Fraction of Cancer Attributable to Carcinogenic Drugs in Korea from 2015 to 2030

Woojin Lim; Soseul Sung; Youjin Hong; Sungji Moon; Sangjun Lee; Kyungsik Kim; Jung Eun Lee; Inah Kim; Kwang-Pil Ko; Sue K. Park

doi:10.4143/crt.2024.644

Articles

Page Path: HOME > Cancer Res Treat > Volume 57(3); 2025 > Article

Original Article General Fraction of Cancer Attributable to Carcinogenic Drugs in Korea from 2015 to 2030: Woojin Lim^1,2,3, Soseul Sung^1,2,3, Youjin Hong^1,2,4, Sungji Moon^1,2,5, Sangjun Lee^1,2,4, Kyungsik Kim^1,2, Jung Eun Lee⁶, Inah Kim⁷, Kwang-Pil Ko⁸, Sue K. Park^1,2,4; Cancer Research and Treatment : Official Journal of Korean Cancer Association 2025;57(3):635-648.
DOI: https://doi.org/10.4143/crt.2024.644
Published online: November 6, 2024

¹Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Korea

²Cancer Research Institute, Seoul National University, Seoul, Korea

³Department of Biomedical Sciences, Seoul National University Graduate School, Seoul, Korea

⁴Integrated Major in Innovative Medical Science, Seoul National University Graduate School, Seoul, Korea

⁵Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine, Seoul, Korea

⁶Department of Food and Nutrition, Seoul National University College of Human Ecology, Seoul, Korea

⁷Department of Occupational and Environmental Medicine, Hanyang University College of Medicine, Seoul, Korea

⁸Clinical Preventive Medicine Center, Seoul National University Bundang Hospital, Seongnam, Korea

Correspondence: Sue K. Park, Department of Preventive Medicine, Seoul National University College of Medicine, 103 Daehak-ro, Jongno-gu, Seoul 03080, Korea
Tel: 82-2-740-8338 E-mail: suepark@snu.ac.kr

• Received: June 28, 2024 • Accepted: November 5, 2024

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 unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

3,396 Views
183 Download

prev next

Full Article

Download PDF

Abstract

Purpose
This study aims to estimate and project the population attributable fraction (PAF) of cancer incidence and death due to carcinogenic drug use in Korea from 2015 to 2030, to estimate the degree of cancer prevention from exposure to carcinogenic drugs in Korea. Selected carcinogenic drugs were immunosuppressive and antineoplastic drugs classified as group I by the International Agency for Research on Cancer.
Materials and Methods
Systematic review and meta-analyses were conducted to estimate the relative risk of cancer associated with carcinogenic drug use. Age was standardized using the annual prevalence rate of the National Health Insurance Service sample cohort (NHIS-NSC) from 2002 to 2013 to calculate the standardized prevalence rate of carcinogenic drug use each year. The PAF of specific cancer incidence and death were calculated using Levin’s formula and Monte Carlo methods. The prevalence rates were extrapolated to estimate the trend of PAF from 2015 to 2030.
Results
In 2015, carcinogenic drugs attributed to 0.003% and 0.002% among the causes of cancer incidence and death in Korea. However, carcinogenic drugs attributed to 1.1% among the causes of both cancer incidence and death in patients with clinical indications of carcinogenic drugs.
Conclusion
The PAF in patients with clinical indications of carcinogenic drugs were significantly high and expected to increase rapidly over time. Since these drugs are listed as essential by the World Health Organization, and may be difficult to replace, a surveillance system on susceptible populations using group I carcinogenic drugs must be discussed and implemented.
Key words: Population attributable fraction, Attributable count, Carcinogenic drugs

Introduction

Since its establishment in 1965, the International Agency for Research on Cancer (IARC), under the World Health Organization, has been conducting research to collect evidence on the risk factors of cancer in humans. Since the publication of its first 16-volume monograph series in 1971, IARC has been continuously updating the contents. Out of the monograph series published by IARC since 1971, a total of 5 were solely about the risk of cancer from various drugs, emphasizing the importance of assessing the cancer risk from these risk factors [1-5].

In the IARC monograph, factors were categorized into groups according to the sufficiency of evidence regarding their carcinogenicity in humans, and each factor’s risk level and need for additional research were provided. Among them, factors with sufficient evidence for causing cancer in humans are defined as group I.

These group I carcinogenic drugs consist of immunosuppressants used for organ transplants, prevention of chronic allograft rejection, and inflammatory bowel diseases (Crohn’s disease, ulcerative colitis). Group I drug also consist of alkylating agents, and anti-cancer drugs used alone or in combination for various solid tumors, hematologic cancers, polycythemia vera, or lupus. In addition, ointments used in combination with ultraviolet therapy for various skin diseases like psoriasis or vitiligo (Methoxsalen), analgesics (Phenacetin), and anti-inflammatory drugs used as astringents for edema and persistent diarrhea (aristolochic acid) are also defined as group I. Long-term or cumulative use of these drugs has been reported in the IARC Monograph to cause hematologic cancers such as lymphoma and leukemia, as well as skin cancer, urinary bladder cancer, lung cancer, and kidney cancer [4,5].

In Korea, the National Cancer Center (NCC) reported the attributable fractions of cancer incidence and death in 2009 due to exposure to risk factors in 1990. In addition, the attributable fractions of cancer incidence and death caused by obesity and physical inactivity was reported in 2014 [6]. However, these reports did not calculate the attributable fractions of cancer incidence and death caused by carcinogenic drugs. Moreover, no previous studies have calculated the prevalence rate of group I carcinogenic drug use within the general population.

Therefore, this study aims to estimate the degree of cancer prevention due to group I carcinogenic drug exposure by estimating the population attributable fraction (PAF) of cancer incidence and death from 2015 to 2030. The findings of this study can be used to establish novel health care guidelines and national cancer prevention policies for patients with indications of group I carcinogenic drugs, providing a basis for further research.

Materials and Methods

1. Risk factors

From monographs on various type of drugs, 24 group I drugs were identified. Of these, five, which include diethylstilbestrol, chlornaphazine, phenacetine, a composition with phenacetine, and semustine (methyl-CCNU), were either prohibited worldwide or set aside for further scrutiny on their carcinogenicity. Four drugs related to hormones, such as combined estrogen-progestogen menopausal therapy, combined estrogen-progestogen oral contraceptives, tamoxifen, and postmenopausal estrogen therapy, were omitted as distinct monographs already covered their carcinogenic aspects. To solely evaluate synthetic drugs, three plant-based medicines—aristolochic acid, plants associated with aristolochic acid, and opium—were not considered. Consequently, after excluding 12 drugs, the research focused on following 12 drugs: three immunosuppressive drugs (cyclosporine, azathioprine, and cyclophosphamide), nine antineoplastic drugs (cyclophosphamide, busulfan, melphalan, chlorambucil, thiotepa, treosulfan, mustargen-oncovin-procarbazine-prednisone [MOPP], bleomycin-etoposide-cisplatin [BEP], and etoposide), and one skin ointment (methoxsalen).

2. Clinical indications and outcomes

The clinical indications selected in this study are diseases confirmed in the IARC monographs, and the outcomes selected in this study are cancers reported in the IARC monographs as having sufficient evidence of associations with group I carcinogenic drug use (S1 Table). For outcomes, the International Classification of Diseases, 10th revision, clinical modification (ICD-10-CM) code was used to identify types of cancer. C67 was used for urinary bladder cancer, C43-C44 were used for skin cancer including melanoma and non-melanoma, C81-C96 were used for hematologic cancer including lymphoma, leukemia, multiple myeloma, and other types of hematopoietic malignancies, and C34 were used for lung cancer.

3. Relative risk estimation

The relative risk (RR) of cancer incidence and death associated with group I carcinogenic drug use couldn’t be determined through raw data analysis, as there were no existing cohort studies or consortium using Korean population [7].

Hence, a literature search was conducted to verify the association between group I carcinogenic drugs and specific cancers proposed by the IARC. Specific search terms for each group I pharmaceutical in the database were used to identify studies that met the inclusion criteria (S2 Table). A comprehensive literature search using both PubMed (MEDLINE) and Embase were conducted. To enhance the sensitivity of search strategy in PubMed, keywords based on Medical Subject Headings (MeSH) terms were used.

After including studies on the association between each carcinogenic drugs and cancer, the RR of cancer derived from each study was meta-analyzed using a random effects model to calculate the integrated risk (summary RR, SRR).

4. Meta-analysis

To calculate the SRR on the incidence of each cancer type, all studies were analyzed in the same manner. The heterogeneity between studies were assessed with Cochran’s Q test and presented as Higgins I² value. Publication bias was evaluated with Begg and Egger test, and the risk of bias was evaluated using the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) and Cochrane’s Risk of Bias (RoB) tool. Subgroup analyses were conducted based on type of clinical indication, study region, publication period, and study design to identify reasons for heterogeneity. Heterogeneity was defined significant if p-value was less than 0.05. In addition, to identify the robustness of the results, sensitivity analysis was conducted with influential meta-analysis, using the inverse variance and DerSimonian-Laird method.

5. Prevalence rate estimation

In this study, 15-year latency period was used for cancer incidence or death due to group I carcinogenic drug exposure. Therefore, it was designed to calculate the attributable fractions of cancer incidence or death in 2015 based on carcinogenic drug exposure in 2000.

There were no references for the prevalence rate of carcinogenic drug use in Korea. Therefore, past and current prevalence rates of carcinogenic drug use within Korean population were determined using the National Health Insurance Service (NHIS) sample cohort (NHIS-NSC). NHIS-NSC allows rate calculation from one million people stratified by sex, age, subscriber category, insurance fee level, and region, who are maintaining the qualifications of health insurance subscribers and medical aid beneficiaries for a year in 2006. Among 12 selected group I drugs, five of the drugs were excluded (MOPP, etoposide, BEP, treosulfan, thiotepa), since they had no main component codes (MOPP, BEP, treosulfan) in the NHIS data or since it was unable to calculate prevalence rate within the NHIS-NSC cohort (etoposide and thiotepa).

The annual prevalence rates by sex among Koreans aged 20 and over were calculated for five carcinogenic drugs (cyclosporine, azathioprine, cyclophosphamide, methoxsa-len, melphalan) which had main component codes (generic name codes) within the NHIS-NSC database. However, for drugs rarely used in Korea, or drugs whose use declined and ceased before 2006, or other drugs for extremely rare indications, the prevalence rates were estimated based on a systematic review. Eventually, the annual prevalence rates for seven carcinogenic drugs (cyclosporine, azathioprine, cyclophosphamide, methoxsalen, melphalan, busulfan, chlorambucil) were confirmed within NHIS-NSC, and included in the study.

To estimate PAF for each group I drug to estimate PAF of total group I drug included in this study, the prevalence rate of each group I drug was calculated through identifying individual prescription records (existence of each record of main component codes) within the NHIS-NSC cohort.

Standardized prevalence rate for annual carcinogenic drug use was calculated by age-standardizing the annual prevalence rate of the NHIS-NSC of Korea from 2002 to 2013, using the mid-year population of the Korean resident registration in 2000 as the standard population. The most adequate regression model was selected using all annual age-standardized rates to estimate the prevalence rate in 2000. The prevalence rate of carcinogenic drugs was estimated using a linear regression model.

6. PAF calculation

Cancer incidence rate, mortality rate, the number of cancer case, and the number of cancer death were derived from the 2015 cancer registry statistics of the NCC.

The optimal exposure causing the minimum cancer risk for group I carcinogenic drug use was defined as ‘non-exposure’, and the attributable fractions for specific cancer incidence and specific cancer death was calculated.

The attributable fractions (%) for specific cancer incidence were calculated using Levin’s formula, and its 95% confidence intervals (CI) were calculated using Monte Carlo methods [8].

(1)

(PAF=∑Pe(RR−1)∑Pe(RR−1)+1)

The calculated attributable fractions for specific cancer incidence imply the number of specific cancer incidence (attributable count, AC) attributable to carcinogenic drug exposure, and by adding all the AC for specific cancer incidence, the total AC of cancer incidence due to carcinogenic drug exposure was calculated. The attributable fractions of cancer incidence due to carcinogenic drug exposure were calculated by dividing the total cancer AC by the total number of cancer incidence. The attributable fraction of cancer death due to exposure to carcinogenic drugs was calculated using the risk and number of specific cancer death, the same methods used in the process of calculating the attributable fractions of cancer incidence.

In addition, the fraction attributable to carcinogenic drugs on cancer incidence and death in patients with clinical indications of group I carcinogenic drugs was also calculated. Sensitivity analysis was performed to observe the difference of PAF by region, using RR of Korean and global studies (excluding Korea), and by study design, using RR of cohort and case-control studies.

7. PAF projection

The prevalence rates up to 2030 were extrapolated using the same method used to calculate the prevalence rate in 2000. For the number of cancer incidence and death, the specific cancer incidence and mortality rate by age and sex at 5-year intervals for adults aged 20 and over from 2000 to 2017 (for cancer incidence) and 2000 to 2018 (for cancer death) were used, and then transformed to the standardization rate of 2000. The average annual percentage change (AAPC) of the standardized incidence and mortality rate and a join point regression model (x=year, Y=Ln (Rate)) was used to estimate the number of cancer incidence and death up to 2030. The U.S. National Cancer Institute’s Join point program was used for the above analysis (ver. 4.7.0.0, http://surveillance.cancer.gov/joinpoint).

The join point regression (x=year, Y=Ln(Rate)) was calculated using

(2)

log⁡(Y)=β0−β1X,

the AAPC using

(3)

AAPC=(eβ1−1)×100,

and the expected number of incidence and death using

(4)

Id=∑j=xyIR2000,j×N2020,j,

and

(5)

Dd=∑j=xyDR2000,j×N2020,j,

respectively [9].

The standardized cancer incidence and mortality rates for specific cancers by year and sex up to 2030 were calculated using the number of populations by sex and age in the respective year and the estimated number of incidence and death for specific cancers by sex and age. The total population in the respective year was calculated using the age-sex population trend data from the National Statistics of Korea. The change in attributable fractions up to 2030 was confirmed using the annual prevalence rate and the estimated annual number of cancer incidence and death. In addition, the fraction attributable to carcinogenic drugs on cancer incidence and death in patients with clinical indications of group I carcinogenic drugs was also projected up to 2030.

Results

1. RR estimation

A total of 14 studies on the association between cyclosporine and skin cancer [10-23], and azathioprine and skin cancer [10,12-16,18-20,22,24-27] were included in the meta-analysis, followed by 12 studies on the association between azathioprine and hematologic cancer [10,24,28-37]. In addition, seven studies on the association between cyclosporine and hematologic cancer were included in the meta-analysis [10,11,28-32], followed by six studies on the association between cyclophosphamide and hematologic cancer [10,34, 38-41], four studies on the association between cyclophosphamide and bladder cancer [10,38,42,43], two studies on the association between methoxsalen and skin cancer [10,44], and two studies on the association between melphalan [10,39], chlorambucil [10,41], busulfan [45,46] and hematologic cancer (Table 1).

All carcinogenic drugs were found to increase the risk of cancers, except for cyclosporine and hematologic cancer. In addition, RR of specific cancers were relatively higher than other cancers as previously reported [47]. The risk of skin cancer due to cyclosporine compared to non-exposure was high (RR, 1.44; 95% CI, 1.15 to 1.81), but the risk of hematologic cancer was found to have no significant association (RR, 1.05; 95% CI, 0.62 to 1.76). The risk of skin cancer due to azathioprine compared to non-exposure was high (RR, 1.81; 95% CI, 1.38 to 2.36), and the risk of hematologic cancer was also high (RR, 1.73; 95% CI, 1.18 to 2.55). The risk of bladder cancer due to cyclophosphamide compared to non-exposure was very high (RR, 2.77; 95% CI, 1.99 to 3.86), and the risk of hematologic cancer was also very high (RR, 2.85; 95% CI, 1.72 to 4.71). The risk of skin cancer due to methoxsalen compared to non-exposure was also observed to be very high (RR, 2.97; 95% CI, 1.25 to 7.01). The risk of hematologic cancer due to melphalan compared to non-exposure was observed to be extremely high (RR, 7.54; 95% CI, 1.49 to 38.23). The risk of hematologic cancer due to chlorambucil compared to non-exposure was also observed to be very high (RR, 3.43; 95% CI, 2.50 to 4.70). The risk of hematologic cancer due to busulfan compared to non-exposure was observed to be extremely high (RR, 6.71; 95% CI, 2.49 to 18.08) (Table 1). The RR of each study by study design is presented in S3-S7 Tables.

2. Prevalence rate estimation

In 2000, the prevalence rate of cyclosporine was 0.003%, while the prevalence rate of azathioprine was 0.005%. The prevalence rate of cyclophosphamide was 0.006%. The prevalence rate of methoxsalen was 0.034%, while the prevalence rate of melphalan was 0.0004%. The prevalence rate of chlorambucil was 0.0001%, and for busulfan, it was 0.0005%. All carcinogenic pharmaceuticals showed decreasing trend over time, except melphalan, which showed a slight increasing trend (S8 Table, S9 Fig.).

3. PAF calculation

Table 2 indicates that exposure to methoxsalen for example, is attributable for 0.0017% of total cancer incidence among Korean population in 2015 (which is presented as 1.7 persons per 100,000 persons). In addition, the PAF of “Skin cancer by drug” indicates that group I drugs are attributable for 0.0714% of skin cancer incidence among Korean population in 2015 (which is presented as 71.4 persons per 100,000 persons). Lastly, the PAF of “All cancers by drug” indicates that group I drugs are attributable for 0.0032% of total cancer incidence among Korean population in 2015 (which is presented as 3.2 persons per 100,000 persons). The term “All cancers” in the Tables indicate all types of malignancies which is defined as C00 to C96 in the ICD-10-CM.

Carcinogenic drugs attributed to 0.003% of all cancer incidence in 2015. These drugs were attributable for 0.002% of all cancer death (Table 2).

When estimated by each carcinogenic drug, cyclosporine attributed to 0% of cancer incidence and death, and azathioprine attributed to 0.0003% of cancer incidence and death. Cyclophosphamide attributed to 0.0008% of cancer incidence and 0.0009% of cancer death. Methoxsalen attributed to 0.0017% of cancer incidence and 0.0004% of cancer death, and melphalan attributed to 0.0003% of cancer incidence and 0.0004% of cancer death. Chlorambucil attributed to 0% of cancer incidence and death, and busulfan attributed to 0.0001% of cancer incidence and 0.0002% of cancer death (Table 2).

The AC for all cancer incidence caused by carcinogenic drugs was 7. On the other hand, the AC for all cancer death caused by carcinogenic drugs was 2. The AC for skin, hematologic, and bladder cancer incidence by carcinogenic drugs were 4, 2, and 1, respectively. On the other hand, the AC for skin, hematologic, and bladder cancer death by carcinogenic drugs were 0, 1, and 0, respectively (Table 2).

The fraction of cancer incidence and death caused by carcinogenic drug, and proportion of each cancer incidence and death among all cancer incidence and death is presented in S10 and S11 Figs.

When sensitivity analysis was performed by study region, PAF estimated using the RR of global studies were 0.0058% for cancer incidence, and 0.0026% for cancer death. On the other hand, PAF estimated using only the RR of Korean studies were 0.0044% for cancer incidence, and 0.004% for cancer death (Table 2).

When sensitivity analysis was performed by study design, PAF estimated using only the RR of cohort studies were 0.0033% for cancer incidence, and 0.0031% for cancer death. On the other hand, PAF estimated using only the RR of case-control studies were 0.0059% for cancer incidence, and 0.0025% for cancer death (Table 2). The PAF and AC by men and women are presented in S12 Table.

In 2020, methoxsalen was most attributable for cancer incidence, and cyclophosphamide was most attributable for cancer death among carcinogenic drugs. Skin cancer had the largest proportion among cancer incidence and hematologic cancer had the largest proportion among cancer death among all cancer incidence and death (Figs. 1 and 2).

4. PAF projection

The attributable fraction of carcinogenic drugs to cancer incidence and death showed a slightly decreasing trend from 2015 to 2030 (Table 3).

In 2020, the attributable fraction of carcinogenic drugs to cancer incidence and death were 0.0023% and 0.0019%, respectively, which was slightly decreasing from 2015. In 2025, the attributable fraction of carcinogenic drugs to cancer incidence and death were 0.0008% and 0.0007%, respectively, which was also slightly decreasing from 2020. In 2030, the attributable fraction of carcinogenic drugs to cancer incidence and death were 0.0005% and 0.0004%, respectively, which showed a consistent decrease from 2015 (Table 3).

In 2020, the AC for cancer incidence and death were 6 and 2, and in 2025, the AC for cancer incidence and death were 2 and 1. In 2030, the AC for cancer incidence and death were 2 and 0, showing similar decreasing trend as PAF from 2015 to 2030 (Table 3). The temporal trends of PAF and AC by men and women are presented in S13 Table.

5. PAF calculation in patients with clinical indications

In patients with clinical indications of group I carcinogenic drugs, carcinogenic drugs attributed to 1.1% of cancer incidence and death in 2015 (Table 4).

When estimated by each carcinogenic drug, cyclosporine attributed to 0.001% of cancer incidence and death, azathioprine attributed to 0.09% of cancer incidence and death, and cyclophosphamide attributed to 0.2% of cancer incidence and death. Methoxsalen attributed to 0.3% of cancer incidence and 0.07% of cancer death, and melphalan attributed to 0.4% of cancer incidence and 0.5% of cancer death. Chlorambucil attributed to 0.02% of cancer incidence and 0.03% of cancer death, and busulfan attributed to 0.2% of cancer incidence and death (Table 4).

The AC for all cancer incidence caused by carcinogenic drugs was 47. On the other hand, the AC for all cancer death caused by carcinogenic drugs was 17. The AC for skin, hematologic, and bladder cancer incidence by carcinogenic drugs were 13, 31, and 3, respectively. On the other hand, the AC for skin, hematologic, and bladder cancer death by carcinogenic drugs were 1, 15, and 1, respectively (Table 4).

6. PAF projection in patients with clinical indications

The attributable fraction of carcinogenic drugs to cancer incidence and death in patients with clinical indications of group I carcinogenic drugs showed an increasing trend from 2015 to 2030 (Table 4).

In 2020, the attributable fraction of carcinogenic drugs to cancer incidence and death were 1.3% and 1.5%, respectively, which was slightly increasing from 2015. In 2025, the attributable fraction of carcinogenic drugs to cancer incidence and death were 1.5% and 1.9%, respectively, which was also increasing from 2020.

In 2030, the attributable fraction of carcinogenic drugs to cancer incidence and death were 2.4% and 3.3%, respectively, showing a dramatic increase from 2025 (Table 4).

In 2020, the AC for cancer incidence and death were 63 and 24, and in 2025, the AC for cancer incidence and death were 91 and 34. In 2030, the AC for cancer incidence and death were 169 and 61, showing similar increasing trend as PAF from 2015 to 2030 (Table 4). The temporal trends of PAF and AC by men and women are presented in S14 Table.

Discussion

The PAF of cancer incidence and death due to exposure to group I carcinogenic drugs was found to be very low (less than 0.01%) compared to other carcinogens. Although the PAF was very low, we found that skin cancer had notably the highest AC among cancer incidence attributed to carcinogenic drugs in women and total, followed by hematologic cancer and bladder cancer. Skin cancer also had the highest AC among cancer death attributed to carcinogenic drugs in women. Skin cancer consisted over 80% of AC among all cancers in cancer incidence in women and over 50% of AC among all cancers in cancer death in women. On the other hand, hematologic cancer had the highest AC among cancer incidence attributed to carcinogenic drugs in men, followed by skin and bladder cancer. Hematologic cancer also had the highest AC among cancer death attributed to carcinogenic drugs in men and total. Hematologic cancer consisted over 50% of AC among all cancers in cancer incidence in men and almost 80% of AC among all cancers in cancer death in men.

When classified by carcinogenic drugs, methoxsalen had the highest AC among carcinogenic drugs attributable to cancer incidence in women and total, followed by cyclophosphamide. On the contrary, cyclophosphamide had the highest AC among carcinogenic drugs attributable to cancer death in men and total.

Specifically, methoxsalen consisted over 80% of AC among all carcinogenic drugs attributable to cancer incidence in women and almost 50% of AC attributable to cancer death in women. On the other hand, cyclophosphamide had the highest AC among carcinogenic drugs attributable to cancer incidence in men. Specifically, cyclophosphamide consisted almost 40% of AC among all carcinogenic drugs attributable to cancer incidence in men and nearly 50% of AC attributable to cancer death in men.

The result indicates that skin cancer incidence and death due to methoxsalen use are more prevalent in women than men in Korea. On the contrary, it also indicates that hematologic cancer incidence and death due to cyclophosphamide use are more prevalent in men than women in Korea. This emphasizes the importance of regulating over prescription which could lead to drug abuse of methoxsalen in women and cyclophosphamide in men with dermatopathy and solid cancers, which are indications of these two drugs.

In addition, prescription of non-group I drugs as fist-line treatments or application of alternative therapeutic regimen in these patients appear to be essential. Difference of PAF and AC between men and women is also notable, since clinical indications such as atopic dermatitis seems to be equally prevalent among both sexes [48]. This difference implies the difference of RR and prevalence rate of carcinogenic drug use between both sexes.

Both cyclosporine and azathioprine had relatively less portion of AC compared to other five drugs, which indicates immunosuppressants are less attributing to cancer than antineoplastic agents or skin ointments. When examining the temporal trends of PAF, the overall projected PAF of cancer incidence and death due to carcinogenic drugs was estimated to be very low at all time, and there was a gradual decreasing trend as the years went by in both men and women. This temporal trend is predicted to continue, since the prevalence rate of carcinogenic drug use and RR of associated cancers are decreasing over time.

On the contrary, when calculated using population of patients with clinical indications of group I carcinogenic drugs, PAF were relatively high compared to the PAF calculated using the general population. In addition, the PAF using population of patients with clinical indications of group I carcinogenic drugs is expected to dramatically increase over time. This result emphasizes the importance of selecting an adequate study population when calculating PAF of carcinogenic drugs.

The sensitivity analysis by study region also indicates notable results, where the PAF of cancer incidence using RR of global studies (excluding Korea) is higher than the PAF using RR of only Korean studies, even though the PAF using RR of Korean studies was higher than the PAF using RR of global studies in men. This might be due to exceedingly higher PAF using RR of global studies than PAF using RR of Korean studies in women, which implies higher prevalence of methoxsalen use in outside of Korea than in Korea.

On the other hand, the PAF of cancer death using RR of Korean studies is higher than the PAF using RR of global studies, even though the PAF using RR of global studies was higher than the PAF using RR of Korean studies in women. This might be due to notably higher PAF using RR of Korean studies than PAF using RR of global studies in men, which implies higher prevalence of cyclophosphamide and melphalan use in Korea than outside of Korea.

The limitations of this study include the use of the main component codes (generic name codes) of the NHIS-NSC to calculate the standardized prevalence rate of carcinogenic drug use by year through age standardization. It was not possible to confirm the prevalence rate in the NHIS-NSC when the drug was non-covered in the past. In addition, among 24 known group I carcinogenic drugs, we could only include seven carcinogenic drugs in our study, due to absence of the main component codes (generic name codes) of other carcinogenic drugs and their relatively low prevalence rates calculated within NHIS-NSC. Since there were no records of main component codes of combined prescriptions of selected group I drugs within the NHIS-NSC cohort, we could not calculate the prevalence rate and estimate PAF of combined prescription of selected group I drugs. In addition, the duration of administration, dosage, and changes in medication of each group I drug was not used in the analysis since only the prevalence rate, RR, and PAF due to ever-use of group I drug use was estimated in this study.

Furthermore, as most of the studies included in the systematic review for the meta-analysis were foreign studies outside of Korea, SRR may not represent actual figures of Korea. We also could not obtain the RR of cancer death due to carcinogenic drug use, as there were no studies or raw data reporting the RR of cancer death. As a result, when calculating the PAF of cancer death due to carcinogenic drugs, we used the RR of cancer incidence instead. Therefore, there may exist a difference between the actual attributable fraction of cancer death due to carcinogenic drug use and attributable fraction of cancer death presented in this study. Moreover, it was not possible to obtain the RR of cancer incidence by each sex, as no studies included in the systematic review and meta-analysis reported RR according to each sex.

On the other hand, the strength of this study was that it was the first to calculate the PAF and AC for carcinogenic drugs, and it was estimated to represent Korean population based on the prevalence rate at the time of reimbursement of the carcinogenic drugs included in the study.

In conclusion, attributable risks of cancer incidence and death in general population were 0.003% and 0.002% respectively, which were relatively low compared to attributable risks of other known risk factors of cancers. However, group I carcinogenic drugs are primarily prescribed to solid organ transplant recipients and patients with various autoimmune diseases, inflammatory disorders, and cancers. In these susceptible populations, the attributable risks of cancer incidence and death were elevated to 1.1%, which was about 344 times higher than the risk of cancer incidence, and 478 times higher than the risk of cancer death compared to the general population. In addition, the attributable risks of cancer incidence and death were estimated to increase and reach 2.4% and 3.3% by 2030, respectively. This is over 2-fold increased risk of cancer incidence and 3-fold increased risk of cancer death compared to the attributable risks in 2015. Therefore, the comparisons between attributable risk in general population and susceptible populations were made in this study to underline the need for discussing and implementing surveillance system on susceptible populations using group I carcinogenic drugs in Korea. The surveillance system should integrate big data (e.g., data mining techniques) monitoring, risk signal detection, epidemiological investigations and causality assessments, and subsequently implement necessary actions and feedbacks. Through continuous actions and feedbacks on the aggregated real-world evidence, discussions among relevant institutions can be facilitated, and these insights can be reflected into actual policy decisions. Ultimately, the surveillance system could lead to discussions on providing financial support for regular cancer screenings for susceptible populations and consideration on establishing a compensation system following the health check-ups.

Electronic Supplementary Material

Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).

crt-2024-644_S1_Table.pdf

crt-2024-644_S2_Table.pdf

crt-2024-644_S3_Table.pdf

crt-2024-644_S4_Table.pdf

crt-2024-644_S5_Table.pdf

crt-2024-644_S6_Table.pdf

crt-2024-644_S7_Table.pdf

crt-2024-644_S8_Table.pdf

crt-2024-644_S9_Fig.pdf

crt-2024-644_S10_Fig.pdf

crt-2024-644_S11_Fig.pdf

crt-2024-644_S12_Table.pdf

crt-2024-644_S13_Table.pdf

crt-2024-644_S14_Table.pdf

NOTES

Ethical Statement

This study was approved by the Institutional Review Board of Seoul National University Hospital (IRB number C-1911-188-1084). The requirement for informed consent was waived.

Author Contributions

Conceived and designed the analysis: Lim W, Sung S, Hong Y, Moon S, Kim K, Lee JE, Kim I, Ko KP, Park SK.

Collected the data: Lim W, Sung S, Hong Y, Moon S, Park SK.

Contributed data or analysis tools: Lim W, Lee S, Park SK.

Performed the analysis: Lim W.

Wrote the paper: Lim W.

Conflict of Interest

Conflict of interest relevant to this article was not reported.

Funding

This study was funded by the Korean Foundation for Cancer Research (grant number. CB-2017-A-2). This study was supported by National Evidence-based healthcare Collaborating Agency (NA21-003).

Fig. 1.

Fraction (%) of cancer incidence caused by carcinogenic drug in Korea, 2020: total (A), men (B), and women (C).

Fig. 2.

Fraction (%) of cancer death caused by carcinogenic drug in Korea, 2020: total (A), men (B), and women (C).

Table 1.

RR of carcinogenic drugs^a) on cancer in all studies

Carcinogenic drugs	Cancer	All studies		All cohort studies
Carcinogenic drugs	Cancer	Study	RR (95% CI)	Study	RR (95% CI)
Cyclosporine	Skin (C43-C44)	14	1.44 (1.15-1.81)	11	1.38 (1.08-1.75)
	Hematologic (C81-C96)	7	1.05 (0.62-1.76)	6	1.06 (0.62-1.82)
Azathioprine	Skin (C43-C44)	14	1.81 (1.38-2.36)	10	1.62 (1.20-2.18)
	Hematologic (C81-C96)	12	1.73 (1.18-2.55)	10	1.80 (1.16-2.81)
Cyclophosphamide	Bladder (C67)	4	2.77 (1.99-3.86)	1	2.69 (1.92-3.78)
	Hematologic (C81-C96)	6	2.85 (1.72-4.71)	1	3.83 (3.20-4.59)
Methoxsalen	Skin (C43-C44)	2	2.97 (1.25-7.01)	1	2.32 (1.36-3.95)
Melphalan	Hematologic (C81-C96)	2	7.54 (1.49-38.23)	1	16.31 (13.41-19.85)
Chlorambucil	Hematologic (C81-C96)	2	3.43 (2.50-4.70)	1	3.51 (2.53-4.87)
Busulfan	Hematologic (C81-C96)	2	6.71 (2.49-18.08)	1	8.64 (2.44-30.60)

RR of incidence is used as RR of death since no studies reported RR of death in accordance with carcinogenic drugs. The difference of RR among men and women is not reported. CI, confidence interval; RR, relative risk.

^a) Drug use vs. non-use (reference).

Table 2.

PAF^a) of cancer due to carcinogenic drugs according to different RRs in 2015

	All study		Cohort study		Case control study^b)		Global study		Korean study
	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC
Incidence
Cyclosporine	0.0	0.09	0.0	0.08	0.1	0.18	0.0	0.06	0.4	0.83
Azathioprine	0.3	0.62	0.3	0.61	0.3	0.64	0.2	0.44	1.0	2.21
Cyclophosphamide	0.8	1.67	1.0	2.22	0.7	1.53	0.7	1.53	1.0	2.22
Methoxsalen	1.7	3.56	1.1	2.39	4.6	9.94	4.6	9.94	1.1	2.39
Melphalan	0.3	0.65	0.7	1.53	0.1	0.21	0.1	0.21	0.7	1.53
Chlorambucil	0.0	0.03	0.0	0.03	0.0	0.02	0.0	0.02	0.0	0.03
Busulfan	0.1	0.27	0.2	0.36	0.1	0.16	0.1	0.27	0.1	0.27
Skin cancer by drug	71.4	3.88	48.5	2.64	194.8	10.60	186.5	10.14	68.0	3.70
Hematologic cancer by drug	27.8	2.44	45.8	4.03	16.7	1.47	19.4	1.71	59.4	5.22
Bladder cancer by drug	14.1	0.58	13.5	0.55	14.9	0.61	14.9	0.61	13.5	0.55
All cancers^c) by drug	3.2	6.91	3.3	7.22	5.9	12.68	5.8	12.47	4.4	9.48
Death
Cyclosporine	0.0	0.01	0.0	0.01	0.0	0.02	0.0	0.01	0.4	0.31
Azathioprine	0.3	0.21	0.3	0.22	0.2	0.12	0.2	0.16	0.8	0.65
Cyclophosphamide	0.9	0.71	1.3	0.98	0.8	0.63	0.8	0.63	1.3	0.98
Methoxsalen	0.4	0.33	0.3	0.22	1.2	0.92	1.2	0.92	0.3	0.22
Melphalan	0.4	0.32	1.0	0.75	0.1	0.10	0.1	0.10	1.0	0.75
Chlorambucil	0.0	0.02	0.0	0.02	0.0	0.01	0.0	0.01	0.0	0.02
Busulfan	0.2	0.13	0.2	0.17	0.1	0.08	0.2	0.13	0.2	0.13
Skin cancer by drug	72.0	0.36	48.9	0.24	196.5	0.98	188.2	0.94	68.2	0.34
Hematologic cancer by drug	27.9	1.19	46.0	1.97	16.7	0.72	19.5	0.83	59.6	2.55
Bladder cancer by drug	13.5	0.18	12.9	0.17	14.3	0.19	14.3	0.19	12.9	0.17
All cancers^c) by drug	2.3	1.73	3.1	2.38	2.5	1.88	2.6	1.96	4.0	3.06

AC, attributable count; PAF, population attributable fraction.

^a) PAF presented as per 100,000 persons,

^b) 1 randomized controlled trial study on busulfan was included in the meta-analysis between case-control studies since no case-control study was found for busulfan,

^c) All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.

Table 3.

PAF^a) of cancer attributable to carcinogenic drugs from 2015 to 2030

	2015			2020			2025			2030
	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)
Incidence
Cyclosporine	215,570	0.09	0.0	246,436	0.14	0.1	300,233	0.17	0.1	358,627	0.18	0.0
Azathioprine	215,570	0.62	0.3	246,436	1.05	0.4	300,233	0.27	0.1	358,627	1.24	0.3
Cyclophosphamide	215,570	1.67	0.8	246,436	0.86	0.3	300,233	0.18	0.1	358,627	0.07	0.0
Methoxsalen	215,570	3.56	1.7	246,436	2.16	0.9	300,233	0.81	0.3	358,627	0.04	0.0
Melphalan	215,570	0.65	0.3	246,436	0.79	0.3	300,233	0.55	0.2	358,627	0.09	0.0
Chlorambucil	215,570	0.03	0.0	246,436	0.02	0.0	300,233	0.01	0.0	358,627	0.00	0.0
Busulfan	215,570	0.27	0.1	246,436	0.60	0.2	300,233	0.30	0.1	358,627	0.04	0.0
Skin cancer by drug	5,440	3.88	71.4	7,075	2.70	38.2	10,878	1.08	9.9	16,425	0.82	5.0
Hematologic cancer by drug	8,798	2.44	27.8	10,708	2.70	25.2	13,484	1.16	8.6	16,735	0.82	4.9
Bladder cancer by drug	4,111	0.58	14.1	4,750	0.22	4.7	5,658	0.04	0.7	6,596	0.01	0.2
All cancers^b) by drug	215,570	6.91	3.2	246,436	5.63	2.3	300,233	2.28	0.8	358,627	1.65	0.5
Death
Cyclosporine	76,621	0.01	0.0	82,036	0.02	0.0	88,713	0.02	0.0	93,690	0.02	0.0
Azathioprine	76,621	0.21	0.3	82,036	0.32	0.4	88,713	0.07	0.1	93,690	0.28	0.3
Cyclophosphamide	76,621	0.71	0.9	82,036	0.37	0.5	88,713	0.07	0.1	93,690	0.03	0.0
Methoxsalen	76,621	0.33	0.4	82,036	0.18	0.2	88,713	0.05	0.1	93,690	0.00	0.0
Melphalan	76,621	0.32	0.4	82,036	0.36	0.4	88,713	0.24	0.3	93,690	0.04	0.0
Chlorambucil	76,621	0.02	0.0	82,036	0.01	0.0	88,713	0.00	0.0	93,690	0.00	0.0
Busulfan	76,621	0.13	0.2	82,036	0.28	0.3	88,713	0.13	0.1	93,690	0.01	0.0
Skin cancer by drug	500	0.36	72.0	579	0.22	38.2	710	0.07	10.1	851	0.04	5.1
Hematologic cancer by drug	4,281	1.19	27.9	4,907	1.23	25.2	5,793	0.50	8.6	6,782	0.33	4.9
Bladder cancer by drug	1,299	0.18	13.5	1,593	0.08	4.8	1,874	0.01	0.7	2,134	0.00	0.2
All cancers^b) by drug	76,621	1.73	2.3	82,036	1.53	1.9	88,713	0.58	0.7	93,690	0.38	0.4

PAF, population attributable fraction.

^a) PAF presented as per 100,000 persons,

^b) All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.

Table 4.

Fraction^a) of cancer attributable to drug use among patients with clinical indications of carcinogenic drugs from 2015 to 2030

	2015			2020			2025			2030
	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)
Incidence
Cyclosporine	4,311	0	1.3	4,929	0	1.9	6,005	0	1.8	7,173	0	2.0
Azathioprine	4,311	4	90.5	4,929	8	156.5	6,005	17	275.0	7,173	35	487.4
Cyclophosphamide	4,311	9	202.1	4,929	5	110.8	6,005	7	119.3	7,173	10	133.3
Methoxsalen	4,311	11	260.5	4,929	9	185.3	6,005	6	94.1	7,173	4	49.7
Melphalan	4,311	15	354.0	4,929	22	446.8	6,005	39	644.6	7,173	81	1,128.8
Chlorambucil	4,311	1	20.7	4,929	1	12.6	6,005	0	7.3	7,173	0	4.2
Busulfan	4,311	7	161.8	4,929	18	359.9	6,005	22	371.5	7,173	40	553.8
Skin cancer by drug	109	13	11,749.2	142	12	8,708.9	218	13	6,076.4	329	21	6,475.2
Hematologic cancer by drug	176	31	17,745.9	214	49	22,896.2	270	76	28,210.2	335	146	43,657.6
Bladder cancer by drug	82	3	3,680.6	95	1	1,496.5	113	2	1,405.0	132	2	1,380.6
All cancers^b) by drug	4,311	47	1,090.9	4,929	63	1,273.7	6,005	91	1,513.6	7,173	169	2,359.2
Death
Cyclosporine	1,532	0	0.6	1,641	0	0.8	1,774	0	0.7	1,874	0	0.7
Azathioprine	1,532	1	85.3	1,641	2	144.1	1,774	4	246.1	1,874	8	423.8
Cyclophosphamide	1,532	4	241.0	1,641	2	143.0	1,774	3	167.8	1,874	4	204.9
Methoxsalen	1,532	1	68.0	1,641	1	45.2	1,774	0	21.2	1,874	0	10.4
Melphalan	1,532	7	487.8	1,641	10	611.6	1,774	17	935.0	1,874	33	1766.1
Chlorambucil	1,532	0	28.5	1,641	0	17.3	1,774	0	10.6	1,874	0	6.7
Busulfan	1,532	3	221.9	1,641	8	494.6	1,774	10	538.6	1,874	16	855.3
Skin cancer by drug	10	1	11,832.4	12	1	8,726.5	14	1	6,229.8	17	1	6,682.4
Hematologic cancer by drug	86	15	17,826.5	98	22	22,824.2	116	33	28,157.5	136	59	43,830.6
Bladder cancer by drug	26	1	3,525.5	32	0	1,535.9	37	1	1,487.4	43	1	1,512.2
All cancers^b) by drug	1,532	17	1,133.0	1,641	24	1,456.7	1,774	34	1,920.0	1,874	61	3,267.9

PAF, population attributable fraction.

^a) PAF presented as per 100,000 persons,

^b) All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.

REFERENCES

1. IARC monographs on the evaluation of the carcinogenic risk of chemicals to man: some aziridines, N-, S- & O-mustards and selenium. IARC Monogr Eval Carcinog Risk Chem Man. 1975;9:1–268. Article PubMed
2. Some pharmaceutical drugs. IARC Monogr Eval Carcinog Risk Chem Hum. 1980;24:1–337. PubMed
3. Some antineoplastic and immunosuppressive agents. IARC Monogr Eval Carcinog Risk Chem Hum. 1981;26:1–411. PubMed
4. Overall evaluations of carcinogenicity: an updating of IARC Monographs volumes 1 to 42. IARC Monogr Eval Carcinog Risks Hum Suppl. 1987;7:1–440. PubMed
5. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Pharmaceuticals. IARC Monogr Eval Carcinog Risks Hum. 2012;100:1–401.
6. Park S, Jee SH, Shin HR, Park EH, Shin A, Jung KW, et al. Attributable fraction of tobacco smoking on cancer using population-based nationwide cancer incidence and mortality data in Korea. BMC Cancer. 2014;14:406.Article PubMed PMC PDF
7. Lee S, Ko KP, Lee JE, Kim I, Jee SH, Shin A, et al. The Korea Cohort Consortium: the future of pooling cohort studies. J Prev Med Public Health. 2022;55:464–74. Article PubMed PMC PDF
8. Lee S, Moon S, Kim K, Sung S, Hong Y, Lim W, et al. A comparison of green, delta, and Monte Carlo methods to select an optimal approach for calculating the 95% confidence interval of the population-attributable fraction: guidance for epidemiological research. J Prev Med Public Health. 2024;57:499–507. Article PubMed PMC PDF
9. Hong Y, Lee S, Moon S, Sung S, Lim W, Kim K, et al. Projection of cancer incidence and mortality from 2020 to 2035 in the Korean population aged 20 years and older. J Prev Med Public Health. 2022;55:529–38. Article PubMed PMC PDF
10. Park SK, Lee NR. Safety and scale evaluation of IARC Group I carcinogenic pharmaceuticals use in Korea. National Evidence-based Healthcare Collaborating Agency Research reports. National Evidence-based Healthcare Collaborating Agency Research; 2022.
11. Bhat M, Mara K, Dierkhising R, Watt KD. Immunosuppression, race, and donor-related risk factors affect de novo cancer incidence across solid organ transplant recipients. Mayo Clin Proc. 2018;93:1236–46. Article PubMed
12. Cahoon EK, Linet MS, Clarke CA, Pawlish KS, Engels EA, Pfeiffer RM. Risk of Kaposi sarcoma after solid organ transplantation in the United States. Int J Cancer. 2018;143:2741–8. Article PubMed PMC PDF
13. Clarke CA, Robbins HA, Tatalovich Z, Lynch CF, Pawlish KS, Finch JL, et al. Risk of merkel cell carcinoma after solid organ transplantation. J Natl Cancer Inst. 2015;107:dju382.Article PubMed PMC
14. Coghill AE, Johnson LG, Berg D, Resler AJ, Leca N, Madeleine MM. Immunosuppressive medications and squamous cell skin carcinoma: nested case-control study within the Skin Cancer after Organ Transplant (SCOT) cohort. Am J Transplant. 2016;16:565–73. Article PubMed PMC
15. Hamandi B, Fegbeutel C, Silveira FP, Verschuuren EA, Younus M, Mo J, et al. Voriconazole and squamous cell carcinoma after lung transplantation: a multicenter study. Am J Transplant. 2018;18:113–24. Article PubMed PDF
16. Ingvar A, Smedby KE, Lindelof B, Fernberg P, Bellocco R, Tufveson G, et al. Immunosuppressive treatment after solid organ transplantation and risk of post-transplant cutaneous squamous cell carcinoma. Nephrol Dial Transplant. 2010;25:2764–71. Article PubMed
17. Jensen P, Hansen S, Moller B, Leivestad T, Pfeffer P, Geiran O, et al. Skin cancer in kidney and heart transplant recipients and different long-term immunosuppressive therapy regimens. J Am Acad Dermatol. 1999;40:177–86. Article PubMed
18. Kasiske BL, Snyder JJ, Gilbertson DT, Wang C. Cancer after kidney transplantation in the United States. Am J Transplant. 2004;4:905–13. Article PubMed
19. Keller B, Braathen LR, Marti HP, Hunger RE. Skin cancers in renal transplant recipients: a description of the renal transplant cohort in Bern. Swiss Med Wkly. 2010;140:w13036.Article PubMed
20. Lange E, Blizzard L, Venn A, Francis H, Jones G. Disease-modifying anti-rheumatic drugs and non-melanoma skin cancer in inflammatory arthritis patients: a retrospective cohort study. Rheumatology (Oxford). 2016;55:1594–600. Article PubMed
21. Mithoefer AB, Supran S, Freeman RB. Risk factors associated with the development of skin cancer after liver transplantation. Liver Transpl. 2002;8:939–44. Article PubMed
22. Molina BD, Leiro MG, Pulpon LA, Mirabet S, Yanez JF, Bonet LA, et al. Incidence and risk factors for nonmelanoma skin cancer after heart transplantation. Transplant Proc. 2010;42:3001–5. Article PubMed
23. Tseng HW, Lu LY, Lam HC, Tsai KW, Huang WC, Shiue YL. The influence of disease-modifying anti-rheumatic drugs and corticosteroids on the association between rheumatoid arthritis and skin cancer: a nationwide retrospective case-control study in Taiwan. Clin Exp Rheumatol. 2018;36:471–8. PubMed
24. Pasternak B, Svanstrom H, Schmiegelow K, Jess T, Hviid A. Use of azathioprine and the risk of cancer in inflammatory bowel disease. Am J Epidemiol. 2013;177:1296–305. Article PubMed
25. Pedersen EG, Pottegard A, Hallas J, Friis S, Hansen K, Jensen PE, et al. Risk of non-melanoma skin cancer in myasthenia patients treated with azathioprine. Eur J Neurol. 2014;21:454–8. Article PubMed
26. Setshedi M, Epstein D, Winter TA, Myer L, Watermeyer G, Hift R. Use of thiopurines in the treatment of inflammatory bowel disease is associated with an increased risk of non-melanoma skin cancer in an at-risk population: a cohort study. J Gastroenterol Hepatol. 2012;27:385–9. Article PubMed
27. Singh H, Nugent Z, Demers AA, Bernstein CN. Increased risk of nonmelanoma skin cancers among individuals with inflammatory bowel disease. Gastroenterology. 2011;141:1612–20. Article PubMed
28. Caillard S, Agodoa LY, Bohen EM, Abbott KC. Myeloma, Hodgkin disease, and lymphoid leukemia after renal transplantation: characteristics, risk factors and prognosis. Transplantation. 2006;81:888–95. Article PubMed
29. Caillard S, Lamy FX, Quelen C, Dantal J, Lebranchu Y, Lang P, et al. Epidemiology of posttransplant lymphoproliferative disorders in adult kidney and kidney pancreas recipients: report of the French registry and analysis of subgroups of lymphomas. Am J Transplant. 2012;12:682–93. Article PubMed
30. Fernberg P, Edgren G, Adami J, Ingvar A, Bellocco R, Tufveson G, et al. Time trends in risk and risk determinants of non-Hodgkin lymphoma in solid organ transplant recipients. Am J Transplant. 2011;11:2472–82. Article PubMed
31. Na R, Laaksonen MA, Grulich AE, Meagher NS, McCaughan GW, Keogh AM, et al. Iatrogenic immunosuppression and risk of non-Hodgkin lymphoma in solid organ transplantation: a population-based cohort study in Australia. Br J Haematol. 2016;174:550–62. Article PubMed
32. O’Regan JA, Prendeville S, McCaughan JA, Traynor C, O’Brien FJ, Ward FL, et al. Posttransplant lymphoproliferative disorders in Irish renal transplant recipients: insights from a national observational study. Transplantation. 2017;101:657–63. Article PubMed
33. Beaugerie L, Brousse N, Bouvier AM, Colombel JF, Lemann M, Cosnes J, et al. Lymphoproliferative disorders in patients receiving thiopurines for inflammatory bowel disease: a prospective observational cohort study. Lancet. 2009;374:1617–25. Article PubMed
34. Bernatsky S, Clarke AE, Suissa S. Hematologic malignant neoplasms after drug exposure in rheumatoid arthritis. Arch Intern Med. 2008;168:378–81. Article PubMed
35. Khan N, Abbas AM, Lichtenstein GR, Loftus EV Jr, Bazzano LA. Risk of lymphoma in patients with ulcerative colitis treated with thiopurines: a nationwide retrospective cohort study. Gastroenterology. 2013;145:1007–15. Article PubMed
36. Mbulaiteye SM, Clarke CA, Morton LM, Gibson TM, Pawlish K, Weisenburger DD, et al. Burkitt lymphoma risk in U.S. solid organ transplant recipients. Am J Hematol. 2013;88:245–50. Article PubMed PMC
37. Zimmermann T, Hoppe-Lotichius M, Tripkovic V, Barreiros AP, Wehler TC, Zimmermann A, et al. Liver transplanted patients with preoperative autoimmune hepatitis and immunological disorders are at increased risk for post-transplant lymphoproliferative disease (PTLD). Eur J Intern Med. 2010;21:208–15. Article PubMed
38. Xu Y, Wang H, Zhou S, Yu M, Wang X, Fu K, et al. Risk of second malignant neoplasms after cyclophosphamide-based chemotherapy with or without radiotherapy for non-Hodgkin lymphoma. Leuk Lymphoma. 2013;54:1396–404. Article PubMed
39. Curtis RE, Boice JD Jr, Stovall M, Bernstein L, Greenberg RS, Flannery JT, et al. Risk of leukemia after chemotherapy and radiation treatment for breast cancer. N Engl J Med. 1992;326:1745–51. Article PubMed
40. Nandakumar A, Davis S, Moolgavkar S, Witherspoon RP, Schwartz SM. Myeloid leukaemia following therapy for a first primary cancer. Br J Cancer. 1991;63:782–8. Article PubMed PMC PDF
41. Travis LB, Curtis RE, Stovall M, Holowaty EJ, van Leeuwen FE, Glimelius B, et al. Risk of leukemia following treatment for non-Hodgkin’s lymphoma. J Natl Cancer Inst. 1994;86:1450–7. Article PubMed
42. Kaldor JM, Day NE, Kittelmann B, Pettersson F, Langmark F, Pedersen D, et al. Bladder tumours following chemotherapy and radiotherapy for ovarian cancer: a case-control study. Int J Cancer. 1995;63:1–6. Article PubMed
43. Travis LB, Curtis RE, Glimelius B, Holowaty EJ, Van Leeuwen FE, Lynch CF, et al. Bladder and kidney cancer following cyclophosphamide therapy for non-Hodgkin’s lymphoma. J Natl Cancer Inst. 1995;87:524–30. Article PubMed
44. Hannuksela-Svahn A, Pukkala E, Laara E, Poikolainen K, Karvonen J. Psoriasis, its treatment, and cancer in a cohort of Finnish patients. J Invest Dermatol. 2000;114:587–90. Article PubMed
45. Finazzi G, Ruggeri M, Rodeghiero F, Barbui T. Second malignancies in patients with essential thrombocythaemia treated with busulphan and hydroxyurea: long-term follow-up of a randomized clinical trial. Br J Haematol. 2000;110:577–83. Article PubMed PDF
46. Finazzi G, Caruso V, Marchioli R, Capnist G, Chisesi T, Finelli C, et al. Acute leukemia in polycythemia vera: an analysis of 1638 patients enrolled in a prospective observational study. Blood. 2005;105:2664–70. Article PubMed
47. Lim W, Moon S, Lee NR, Shin HG, Yu SY, Lee JE, et al. Group I pharmaceuticals of IARC and associated cancer risks: systematic review and meta-analysis. Sci Rep. 2024;14:413.Article PubMed PMC PDF
48. Johansson EK, Bergstrom A, Kull I, Melen E, Jonsson M, Lundin S, et al. Prevalence and characteristics of atopic dermatitis among young adult females and males-report from the Swedish population-based study BAMSE. J Eur Acad Dermatol Venereol. 2022;36:698–704. Article PubMed PMC PDF

Figure & Data

REFERENCES

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Fraction of Cancer Attributable to Carcinogenic Drugs in Korea from 2015 to 2030

Cancer Res Treat. 2025;57(3):635-648. Published online November 6, 2024

DOI: https://doi.org/10.4143/crt.2024.644

XML Download

Abstract
Introduction
Materials and Methods
Results
Discussion
Electronic Supplementary Material
NOTES
REFERENCES

Fraction of Cancer Attributable to Carcinogenic Drugs in Korea from 2015 to 2030

Fig. 1. Fraction (%) of cancer incidence caused by carcinogenic drug in Korea, 2020: total (A), men (B), and women (C).

Fig. 2. Fraction (%) of cancer death caused by carcinogenic drug in Korea, 2020: total (A), men (B), and women (C).

Fig. 1.

Fig. 2.

Fraction of Cancer Attributable to Carcinogenic Drugs in Korea from 2015 to 2030

Carcinogenic drugs	Cancer	All studies		All cohort studies
Carcinogenic drugs	Cancer	Study	RR (95% CI)	Study	RR (95% CI)
Cyclosporine	Skin (C43-C44)	14	1.44 (1.15-1.81)	11	1.38 (1.08-1.75)
	Hematologic (C81-C96)	7	1.05 (0.62-1.76)	6	1.06 (0.62-1.82)
Azathioprine	Skin (C43-C44)	14	1.81 (1.38-2.36)	10	1.62 (1.20-2.18)
	Hematologic (C81-C96)	12	1.73 (1.18-2.55)	10	1.80 (1.16-2.81)
Cyclophosphamide	Bladder (C67)	4	2.77 (1.99-3.86)	1	2.69 (1.92-3.78)
	Hematologic (C81-C96)	6	2.85 (1.72-4.71)	1	3.83 (3.20-4.59)
Methoxsalen	Skin (C43-C44)	2	2.97 (1.25-7.01)	1	2.32 (1.36-3.95)
Melphalan	Hematologic (C81-C96)	2	7.54 (1.49-38.23)	1	16.31 (13.41-19.85)
Chlorambucil	Hematologic (C81-C96)	2	3.43 (2.50-4.70)	1	3.51 (2.53-4.87)
Busulfan	Hematologic (C81-C96)	2	6.71 (2.49-18.08)	1	8.64 (2.44-30.60)

	All study		Cohort study		Case control study^b)		Global study		Korean study
	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC	PAF^a)	AC
Incidence
Cyclosporine	0.0	0.09	0.0	0.08	0.1	0.18	0.0	0.06	0.4	0.83
Azathioprine	0.3	0.62	0.3	0.61	0.3	0.64	0.2	0.44	1.0	2.21
Cyclophosphamide	0.8	1.67	1.0	2.22	0.7	1.53	0.7	1.53	1.0	2.22
Methoxsalen	1.7	3.56	1.1	2.39	4.6	9.94	4.6	9.94	1.1	2.39
Melphalan	0.3	0.65	0.7	1.53	0.1	0.21	0.1	0.21	0.7	1.53
Chlorambucil	0.0	0.03	0.0	0.03	0.0	0.02	0.0	0.02	0.0	0.03
Busulfan	0.1	0.27	0.2	0.36	0.1	0.16	0.1	0.27	0.1	0.27
Skin cancer by drug	71.4	3.88	48.5	2.64	194.8	10.60	186.5	10.14	68.0	3.70
Hematologic cancer by drug	27.8	2.44	45.8	4.03	16.7	1.47	19.4	1.71	59.4	5.22
Bladder cancer by drug	14.1	0.58	13.5	0.55	14.9	0.61	14.9	0.61	13.5	0.55
All cancers^c) by drug	3.2	6.91	3.3	7.22	5.9	12.68	5.8	12.47	4.4	9.48
Death
Cyclosporine	0.0	0.01	0.0	0.01	0.0	0.02	0.0	0.01	0.4	0.31
Azathioprine	0.3	0.21	0.3	0.22	0.2	0.12	0.2	0.16	0.8	0.65
Cyclophosphamide	0.9	0.71	1.3	0.98	0.8	0.63	0.8	0.63	1.3	0.98
Methoxsalen	0.4	0.33	0.3	0.22	1.2	0.92	1.2	0.92	0.3	0.22
Melphalan	0.4	0.32	1.0	0.75	0.1	0.10	0.1	0.10	1.0	0.75
Chlorambucil	0.0	0.02	0.0	0.02	0.0	0.01	0.0	0.01	0.0	0.02
Busulfan	0.2	0.13	0.2	0.17	0.1	0.08	0.2	0.13	0.2	0.13
Skin cancer by drug	72.0	0.36	48.9	0.24	196.5	0.98	188.2	0.94	68.2	0.34
Hematologic cancer by drug	27.9	1.19	46.0	1.97	16.7	0.72	19.5	0.83	59.6	2.55
Bladder cancer by drug	13.5	0.18	12.9	0.17	14.3	0.19	14.3	0.19	12.9	0.17
All cancers^c) by drug	2.3	1.73	3.1	2.38	2.5	1.88	2.6	1.96	4.0	3.06

	2015			2020			2025			2030
	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)
Incidence
Cyclosporine	215,570	0.09	0.0	246,436	0.14	0.1	300,233	0.17	0.1	358,627	0.18	0.0
Azathioprine	215,570	0.62	0.3	246,436	1.05	0.4	300,233	0.27	0.1	358,627	1.24	0.3
Cyclophosphamide	215,570	1.67	0.8	246,436	0.86	0.3	300,233	0.18	0.1	358,627	0.07	0.0
Methoxsalen	215,570	3.56	1.7	246,436	2.16	0.9	300,233	0.81	0.3	358,627	0.04	0.0
Melphalan	215,570	0.65	0.3	246,436	0.79	0.3	300,233	0.55	0.2	358,627	0.09	0.0
Chlorambucil	215,570	0.03	0.0	246,436	0.02	0.0	300,233	0.01	0.0	358,627	0.00	0.0
Busulfan	215,570	0.27	0.1	246,436	0.60	0.2	300,233	0.30	0.1	358,627	0.04	0.0
Skin cancer by drug	5,440	3.88	71.4	7,075	2.70	38.2	10,878	1.08	9.9	16,425	0.82	5.0
Hematologic cancer by drug	8,798	2.44	27.8	10,708	2.70	25.2	13,484	1.16	8.6	16,735	0.82	4.9
Bladder cancer by drug	4,111	0.58	14.1	4,750	0.22	4.7	5,658	0.04	0.7	6,596	0.01	0.2
All cancers^b) by drug	215,570	6.91	3.2	246,436	5.63	2.3	300,233	2.28	0.8	358,627	1.65	0.5
Death
Cyclosporine	76,621	0.01	0.0	82,036	0.02	0.0	88,713	0.02	0.0	93,690	0.02	0.0
Azathioprine	76,621	0.21	0.3	82,036	0.32	0.4	88,713	0.07	0.1	93,690	0.28	0.3
Cyclophosphamide	76,621	0.71	0.9	82,036	0.37	0.5	88,713	0.07	0.1	93,690	0.03	0.0
Methoxsalen	76,621	0.33	0.4	82,036	0.18	0.2	88,713	0.05	0.1	93,690	0.00	0.0
Melphalan	76,621	0.32	0.4	82,036	0.36	0.4	88,713	0.24	0.3	93,690	0.04	0.0
Chlorambucil	76,621	0.02	0.0	82,036	0.01	0.0	88,713	0.00	0.0	93,690	0.00	0.0
Busulfan	76,621	0.13	0.2	82,036	0.28	0.3	88,713	0.13	0.1	93,690	0.01	0.0
Skin cancer by drug	500	0.36	72.0	579	0.22	38.2	710	0.07	10.1	851	0.04	5.1
Hematologic cancer by drug	4,281	1.19	27.9	4,907	1.23	25.2	5,793	0.50	8.6	6,782	0.33	4.9
Bladder cancer by drug	1,299	0.18	13.5	1,593	0.08	4.8	1,874	0.01	0.7	2,134	0.00	0.2
All cancers^b) by drug	76,621	1.73	2.3	82,036	1.53	1.9	88,713	0.58	0.7	93,690	0.38	0.4

	2015			2020			2025			2030
	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)	Observed cancer	Attributable cancer	PAF^a)
Incidence
Cyclosporine	4,311	0	1.3	4,929	0	1.9	6,005	0	1.8	7,173	0	2.0
Azathioprine	4,311	4	90.5	4,929	8	156.5	6,005	17	275.0	7,173	35	487.4
Cyclophosphamide	4,311	9	202.1	4,929	5	110.8	6,005	7	119.3	7,173	10	133.3
Methoxsalen	4,311	11	260.5	4,929	9	185.3	6,005	6	94.1	7,173	4	49.7
Melphalan	4,311	15	354.0	4,929	22	446.8	6,005	39	644.6	7,173	81	1,128.8
Chlorambucil	4,311	1	20.7	4,929	1	12.6	6,005	0	7.3	7,173	0	4.2
Busulfan	4,311	7	161.8	4,929	18	359.9	6,005	22	371.5	7,173	40	553.8
Skin cancer by drug	109	13	11,749.2	142	12	8,708.9	218	13	6,076.4	329	21	6,475.2
Hematologic cancer by drug	176	31	17,745.9	214	49	22,896.2	270	76	28,210.2	335	146	43,657.6
Bladder cancer by drug	82	3	3,680.6	95	1	1,496.5	113	2	1,405.0	132	2	1,380.6
All cancers^b) by drug	4,311	47	1,090.9	4,929	63	1,273.7	6,005	91	1,513.6	7,173	169	2,359.2
Death
Cyclosporine	1,532	0	0.6	1,641	0	0.8	1,774	0	0.7	1,874	0	0.7
Azathioprine	1,532	1	85.3	1,641	2	144.1	1,774	4	246.1	1,874	8	423.8
Cyclophosphamide	1,532	4	241.0	1,641	2	143.0	1,774	3	167.8	1,874	4	204.9
Methoxsalen	1,532	1	68.0	1,641	1	45.2	1,774	0	21.2	1,874	0	10.4
Melphalan	1,532	7	487.8	1,641	10	611.6	1,774	17	935.0	1,874	33	1766.1
Chlorambucil	1,532	0	28.5	1,641	0	17.3	1,774	0	10.6	1,874	0	6.7
Busulfan	1,532	3	221.9	1,641	8	494.6	1,774	10	538.6	1,874	16	855.3
Skin cancer by drug	10	1	11,832.4	12	1	8,726.5	14	1	6,229.8	17	1	6,682.4
Hematologic cancer by drug	86	15	17,826.5	98	22	22,824.2	116	33	28,157.5	136	59	43,830.6
Bladder cancer by drug	26	1	3,525.5	32	0	1,535.9	37	1	1,487.4	43	1	1,512.2
All cancers^b) by drug	1,532	17	1,133.0	1,641	24	1,456.7	1,774	34	1,920.0	1,874	61	3,267.9

Table 1. RR of carcinogenic drugsa) on cancer in all studies

Drug use vs. non-use (reference).

Table 2. PAFa) of cancer due to carcinogenic drugs according to different RRs in 2015

AC, attributable count; PAF, population attributable fraction.

PAF presented as per 100,000 persons,

1 randomized controlled trial study on busulfan was included in the meta-analysis between case-control studies since no case-control study was found for busulfan,

All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.

Table 3. PAFa) of cancer attributable to carcinogenic drugs from 2015 to 2030

PAF, population attributable fraction.

PAF presented as per 100,000 persons,

All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.

Table 4. Fractiona) of cancer attributable to drug use among patients with clinical indications of carcinogenic drugs from 2015 to 2030

PAF, population attributable fraction.

PAF presented as per 100,000 persons,

All cancers indicate all types of malignancies which is defined as C00 to C96 in the International Classification of Diseases, 10th revision, clinical modification.