Ranitidine Use and Risk of Upper Gastrointestinal Cancers

In 2019, contamination with N-nitrosodimethylamine (NDMA) was reported in the acid-reducing histamine-2 receptor blocker (H2RB) ranitidine, but not in other H2RBs (e.g., cimetidine and famotidine; ref. 1). NDMA, an N-nitroso compound, is classified by the International Agency for Research on Cancer as “probably carcinogenic to humans,” based on “sufficient evidence” of carcinogenicity in animals and “no adequate data” in humans (2). In light of the hypothesis that NDMA may contribute to the development of cancer in the gastrointestinal tract (3), ranitidine was recalled by several agencies. The degree of NDMA contamination in various ranitidine products was found by the FDA to range between 0.004 and 0.86 μg per tablet or oral dose (4), although many companies have not tested their drug products for the presence of NDMA (5). Oral ranitidine, compared with placebo, was found in a small randomized controlled trial not to increase urinary excretion of NDMA (6).

However, at least two areas of profound uncertainty remain. First, the reported levels of NDMA contamination in ranitidine tablets differ by several orders of magnitude between laboratories (2), and the variation among batches and the impact of storage, transportation, temperature, and endogenous metabolism remain to be established. Although it is unclear how long NDMA has been present as an impurity in ranitidine products, its source is not known to be a recent change in the manufacturing process (7). Second, although NDMA is an animal carcinogen, decades of research have not documented a causal association with any human cancer type or site (2), leaving unresolved questions about the impact of NDMA on human health, especially at concentrations reportedly detected in ranitidine.

Given these uncertainties, analyses of cancer risk following ranitidine use per se—rather than studies based on debatable estimates of NDMA exposure—are more informative about possibly increased cancer risk from use of this drug. Analyses focusing on upper gastrointestinal tract cancer, which is directly exposed to the drug, are also more likely to yield informative findings than analyses of other cancer sites. Research on ranitidine use in relation to risk of cancers of the esophagus, stomach, pancreas, and liver has been inconclusive (8–15). This lack of clarity stems in part from potential confounding by indication and reverse causality, which favor spurious positive association (8, 9, 12, 14). In several studies of ranitidine and cancer risk, additional limitations include self-reported drug use, incomplete confounder adjustment, limited sample size, short follow-up, and conflation of disparate cancer types.

To reduce these concerns, we undertook a cohort study based on data from the Danish Prescription Registry. Its design emulated a randomized trial and thereby reduced confounding by indication (16) by comparing cancer risk between ranitidine users and an active comparator group comprising individuals prescribed other H2RBs. We included a second active comparator group of individuals prescribed proton pump inhibitors (PPI), which are used for similar indications.

We conducted this study within the entire Danish population of approximately 5.8 million persons. All Danish residents are provided free access to health care (17). Since 1968, the Civil Registration System has assigned a unique civil registration number to all residents at birth or upon immigration (18). The system also records sex and date of birth, and tracks changes in vital status, and migration for the entire population. The registration number allows unambiguous record linkage at the individual level.

The Danish National Prescription Registry contains complete data on prescription medications dispensed from community pharmacies in Denmark since 1995 (19). It does not include medications used in-hospital. We used this registry (codes defined in Supplementary Table S1) to identify all adult (18 years or older) first-time users of ranitidine. We defined first-time users as persons who redeemed their first prescription for ranitidine between 1996 and 2008 and had no previous prescriptions for nizatidine, because nizatidine oral solution may contain high levels of NDMA (1). As active comparators, we identified all adult first-time users of cimetidine or famotidine, hereafter categorized as other H2RBs, and PPIs, defined as persons who redeemed their first prescription for these drugs between 1996 and 2008 and had no previous prescriptions for ranitidine or nizatidine. Specific PPIs were combined due to the overlap in indications for this drug class. We did not include a comparison group of adults taking neither PPIs nor other H2RBs, because comparisons with such a group would be strongly biased due to confounding by indication and reverse causality.

We defined the index date as the date of the first prescription. We allowed ranitidine users to redeem prescriptions for other H2RBs or PPIs, but censored users of other H2RBs and PPIs if they redeemed a ranitidine prescription during follow-up, with a 2-year lag-time. As a sensitivity analysis, we did not allow ranitidine users to have redeemed prescriptions for other H2RBs or PPIs. Statistics on over-the-counter use of H2RBs and PPIs have been available since 1999 in Denmark. The proportion of ranitidine defined daily doses (DDD) sold by prescription was 84% in 1999, declining to approximately 50% in 2004 to 2011, and to approximately 20% in 2012 to 2017 (Supplementary Table S2). Similarly, the proportion of total H2RB DDDs sold by prescription was 83% in 1999, declining to 60% in 2004 to 2007; data are not available for 2010 to 2017. In contrast, 97% to 98% of PPIs were sold by prescription throughout the period.

We also retrieved information on use of selected drugs (defined in Supplementary Table S1) potentially associated with risk of upper gastrointestinal cancer and thiazolidinedione recorded before or on the index date with a 90-day look-back period. We also identified use of drugs associated with bleeding risk, including platelet inhibitors, and anticoagulants, as such drugs could alter the threshold for diagnosis of cancer.

The Danish National Patient Registry has recorded all inpatient admissions to all Danish hospitals since 1977, and hospital outpatient clinic and emergency room visits since 1995 (20). We retrieved all diagnoses of gastrointestinal disease, indicating H2RB treatment, including Barrett's esophagus, gastroesophageal reflux disease, and peptic ulcers, recorded before or on the index date, with a 10-year look-back period. We also retrieved data on other chronic diseases, including diabetes, chronic obstructive pulmonary disease, ischemic heart disease, and alcohol-related disease (see Supplementary Table S3 for definitions and disease codes). The Danish Pathology Registry contains information coded using SNOMED on all specimens examined in Danish departments of pathology. From this registry, we retrieved information on Barrett's esophagus diagnosed before or on the index date.

We followed members of the three cohorts starting one year after their index date (to avoid reverse causation) for an incident diagnosis of cancer recorded in the Danish Cancer Registry (21) that has recorded cancer diagnoses since 1943, with accurate and nearly complete case ascertainment. Cancer diagnoses are recorded using International Classification of Diseases, Tenth Revision (ICD-10) codes, and codes from the ICD for Oncology, Third Revision (ICD-O-3) for topography and morphology. The outcomes of interest were esophageal cancer (any, adenocarcinoma, and squamous cell carcinoma), stomach cancer (any, proximal, distal, and unknown/several regions), hepatocellular carcinoma, and pancreatic cancer. If a subject had cancer at more than one site, all were included in the analysis. We excluded subjects with a cancer diagnosis recorded before the start of follow-up, except for non-melanoma skin cancer.

The Danish Register of Causes of Death includes age, place, and cause of death (coded according to ICD-10 since 1994; ref. 22). We obtained information on deaths due to cancer. We included cancer-specific deaths in the outcome definition if the cancer was not identified through the Cancer Registry.

The study was reported to the Danish Data Protection Agency through registration at Aarhus University, with exemption from informed consent (record number KEA2017‐36/812).

Starting one year after the date of the first prescription, we followed cohort members until the diagnosis date of a cancer of interest for a given analysis (i.e., without censoring for other cancer types), date of death, emigration, or end of the period (December 31, 2018), using an intention-to-treat approach. We censored users of other H2RBs and PPIs if they redeemed a ranitidine prescription, with a 2-year lag-time. We computed the number of events (incident cancer or cancer-specific death) per person-years at risk and the 10-year cumulative risk (with death from other causes as a competing risk) of each cancer type by exposure status.

We used logistic regression, including all available covariates, to compute propensity scores for exposure to ranitidine. In the models, we included age and index dates as splines, along with use of PPIs in the previous two years when comparing with H2RBs and use of other H2RBs in the previous two years when comparing with PPIs. To control for potential confounding, we then used the propensity scores to compute stabilized inverse-probability-of-treatment (sIPT) weights (23). We assessed the covariate balance after weighting using standardized mean differences. The use of sIPT weighting permitted estimation of the average treatment effect through comparison of exposed versus unexposed populations with covariate distributions resembling the distribution in the overall population.

We compared the ranitidine-exposed cohort with each of the two comparison cohorts by estimating crude and sIPT-weighted risk curves, considering death from other causes as a competing risk. We obtained the 10-year cumulative risk and calculated crude and weighted 10-year risk differences with 95% confidence intervals (CI) estimated by bootstrapping. We calculated hazard ratios (HR) using Cox regression analysis, including the crude and sIPT-weighted observations, with 95% CIs estimated by bootstrapping. We confirmed the proportionality assumption using log–log plots.

To address treatment duration, we conducted analyses restricted to persons who redeemed at least five or at least 10 prescriptions for ranitidine, other H2RBs, or PPIs. We started follow-up one year after the fifth prescription or the tenth prescription, respectively, and we included time between the first and the fifth/tenth prescription as a variable when computing propensity scores. In addition, in the analysis of subjects who redeemed at least 10 prescriptions, we separately analyzed the first 10 years of follow-up and follow-up after 10+ years, to focus on the subpopulation with relatively high exposure and long follow-up time.

We used SAS version 9.4 (SAS Institute) for analyses.

Our study encompassed 103,565 first time users of ranitidine, 182,497 incident users of other H2RBs, and 807,725 incident users of PPIs. Table 1 summarizes characteristics of the three analytic cohorts. During median follow-up of 14 years [interquartile range (IQR), 10–18] among ranitidine users, we ascertained a total of 276 esophageal, 342 stomach, 133 hepatocellular, and 517 pancreatic cancers (Table 1).

Figure 1 and Supplementary Fig. S1 show sIPT weighted and crude HRs for each cancer site. In the primary analyses, which included individuals with at least one prescription of interest during the entire follow-up period, all HRs except one were clustered around the null value of 1.0, whether ranitidine was compared with other H2RBs or with PPIs. In both comparisons, we found an approximately 30% increased risk of esophageal adenocarcinoma, but not esophageal squamous cell carcinoma, among ranitidine users. However, the 10-year cumulative risk difference between the ranitidine and the comparator cohorts was 0.02% (95% CI, 0.00–0.03; Fig. 1).

The pattern of results was virtually identical in analyses restricted to individuals with at least five prescriptions of ranitidine, other H2RBs, or PPIs (Supplementary Fig. S2), as well as after restriction to those with at least 10 prescriptions, although statistical precision was lower for the latter comparisons (Fig. 2). We found no evidence of an increase in risk with a larger number of prescriptions; on the contrary, HRs remained clustered around 1.0 in comparisons with other H2RBs and with PPIs, and no single HR was statistically significant. In the analysis of subjects with 10 or more prescriptions, the sIPT-weighted HR for adenocarcinoma of the esophagus was 0.97 (95% CI, 0.53–1.76) for ranitidine compared with other H2RBs, and 0.62 (95% CI, 0.38–1.00) for ranitidine compared with PPIs.

We also analyzed the impact of follow-up time (i.e., latency) among subjects with 10 or more prescriptions. Because of small numbers, we analyzed each of the four cancer sites without further subgrouping of esophageal and stomach cancers. Case numbers were insufficient (n < 10 among ranitidine users) to estimate the risk of liver cancer after more than 10 years of follow-up. We found no evidence of higher risk for esophageal, stomach, or pancreatic cancer in association with long-term use of ranitidine versus other H2RBs or PPIs (Fig. 3).

The sensitivity analysis restricting to ranitidine users without previous use of other H2RBs or PPIs yielded virtual similar results as the main analysis (Supplementary Fig. S3)

In this large, population-based study using high-quality exposure and outcome data, we consistently observed HRs and risk differences close to unity when use of ranitidine was compared with use of either other H2RBs or PPIs. Analyses by number of prescriptions and duration of follow-up showed no evidence of trends. Although analyses restricted to participants with at least five or 10 prescriptions relied on smaller numbers of incident cancers, the estimates remained clustered around one, with no evidence of significantly increased risk of any evaluated malignancy.

Because we calculated many estimates, some (on average 1 in 20) are expected to be significant due to chance. Thus, chance might explain the modest excess risk of esophageal adenocarcinoma among ranitidine users compared with users of other H2RBs or PPIs; such an explanation is supported by the lack of consistent observation of this association across other studies. The observed associations are unlikely to be due to confounding by indication, which was similar among comparison groups, or by lifestyle factors such as smoking and alcohol use, which are causes of esophageal squamous cell carcinoma but not adenocarcinoma. If a causal association existed, we would expect to observe stronger associations with a larger number of prescriptions and, most likely, with longer follow-up, yet such patterns were not evident.

Comparison of our results with those in the literature on ranitidine and risk of upper gastrointestinal cancers is hampered by differences in study design. The few earlier studies were limited by evidence of reverse causality, confounding by indication, small sample size, shorter follow-up time, or combination of malignancies with different etiologies (8–14). Two studies detected a higher risk of upper gastrointestinal cancers following use of PPIs compared with H2RBs (12, 14), whereas one study found a similar risk of liver cancer among users and non-users of ranitidine and among users and non-users of other H2RBs (12).

Only two cohort studies attempted to emulate a randomized trial by, including an active comparator cohort and restricting the analyses to incident users of PPIs or H2RBs. In the first study, which followed US veterans for cause-specific mortality over a median duration of 10 years (14), a 41% higher mortality from all upper gastrointestinal cancers combined was observed among PPI users compared with H2RB users (90% of whom used ranitidine). Hence, this study provided no information on whether ranitidine itself or NDMA contamination increased the risk of any cancer by type or at any particular site. However, its findings suggest that if a causal effect exists, the risk is further increased following use of PPIs. The second study used a claims database in Japan (5) to follow users of ranitidine/nizatidine or other H2RBs for cancer incidence over a median duration of 2.4 years. Despite its careful design, it found no significant differences in the risk of overall or any site-specific cancers, including stomach, pancreatic, and other sites (but not esophagus or liver, which were not analyzed separately). However, the short duration of follow-up limited inferences based on these findings.

Strengths of our study include its population-based design, with two active comparator groups that emulate a randomized trial; unbiased assessment of prescriptions; the ability to identify first time rather than prevalent users of the drugs by application of a 1-year lag time; the separate assessment of etiologically distinct cancer types identified through linkage to the Cancer Registry; the use of propensity scores to adjust for confounding at baseline (23); and complete follow-up for cancer. Although approximately half of ranitidine users had only one prescription, a substantial number had five or even 10 prescriptions, enabling analyses of long-term exposure with up to 20 years of follow-up.

Our study also has limitations. Because malignant transformation of cancer may take longer than 20 years, the most severe shortcoming of our study is the limited number of participants with long-term follow-up. Second, the Prescription Registry provides no information about drugs removed from the Danish market, nor about compliance with prescriptions or use of drugs sold over-the-counter. Because the drugs examined relieve severe gastrointestinal symptoms, compliance with prescriptions was probably high, at least for chronic use. Although we were able to rely on the Prescription Registry to identify most ranitidine users with the longest follow-up (i.e., those first prescribed ranitidine in 1999 to 2001; Supplementary Table S3), over-the-counter use inevitably led to underestimation of ranitidine exposure, as well as exposure to other H2RBs, in more recent years. The potential direction of bias in HR estimates is unclear, but the magnitude of any bias would be limited in the comparison between users of ranitidine and other H2RBs, because all major H2RBs were similarly impacted by over-the-counter use over time.

We conclude that our study provides little evidence that ranitidine, whether through NDMA contamination or any other reason, increases risk of upper gastrointestinal cancers. This conclusion is based on the assumption that no other relevant differences than NDMA content exist between ranitidine (and also nizatidine) and other H2RBs. And it pertains to the prevailing duration of treatment in Denmark and presumably also in other countries. Studies of individuals with longer exposure may not be feasible because ranitidine now has been removed from several markets worldwide. Hence, extended follow-up of our cohort may be the most realistic and informative approach to understanding the human health impact, if any, of NDMA contamination of ranitidine.

H.-O. Adami and E.T. Chang report consulting for Sanofi in ranitidine litigation, and E.T. Chang has provided ongoing consulting support for Sanofi in other matters. Sanofi had no involvement in their work on this publication. The company also did not provide any financial or other support for this study. Hence, the analytical approach, conduct of the analyses, interpretation of the results, and conclusions drawn are exclusively the work of the authors. No parties other than the authors had any role in the study design, data collection and analysis, article preparation, or decision to publish, and no outside parties reviewed any portion of the article before its submission for publication. E.T. Chang reports other support from Sanofi outside the submitted work. No disclosures were reported by the other authors.

H.-O. Adami: Writing–original draft. I. Trolle Andersen: Formal analysis. U. Heide-Jørgensen: Formal analysis. E.T. Chang: Writing–original draft. M. Nørgaard: Formal analysis. H. Toft Sørensen: Writing–original draft.

This study received no funding.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.