No Association Between Chronic use of Ranitidine, Compared With Omeprazole or Famotidine, and Gastrointestinal Malignancies

Yeseong D. Kim; Jiasheng Wang; Fahmi Shibli; Kamrine E. Poels; Stephen J. Ganocy; Ronnie Fass

Disclosures

Aliment Pharmacol Ther. 2021;54(5):606-615. 

In This Article

Results

Demographics and Baseline Patient Characteristics

Incidence data were queried for new malignancy diagnoses in the ranitidine, famotidine, and omeprazole cohorts. A total of 581 028 patients were prescribed ranitidine, 909 970 were prescribed famotidine, and 2 179 048 were prescribed omeprazole, and continued in various intervals throughout the 2009–2018 10-year window. Demographic data were collected for all study and control cohorts over a 10-year window, as depicted in Table 1.

Statistical analyses of significance of differences in proportions were performed for 16 baseline patient characteristics between ranitidine and famotidine, as well as ranitidine and omeprazole, as presented in Table 2. In the ranitidine vs famotidine analyses, race (Asian and other), tobacco use, DM, obesity, cirrhosis, IBD, and atrophic gastritis showed significant differences, with famotidine showing higher rates in all categories except other race. In the ranitidine vs omeprazole comparisons, age (below and over 65), sex (male and female), race (African American and Caucasian), DM, cirrhosis, and BO demonstrated significant differences, with ranitidine showing higher rates in age below 65, female sex, African American race, and lower rates in DM, cirrhosis, and BO.

Incidences of New Gastrointestinal Malignancies

In the ranitidine cohort, 180 (0.031%), 170 (0.029%), 880 (0.15%), 320 (0.055%), and 1370 (0.24%) developed new oesophageal, gastric, hepatocellular, pancreatic, and colorectal malignancies, respectively. In the famotidine cohort, 550 (0.060%), 670 (0.074%), 4520 (0.50%), 1120 (0.12%), and 5330 (0.59%) developed new oesophageal, gastric, hepatocellular, pancreatic, and colorectal malignancies, respectively. In the omeprazole cohort, 1260 (0.058%), 1160 (0.053%), 5440 (0.25%), 2060 (0.094%), and 7190 (0.33%) developed new oesophageal, gastric, hepatocellular, pancreatic, and colorectal malignancies, respectively. Combined incidences of new GI malignancies were highest in the first 1–2 years after starting anti-reflux therapy and consistently decreased throughout the study period (Figure 2). Patient characteristics, demographics, and risk factor data are included in Table 1.

Figure 1.

Flow chart diagram of study algorithm describing patient selection rationale and sample sizes of target cohorts

Figure 2.

Combined incidences of all GI malignancies for ranitidine, famotidine, and omeprazole queried over a 10-year period

Multivariable Regression Modeling Using Selected Risk Factors for Each Cancer

Table 3 shows the OR and confidence intervals (CI) derived from the multivariable logistic regression modelling which corrected for common risk factors for each cancer. Patients on ranitidine compared to patients on famotidine had ORs of 0.51, 0.43, 0.39, 0.54, and 0.46 of developing oesophageal, gastric, hepatocellular, pancreatic, and colorectal cancers, respectively. Compared to patients on omeprazole, patients on ranitidine had ORs of 0.62, 0.58, 0.81, 0.68, and 0.66 of developing oesophageal, gastric, hepatocellular, pancreatic, and colorectal cancers, respectively. As seen, ORs were significantly below 1 for all malignancies analysed, with ranitidine showing a further reduced OR compared to famotidine than to omeprazole. P values were <0.001 for all ORs calculated.

Multivariable Regression Modeling Using Demographic Factors

Table 4 describes the ORs and CI derived from the second regression analysis which corrected for demographic factors and one common cancer risk factor with the largest weight for each cancer. Patients on ranitidine compared to patients on famotidine had ORs of 0.56, 0.49, 0.38, 0.50, and 0.49 of developing oesophageal, gastric, hepatocellular, pancreatic, and colorectal cancers, respectively. Compared to patients on omeprazole, patients on ranitidine had ORs of 0.80, 0.93, 0.97, 0.84, and 0.97 of developing oesophageal, gastric, hepatocellular, pancreatic, and colorectal cancers, respectively. P values were <0.001 for all ORs calculated. The ORs and CIs of the common cancer risk factors and demographic predictors included in the regression analyses are included in Table S1.

The comparison against famotidine were similar to the demographic regression model, in that the ORs were well below 1. The comparison against omeprazole, however, revealed that the odds of developing cancers of the stomach, liver, or colon/rectum in patients taking ranitidine were equivalent to the odds in patients taking omeprazole. These results suggest that as compared with famotidine and omeprazole, patients consuming ranitidine are at a lower risk when compared to patients consuming famotidine, and a relatively equivalent risk when compared to patients consuming omeprazole, of developing gastrointestinal malignancies.

Data Validation

To validate our data, we obtained annual incidence data of the five gastrointestinal malignancies studied from 2009 to 2018 (Figure S1). Average incidence ratios (cases per 100 000) were 4.66, 5.54, 9.03, 7.35, and 3.84 for oesophageal, gastric, hepatocellular, pancreatic, and colorectal cancers each, respectively (Table 5). These values show reasonable consistency with rates reported in the literature,[19–23] suggesting that the search strategy utilised and cohorts generated contain accurate data.

processing....