Impact of proton pump inhibitors on the onset of gastrointestinal immune‐related adverse events during immunotherapy

Abstract Introduction The gut microbiota (GM) can influence the pathogenesis of immune‐mediated adverse events (irAEs). Proton pump inhibitors (PPIs) can affect the integrity of GM, but their role in promoting irAEs is still poorly understood. Methods In this retrospective single‐center cohort study, the primary endpoint was the evaluation of the incidence of gastrointestinal (GI) irAEs in cancer patients on PPIs (exposed) versus cancer patients who were not on PPIs (unexposed). Results Three hundred and sixty three patients' records (248 M/115F, median age 69) were reviewed. Twenty‐three exposed patients (92%) developed GI irAEs while only two unexposed patients (8%) developed GI irAEs (hazard ratio [HR] 13.22, 95% confidence interval [CI] 3.11–56.10, p < 0.000). This HR was confirmed after weighting for the propensity score (HR15.13 95% CI 3.22–71.03, p < 0.000). Conclusion Chronic PPI use is associated with an increased risk of GI irAES.


| INTRODUCTION
Immune checkpoint inhibitors (ICIs) play an increasingly relevant role in the management of solid tumors.Although the efficacy and durability of response with ICIs have been well established, one of the major issues is the high rate of immune-related adverse events (irAEs) during the treatment. 1The mechanism of toxicity varies according to ICI, and may ultimately affect the overall management.Gastrointestinal (GI) toxicity is among the most frequently reported irAEs.Overt colitis occur in 8%-27% of the patients, while the incidence of diarrhea alone has been reported to be as high as 54% among the patients on ICIs. 2 Hepatitis is described in 5%-10% of the patients during ICI monotherapy and in 25%-30% during a combination of anti-CTLA-4 and anti-PD (L) 1. 3 The intestinal toxicity of ICIs has been associated with gut microbiota (GM) alterations, particularly to a significant increase of Bacteroides intestinalis. 4More in depth, Firmicutes have been implicated in a higher incidence of irAEs while Bacteroidetes positively correlated with a lower incidence. 5The GM may therefore directly influence the pathogenesis of irAEs, and indirectly through the regulation of metabolites, cytokines, and immune cells. 6Moreover, the imbalance of gut-liver axis caused by GM dysbiosis and/or gut mucosal barrier damage leads to various types of liver diseases, 7 while a higher GM diversity seems to be a protective factor against irAEs. 6Indeed, the concomitant drugs given to these patients may alter GM diversity.For example, antibiotics can modify the GM composition, increasing inflammasome signaling, and thus promoting a pro-inflammatory state, susceptible to GI irAEs. 8roton pump inhibitors (PPIs) are among the most commonly prescribed drugs worldwide, although without evidence-based indication in many cases. 9PPIs may favor the onset of immune-mediated disorders via multiple mechanisms, including GM alterations, malabsorption of nutrients and vitamins, and possibly via other unknown effects. 10Indeed, PPI treatment was associated with distinct taxonomic alterations: in the upper GI tract, PPIs users showed an overgrowth of orally derived bacteria, mostly Streptococcaceae.In fecal samples, PPI increased multiple taxa from the orders Bacillales and Lactobacillales, the families Pasteurellaceae and Enterobacteriaceae and the genus Veillonella. 11Growing evidence shows that PPIs affect the integrity of GM and their use is consistently associated with profound changes in GM with a reduced α diversity. 12espite the abovementioned premises, the role of PPIs in promoting irAEs has been poorly addressed.Hence, we sought to describe the incidence of GI irAEs in patients with solid tumors undergoing immunotherapy and concomitant PPIs.

| Study design and setting
In this retrospective single-center cohort study, we reviewed all the patients with solid tumor who had received immunotherapy with or without chemotherapy at Medical Oncology Unit of Fondazione "IRCCS Policlinico San Matteo di Pavia" between January 2016 and October 2022.The study was conducted according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement for reporting observational studies 13 and was approved by the local Ethics Committee (Comitato Etico Area Pavia) and Institutional Review Board (P-0004914/23) according to the ethical guidelines of the 1975 Declaration of Helsinki.All the subjects signed, before the initiation of treatment, an informed consent provided by the Fondazione IRCCS Policlinico San Matteo at hospitalization.
We defined "exposed" all the cancer patients who had received immunotherapy and concomitant PPIs, conversely all the cancer patients who had received immunotherapy without concomitant PPIs were defined 'unexposed'.
The primary endpoint was the evaluation of the incidence of GI irAEs in the cancer patients who were on PPIs (exposed) versus the cancer patients who were not on PPIs (unexposed).The secondary objectives were: (i) evaluation of Overall survival (OS) between exposed and unexposed; (ii) evaluation of Progression Free Survival (PFS) between exposed and unexposed.

| Data collection
Data were collected from the hospital's electronic patient records, including sociodemographic (age, sex) and clinical characteristics, such as type of cancer, TMN stage, treatment, the use of PPIs and/or antibiotics, the onset of IRAEs, the comorbidities (diabetes mellitus, heart disease, autoimmune diseases).
The inclusion criteria were: (i) patients aged 18 and older, regardless of gender; (ii) treatment with immunotherapy alone or in combination with chemotherapy; (iii) patients who received at least 3 months of ICIs; (iv) signing of informed consent.Patients with unavailable and incomplete basic characteristics, laboratory data, and follow-up information under 3 months were excluded from the study.

| Statistical analysis
All statistical analyses were performed with the Stata 17 (StataCorp.).A 2-sided p-value<0.05was considered statistically significant.Continuous data were reported as median and quartiles (IQR); categorical data were reported as counts and percent.They were compared between cohorts with the Mann Whitney U test and the Fisher exact test respectively.
The incidence of GI irAEs was computed for each cohort as number of events per 100-person year.The cumulative irAE-free survival (defined as the time free from irAEs from the start of ICIs therapy) was computed and plotted using the Kaplan Meier method and compared using the logrank test.Hazard ratio (HR) and 95% Confidence Interval (CI) were derived from a Cox model.The proportional hazard assumption was satisfied.To adjust for the bias by indication a propensity score for using PPIs was estimated via logistic regression, including the following baseline patients' characteristics: age, BMI.History HIV, HBV or HCV infection, presence of comorbidities, type of tumor, metastases, tumor stage and oncologic treatment.The Hoteling test to compare distribution was nonsignificant (p = 0.559).The Cox model was then weighted using the inverse probability of PPI administration derived from the PS.
OS and PFS were analyzed as described for the primary endpoint.
The dataset generated from this study is not publicly available due to data protection compliance.The corresponding author upon reasonable request can share the raw data.A medical indication for the use of PPIs was not clearly reported in most of cases (130 patients, 68.8%), while the most common clinical indications were the age over 75 (29 patients, 58.5%), gastroesophageal reflux disease (GERD) (10 patients, 17%) and functional dyspepsia (10 patients, 17%).No patients at the time of the start of ICI were taking aspirin.The patients' characteristics according to PPI exposition are described in Table 3 and in Figure 1.

| Characteristics of the study population
Eight patients (0.6%) received antibiotic therapy within 30 days prior to ICI initiation.Six of these patients (75%) received both PPI and antibiotics.

| DISCUSSION
This is one of the first studies assessing the impact of PPIs on the onset of GI irAES in patients with solid tumors undergoing ICIs.PPIs have a well-known dysbiotic effect, as demonstrated in previous reports. 13,14Due to these alterations in the composition of the GM, PPIs might be able to modify the response to ICIs. 15 In contrast, other authors have not demonstrated an impact of PPIs on the response to ICIs. 16ince PPIs can modulate GM and GI irAEs depending on altered bio-diversity of GM we hypothesized that patients exposed to PPIs had a higher risk of developing GI irAEs.During the period of the collection of data 25 GI irAES were reported.The majority of our patients (92%) were PPI exposed (HR 13.22, 95% CI 3.11-56.10,p < 0.0001).Importantly, this HR is confirmed after weighting for the propensity score (HR 15.13 95% CI 3.22-71.03,p < 0.0001).Our results are in line with those from a recent study that demonstrated the PPI use as a risk factor for chronic immune-mediated diarrhea and colitis. 17PPIs can modify the composition of GM by, altering pH and modulating the immune response through their effect to on neutrophils, cytokines and natural killer cells (NK cells). 18The compositional and functional alterations in GM lead to intestinal barrier breakage with an increased translocation of toxins and inflammatory factors that may be able to alter dynamically the immunological profile in a pro-inflammatory direction.These cytokines, including TNFα and IL-10, may shift the threshold of immune subsets activation within the tumor micro-environment, thereby resulting in augmented adaptive immune responses. 19The exact mechanism of PPI exposure and increased risk of GI irAEs remains unknown, but if these data will be confirmed by larger prospective cohorts with GM analysis, the clinicians should use PPIs more carefully and only if strictly indicated.In fact, there is a high percentage of inappropriate PPI prescription.More than half of PPIs prescribed among our patients were not clearly motivated.This evidence is in line with other papers. 20,21 our cohort, concomitant PPI use seems to have a detrimental impact on PFS and OS, even if statistical significance is not reached.Two recent meta-analyses suggested that PPIs are significantly associated with poorer OS and PFS for cancer patients treated with ICIs. 22,23he dysbiosis caused by PPIs may modulate the antitumor immunity and inflammation, as reported by other authors. 24Indeed, other comorbidities and polypharmacy, increasing the risk of drug-drug interactions, may constitute additional detrimental factors in PPI-exposed individuals.
Some limitations of the study must be mentioned.First, this is a retrospective single-center study that might have been affected by reporting bias and some missing data regarding baseline and previous co-medications.Second, we did not perform a GM analyses, as this will be part of a larger, prospective study that will soon start.Lastly, our results should be carefully interpreted in the light of the enrolling center characteristics (i.e., tertiary referral center), the relatively small sample size, and the tumor heterogeneity.We recognize that the number of patients in the final analysis is really too small to draw significant conclusions, nonetheless, despite these limitations, our data seem to suggest that chronic PPI use may be associated with an increased risk of GI irAES.

| CONCLUSIONS
In conclusion, in this study we have shown that PPI use is associated with an increased risk of GI irAEs.Prospective studies are needed to confirm these findings.In all cases, we recommend a careful use of PPIs in this setting, only in case a clear indication is evident.

3. 2 |
Exposure to PPIs and antibiotics therapyOne hundred and eighty-nine (52.2%) patients were exposed to PPIs during the study period.The most common type of PPIs was pantoprazole (124 patients, 65.6%).

4 F I G U R E 2
Toxicity events, by type and grade.Kaplan Meier IRAES free survival estimate, by PPI therapy.F I G U R E 3 Kaplan Meier survival estimate, by PPI therapy.

F I G U R E 4
Kaplan Meier PFS estimate, by PPI therapy.
T A B L E 1Abbreviations: HIV, human immunodeficiency virus; HBV, hepatitis B virus; HCV, hepatitis C virus.T A B L E 2Abbreviations: ICI, immune checkpoint inhibitors; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer.