Gastroenterology · Medical Literature
Clinical Trial: Predicting Response to Iron Therapy in Patients With Active Inflammatory Bowel Disease Using Hepcidin and Functional Iron Indices: A Multicentre Randomised Trial.
Abstract
Iron deficiency anaemia (IDA) is common in inflammatory bowel disease (IBD) and impairs quality of life. Hepcidin, the regulator of systemic iron homeostasis, may predict response to iron therapy; however, its utility in IBD remains unclear. This study evaluated whether baseline hepcidin predicts response to oral and intravenous (IV) iron in active IBD to guide personalised treatment. Ninety adults with active IBD and iron deficiency (with or without anaemia) from two randomised trials received IV iron, oral ferrous fumarate (FF) or oral ferric maltol (FM). Response at 12 weeks was defined as haemoglobin increase ≥ 1.2 mmol/L (19.3 g/L) or normalisation in IDA or ferritin > 100 μg/L and transferrin saturation > 20% in iron deficient patients. Baseline iron indices including ferritin, hepcidin, and soluble transferrin receptor (sTfR) were measured. Logistic regression and receiver operating curve analyses evaluated predictive performance. Baseline hepcidin strongly predicted response to iron therapy: AUC(FF): 0.86, 95% CI: 0.71-1.00 and AUC(IV): 0.63, 95% CI: 0.30-0.96. Hepcidin > 2.68 μg/mL identified non-responders to FF with 89% sensitivity and 77% specificity. Each twofold increase in baseline hepcidin or ferritin reduced the odds of response [log2(hepcidin) OR: 0.71, 95% CI: 0.56-0.89; log2(ferritin) OR: 0.38, 95% CI: 0.21-0.69]. Higher transferrin/log10(ferritin) (OR: 2.95, 95% CI: 1.38-6.30) and sTfR/log10(ferritin) ratios (OR: 1.29, 95% CI: 1.05-1.59) increased likelihood of response. Baseline hepcidin supports route selection for iron therapy: higher levels favour IV iron, while lower levels indicate likely oral response. Ferritin-based indices, notably transferrin/log10(ferritin), offer pragmatic alternatives where hepcidin testing is unavailable. ClinicalTrials.gov identifier: NCT05581420 and NCT05456932.
Clinical relevance
Does baseline serum hepcidin predict response to oral ferrous fumarate, oral ferric maltol, or intravenous iron therapy at 12 weeks in adults with active inflammatory bowel disease and iron deficiency? __SNAPSHOT__ {"studyDesign":"Pooled analysis of two open-label, multicentre randomised controlled trials (OVI-IBD and PRIme); secondary predictive biomarker study using logistic regression and receiver operating characteristic (ROC) curve analysis","setting":"Multiple general, teaching, and academic hospitals in the Netherlands (including Leiden University Medical Center, Amsterdam UMC, Haaglanden Medical Center, Erasmus UMC, and University Medical Center Groningen); patients enrolled 2021–July 2024","sampleSize":"n=90 (IV iron: n=39; oral ferrous fumarate [FF]: n=38; oral ferric maltol [FM]: n=13)","population":"Adults (≥18 years) with active IBD (defined by clinical, endoscopic, or biochemical activity including CRP >5 mg/L or fecal calprotectin >150 mg/kg) and iron deficiency with or without anaemia; median age 40 years; 58.9% women. Key exclusion criteria included recent blood transfusion, contraindications to iron supplementation, and significant comorbidities affecting iron metabolism.","intervention":"Baseline measurement of serum hepcidin, ferritin, and soluble transferrin receptor (sTfR) as predictors of response; patients received one of three iron supplementation strategies: IV iron (dosed by Hb and body weight), oral ferrous fumarate 200 mg daily (with dose adjustment after Hb normalisation in some patients), or oral ferric maltol 30 mg twice daily for 12 weeks","comparator":"No separate control arm for biomarker prediction; treatment arms compared for response rates and biomarker predictive performance across three active iron formulations","primaryOutcome":"Predictive value of baseline hepcidin for treatment response to iron therapy at 12 weeks, assessed by AUC from ROC analysis. Response defined as: Hb increase ≥1.2 mmol/L (19.3 g/L) or Hb normalisation in iron deficiency anaemia (IDA), OR ferritin >100 µg/L plus transferrin saturation >20% in iron-deficient patients without anaemia.","followUp":"12 weeks","fundingSource":"Mixed: one trial funded by ZonMw (Netherlands Organisation for Health Research and Development; government); one trial partially funded by Norgine Ltd. (industry). Some authors declared speaker's fees, advisory board honoraria, and unrestricted grants from pharmaceutical and biotech companies."}
Key findings
At 12 weeks, 60.9% of all patients responded to iron therapy overall. Response rates were markedly higher with IV iron (84.4%) than with either oral formulation (ferrous fumarate 37.5%; ferric maltol 37.5%). Baseline hepcidin demonstrated overall good discriminative ability for predicting response (AUC 0.76; exact 95% CI not reported in abstract). Stratified by treatment, hepcidin performed strongly for predicting response to ferrous fumarate (AUC 0.86, 95% CI 0.71–1.00) but showed only limited discrimination in the IV iron group (AUC 0.63, 95% CI 0.30–0.96). An overall hepcidin threshold of >9.20 µg/mL was associated with non-response (sensitivity 83%, specificity 65%). Within the ferrous fumarate group specifically, hepcidin >2.68 µg/mL identified non-responders with 89% sensitivity and 77% specificity. Each twofold increase in baseline hepcidin reduced the odds of response (OR 0.71, 95% CI 0.56–0.89; p=0.003), as did each twofold increase in ferritin (OR 0.38, 95% CI 0.21–0.69; p=0.002). Conversely, higher transferrin/log₁₀(ferritin) ratio (OR 2.95; p=0.005) and higher sTfR/log₁₀(ferritin) ratio (OR 1.29; p=0.015) were associated with increased odds of response. Predictive performance could not be assessed for the ferric maltol group due to its small sample size (n=13). NNT and NNH were not reported and cannot be calculated from available data. Adverse events by iron formulation were not reported in the abstract.
Practice implications
This study establishes that baseline serum hepcidin is a strong predictor of response to oral ferrous fumarate in adults with active IBD and iron deficiency, with an AUC of 0.86 and a threshold of >2.68 µg/mL identifying likely non-responders with high sensitivity (89%) and acceptable specificity (77%). These data support a clinically actionable inference: patients with elevated hepcidin are less likely to absorb and respond to oral iron, and may preferentially benefit from IV administration — consistent with hepcidin's known role in suppressing ferroportin-mediated intestinal iron absorption during inflammation. The substantially higher overall response rate with IV iron (84.4%) versus oral therapies (37.5% each) reinforces this pharmacologic rationale. However, the wide confidence interval for the IV iron AUC (0.30–0.96) indicates that hepcidin does not reliably discriminate IV iron responders from non-responders, which is mechanistically expected. Ferritin-based indices (transferrin/log₁₀[ferritin] and sTfR/log₁₀[ferritin] ratios) showed independent predictive value and may serve as practical alternatives in settings where hepcidin assays are unavailable. Important caveats: the ferric maltol arm was too small (n=13) to draw conclusions; the study is a pooled secondary analysis of two trials with differing FF dosing protocols, introducing potential heterogeneity; the open-label design introduces ascertainment and performance bias; the population was exclusively Dutch and predominantly female, limiting direct generalisability to other ethnicities or healthcare contexts; hepcidin assay standardisation across centres was not explicitly discussed; and hepcidin may be elevated by non-IBD comorbidities (e.g., obesity, infections, CKD), reducing its specificity in complex patients. A correction to this paper was published on 14 July 2026 (PMID 42444579); the nature of that correction is not detailed in the available abstract, and clinicians should consult the corrected final version. Only the abstract was reviewed; full data tables, subgroup analyses, and safety data were not accessible for this summary. __LIMITATIONS__ Small and unequal treatment arm sizes, particularly the ferric maltol arm (n=13), precluding predictive analysis for that formulation and reducing overall statistical power.,Pooled design combining two trials with different ferrous fumarate dosing regimens, introducing treatment heterogeneity that may confound response estimates.,Open-label design with risk of performance and ascertainment bias; adherence to oral iron was not controlled and may have independently contributed to reduced oral response rates.,Predominantly Dutch, single-ethnicity cohort (58.9% female, median age 40 years), limiting generalisability to more diverse populations.,Hepcidin is an acute-phase reactant influenced by comorbidities beyond IBD (obesity, infection, chronic kidney disease), potentially reducing its diagnostic specificity in clinically complex patients.,The wide 95% CI for the IV iron AUC (0.30–0.96) crosses the null (0.50), indicating hepcidin cannot reliably identify IV iron non-responders.,The hepcidin cutoff thresholds (2.68 µg/mL and 9.20 µg/mL) are derived from a single data-driven ROC analysis without external validation, raising concern for overfitting and limited generalisability.,Hepcidin assay standardisation across participating centres was not reported; inter-laboratory variability may affect reproducibility of the proposed thresholds.,A published correction exists (PMID 42444579; July 14, 2026); the specific nature and clinical impact of the correction are not detailed and must be reviewed before clinical application.,No patient-reported outcomes, quality-of-life data, or adverse event comparisons were reported in the abstract.,Partial industry funding (Norgine Ltd.) and multiple author financial disclosures warrant consideration in interpreting conclusions. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In adults with active IBD and iron deficiency, baseline serum hepcidin strongly predicts non-response to oral ferrous fumarate (AUC 0.86), with a threshold of >2.68 µg/mL identifying likely non-responders with 89% sensitivity and 77% specificity — supporting a hepcidin-guided, personalised approach to iron route selection. Overall IV iron response rates were markedly superior (84.4% vs 37.5% for oral agents), consistent with hepcidin-mediated impairment of oral absorption during active disease. These findings are hypothesis-confirming and clinically plausible, but external validation in larger, prospective, blinded trials with standardised hepcidin assays and correction review is needed before routine threshold-based clinical use can be recommended.