Acid-Blocking Drug Plus Iron Chelator Shows Added Effect on Iron Overload in Hereditary Anaemias

A combination of the acid-suppressing drug esomeprazole and the iron chelator deferasirox was associated with a substantially greater reduction in liver iron stores than either treatment alone in patients with hereditary anaemias and secondary iron overload, according to an exploratory re-analysis of a placebo-controlled cross-over trial published in a haematology journal. Among the eleven trial participants who were taking deferasirox, twelve months of esomeprazole treatment produced a mean decrease in liver iron concentration of 1.52 mg iron per gram of dry weight, compared with a decrease of 0.22 mg iron per gram of dry weight during the placebo period. In participants who were not taking the chelator, esomeprazole showed no significant effect on liver iron. The finding points to a potential additive interaction between two established drugs, although the small number of patients involved and the design of the re-analysis mean the result must be interpreted with considerable caution.

Secondary iron overload is a serious and progressive complication of hereditary anaemias, a diverse group of inherited blood disorders that includes non-transfusion-dependent thalassaemia, congenital sideroblastic anaemia, congenital dyserythropoietic anaemia, pyruvate kinase deficiency, hereditary elliptocytosis and sickle cell disease. In patients who do not receive regular blood transfusions, iron accumulates in the body largely because of increased uptake from the gut. This intestinal iron loading is driven by hepcidin, the hormonal regulator of iron metabolism, which is suppressed in these disorders. Because therapeutic options beyond iron chelation therapy remain limited, the intestinal absorption pathway has long been viewed as an attractive target for intervention, and the new analysis examined whether blocking gastric acid could exploit that pathway.

Proton pump inhibitors such as esomeprazole reduce the secretion of gastric acid, and this matters for iron handling because gastric acid is required to solubilise dietary non-haem ferric iron and to reduce it to the ferrous form that the intestine can absorb. By raising gastric pH, these drugs impair non-haem iron absorption. Several studies have suggested that proton pump inhibitor treatment lowers ferritin levels and reduces the need for therapeutic phlebotomy in patients with HFE-related hereditary haemochromatosis. More directly relevant to the current work, a randomised controlled trial in patients with thalassaemia reported reductions in ferritin among those receiving proton pump inhibitors together with iron chelators, while no changes were observed in patients receiving iron chelation monotherapy, hinting at a possible additive effect of the two treatments.

The underlying trial behind the new analysis enrolled thirty patients with a heterogeneous mix of non-transfusion-dependent hereditary anaemias and mild-to-moderate iron overload. In this placebo-controlled cross-over study, each participant received esomeprazole 40 mg and placebo twice daily, each for twelve months. Measured by magnetic resonance imaging, liver iron concentration fell from baseline by 0.57 mg iron per gram of dry weight during esomeprazole treatment, compared with a reduction of 0.11 mg iron per gram of dry weight during placebo, a difference that reached statistical significance with a p-value of 0.033. A pre-planned subgroup analysis found that participants with a low hepcidin-to-ferritin ratio, below the cohort median, showed a reduction of 1.20 mg iron per gram of dry weight on esomeprazole, against an increase of 0.08 mg iron per gram of dry weight on placebo, with a p-value of 0.003. In participants above the median ratio, liver iron reductions were comparable between the two arms. Similar results emerged when patients were stratified by hepcidin alone, and inflammatory markers were normal.

The interpretation of those findings has been debated. One research group hypothesised that proton pump inhibitors suppress hepcidin because less iron becomes available in the small intestine, and that this hepcidin suppression subsequently increases iron absorption, thereby negating the benefit of the acid-blocking treatment. However, in patients whose hepcidin levels are already substantially suppressed, hepcidin may not be capable of being suppressed any further. In such patients, the counteracting mechanism cannot operate, and the net effect is a reduction in intestinal iron uptake. This hypothesis could explain why the largest liver iron reductions in the trial were observed in patients with low baseline hepcidin. Yet the trial investigators proposed an additional possibility: given the earlier evidence that proton pump inhibitors may enhance the effect of iron chelators in thalassaemia, the two therapies might act together additively on the body’s iron balance.

To explore whether that additive hypothesis was plausible, the researchers re-analysed the trial data after stratifying patients by iron chelation therapy. All patients on chelation were receiving deferasirox film-coated tablets at a fixed dose, with no dose alterations permitted during the study. The median dose was 5.4 mg per kilogram per day over the twelve-month treatment period. One participant lacked dosage information and one used deferasirox for only four months in each treatment period, and because adherence was not recorded, actual treatment exposure could not be assessed. The stratification revealed important baseline differences: patients receiving deferasirox had higher transferrin saturation, a median of 74.5 percent compared with 45.0 percent, and lower plasma hepcidin, a median of 4.2 micrograms per litre compared with 11.5 micrograms per litre. Critically, nine of the eleven patients on deferasirox had also been placed in the low-hepcidin subgroup of the original analysis, making it impossible to separate the contribution of hepcidin status from that of combined treatment.

The re-analysis results were striking. In the intention-to-treat analysis of patients receiving deferasirox, esomeprazole treatment led to a decrease of 1.52 mg iron per gram of dry weight in liver iron concentration, with a standard deviation of 1.31, compared with a decrease of 0.22 mg iron per gram of dry weight, with a standard deviation of 0.82, in the placebo group. The mean difference was minus 1.31 mg iron per gram of dry weight, with a 95 percent confidence interval from minus 2.16 to minus 0.45 and a p-value of 0.007. In contrast, among patients not receiving deferasirox, no significant difference between the treatment arms was observed, with a mean difference of minus 0.19 mg iron per gram of dry weight and a p-value of 0.945. Because hepcidin levels differed between the groups, the investigators performed a sensitivity analysis adding hepcidin as a non-time-varying covariate in the primary mixed model, which changed the effect estimates by less than five percent and preserved statistical significance. A treatment-by-chelation interaction analysis produced a statistically significant interaction term of minus 1.20 in the intention-to-treat cohort, with a 95 percent confidence interval from minus 2.20 to minus 0.20 and a p-value of 0.020, supporting an association between combined treatment and liver iron change. The comparable per-protocol estimate of minus 0.83 was not statistically significant, likely owing to limited statistical power. No differences in ferritin or transferrin saturation changes were observed between treatment arms in either chelation group.

The authors distilled the findings into three observations. First, after twelve months of deferasirox alone, no effect on liver iron concentration was detected, with a mean change of minus 0.22 mg iron per gram of dry weight. Second, adding esomeprazole to the regimen produced a considerable reduction, with a mean change of minus 1.52 mg iron per gram of dry weight. Third, patients without concomitant deferasirox showed no effect of esomeprazole, with a mean change of 0.00 mg iron per gram of dry weight. Neither treatment alone was associated with a decrease in iron overload in this cohort, suggesting that their individual effects are limited, whereas co-administration coincided with a meaningful reduction. The magnitude of the reduction in the combined-treatment subgroup is comparable with results from the large THALASSA and THETIS trials of deferasirox in secondary iron overload. In patients with baseline liver iron concentrations at or below 7 mg iron per gram of dry weight, similar to this cohort’s median baseline of 5.0 mg iron per gram of dry weight, annual liver iron concentration fell by 2.5 mg iron per gram of dry weight in THALASSA and 1.8 mg iron per gram of dry weight in THETIS, while patients received deferasirox dispersible tablets at 10 mg per kilogram per day, a higher dose than the roughly 5 mg per kilogram per day used in the cross-over trial. Patients with greater baseline iron burden in those trials showed larger decreases while also receiving higher chelator doses.

Several limitations temper these conclusions. The stratification produced very small sample sizes, particularly in the per-protocol analyses, and the limited statistical power warrants cautious interpretation. The cohort’s heterogeneity, spanning disorders with distinct pathophysiological backgrounds and differing mechanisms and degrees of iron loading, further complicates interpretation, although the investigators argue that a shared mechanism contributes to iron overload across these conditions and that the rarity of some disorders justified the basket trial design. Adherence data for deferasirox were absent, so the influence of treatment adherence on the outcome cannot be assessed. Most importantly, the heavy overlap between the chelation group and the low-hepcidin subgroup means the analysis cannot determine whether the observed liver iron reduction was driven by the combined therapy itself or by the fact that acid blockade is particularly effective in patients with marked hepcidin suppression. The investigators speculate that increased intestinal iron uptake driven by suppressed hepcidin in this subgroup normally offsets the iron removal achieved by chelation, and that blocking absorption tips the balance.

Despite these caveats, the re-analysis suggests that combining esomeprazole with deferasirox produced substantially greater reduction of secondary iron overload in hereditary anaemias than either therapy achieved alone, and that deferasirox monotherapy at the doses used here was not associated with a significant liver iron reduction in this small subgroup. The potential additive effect of the two treatments now requires confirmation in larger, prospective, randomised controlled trials before any change to clinical practice can be considered. For now, the findings remain an exploratory signal from a small and heterogeneous population rather than an established treatment standard.

Subject of Research: Combined proton pump inhibitor and iron chelator therapy for secondary iron overload in hereditary anaemias

Article Title: Combination Therapy With Proton Pump Inhibitors and Iron Chelators in Hereditary Anaemias With Secondary Iron Overload

Article References: Doeven, T., van Vuren, A., Glenthøj, A., Kerkhoffs, J. L., Schols, S. E. M., Rijneveld, A., Nur, E., Peereboom, D., Welsing, P., Schutgens, R., Biemond, B. J., & van Beers, E. J. (2026). Combination Therapy With Proton Pump Inhibitors and Iron Chelators in Hereditary Anaemias With Secondary Iron Overload. eJHaem, 7(5), Article e70429. https://doi.org/10.1002/jha2.70429

Image Credits: AI Generated

DOI: 10.1002/jha2.70429

Keywords: hereditary anaemias, secondary iron overload, proton pump inhibitors, esomeprazole, deferasirox, iron chelation, hepcidin, liver iron concentration, thalassaemia, cross-over trial, intestinal iron absorption, re-analysis