Autoimmune-polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED, OMIM240300) syndrome is a rare monogenic disease due to biallelic mutations in autoimmune regulator (AIRE) gene (1). Classic diagnostic criteria are the presence of two of the three main APECED symptoms: chronic mucocutaneous candidiasis (CMC), chronic hypoparathyroidism (CH), and Addison’s disease (AD). Non-endocrine autoimmunity involving the liver, stomach, intestine, eyes, and kidneys is generally reported in a minority of patients (1).

Although it depends on the series analyzed, approximately 25% of APECED patients in different populations are affected by gastro-enteropathy (GE) manifestations (2) including celiac disease, cystic fibrosis, pancreatic exocrine insufficiency (PEI), autoimmune intestinal disease (2) and loss of enterochromaffin cells (3, 4), intestinal infections, gastritis due to an autoimmune attack against gastric parietal cells leading to gastric atrophy, and pernicious anemia (5). Circulating autoantibody (Abs) against parietal cells (anti-gastric parietal cells (PCAs)), i.e., anti-sodium potassium channel molecule and anti-intrinsic factor (IFA), are detectable. Candida esophagitis is characterized by pseudomembranotic lesions, erosion, and ulceration with retrosternal pain while swallowing and dysphagia (6, 7). Intestinal candidiasis may also be responsible for watery diarrhea and malabsorption (7). Autoimmune enteropathy is characterized by small intestinal villous atrophy, protracted diarrhea, no response to exclusion diets, and no severe immunodeficiency (2, 3, 8). Abs to tryptophan hydroxylase (TPHAb), an intestinal autoantigen expressed in serotonin-producing cells in the central nervous system and the intestine, were detected in APECED (8). Lymphangiectasia of the small intestine may also be responsible for steatorrhea in APECED patients (9).

Among the other non-endocrine manifestations, APECED-associated hepatitis (APAH) has been described to occur at different prevalence in large series of APECED patients among different ethnicities (reviewed (rev) in 10). Further, APAH is the first manifestation in less than 2% of APECED patients (rev in 11). Abs are detected against hepatic aromatic l-amino acid decarboxylase (AADC), cytochrome P450 family 1 subfamily A member 2 (CYP1A2), histidine decarboxylase (HDC), bactericidal permeability-increasing fold-containing B1, TPH, and p450 21-hydroxylase (21-OH, CYP21A2). No liver-specific autoantigen has been so far identified in all patients with APAH, and it has not been possible to specifically distinguish APAH from classical autoimmune hepatitis (AIH) (10). The homozygous c.967_979del13 AIRE mutation is the most frequently associated genotype to APAH development (10).

American APECED patients showed enrichment of organ-specific non-endocrine manifestations early in life when compared to European cohorts (11). Between 40% and 80% of the patients had non-major disorders at onset, while the occurrence of classical manifestations was delayed (11). This observation led to the revision of the diagnostic clinical criteria (11), which would permit earlier diagnosis based on the appearance of one classic triad (CMC, AD, and CH) (occurring below 20 years of age) and one non-classical manifestation (enamel hypoplasia, periodic fever with rash, non-infectious keratitis, or autoimmune hepatitis) at a young age in presence of IFNωAb or AIRE mutations (11, 12). In light of the foregoing, in this manuscript, we report and discuss the clinical, genetic, and Ab profile in a series of 14 pediatric Italian APECED patients with APAH or gastrointestinal manifestations highlighting common features and evaluating whether their early enrichment would allow an expedited clinical diagnosis of APECED similar to the American cohort.

In the present series, all APECED patients presented APAH or autoimmune enteropathies (AIEs) ( Tables 1 , 2 , Figure 1 , Supplementary Figure 1 ) (13–19). A description of unpublished case reports is reported in the Supplementary Text , Supplementary Figure 1 and Table 1 .

Nine patients (Nos. 1–5, 8, and 12–14) presented APAH ( Table 1 ). The mean age of APAH diagnosis was 8.1 years (range, 1.5–13). In three patients (Nos. 1, 3, and 5), APAH occurrence was the first disease manifestation, while three patients (Nos. 2, 4, and 14) developed it following CH, two (Nos. 8 and 12) following CMC, and one (No. 13) developed APAH following CH and simultaneously with CMC. APAH-related specificities LKMAb, which target cytochrome P4502D6 (cP4502D6 and CYP2D6) autoantigen (27), tested positive in 3/9 patients (Nos. 4, 12, and 13) and TPH in 4/8 (Nos. 2, 5, 8, and 12); AADC was enriched in 6/7 (Nos. 1–3, 5, 8, and 12), cP450c21Ab in 4/6 (Nos. 1–4), and cP4501A2Ab in one patient (No. 12). Regarding other APAH-related Abs, ANA was present in 3/10 (Nos. 2, 3, and 14), SMA was present in 3/10 patients (Nos. 1, 3, and 12), ANCA and LC1Ab tested negative in all those tested (Nos. 1–5, 13, and 14 and Nos. 1–4, 13, and 14, respectively), and lastly, SLA-IgG was absent in all those tested (Nos. 1–4, 13, and 14) ( Table 1 ). The most frequent AIRE mutation was c.967_979del13 (4/9 patients, Nos. 1, 3, 4, and 14) (10).

Seven patients (Nos. 5–11) were affected by signs of constipation, diarrhea, and/or malabsorption ( Table 1 ). The mean age of intestinal dysfunction diagnosis was 6.1 years (range, 3.9–9). Two patients (Nos. 8 and 11) were also affected by iron-deficiency anemia and two (Nos. 5 and 10) by atrophic gastritis, and of them was affected with pernicious anemia (No. 10). The sera of five patients (Nos. 5, 7, 8, 10, and 11) tested positive for TPHAb and five (Nos. 5, 6, 8, 10, and 11) for AADC Abs known to be associated with AIE in APECED. AIE-75 and VillinAb tested negative in Nos. 5–10. cP450c21Ab tested positive in 3/6 patients (Nos. 6, 7, and 9) with gastrointestinal dysfunction. PCA was positive in 1/7 (No. 5), while IFA was present in 1/5 (No. 10) patients ( Table 1 ). Six of six patients’ sera (Nos. 5–10) tested positive for IFNωAb and IFN-α4Ab (2). None of the patients presented celiac-related Abs. No specific AIRE genotype was prevalent.

Eleven of 12 (Nos. 1–8 and 10–12) of the whole series were positive for IFNω and 11/12 (Nos. 1–11) for IFNα4Ab.

Two patients (Nos. 5 and 8) with AIE also presented clinical laboratory parameters indicative of autoimmune hepatitis (vide supra).

In this study, we have characterized the clinical presentation and autoantibody profile of APAH and gastrointestinal manifestations in a series of 14 Italian APECED patients. Of these, nine patients (Nos. 1, 3–8, 11, and 12) were previously reported by our group (13–19), while five patients were newly described in the present study (Nos. 2, 9, 10, 13, and 14) and were added to the published series of 158 APECED patients recruited from across Italy (14).

In the analysis conducted by Garelli et al. (14) on 568 patients from different ethnicities, APAH occurs with an overall prevalence of between 4% and 43% of patients. In particular, in the Italian APECED series, APAH was diagnosed in 21.5% of patients by the end of the follow-up (14). Autoimmune intestinal diseases were diagnosed between 5% and 54% of APECED patients and autoimmune gastritis/pernicious anemia in 4% to 45% of patients. In the Italian APECED series, 29.7% of patients developed autoimmune intestinal dysfunction; autoimmune gastritis was diagnosed in 25.9% of APECED patients, and pernicious anemia was diagnosed in 21/41 of patients.

The analysis of the present Italian APECED series showed a significant expansion of disease manifestations with 35 clinical entities in varying frequencies (from 7% to 78%, Figure 1A , Table 1 ) with a median of 5.5 manifestations per patient (mean 6.6; range, 1–18; Figure 1B ). The observed spectrum and frequency of endocrine and non-endocrine manifestations (median of 1.5 endocrine manifestations per patient; mean 2; range, 0–5 ( Figure 1C ); median of 3.5 non-endocrine manifestations; mean 4.6; range, 1–13 ( Figure 1D ), respectively) were similar to those of published cohorts (rev in 12). Further, the putative association between eight manifestations and APECED/AIRE is highly debatable; these include IgA deficit, septic arthritis, periodic rash-associated fever, kidney cysts, headaches, polyps, posterior reversible encephalopathy syndrome (PRES), and abscesses.

Of note, organ-specific non-endocrine manifestations were enriched in the present series with autoimmune hepatitis and gastrointestinal dysfunction seen in 57% (814) and 50% (7/14) of patients, respectively ( Supplementary Figure 1 ). Remarkably, in published cohorts, circulating autoantibodies were found in approximately 80% of non-APECED patients affected by AIH (28). Two clinically distinct subtypes of AIH can be distinguished based on Ab patterns: AIH-1 and AIH-2 (29). AIH-1 patients had the presence of circulating ANA and/or SMA. Antibodies directed against SLA/liver-pancreas antigen (LP) (30, 31) specifically occurred in AIH patients. Abs against neutrophilic antigens can be at high prevalence (32). A little overlap was found between AIH-1 and AIH-2 patients (28, 29). Patients affected by AIH-2 were characterized by LKM-1Ab. Often associated with AIH-2 was LC1 and less frequently Abs against UDP-glucuronosyltransferases (LKM3Ab). The prevalence and diagnostic value of autoantibodies associated with idiopathic AIH remain to be determined for patients with APAH. In APECED, molecular targets of different hepatic autoantigens have been identified: P4501A2 (33, 34), cytochrome P4502A6 (CYP2A6) (35/36), and AADC (35) (vide supra). Remarkably, AADCs were detected in 92% of Finnish patients with APAH, but they were also frequent in those with no other signs of APAH (35).

Overall autoantibodies against TPHAb and AADC were significantly associated in APECED patients of the present series ( Table 1 ). The last AADC specificity was also significantly associated with autoimmune biopsy-proven hepatopathy. CYP1A2 and AADC are known hepatic autoantigens in APECED (34). CYP1A2Ab was detected in the serum of 4/8 patients with APAH and CYP2A6Ab in the serum of 3/8 patients. Overall, these data confirm the association with APAH discovered in the previous series (27). As regards CYP2A6 and CYP1A2, they are two of six cytochromes P450 that are highly expressed in human liver microsomes (36). CYP1A2 seems to be not expressed in extrahepatic tissues, while CYP2A6 is also expressed in human kidneys (37). CYP2D6Ab was detected in the serum of two patients with APAH (Nos. 3 and 12); in one case, it was associated with CYP1A2Ab. As regards CYP2D6, it is the major autoantigen of LKM-1Ab widely used as a diagnostic marker of AIH-2, which was detectable in 95%–100% of patients (28).

Clinical enteropathy was diagnosed in seven patients; AADC Abs tested positive in 5/7 patients, indicating gastrointestinal dysfunction in APECED. Consistent with published literature (38), the presence of TPHAb was significantly associated with APECED being present in the serum of 5/7 patients. In contrast, VillinAb and HarmoninAb tested negative, as they were reported to be detectable only in a few APECED patients, while these are specific markers of IPEX (39).

Among other cP450 disease-related specificities, adrenal and steroid cell antibodies are detectable in APECED patients and directed against cP450ssc and cP450c17 (17αOH) (40–42). In the present study, 5/12 patients had cP450sscAb and 2/12 had cP450c17Ab. In the literature, there is some disagreement on whether cP450c21 is an autoantigen in APECED. As regards this matter, Uibo et al. (1994) reported that this specificity is present in 15/50 APECED patients (42), while Winqvist (1995) (43, 44) and Clemente in the Sardinian cohort failed to detect it (33). In our series, we confirmed the presence of cP450c21Ab in 9/12 APECED patients.

Almost all patients with APECED usually develop high levels of neutralizing Ab anti-type 1 interferons, which are often detectable before the appearance of clinical symptoms of autoimmunity (40). Indeed, 11/14 patients of the present series (except for patient Nos. 9) showed circulating IFNωAb, and 11/14 patients showed circulating IFNα4Ab; in two patients, IFNAb was not investigated.

On a general ground, a genotype/phenotype correlation has been rarely observed in the APECED series (rev in 10). In the present small series of APECED patients, the most frequent AIRE mutation was c.967_979del13, which was detected in the DNA of 4/9 patients with APAH as previously reported (10). No specific AIRE genotype was prevalent in APECED patients with gastrointestinal manifestations: this does not allow to establish a clear genotype/phenotype correlation.

Although the application of Ferre-Lionakis expanded criteria (11, 12) would require validation in independent large cohorts of patients, the early enrichment of organ-specific non-endocrine manifestations, similar to the American cohort investigated, has allowed an expedited clinical diagnosis in 11/14 patients (Nos. 1–5, 8–12, and 14) of the present series. On a general ground based on Ferre-Lionakis criteria, the presence of APAH or gastrointestinal manifestations in association with one symptom of the classical triad could lead to a clinical diagnosis that should be further confirmed by AIRE gene sequencing and detection of IFNωAb (11). This emphasizes the importance in the clinical practice of the definition of APAH and gastrointestinal dysfunction for an earlier diagnosis of APECED even in European cohorts. An earlier diagnosis would also allow the prevention of episodes of life-threatening hypocalcemic seizures and adrenal crisis, which are the main manifestations of undiagnosed APECED.

The datasets presented in this article are not readily available because of ethical consent for patients. All reported mutations of the AIRE gene are already described in the literature. Requests to access the datasets should be directed to corresponding authors and clinicians Marco Cappa, Maria Felicia Faienza, Giuseppe Maggiore, Mariacarolina Salerno.

The studies involving human participants were reviewed and approved by Bambino Gesù Children’s Hospital, IRCCS, Rome, Italy. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

GP, OP, AC, and EC analyzed data and conducted genetic screening. MF, MC, AP, DC, MV, GM, and MS provided patients with clinical data and contributed to writing the manuscript. VL conducted hepatopathy and enteropathy-related Ab determinations. CB contributed to writing the manuscript and provided critical revision. AF designed the study; analyzed, interpreted, and discussed the clinical and laboratory data; and wrote the manuscript. All authors contributed to the article and approved the submitted version.