Next Article in Journal
COVID-19 Pandemic: Impact on Cancer Patients
Previous Article in Journal
Effectiveness of Dry Needling and Ischemic Trigger Point Compression in the Gluteus Medius in Patients with Non-Specific Low Back Pain: A Randomized Short-Term Clinical Trial
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Pharmacoepidemiological Research on N-Nitrosodimethylamine-Contaminated Ranitidine Use and Long-Term Cancer Risk: A Population-Based Longitudinal Cohort Study

1
Department of Hepatogastroenterology, Tainan Municipal Hospital (Managed by Show Chwan Medical Care Corporation), Tainan 701033, Taiwan
2
Department of Optometry, Chung Hwa Medical University, Tainan 701033, Taiwan
3
Department of Gastroenterology, Chang Bing Show Chwan Memorial Hopital, Changhua 505029, Taiwan
4
Committee of Medical Research, Tainan Municipal Hospital (Managed by Show Chwan Medical Care Corporation), Tainan 701033, Taiwan
*
Author to whom correspondence should be addressed.
These author contributed equally to this work.
Int. J. Environ. Res. Public Health 2022, 19(19), 12469; https://doi.org/10.3390/ijerph191912469
Submission received: 27 July 2022 / Revised: 24 September 2022 / Accepted: 27 September 2022 / Published: 30 September 2022
(This article belongs to the Section Public Health Statistics and Risk Assessment)

Abstract

:

Simple Summary

There is a lack of published data regarding the association between N-nitrosodimethylamine and human cancer risks. Hence, this study answers several questions using propensity score matching with a large population size selected from a high-quality nationwide and population-based database with a long follow-up period to assess the relationship between the cumulative individual cancer incidence and long-term ranitidine use.

Abstract

N-Nitrosodimethylamine (NDMA), a carcinogenic chemical, has recently been identified in ranitidine. We conducted a population-based study to explore ranitidine use and cancer emergence over time. Using the Taiwan National Health Insurance Research Database, a population-based cohort study was conducted. A total of 55,110 eligible patients who received ranitidine between January 2000 and December 2018 were enrolled in the treated cohort. We conducted a 1:1 propensity-score-matching procedure to match the ranitidine-treated group with the ranitidine-untreated group and famotidine controls for a longitudinal study. The association of ranitidine exposure with cancer outcomes was assessed. A multivariable Cox regression analysis that compared cancer risk with the untreated groups revealed that ranitidine increased the risk of liver (hazard ratio (HR): 1.22, 95% confidence interval (CI): 1.09–1.36, p < 0.001), lung (HR: 1.17, CI: 1.05–1.31, p = 0.005), gastric (HR: 1.26, CI: 1.05–1.52, p = 0.012), and pancreatic cancers (HR 1.35, CI: 1.03–1.77, p = 0.030). Our real-world observational study strongly supports the pathogenic role of NDMA contamination, given that long-term ranitidine use is associated with a higher likelihood of liver cancer development in ranitidine users compared with the control groups of non-ranitidine users treated with famotidine or proton-pump inhibitors.

1. Introduction

Ranitidine, a histamine-2 receptor antagonist, inhibits gastric acid secretion when treating gastroesophageal reflux disease and peptic ulcers [1]. Additionally, according to the data from the Food and Drug Administration (FDA) Adverse Event Reporting System, elevated and significant proportional reporting ratios (PRRs) were observed for pharyngeal, esophageal, stomach, colorectal, liver, and pancreatic cancers, including elevated PRRs for anal and gallbladder cancers [2].
In 2019, the U.S. FDA declared that N-nitrosodimethylamine (NDMA), with the formula (CH3)2NNO, identified in medicines containing valsartan and ranitidine, is a member of N-nitrosamines and a known carcinogen, according to laboratory results [3,4,5]. The FDA’s testing of ranitidine products revealed that NDMA levels were nine times greater than the FDA’s recommended limit, resulting in global recalls [6,7]. Several studies have also reported that NDMA could be oncogenic in animals [8]. According to the International Agency for Research on Cancer report, NDMA has been proven to belong to group 2A and to be “probably carcinogenic to humans” [9].
A study also reported that high ranitidine doses combined with nitrite produced DNA fragmentation in rodents’ livers and gastric mucosa [10]. Many observational human studies have reported that consuming a high number of NDMA-contaminated foods may be linked to an increased risk of stomach and colon cancers [11,12]. Additionally, detailed experimental animal studies showed that cancer risk may increase with NDMA exposure through inhalation or oral delivery and that tumors developed in the lungs, liver, kidneys, and bile ducts in animals [13]. Several studies previously examined the carcinogenic effects of NDMA on humans, although they were equivocal. Some authors reported no association between ranitidine use and cancer risk [14,15,16], whereas others supported the connection [17,18,19,20]. In addition to cancers caused by NDMA contamination, multiple studies reported that acid-suppressive agents, such as proton-pump inhibitors (PPIs) and histamine-2 receptor antagonists (H2RAs), were linked to gastric [21,22] and liver cancers [23,24,25,26]. However, these reports were contradictory, and the data were not sufficient to reach definite conclusions. The conflicting results of studies underlie the lack of concrete evidence supporting the role of ranitidine in cancer development. Therefore, we aimed to conduct a large-scale, long-term follow-up cohort study to investigate ranitidine use and the subsequent emergence of cancer over time in a real-life setting.

2. Materials and Methods

2.1. Data Source

This study used Taiwan’s National Health Insurance Database (NHIRD), which is a population-based claims database, and a cross-sectional survey participated in by over 99% of Taiwan’s population. We included all medical services, procedures, and prescription medication data from 1 January 2000 to 31 December 2018. The diagnoses recorded in the NHIRD are coded in accordance with the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM), and Tenth Revision, Clinical Modification (ICD-10-CM).
Considering that the NHIRD dataset consists of encrypted secondary data, each person is impossible to identify; thus, the informed consent requirement was waived. The Research Ethics Committee of Show Chwan Memorial Hospital approved the study protocol on 14 December 2021 (IRB-No: 1101105).

2.2. Study Design and Study Participants

The total doses for each ranitidine prescription during the follow-up period were calculated to indicate the duration of ranitidine exposure. As the World Health Organization proposed, one defined daily dose (DDD) of ranitidine was 300 mg/day [27]. We defined 90 DDDs as the valid treatment for patients with reflux esophagitis and peptic ulcer disease treated with 300 mg ranitidine daily for 3 months [28,29]. Patients prescribed ranitidine at ≥90 DDDs were assigned to the ranitidine cohort, whereas those who never used ranitidine belonged to the non-ranitidine cohort.
We investigated whether a dose–response relationship exists between ranitidine and cancer diagnosis. For the sensitivity analysis, we grouped the patient follow-up period into four intervals according to the cumulative dose, starting from the first prescription: 90–180, 181–270, 271–360, and >360 DDDs.

2.3. Potential Confounders

The exclusion criteria were as follows: age < 40 years, diagnosis with cancer before the index date, ranitidine use <90 DDDs, and follow-up <1 year.
We enrolled ranitidine users who were matched for exact age, sex, the Charlson comorbidity index (CCI), comorbidities (hypertensive cardiovascular disease (HCD0, hyperlipidemia, diabetes mellitus (DM), and chronic kidney disease (CKD)), medications (aspirin, statins, angiotensin-converting enzyme inhibitors (ACEIs), β-blockers, spironolactone, glucocorticoids, and selective serotonin reuptake inhibitors (SSRIs), and antiviral therapy for hepatitis B or C virus (HBV and HCV, respectively) infection), and the index date.
The final matched cohort consisted of 55,110 patients who were evaluated from the index date until the target cancer onset, death, or the end of the study period (31 December 2018).

2.4. Covariate Assessment

The main confounding factors were based on a recent study investigating the association between N-nitrosodimethylamine and cancer. We adjusted for the following covariates that potentially affect cancer incidence: age, sex, and medication history (low-dose aspirin, statins, ACEIs, β-blockers, spironolactone, glucocorticoids, SSRIs, and antiviral treatment for chronic hepatitis B or C).
We also considered the following comorbidities: HCD (ICD-9 codes 401–405; ICD-10 codes I10–I15), hyperlipidemia (ICD-9 code 272; ICD-10 code E78), DM (ICD-9 code 250; ICD-10 codes E10.0, E10.1, E10.9, E11.0, E11.1, and E11.9), CKD (ICD-9 code 585; ICD-10 code N18), and CCI.

2.5. Main Outcome Measurements

We assessed the following cancer categories for the first cancer diagnosis: liver cancer (ICD-9 code 155; ICD-10 code C22), oral cancer (ICD-9 codes 140–149; ICD-10 codes C00–C14), esophageal cancer (ICD-9 code 150; ICD-10 code C15), gastric cancer (ICD-9 code 151; ICD-10 code C16), colon cancer (ICD-9 code 153; ICD-10 code C18), rectal cancer (ICD-9 code 154; ICD-10 codes C19–C21), pancreatic cancer (ICD-9 code 157; ICD-10 code C25), lung cancer (ICD-9 code 162; ICD-10 codes C33 and C34), bone cancer (ICD-9 code 170; ICD-10 codes C40 and C41), bladder cancer (ICD-9 code 188; ICD-10 code C67), renal cancer (ICD-9 code 189; ICD-10 codes C65, C66, and C68), thyroid cancer (ICD-9 code 193; ICD-10 code C73), skin cancer (ICD-9 codes 172–173; ICD-10 code C44), breast cancer in females (ICD-9 code 174; ICD-10 code C50), uterine cancer (ICD-9 code 179; ICD-10 code C55), cervical cancer (ICD-9 code 180; ICD-10 code C53), ovarian cancer (ICD-9 code 183; ICD-10 codes C56 and C570–C574), prostate cancer (ICD-9 code 185; ICD-10 codes C61), and overall cancer. For breast, uterine, cervical, and ovarian cancers, the analysis was restricted to females, and prostate cancer was restricted to males.

2.6. Exposure Definition and Follow-Up

In this study, we adopted a new-user design with 1 year as the washout period to eliminate the influence of external factors in patients with newly diagnosed cancer [30]. Patients with no predefined outcomes or those who died during the follow-up were censored. We defined the index date for each ranitidine user as the date of their first prescription. For their corresponding matched comparison group, the index date was set to be that of their matched individual with ranitidine use. All patients were followed up from the index date to 2018. The mean follow-up period was 9.56 ± 5.96 (median: 8.42) years for the ranitidine cohort and 9.70 ± 5.96 (median: 8.58) years for the non-ranitidine cohort. All prescriptions, diagnostic outcomes, and deaths were ascertained until 31 December 2018. For all groups, the follow-up duration was defined as the interval from the date of enrollment to the date of cancer diagnosis, death, or the end of the follow-up period, whichever came first.

2.7. Statistical Analysis

Categorical variables were compared using the McNemar test. Continuous variables such as the prescription and medical records as baseline characteristics were compared using paired t-tests. To reduce potential selection bias, we used propensity score matching (PSM) to balance the differences in proportions, such as comorbidities, between the ranitidine and non-ranitidine cohorts.
For a robust propensity score matching, 1:1 full matching without replacement was performed. Therefore, the regression model was specified correctly relative to the population regression function of the outcome variable on the treatment and all covariates used for matching. Therefore, trimming techniques were not employed in the weighting approach [31,32]. PSM was performed using multivariate logistic regression analysis and nearest-neighbor matching with the R package “MatchIt” (version 4.3.4).
We used Cox proportional hazards regression models to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) of cancer risk for ranitidine users in comparison with non-ranitidine users. To confirm the stability and robustness of our model, all hazard ratios and their 95% CIs were modified using the bootstrapping method [33,34]. The outcomes of the different study cohorts were estimated using the Kaplan–Meier method, and the differences in curves were examined using the log-rank test. Sensitivity analysis was conducted to validate the individual events from the index date to the end of the study in different DDD exposure groups using Cox proportional hazards regression models and the Kaplan–Meier method.
The competing risks of death were adjusted using the R package “cmprsk” (version 2.2–11), and the regression model was assessed according to Fine and Gray. All data management procedures were performed using SPSS 21.0 (SPSS Inc., Chicago, IL, USA) and R version 3.4.3 (R Core Team, 2017). A p-value of <0.05 was considered statistically significant.

3. Results

In the NHIRD database, we found 290,990 ranitidine users and 1,709,128 non-ranitidine users within the study period (Figure 1). In accordance with the exclusion criteria, 75,715 and 1,022,217 patients were finally included in the ranitidine and non-ranitidine cohorts, respectively. Using PSM, we matched the ranitidine cohort (n = 55,110) with the non-ranitidine cohort (n = 55,110) in a 1:1 model. Figure 1 illustrates the flowchart of patient selection.
Patients from the non-ranitidine cohort who were statistically matched with those from the ranitidine cohort were selected in consideration of the following factors: age, sex, CCI, and comorbidities, including HCD, hyperlipidemia, DM, and CKD.
Certain medications (aspirin, statins, ACEIs, β-blockers, spironolactone, glucocorticoids, SSRIs, and antiviral therapy for HBV or HCV infection) and the index date (the exact date of diagnosis) were potential confounders for most cancers. Table 1 shows the baseline characteristics of the well-balanced, 1:1 matched cohort. Sex, age, CCI scores, and the follow-up period were fully matched between the ranitidine and non-ranitidine cohorts. Overall, the male–female ratio was 47.8:52.2 (p = 1.000), with mean values of 66.8 ± 14.1 (p = 1.000), 3.77 ± 2.78 (p = 1.000), and 9.2 ± 5.9 years (p = 1.000) for age, CCI, and follow-up duration, respectively. The median follow-up duration was 8.3 years.
As shown in Figure 2, the ranitidine cohort showed a significantly higher prevalence of liver cancer (1.1% vs. 1.3%; p = 0.012), gastric cancer (0.4% vs. 0.5%; p = 0.037), and lung cancer (1.0% vs. 1.2%; p = 0.033) and a higher overall cancer rate (8.0% vs. 8.5%; p = 0.001) than the non-ranitidine cohort. In terms of the cumulative incidence rates and HRs, ranitidine use was associated with overall cancer (HR, 1.10; 95% CI, 1.06–1.15), liver cancer (HR, 1.22; 95% CI, 1.09–1.36), gastric cancer (HR, 1.26; 95% CI, 1.05–1.52), pancreatic cancer (HR, 1.35; 95% CI, 1.03–1.77), and lung cancer (HR, 1.17; 95% CI, 1.05–1.31) compared with non-ranitidine use. No significant associations were observed for the 14 other cancers. In the Kaplan–Meier analysis, the ranitidine cohort exhibited a significantly higher risk of developing liver cancer (log-rank test, p = 0.005), lung cancer (p = 0.016), gastric cancer (p = 0.025), and pancreatic cancer (p = 0.043) than the non-ranitidine cohort (Figure 3).
Table 2 shows the incidence of each cancer. The incidence rate was 9.19 per 1000 person-years in the ranitidine group and 8.49 per 1000 person-years in the non-ranitidine group. The incidence rates of certain cancers were greater in the group exposed to ranitidine than in the unexposed group. Ranitidine use was associated with some individual cancers with a high incidence rate (per 1000 person-years); these cancers were liver (1.35 vs. 1.16), lung (1.23 vs. 1.07), gastric (0.48 vs. 0.39), and pancreatic (0.23 vs. 0.17) cancers.

3.1. Ranitidine Duration Effect on Cancer Development

The effect of ranitidine use on the risk of progression to significant individual cancers (e.g., liver cancer, gastric cancer, lung cancer, and pancreatic cancer) was assessed by multivariate Cox regression analysis adjusted for age, sex, CCI, co-medications (aspirin, statins, ACEIs, β-blockers, spironolactone, glucocorticoids, SSRIs, and antiviral therapy for HBV or HCV infection), comorbidities (HCD, hyperlipidemia, DM, and CKD), and the calendar date at the start of follow-up.
Ranitidine users were divided into groups according to drug exposure: 90–180 DDDs, 181–270 DDDs, 271–360 DDDs, >360 DDDs, and an unexposed group. After adjusting for potential confounders, we found that ranitidine use was a potential risk factor for liver cancer development. For patients with relatively limited exposure to ranitidine (<360 DDDs), ranitidine did not significantly affect the risk of developing liver cancer compared with nonusers (Figure 4). However, increased ranitidine exposure was associated with liver cancer risk. For patients with >360 DDDs, the adjusted HR of liver cancer development was 1.42 (95% CI: 1.22–1.66; p < 0.001) compared with that in the non-ranitidine group (Table 3).
Regression was adjusted for age, sex, the Charlson comorbidity index, co-medications (aspirin, statins, angiotensin-converting enzyme inhibitors, β-blockers, spironolactone, glucocorticoids, selective serotonin reuptake inhibitors, and antiviral therapy for hepatitis B or C), comorbidities (hypertensive cardiovascular disease, hyperlipidemia, diabetes mellitus, and chronic kidney disease), and calendar year at the start of follow-up.

3.2. Comparison between Ranitidine and Famotidine for the Association with Patient Outcomes

To avoid potential indication bias, we selected non-ranitidine users (control subjects) by PSM for the famotidine cohort, with a ranitidine–famotidine ratio of 1:1. This subgroup was added to determine whether the use of ranitidine increases the risk of developing cancers due to the related indication.
We screened the risk for cancer in the ranitidine (n = 35,269) and famotidine (n = 35,269) cohorts (Figure 1). Ranitidine users who were statistically matched with famotidine users were selected, adjusting for the following factors: age, sex, indications, co-medications, and comorbidities.
The prevalence of overall cancer was 3052 (8.7%) in the ranitidine group and 2924 (8.3%) in the famotidine group. The overall cancer risk was statistically different between these two groups (adjusted HR, 1.07; 95% CI, 1.02–1.12, p = 0.010). Significant differences were also observed in liver (adjusted HR, 1.22; 95% CI, 1.06–1.40, p = 0.005) and renal cancer (adjusted HR, 1.33; 95% CI, 1.02–1.73, p = 0.034) outcomes between the two groups (Table 4).
In the Kaplan–Meier analysis, we found that liver cancer risk was significantly different between the ranitidine and famotidine cohorts with a balanced model (p = 0.019, Figure 5A). Moreover, the liver cancer risk was significantly higher in the ranitidine cohort than in non-ranitidine with famotidine and non-ranitidine without famotidine cohorts (p < 0.001 and p = 0.02, respectively; Figure 5B). Therefore, the pattern of the cumulative incidence of liver cancer was the same in the non-ranitidine groups, regardless of famotidine use.

3.3. Comparison between Ranitidine and PPIs for Their Association with Liver Cancer

Considering another potential indication bias, we categorized the non-ranitidine users (control subjects) into those with and without PPI use. This subgroup analysis aimed to determine whether ranitidine use increased liver cancer risk due to an alternative medicine with a related indication.
We screened the risk for cancer in the ranitidine (n = 55,110), ranitidine without PPI (n = 51,361), and ranitidine with PPI (n = 3749) cohorts (Figure 6). Ranitidine users were selected to adjust for the following factors: age, sex, indications, co-medications, and comorbidities.
The liver cancer risk was significantly higher in the ranitidine group than in the non-ranitidine without PPI group (adjusted HR, 1.16; 95% CI, 1.04–1.30, p = 0.006). Furthermore, liver cancer risk was significantly lower in the non-ranitidine with PPI group than in the non-ranitidine without PPI group (adjusted HR, 0.49; 95% CI, 0.33–0.75, p = 0.001). In the Kaplan–Meier analysis, we found that liver cancer risk was significantly higher in the ranitidine group than in the non-ranitidine without PPI and non-ranitidine with PPI groups (p = 0.032 and p < 0.001, respectively; Figure 6).

4. Discussion

The current research, a population-level epidemiologic study, evaluated cancer risk attributed to long-term ranitidine use with NDMA exposure, which was linked to a higher liver cancer risk than the non-ranitidine group and the famotidine group. Several epidemiological analyses have reported the public health concern of NDMA exposure, which has been linked to an increased risk of stomach and colon cancers [11,12,35]. The carcinogenic effects of NDMA theoretically result from inducing DNA-damaging metabolites in the gastrointestinal tract and liver, as suggested by animal studies. Notably, NDMA is metabolized in the liver by CYP2E1 to methyl diazonium, leading to mutations caused by methylation and the development of liver cancer [3,36]. Several studies have reported that hypoacidity due to acid-suppressive medication use also plays a critical role in the development of liver and gastric cancers. The hypothesized mechanisms include bacterial overgrowth, the formation of N-nitroso compounds, lipopolysaccharides, and deoxycholic acid, which have been linked to the development of liver cancer [37,38,39,40,41,42,43]. Additionally, higher gastrin levels following PPI or H2RA use may be associated with gastrointestinal malignancies [44,45]. Therefore, it is reasonable to assume that a link between acid-suppressive medication use and cancer development may be based on differing mechanisms. However, the clear data from our real-world observational study strongly support the pathogenic role of NDMA contamination, given that long-term ranitidine use is associated with a higher likelihood of cancer development in ranitidine users compared to the control groups of non-ranitidine users who were treated with PPIs or famotidine. Conversely, an increasing number of recent clinical and epidemiological studies [14,15] concluded that there is no convincing evidence of the carcinogenic potency of ranitidine. Nevertheless, the limitations of the two studies mentioned above should be considered since their small sample size and short follow-up duration may cause statistical bias and inaccurate conclusions. One notable strength of our study is the huge population size selected from a high-quality nationwide and population-based database with a long follow-up period of 18 years. Specifically, it was based on a cohort design of a seemingly prospective technique to explore ranitidine exposure and cancer outcomes. Additionally, outcome data were retrieved from formal cancer registries, which are more accurate than other sources. Using PSM, our study constructed an artificial control group (non-ranitidine users) with similar characteristics by combining it with additional matching for multiple prognostic factors or regression adjustment. Using these matches, we estimated the impact of ranitidine intervention on cancer risk, which showed increased odds of developing liver, lung, pancreatic, and gastric cancers. The Kaplan–Meier analysis of our 18-year dataset confirmed these findings.
We included a second active comparator group of individuals who were also prescribed famotidine, containing no NDMA and used for an almost identical indication, which might minimize potential bias to clarify potential confounding by indication.
The overall cancer risk was statistically different between these two groups compared with famotidine or non-ranitidine users. Notably, liver and renal cancers were more common among ranitidine users. Furthermore, the Kaplan–Meier analysis revealed that liver cancer risk was significantly higher in the ranitidine cohort than in the famotidine cohort. Our study observed this outcome using non-ranitidine users as a control group. Additionally, this result contradicts other reports [14,15]. Therefore, based on a direct comparison with either the non-ranitidine group or the famotidine group (similar indication to ranitidine users), only liver cancer displayed a significant association with long-term ranitidine use. This approach was used to ameliorate the implicit indication bias that occurs when the cancer risk is related to the indication for medication use but not to the use of the medication itself [46].
Another comparative approach in our study revealed the association of ranitidine usage with four individual cancers with a high incidence rate (per 1000 person-years), including liver, lung, gastric, and pancreatic cancers, compared with the general population. However, as stated by Roberts et al. [27,47,48], the preference for a pharmacoepidemiological study of drug safety is to develop a new-user design rather than a prevalent-user design in which patients have already been receiving therapy for some time before the study follow-up begins. Therefore, our study used 1 year as the washout period, during which the participants were taken off ranitidine, to remove the effects of treatment before the study initiation. Nonetheless, ranitidine still showed similar results after excluding “protopathic bias” [49], meaning that ranitidine use sometimes precedes cancer development before it is diagnosed.
A positive quality of our study is the use of PSM [50] in a population-based cohort design, which imitates a randomized trial to control for confounding factors that mostly depend on selecting the documented confounders used in the matching model [51]. These potential confounders in our study, which were causally associated with cancer development, include age, sex, CCI, comorbidities, and medications. Nonetheless, PSM cannot specifically balance unknown factors as randomized controlled trials (RCTs) do. Therefore, some experts [52] argued that substantial bias exists in a PSM study, which is one of our study limitations. However, despite having this limitation, PSM potentially takes advantage of the ability to generate a huge sample size from a large database within a short time. Moreover, it is impractical and unethical to conduct an RCT to test the carcinogenicity of ranitidine in a patient.
In an additional dose–response subanalysis, given that drug–cancer associations are mostly dose-dependent, we further stratified the extent of NDMA exposure by cumulative ranitidine usage based on drug exposure: 90–180 DDDs, 181–270 DDDs, 271–360 DDDs, >360 DDDs, and an unexposed group. Notably, when considering the dose–response of ranitidine usage, there were significant trends of increased liver cancer risk with an increasing dose of ranitidine. However, there was no continuous dose–response relationship among the other individual cancers. Additionally, Iwagami et al. [14] reported contrary results, although they acknowledged a weakness in the study design due to the limited sample size and statistical power.
The conclusive results of our study after gathering data emphasize that consuming high levels of NDMA due to ranitidine use is linked to liver cancer development. Many current pieces of evidence based on several animal studies propose that NDMA affects liver cancer development, mostly originating from a detailed exploration of the molecular basis of NDMA’s carcinogenic action [53,54,55,56,57]. For example, Souliotis et al. [58] reported that rats exposed to hepatocarcinogenic NDMA (0.2–2.64 ppm in the drinking water) for up to 180 days had a rapid accumulation of N7- and O6-methylguanine in the liver and white blood cells. The analysis of DNA, with the maximum adduct levels reached within 1–7 days dose-dependently, indicates that the accumulation of DNA damage and alterations in hepatocyte DNA replication during chronic NDMA exposure may influence the dose dependence of its carcinogenic efficacy. Notably, in the actual scenario, our result agrees with the above experimental data on the cumulative dose of ranitidine usage, which plays a vital role in hepatocarcinogenesis.
The present study has several limitations. First, it was constrained by the study design since we could not accurately estimate the NDMA level. Second, there were no data available in our database regarding certain confounders, such as alcohol consumption and cigarette smoking. Third, the patients’ medication compliance cannot be detected from the NHIRD. Fourth, there was scarce information regarding over-the-counter ranitidine usage, which caused the underestimation of ranitidine exposure. Fifth, the NHIRD data used in our study was not the most recent. Sixth, the NHIRD lacks specific laboratory information. Finally, potential misdiagnosis, including comorbidities and cancer categories, is possible in the NHIRD due to the potential misclassification of ICD-9-CM and ICD-10-CM codes.

5. Conclusions

To conclude, the clinically meaningful results of this large-scale, longitudinal population-based cohort study using an excellent prescription and cancer database provide concrete evidence with very convincing long-term follow-up information for exploring the causative role of ranitidine in increasing the risk of carcinogenic effects on the liver, which was primarily caused by increasingly heavier ranitidine usage. However, to elucidate the underlying mechanisms of its causal association, further studies are necessary.

Author Contributions

Conceptualization, C.-H.W. and I.-I.C.; methodology, Y.-T.T.; software, Y.-T.T.; validation, Y.-T.T.; formal analysis, Y.-T.T.; investigation, C.-H.W. and I.-I.C.; resources, C.-H.W. and C.-H.C.; data curation, C.-H.W.; writing—original draft preparation, C.-H.W.; writing—review and editing, I.-I.C.; visualization, Y.-T.T.; supervision, C.-H.W.; project administration, C.-H.W.; funding acquisition, C.-H.W. and C.-H.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by grant number RD-111-05, and the APC was funded by Tainan Municipal Hospital.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Show Chwan Memorial Hospital (IRB-No: 1101105) for studies involving humans.

Informed Consent Statement

Considering that the NHIRD dataset consists of encrypted secondary data, each person is impossible to identify; thus, the informed consent requirement was waived.

Data Availability Statement

The NHIRD protects personal electronic data by rigorous confidentiality guidelines. The results presented in the study are available from the NHIRD of Taiwan for researchers who meet the criteria for access to confidential data, which cannot be shared publicly because of legal restrictions imposed by the government of Taiwan under the “Personal Information Protection Act”. Requests for data can be sent as a formal proposal to the NHIRD (https://dep.mohw.gov.tw/dos/np-2497-113.html, accessed on 1 August 2022). The contact information for needed data is: 886-2-85906828; Email: [email protected].

Conflicts of Interest

The authors declare no conflict of interest. The funding sources had no role in the design, analysis, interpretation, or reporting of results or in the decision to submit the manuscript for publication.

References

  1. Lipsy, R.J.; Fennerty, B.; Fagan, T.C. Clinical review of histamine2 receptor antagonists. Arch. Intern. Med 1990, 150, 745–751. [Google Scholar] [CrossRef] [PubMed]
  2. Vermeer, N.S.; Straus, S.M.; Mantel-Teeuwisse, A.K.; Domergue, F.; Egberts, T.C.; Leufkens, H.G.; De Bruin, M.L. Traceability of biopharmaceuticals in spontaneous reporting systems: A cross-sectional study in the FDA Adverse Event Reporting System (FAERS) and EudraVigilance databases. Drug Saf. 2013, 36, 617–625. [Google Scholar] [CrossRef] [PubMed]
  3. Tricker, A.R.; Preussmann, R. Carcinogenic N-nitrosamines in the Diet: Occurrence, Formation, Mechanisms and Carcinogenic Potential. Mutat. Res. Genet. Toxicol. 1991, 259, 277–289. [Google Scholar] [CrossRef]
  4. U.S. Food and Drug Administration. FDA Updates and Press Announcements on NDMA in Zantac (Ranitidine). Available online: https://www.fda.gov/drugs/drug-safety-and-availability/fda-updates-and-press-announcementsndma-zantac-ranitidine (accessed on 16 April 2020).
  5. European Medicines Agency. EMA Reviewing Medicines Containing Valsartan from Zhejiang Huahai Following the Detection of an Impurity. 2018. Available online: http://www.ema.europa.eu/ema/index.jsp?curl=pages/news_and_events/news/2018/07/news_detail_002984.jsp&mid=WC0b01ac058004d5c1 (accessed on 24 August 2018).
  6. Mahase, E. FDA recalls ranitidine medicines over potential cancer causing impurity. BMJ 2019, 367, I5832. [Google Scholar] [CrossRef]
  7. Medicines and Healthcare Products Regulatory Agency. Class 2 Medicines Recall: Zantac Injection 50 mg/2 mL, Zantac Syrup 150 mg/10 mL, Zantac Tablets 150 mg, Zantac Tablets 300 mg (EL (19)A 24). 8 October 2019. Available online: https://www.gov.uk/drug-devic-alerts/class-2-medicines-recall-zantac-injection-50mg-2ml-zantac-syrup-150mg-10ml-zantac-tablets-150mg-zantac-tablets-300mg-el-19-a-24 (accessed on 22 June 2020).
  8. Lijinsky, W.; Reuber, M.D. Carcinogenesis in rats by nitrosodimethylamine and other nitrosomethylalkylamines at low doses. Cancer Lett. 1984, 22, 83–88. [Google Scholar] [CrossRef]
  9. International Agency for Research on Cancer. Overall Evaluations of Carcinogenicity: And Updating of IARC Monographs, Volume 1 to 42. In IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans: Supplement 7; IARC: Lyon, France, 1987; Volume 7, pp. 1–440. [Google Scholar]
  10. Brambilla, G.; Cavanna, M.; Faggin, P.; Maura, A.; Pino, A.; Ricci, R.; Robbiano, L. Genotoxic effects in rodents given high oral doses of ranitidine and sodium nitrite. Carcinogenesis 1983, 4, 1281–1285. [Google Scholar] [CrossRef]
  11. Loh, Y.H.; Jakszyn, P.; Luben, R.N.; Mulligan, A.A.; Mitrou, P.N.; Khaw, K.T. N-Nitroso compounds and cancer incidence: The European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk Study. Am. J. Clin. Nutr. 2011, 93, 1053–1061. [Google Scholar] [CrossRef]
  12. Zhu, Y.; Wang, P.P.; Zhao, J.; Green, R.; Sun, Z.; Roebothan, B.; Squires, J.; Buehler, S.; Dicks, E.; Zhao, J.; et al. Dietary N-nitroso compounds and risk of colorectal cancer: A case-control study in Newfoundland and Labrador and Ontario, Canada. Br. J. Nutr. 2014, 111, 1109–1117. [Google Scholar] [CrossRef]
  13. Habel, L.A.; Levin, T.R.; Friedman, G.D. Cimetidine use and risk of breast, prostate, and other cancers. Pharmacoepidemiol. Drug Saf. 2000, 9, 149–155. [Google Scholar] [CrossRef]
  14. Iwagami, M.; Kumazawa, R.; Miyamoto, Y.; Ito, Y.; Ishimaru, M.; Morita, K.; Hamada, S.; Tamiya, N.; Yasunaga, H. Risk of Cancer in Association with Ranitidine and Nizatidine vs Other H2 Blockers: Analysis of the Japan Medical Data Center Claims Database 2005–2018. Drug Saf. 2021, 44, 361–371. [Google Scholar] [CrossRef]
  15. Yoon, H.J.; Kim, J.-H.; Seo, G.H.; Park, H. Risk of Cancer Following the Use of N-Nitrosodimethylamine (NDMA) Contaminated Ranitidine Products: A Nationwide Cohort Study in South Korea. J. Clin. Med. 2021, 10, 153. [Google Scholar] [CrossRef]
  16. Kantor, E.D.; O’Connell, K.; Du, M.; Mendelsohn, R.B.; Liang, P.S.; Braunstein, L.Z. Ranitidine use and cancer risk: Results from UK Biobank. Gastroenterology 2021, 160, 1856–1859. [Google Scholar] [CrossRef]
  17. Cardwell, C.R.; McDowell, R.D.; Hughes, C.M.; Hicks, B.; Murchie, P. Exposure to Ranitidine and Risk of Bladder Cancer: A Nested Case-Control Study. Am. J. Gastroenterol. 2021, 116, 1612–1619. [Google Scholar] [CrossRef]
  18. McGwin, G. The Associa-tion between Ranitidine Use and Gas-trointestinal Cancers. Cancers 2021, 13, 24. [Google Scholar] [CrossRef]
  19. Song, P.; Wu, L.; Guan, W. Dietary nitrates, nitrites, and nitrosamines intake and the risk of gastric cancer: A meta-analysis. Nutrients 2015, 7, 9872–9895. [Google Scholar] [CrossRef]
  20. Hidajat, M.; McElvenny, D.M.; Ritchie, P.; Darnton, A.; Mueller, W.; Van Tongeren, M.; De Vocht, F. Lifetime exposure to rubber dusts, fumes and N-nitrosamines and cancer mortality in a cohort of British rubber workers with 49 years follow-up. Occup. Environ. Med. 2019, 76, 250–258. [Google Scholar] [CrossRef]
  21. Tran-Duy, A.; Spaetgens, B.; Hoes, A.W.; de Wit, N.J.; Stehouwer, C.D. Use of proton pump inhibitors and risks of fundic gland polyps and gastric cancer: Systematic review and meta-analysis. Clin. Gastroenterol. Hepatol. 2016, 14, 1706–1719. [Google Scholar] [CrossRef]
  22. Cheung, K.S.; Chan, E.W.; Wong, A.Y.; Chen, L.; Wong, I.C.; Leung, W.K. Long-term proton pump inhibitors and risk of gastric cancer development after treatment for Helicobacter pylori: A population-based study. Gut 2018, 67, 28–35. [Google Scholar] [CrossRef]
  23. Li, D.K.; Yan, P.; Abou-Samra, A.B.; Chung, R.T.; Butt, A.A. Proton pump inhibitors are associated with accelerated development of cirrhosis, hepatic decompensation and hepatocellular carcinoma in noncirrhotic patients with chronic hepatitis C infection: Results from ERCHIVES. Aliment. Pharmacol. Ther. 2018, 47, 246–258. [Google Scholar] [CrossRef]
  24. Lai, S.W.; Liao, K.F.; Lai, H.C.; Lin, C.L.; Sung, F.C. Proton pump inhibitors and risk of hepatocellular carcinoma: A case-control study in Taiwan. Acta Gastro-Enterol. Belg. 2013, 76, 348–350. [Google Scholar]
  25. Kao, W.-Y.; Su, C.-W.; Tan, E.C.-H.; Lee, P.-C.; Chen, P.-H.; Tang, J.-H.; Huang, Y.-H.; Huo, T.-L.; Chang, C.-C.; Hou, M.-C.; et al. Proton pump inhibitors and risk of hepatocellular carcinoma in patients with chronic hepatitis B or C. Hepatology 2019, 69, 1151–1164. [Google Scholar] [CrossRef] [PubMed]
  26. Kim, S.; Jeong, S.; Park, S.J.; Chang, J.; Choi, S.; Cho, Y.; Ahn, J.C.; Lee, G.; Son, J.S.; Park, S.M. Association between Proton Pump Inhibitor Use and Risk of Hepatocellular Carcinoma: A Korean Nationally Representative Cohort Study. J. Clin. Med. 2022, 11, 2865. [Google Scholar] [CrossRef] [PubMed]
  27. World Health Organization. ATC/DDD Index 2020. Available online: https://www.whocc.no/atc_ddd_index/ (accessed on 22 June 2020).
  28. Wesdorp, I.C.; Dekker, W.; Klinkenberg-Knol, E.C. Treatment of reflux oesophagitis with ranitidine. Gut 1983, 24, 921–924. [Google Scholar] [CrossRef]
  29. Ashton, M.G.; Holdsworth, C.D.; Ryan, F.P.; Moore, M. Healing of gastric ulcers after one, two, and three months of ranitidine. Br. Med. J. (Clin. Res. Ed.) 1982, 284, 467–468. [Google Scholar] [CrossRef] [PubMed]
  30. Roberts, A.W.; Dusetzina, S.B.; Farley, J.F. Revisiting the washout period in the incident user study design: Why 6–12 months may not be sufficient. J. Comp. Eff. Res. 2015, 4, 27–35. [Google Scholar] [CrossRef]
  31. Greifer, N. Estimating Effects After Matching. MatchIt: Nonparametric Preprocessing for Parametric Causal Inference. Available online: https://cran.r-project.org/web/packages/MatchIt/vignettes/estimating-effects.html (accessed on 1 August 2022).
  32. Abadie, A.; Spiess, J. Robust post-matching inference. J. Am. Stat. Assoc. 2022, 117, 983–995. [Google Scholar] [CrossRef]
  33. Higbee, J.D.; Lefler, J.S.; Burnett, R.T.; Ezzati, M.; Marshall, J.D.; Kim, S.-Y.; Bechle, M.; Robinson, A.L.; Pope, C.A., III. Estimating long-term pollution exposure effects through inverse probability weighting methods with Cox proportional hazards models. Environ. Epidemiol. 2020, 4, e085. [Google Scholar] [CrossRef]
  34. Davison, A.C.; Hinkley, D.V. Bootstrap Methods and Their Application; Cambridge University Press: Cambridge, UK, 2013. [Google Scholar]
  35. Knekt, P.; Järvinen, R.; Dich, J.; Hakulinen, T. Risk of colorectal and other gastro-intestinal cancers after exposure to nitrate, nitrite and N-nitroso compounds: A follow-up study. Int. J. Cancer 1999, 80, 852–856. [Google Scholar] [CrossRef]
  36. Bartsch, H.; KO’Neill, I. Ninth International Meeting on N-Nitroso Compounds: Exposures, Mechanisms, and Relevance to Human Cancer. Cancer Res. 1988, 48, 4711–4714. [Google Scholar]
  37. Thorens, J.; Froehlich, F.; Schwizer, W.; Saraga, E.; Bille, J.; Gyr, K.; Duroux, P.; Nicolet, M.; Pignatelli, B.; Blum, A.L.; et al. Bacterial overgrowth during treatment with omeprazole compared with cimetidine: A prospective randomised double blind study. Gut 1996, 39, 54–59. [Google Scholar] [CrossRef]
  38. Laine, L.; Ahnen, D.; McClain, C.; Solcia, E.; Walsh, J.H. potential gastrointestinal effects of long-term acid suppression with proton pump inhibitors. Aliment. Pharmacol. Ther. 2000, 14, 651–668. [Google Scholar] [CrossRef]
  39. Tao, X.; Wang, N.; Qin, W. Gut microbiota and hepatocellular carcinoma. Gastrointest. Tumors 2015, 2, 33–40. [Google Scholar] [CrossRef]
  40. Tannenbaum, S.R. N-nitroso compounds: A perspective on human exposure. Lancet 1983, 1, 629–632. [Google Scholar] [CrossRef]
  41. Mitacek, E.J.; Brunnemann, K.D.; Suttajit, M.; Martin, N.; Limsila, T.; Ohshima, H.; Caplan, L.S. Exposure to N-nitroso compounds in a population of high liver cancer regions in Thailand: Volatile nitrosamine (VNA) levels in Thai food. Food Chem. Toxicol. 1999, 37, 297–305. [Google Scholar] [CrossRef]
  42. Jirillo, E.; Caccavo, D.; Magrone, T.; Piccigallo, E.; Amati, L.; Lembo, A.; Kalis, C.; Gumenscheimer, M. The role of the liver in the response to LPS: Experimental and clinical findings. J. Endotoxin Res. 2002, 8, 319–327. [Google Scholar] [CrossRef]
  43. Darnaud, M.; Faivre, J.; Moniaux, N. Targeting gut flora to prevent progression of hepatocellular carcinoma. J. Hepatol. 2013, 58, 385–387. [Google Scholar] [CrossRef]
  44. Orlando, L.A.; Lenard, L.; Orlando, R.C. Chronic hypergastrinemia: Causes and consequences. Dig. Dis. Sci. 2007, 52, 2482–2489. [Google Scholar] [CrossRef]
  45. Moore, T.C.; Jepeal, L.I.; Boylan, M.O.; Singh, S.K.; Boyd, N.; Beer, D.G.; Chang, A.J.; Wolfe, M.M. Gastrin stimulates receptor-mediated proliferation of human esophageal adenocarcinoma cells. Regul. Pept. 2004, 120, 195–203. [Google Scholar] [CrossRef]
  46. Salas, M.; Hotman, A.; Stricker, B.H. Confounding by indication: An example of variation in the use of epidemiologic terminology. Am. J. Epidemiol. 1999, 149, 981–983. [Google Scholar] [CrossRef]
  47. Ray, W.A. Evaluating medication effects outside of clinical trials: New-user designs. Am. J. Epidemiol. 2003, 158, 915–920. [Google Scholar] [CrossRef]
  48. Perrio, M.; Waller, P.C.; Shakir, S.A. An analysis of the exclusion criteria used in observational pharmacoepidemiological studies. Pharmacoepidemiol. Drug Saf. 2007, 16, 329–336. [Google Scholar] [CrossRef]
  49. Horwitz, R.I.; Feinstein, A.R. The problem of “protopathic bias” in case-control studies. Am. J. Med. 1980, 68, 255–258. [Google Scholar] [CrossRef]
  50. Stürmer, T.; Joshi, M.; Glynn, R.J.; Avorn, J.; Rothman, K.J.; Schneeweiss, S. A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods. J. Clin. Epidemiol. 2006, 59, 437.e1–437.e24. [Google Scholar] [CrossRef] [PubMed]
  51. Hallas, J.; Pottegård, A. Use of self-controlled designs in pharmacoepidemiology. J. Intern. Med. 2014, 275, 581–589. [Google Scholar] [CrossRef] [PubMed]
  52. Peikes, D.N.; Moreno, L.; Orzol, S.M. Propensity score matching: A note of caution for evaluators of social programs. Am. Stat. 2008, 62, 222–231. [Google Scholar] [CrossRef]
  53. Swenberg, J.A.; Bedell, M.A.; Billings, K.C.; Umbenhauer, D.R.; Pegg, A.E. Cell-specific differences in O6-alkylguanine DNA repair activity during continuous exposure to carcinogen. Proc. Natl. Acad. Sci. USA 1982, 79, 5499–5502. [Google Scholar] [CrossRef]
  54. Bedell, M.A.; Lewis, J.G.; Billings, K.C.; Swenberg, J.A. Cell specificity in hepatocarcinogenesis: 06-methylguanine preferentially accumulates in target cell DNA during continuous exposure of rats to 1,2-dimethylhydrazine. Cancer Res. 1982, 42, 3079–3083. [Google Scholar]
  55. Dumenco, L.L.; Allay, E.; Norton, K.; Gerson, S.L. The prevention of thymic lymphomas in transgenic mice by human O6-alkylguanine-DNA alkyltransferase. Science 1993, 259, 219–222. [Google Scholar] [CrossRef]
  56. Nakatsuru, Y.; Matsukuma, S.; Nemoto, N.; Sugano, H.; Sekiguchi, M.; Ishikawa, T. O6-methylguanine-DNA methyltransferase protects against nitrosamine-induced hepatocarcinogenesis. Proc. Natl. Acad. Sci. USA 1993, 90, 6468–6472. [Google Scholar] [CrossRef]
  57. Zaidi, N.H.; Pretlow, T.P.; O’Riordan, M.A.; Dumenco, L.L.; Allay, E.; Gerson, S.L. Transgenic expression of human MGMT protects against azoxymethane-induced aberrant crypt foci and G to A mutations in the K-ras oncogene of mouse colon. Carcinogenesis 1995, 16, 451–456. [Google Scholar] [CrossRef]
  58. Souliotis, V.L.; Henneman, J.R.; Reed, C.D.; Chhabra, S.K.; Diwan, B.A.; Anderson, L.M.; Kyrtopoulos, S.A. DNA adducts and liver DNA replication in rats during chronic exposure to N-nitrosodimethylamine (NDMA) and their relationships to the dose-dependence of NDMA hepatocarcinogenesis. Mutat. Res. Fundam. Mol. Mech. Mutagen. 2002, 500, 75–87. [Google Scholar] [CrossRef]
Figure 1. Flowchart of the selection of study patients. 1 Comorbidities: hypertensive cardiovascular disease, hyperlipidemia, diabetes mellitus, and chronic kidney disease. 2 Medications: aspirin, statins, angiotensin-converting enzyme inhibitors, β-blockers, famotidine, spironolactone, glucocorticoids, selective serotonin reuptake inhibitors, and antiviral therapy for hepatitis B or C. 3 Index date: exact date of the first prescription.
Figure 1. Flowchart of the selection of study patients. 1 Comorbidities: hypertensive cardiovascular disease, hyperlipidemia, diabetes mellitus, and chronic kidney disease. 2 Medications: aspirin, statins, angiotensin-converting enzyme inhibitors, β-blockers, famotidine, spironolactone, glucocorticoids, selective serotonin reuptake inhibitors, and antiviral therapy for hepatitis B or C. 3 Index date: exact date of the first prescription.
Ijerph 19 12469 g001
Figure 2. Number and proportion of 18 different cancer reports in the ranitidine and non-ranitidine cohorts and the associated proportional reporting ratios and 95% confidence intervals in multivariate Cox proportional hazards regression adjusted for competing mortality.
Figure 2. Number and proportion of 18 different cancer reports in the ranitidine and non-ranitidine cohorts and the associated proportional reporting ratios and 95% confidence intervals in multivariate Cox proportional hazards regression adjusted for competing mortality.
Ijerph 19 12469 g002
Figure 3. Cumulative incidences of single cancers after adjustment for competing risks. (A) Liver cancer, (B) lung cancer, (C) gastric cancer, and (D) pancreatic cancer.
Figure 3. Cumulative incidences of single cancers after adjustment for competing risks. (A) Liver cancer, (B) lung cancer, (C) gastric cancer, and (D) pancreatic cancer.
Ijerph 19 12469 g003
Figure 4. Cumulative incidences of liver cancer for different prescription durations after adjustment for competing risks. Abbreviations: DDDs, defined daily doses.
Figure 4. Cumulative incidences of liver cancer for different prescription durations after adjustment for competing risks. Abbreviations: DDDs, defined daily doses.
Ijerph 19 12469 g004
Figure 5. (A) Cumulative incidences of liver cancer between ranitidine and famotidine users after adjustment for competing risks. (B) Cumulative incidences of liver cancer among ranitidine, non-ranitidine with famotidine, and non-ranitidine without famotidine users after adjustment for competing risks.
Figure 5. (A) Cumulative incidences of liver cancer between ranitidine and famotidine users after adjustment for competing risks. (B) Cumulative incidences of liver cancer among ranitidine, non-ranitidine with famotidine, and non-ranitidine without famotidine users after adjustment for competing risks.
Ijerph 19 12469 g005
Figure 6. Cumulative incidences of liver cancer among ranitidine, non-ranitidine with PPI, and non-ranitidine without PPI users after adjustment for competing risks.
Figure 6. Cumulative incidences of liver cancer among ranitidine, non-ranitidine with PPI, and non-ranitidine without PPI users after adjustment for competing risks.
Ijerph 19 12469 g006
Table 1. Baseline characteristics of the non-ranitidine cohort and ranitidine cohort with over 90 DDDs.
Table 1. Baseline characteristics of the non-ranitidine cohort and ranitidine cohort with over 90 DDDs.
Characteristics Untreated n = 55,110%Ranitidine n = 55,110%p-Value
SexFemale28,79452.2%28,79452.2%1.000
male26,31647.8%26,31647.8%
Age (mean ± SD) 66.8 ± 14.166.8 ± 14.11.000
CCI0–112,44422.6%12,44422.6%1.000
2–316,20529.4%16,20529.4%
4–512,79923.2%12,79923.2%
>513,66224.8%13,66224.8%
HCDNo22,20440.3%22,20440.3%1.000
Yes32,90659.7%32,90659.7%
HyperlipidemiaNo26,07247.3%26,07247.3%1.000
Yes29,03852.7%29,03852.7%
DMNo37,59868.2%37,59868.2%1.000
Yes17,51231.8%17,51231.8%
CKDNo49,70490.2%49,70490.2%1.000
Yes54069.8%54069.8%
AspirinNo25,81046.8%25,81046.8%1.000
Yes29,30053.2%29,30053.2%
StatinsNo32,08958.2%32,08958.2%1.000
Yes23,02141.8%23,02141.8%
ACEIsNo31,21456.6%31,21456.6%1.000
Yes23,89643.4%23,89643.4%
β-BlockersNo16,04729.1%16,04729.1%1.000
Yes39,06370.9%39,06370.9%
FamotidineNo19,84136.0%19,84136.0%1.000
Yes35,26964.0%35,26964.0%
SpironolactoneNo48,32087.7%48,32087.7%1.000
Yes679012.3%679012.3%
GlucocorticoidsNo574810.4%574810.4%1.000
Yes49,36289.6%49,36289.6%
SSRIsNo44,74981.2%44,74981.2%1.000
Yes10,36118.8%10,36118.8%
Antiviral therapyNo54,27198.5%54,27198.5%1.000
Yes8391.5%8391.5%
Abbreviations: DDDs, defined daily doses; SD, standard deviation; CCI, Charlson comorbidity index; HCD, hypertensive cardiovascular disease; DM, diabetes mellitus; CKD, chronic kidney disease; ACEI, angiotensin-converting enzyme inhibitor; SSRIs, selective serotonin reuptake inhibitors.
Table 2. Incidence rates of individual cancers (per 1000 person-years).
Table 2. Incidence rates of individual cancers (per 1000 person-years).
CancersUntreatedIncidence Rate *(95% CI)RanitidineIncidence Rate *(95% CI)
Liver cancer6191.16(1.07–1.25)7111.35(1.25–1.45)
Oral Cancer1910.36(0.30–0.41)1610.30(0.26–0.36)
Esophageal cancer820.15(0.12–0.19)1010.19(0.16–0.23)
Gastric cancer2100.39(0.34–0.44)2550.48(0.43–0.55)
Colon cancer5270.99(0.90–1.07)4920.93(0.85–1.02)
Rectal cancer2860.53(0.47–0.60)3040.58(0.51–0.64)
Pancreas cancer930.17(0.14–0.21)1210.23(0.19–0.27)
Lung cancer5751.07(0.98–1.16)6491.23(1.14–1.33)
Bone cancer30.01(0.00–0.02)40.01(0.00–0.02)
Bladder cancer1770.33(0.28–0.38)1810.34(0.30–0.40)
Renal cancer1540.29(0.25–0.33)1810.34(0.30–0.40)
Thyroid cancer960.18(0.14–0.21)900.17(0.14–0.21)
Skin cancer2030.38(0.33–0.43)1890.36(0.31–0.41)
Breast cancer (female)4070.76(0.69–0.83)4550.86(0.79–0.95)
Uterine cancer670.12(0.09–0.15)530.10(0.08–0.13)
Cervix cancer1660.31(0.27–0.36)1550.29(0.25–0.34)
Ovarian cancer540.10(0.08–0.13)420.08(0.06–0.11)
Prostate cancer3060.57(0.51–0.63)3130.59(0.53–0.66)
All cancers43998.49(8.24–8.73)46829.19(8.93–9.45)
* Incidence rate (per 1000 person-years).
Table 3. Estimates for the association between ranitidine use duration and cancer risk compared with non-ranitidine use by multivariate Cox proportional hazards regression.
Table 3. Estimates for the association between ranitidine use duration and cancer risk compared with non-ranitidine use by multivariate Cox proportional hazards regression.
Liver CancerpGastric CancerpLung CancerpPancreatic Cancerp
Never used1.00 1.00 1.00 1.00
90–180 DDDs *1.03 (0.89–1.18)0.6901.26 (1.00–1.59)0.0491.25 (1.09–1.44)0.0021.64 (1.19–2.26)0.003
181–270 DDDs1.12 (0.93–1.34)0.2201.13 (0.82–1.54)0.4521.09 (0.90–1.32)0.4031.10 (0.69–1.77)0.682
271–360 DDDs1.26 (0.99–1.61)0.0641.27 (0.84–1.93)0.2521.31 (1.02–1.68)0.0320.92 (0.45–1.89)0.816
Over 360 DDDs1.42 (1.22–1.66)<0.0011.33 (1.02–1.74)0.0371.04 (0.87–1.24)0.6581.22 (0.80–1.85)0.358
* Abbreviations: DDDs, defined daily doses.
Table 4. Risk of cancer between famotidine and ranitidine users.
Table 4. Risk of cancer between famotidine and ranitidine users.
CancersFamotidine%Ranitidine%Totalp-ValueHR (95% CI)p-Value
Liver cancer3801.1%4421.3%8220.0321.22(1.06–1.40)0.005
Oral cancer1250.4%1070.3%2320.2370.87(0.67–1.12)0.286
Esophageal cancer520.1%600.2%1120.4511.19(0.82–1.72)0.364
Gastric cancer1420.4%1650.5%3070.2081.19(0.95–1.49)0.122
Colon cancer3651.0%3090.9%6740.0330.86(0.74–1.01)0.059
Rectal cancer1930.5%1950.6%3880.9191.03(0.84–1.26)0.768
Pancreas cancer660.2%800.2%1460.2811.25(0.90–1.73)0.186
Lung cancer3561.0%4001.1%7560.1161.15(1.00–1.33)0.052
Bone cancer *n/an/an/an/an/an/a1.51(0.25–9.03)0.652
Bladder cancer1160.3%1170.3%2330.9481.03(0.80–1.33)0.830
Renal cancer980.3%1280.4%2260.0531.33(1.02–1.73)0.034
Thyroid cancer710.2%630.2%1340.5450.89(0.64–1.25)0.514
Skin cancer1370.4%1370.4%2741.0001.02(0.81–1.30)0.842
All cancers29248.3%30528.7%59760.0861.07(1.02–1.12)0.010
* According to the data protection policy of NHIRD, the data on cancers with <3 cases cannot be provided.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Wang, C.-H.; Chen, I.-I.; Chen, C.-H.; Tseng, Y.-T. Pharmacoepidemiological Research on N-Nitrosodimethylamine-Contaminated Ranitidine Use and Long-Term Cancer Risk: A Population-Based Longitudinal Cohort Study. Int. J. Environ. Res. Public Health 2022, 19, 12469. https://doi.org/10.3390/ijerph191912469

AMA Style

Wang C-H, Chen I-I, Chen C-H, Tseng Y-T. Pharmacoepidemiological Research on N-Nitrosodimethylamine-Contaminated Ranitidine Use and Long-Term Cancer Risk: A Population-Based Longitudinal Cohort Study. International Journal of Environmental Research and Public Health. 2022; 19(19):12469. https://doi.org/10.3390/ijerph191912469

Chicago/Turabian Style

Wang, Chun-Hsiang, I-I Chen, Chung-Hung Chen, and Yuan-Tsung Tseng. 2022. "Pharmacoepidemiological Research on N-Nitrosodimethylamine-Contaminated Ranitidine Use and Long-Term Cancer Risk: A Population-Based Longitudinal Cohort Study" International Journal of Environmental Research and Public Health 19, no. 19: 12469. https://doi.org/10.3390/ijerph191912469

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop