) | 3330 (37) | 5743 (63) | 0.35 (0.34 to 0.36) |
| Size ≥4 cm or lymph node involvement | 2554 (75) | 855 (25) | 2764 (76) | 891 (24) | ||
| Unknown | 3061 (40) | 4626 (60) | 4156 (38) | 6915 (62) | ||
Abbreviations: NA, not applicable; SEER, US National Cancer Institute’s Surveillance, Epidemiology, and End Results Program database, eighth release (1975-2017); SPM, second primary malignant neoplasm; TNM, tumor, nodule, and metastases staging per the American Joint Committee on Cancer Staging Manual, eighth edition.
Race categories are reported as coded in the SEER database; other includes American Indian/Alaska Native and Asian/Pacific Islander.
Second Primary Malignant Neoplasms
Among the 23 799 patients who were 2-year survivors, 149 hematologic SPMs were observed over the surveillance period. The overall risk of hematologic SPMs was significantly higher (RR, 1.73; 95% CI, 1.14-2.60) in the RAI group (Figure 1). Specifically, the risk of nodal non-Hodgkin lymphoma was increased (RR, 2.81; 95% CI, 1.34-5.89).
Figure 1. Solid and Hematologic Malignant Neoplasms in Adults Aged 45 to 64 Years, Adjusted for Age at Differentiated Thyroid Cancer (DTC) Diagnosis, Sex,a and Latency.
NHL indicates non-Hodgkins lymphoma; RAI, radioactive iodine treatment; SPM, second primary malignant neoplasms.
aProstate relative risk analysis was restricted to male patients, and ovary and corpus uteri cancer analyses were restricted to female patients.
Among the 19 695 patients who were 5-year survivors, 1040 solid SPMs were observed over the surveillance period. The overall risk of solid SPMs was not significantly higher for patients treated with RAI (RR, 0.94; 95% CI, 0.76-1.16). However, risks were significantly higher for 2 specific solid cancers, prostate (RR, 1.61; 95% CI, 1.10-2.37) and salivary gland (RR, 10.22; 95% CI, 1.27-82.24). Conversely, the risk of brain cancer was significantly decreased in patients treated with RAI (RR, 0.21; 95% CI, 0.06-0.74).
Of note, all 9 other endocrine cancers, 6 acute lymphocytic leukemia, and 2 Hodgkin-nodal cases occurred in the RAI group vs 0 in the non-RAI group, whereas all 16 vulvar cancers and 4 acute myeloid leukemia cases occurred in the non-RAI group vs 0 in the RAI group. No significantly increased risks were found for 21 other solid and 4 other hematologic cancers. We estimated that 24.66 (95% CI, 6.95 to 42.37) excess prostate cancers, 7.29 (95% CI, 4.22 to 37.33) salivary gland cancers, and 7.86 (95% CI, −2.00 to 17.72) nodal non-Hodgkin lymphoma cases were attributable to RAI in this group.
Age 65 Years or Older Group
We identified 9971 adults aged 65 years and older diagnosed with nonmetastatic DTC from 1975 to 2021. This group was composed of 6936 female (69.6%) and 3035 male (30.4%) participants, including 442 Black (4.4%), 8135 White (81.6%), and 1394 individuals of other race (14.0%). Of these patients, 8115 were 2-year survivors and 5908 were 5-year survivors. Of the 2-year survivors, 70% (5715 of 8115) were female patients, and only 38% (3107 of 8115) were treated with RAI, compared to the 43% in the younger cohort (Table 2). RAI use was highest in patients diagnosed between ages 65 and 69 years (1533 of 3648; 42%) compared to those diagnosed after age 85 years (90 of 345; 26%) (Cramer V = 0.14; 95% CI, 0.12-0.17). Similar patterns were observed in 5-year survivors.
Table 2. Characteristics of 2- and 5-Year Survivors of Differentiated Thyroid Cancer Aged 65 Years or Older, by Radioactive Iodine Treatment (RAI) Status, SEER 8.
| Characteristic | Received RAI, No. (%) | |||||
|---|---|---|---|---|---|---|
| 5-y Survivors | 2-y Survivors | |||||
| Yes | No/unknown | Effect size (95% CI) | Yes | No/unknown | Effect size (95% CI) | |
| Total | 2360 (40) | 3548 (60) | 0.41 (0.35 to 0.46) | 3107 (38) | 5008 (62) | 0.47 (0.43 to 0.52) |
| Sex | ||||||
| Female | 1626 (38) | 2674 (62) | 0.04 (0.01 to 0.06) | 2056 (36) | 3659 (64) | 0.07 (0.05 to 0.09) |
| Male | 734 (46) | 874 (54) | 1051 (44) | 1349 (56) | ||
| Age at diagnosis, y | ||||||
| 65 to <70 | 1213 (44) | 1564 (56) | 0.11 (0.09 to 0.14) | 1533 (42) | 2115 (58) | 0.14 (0.12 to 0.17) |
| 70 to <75 | 639 (39) | 983 (61) | 837 (38) | 1378 (62) | ||
| 75 to <80 | 335 (36) | 585 (64) | 451 (35) | 829 (65) | ||
| 80 to <85 | 122 (30) | 285 (70) | 196 (31) | 431 (69) | ||
| ≥85 | 51 (28) | 131 (72) | 90 (26) | 255 (74) | ||
| Racea | ||||||
| Black | 71 (31) | 156 (69) | 0.08 (0.05 to 0.10) | 109 (32) | 237 (68) | 0.07 (0.05 to 0.09) |
| White | 1891 (39) | 2971 (61) | 2485 (37) | 4176 (63) | ||
| Other | 398 (49) | 421 (51) | 513 (46) | 595 (54) | ||
| Attained age at SPM, y | ||||||
| 65 to <70 | NA | NA | 0.03 (0.005 to 0.06) | 5 (63) | 3 (37) | 0.05 (0.02 to 0.07) |
| 70 to <75 | 54 (47) | 60 (53) | 22 (65) | 12 (35) | ||
| 75 to <80 | 88 (38) | 143 (62) | 14 (33) | 29 (67) | ||
| 80 to <85 | 95 (44) | 121 (56) | 14 (39) | 22 (61) | ||
| ≥85 | 75 (36) | 133 (64) | 7 (23) | 24 (77) | ||
| Time to SPM, median (IQR), mo | 111 (85 to 149) | 107 (80 to 146) | 0.08 (0.03 to 0.14) | 76.5 (47 to 99) | 80.5 (57 to 149) | −0.07 (−0.12 to −0.03) |
| Follow-up time, median (IQR), y | 9.5 (7.1 to 12.7) | 9.1 (6.6 to 12.6) | 0.09 (0.04 to 0.15) | 6.2 (3.8 to 8.1) | 6.6 (4.6 to 12.3) | −0.09 (−0.13 to −0.04) |
| Follow-up time, y | ||||||
| 2 to <5 | NA | NA | 0.05 (0.02 to 0.07) | 766 (34) | 1488 (66) | 0.07 (0.05 to 0.09) |
| 5 to <10 | 1150 (39) | 1764 (61) | 1140 (39) | 1750 (61) | ||
| 10 to <20 | 1056 (42) | 1471 (58) | 1047 (42) | 1458 (58) | ||
| ≥20 | 154 (33) | 313 (67) | 154 (33) | 312 (67) | ||
| Time period of diagnosis, y | ||||||
| 1975-2009 | 1276 (39) | 2007 (61) | 0.03 (0.004 to 0.06) | 1496 (39) | 2389 (61) | 0.06 (0.04 to 0.08) |
| 2009-2015 | 818 (42) | 1128 (58) | 919 (42) | 1268 (58) | ||
| 2015-2021 | 266 (39) | 413 (61) | 692 (34) | 1351 (66) | ||
| Tumor histologic findings | ||||||
| Papillary | 1991 (39) | 3062 (61) | 0.03 (0.001 to 0.05) | 2620 (38) | 4306 (62) | 0.02 (0.0008 to 0.04) |
| Follicular | 369 (43) | 486 (57) | 487 (41) | 702 (59) | ||
| Thyroid cancer stage per TNM | ||||||
| Size <4 cm and no lymph node involvement | 721 (31) | 1637 (69) | 0.32 (0.30 to 0.34) | 797 (30) | 1883 (70) | 0.29 (0.27 to 0.31) |
| Size ≥4 cm or lymph node involvement | 735 (71) | 299 (29) | 962 (68) | 457 (32) | ||
| Unknown | 904 (36) | 1612 (64) | 1348 (34) | 2668 (66) | ||
Abbreviations: NA, not applicable; SEER, US National Cancer Institute’s Surveillance, Epidemiology, and End Results Program database, eighth release (1975-2017); SPM, second primary malignant neoplasm; TNM, tumor, nodule, and metastases staging per the American Joint Committee on Cancer Staging Manual, eighth edition.
Race categories are reported as coded in the SEER database; other includes American Indian/Alaska Native and Asian/Pacific Islander.
Among the 8115 patients who were 2-year survivors, 152 hematologic SPMs were observed during the surveillance period. Among the 5908 patients who were 5-year survivors, 769 solid SPMs were observed over the surveillance period. The overall risk of solid SPMs was significantly higher for patients who received RAI vs patients who did not (RR, 1.88; 95% CI, 1.59-2.21), whereas the risk of hematologic SPMs was not significantly different between groups (RR, 1.35; 95% CI, 0.97-1.87; Figure 2).
Figure 2. Solid and Hematologic Malignant Neoplasms in Adults Aged 65 Years or Older, Adjusted for Age at Differentiated Thyroid Cancer (DTC) Diagnosis, Sex,a and Latency.
NHL indicates non-Hodgkins lymphoma; NOS, not otherwise specified; RAI, radioactive iodine treatment; SPM, second primary malignant neoplasms.
aProstate relative risk analysis was restricted to male patients, and ovary and corpus uteri cancer analyses were restricted to female patients.
SPM risks were significantly increased specifically for stomach (RR, 4.06; 95% CI, 1.05-15.81), esophagus (RR, 11.42; 95% CI, 1.40-93.3), nonepithelial skin cancers (RR, 10.52; 95% CI, 1.09-88.77), and acute myeloid leukemia (RR, 3.26; 95% CI, 1.15-9.24). We did not find increased risks of 20 other solid and 9 other hematologic cancers. We estimated that 5.07 (95% CI, −0.12 to 10.26) excess stomach, 6.36 (95% CI, 1.17 to 11.54) esophageal, 5.36 (95% CI, 0.56 to 10.16) nonepithelial skin cancers, and 5.14 (95% CI, −1.50 to 11.78) acute myeloid leukemia cases were attributable to RAI in this group.
Discussion
Our findings contribute to the growing body of evidence regarding age-specific risks of RAI treatment. RAI treatment for DTC has been associated with an elevated risk of SPMs in the general population as well as in pediatric and young adults.4,8,10,14,17 Our study found that different age groups had varying risks of SPMs at different sites. Among adults aged 45 to 64 years, the overall risk of solid SPMs was not elevated, while the overall risk of hematologic SPMs was elevated. In contrast, individuals aged 15 to 44 years and 65 years and older showed the opposite pattern. Given that both life expectancy and radiation sensitivity decline with age, different age groups likely experience different patterns of SPM risks,18,19 and it is crucial to analyze risks separately across different age groups.
It is interesting to compare our results to a research article by Pasqual et al14 assessing the RR of SPMs after RAI treatment in pediatric and young adults (aged younger than 45 years; eTable 2 in Supplement 1). They found significantly increased risks of all solid cancers, cancers of the corpus uteri, and all hematologic cancers, including leukemia and nonlymphocytic leukemia. Specifically, they reported an increased risk of solid malignant neoplasms (RR, 1.23; 95% CI, 1.11-1.37) and an increased risk of hematologic malignant neoplasms (RR, 1.51; 95% CI, 1.08-2.01) in patients treated with RAI. Their findings were particularly strong in patients younger than 25 at diagnosis, with an RR of 1.6 (95% CI, 1.07-2.40) for solid cancers, and showed the highest risks after 20 years of follow-up (RR, 1.47; 95% CI, 1.24-1.74).
In our updated analysis of SEER-8 data through 2021, we reanalyzed data for adults aged 15 to 44 years using methods identical to those used by Pasqual et al.14 We did not observe elevated risks for any individual solid malignant neoplasms, and the overall RR for solid SPMs in this group was also not elevated. While Pasqual et al14 reported increased leukemia risk, the SEER database now separates leukemia into subtypes, which limited direct comparison. Furthermore, Pasqual et al14 did not report an elevated myeloma risk (RR, 2.36; 95% CI 0.7-7.9) unlike our study (RR, 4.22; 95% CI, 1.68-10.62). The extra 4 years of follow-up in our study may have allowed more SPMs to accrue, as there were 11 total myeloma cases in the study by Pasqual et al14 compared to 20 in our updated analysis. Both studies had wide confidence intervals due to low counts. Nonetheless, both studies consistently found an overall elevated risk of hematologic SPMs in patients younger than 45 years treated with RAI (RR, 1.34; 95% CI, 1.68-10.62).
In contrast, our study showed no significant overlap in specific solid SPM sites. While Pasqual et al14 found an elevated risk of uterine cancer in pediatric and young adults, our updated analysis did not find any significantly elevated risks for specific solid malignant neoplasms in the 15- to 44-years group. The discrepancy in the results may be due to several factors. First, while Pasqual et al14 assessed data of all patients younger than 45 years, our analysis restricted the analysis to ages 15 to 44 years to focus on the adult population. Pasqual et al14 used SEER 9 (1975-2017), whereas we used SEER 8 (1975-2021), extending the analysis by 4 additional years. However, SEER 8 lacks data from Detroit, which is included in SEER 9; therefore, these differences in time period and geographic coverage may have influenced the results.
Regarding hematologic malignant neoplasms, both studies observed elevated risks in the youngest cohort, although with different patterns. Pasqual et al14 observed increased risks in all hematologic cancers including leukemia and nonlymphocytic leukemia, while our 15- to 44-year group showed an increased risks of hematologic malignant neoplasms overall, and a significantly elevated risk of myeloma. We also found age-specific patterns: an increased risk of nodal non-Hodgkin lymphoma in the middle-aged group and acute myeloid leukemia in the older group. These differences reinforce the importance of age-stratified analyses in understanding RAI-associated cancer risks.
The middle-aged group (45-64 years) showed increased risks specifically in prostate and salivary gland cancers, and our older cohort (≥65 years) demonstrated elevated risks in stomach, esophageal, and nonepithelial skin cancers. These findings are consistent with known patterns of radiation dose absorption given that both the salivary glands and stomach exhibit early active iodine uptake.20 Although the skin is not typically considered a high-dose organ in standard RAI dosimetry, a recent SEER-based study by Rezaei et al21 also found an increased risk of melanoma and other nonkeratinocyte skin cancers after RAI, particularly in the head and neck region, suggesting a possible association despite unclear biological mechanisms.
The increased risk of prostate cancer that we found in the age group 45 to 64 years has been previously reported.10 This may be partly due to detection bias because patients who received RAI treatment may be more likely to undergo routine screening. Additionally, molecular pathways implicated in both thyroid and prostate cancer, such as RET protooncogene amplification22,23 and the CHEK2 protein kinase mutation,24,25 could contribute. We also found an increased risk of salivary gland cancer, which is consistent with previous findings in the general population as well as pediatric and young adults younger than 25 years.17,26,27 This pattern may be due to Na+/I− symporter expression driving RAI accumulation in the salivary gland.28
We observed an increased risk of nodal non-Hodgkin lymphoma in adults aged 45 to 64 years, but not in those 65 years and older. While the association between radiation and lymphomas remains unclear, our results suggest this risk may be age-dependent and stronger for non-Hodgkin lymphomas in particular.29,30 All cases of other endocrine cancers, acute lymphocytic leukemia, and nodal Hodgkin cases occurred in the RAI group. All vulvar cancers and acute myeloid leukemia cases occurred in the non-RAI group. Due to the small number of cases of these rare cancers, these results may be due to random variation rather than a true association, which reflects a limitation in our study.
Among adults aged 65 years and older, the risks of rarer cancers (<43 cases per 100 000), including cancers of the stomach, esophagus, nonepithelial skin, and acute myeloid leukemia, were elevated. The higher risks we observe in this group most likely reflects the greater baseline cancer incidence that accompanies advancing age.31 Previous studies in the general population have also reported a significant association between RAI and leukemias,10,11,26,27 particularly acute myeloid leukemia.8,32,33 Although we did not observe enough cases to calculate the RR for acute myeloid leukemia in the 45- to 64-year age group, its increased risk has been well documented and warrants further investigation. We also observed an elevated risk of stomach cancer, consistent with previous reports.17,34 Similar to the salivary gland, the stomach expresses high concentrations of Na+/I− symporters and absorb high doses of radiation, which may account for this increased risk.17,35,36 The increased risks of other rare cancers are more difficult to explain and may be partly due to detection bias.
Strengths and Limitations
This study had notable strengths. The large sample size and total person-time years calculated provide a more comprehensive view of the long-term risks associated with RAI treatment for DTC. By stratifying our patients by age, we were able to assess how the risks of SPMs vary across different age groups. This is particularly important in identifying age-dependent patterns that may not have been apparent in studies without age stratification.
Our study also has several limitations that could bias the observed associations. First, many confounders, such as smoking, comorbidities, concomitant medications, and prior exposure to environmental or therapeutic radiation, were unavailable in SEER. These factors affect both baseline cancer risk as well as the decision to administer RAI.37,38,39,40 For instance, a history of radiation therapy raises the underlying risk of SPMs yet may discourage additional RAI, thereby underestimating the true RAI effect. In contrast, chronic kidney disease often leads clinicians to withhold RAI, but if RAI is given, impaired iodine clearance can prolong exposure,40 potentially overestimating risk. Because such mechanisms operate in opposite directions, the net bias is difficult to estimate.
Second, patients who were treated with RAI often had more aggressive disease pathology based on tumor, node, and metastases staging (eg, larger tumors or lymph node involvement), but had sufficient functional status to tolerate RAI treatment.41 Greater disease severity could raise underlying SPM risk, whereas better functional status could lower it, biasing the association in the opposite direction.
Third, SEER does not capture RAI dose, an important factor that could affect SPM risk. Previous studies have indicated that the risk of SPMs increases with higher doses of RAI,7 suggesting that the lack of dose-specific data in our study may limit our understanding of the dose-response relationship. The SEER registry has inherent known limitations, including missing data, potential reporting biases, and patients lost to follow-up, which could produce underreporting of SPMs and subsequent inaccuracies in observed RRs. Additionally, patients treated with RAI may have received more intensive follow-up care, potentially allowing enhanced detection of SPMs and higher observed RRs.
Many cancers with significant associations in our study were rare, and although our sample size was relatively large, the overall number of hematologic cancers was small, limiting the statistical power for detecting subtle associations for these rarer cancers. Several of the statistically significant RRs observed in this study had confidence intervals with lower bounds close to 1.0, suggesting that some associations, while statistically significant, may not be clinically meaningful. Therefore, these findings should be interpreted with caution in the context of potential clinical relevance.
Due to these limitations and the observational nature of the study, we cannot establish a causal relationship between RAI and SPMs. Our findings reflect associations that may be influenced by both measured and unmeasured confounders.
Conclusions
This cohort study found that RAI treatment appears to be associated with an elevated risk of different SPMs depending on age, suggesting that DTC survivors should be monitored with a focus on age-specific risks for SPMs. These findings have important clinical implications; clinicians should consider the patient’s age at the time of DTC diagnosis and RAI treatment when evaluating their potential risks of developing SPMs. This information can guide decisions on appropriate screenings for specific malignant neoplasms, especially rarer cancers that are not usually captured by traditional screening. Additionally, our findings support the need for tailoring RAI use to individual risk profiles. By considering factors such as age and potential comorbidities, clinicians can make more informed decisions that minimize the long-term risks associated with RAI treatment.
Several important questions remain unanswered and warrant further investigation. Primary areas requiring further research include assessing how different RAI doses affect age-specific SPM risk, understanding genetic susceptibility to specific RAI-associated SPMs, and evaluating the cost-effectiveness of age-specific surveillance protocols. With this additional knowledge, clinicians would be better enabled to optimize RAI treatment protocols and improve long-term outcomes for DTC survivors through more personalized risk assessment and monitoring strategies.
eTable 1. Characteristics of Differentiated Thyroid Cancer Patients Aged 15-44 Years in SEER 8 Database
eFigure. Relative Risks (RR) and 95% Confidence Intervals for Solid and Hematologic Malignancies in Individuals Aged 15-44, Adjusted for Age At DTC Diagnosis, Sex, and Latency
eTable 2. Relative Risks of Second Primary Malignancies by Site in Different Age Groups
Data Sharing Statement
References
- 1.Cancer Stat Facts . Thyroid Cancer. Accessed November 6th, 2024. https://seer.cancer.gov/statfacts/html/thyro.html
- 2.Fagin JA, Wells SA Jr. Biologic and clinical perspectives on thyroid cancer. N Engl J Med. 2016;375(23):2307. doi: 10.1056/NEJMra1501993 [DOI] [PubMed] [Google Scholar]
- 3.Cabanillas ME, McFadden DG, Durante C. Thyroid cancer. Lancet. 2016;388(10061):2783-2795. doi: 10.1016/S0140-6736(16)30172-6 [DOI] [PubMed] [Google Scholar]
- 4.Wu W, Li S, Xu K, Meng Z. Hazard ratios of second primary malignancy after radioiodine for differentiated thyroid carcinoma: a large-cohort retrospective study. Endokrynol Pol. 2023;74(3):260-270. doi: 10.5603/EP.a2023.0028 [DOI] [PubMed] [Google Scholar]
- 5.Fard-Esfahani A, Emami-Ardekani A, Fallahi B, et al. Adverse effects of radioactive iodine-131 treatment for differentiated thyroid carcinoma. Nucl Med Commun. 2014;35(8):808-817. doi: 10.1097/MNM.0000000000000132 [DOI] [PubMed] [Google Scholar]
- 6.Lee SL. Complications of radioactive iodine treatment of thyroid carcinoma. J Natl Compr Canc Netw. 2010;8(11):1277-1286. doi: 10.6004/jnccn.2010.0094 [DOI] [PubMed] [Google Scholar]
- 7.Khang AR, Cho SW, Choi HS, et al. The risk of second primary malignancy is increased in differentiated thyroid cancer patients with a cumulative (131)I dose over 37 GBq. Clin Endocrinol (Oxf). 2015;83(1):117-123. doi: 10.1111/cen.12581 [DOI] [PubMed] [Google Scholar]
- 8.Molenaar RJ, Sidana S, Radivoyevitch T, et al. Risk of hematologic malignancies after radioiodine treatment of well-differentiated thyroid cancer. J Clin Oncol. 2018;36(18):1831-1839. doi: 10.1200/JCO.2017.75.0232 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Seo GH, Kong KA, Kim BS, et al. Radioactive iodine treatment for children and young adults with thyroid cancer in South Korea: a population-based study. J Clin Endocrinol Metab. 2021;106(7):e2580-e2588. doi: 10.1210/clinem/dgab192 [DOI] [PubMed] [Google Scholar]
- 10.Teng CJ, Hu YW, Chen SC, et al. Use of radioactive iodine for thyroid cancer and risk of second primary malignancy: a nationwide population-based study. J Natl Cancer Inst. 2015;108(2):djv314. doi: 10.1093/jnci/djv314 [DOI] [PubMed] [Google Scholar]
- 11.Yu CY, Saeed O, Goldberg AS, et al. A systematic review and meta-analysis of subsequent malignant neoplasm risk after radioactive iodine treatment of thyroid cancer. Thyroid. 2018;28(12):1662-1673. doi: 10.1089/thy.2018.0244 [DOI] [PubMed] [Google Scholar]
- 12.Zhang Y, Liang J, Li H, Cong H, Lin Y. Risk of second primary breast cancer after radioactive iodine treatment in thyroid cancer: a systematic review and meta-analysis. Nucl Med Commun. 2016;37(2):110-115. doi: 10.1097/MNM.0000000000000419 [DOI] [PubMed] [Google Scholar]
- 13.US National Research Council of the National Academies . Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2. 2006. Accessed July 21, 2025. https://nap.nationalacademies.org/catalog/11340/health-risks-from-exposure-to-low-levels-of-ionizing-radiation
- 14.Pasqual E, Schonfeld S, Morton LM, et al. Association between radioactive iodine treatment for pediatric and young adulthood differentiated thyroid cancer and risk of second primary malignancies. J Clin Oncol. 2022;40(13):1439-1449. doi: 10.1200/JCO.21.01841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Berrington de González A BA, Rajaraman P, Schubauer-Berigan M, Thun M, Linet MS, Cerhan JR, et al. Cancer epidemiology and prevention, 4th edition. Accessed July 21, 2025. https://www.scribd.com/document/384238857/Cancer-Epidemiology-and-Prevention-4th-Edition
- 16.Yasui Y, Liu Y, Neglia JP, et al. A methodological issue in the analysis of second-primary cancer incidence in long-term survivors of childhood cancers. Am J Epidemiol. 2003;158(11):1108-1113. doi: 10.1093/aje/kwg278 [DOI] [PubMed] [Google Scholar]
- 17.Corrêa NL, de Sá LV, de Mello RC. Estimation of second primary cancer risk after treatment with radioactive iodine for differentiated thyroid carcinoma. Thyroid. 2017;27(2):261-270. doi: 10.1089/thy.2016.0266 [DOI] [PubMed] [Google Scholar]
- 18.Brenner DJ, Hall EJ. Computed tomography–an increasing source of radiation exposure. N Engl J Med. 2007;357(22):2277-2284. doi: 10.1056/NEJMra072149 [DOI] [PubMed] [Google Scholar]
- 19.Chodick G, Ronckers CM, Shalev V, Ron E. Excess lifetime cancer mortality risk attributable to radiation exposure from computed tomography examinations in children. Isr Med Assoc J. 2007;9(8):584-587. [PubMed] [Google Scholar]
- 20.Mattsson S, Johansson L, Leide Svegborn S, et al. ; ICRP . Radiation dose to patients from radiopharmaceuticals: a compendium of current information related to frequently used substances. Ann ICRP. 2015;44(2)(suppl):7-321. doi: 10.1177/0146645314558019 [DOI] [PubMed] [Google Scholar]
- 21.Rezaei SJ, Chen ML, Kim J, John EM, Sunwoo JB, Linos E. Development of melanoma and other nonkeratinocyte skin cancers after thyroid cancer radiation. JAMA Netw Open. 2024;7(9):e2434841. doi: 10.1001/jamanetworkopen.2024.34841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Desilets A, Repetto M, Yang SR, Sherman EJ, Drilon A. RET-altered cancers-a tumor-agnostic review of biology, diagnosis and targeted therapy activity. Cancers (Basel). 2023;15(16):4146. doi: 10.3390/cancers15164146 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Salvatore D, Santoro M, Schlumberger M. The importance of the RET gene in thyroid cancer and therapeutic implications. Nat Rev Endocrinol. 2021;17(5):296-306. doi: 10.1038/s41574-021-00470-9 [DOI] [PubMed] [Google Scholar]
- 24.Brock P, Liynarachchi S, Nieminen TT, et al. CHEK2 founder variants and thyroid cancer risk. Thyroid. 2024;34(4):477-483. doi: 10.1089/thy.2023.0529 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hale V, Weischer M, Park JY. CHEK2 (∗) 1100delC mutation and risk of prostate cancer. Prostate Cancer. 2014;2014:294575. doi: 10.1155/2014/294575 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Marti JL, Jain KS, Morris LG. Increased risk of second primary malignancy in pediatric and young adult patients treated with radioactive iodine for differentiated thyroid cancer. Thyroid. 2015;25(6):681-687. doi: 10.1089/thy.2015.0067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Iyer NG, Morris LG, Tuttle RM, Shaha AR, Ganly I. Rising incidence of second cancers in patients with low-risk (T1N0) thyroid cancer who receive radioactive iodine therapy. Cancer. 2011;117(19):4439-4446. doi: 10.1002/cncr.26070 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Dohán O, De la Vieja A, Paroder V, et al. The sodium/iodide Symporter (NIS): characterization, regulation, and medical significance. Endocr Rev. 2003;24(1):48-77. doi: 10.1210/er.2001-0029 [DOI] [PubMed] [Google Scholar]
- 29.Harbron RW, Pasqual E. Ionising radiation as a risk factor for lymphoma: a review. J Radiol Prot. 2020;40(4). doi: 10.1088/1361-6498/abbe37 [DOI] [PubMed] [Google Scholar]
- 30.Hsu WL, Preston DL, Soda M, et al. The incidence of leukemia, lymphoma and multiple myeloma among atomic bomb survivors: 1950-2001. Radiat Res. 2013;179(3):361-382. doi: 10.1667/RR2892.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.US Centers for Disease Control and Prevention . 2025 US Cancer Statistics. https://www.cdc.gov/united-states-cancer-statistics/publications/uscs-highlights.html
- 32.Little MP, Wakeford R, Borrego D, et al. Leukaemia and myeloid malignancy among people exposed to low doses (<100 mSv) of ionising radiation during childhood: a pooled analysis of nine historical cohort studies. Lancet Haematol. 2018;5(8):e346-e358. doi: 10.1016/S2352-3026(18)30092-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Teepen JC, Curtis RE, Dores GM, et al. Risk of subsequent myeloid neoplasms after radiotherapy treatment for a solid cancer among adults in the United States, 2000-2014. Leukemia. 2018;32(12):2580-2589. doi: 10.1038/s41375-018-0149-2 [DOI] [PubMed] [Google Scholar]
- 34.Lee Y, Chung CH, Lin LF, et al. Radioactive iodine treatment for thyroid cancer patients increases the risk of long-term gastrointestinal disorders: a nationwide population-based cohort analysis. Cancers (Basel). 2022;14(10):2505. doi: 10.3390/cancers14102505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bruno R, Giannasio P, Ronga G, et al. Sodium iodide symporter expression and radioiodine distribution in extrathyroidal tissues. J Endocrinol Invest. 2004;27(11):1010-1014. doi: 10.1007/BF03345302 [DOI] [PubMed] [Google Scholar]
- 36.Johansson L, Leide-Svegborn S, Mattsson S, Nosslin B. Biokinetics of iodide in man: refinement of current ICRP dosimetry models. Cancer Biother Radiopharm. 2003;18(3):445-450. doi: 10.1089/108497803322285206 [DOI] [PubMed] [Google Scholar]
- 37.Newman SK, Patrizio A, Boucai L. Decision variables for the use of radioactive iodine in patients with thyroid cancer at intermediate risk of recurrence. Cancers (Basel). 2024;16(17):3096. doi: 10.3390/cancers16173096 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Morris LG, Sikora AG, Tosteson TD, Davies L. The increasing incidence of thyroid cancer: the influence of access to care. Thyroid. 2013;23(7):885-891. doi: 10.1089/thy.2013.0045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Verburg FA, Hänscheid H, Luster M. Radioactive iodine (RAI) therapy for metastatic differentiated thyroid cancer. Best Pract Res Clin Endocrinol Metab. 2017;31(3):279-290. doi: 10.1016/j.beem.2017.04.010 [DOI] [PubMed] [Google Scholar]
- 40.Mayson SE, Chan CM, Haugen BR. Tailoring the approach to radioactive iodine treatment in thyroid cancer. Endocr Relat Cancer. 2021;28(10):T125-T140. doi: 10.1530/ERC-21-0161 [DOI] [PubMed] [Google Scholar]
- 41.Papaleontiou M, Banerjee M, Yang D, Sisson JC, Koenig RJ, Haymart MR. Factors that influence radioactive iodine use for thyroid cancer. Thyroid. 2013;23(2):219-224. doi: 10.1089/thy.2012.0380 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Characteristics of Differentiated Thyroid Cancer Patients Aged 15-44 Years in SEER 8 Database
eFigure. Relative Risks (RR) and 95% Confidence Intervals for Solid and Hematologic Malignancies in Individuals Aged 15-44, Adjusted for Age At DTC Diagnosis, Sex, and Latency
eTable 2. Relative Risks of Second Primary Malignancies by Site in Different Age Groups
Data Sharing Statement
