Background: Autosomal recessive congenital ichthyoses (ARCIs) are a clinically heterogeneous group of keratinization disorders characterized by generalized skin scaling due to mutations in at least 12 genes. The aim of our study was to assess disease severity, phenotypic, and ultrastructural features and to evaluate their association with genetic findings in ARCI patients. Methods: Clinical signs and symptoms, and disease severity were scored in a single-center series of patients with a genetic diagnosis of ARCI. Skin ultrastructural findings were reviewed. Results: Seventy-four consecutive patients (mean age 11.0 years, range 0.1–48.8) affected with lamellar ichthyosis (50/74, 67.5%), congenital ichthyosiform erythroderma (18/74, 24.3%), harlequin ichthyosis (two/74, 2.7%), and other minor ARCI subtypes (four/74, 5.4%) were enrolled. Mutated genes were as follows: TGM1 in 18/74 (24.3%) patients, ALOX12B in 18/74 (24.3%), CYP4F22 in 12/74 (16.2%), ABCA12 in nine/74 (12.2%), ALOXE3 in seven/74 (9.5%), NIPAL4 in seven/74 (9.5%), and CERS3, PNPLA1, and SDR9C7 in 1 patient each (1.4%). Twenty-five previously undescribed mutations in the different ARCI causative genes, as well as two microduplications in TGM1, and two microdeletions in CYP4F22 and NIPAL4 were identified. The mean ichthyosis severity score in TGM1- and ABCA12-mutated patients was significantly higher than in all other mutated genes, while the lowest score was observed in CYP4F22-mutated patients. Alopecia, ectropion, and eclabium were significantly associated with TGM1 and ABCA12 mutations, and large, thick, and brownish scales with TGM1 mutations. Among specific phenotypic features, psoriasis-like lesions as well as a trunk reticulate scale pattern and striated keratoderma were present in NIPAL4-mutated patients. Ultrastructural data available for 56 patients showed a 100% specificity of cholesterol clefts for TGM1-mutated cases and revealed abnormal lamellar bodies in SDR9C7 and CERS3 patients. Conclusion: Our study expands the phenotypic and genetic characterization of ARCI by the description of statistically significant associations between disease severity, specific clinical signs, and different mutated genes. Finally, we highlighted the presence of psoriasis-like lesions in NIPAL4-ARCI patients as a novel phenotypic feature with diagnostic and possible therapeutic implications.

Autosomal recessive congenital ichthyoses (ARCIs) are a clinically and genetically heterogeneous group of rare disorders of keratinization, characterized by widespread skin scaling with or without erythema [1]. ARCI manifest at birth with either a translucent membrane covering the body, known as collodion baby, or, less frequently, with widespread erythema and scaling (ichthyosiform erythroderma). In rare cases, the newborn is encased in a thick and fissured armor that limits breathing and movements, a picture that corresponds to the most severe life-threatening ARCI form, harlequin ichthyosis (HI) [2, 3]. Over time, patients can develop the typical features of one of the major subtypes of ARCI: lamellar ichthyosis (LI), i.e., diffuse large coarse and brownish-dark scales with no or minimal erythema, or congenital ichthyosiform erythroderma (CIE), i.e., small-thin and whitish-gray scaling with diffuse erythema. Less commonly, disease course features one of the minor ARCI variants: self-healing collodion baby, acral self-healing collodion baby, and bathing suit ichthyosis. Additional clinical features observed in ARCI patients are palmoplantar keratoderma (PPK), ectropion, eclabium, ear deformities, hearing impairment, painful fissures, joint flexion contractures of extremities, alopecia, hypohidrosis, and heat intolerance. Most patients also complain pruritus which may be severe and contribute to the poor quality of life of these individuals [4, 5]. At least 12 genes have been associated with ARCI: ABCA12, ALOX12B, ALOXE3, CASP14, CERS3, CYP4F22, LIPN, NIPAL4, PNPLA1, SDR9C7, SULT2B1, and TGM1 [6, 7]. They encode for enzymes and cofactors involved in the synthesis, storage, and transportation of lipids implicated in epidermal barrier formation and maintenance.

The diagnosis of ARCI is made on clinical features and nowadays can be confirmed by molecular genetic testing in most patients. Whenever available, ultrastructural skin examination can provide additional clues to diagnosis [1], in particular cases where no mutations are identified or when a variant of unknown significance is detected. Disease rarity and genetic heterogeneity have hampered the establishment of genotype-phenotype correlations for long. In the last years, the widespread use of next-generation sequencing (NGS) technologies has allowed to identify causative mutations in an ever-growing number of ARCI cases [8‒16]. However, most large series of molecularly characterized ARCI patients did not report detailed phenotypic and ultrastructural information. We present the findings of a cross-sectional study correlating disease severity, phenotypic, and ultrastructural features with genetic findings in 74 ARCI patients diagnosed and followed-up in our center.

Study Design

This is as a single-center, cross-sectional study correlating disease severity, clinical, and ultrastructural features with genotypic findings in Italian patients affected with ARCI. Patients diagnosed with ARCI according to the current ichthyosis classification [1], and confirmed by genetic testing between September 2013 and June 2021, were recruited. Exclusion criteria were a diagnosis of an ichthyosis form other than ARCI, and refusal to provide consent to participate in the study. The study was approved by the Institutional Ethical Committee (protocol n. 2579_OPBG_2021) and conducted in accordance with the Declaration of Helsinki. All participants or their parents/legal guardians/next to kin provided written informed consent before entering the study.

Clinical Evaluation

A standardized case-report form was used to collect information on patient demographics and history during the last follow-up visit. Clinical severity was evaluated using a previously published ichthyosis severity score based on the SCORing Atopic Dermatitis (SCORAD) index and taking into account the percentage of body area affected with scales and erythema, scores for 10 typical ichthyosis features (e.g., ectropion, eclabium, and fissures), and VAS scale for itch and pain [17]. Ten additional clinical features were also scored: scale size and color, hypohidrosis, heat intolerance, alopecia, focal PPK, reticulate scale pattern, ear deformities, height, and weight.

Transmission Electron Microscopy

Transmission electron microscopy reports of patients’ skin biopsies and corresponding images of the epidermis were reviewed. Ultrastructural classification into the four major types was based on the presence of: (1) numerous lipid droplets in the horny layer in the absence of other markers (type I), (2) crystalloid structures, so-called cholesterol clefts in the horny layer (type II), (3) elongated membranous structures and vesicular complexes in the granular and horny layers (type III), and (4) trilaminar membrane structures in upper epidermal and horny cells (type IV) [18]. The morphology of lamellar bodies was also evaluated.

Molecular Genetic Diagnosis

All enrolled patients had received a molecular diagnosis on blood genomic DNA in our hospital. Mutation analysis of samples collected until 2018 was performed through the targeted NGS approach using a regularly updated customized ichthyosis gene panel (Nextera Rapid Capture Custom Enrichment Kit, Illumina, San Diego, CA, USA; NimbleGen SeqCap Target Enrichment, Roche, Madison, WI, USA) and MiSeq or NextSeq550 sequencing platforms (Illumina). More recently (2019–2021), mutations were identified through NGS analysis by Clinical Exome (Twist Bioscience, San Francisco, CA, USA), filtered for ARCI genes, and NovaSeq6000 sequencing platforms (Illumina). Identified variants were evaluated by the web-based tools VarSome [19], Human Gene Mutation Database, and Leiden Open Variation Database. Variants were validated by Sanger sequencing in the patients and their parents. The mutations identified in 13 patients diagnosed till 2015 have been previously reported by our group [9]. Quantitative real-time PCR and/or microarray-based comparative genomic hybridization were employed to assess small genomic deletions or duplications (GRCh37).

Statistical Analyses

Categorical variables were described as numbers and percentages, and continuous variables as means and standard deviations. Then, for each level of the variables of interest (e.g., scale size, color, and thickness; alopecia; ear deformities), mean (standard deviation) and median values of the patient clinical severity score were computed.

The total clinical severity score was analyzed within the main genes of interest, and pairwise comparisons were carried out to verify whether the observed differences were statistically significant. No adjustment was performed for multiple comparisons, according to the indications of Rothman, and Savitz and Olshan [20, 21]. The distribution of the main variables of interest (e.g., scale size, thickness, and color; ectropion, eclabium, alopecia) was studied in detail comparing the different frequencies within the different genes.

Differences were tested using the Mann-Whitney U test for two samples, and the Kruskal-Wallis one-way analysis of variance (ANOVA) for three or more samples. Bivariate correlations, between age and the total clinical score, were performed for the whole sample as well as separately for each mutated gene; we reported the nonparametric Spearman’s rho. All analyses were performed using the statistical package IBM SPSS Statistics for Windows, Version 26.0.0.1 (IBM Corp., Armonk, NY, USA).

Demographic and Clinical Features

All 74 consecutive eligible patients accepted to participate. The series included 43 females and 31 males, from 73 families. Mean age was 11.0 years (median age 5.1 years, minimum-maximum age 0.1–48.8 years). Specifically, 57/74 (77.0%) patients were aged <18 years at the time of enrolment (Table 1). Pediatric patients included 15 infants, 15 children aged 1–3 years, 22 aged 4–10 years, and five adolescents aged 11–17 years. Seventy/74 (93.3%) patients were Caucasian, two were Pakistani and two Moroccan. Consanguinity was reported in 13/73 (17.8%) families. Presentation at birth was known for 70/74 (94.6%) patients as detailed in Table 1. The most common clinical type was LI, followed by CIE, whilst HI, self-healing collodion baby and bathing suit ichthyosis included two cases each (Table 1). HI patients presented a much severe phenotype at last follow-up visit as denoted by a markedly higher clinical severity score (Table 1).

Table 1.

Association between ichthyosis severity score and demographic, clinical, and molecular findings in 74 patients affected with ARCI

Variablen (%)Mean severity score ± SDMedian severity scorep valuea
Overall 74 (100) 29.8±15.7 28.0 na 
Sex 
 Male 31 (41.9) 30.6±16.3 29.3 0.618 
 Female 43 (58.1) 28.7±15.0 26.2 
Age at last follow-upb 
 <18 57 (77.0) 30.3±16.0 27.8 0.695 
 ≥18 17 (23.0) 28.2±14.7 29.0 
Presentation at birthc 
 CB 54 (77.1) 29.7±15.0 28.8 0.076 
 CIE 14 (20.0) 28.4±14.1 27.5 
 HI 2 (2.9) 65.0±12.7 65.0 
Clinical type 
 LI 50 (67.5) 29.1±13.5 27.6 0.008 
 CIE 18 (24.3) 32.0±16.6 28.8 
 HI 2 (2.7) 65.0±12.7 na 
 SHCB 2 (2.7) 9.4±2.3 na 
 BSI 2 (2.7) 12.0±5.0 na 
Mutated gene 
 ABCA12 9 (12.2) 43.6±20.3 48.0 <0.001 
 ALOX12B 18 (24.3) 23.2±11.3 20.4 
 ALOXE3 7 (9.5) 29.1±16.1 28.1 
 CERS3 1 (1.4) 14.5 na 
 CYP4F22 12 (16.2) 17.9±5.1 18.2 
 NIPAL4 7 (9.5) 28.1±8.8 28.2 
 PNPLA1 1 (1.4) 27.8 na 
 SDR9C7 1 (1.4) 11.8 na 
 TGM1 18 (24.3) 40.3±14.1 40.6 
Hypohidrosisd 
 No 20 (27.0) 28.0±16.5 26.2 0.425 
 Yes 54 (73.0) 30.5±15.5 28.0 
Heat intoleranced 
 No 28 (37.8) 24.7±15.5 18.5 0.008 
 Yes 46 (62.2) 32.9±15.1 29.2 
Alopeciad 
 No 58 (78.4) 25.1±12.7 21.5 <0.001 
 Yes 16 (21.6) 46.8±13.9 50.2 
Ear deformitiesd 
 No 52 (70.3) 22.8±9.9 20.3 <0.001 
 Yes 22 (29.7) 46.3±14.4 60.0 
Scales sized 
 Small 41 (55.4) 23.5±12.1 20.1 <0.001 
 Medium 16 (21.6) 32.3±20.1 27.7 
 Large 17 (23.0) 42.7±9.4 41.0 
Scales colord 
 White-gray 47 (63.5) 25.3±15.0 20.1 <0.001 
 Brown 27 (36.5) 37.6±13.8 38.0 
Scale thicknesse 
 Thin 28 (37.8) 18.7±8.2 17.4 <0.001 
 Medium 28 (37.8) 30.8±14.2 28.6 
 Thick 18 (24.3) 45.4±12.8 43.1 
Ectropion 
 No 55 (74.3) 22.8±10.0 20.5 <0.001 
 Yes 19 (25.7) 50.2±10.2 52.4 
Eclabium 
 No 63 (85.1) 25.5±12.3 22.1 <0.001 
 Yes 11 (14.9) 54.2±9.2 55.9 
Fissuresf 
 No 53 (71.6) 23.3±10.6 20.5 <0.001 
 Yes 21 (28.4) 46.3±14.2 49.5 
Hand contractures 
 No 53 (71.6) 23.2±11.2 20.1 <0.001 
 Yes 21 (28.4) 46.4±12.9 48.0 
Foot contractures 
 No 53 (71.6) 23.5±10.8 20.5 <0.001 
 Yes 21 (28.4) 45.7±14.9 48.0 
Variablen (%)Mean severity score ± SDMedian severity scorep valuea
Overall 74 (100) 29.8±15.7 28.0 na 
Sex 
 Male 31 (41.9) 30.6±16.3 29.3 0.618 
 Female 43 (58.1) 28.7±15.0 26.2 
Age at last follow-upb 
 <18 57 (77.0) 30.3±16.0 27.8 0.695 
 ≥18 17 (23.0) 28.2±14.7 29.0 
Presentation at birthc 
 CB 54 (77.1) 29.7±15.0 28.8 0.076 
 CIE 14 (20.0) 28.4±14.1 27.5 
 HI 2 (2.9) 65.0±12.7 65.0 
Clinical type 
 LI 50 (67.5) 29.1±13.5 27.6 0.008 
 CIE 18 (24.3) 32.0±16.6 28.8 
 HI 2 (2.7) 65.0±12.7 na 
 SHCB 2 (2.7) 9.4±2.3 na 
 BSI 2 (2.7) 12.0±5.0 na 
Mutated gene 
 ABCA12 9 (12.2) 43.6±20.3 48.0 <0.001 
 ALOX12B 18 (24.3) 23.2±11.3 20.4 
 ALOXE3 7 (9.5) 29.1±16.1 28.1 
 CERS3 1 (1.4) 14.5 na 
 CYP4F22 12 (16.2) 17.9±5.1 18.2 
 NIPAL4 7 (9.5) 28.1±8.8 28.2 
 PNPLA1 1 (1.4) 27.8 na 
 SDR9C7 1 (1.4) 11.8 na 
 TGM1 18 (24.3) 40.3±14.1 40.6 
Hypohidrosisd 
 No 20 (27.0) 28.0±16.5 26.2 0.425 
 Yes 54 (73.0) 30.5±15.5 28.0 
Heat intoleranced 
 No 28 (37.8) 24.7±15.5 18.5 0.008 
 Yes 46 (62.2) 32.9±15.1 29.2 
Alopeciad 
 No 58 (78.4) 25.1±12.7 21.5 <0.001 
 Yes 16 (21.6) 46.8±13.9 50.2 
Ear deformitiesd 
 No 52 (70.3) 22.8±9.9 20.3 <0.001 
 Yes 22 (29.7) 46.3±14.4 60.0 
Scales sized 
 Small 41 (55.4) 23.5±12.1 20.1 <0.001 
 Medium 16 (21.6) 32.3±20.1 27.7 
 Large 17 (23.0) 42.7±9.4 41.0 
Scales colord 
 White-gray 47 (63.5) 25.3±15.0 20.1 <0.001 
 Brown 27 (36.5) 37.6±13.8 38.0 
Scale thicknesse 
 Thin 28 (37.8) 18.7±8.2 17.4 <0.001 
 Medium 28 (37.8) 30.8±14.2 28.6 
 Thick 18 (24.3) 45.4±12.8 43.1 
Ectropion 
 No 55 (74.3) 22.8±10.0 20.5 <0.001 
 Yes 19 (25.7) 50.2±10.2 52.4 
Eclabium 
 No 63 (85.1) 25.5±12.3 22.1 <0.001 
 Yes 11 (14.9) 54.2±9.2 55.9 
Fissuresf 
 No 53 (71.6) 23.3±10.6 20.5 <0.001 
 Yes 21 (28.4) 46.3±14.2 49.5 
Hand contractures 
 No 53 (71.6) 23.2±11.2 20.1 <0.001 
 Yes 21 (28.4) 46.4±12.9 48.0 
Foot contractures 
 No 53 (71.6) 23.5±10.8 20.5 <0.001 
 Yes 21 (28.4) 45.7±14.9 48.0 

LI, Lamellar ichthyosis; CIE, congenital ichthyosiform erythroderma; HI, harlequin ichthyosis; SHCB, self-healing collodion baby; BSI, bathing suit ichthyosis.

aIndependent sample Mann-Whitney U test for two samples and Kruskal-Wallis one-way analysis of variance (ANOVA) for three or more samples; p values <0.05 in bold.

bYears.

cTotals may vary due to missing values.

dClinical findings not included in the ichthyosis severity score (17) and evaluated independently.

eBody scales.

fAt body sites other than palmoplantar surfaces.

Most patients (63/73, 86.3%) complained some degree of pruritus, with a mean VAS itch value of 5.7 ± 2.0. On the other hand, pain was reported in 11/73 (15.1%) patients only (mean VAS score 5.2 ± 2.7). Almost all patients (72/74, 97.3%) had PPK, which was mild in half of them (36/72, 50.0%); the pattern was diffuse in most of the cases, with only four/72 (5.6%) patients showing a focal/striated appearance. Fissures located at sites other than palmoplantar surfaces were detected in 21/74 (28.4%) patients. Nineteen/74 (25.7%) patients had ectropion and 11/74 (14.9%) mild eclabium. Alopecia was present in 16/74 (21.6%) patients. Twenty-five/74 (33.8%) patients showed some degree of flexion contractures of hands and/or feet. Ear deformities were seen in 22/74 (29.7%) patients. Clinical signs significantly associated with greater ichthyosis severity score comprised fissures, ectropion, eclabium, hand/foot contractures, alopecia, ear deformities, scale size, color, and thickness (p < 0.001, Table 1). Although hypohidrosis was a feature reported in the vast majority of the patients (54/74, 73.0%), only overt heat intolerance was significantly associated with a higher clinical severity score (p = 0.008, Table 1).

Molecular Genetic Analysis

The distribution of mutated genes in our series was TGM1 and ALOX12B in 18/74 (24.3%) patients each, CYP4F22 in 12/74 (16.2%) patients, ABCA12 in 9/74 (12.2%) patients, ALOXE3 and NIPAL4 in 7/74 (9.5%) patients each (Table 1). Finally, CERS3, PNPLA1, and SDR9C7 were mutated in 1 patient each. Globally, 83 distinct sequence variants (online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000536366), including 25 previously undescribed (Table 2), were identified. In addition, two microduplications in TGM1 gene and two microdeletions affecting CYP4F22 and NIPAL4 genes, respectively, were detected (online suppl. Table S2). The presentation at birth according to the mutated gene is summarized in online supplementary Table S3.

Table 2.

List of the 25 previously unrecognized sequence variants in ARCI genes reported in this study

GenecDNA levelProtein levelConsequencesZygosityVariant interpretationaOther allele variant
ABCA12 c.723dupA p.Val242Serfs*27 PTC Compound heterozygous Likely Pathogenic c.6875A>G (p.Asn2292Ser)b 
c.2984C>T p.Thr995Ile Missense Homozygous VUSe  
c.3022C>G p.His1008Asp Missense Homozygous VUS  
c.4212delATinsGGG p.Tyr1405Glyfs*32 PTC Compound heterozygous Likely Pathogenic c.4412A>G (p.His1471Arg)b 
c.4412A>G p.His1471Arg Missense Compound heterozygous Likely Pathogenic c.4212delATinsGGG (p.Tyr1405Glyfs*32)b 
c.4543C>G p.Arg1515Gly Missense Compound heterozygous Likely pathogenic c.487C>T (p.Arg163*)c 
c.6875A>G p.Asn2292Ser Missense Compound heterozygous Likely pathogenic c.723dupA (p.Val242Serfs*27)b 
c.7343+8G>A p.? Splice site Compound heterozygous Likely pathogenic c.7715A>G (p.Tyr2572Cys)b 
c.7715A>G p.Tyr2572Cys Missense Compound heterozygous VUS c.7343+8G>A (p.?)b 
ALOX12B c.406_408delGAG p.Glu136del In-frame deletion Compound heterozygous Likely pathogenic c.1907G>T (p.Ser636Ile)b 
c.1907G>Tc p.Ser636Ile Missense Compound heterozygous Likely pathogenic c.1579G>A (p.Val527Met)d 
c.1907G>Tc p.Ser636Ile Missense Compound heterozygous Likely pathogenic c.406_408delGAG (p.Glu136del)b 
ALOXE3 c.543+5delG p.? Splice site Compound heterozygous VUS c.2461 C>T (p.Arg821Trp)b 
c.2461C>T p.Arg689Trp Missense Compound heterozygous VUS c.543+5delG (p.?)b 
c.1219G>A p.Gly407Arg Missense Compound heterozygous Likely pathogenic c.1096C>T (p.Arg366*)d 
CYP4F22 c.73_82delTCCACCCTTC p.Thr26Serfs*15 PTC Compound heterozygous Likely pathogenic c.1523G>C (p.Arg508Pro)b 
c.384G>C p.Lys128Asn Missense Compound heterozygous VUSe c.367+1G>A (p.?)d 
c.1254_1255delGC p.Arg419Hisfs*24 PTC Compound heterozygous Pathogenic c.421+1G>A (p.?)d 
c.1523G>C p.Arg508Pro Missense Compound heterozygous VUSe c.73_82delTCCACCCTTC (p.Thr26Serfs*15)b 
c.1544G>T p.Arg515Leu Missense Homozygous VUS  
NIPAL4 c.335–2A>G p.? Splice site Compound heterozygous Likely pathogenic c.395C>T (p.Thr132Met)b 
c.395C>T p.Thr132Met Missense Compound heterozygous VUS c.335–2A>G (p.?)b 
SDR9C7 c.768_769delAG p.Arg256Serfs*36 PTC Homozygous Likely pathogenic  
TGM1 c.1134G>C p.Trp378Cys Missense Homozygous VUSe  
c.1443G>C p.Lys481Asn Missense Homozygous VUS  
c.1645G>T p.Gly549Cys Missense Compound heterozygous Likely pathogenic Microduplication 
Exon 11–12 
GenecDNA levelProtein levelConsequencesZygosityVariant interpretationaOther allele variant
ABCA12 c.723dupA p.Val242Serfs*27 PTC Compound heterozygous Likely Pathogenic c.6875A>G (p.Asn2292Ser)b 
c.2984C>T p.Thr995Ile Missense Homozygous VUSe  
c.3022C>G p.His1008Asp Missense Homozygous VUS  
c.4212delATinsGGG p.Tyr1405Glyfs*32 PTC Compound heterozygous Likely Pathogenic c.4412A>G (p.His1471Arg)b 
c.4412A>G p.His1471Arg Missense Compound heterozygous Likely Pathogenic c.4212delATinsGGG (p.Tyr1405Glyfs*32)b 
c.4543C>G p.Arg1515Gly Missense Compound heterozygous Likely pathogenic c.487C>T (p.Arg163*)c 
c.6875A>G p.Asn2292Ser Missense Compound heterozygous Likely pathogenic c.723dupA (p.Val242Serfs*27)b 
c.7343+8G>A p.? Splice site Compound heterozygous Likely pathogenic c.7715A>G (p.Tyr2572Cys)b 
c.7715A>G p.Tyr2572Cys Missense Compound heterozygous VUS c.7343+8G>A (p.?)b 
ALOX12B c.406_408delGAG p.Glu136del In-frame deletion Compound heterozygous Likely pathogenic c.1907G>T (p.Ser636Ile)b 
c.1907G>Tc p.Ser636Ile Missense Compound heterozygous Likely pathogenic c.1579G>A (p.Val527Met)d 
c.1907G>Tc p.Ser636Ile Missense Compound heterozygous Likely pathogenic c.406_408delGAG (p.Glu136del)b 
ALOXE3 c.543+5delG p.? Splice site Compound heterozygous VUS c.2461 C>T (p.Arg821Trp)b 
c.2461C>T p.Arg689Trp Missense Compound heterozygous VUS c.543+5delG (p.?)b 
c.1219G>A p.Gly407Arg Missense Compound heterozygous Likely pathogenic c.1096C>T (p.Arg366*)d 
CYP4F22 c.73_82delTCCACCCTTC p.Thr26Serfs*15 PTC Compound heterozygous Likely pathogenic c.1523G>C (p.Arg508Pro)b 
c.384G>C p.Lys128Asn Missense Compound heterozygous VUSe c.367+1G>A (p.?)d 
c.1254_1255delGC p.Arg419Hisfs*24 PTC Compound heterozygous Pathogenic c.421+1G>A (p.?)d 
c.1523G>C p.Arg508Pro Missense Compound heterozygous VUSe c.73_82delTCCACCCTTC (p.Thr26Serfs*15)b 
c.1544G>T p.Arg515Leu Missense Homozygous VUS  
NIPAL4 c.335–2A>G p.? Splice site Compound heterozygous Likely pathogenic c.395C>T (p.Thr132Met)b 
c.395C>T p.Thr132Met Missense Compound heterozygous VUS c.335–2A>G (p.?)b 
SDR9C7 c.768_769delAG p.Arg256Serfs*36 PTC Homozygous Likely pathogenic  
TGM1 c.1134G>C p.Trp378Cys Missense Homozygous VUSe  
c.1443G>C p.Lys481Asn Missense Homozygous VUS  
c.1645G>T p.Gly549Cys Missense Compound heterozygous Likely pathogenic Microduplication 
Exon 11–12 

PTC, premature termination codon; VUS, variant of unknown significance.

aAccording to ACMG criteria based on VARSOME version 11.4.0, dated September 23rd, 2022.

bPreviously undescribed variant.

cPreviously undescribed variant detected in 2 unrelated patients in our case series.

dKnown mutation.

eWith minor pathogenic evidence.

TGM1

Sixteen presented a LI phenotype, and two had bathing suit ichthyosis (online suppl. Fig. S1). The mean ichthyosis severity score in TGM1-mutated subgroup was 40.3 ± 14.1 (Table 1), similar to that found in patients with ABCA12 mutations, while significantly higher than in patients carrying ALOX12B, CYP4F22, NIPAL4, and ALOXE3 mutations, as shown in Table 3. Although not statistically significant, we observed a trend toward a reduced disease severity score with increasing patient age (Spearman’s rho = −0.205). All patients also had PPK of varying degree, associated in 12/18 (66.7%) cases with hand/foot contractures (Table 4). The probability of presenting with large, thick, and brownish scales was significantly greater, for each of these three parameters, in patients with TGM1 mutations compared to those with mutations in all other genes taken together and in each of the other frequently mutated genes, with the exclusion of NIPAL4 (Table 4). A majority of patients had fissures at body sites other than palmoplantar surfaces, the probability of presenting such feature being higher as compared to all other patients taken together (Table 4). As shown in Figure 1, the majority of TGM1-mutated patients presented alopecia and/or ectropion, a percentage significantly higher compared to each of all other frequently mutated genes, except ABCA12. In addition, mild eclabium was observed in 6/18 (33.3%) of TGM1-mutated patients (Fig. 1). Finally, 9/18 (50.0%) patients had ear deformities, including an anteriorly overfolded ear in four (Fig. 2a).

Table 3.

Statistical significance (p values) of the difference in ichthyosis severity score for the pairwise comparisons between patients with different mutated genes

Table 3.

Statistical significance (p values) of the difference in ichthyosis severity score for the pairwise comparisons between patients with different mutated genes

Close modal
Table 4.

Frequency distribution of patients by scale features (color, thickness, and size), fissures at sites other than palmoplantar surfaces, and hand/foot contractures for each mutated gene

Clinical featureGene
TGM1ALOX12BCYP4F22ABCA12ALOXE3NIPAL4total
Scale color 
 White-gray 17 12 45 
 Brown 18 26 
Scale thickness 
 Thin 12 26 
 Medium 27 
 Thick 13 18 
Scale size 
 Small 16 10 39 
 Medium 15 
 Large 13 17 
Fissures 
 No 15 12 51 
 Yes 10 20 
Contractures 
 No 13 12 46 
 Yes 12 25 
Clinical featureGene
TGM1ALOX12BCYP4F22ABCA12ALOXE3NIPAL4total
Scale color 
 White-gray 17 12 45 
 Brown 18 26 
Scale thickness 
 Thin 12 26 
 Medium 27 
 Thick 13 18 
Scale size 
 Small 16 10 39 
 Medium 15 
 Large 13 17 
Fissures 
 No 15 12 51 
 Yes 10 20 
Contractures 
 No 13 12 46 
 Yes 12 25 

p ≤ 0.001 TGM1 versus all other genes taken together (scale thickness, color, and size); TGM1 versus ALOXE3 (scale thickness, color, and size), ABCA12 (scale size and color), ALOX12B (scale thickness, color, and size), and CYP4F22 (scale thickness, color and size, fissures, and contractures).

p ≤ 0.005 TGM1 versus all other genes taken together (fissures and contractures); TGM1 versus ABCA12 (scale thickness), and ALOXE3 (contractures).

p < 0.05 TGM1 versus ALOX12B (fissures and contractures).

Fig. 1.

Frequency distribution of patients presenting with alopecia (a), ectropion (b), and eclabium (c), by each mutated gene. Alopecia was significantly more frequent in TGM1-mutated patients as compared to ALOX12B (p < 0.001), ALOXE3 (p < 0.006), CYP4F22 (p < 0.004), and NIPAL4 (p < 0.006). ABCA12-mutated patients had alopecia more frequently than patients carrying ALOX12B (p < 0.004), ALOXE3 (p < 0.035), CYP4F22 (p < 0.0036), and NIPAL4 (p < 0.0035) mutations. Ectropion was more frequently present in TGM1-mutated patients as compared to ALOX12B and CYP4F22 (p < 0.001), NIPAL4 and ALOXE3 (p = 0.019). Eclabium was overall rarer, and with one exception (in ALOXE3) all cases were concentrated in ABCA12- and TGM1-mutated patients. The frequencies observed for these two genes were statistically significantly different when each was compared to the frequency in all other genes combined (p = 0.011, and p = 0.016, respectively).

Fig. 1.

Frequency distribution of patients presenting with alopecia (a), ectropion (b), and eclabium (c), by each mutated gene. Alopecia was significantly more frequent in TGM1-mutated patients as compared to ALOX12B (p < 0.001), ALOXE3 (p < 0.006), CYP4F22 (p < 0.004), and NIPAL4 (p < 0.006). ABCA12-mutated patients had alopecia more frequently than patients carrying ALOX12B (p < 0.004), ALOXE3 (p < 0.035), CYP4F22 (p < 0.0036), and NIPAL4 (p < 0.0035) mutations. Ectropion was more frequently present in TGM1-mutated patients as compared to ALOX12B and CYP4F22 (p < 0.001), NIPAL4 and ALOXE3 (p = 0.019). Eclabium was overall rarer, and with one exception (in ALOXE3) all cases were concentrated in ABCA12- and TGM1-mutated patients. The frequencies observed for these two genes were statistically significantly different when each was compared to the frequency in all other genes combined (p = 0.011, and p = 0.016, respectively).

Close modal
Fig. 2.

Ear deformities in ARCI patients. Three infants carrying mutations in TGM1 (a), ABCA12 (c) and ALOXE3 (d) genes, and an ALOX12B-mutated child (b) are shown. Note anterior overfolding in all cases.

Fig. 2.

Ear deformities in ARCI patients. Three infants carrying mutations in TGM1 (a), ABCA12 (c) and ALOXE3 (d) genes, and an ALOX12B-mutated child (b) are shown. Note anterior overfolding in all cases.

Close modal

Overall, 17 distinct sequence variants, the majority being missense (11/17), and 2 microduplications were identified in TGM1. Three previously undescribed TGM1 missense variants, all located within the catalytic core protein domain, were identified and classified as variant of uncertain significance with minor evidence of pathogenicity, of uncertain significance, and likely pathogenic, respectively, according to the American College of Medical Genetics (ACMG) guidelines [22] (Table 2). Finally, 2 patients carried a paternally inherited microduplication involving exons 11–12 and 10 (online suppl. Table S2), respectively, in compound heterozygosity with a maternal missense variant.

ALOX12B

Ten/18 (55.6%) patients had a LI phenotype, seven/18 (38.9%) CIE, and one self-healing collodion baby (online suppl. Table S1). The mean ichthyosis severity score in this subgroup was 23.2 ± 11.3 (Table 1), i.e., significantly less severe than in patients carrying TGM1 and ABCA12 mutations (Table 3). All patients but one presented erythema, which was minimal or mild in most cases. All patients had PPK, mostly mild (13/18, 72.2%) and associated with hand/foot contractures in 5/18 (27.8%) cases only (Table 4). ALOX12B-mutated patients had a greater probability of presenting small, thin and whitish scales as compared to all the other genetic groups taken together (p = 0.003, p = 0.01, p = 0.006, respectively). Ectropion, eclabium, and alopecia were not detected in this subgroup of patients (Fig. 1a–c). Five/18 (27.8%) patients presented ear deformities, including anterior ear overfolding in three (Fig. 2b).

The most common type of sequence variants in ALOX12B was by far missense (15/19 distinct mutations, 78.9%) (online suppl. Table S1), which were mainly located (11/15, 73.3%) in the second half of the protein from exon 9 to 15 (amino acids 358 to 701). Two novel ALOX12B variants were identified: a missense, p.Ser636Ile, and a single amino acid in-frame deletion, p.Glu136del, both classified as likely pathogenic according to ACMG criteria (Table 2).

CYP4F22

Nine patients had mild LI and 2 patients CIE (online suppl. Table S1). The mean ichthyosis severity score was 17.9 ± 5.1, the lowest score among genes mutated in more than 1 patient (Tables 1, 3). All patients had white-gray scales, which were small and thin in the majority of them (Table 4). Mild to moderate PPK was observed in 10 cases. No ectropion, eclabium, fissures, hand/foot contractures, or ear deformities were detected in this subgroup, and alopecia was present in a single case (Fig. 1; online suppl. Fig. S1c).

Overall, 16 different sequence variants were identified in the CYP4F22 gene, most being detected in a single family (online suppl. Table S1). Variant types comprised missense (eight/16, 50.0%), followed by frameshift (three/16, 18.7%), nonsense (three/16, 18.7%), and splice site (two/16, 12.6%). Previously undescribed sequence variants comprised three missense, and two frameshift variants (Table 2). Finally, 1 patient carried a paternally inherited microdeletion involving CYP4F22 gene in correspondence of exon 10 in trans with a known nonsense mutation located in exon 10 (online suppl. Table S2).

ABCA12

Two patients presented HI phenotype and developed arthritis as previously reported [23]. Six of the remaining patients developed a CIE phenotype, and one LI. The mean ichthyosis severity score in ABCA12-mutated subgroup was 43.6 ± 20.3, the highest observed among all mutated genes (Table 1). Similar to TGM1-mutated patients, a trend toward a reduced disease severity score at increasing age was observed (Spearman’s rho = −0.517). Whitish small scales were present in most cases (Table 4). All patients manifested PPK. Hand and foot flexion contractures were detected in 6/9 (66.7%) patients (Table 4). Alopecia was present in 4/9 (44.4%) patients (Fig. 1a). Moreover, 5/9 (55.6%) patients had ectropion (Fig. 1b), and four of them also eclabium (Fig. 1c). Finally, 3 patients manifested anteriorly overfolded ears (Fig. 2c).

In ABCA12-mutated patients, 15 distinct sequence variants were identified, two at the homozygous state and the remaining ones in compound heterozygosity. Nine/15 variants were not previously reported in the literature and comprised six missenses, two frameshifts, and one splice site (Table 2). Among them, two missense, p.Thr995Ile and p.His1008Asp, were found in homozygosity in 2 patients with a mild CIE phenotype (severity score 14.1 and 14.5, respectively), one presenting with ichthyosiform erythroderma at birth.

ALOXE3

Five/seven patients developed a LI phenotype, while the remaining two had CIE. The mean ichthyosis severity score was 29.1 ± 16.1 (Table 1). Scales were white-gray in all patients and small in 6/7 (Table 4). All patients had PPK, usually mild, as well as some degree of body erythema, which was minimal in five cases, moderate in one, and severe in one. No patient manifested alopecia (Fig. 1b) or hand/foot contractures (Table 4). A single case who was 1-month-old had mild ectropion and eclabium as well as painful fissures at sites other than palmoplantar surfaces. The same patient also presented ear deformities including anterior overfolding (Fig. 2d). Nine distinct sequence variants were detected in ALOXE3 (online suppl. Table S1), including three novel mutations: two missense and one splice site (Table 2).

NIPAL4

The seven NIPAL4-mutated patients all had a LI phenotype. The mean ichthyosis severity score was 28.1 ± 8.8 (Table 1). PPK was focal/striated with a yellowish hue in four cases and diffuse in the remaining three. Two patients also had hand flexion contractures (Table 4). Four patients presented a reticulate scale pattern on the thorax and abdomen. Both focal/striated PPK and reticulate scale pattern were not seen in ARCI due to mutations in other genes. Notably, 4 patients developed psoriasiform lesions predominantly localized to the extensor surfaces of lower limbs from childhood (Fig. 3a, b) and one had a previous diagnosis of psoriasis. Ear deformities were present in 2 patients, and ectropion (Fig. 1b) as well as localized painless fissures (Table 4) in a single patient.

Fig. 3.

Clinical appearance of lower limbs lesions in a 4-year-old child (a) and an adult patient (b), both carrying NIPAL4 mutations. The insets show details of patchy psoriasiform lesions in the child at 2 years of age and in the adult patient.

Fig. 3.

Clinical appearance of lower limbs lesions in a 4-year-old child (a) and an adult patient (b), both carrying NIPAL4 mutations. The insets show details of patchy psoriasiform lesions in the child at 2 years of age and in the adult patient.

Close modal

In addition to being detected in homozygosity in 5 patients, the known missense mutation c.527C>A (p.Ala176Asp) was found to have arisen de novo in a sixth patient who also carried in trans a large deletion involving the entire NIPAL4 gene (online suppl. Fig. S2). The seventh patient was a compound heterozygous for two previously undescribed variants, a splice site and a missense (Table 2).

SDR9C7, CERS3, PNPLA1

A previously undescribed homozygous frameshift sequence variant (c.768_769delAG; p.Arg256Serfs*36) was identified in the SDR9C7 gene in a 42-year-old male Caucasian patient born to consanguineous parents. The sequence variant is not annotated in the GnomAD database of human variations. It causes a frameshift and was therefore considered likely pathogenic according to ACMG criteria. The parents did not report a collodion membrane at birth. Ichthyosis was diagnosed in the first month of life. Physical examination showed mild thin whitish scaling on the scalp and trunk with follicular keratosis of extensor limbs and hyperkeratosis of knees and elbows (Fig. 4b). Large thin light-brown scales were present on the legs (Fig. 4a) and hand dorsum. The patient also had mild palmar and focal plantar keratoderma, while nails were normal. No history of onychomycosis was reported.

Fig. 4.

Clinical features of SDR9C7- and CERS3-mutated patients. The SDR9C7-mutated adult patient shows large thin adherent light-brown scales on the leg (a), and small whitish scales on the arms with elbow hyperkeratosis and follicular keratosis (b). The infant carrying CERS3 homozygous mutation presents generalized mild erythema with thin whitish scales (c, d).

Fig. 4.

Clinical features of SDR9C7- and CERS3-mutated patients. The SDR9C7-mutated adult patient shows large thin adherent light-brown scales on the leg (a), and small whitish scales on the arms with elbow hyperkeratosis and follicular keratosis (b). The infant carrying CERS3 homozygous mutation presents generalized mild erythema with thin whitish scales (c, d).

Close modal

A 3-month-old Caucasian patient born to consanguineous parents was shown to carry a known homozygous missense mutation in CERS3 gene. She was born with a collodion membrane, which rapidly peeled-off leaving mild erythema with thin whitish scales (Fig. 4c, d). Finally, our series included a case of a three-year-old female ARCI patient due to PNPLA1 mutation, who has been previously reported [24].

Ultrastructural Findings

Ultrastructural examination of a skin biopsy was available for 56/74 (75.7%) patients. Sixteen of them were classified as type I based on the presence of numerous lipid droplets in the horny layer, seven as type II for the presence of cholesterol clefts in the stratum corneum and 29 as type III due to the finding of elongated membranous structures and/or vesicular complexes in the granular and horny layers (online suppl. Fig. S3). One case carrying a PNPLA1 homozygous mutation was considered a mixed type I and III, as previously reported [24]. Three cases that lacked distinctive characteristics could not be classified. Interestingly, all type III cases presented vesicular complexes and vacuoles of variable size sometimes containing a few small vesicles, while perinuclear elongated membranes in the granular layer were visualized in a minority of them at routine examination. As to associations between the ultrastructural findings and mutated gene, all seven cases with features of type II ARCI were due to mutations in TGM1, resulting in a 100% specificity of cholesterol cleft presence for TGM1 as a causative gene. However, cholesterol clefts were not seen in five/12 (41.7%) TGM1-mutated cases indicating that this ultrastructural feature has a low sensitivity (58.3%) (online suppl. Table S4). Out of the four NIPAL4-mutated patients for whom EM was available, three had type III ultrastructural features. No additional associations were detected.

Among the unclassified cases, the SDR9C7-mutated one showed a thickened stratum corneum with no lipid droplets or cholesterol clefts. Several perinuclear membrane-bound small vacuoles (100–600 nm in size), suggestive for abnormal lamellar bodies (online suppl. Fig. S4), were observed in the granular layer. On the other hand, a few recognizable lamellar bodies were seen, most of them displaying an abnormal lamellar content. Finally, no vesicular complexes and elongated membranous structures were present. Another case that could not be classified was CERS3-mutated and presented several short lamellar inclusions, isolated or in stacks, and rare lipid droplets within horny cells. In the granular layer, lamellar bodies appeared sparse and frequently empty or with an abnormal lamellar content (not shown).

The present study describes the phenotypic and genotypic characteristics of a cohort of 74 ARCI patients molecularly diagnosed in our Rare Skin Disease Centre. Most patients were pediatric (57/74, 77.0%) and followed-up since birth. Of note, an ichthyosis severity score [17] and defined additional clinical findings were collected for all patients during the last visit, thus allowing for a detailed phenotypic characterization. In addition, the scoring was performed by the same dermatologists (AD and MC) and, for the vast majority of patients, at age ≥1 year when the skin phenotype rarely undergoes major changes. Finally, molecular findings were correlated to ultrastructural features in 56/74 (75.7%) patients.

A collodion baby was observed in 54/70 (77.1%) of our cases at birth, in particular in all TGM1- and ALOX12B-mutated patients. The fraction of ARCI cases born as collodion baby is higher compared to what previously reported [8, 13]. This discrepancy likely depends on a referral bias, as our hospital is a tertiary care center, where severely affected newborns are referred from all over the country. Pruritus and hypohidrosis were common symptoms across multiple genotypes, in line with previous reports [4, 8, 13]. As for scale features, the strong association of white-gray color with small-size and thinness further supports the differentiation between CIE and LI phenotypes. At last follow-up, ectropion and alopecia were present in 19/74 (25.7%) and 16/74 (21.6%) cases, respectively, in keeping with the percentages reported in a recent large ARCI cohort [13]. Eclabium appeared more frequent in our population, possibly related to the higher proportion of severe cases included. Interestingly, we observed that several clinical features are significantly associated with greater ichthyosis severity score. These comprised fissures, ectropion, eclabium, hand/foot contractures, and brownish, large thick scales, but also alopecia, ear deformities, and heat intolerance.

TGM1-mutated patients were 18/74 (24.3%) in our study population. Thus, TGM1 mutations have been detected in a percentage slightly lower than in recent literature reports on large ARCI case series, where the percentage of TGM1-mutated patients ranged from 33.6% to 40.8% of cases with identified causative mutations [11, 13, 25]. The screening by Sanger sequencing, limited to the TGM1 gene and historically performed in other Italian reference centers, may partly explain the lower frequency of TGM1 mutations among patients referred to us for molecular diagnosis [26, 27]. Conversely, we observed a higher proportion of patients carrying ALOX12B (18/74, 24.3%) and CYP4F22 (12/74, 16.2%) mutations than that detected in previous series [8, 10, 11, 13, 25]. In these reports, the proportion of ALOX12B-mutated patients ranged from 9.0% to 13.2% and that of CYP4F22 from 2.6% to 8.0%. However, a study from the Middle-East identified ALOX12B mutations in 26.0% of Muslim patients [16], and in another one from the Czech Republic ALOX12B was the most frequently mutated gene (11 of 34 cases, 32.0%) [28]. Further studies are needed to ascertain if the diverse percentages in ARCI-mutated genes may be related to the patient recruitment criteria and mutation search methodologies employed or reflect a true different distribution of ARCI-mutated genes in specific populations. Indeed, the latter seems the case for mutations in the PNPLA1 and CERS3 genes which have been rarely detected in our ARCI case series (1 patient each) or in the English cohort by Simpson and co-workers (three PNPLA1 cases, 2%, and four CERS3, 3%) [13] but were identified in 19 and seven of 125 (15.2% and 5.6%, respectively) ARCI Iranian families, a population typified by high endogamy levels [11]. Finally, our findings confirm that ABCA12, ALOXE3, and NIPAL4 represent, together with TGM1, ALOX12B, and CYP4F22, the genes commonly mutated in ARCI [8, 13, 25].

Although the majority of mutations identified in our study were found in a single family, some recurrent mutations were detected. Among known TGM1 mutations, the recurrent splice site c.877–2A>G [29] was found in four alleles (2 patients), and the nonsense c.788G>A (p.Trp263*), originally described in Tunisian families [30, 31], was present in four alleles from 3 Italian patients. Missense mutations affecting codon 307 included the recurrent variant c.919C>T (Arg307Trp), identified in 1 patient at the homozygous state, and c.919C>G (Arg307Gly) detected in compound heterozygosity in the 2 patients affected with bathing suit ichthyosis. Indeed, the missense mutation Arg307Gly was described for the first time by Oji et al. [32] in a bathing suit ichthyosis case, and subsequently detected in 4 out of 16 patients presenting this peculiar phenotype [33]. Thus, our findings confirm that the temperature-sensitive TGM1 missense Arg307Gly is typically associated with bathing suit ichthyosis. As to ALOX12B, the recurrent mutation p.Tyr521Cys, reported to account for 61/282 (21.6%) alleles in a large series [15], was identified at the heterozygous state in 2 patients (two alleles, 5.5%) only. Among recurrent mutations in CYP4F22 [10], the missense p.Arg156Cys was identified in four alleles (2 patients), and the nonsense p.Arg362* in three alleles (2 patients). Finally, the highly recurrent NIPAL4 missense mutation c.527C>A, p.Ala176Asp was detected in 6/7 patients supporting the notion that it likely represents a mutational hotspot [12, 34].

In our series, we identified a relatively high frequency of patients (4/74, 5.4%) carrying microdeletions/microduplications affecting one allele of TGM1, CYP4F22, and NIPAL4 genes. Among ARCI causing genes, exon deletions have been described mainly in the ABCA12 gene [35‒37] and only rarely in CYP4F22, ALOX12B, and ALOXE3 [10, 15, 38‒41]. In addition, a single case carrying a microduplication in the TGM1 gene has been reported [8]. As to deletions, in case of NGS identification of a single point homozygous mutation that turns out to be carried by one parent only, we regularly perform screening for copy number variations by microarray-based comparative genomic hybridization. In case of negative results, quantitative real-time PCR is then done to exclude apparent homozygosity and thus avoid diagnostic pitfalls. Specific to the 2 patients carrying the TGM1 microduplications, the allele frequency of the single maternal variant identified by NGS was 25–30%, i.e., lower than the expected 50%. This finding prompted us to perform quantitative real-time PCR of the involved region to evaluate the presence of an actual duplication. Our results suggest that microduplications/microdeletions may be somewhat underestimated because of the complex diagnostic procedures required, and further highlight the need to always analyze in parallel patients and their relatives and to complement NGS technology with additional quantitative assays in cases with unexpected molecular findings.

The phenotypic global severity of TGM1- and ABCA12-mutated patients has been quantified in our series for the first time and shown to be greater than in other ARCI-mutated genes, in particular CYP4F22, ALOX12B, NIPAL4, and ALOXE3. The probability of presenting with brownish, thick, and large scales was significantly greater for cases with TGM1 mutations as compared to all the remaining cases, in keeping with previous results [13, 29]. In addition, the difference remained statistically significant when comparing TGM1-mutated patients with cases carrying mutations in ALOX12B, CYP4F22, ABCA12, and ALOXE3 genes. These findings represent the first statistically significant association between scale characteristics and single-mutated genes. Other phenotypic features, specifically alopecia, ectropion, and eclabium were also significantly associated with mutations in TGM1 and ABCA12 genes as compared with other frequently mutated genes (ALOX12B, ALOXE3, CYP4F22, and NIPAL4). Moreover, ABCA12-mutated patients had high rates of hand and foot contractures in our series. Interestingly, a recent study also detected a significant association between hair loss and mutations in TGM1 or ABCA12 genes, and between hand deformities, ectropion, and eclabium and ABCA12-mutated cases [13]. Taken together, these findings reinforce the notion that severe and disabling phenotypic features are characteristic of most TGM1 and ABCA12 mutations. On the other hand, ear deformities, including anteriorly overfolded ears, were seen in patients carrying mutations in all frequently mutated genes except CYP4F22. Thus, in our series, this sign does not appear specifically associated with ALOX12B mutations as originally reported [13]. In line with our findings, such type of ear deformity was not exclusive of the ALOX12B-mutated cases [42]. Finally, among ABCA12-mutated patients the two who presented with the mildest clinical phenotype carried two novel homozygous missense variants, in agreement with previous reports describing a CIE presentation in patients with at least one ABCA12 missense mutation [35, 43].

Seventeen/18 (94.4%) ALOX12B-mutated patients were born as collodion baby and then developed a relatively mild phenotype with modest but constant erythema, small-thin and whitish scales, minimal PPK and no ectropion, eclabium, or alopecia. Our quantitative findings confirm and extend data of recent case series [13, 15]. ALOXE3-mutated patients also presented light small scales and variable degrees of erythema in the absence of alopecia, hand/foot contractures, in line with the notion that ALOXE3 is associated with a mild phenotype in most cases [15]. CYP4F22-mutated patients showed the lowest disease severity score among all ARCI patients with commonly mutated genes. Although frequently born as collodion baby (75%), later in life they did not manifest ectropion, eclabium, fissures, hand/foot contractures or ear deformities and presented thin, white-gray scales. Our findings provide a quantification for the phenotypic description given in a large case series of CYP4F22-mutated patients who commonly presented collodion baby at birth and mild disease features later in life [10].

As to NIPAL4-mutated patients, noticeably 4/7 had psoriasiform lesions, which overtime could appear as circumscribed erythematous patches or frankly erythematous-squamous plaques usually localized to extensor limb surfaces. Interestingly, well-circumscribed plaque-like erythematous lesions have been recently associated with NIPAL4 mutations [13] and represented a phenotypic feature specific of these patients. Moreover, 4/18 NIPAL4-mutated patients described by Simpson et al. [13] were reported to have a past medical history of psoriasis. This was also the case of one of our NIPAL4-mutated patients. An overexpression of psoriasis-related genes (e.g., VNN3, LCE3D, PLA2G4D) has been described in NIPAL4-mutated patients but also in other ARCI cases [44]. Interestingly, a dramatic response to the anti-IL-12/23 antibody ustekinumab, employed in the treatment of psoriasis, has been reported in a NIPAL4-mutated patient [45]. Although the pathogenetic mechanisms, including cytokine profile, underlying this peculiar phenotypic trait remain to be investigated, the presence of psoriasiform lesions seems exclusive to patients carrying NIPAL4 mutations. Thus, it could be considered predictive of mutations in this gene. A reticulate scaling pattern on the abdomen and a focal yellowish keratoderma have been also described in NIPAL4-mutated patients [12, 13, 38]. In our series, these two features were specifically associated with NIPAL4 mutations, pointing to their interest as a clue for the diagnosis. Finally, in one NIPAL4-mutated patient the recurrent c.527C>A missense mutation occurred as a de novo event supporting the notion that it represents a mutational hotspot [12].

Among ARCI cases due to rarely mutated genes the clinical presentation of our patient carrying a previously undescribed SDR9C7 homozygous frameshift appears of interest for the presence of keratosis follicularis in the absence of any atopy history. This feature has not been previously reported in patients carrying SDR9C7 mutations [46‒49].

Concerning electron microscopy findings, we observed abnormal lamellar bodies in SDR9C7- and CERS2-mutated patients. Similar alterations of lamellar body structure have been reported in a single SDR9C7-mutated case [50], while no literature data are available for CERS2-mutated patients. Further electron microscopy studies are needed to better characterize epidermal subcellular abnormalities in SDR9C7- and CERS2-associated ARCI. Our findings also confirm the absolute specificity of cholesterol clefts (i.e., ultrastructural type II) for the identification of TGM1 as the causative gene [1, 51], but also the limited sensitivity of this feature. Finally, no other gene-specific features were found in our series.

The main limitation of our study is the small number of patients included for less frequently mutated genes, in particular SDR9C7, PNPLA1, and CERS2, for which no statistical evaluation could be performed. In addition, the cross-sectional design of our study precluded appreciation of disease course mainly in the few infants examined.

In conclusion, we identified 25 novel variants and four microdeletions/microduplications in causative genes and found significant associations between disease severity, specific clinical signs, and different mutated genes, thus expanding the phenotypic and genetic spectrum of ARCI. We also highlighted the presence of psoriasis-like lesions in NIPAL4-ARCI patients as a novel phenotypic feature with diagnostic and possibly therapeutic implications.

Our study expands the phenotypic and genetic characterization of ARCI by the description of significant associations between disease severity, clinical signs, and different mutated genes in ARCI.

We wish to thank the Italian Patient Association of Ichthyosis (UNITI), which collaborated with our reference center for patient referral. We also thank Mr. Gabriele Bacile for iconography preparation and Mr. Riccardo Mariani for skillful technical assistance in transmission electron microscopy.

Our study complies with the guidelines for human studies and was conducted in accordance with the World Medical Association Declaration of Helsinki. The study protocol was approved by Bambino Gesù Children’s Hospital Ethics Committee (protocol n. 2579_OPBG_2021). All patients or their parents/legal guardians signed written informed consents, including permission to publication of images, approved by the Ethics Committee.

The authors have no conflicts of interest to declare.

A.D., A.G.C., S.G., G.Z., and M.E.H. as well as T.S. and D.A. were supported, in part, by the “Progetto Ricerca Corrente” of the Italian Ministry of Health, Rome, Italy.

Conceptualization: all authors; investigations: A.D., M.C., S.R., E.P., C.C., A.G.C., and S.G.; data curation: A.D., T.S., E.P., C.C., A.G.C., G.Z., and D.A.; writing – original draft preparation: A.D., M.C., E.P., C.C., G.Z., and D.A.; writing – review and editing: A.D., S.R., E.P., C.C., S.G., G.Z., A.A., A.N., R.A., D.A., and M.E.H.; supervision: A.D. and M.E.H. All authors have read and agreed to the submitted version of the manuscript.

Additional Information

Andrea Diociaiuti, Marialuisa Corbeddu, Damiano Abeni, and May El Hachem contributed equally to this work.

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon reasonable request.

1.
Oji
V
,
Tadini
G
,
Akiyama
M
,
Blanchet Bardon
C
,
Bodemer
C
,
Bourrat
E
, et al
.
Revised nomenclature and classification of inherited ichthyoses: results of the first ichthyosis consensus conference in sorèze 2009
.
J Am Acad Dermatol
.
2010
;
63
(
4
):
607
41
.
2.
Rajpopat
S
,
Moss
C
,
Mellerio
J
,
Vahlquist
A
,
Gånemo
A
,
Hellstrom-Pigg
M
, et al
.
Harlequin ichthyosis: a review of clinical and molecular findings in 45 cases
.
Arch Dermatol
.
2011
;
147
(
6
):
681
6
.
3.
Glick
JB
,
Craiglow
BG
,
Choate
KA
,
Kato
H
,
Fleming
RE
,
Siegfried
E
, et al
.
Improved management of harlequin ichthyosis with advances in neonatal intensive care
.
Pediatrics
.
2017
;
139
(
1
):
e20161003
.
4.
De Palma
AM
,
Mazereeuw-Hautier
J
,
Giehl
K
,
Hernández-Martin
A
,
Merlos
M
,
Moons
P
, et al
.
Burden of itch in ichthyosis: a multicentre study in 94 patients
.
J Eur Acad Dermatol Venereol
.
2019
;
33
(
11
):
2095
100
.
5.
Abeni
D
,
Rotunno
R
,
Diociaiuti
A
,
Giancristoforo
S
,
Bonamonte
D
,
Filoni
A
, et al
.
The burden of autosomal recessive congenital ichthyoses on patients and their families: an Italian multicentre study
.
Acta Derm Venereol
.
2021
;
101
(
6
):
adv00477
.
6.
Richard
G
.
Autosomal recessive congenital ichthyosis
. In:
Adam
MP
,
Everman
DB
,
Mirzaa
GM
,
Pagon
RA
,
Wallace
SE
,
Bean
LJH
, et al
, editors.
GeneReviews®
.
Seattle (WA)
:
University of Washington
;
2022
.
7.
Fischer
J
,
Bourrat
E
.
Genetics of inherited ichthyoses and related diseases
.
Acta Derm Venereol
.
2020
;
100
(
7
):
adv00096
.
8.
Pigg
MH
,
Bygum
A
,
Gånemo
A
,
Virtanen
M
,
Brandrup
F
,
Zimmer
AD
, et al
.
Spectrum of autosomal recessive congenital ichthyosis in scandinavia: clinical characteristics and novel and recurrent mutations in 132 patients
.
Acta Derm Venereol
.
2016
;
96
(
7
):
932
7
.
9.
Diociaiuti
A
,
El Hachem
M
,
Pisaneschi
E
,
Giancristoforo
S
,
Genovese
S
,
Sirleto
P
, et al
.
Role of molecular testing in the multidisciplinary diagnostic approach of ichthyosis
.
Orphanet J Rare Dis
.
2016
;
11
:
4
.
10.
Hotz
A
,
Bourrat
E
,
Küsel
J
,
Oji
V
,
Alter
S
,
Hake
L
, et al
.
Mutation update for CYP4F22 variants associated with autosomal recessive congenital ichthyosis
.
Hum Mutat
.
2018
;
39
(
10
):
1305
13
.
11.
Youssefian
L
,
Vahidnezhad
H
,
Saeidian
AH
,
Touati
A
,
Sotoudeh
S
,
Mahmoudi
H
, et al
.
Autosomal recessive congenital ichthyosis: genomic landscape and phenotypic spectrum in a cohort of 125 consanguineous families
.
Hum Mutat
.
2019
;
40
(
3
):
288
98
.
12.
Ballin
N
,
Hotz
A
,
Bourrat
E
,
Küsel
J
,
Oji
V
,
Bouadjar
B
, et al
.
Genetical, clinical, and functional analysis of a large international cohort of patients with autosomal recessive congenital ichthyosis due to mutations in NIPAL4
.
Hum Mutat
.
2019
;
40
(
12
):
2318
33
.
13.
Simpson
JK
,
Martinez-Queipo
M
,
Onoufriadis
A
,
Tso
S
,
Glass
E
,
Liu
L
, et al
.
Genotype-phenotype correlation in a large English cohort of patients with autosomal recessive ichthyosis
.
Br J Dermatol
.
2020
;
182
(
3
):
729
37
.
14.
Cheng
R
,
Liang
J
,
Li
Y
,
Zhang
J
,
Ni
C
,
Yu
H
, et al
.
Next-generation sequencing through multi-gene panel testing for diagnosis of hereditary ichthyosis in Chinese
.
Clin Genet
.
2020
;
97
(
5
):
770
8
.
15.
Hotz
A
,
Kopp
J
,
Bourrat
E
,
Oji
V
,
Komlosi
K
,
Giehl
K
, et al
.
Meta-analysis of mutations in ALOX12B or ALOXE3 identified in a large cohort of 224 patients
.
Genes
.
2021
;
12
(
1
):
80
.
16.
Mohamad
J
,
Samuelov
L
,
Malchin
N
,
Rabinowitz
T
,
Assaf
S
,
Malki
L
, et al
.
Molecular epidemiology of non-syndromic autosomal recessive congenital ichthyosis in a Middle-Eastern population
.
Exp Dermatol
.
2021
;
30
(
9
):
1290
7
.
17.
Bodemer
C
,
Bourrat
E
,
Mazereeuw-Hautier
J
,
Boralevi
F
,
Barbarot
S
,
Bessis
D
, et al
.
Short- and medium-term efficacy of specific hydrotherapy in inherited ichthyosis
.
Br J Dermatol
.
2011
;
165
(
5
):
1087
94
.
18.
Niemi
KM
,
Kanerva
L
,
Kuokkanen
K
,
Ignatius
J
.
Clinical, light and electron microscopic features of recessive congenital ichthyosis type I
.
Br J Dermatol
.
1994
;
130
(
5
):
626
33
.
19.
Kopanos
C
,
Tsiolkas
V
,
Kouris
A
,
Chapple
CE
,
Albarca Aguilera
M
,
Meyer
R
, et al
.
VarSome: the human genomic variant search engine
.
Bioinformatics
.
2019
;
35
(
11
):
1978
80
.
20.
Rothman
KJ
.
No adjustments are needed for multiple comparisons
.
Epidemiology
.
1990
;
1
(
1
):
43
6
.
21.
Savitz
DA
,
Olshan
AF
.
Multiple comparisons and related issues in the interpretation of epidemiologic data
.
Am J Epidemiol
.
1995
;
142
(
9
):
904
8
.
22.
Richards
S
,
Aziz
N
,
Bale
S
,
Bick
D
,
Das
S
,
Gastier-Foster
J
, et al
.
Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of medical genetics and genomics and the association for molecular pathology
.
Genet Med
.
2015
;
17
(
5
):
405
24
.
23.
Auriti
C
,
Rotunno
R
,
Diociaiuti
A
,
Manzoni
SM
,
Uva
A
,
Bersani
I
, et al
.
Juvenile idiopathic arthritis in infants with Harlequin Ichthyosis: two cases report and literature review
.
Ital J Pediatr
.
2020
;
46
(
1
):
44
.
24.
Diociaiuti
A
,
Pisaneschi
E
,
Zambruno
G
,
Angioni
A
,
Novelli
A
,
Boldrini
R
, et al
.
Novel PNPLA1 mutations in two Italian siblings with autosomal recessive congenital ichthyosis
.
J Eur Acad Dermatol Venereol
.
2018
;
32
(
3
):
e110
2
.
25.
Fischer
J
.
Autosomal recessive congenital ichthyosis
.
J Invest Dermatol
.
2009
;
129
(
6
):
1319
21
.
26.
Terrinoni
A
,
Serra
V
,
Codispoti
A
,
Talamonti
E
,
Bui
L
,
Palombo
R
, et al
.
Novel transglutaminase 1 mutations in patients affected by lamellar ichthyosis
.
Cell Death Dis
.
2012
;
3
(
10
):
e416
.
27.
Esposito
G
,
De Falco
F
,
Neri
I
,
Graziano
C
,
Toschi
B
,
Auricchio
L
, et al
.
Different TGM1 mutation spectra in Italian and Portuguese patients with autosomal recessive congenital ichthyosis: evidence of founder effects in Portugal
.
Br J Dermatol
.
2013
;
168
(
6
):
1364
7
.
28.
Bučková
H
,
Nosková
H
,
Borská
R
,
Réblová
K
,
Pinková
B
,
Zapletalová
E
, et al
.
Autosomal recessive congenital ichthyoses in the Czech Republic
.
Br J Dermatol
.
2016
;
174
(
2
):
405
7
.
29.
Farasat
S
,
Wei
MH
,
Herman
M
,
Liewehr
DJ
,
Steinberg
SM
,
Bale
SJ
, et al
.
Novel transglutaminase-1 mutations and genotype-phenotype investigations of 104 patients with autosomal recessive congenital ichthyosis in the USA
.
J Med Genet
.
2009
;
46
(
2
):
103
11
.
30.
Hennies
HC
,
Raghunath
M
,
Wiebe
V
,
Vogel
M
,
Velten
F
,
Traupe
H
, et al
.
Genetic and immunohistochemical detection of mutations inactivating the keratinocyte transglutaminase in patients with lamellar ichthyosis
.
Hum Genet
.
1998
;
102
(
3
):
314
8
.
31.
Louhichi
N
,
Hadjsalem
I
,
Marrakchi
S
,
Trabelsi
F
,
Masmoudi
A
,
Turki
H
, et al
.
Congenital lamellar ichthyosis in Tunisia is caused by a founder nonsense mutation in the TGM1 gene
.
Mol Biol Rep
.
2013
;
40
(
3
):
2527
32
.
32.
Oji
V
,
Hautier
JM
,
Ahvazi
B
,
Hausser
I
,
Aufenvenne
K
,
Walker
T
, et al
.
Bathing suit ichthyosis is caused by transglutaminase-1 deficiency: evidence for a temperature-sensitive phenotype
.
Hum Mol Genet
.
2006
;
15
(
21
):
3083
97
.
33.
Marukian
NV
,
Hu
RH
,
Craiglow
BG
,
Milstone
LM
,
Zhou
J
,
Theos
A
, et al
.
Expanding the genotypic spectrum of bathing suit ichthyosis
.
JAMA Dermatol
.
2017
;
153
(
6
):
537
43
.
34.
Dahlqvist
J
,
Klar
J
,
Hausser
I
,
Anton-Lamprecht
I
,
Pigg
MH
,
Gedde-Dahl
T
Jr
, et al
.
Congenital ichthyosis: mutations in ichthyin are associated with specific structural abnormalities in the granular layer of epidermis
.
J Med Genet
.
2007
;
44
(
10
):
615
20
.
35.
Akiyama
M
.
ABCA12 mutations and autosomal recessive congenital ichthyosis: a review of genotype/phenotype correlations and of pathogenetic concepts
.
Hum Mutat
.
2010
;
31
(
10
):
1090
6
.
36.
Scott
CA
,
Plagnol
V
,
Nitoiu
D
,
Bland
PJ
,
Blaydon
DC
,
Chronnell
CM
, et al
.
Targeted sequence capture and high-throughput sequencing in the molecular diagnosis of ichthyosis and other skin diseases
.
J Invest Dermatol
.
2013
;
133
(
2
):
573
6
.
37.
Shibata
A
,
Sugiura
K
,
Suzuki
A
,
Ichiki
T
,
Akiyama
M
.
Apparent homozygosity due to compound heterozygosity of one point mutation and an overlapping exon deletion mutation in ABCA12: a genetic diagnostic pitfall
.
J Dermatol Sci
.
2015
;
80
(
3
):
196
202
.
38.
Lefèvre
C
,
Bouadjar
B
,
Ferrand
V
,
Tadini
G
,
Mégarbané
A
,
Lathrop
M
, et al
.
Mutations in a new cytochrome P450 gene in lamellar ichthyosis type 3
.
Hum Mol Genet
.
2006
;
15
(
5
):
767
76
.
39.
Ullah
R
,
Ansar
M
,
Durrani
ZU
,
Lee
K
,
Santos-Cortez
RL
,
Muhammad
D
, et al
.
Novel mutations in the genes TGM1 and ALOXE3 underlying autosomal recessive congenital ichthyosis
.
Int J Dermatol
.
2016
;
55
(
5
):
524
30
.
40.
Fioretti
T
,
Auricchio
L
,
Piccirillo
A
,
Vitiello
G
,
Ambrosio
A
,
Cattaneo
F
, et al
.
Multi-gene next-generation sequencing for molecular diagnosis of autosomal recessive congenital ichthyosis: a genotype-phenotype study of four Italian patients
.
Diagnostics
.
2020
;
10
(
12
):
995
.
41.
Zhang
L
,
Hu
Y
,
Lu
J
,
Zhao
P
,
Zhang
X
,
Tan
L
, et al
.
Identification of the first congenital ichthyosis case caused by a homozygous deletion in the ALOX12B gene due to chromosome 17 mixed uniparental disomy
.
Front Genet
.
2022
;
13
:
931833
.
42.
Hake
L
,
Süßmuth
K
,
Komlosi
K
,
Kopp
J
,
Drerup
C
,
Metze
D
, et al
.
Quality of life and clinical characteristics of self-improving congenital ichthyosis within the disease spectrum of autosomal-recessive congenital ichthyosis
.
J Eur Acad Dermatol Venereol
.
2022
;
36
(
4
):
582
91
.
43.
Wada
Y
,
Kusakabe
M
,
Nagai
M
,
Yamamoto
M
,
Imai
Y
,
Ide
YH
, et al
.
Mild case of congenital ichthyosiform erythroderma with periodic exacerbation: novel mutations in ABCA12 and upregulation of calprotectin in the epidermis
.
J Dermatol
.
2017
;
44
(
11
):
e282
3
.
44.
Murase
Y
,
Takeichi
T
,
Kawamoto
A
,
Tanahashi
K
,
Okuno
Y
,
Takama
H
, et al
.
Reduced stratum corneum acylceramides in autosomal recessive congenital ichthyosis with a NIPAL4 mutation
.
J Dermatol Sci
.
2020
;
97
(
1
):
50
6
.
45.
Poulton
C
,
Gration
D
,
Murray
K
,
Baynam
G
,
Halbert
A
.
Autosomal recessive congenital ichthyosis due to homozygous variants in NIPAL4 with a dramatic response to ustekinumab
.
Pediatr Dermatol
.
2019
;
36
(
6
):
1002
3
.
46.
Shigehara
Y
,
Okuda
S
,
Nemer
G
,
Chedraoui
A
,
Hayashi
R
,
Bitar
F
, et al
.
Mutations in SDR9C7 gene encoding an enzyme for vitamin A metabolism underlie autosomal recessive congenital ichthyosis
.
Hum Mol Genet
.
2016
;
25
(
20
):
4484
93
.
47.
Hotz
A
,
Fagerberg
C
,
Vahlquist
A
,
Bygum
A
,
Törmä
H
,
Rauschendorf
MA
, et al
.
Identification of mutations in SDR9C7 in six families with autosomal recessive congenital ichthyosis
.
Br J Dermatol
.
2018
;
178
(
3
):
e207
9
.
48.
Seidl-Philipp
M
,
Schossig
AS
,
Moosbrugger-Martinz
V
,
Zschocke
J
,
Schmuth
M
,
Gruber
R
.
Impaired epidermal barrier in autosomal recessive congenital ichthyosis (ARCI) caused by missense mutations in SDR9C7 in two Austrian sisters
.
J Dtsch Dermatol Ges
.
2019
;
17
(
7
):
742
5
.
49.
Mazereeuw-Hautier
J
,
Severino-Freire
M
,
Gaston
V
,
Texier
H
,
Vincent
M
,
Aubert
H
, et al
.
Identification of mutations in SDR9C7 in three patients with autosomal recessive congenital ichthyosis
.
Acta Derm Venereol
.
2020
;
100
(
4
):
5624
.
50.
Takeichi
T
,
Nomura
T
,
Takama
H
,
Kono
M
,
Sugiura
K
,
Watanabe
D
, et al
.
Deficient stratum corneum intercellular lipid in a Japanese patient with lamellar ichthyosis with a homozygous deletion mutation in SDR9C7
.
Br J Dermatol
.
2017
;
177
(
3
):
e62
4
.
51.
Laiho
E
,
Niemi
KM
,
Ignatius
J
,
Kere
J
,
Palotie
A
,
Saarialho-Kere
U
.
Clinical and morphological correlations for transglutaminase 1 gene mutations in autosomal recessive congenital ichthyosis
.
Eur J Hum Genet
.
1999
;
7
(
6
):
625
32
.