Proton pump inhibitors and mortality in patients with cancer: Unraveling the complex link

Abstract

Proton pump inhibitors (PPIs) are widely utilized in oncology to treat gastroesophageal reflux disease, peptic ulcer disease, and chemotherapy-induced mucosal injury. Emerging evidence has suggested that prolonged use of PPIs is associated with increased mortality in patients with cancer. The findings of Krishnan et al are proof of this but need to be considered cautiously due to potential confounding factors. This letter raised important methodological concerns, including confounding by indication, reverse causality, and polypharmacy that can influence the observed association. While PPIs are important in oncology, their advantages and disadvantages should be weighed judiciously by clinicians. Future prospective studies with robust analytical approaches will be required to ascertain more definite causality.

Key Words: Proton pump inhibitors; Cancer mortality; Confounding by indication; Reverse causality; Polypharmacy; Gastrointestinal complications; Acid suppression therapy; Clinical pharmacology; Oncology treatment; Observational study

Core Tip: The original study by Krishnan et al proposed a link between proton pump inhibitor use and increased all-cause mortality. Our critique highlighted methodological limitations, including residual confounding, indication bias, and a lack of causality assessment. We drew attention to the need for cautious interpretation and advocated for well-designed prospective studies before drawing definitive clinical conclusions.

TO THE EDITOR

We have read with considerable interest the latest study from Krishnan et al[1] on the relationship between proton pump inhibitors (PPIs) and all-cause mortality risk in patients with cancer[1]. The authors performed an exhaustive analysis using stringent methods to test a question of significant clinical concern. They described a seeming link between long-term use of PPIs and higher mortality in patients with cancer, a finding that is of significant concern. Nonetheless, several methodological flaws call for careful interpretation of the results.

PPIs such as esomeprazole, lansoprazole, omeprazole, pantoprazole, and rabeprazole are used frequently in the treatment of gastroesophageal reflux disease, peptic ulcers, and chemotherapy-induced gastrointestinal adverse effects[2]. PPIs are valuable in upper gastrointestinal bleeding prevention and facilitating quality of life in patients with cancer. The latest research has, however, indicated risks such as cardiovascular side effects, renal function impairment, and susceptibility to infection associated with long-term PPI use[3]. The link between PPIs and the mortality of patients with cancer deserves critical examination.

Krishnan et al[1] compared recent initiators of PPI with non-users and H2 receptor antagonist users like cimetidine, famotidine, nizatidine, and ranitidine[1]. The findings revealed that PPI users were at a higher risk of death, with a median follow-up of 5.4 years in PPI users and 6.5 years in non-users. Reported hazard ratios were 2.34 and 2.72 compared with non-users, and even compared with H2 receptor antagonist users, the risk of mortality was higher (1.51-fold). While these findings are alarming, several confounders must be considered before drawing definitive conclusions.

One of the most significant limitations is confounding by indication, as PPIs are often prescribed for patients with cancer and gastroesophageal reflux, peptic ulcers, or mucosal injury from chemotherapy[4]. These conditions, not PPI use per se, might explain the increased mortality. Without adequate correction for these confounders, the association between PPIs and mortality may be overestimated. Previous studies have already shown that patients requiring PPIs tend to have more severe comorbid conditions, which can independently result in increased risk of mortality[5].

The second significant concern is reverse causality, which means that PPI use is more frequent in patients with advanced cancer or increased symptom burden and is not a cause of increased mortality by itself. Research has established that PPIs are frequently used in palliative care for symptom control of nausea, dyspepsia, and stress due to pain[6]. This suggests that the association in the study could be a consequence of the advanced stage of disease and not a causal impact of PPI exposure. The study conclusions are questionable unless the possibility is ruled out using proper sensitivity analyses.

Also, the impact of polypharmacy must be scrutinized with caution. Patients with cancer are administered more than a single drug, like corticosteroids, opioids, and anticoagulants, that can interact with PPIs and impact overall health outcomes. Polypharmacy was identified as an independent predictor of adverse outcomes in patients with cancer, and drug-drug interaction is a probable cause of higher morbidity and mortality[7]. Previous studies have also shown that exposure to PPIs in combination with other drugs would change drug metabolism and even compound existing risk factors[8]. Levelling the exposure to and interaction between PPIs at an equal level is necessary to reveal the actual link between PPI exposure and death.

Clinically, it is important to balance the risks and benefits of PPI therapy. Withdrawing PPIs solely on mortality concern grounds has the potential for unforeseen outcomes, like possibly escalating the risk of upper gastrointestinal bleeding, esophagitis, and worse quality of life[9]. Clinicians should judiciously consider the need for PPI therapy on an individual basis, so that they do not withhold appropriate therapy from patients who benefit from acid suppression. More robust study designs are needed to enhance the validity of further research on this topic. Propensity score-matched prospective cohort studies and specific sensitivity analyses must be conducted to reduce the impact of confounding factors[10]. Where possible, randomized controlled trials would yield the best level of evidence to estimate the authentic effect of PPIs on mortality among patients with cancer.

In conclusion, the study by Krishnan et al[1] is valuable and asks an important clinical question regarding the safety of long-term PPI use, but the following methodological limitations highlight the need for cautious interpretation[1]. Confounding by indication, reverse causality, and polypharmacy may have been responsible for the associations observed, and hence additional, better-designed prospective studies are indicated. Until more persuasive data are forthcoming, PPIs should still be prescribed judiciously, with close weighing of their risks vs benefits in patients with cancer.

ACKNOWLEDGEMENTS

We thank the reviewers for their comments that helped to improve the manuscript.