Whole Goat Milk versus Cow Milk Formula and Atopic Dermatitis in Infants: A Randomised Clinical Trial

Overview

Atopic dermatitis (AD), commonly known as eczema, is a common, chronic inflammatory skin disease that is most prevalent in early childhood, with around 80% of symptom onset occurring within the first year of life^2^. AD is characterised by intense itching, recurrent eczematous lesions, and a relapsing course, and it significantly affects the quality of life for both children and families. AD also predisposes to other inflammatory conditions such as food allergies, asthma, and allergic rhinoconjunctivitis^3^. With this significant burden to the individual, their family, community and healthcare system, nutritional interventions that can help prevent the onset of AD in the first place are increasingly important. Breastfeeding has numerous benefits and has been shown to support the prevention of eczema in babies, and should always be encouraged. For infants who are unable to be breastfed, there is limited high-quality, consistent evidence on different nutritional interventions ^4^.

• The GIraFFE Study is a large, double-blind, randomised controlled trial (n=2,132) aimed to investigate whether whole goat milk formula (WGF) could prevent AD in infants up to 12 months of age vs cow milk formula (CF)^5^. The study was designed following earlier evidence that indicated a one-third lower incidence of AD in infants consuming WGF^6^.

The GIraFFE Study^TM demonstrates that WGF can reduce the incidence of AD in formula-fed infants in the first year of life, especially in the presence of a parental history of AD.

• When AD was diagnosed by a doctor at any time over the 12-month intervention period, there was a significant 34% lower risk of AD in infants who received the WGF and had consumed it continuously throughout the study.
• A 64% significantly lower risk of developing AD, as diagnosed by a doctor at any time in the 12 months, in infants who received WGF and had a family history of AD.

*p<0.05; **p<0.01

Methods

Principal Investigator: Prof. Berthold Koletzko, Division of Metabolic and Nutritional Medicine, Department of Pediatrics, Dr. von Hauner Children’s Hospital, University Hospital, LMU Munich, Germany.

Ethics approval: Registered at ClinicalTrials.gov (NCT04599946) GIraFFE Study^TM^

Funding: Co-funded by the New Zealand Ministry for Primary Industries and the Dairy Goat Co-operative.

Formulations:

Intervention: The whole goat milk formula (Capricare; WGF) was made from whole goat milk as the only source of protein, with no added whey and contained approximately 50% of the fat from goat milk.

Control: The cow milk formula (CF) was a standard whey-adjusted CF, based on skim milk with added whey, with vegetable oils as the predominant fat source and only 5% of the fat from milk.

Recruitment: The study was a two-arm, parallel, randomised, double-blind, controlled trial. Healthy term infants aged up to 3 months were enrolled across 10 study centres in Spain and Poland, without selection based on AD risk. The power calculation for the primary outcome, AD_Primary, was based on an expected AD incidence of 15% and a 30% risk reduction^5^.

Measurements of Atopic Dermatitis(Figure 1)

Primary endpoint:

• AD_Primary: AD diagnosis by trained study personnel at scheduled visits using the UK Working Party diagnostic criteria, modified for infants and applied at three visits at 4, 6, and 12 months of age.

Secondary endpoints:

• AD_UKWP: AD_Primary plus two additional time points, where UK Working Party diagnostic criteria were applied by trained staff who led the parents through the examination of their infant during a phone call.
• AD_Doctor: AD diagnosis by a doctor at any time during the study, reported by parents during study visits.
• AD_Any: AD_UKWP and/or AD_Doctor. AD_Any represents the total number of infants identified as having AD by one of the measures. Infants diagnosed by two or more methods are only counted once.
• Severity of AD: AD severity was rated using POEM (Patient-Oriented Eczema Measure) and SCORAD (SCORing Atopic Dermatitis).
• Intention-to-treat: Intention-to-treat (ITT) analysis included all infants who had at least one serving of study formula.
• Per-protocol: Per-protocol (PP) analysis included all infants randomised who were compliant (no breaks of formula intake >3 days, no complementary feeding before 4 months) and no major protocol violations (all study visits)

Figure 1 Overview of all AD endpoints with AD assessment

Detailed Results

Characteristics of the Study Population

• A total of 2,132 infants were randomised, with 1,065 allocated to the WGF group and 1,067 to the CF group (Figure 2). Overall attrition by age 12 months was 12.6%, with higher retention rates than expected.
• Baseline characteristics were comparable between groups, with nearly all infants commencing formula feeding prior to enrolment (mostly cow milk-based formula).

Figure 2 Flow chart of the study population by formula group

Atopic Dermatitis Cases and Incidence Rates (Figure 3; Table 1)

AD_Primary: The cumulative incidence rate (IR) of AD_Primary in the first year of life was 11.6 per 100 person-years. There was no difference in AD_Primary incidence between WGF and CF groups in the ITT (IRR: 1.00; 95%CI: 0.75, 1.32; p=0.991) or PP populations (IRR: 1.00; 95%CI: 0.75, 1.32; p=0.991).

The use of the UKWP diagnostic criteria enables a practical, standardised, AD assessment. A systematic review has reported reduced sensitivity of the UKWP criteria in infants, particularly in populations with mild or early AD manifestations. A more recent study reported that only 30% of infants with clinical eczema fulfilled the UKWP criteria^7^.

AD_Doctor: The incidence of AD was lower as assessed by AD_Doctor in the WGF group (ITT population; IRR: 0.79; 95% CI: 0.62, 1.02; p=0.073; trend) and was significantly lower for AD_Doctor in the PP population (IRR: 0.66; 95% CI: 0.49, 0.90; p=0.008). The larger observed effect in the PP population suggests that better adherence to WGF feeding is associated with a greater reduction in risk.

AD_Any: The incidence of AD_Any was not different between the WGF and CF groups (IRR: 0.84; 95% CI: 0.68, 1.05; p=0.120). The incidence of AD was significantly lower in the WGF PP population (IRR: 0.75; 95% CI: 0.58, 0.96; p=0.025).

Figure 3 Incidence rate ratios (IRR) for AD (WGF/CF) in all infants in the ITT and PP populations

Effect in high-risk infants (Figure 4; Table 1)

Eczema is a heritable disease. A family history of AD is one of the strongest predictors of AD risk in offspring, increasing the risk by 2-3 times^8-9^. In the GIraFFE Study, a parental history of doctor‑diagnosed AD (at least one parent with a history of AD) was reported in 9.5% of participants.

Among infants with a parental history of AD (high risk), WGF compared with CF was associated with a strong protective effect across all AD outcomes.

•  AD_Primary: A protective effect from AD was observed in the WGF group with an IRR of 0.48 (95% CI: 0.22–1.03; p = 0.06; trend).
• AD_Doctor: The protective effect of WGF reached significance by doctor diagnosis, with a 64% reduction of AD in the WGF group (IRR of 0.36; 95% CI: 0.17–0.76; p = 0.007).

These associations were more pronounced in the PP population.

•  AD_Primary: The incidence of AD was significantly lower in the WGF group as assessed by AD_Primary, with an IRR of 0.24 (95% CI: 0.08–0.69; p <0.01).
• AD_Doctor: The incidence of AD was significantly lower in the WGF group as assessed by AD_Doctor, with an IRR of 0.21 (95% CI: 0.07–0.62; p <0.01).

The clearer protective effect observed in high‑risk children is consistent with previous nutrition studies. Evidence from meta‑analyses and intervention trials suggests that the benefits of breastfeeding, probiotics, and hydrolysed formulas on AD risk are mainly confined to high‑risk infants^10-11^.

Figure 4 Incidence rate ratio (IRR) for AD (WGF/CF) in infants with a parental history of AD for the ITT and PP populations.

Table 1 The incidence rate (IR)* in the first year of life by intervention group for the intention-to-treat total population and the subgroup of infants with parental AD

*New cases per 100 person-years

Severity of Atopic Dermatitis (AD)

There was no difference between WGF and CF in the severity of AD in either the SCORAD or POEM assessments. Overall, 30% of AD_Primary cases were classified as moderate and 3% as severe. Among infants with a parental history of AD, 10% of AD_Primary cases were classified as severe. Median POEM scores were comparable between study groups (WGF: 8 vs CF: 8; p=0.37), and 62% of infants with AD were classified as having moderate-to-very-severe disease by this measure.

Safety and growth

Growth: At 12 months, mean length and weight were similar between WGF and CF, with no significant differences in length‑for‑age or weight‑for‑age z‑scores. The finding that anthropometric data do not differ between WGF and CF is consistent with previous studies showing adequate growth with WGF.

Adverse events (AE): Overall AE incidence was comparable between groups (2,066 total AEs; IRR 1.03, p=0.48),
Growth: At 12 months, mean length and weight were similar between WGF and CF, with no significant differences in length‑for‑age or weight‑for‑age z‑scores. The finding that anthropometric data do not differ between WGF and CF is consistent with previous studies showing adequate growth with WGF.

GIraFFE Findings Support Previous Whole Goat Milk Formula Research

• An earlier clinical study investigating infant growth suggested a one-third trend in a lower incidence of AD in infants consuming whole goat milk formula (WGF) compared to those consuming cow formula (CF)^6^.
• Compositional elements of WGF support a lower allergenic exposure. Milk-based formulas contain major milk allergens, including β‑lactoglobulin, which is not present in breast milk. WGF contains less β‑lactoglobulin as it is not whey-adjusted, unlike most cow milk-based formulas.
• Softer curd formation of WGF for enhanced hydrolysis that may lower allergenic potential. Like breast milk, WGF also contains less αS1 casein than CF, resulting in softer curd formation. The softer curd has been associated with enhanced hydrolysis of milk proteins^12^, which may further lower the allergenic potential of WGF.
• Due to the high milk fat content in WGF, bioactive polar lipids from the milk fat globule membrane are retained (e.g. glycophospholipids, sphingolipids, and cholesterol), which may influence skin lipid composition and reduce AD risk^13-14^.

Conclusion

This clinical trial demonstrates that WGF can prevent the occurrence of AD in the first year of life, especially if there is a parental history of AD.

• A protective effect of WGF on AD was found when AD was diagnosed by a doctor with a significant risk reduction of 34% (PP).
• In infants with a parental history of AD, there was a significant protective effect on AD diagnosed by doctors in both the intention-to-treat (risk reduction of 64%) and per-protocol populations (risk reduction of 79%). The protective effect was observed across all methods of AD diagnosis.
• No difference in the incidence rates was observed between WGF and CF groups for AD diagnosed by the standardised UKWP diagnostic criteria at 3 study visits in all infants in the intention-to-treat population.

Ongoing clinical study

The GIraFFE study continues to evaluate atopic dermatitis (AD) outcomes in children who consumed WGF and CF until they are 5 years old. In addition to dermatological outcomes, the study incorporates comprehensive assessments of growth and overall well-being, including analyses of infant metabolism and gut health.

Key objectives include identifying risk factors for AD, advancing understanding of the immune system and the microbiome, and addressing broader questions about early-life nutrition and child development.