Longitudinal analysis of clinical and biochemical outcomes during seven years of galactose supplementation in a patient with SLC35A2-CDG

Abstract

Background

SLC35A2-CDG is a rare X-linked congenital disorder of glycosylation caused by impaired UDP-galactose transport into the Golgi apparatus, resulting in multisystem involvement with prominent neurological manifestations. Although oral galactose supplementation has been proposed as a targeted therapy, its long-term efficacy and safety remain unclear.

Methods

We report a longitudinal study of a female patient with SLC35A2-CDG treated with oral D-galactose supplementation over seven years. Clinical outcomes were assessed using the Nijmegen Pediatric CDG Rating Scale (NPCRS), and glycosylation abnormalities were monitored via N-glycomics profiling. Glycosylation biomarkers associated with clinical severity were subsequently applied retrospectively to evaluate eight additional SLC35A2-CDG patients with absent or uncertain responses to galactose therapy.

Results

Our index patient demonstrated significantly clinical and biochemical improvements, with the NPCRS score improving from 22 to 12. Plasma N-glycan profiling showed improvement and normalization in both under-galactosylated glycans and normally formed disialylated glycans. Fractionated N-glycan analysis confirmed consistent improvement in both affinity-purified transferrin and the triple-depleted plasma fraction with IgG, transferrin and albumin removed. Strong correlations were also identified between glycosylation markers and disease severity. In contrast, seven non-responder patients showed no obvious clinical or biochemical response, while one displayed partial biochemical improvement and steady clinical course.

Conclusions

This study provides the first long-term evidence that galactose supplementation is effective with meaningful clinical and biochemical improvement observed in a SLC35A2-CDG patient. Glycosylation profiling may serve as a supportive tool for monitoring treatment response and guiding individualized therapy.

1. Introduction

SLC35A2-CDG (OMIM #300896), previously classified as Congenital disorder of glycosylation (CDG) type IIm, is an X-linked disorder resulting from pathogenic variants in SLC35A2 [1]. This gene encodes a transporter responsible for the import of UDP-galactose into the Golgi apparatus, a critical step for the synthesis of N- and O-linked glycans, glycosaminoglycans, and glycosphingolipids [1], [2]. Deficiencies in these processes lead to multisystem manifestations, including developmental delay, epilepsy, dysmorphic features, liver dysfunction, and skeletal abnormalities [3], [4], [5].

Currently, galactose supplementation has been considered a potential therapeutic approach for SLC35A2-CDG [2], [6]. However, not all the patients respond to the therapy, and the data on the long-term therapeutic outcomes remain unclear [3].

In this study, we present a longitudinal study of a female patient with SLC35A2-CDG who received oral D-galactose supplementation for over seven years, providing strong evidence of effectiveness to galactose therapy in this disorder. Using glycosylation biomarkers and clinical severity assessments, we also retrospectively evaluated additional eight female patients with SLC35A2-CDG.

2. Subjects and methods

2.1. Patient

This study was approved by the institutional review board of the Children's Hospital of Philadelphia and written informed consent was obtained from the proband's guardians. The initial 6 month of clinical and biochemical data of our index patient (patient 1) was previously reported as Patient 5 by Witters P in Genet Med [6].

For comparison, we included eight additional SLC35A2-CDG patients. Four of them, Patient 3 and 7 were previously reported by Ng BG in Hum. Mutat. as CDG0449, CDG0059 [4], and Patient 8 and 9, by Witters P in Genet. Med. as Patient 7 and Patient 6 [6]. The long-term biochemical and clinical evaluation of these patients provided natural history data against which the galactose responsiveness's were evaluated using the same clinical and biomarkers assessment. The data from the other four, Patient 2, 4, 5, 6 has not been published before.

2.2. Clinical assessment

The clinical characteristics and disease progression score were documented using the Nijmegen Progressive CDG Rating Scale (NPCRS), which assesses Section I: current function; Section II: system specific involvement and Section III: the current clinical assessment. The total scores were categorized by severity: mild (0–14), moderate (15–25), and severe (>26), as calculated by physicians [7].

2.3. Glycan analysis

Patients' samples were collected before and after the galactose supplementation for semi-quantitative N-glycan analysis using a liquid chromatography-electrical spray ionization-Time of flight/mass spectrometry (LC-ESI-TOF/MS) [8]. Carbohydrate transferrin testing (CDT) and plasma N-glycan tests were performed as routine clinical tests [9]. To evaluate the glycosylation changes in different plasma glycoproteins, we performed the N-glycan profiles of 50 mg affinity purified immunoglobulins (IgG), 15 mg transferrin (Tf), 50 mg the triple depleted plasma fraction (TD), a fraction with IgG, Tf and albumin removed as described previously [10].

Under-galactosylation (under-gal) index is calculated as the sum of ratios of each of the two under-galactosylated glycans to the upper limit of its respective normal range multiplied with the number of missing galactose (x2 for asialo, agalacto biantennary glycans (A-gal), and x1 for monosialo, monogalacto, biantennary glycan (M-gal)) with normal index ≤3.

2.4. Statistical analysis

Data analysis was conducted using IBM-SPSS Statistics version 29.0. Normality of continuous variables was tested using the Kolmogorov–Smirnov or Shapiro–Wilk test, as appropriate. Pearson correlation coefficients were used to assess associations between NPCRS scores and glycan biomarkers. A two-tailed, adjusted p-value <0.0125 (0.05 divided by the number of parameters of each analysis (n = 4, A-gal, M-gal, Disialo-glycans and Under-gal index)) was considered statistically significant.

3. Results

3.1. Clinical effect after 7-year galactose treatment

Our proband (Patient 1) was born at 37 weeks' gestation to healthy, non-consanguineous parents. Birth weight was 2.95 kg (25th percentile). Hypopigmented patches of skin on the face and legs were noted at 1 week of age. At 6-month, developmental delay was observed as she was not reaching for objects and had motor delay. The patient was referred to our hospital for further evaluation at 1 year and 7 months due to global developmental delay without history of seizure, failure to thrive, chronic constipation and an abnormal brain MRI with white matter changes involved the posterior limbs of the internal capsule, bilateral corticospinal tracts, bilateral red nuclei, bilateral lateral medullary periods, and bilateral dentate nuclei. Physical exam is significant for microcephaly (43.5 cm), poor weight gain, flat midface, flat nasal bridge, slightly low-set ears, prominent philtrum, extensive pigmentary mosaicism and hypotonia. Laboratory tests showed mildly elevated serum AST at 56–140 U/L (normal value <60 U/L) and creatine kinase at422 U/L (normal range: 60–365 U/L). Alanine aminotransferase, coagulation factors, kidney function and electrolyte were normal.

Initial plasma CDT analysis showed an essentially normal profile due to limited resolution and sensitivity for measuring a large intact protein (Supplemental Table 1). In contrast, N-glycan profiles demonstrated increased A-gal at 0.33% (normal range: 0.04–0.28%) and M-gal at 2.74% (0.51–1.78%) of total glycans. Further fractionated plasma N-glycan analysis demonstrated significant under-galactosylation in this patient with the abundance of M-gal at 3.39% (< 0.96%) accompany with high normal of A-gal at 0.60% (< 0.69%) in the Tf fraction. In the TD fraction, M-gal was increased at 3.13% (< 1.88%). A-gal was also increased at 0.18% (< 0.09%). IgG glycosylation was only minimally affected with high normal of M-gal at 0.60% (< 0.68%) and normal A-gal and Di-sialo biantennary glycans (Supplemental Table 2). Whole genome sequencing identified a de novo variant of uncertain significance (VUS) in the SLC35A2 gene (c.211G > A, p.Val71Met) as reported previously [6]. The pathogenicity of this variant was ascertained due to chronic isolated abnormal under-galactosylation detected in plasma total glycoproteins using a clinically validated semi-quantitative plasma N-glycan analysis.

She started on oral galactose supplementation at the age of 1 year and 10 months, titrated to the therapeutic dose at 1.5 g/kg/day (maximum: 50 g/day) as following weeks 0–6: 0.5 g/kg/day; weeks 7–12: 1.0 g/kg/day; weeks 13–18: 1.5 g/kg/day. Galactose supplementation has continued to date at 1.5 g/kg/day with interval dosing adjustments for weight increases. Clinical improvements were observed as early as at 6 weeks post treatment. She is now 9 years old, in 3rd grade mostly integrated in the regular classroom. To date, the total NPCRS score improved from 22 to 12. Significant improvement was observed in the current function with Section I score reduced from 12 to 5. This change was primarily driven by marked improvement in mobility, which improved from 3 to 0. Additional gains were observed in communication and self-care abilities. In systematic assessment, the improvement was mild with section II score reduced from 3 to 1. The current clinical assessment score showed minimum improvement with section III score reduced from 7 to 6. Throughout the treatment period, the patient underwent regular clinical and laboratory monitoring for safety assessment of galactose therapy, including liver function tests and assessment of galactose-related biochemical markers. Blood galactose-1-phosphate at median: 0.84 mg/dLl (normal <1.0 mg/dL and urinary galactitol levels at median: 53 mmol/mol creatinine (normal <36 mg/mol; therapeutic range < 98 mg/mol creatinine) remained below the therapeutic rangeduring the seven-year follow-up. No clinical adverse events were observed. Notably, AST levels trended toward normalization as early as 13 weeks after initiation of galactose therapy. A single high urine galactitol read was noted which was an outlier and likely measured right after taking galactose (Fig. 1, Table 1, Supplemental Table 3–5).

Fig. 1.

Comparison of NPCRS scores and liver enzyme levels in the responder and non-responders over time.

A. NPCRS total scores for overall clinical severity, B. Section I score for current function accessment, C. Section II score for system specific accessment, D. Section III score for clinical accessement, E. AST value for liver functions are ploted over time (in months).

Black arrows indicate initiation of galactose supplementation; red arrows indicate when the galactose dose reached 1.5 g/kg/day; green arrows indicate dosage adjustments for weight gain. Patient 1 (responder) is shown with a blue line; Patient 4 (suspected partial responder based on biochemical improvement) with a yellow line; Patients 2 and 3 with red and grey lines, respectively; and Patients 5, 6, 7, 8, and 9 with brown, pink, green, black, and purple lines, respectively. Data points from Patients 1, 4, 5, 6, and 7, who were continuously receiving galactose supplementation, are indicated by triangles.

Table 1.

Clinical and molecular characteristics of nine patients with SLC35A2-CDG.

	1[6]	2	3[4]	4	5	6	7[4]	8[6]	9[6]
Ancestry	Asian	White	White	White	Asian, White	White	Other	Asian	White
Gender	F	F	F	F	F	F	F	F	F
Age at onset	6 m	5w	2.5 m	2 m	6 m	7w	newborn	newborn	6w
Age at assessed	10 m	2 m	4 m	3 m	7 m	2 m	newborn	5 m	20y
Clinical symptoms
Hypopigmented skin	+				+			+
Microcephaly	+				+			+	+
Facial dysmorphism	+				+			+
Torticollis		+			+
Developmental delay	+	+	+	+	+	+	+	+	+
Dystonia		+	+	+	+		+	+	+
Epilepsy		+	+	+		+	+
Stridor			+
Nystagmus			+
Vision impairment			+	+				+
Short stature					+			+
Dysphagia							+	+
Prematurity							+	+
Other									long bone fractures, scoliosis
High AST	+	−	+	−	+	−	+	+	+
EEG	−	infantile spasms	infantile spasms	infantile spasms	−	infantile spasms	infantile spasms	−	−
Brain MRI	white matter abnormalities	−	ventriculomegaly	−	−	−	intraventricular hemorrhage	−	frontal atrophic
Gene test
Variant	211G > A	173G > C	1 A > G	844G > A	2 T > C	653_660del	274 + 2 T > C	523C > T	935C > A
Protein	Val71Met	Arg58Pro	Met1?	Gly282Arg	Met1?	Val218Glyfs*34		Leu175Phe	Ser312Tyr
Zygosity	Heterozygous de novo	Mosaic (23%)	Heterozygous de novo	Heterozygous de novo	Heterozygous de novo	Mosaic (10%)	Heterozygous de novo	Heterozygous de novo	Heterozygous de novo

3.2. Glycosylation changes after treatment

With 7 years of galactose supplementation, plasma N-glycan profiling in this child demonstrated a significant increase in normally glycosylated disialo-glycan, while truncated glycans, M-gal decreased from 2.74% to 1.21% and A-gal decreased from 0.33% to 0.22% of total glycan.

Fractionated N-glycan analysis of the same set of samples also showed overall improvement in protein glycosylation, especially in the Tf fraction, the M-gal decreased from 3.39% to 1.57%, representing an approximately 53.7% reduction from the pre-treatment baseline. Similarly, A-gal decreased from 0.60% to 0.38%, Di-sialo was normalized, and increased to 77.23% (< 71.7%). In the TD fraction, M-gal decreased from 3.13% to 1.77%, an improvement of 43.5%. A-gal was reduced from 0.18% to 0.12%. In the IgG fraction, the A-gal, M-gal and Di-sialo glycans have all been normal in this child as shown in Supplemental Table 2 and Fig. 2.

Fig. 2.

Changes of N-glycan abundance following galactose therapy in transferrin, Triple depleted fraction and total plasma in the responder and non-responders.

A. Trends in a-sialo, a-galacto, biantennary glycan (A-gal), mono-sialo, mono-galacto, biantennary glycan (M-Gal) and disialo biantennary (Di-sialo) glycan levels in transferrin pre- and post galactose treatment, B. Trends in A-gal, M-gal and Di-sialo glycan levels in triple depleted fraction pre- and post galactose treatment, C. Comparison of A-gal, M-gal, and Di-sialo glycan in plasma between responder, partial response and non-responders, highlighting differences in glycomic responses to galactose supplementation.

Black arrows indicate initiation of galactose supplementation; red arrows indicate when the galactose dose reached 1.5 g/kg/day; green arrows indicate dosage adjustments for weight gain. Patient 1 (responder) is shown with a blue line; Patient 4 (suspected partial responder based on biochemical improvement) with a yellow line; Patients 2 and 3 with red and grey lines, respectively; and Patients 5, 6, 7, 8, and 9 with brown, pink, green, black, and purple lines, respectively. Blue squares indicate N-acetylglucosamine (GlcNAc); green circles indicate mannose (Man); yellow circles indicate galactose (Gal); and purple diamonds indicate sialic acid (NeuNAc). Data points from Patients 1, 4, 5, 6, and 7, who were continuously receiving galactose supplementation, are indicated by triangles.

3.3. Correlation between glycosylation and clinical severity

Pearson correlation analysis of NPCRS scores including Section I,II, III and total scores and 4 under-galactosylation parameters including the abundance of A-gal, M-gal, disialo glycans and the under-gal index in each three plasma fractions (plasma, TD and TF) was carried out using individual longitudinal measurements with non-independent measurements. The results.revealed a strong relationship between glycosylation abnormalities and disease severity as measured by the NPCRS. The strongest positive correlation was observed between M-gal abundance in Tf and Section I score (r = 0.887, p = 2.35E-0.5), followed by correlations between the total plasma under-gal index and the total score (r = 0.886, p = 2.49E-0.5), total plasma A-gal abundance and the total score (r = 0.878, p = 3.69E-05), total plasma under-gal index and the Section I score (r = 0.862, p = 7.47E-05), the under-gal index in TD and the Section I score (r = 0.856, p = 1.86E-04), and total plasma M-gal abundance and the Section I score (r = 0.852, p = 1.10E-04). Conversely, Di-sialo glycan abundance in Tf were negatively correlated with Section I and II sub-scores, further supporting the utility of glycosylation profiles as biomarkers of clinical severity (r = −0.787, p = 8.35E-04; r = −0.779, p = 1.03E-03). No significant correlations were found with the Section III score (p > 0.0125 for all comparisons) (Fig. 3, Table 2).

Fig. 3.

Top glycosylation biomarkers that significantly correlated with clinical severity in plasma, transferrin and triple depleted plasma fraction.

A. The plasma under-galactosylation index showed the strongest correlation with the total NPCRS score, B. The abundance of the mono-sialo, mono-galacto, biantennary glycan (M-Gal) in affinity purified transferrin showed the strongest correlation with the Section I score, C. The under-galactosylation index in triple depleted plasma fraction was positively correlated with the Section I score. Blue squares indicate N-acetylglucosamine (GlcNAc); green circles indicate mannose (Man); yellow circles indicate galactose (Gal); and purple diamonds indicate sialic acid (NeuNAc).

Table 2.

Correlative analysis of glycosylation and clinical severity.

		P A-gal	P M-gal	P Di-sialo	P under-gal index	Tf A-gal	Tf M-gal	Tf Di-sialo	Tf under-gal index	TD A-gal	TD M-gal	TD Di-sialo	TD under-gal index
Total score	r value	0.878⁎⁎	0.695⁎	−0.285	0.886⁎⁎	0.644	0.830⁎⁎	−0.763⁎⁎	0.785⁎	0.299	0.731⁎	0.557	0.690⁎
	p value	0.000	0.006	0.322	0.000	0.013	0.000	0.002	0.004	0.299	0.003	0.038	0.010
Section I score	r value	0.780⁎⁎	0.852⁎⁎	−0.203	0.862⁎⁎	0.645	0.887⁎⁎	−0.787⁎⁎	0.822⁎⁎	0.501	0.832⁎⁎	0.283	0.856⁎⁎
	p value	0.001	0.000	0.487	0.000	0.013	0.000	0.001	0.002	0.068	0.000	0.326	0.000
Section II score	r value	0.427	0.561	−0.067	0.488	0.591	0.554	−0.779⁎⁎	0.592	−0.235	0.628	0.597	0.372
	p value	0.128	0.037	0.821	0.077	0.026	0.040	0.001	0.026	0.419	0.016	0.024	0.190
Section III score	r value	0.134	−0.426	−0.151	−0.032	−0.154	−0.231	0.193	−0.208	−0.374	−0.344	0.465	−0.410
	p value	0.648	0.129	0.607	0.914	0.598	0.427	0.509	0.475	0.187	0.229	0.094	0.145

Section I, current function; Section II, system specific involvement; Section III, the current clinical assessment; P, plasma; Tf, transferrin; TD, triple depleted fraction; A-gal, a-sialo, a-galacto, biantennary glycan; M-gal, mono-sialo, mono-galacto, biantennary glycan; Di-sialo, disialo biantennary glycan. ^⁎⁎Correlation is significant with p value <0.0025(2-tailed, adjusted for multiple hypothesis n = 4); ^⁎Correlation is significant with p value <0.0125 (2-tailed, adjusted).

3.4. Comparative analysis of additional SLC35A2-CDG patients

We reanalyzed eight additional SLC35A2-CDG female patients who with no demonstrable response during the available follow-up period. (Patient 2–9) by clinical assessment using the same set of glycan biomarkers identified by the above association study in the proband. Pre-treatment NPCRS total scores among 8 patients ranged from 15 to 40, with Section I score ranging from 5 to 17, Section II score from 2 to 8, and Section III score from 5 to 15. The age at galactose treatment initiation ranged from 5 months to 20 years, and treatment duration varied between 6 and 40 months. Following treatment, total NPCRS scores ranged from 16 to 39, with Section I score from 6 to 17, Section II score from 2 to 8, and Section III score from 8 to 17 (Fig. 1, Table 1, Supplemental Table 3). NPCRS total scores of 7 patients were increased with time suggesting a stable and slowly regressive natural history of this disease. The clinical severity score in Patient 4 showed moderate improvement reduced from 26 to 22 due to reduction in Section I (14 to 11) and Section II (4 to 2) scores, while Section III (8 to 8) score by clinical assessment unchanged. Interestingly, galactose therapy (1.5 g/kg/day) was initiated in Patient 4 at the age of 10 months, briefly reduced dose to 0.5 g/kg/day at the age of 13 months to accommodate ketogenic diet, then increased back to the full dose at 20 months. The down-trending of Section I score was observed afterwards (Fig. 1B) mirroring what we observed in our index patient. The patient has remained on galactose therapy to date for over 5 years. She is currently 5 years and 11 months old.

Biochemically, plasma N-glycan profiling in Patient 4 also showed rapid improvement, with A-gal decreasing from 0.65% to 0.10% in total plasma N-glycans, while M-gal and Di-sialo glycans remained within normal ranges after 40 months of galactose supplementation. The remaining seven patients showed no meaningful improvement in glycomic profiles or NPCRS scores (Fig. 2), and four of the seven patients stopped galactose therapy after the initial 6-9 M trial with exception of Patient 5, 6, and 7 who started galactose therapy recently and have remained on.

4. Discussion

This study provides evidence that long-term oral galactose therapy can lead to significant clinical and biochemical improvement in a female patient with SLC35A2-CDG. Over seven years of continuous therapy, the index patient showed progressive normalization of N-glycan profiles in a dose dependent manner and accompanied by clear functional gains on the NPCRS score (mostly in Section I score). These findings indicate that galactose supplementation may contribute to long-term stabilization and improvement of clinical outcomes in selected individuals.

Previous reports on galactose therapy in SLC35A2-CDG have described inconsistent or absent responses, leading to uncertainty about its long-term clinical benefit [2], [6]. The extended follow-up and longitudinal natural history data of non-responders (most were not on galactose) in this study demonstrates that therapeutic effects may emerge gradually and persist over years, suggesting that long-term clinical and biochemical monitoring and better understanding of disease natural history is essential before concluding non-responsiveness. In our patients, both biochemical and clinical improvements occurred in parallel, separating well from natural history of biochemical and clinical trends in non-responders.

The strong correlations observed between glycosylation abnormalities and clinical severity indicate that higher levels of under-galactosylated glycans are associated with greater functional impairment and overall disease burden. These relationships suggest that specific glycans may serve as quantitative biomarkers with predictive value. Among them, M-gal in Tf, and plasma A-gal abundance showed the closest association with clinical severity score, suggesting that they can potentially be sensitive indicators of disease severity and reliable parameters for therapeutic monitoring. To evaluate two N-glycan biomarkers together, we developed a plasma under-galactosylation index that can be conveniently applied to evaluate under-galactosylation in plasma without fractionation (Fig. 3).

By applying the same clinical and biochemical assessments retrospectively to eight additional SLC35A2-CDG patients initially classified as non-responders, we identified a spectrum of responses ranging from partial biochemical improvement to complete non-responsiveness. Notably, Patient 4, although clinically considered as a non-responder due to minimal change in Section III score, demonstrated clear biochemical responsiveness, with rapid reduction and normalization of A-gal and stable M-gal and Di-sialo levels after 40 months of therapy. Re-evaluation of functional assessment revealed a mild improvement in Section I score following attainment of therapeutic dosing (Fig. 1B), and treatment was continued based on parental-reported clinical benefit. This case highlights that biochemical improvement may precede or be more sensitive than conventional clinical assessment in detecting therapeutic response. Importantly, section I score (functional assessment) of the NPCRS, is a better clinical biomarker than section III score. In both index patient and patient 4, Section III scores were obviously not improved.

In contrast, three patients (Patient 2, 3, 8) exhibited progressive worsening in NPCRS, particularly in Section I scores, without measurable biochemical improvement and subsequently discontinued therapy. Another three patients (Patients 5, 6, and 7), who initiated therapy more recently, have not yet demonstrated clear clinical benefit, although early downward trends in under-galactosylation index were observed and Section I scores have remained stable to date with no regression (Supplemental Table 3). Overall, these findings underscore the marked heterogeneity of response to galactose therapy and suggest that treatment outcomes are influenced by multiple factors, including residual SLC35A2 activity, somatic mosaicism, genotype, timing of treatment initiation, and treatment duration.

Five of the 8 partial and non-responders (Patient 2, 3, 4, 6, 7) presented with early-onset severe epilepsy prior to treatment initiation, a feature reported in approximately 83% of individuals with SLC35A2-CDG [11]. Although epilepsy has been proposed as a potential predictor of treatment response [12], our findings do not support this as a reliable determinant. Patient 4 demonstrated partial responsiveness despite severe epilepsy, whereas Patient 8, without epilepsy, showed no response. These observations indicate that epilepsy history alone is insufficient to predict therapeutic outcome and should not be used as an exclusion criterion for galactose therapy. Nonetheless, early initiation of galactose treatment, before the onset of irreversible neurological damage, may still be critical for achieving significant clinical benefit.

Notably the improvement in the proband and one partial responder is mostly seen in the functional assessment suggesting Section I score be a strong candidate as a clinical biomarker for an outcome measurement. Importantly, parents' input sometimes are essential when assessing Section I score. Developing a more comprehensive functional assessment customized for SLC35A2-CDG may further improve our understanding of galactose therapy responsiveness among the patients.

In both our index patient and Patient 4, glycosylation biomarkers can detect early biochemical responses, sometimes preceding measurable clinical changes. Similar observation was also made previously [6]. Changes in glycan profiles, particularly a reduction in A-Gal glycan, were detectable when the lowest dose of galactose was used before any major clinical gains in both the proband and Patient 4, while in our index patient, a rapid reduction in Mono-Gal glycan in both Tf and TD fraction was the first indicator. Mono-Gal glycan levels are normal in Patient 4. Thus, we developed a plasma under-galactosylation index that facts in changes in both glycans and showed the best correlation with NPCRS score. This suggests that plasma under-galactosylation index may serve as early supportive indicators of therapeutic response, helping to guide treatment decisions. This is especially relevant given the high proportion of non-responders among affected patients.

Although it had been reported in the past that non-responders may also show biochemical improvement [6], our longitudinal studies here suggest clinical improvements may emerge slowly or subtly, mild benefits can be difficult to capture in these previously defined “non-responders” without a good collection of natural history data. Therefore, comprehensive characterization of both biochemical and clinical trajectories remains essential for understanding disease progression and for improving clinical trial readiness.

5. Conclusion

This work expands current understanding of the therapeutic potential of galactose in SLC35A2-CDG. While the evidence supports meaningful improvement in selected patients, it also highlights the importance of individualized assessment and long-term observation. Glycosylation profile represents a supportive tool for assessing therapeutic response and may guide optimization of treatment strategies and the pre-selection of clinical trial candidates in future studies. These findings underscore the need to interpret biochemical and clinical outcomes together, given the variability in response among patients.

Ethics approval and consent statement

The subjects' guardians consented to the Frontiers in Congenital Disorders of Glycosylation natural history protocol that was approved by Institutional Review Board (IRB) at the Children's Hospital of Philadelphia.

CRediT authorship contribution statement

Yupeng Liu: Writing – original draft, Visualization. Earnest J.P. Daniel: Methodology, Data curation. Wenyue Zhang: Methodology, Data curation. Andrew Edmondson: Writing – review & editing, Resources, Data curation. Miao He: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Ethics declaration

Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed.

This study included organ or tissue donors.

This study includes human biological material and consent was obtained by donors, or their next of kin or legal representatives, for use in this study and for publication of the article. The samples used in this research were not sourced from executed prisoners or prisoners of conscience.

This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the IRB at children's hospital of Philadelphia (Approval No. 14-011223).

This research follows the CARE guidelines and the CARE checklist.

Funding

This work was partially supported by the Rose Fund (MH), U54 NS115198 (ACE, EJPD, SM and MH).

Declaration of competing interest

No author has any competing interests.

Appendix A. Supplementary data

Supplemental Table: Red blood cell galactose-1-P and urine galactitol measurement for safety assessment of galactose therapy.

Data availability

Data will be made available on request.