The impact of proton-pump inhibitor use on the incidence and severity of gastrointestinal toxicity to checkpoint inhibition.

2657 Background: Checkpoint inhibitors (ICIs) for cancer come with the risk of immune-related adverse events (irAE). Upper and lower gastrointestinal (UGI and LGI, respectively) tract inflammation are common toxicities. Proton pump inhibitors (PPIs) are frequently used to treat immune-mediated UGI irAEs. Studies have suggested that these agents may worsen LGI irAEs. Our study aims to observe the effect of PPI use on the incidence and severity of GI irAEs. Methods: This was a single-center retrospective chart review including all patients who received ICIs. Patients were categorized based on PPI usage and screened for upper or lower GI toxicities. PPI use was counted if the medication was received at least three months before starting ICI and one year after the last ICI dose. Results: 16,849 patients were included. 2,114 (12.5%) patients received PPIs (PPI group – PG), while 14,735 (87.5%) did not (non-PPI group - NPG). 46 (2.2%) PG patients developed UGI irAE compared to 43 (0.3%) in the NPG (p0.05). PG patients were more likely to be hospitalized however (p=0.021). 115 (5.4%) PG patients developed an LGI irAE compared to 173 (1.2%) in the nPG (p<0.001). 80.7% of the PG presented with ≥grade 2 diarrhea and 52.0% with ≥grade 2 colitis, compared to 69.1% and 38.5% of the NPG, respectively (p<0.05). PG patients more frequently needed steroid treatment (60.0% vs. 42.2% in the NPG, p=0.004) and SIT (41.7% vs. 24.9%, p=0.003), and were less likely to have symptom resolution (78.5% in the PG vs. 89.6% in the NPG, p=0.020). A dose-dependent effect was observed. Every 10mg increase in PPI dose led to more frequent hospitalization for UGI irAEs (OR: 1.6, p=0.049) and more reliance on steroids (OR:1.3, p=0.044) and SIT (OR: 1.3, p=0.037) for LGI irAEs. This was also associated with less hospitalization and recurrence rates for LGI irAEs (p<0.05). Conclusions: Our study shows that PPI use during immunotherapy may increase the risk of developing a GI irAE. PPI use does not seem to impact the severity of upper GI irAEs aside from a higher hospitalization rate. However, PPI use may lead to a more severe disease course for lower GI toxicities. Prospective studies are needed to validate these results. Table: see text