Challenges to Assessing the True Burden of Iron Deficiency in Cystic Fibrosis

Summary:

Iron deficiency is common in cystic fibrosis (CF), but diagnosis is complicated by a lack of standard guidelines. Consensus is needed.

Iron deficiency (ID) is an important and frequent extrapulmonary comorbidity in people with cystic fibrosis (pwCF). Despite its clinical relevance, it remains poorly defined, both in clinical care and in the previously published literature¹. Currently, the only existing care guidance from the European CF Society (ECFS) recommends annual ID screening, but without defined diagnostic biomarker thresholds or treatment guidance². The Cystic Fibrosis Foundation (CFF), which largely direct management of CF in the United States, provides no guidance.

In this issue of JPGN, Twomey et al.³ add new insights on ID in pediatric pwCF, a population with limited existing data. ID was common in this study population, present in 44–55% of patients (prior reported prevalence 40–60%^1,4–8). As the authors point out, there are no agreed-upon standards on how to define ID or adjust for inflammation in pwCF. Studies of this population have used a wide variety of definitions for ID, with some including serum ferritin <12, 20, 30, or 100mcg/L; percent transferrin saturation (TSAT) <16 or 20%; serum iron <60mcg/dL; or combinations of multiple parameters^1,4–10. This lack of standardization leads to significant heterogeneity in data reporting, making it difficult to estimate its true prevalence. The chronic inflammatory nature of CF and punctuated exacerbations of intensified immune response further complicate the diagnosis^5,9. Furthermore, the diagnostic thresholds in adult and children likely differ, with unclear criterion by young children, adolescents, and adult males or females of childbearing potential^3,4. The WHO suggests using a ferritin cutoff of <12–15 ng/mL (age-dependent) to diagnose ID in the general population; in pwCF, these thresholds have been shown to underestimate ID and correlate poorly with other ID markers^1,4. Twomey et al. redemonstrate this phenomenon: when the authors re-calculate 2021 data using WHO ferritin thresholds, ID rates dropped from 51 to 34%³.

In the absence of standard cutoffs, many studies resort to reporting multiple ID parameters, highlighting how prevalence varies depending on the definition used and the state of inflammation^1,5 (Table 1). A major challenge with using ferritin as a sole ID biomarker in pwCF is its strong association with inflammation rather than true iron stores. Potential adjustments for ferritin interpretation in the inflamed state may include using higher thresholds (based on guidelines for other inflammatory conditions [e.g. IBD]), using different thresholds for those thought to be in an acutely “pro-inflammatory state” (e.g. patients with C reactive protein [CRP] ≥10mg/L⁶, current or recent CF exacerbation^1,5,6,8,10), or utilizing specialized analytic models (e.g. the BRINDA regression approach used by Twomey et al.³). Jia et al. show correlations between higher ferritin and lower TSAT, serum iron, and hemoglobin (Hgb) in pwCF, underscoring the confounding effect of inflammation on ferritin as a biomarker of ID⁴. Lastly, there is some evidence to suggest that soluble transferrin receptor (sTfR) is a more reliable measurement of iron status in pwCF in a pro-inflammatory state⁵.

Table 1:

Comparison of study parameters used to define iron deficiency and inflammation in people with cystic fibrosis.

Study	Population	n	ID Definition	ID Prevalence*	Inflamed State Definition
Gettle [1]	Adult	67	Ferritin < 12 mcg/L or TSAT < 16%	42%	Acute illness, not defined further (these patients were excluded)
Twomey [3]	Children	190	Ferritin <20mcg/L	2021: 51% 2022: 44% Supplemental analysis of 2002 cohort: 49% (n=201)	CRP >10mg/L
Jia [4]	Adult	568	TSAT <20% or Iron <60mcg/dL	39%	NA – not studied
Khalid [5]	Adult	127 Annual review (AR) = 76 Infective exacerbation (IE) = 51	TSAT ≤15%	TSAT: 43% (AR 28%, IE 65%)	IE if ≥ 2 present: ESR >20 mm/h Neutrophil count >10×109/L CRP >20 mg/L Clinical signs of IE CRP >10mg/l considered possible inflammation
			sTfR >1.74 mg/L	sTfR: 15% (AR 12%, IE 20%)
			Ferritin <12 (women) or <20 (men) mcg/L	Ferritin: 19% (AR 20%, IE 18%)
			Iron <12 mcg/dL	Iron: 50% (AR 30%, IE 78%)
Lobbes [6]	Adult	165	Ferritin ≤20 (women) or ≤ 30 (men) mcg/L; ≤100 mcg/L in the case of systemic inflammation or TSAT ≤ 16%	44%	CRP ≥ 10 mg/L or CF exacerbation in the past 7 days (these patients were excluded)
Lobbes [7]	Adult	220	Ferritin ≤20 (women) or ≤ 30 (men) mcg/L; ≤100 mcg/L in the case of systemic inflammation or TSAT ≤ 16%	Time 0: 58% 12 months: 31%	CRP ≥ 10 mg/L
Uijterschout [8]	Children	53	Ferritin <12 (<5 years old) or <15 (>5 years old) mcg/L	60%	Pulmonary exacerbation in last 30 days
Uijterschout [10]	Children	36	Ferritin <12 (<5 years old) or <15 (>5 years old) mcg/L	14%	“Acute” if either present (these patients were excluded): CRP >10 mg/L Pulmonary exacerbation in last 30 days

ID: iron deficiency; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; TSAT: transferrin saturation; sTfR: soluble transferrin receptor

^*Rounded to nearest whole %

Secondly, this study’s³ assertion that HEMT (highly effective modulator therapy) use has little impact on ID differs from other reports. There is a growing body of evidence that HEMT use is associated with improvement in iron status, anemia, and inflammatory markers^4,7,9. While there are certainly physiologic differences between children and adults that could affect ID and HEMT in pwCF, it seems more likely that the statistical modeling of this study³ was limited compared some prior reports. Jia et al. performed the largest study of ID in pwCF to date, with multivariable models including both type of modulator therapy and duration of modulator use⁴. This study showed HEMT use was significantly associated with improvements in TSAT and serum iron, and modulator use of any kind was associated with higher Hgb (with larger degrees of improvement on HEMT compared to moderately-effective modulator therapy [MEMT]). A 2022 study by Lobbes et al.⁶ is mentioned by the authors as an example of a study demonstrating no significant impact of CFTR modulator therapy on ID; however, they did not differentiate between type of modulators (HEMT vs other), nor did they account for duration of modulator use. A follow-up prospective study including these confounders was later performed by the same group (2025)⁷ showing one year of HEMT treatment correlated with improvements in ferritin and TSAT.

In the present study, prevalence of HEMT use increased from 59 (51% of overall modulators used) in 2021 to 114 (79%) in 2022 (Twomey et al, supplemental table 2)³. This sizeable difference in use may confound any unadjusted results. Those on HEMT in 2021 had higher rates of ID erythropoiesis (IDE) (defined as low MCV and low ferritin) compared to those off modulators; in 2022, there was no difference in IDE between groups³. Perhaps increased prevalence or prolonged use of HEMT led to improved iron status (Twomey et al, supplemental table 3; 2021 IDE 15%, 2022 IDE 5%). Likewise, HEMT use was associated with improved MCV (Twomey et al, supplemental table 4) and lower ID anemia in 2022 (Twomey et al, supplemental table 3).

Lastly, physiologic differences between adults and children may impact this study’s interpretation in the context of prior works, which have identified significant associations between ID and female sex, lower FEV1, and lower BMI in adult pwCF^7,4. Additional considerations in pediatric patients–metabolic demand, dietary dairy, menstrual status, etc.–likely contribute to differences in iron status across age subgroups. This study³ population ranged from 1–16 years old; authors include age-based parameters for Hgb and MCV (Twomey et al, supplemental Table 6), though remaining model variables use the same parameters for all ages. Future studies will need to account for subpopulations of age ranges and age-specific covariates.

The true significance of ID in pwCF may be underestimated in the absence of standardized criteria, variability in laboratory thresholds, and confounding effects of chronic inflammation and age-related changes in iron metabolism. Twomey et al.³ not only highlight the prevalence of ID in children, but also a growing need for prospectively collected data to truly elucidate its clinical burden. Longitudinal data with inclusion of multiple metrics of iron status and inflammation with CF-specific outcomes will be key in the creation of evidence-based recommendations for diagnostic thresholds and management.