Helicobacter pylori eradication treatment and the risk of Barrett's esophagus and esophageal adenocarcinoma

Helicobacter pylori (H. pylori) is associated with lower risks of Barrett's esophagus and esophageal adenocarcinoma, but whether H. pylori eradication increases the risk of these conditions is unknown. This study aimed to test the hypothesis that H. pylori eradication leads to gradually increased risks of Barrett's esophagus and esophageal adenocarcinoma over time, while esophageal squamous cell carcinoma was assessed for comparison reasons.


| INTRODUC TI ON
The incidence of esophageal adenocarcinoma has increased markedly in Western countries since the 1970s. 1 In most Western countries (including Sweden), the incidence of esophageal adenocarcinoma has surpassed that of esophageal squamous cell carcinoma, which is the most common histologic type of esophageal cancer globally. 2 For esophageal adenocarcinoma and its precursor lesion Barrett's esophagus, the main risk factors are gastroesophageal reflux disease (GERD) and obesity, while presence of Helicobacter pylori (H. pylori) in the stomach is associated with a 32%-56% decreased risk of Barrett's esophagus, 3 and a 36%-44% decreased risk of esophageal adenocarcinoma. 3 These inverse associations with H. pylori are thought to be the result of H. pylori-associated gastric atrophy, which leads to lower gastric acid production and thus lower prevalence of GERD. Yet, it is not studied if eradication treatment for H. pylori increases the risks of Barrett's esophagus or esophageal adenocarcinoma. H. pylori infection is otherwise a well-established risk factor of noncardia gastric adenocarcinoma, and its eradication reduces the risk of this cancer by around 50%. [4][5][6] For esophageal squamous cell carcinoma, the main risk factors are tobacco smoking and overconsumption of alcohol, and no association has been found with H. pylori.
To test the hypothesis that H. pylori eradication increases the risks of Barrett's esophagus and esophageal adenocarcinoma, and does not influence the risk of esophageal squamous cell carcinoma, we conducted a nationwide Swedish cohort study.

| Design
This population-based and nationwide Swedish cohort study included patients aged 18 years and older who had eradication treatment for H. pylori during the study period July 1, 2005, to December 31, 2012.
The source cohort has been described in detail previously. 5,7 For the purpose of the present study, the risk of Barrett's esophagus, esophageal adenocarcinoma, and esophageal squamous cell carcinoma among the cohort participants was compared to the risk in the corresponding Swedish background population. The data were retrieved from well-maintained nationwide Swedish registers that were linked for each study participant by their unique personal identity number assigned to all residents in Sweden at birth or immigration. The study was approved by the Regional Ethical Review Board in Stockholm (2014/1291-31/4), and the need for informed consent was waived.

| Exposure
The study exposure was eradication treatment for H. pylori using a proton-pump inhibitor (PPI) in combination with at least two of the antibiotics clarithromycin, amoxicillin, or metronidazole, as described earlier. 7 The data on eradication treatment were retrieved from the Swedish Prescribed Drug Registry, which started on July 1, 2005, and contains information on all prescribed and dispensed medications that are used outside in-hospital care for the whole Swedish population. 8 The medications (with their Anatomical Therapeutic Chemical codes) representing H. pylori eradication were a combination package used for H. pylori eradication containing the PPI esomeprazole and the antibiotics clarithromycin and amoxicillin (A02BD06), and prescriptions of a PPI (A02BC) together with clarithromycin (J01FA09), amoxicillin (J01CA04), or metronidazole (J01XD01). The separate prescriptions had to include at least two antibiotics that were prescribed on the same date, and a PPI prescribed within a window of 60 days before to 5 days after the antibiotics. This was done in order to include individuals already using a PPI and to take nonavailability in the pharmacy into account.

| Outcomes
The main outcome was Barrett's esophagus, and the secondary outcome was esophageal adenocarcinoma. Esophageal squamous The histologic code 096 from the C24 WHO classification of histology defined adenocarcinoma, and the code 146 defined squamous cell carcinoma. All outcomes occurring within 1 year of eradication treatment were excluded to avoid detection bias, that is, earlier detection of the outcomes due to examinations preceding the eradication treatment. Therefore, the start of follow-up was 1 year after H. pylori eradication treatment for all participants. In each individual, only the first ever cancer episode was considered. Thus, participants were eligible only if they had no history of cancer. Subgroup analyses were performed to assess the risk of the outcomes over time after eradication treatment (categorized into 1-2, 3-4, or 5-7.5 years) and number of eradication treatments (1, 2, or ≥ 3). Multiple eradication treatments indicated that H. pylori was present for a prolonged amount of time.

| Statistical analysis
The statistical software STATA (Stata Corp v. 13.0) was used for all data management and analyses.

| Participants
The study cohort included 81 919 individuals with at least one dispensed prescription for H. pylori eradication treatment during the study period. Of these participants, 53.9% were women, 60.9% were 59 years of age or younger, and 74.8% resided in urban areas (

| Helicobacter pylori eradication treatment and risk of Barrett's esophagus
For Barrett's esophagus, the overall SIR was increased (3.67, 95% CI 3.15-4.25), but the SIRs did not increase over time after eradication treatment, but rather decreased, from 4.32 (95% CI 3.53-5.23) at 1-2 years-3.09 (95% CI 1.98-4.59) at 5-7.5 years after eradication ( Table 2). Analysis from the date of last eradication treatment showed a similar trend apart from the analysis by number of eradications (Table 3). There was no clear trend in risk depending on the number of eradication episodes (Table 2), and there were no clear differences in risk between the sexes (Table 4). Individuals aged 59 years or younger had a higher SIR of Barrett's esophagus than the oldest age group at any follow-up time (Table 4).

| Helicobacter pylori eradication treatment and risk of esophageal adenocarcinoma
The overall SIR of esophageal adenocarcinoma was slightly increased after H. pylori eradication treatment although not statistically

| Helicobacter pylori eradication treatment and risk of esophageal squamous cell carcinoma
The overall SIR of esophageal squamous cell carcinoma was slightly increased after H. pylori eradication treatment, but the result was not statistically significant, and the SIRs remained stable over time after eradication treatment ( Table 2). Analysis from the date of last eradication treatment showed similar results (Table 3).

| D ISCUSS I ON
This study provided no support for the hypothesis of a gradually in-  To the best of our knowledge, this was the first study that examined the association between eradication treatment for H. pylori Although several studies suggested that H. pylori is protective against esophageal cancer, a recent meta-analysis did not provide strong and consistent evidence for this hypothesis. 13 Although most sub-analyses in that meta-analysis suggested lower risks compared to the comparison groups in the individual studies, methodological heterogeneity was high and there were important variations based on study characteristics (including etno-geographical differences). 13 Our study uses the entire Swedish population as a comparison group, and we have excluded the first year after eradication to reduce the risk of reverse causality (ie, individuals being diagnosed and treated for early signs of yet undiagnosed cancer). Yet, individuals who received eradication treatment may also be followed up more closely leading to earlier detection, and therefore creating a risk of detec-

| CON CLUS ION
To conclude, this first study assessing the risk of Barrett's esophagus and esophageal adenocarcinoma after eradication treatment for H. pylori did not show any gradually increased risk over time after treatment. These findings do not provide any strong evidence to refrain from H. pylori eradication whenever indicated, even in individuals at increased risk of esophageal adenocarcinoma.

CO N FLI C T O F I NTE R E S T
The authors have nothing to disclose. The funding sources have not been involved in the design, conduct, or presentation of the study.

AUTH O R S ' CO NTR I B UTI O N S
All authors contributed to the study conception and design. Data