Clinical Findings, Response to Steroid Treatment, and Recurrence Rate in Alopecia Areata Patients with or without a Nonsynonymous Variant of Coiled-Coil Alpha-Helical Rod Protein 1

Alopecia areata (AA) is a disease in which oval nonscarring hair loss occurs on the head or the whole body, sometimes recurrently1. Although factors such as autoimmunity and heredity are thought to be involved, the precise cause of AA remains unknown. A study investigating the incidence of AA and the predominant age and sex groups affected by the disease in Olmsted County, Minnesota, United States, reported an incidence of 20.2 per 10,000 people (0.2%), a mean age of the affected population of 33 years, and no sex differences in incidence2. In addition, genome-wide association studies have suggested an association between AA and the gene encoding human leukocyte antigen (HLA) located on the short arm of chromosome3, which is known to be genetically associated with number of autoimmune diseases4, 5, 6.

We recently discovered a variant (rs142986308, p. Arg-587Trp) in the coiled-coil alpha-helical rod protein 1 gene (CCHCR1) as the only non-synonymous variant un the AA risk haplotype3. The mice engineered to have this risk allele showed AA-like hair loss. These results indicate the possibility of elucidating the pathogenesis of AA, the cause of which is unknown, and developing new diagnostic methods and treatment methods for each type based on the presence or absence of risk alleles. The variant rs142986308 was detected only in East Asian individuals (allele frequency: 0.0277), suggesting that it is population specific. According to the alle frequency aggregator project, which provides total allele frequencies from the database of Genotypes and Phenotypes, the minor allele frequency of rs142986308 is A=0.001434/280. We also reported that approximately 15% of AA patients carry a mutated CCHCR1 gene, that CCHCR1-variant genome-edited mice and CCHCR1 knockout mice show AA-like hair loss, and that expression of keratinization-related proteins involved in hair shaft formation is altered in both mice and AA patients carrying the variant3, 7. These findings suggest involvement of CCHCR1 in the pathogenesis of AA. CCHCR1 has been reported to be related to psoriasis vulgaris, and there are some studies suggesting a relationship with keratinizing-related protein, keratin, and hair-related protein. In this study, we statistically analyzed the clinical findings of AA in patients with the CCHCR1 variant. Patients with the variant responded well to steroid treatment but showed a significantly higher recurrence rate than the variant-negative group.

All 227 patients diagnosed with AA were of Japanese origin. Of the 227, the variant was identified by DNA analysis in 25 (11.0%) patients. We conducted a retrospective analysis of the data of 142 patients who could be followed up. Of the 142 patients, 20 (14.1%) had the variant, while the remaining 122 (85.9%) did not.

The group with the variant included 8 males (40.0%) and 12 females (60.0%), and the group without the variant consisted of 48 males (39.3%) and 74 females (60.7%) (Fig. 1A).

Fig. 1
The relationship between each parameter with and without the variant. There was a significant difference between the patients with or without a variant in terms of recurrence. On the other hand, there was no significant difference in other clinical features. In each figure, the group with Variant is on the left side, and the group without Variant is on the right side. (A) Sex. (B) Age distribution. (C) Pattern of AA. (D) Family history of Alopecia areata (AA). (E) Atopic dermatitis complication. (F) Oral corticosteroids. (G) Therapeutic effect. (H) Recurrence. N.S.: nonsignificant, AU: alopecia universalis, AT: alopecia totalis, MAA: multifocal AA, SAA: solitary AA. ^*Statistical significance by Fisher’s exact test (p<0.05).

• Click for larger image
• Download as PowerPoint slide

The age at onset of hair loss (mean±standard deviation) was 33.45±10.52 years in the group with the variant and 27.52±2.84 years in the group without the variant. Thus, while the age at onset of hair loss was not significantly different between the two groups, it tended to be higher in the group with the variant. Fig. 1B shows the age distribution in the two groups. In the group with the variant, although AA was experienced by males of all ages, in the female subgroup, the disease mainly manifested in women who were in their 30s or older. On the other hand, in the group without the variant, AA was not experienced by any male patients in their 40s or older, while it manifested in females of all ages.

Next, regarding the types of hair loss in patients with AA, multifocal AA (MAA) was the most common (n=12) type, accounting for 60.0% of all cases, followed by alopecia universalis (AU) (n=4; 20.0%) and alopecia totalis (AT) (n=4, 20.0%) in the group with the variant. None of the patients had solitary AA (SAA) or ophiasis AA (OAA). On the other hand, 59 (48.4%) had MAA, 45 (36.9%) had AU, 10 (8.2%) had AT, 4 (3.3%) had OAA, and 4 (3.3%) had SAA in the patient group without the variant. (Fig. 1C). Five (25.0%) of the 20 patients with the variant and 32 of the 122 patients (26.2%) without the variant had a positive family history (Fig. 1D). Five of the 20 (25.0%) patients with the variant and 29 of the 122 patients (23.8%) without the variant had atopic dermatitis (including childhood atopic dermatitis alone) (Fig. 1E).

Of the 20 patients with the variant, 15 (75.0%) received systemic steroid therapy (prednisolone 0.2~0.4 mg/kg/day). Of the remaining 5 patients in this group who did not receive systemic steroid therapy, 4 had MAA, and the remaining 1 had AT. These patients received a local injection of betamethasone sodium phosphate and topical betamethasone butyrate propionate application. Of the 122 patients without the variant, 68 (55.7%) received systemic steroid therapy (Fig. 1F), and the remaining received local injection of betamethasone sodium phosphate injection, topical betamethasone butyrate propionate application, oral glycyrrhizin/glycine/DL-methionine combination tablets, oral cepharanthine tablets, or other treatments.

Regarding the therapeutic efficacy, all 20 (100%) patients with the variant responded to systemic or topical steroid therapy; however, all 20 (100%) patients experienced recurrence within a relatively short time after therapy discontinuation. On the other hand, in the group without the variant, the symptoms improved or resolved in 65 of the 68 patients (95.6%) who had received steroid therapy and 47 of the 54 patients (87.0%) who had not. Of the 112 who responded to treatment, 48 (70.6%) of the patients who had received steroid therapy and 42 (77.8%) of the patients who had not received steroid therapy developed recurrence of the disease (Fig. 1G, H). Two-tailed Fisher’s test revealed that the recurrence rate was significantly higher in AA patients with the variant than in those without the variant (p=0.0248) (Fig. 1H). In addition, no risk factors such as sex, age at onset, type of hair loss (solitary, universalis, etc.), family history of AA, and/or atopic dermatitis complications were identified. Electron microscopic observations of hair samples revealed disorganized hair epithelium in the close skin surface and intermediate region in patients with the variant compared to those without the variant. Moreover, the long-axis/short-axis ratio was close to 1 in the patient without the variant, but it was significantly larger in the group with the CCHCR1 variant (p=0.00174). This result suggests that the transverse hair sections were circular in patients without the variant but elliptical in those with the variant (Fig. 2, 3).

Fig. 2
Electron microscopic findings of hair. Electron microscopic findings of hair were performed using AA patients. Electron microscopic findings showed abnormalities in hair shaft structure and hair epithelium in the hair of patients with CCHCR1 variants but no abnormalities in the hair morphology of patients without CCHCR1 variants. CCHCR1: coiled-coil alpha-helical rod protein 1.

• Click for larger image
• Download as PowerPoint slide

Fig. 3
Comparison of the long-axis/short-axis ratio of hair. Comparison of the long-axis/short-axis ratio of hair between the patients with the variant and the patient without the variant. The ratio was significantly larger in the group with the CCHCR1 variant (p=0.00174). ^*Significance by Fisher’s exact test (p<0.05).

• Click for larger image
• Download as PowerPoint slide

The prevalence and lifetime incidence of AA are thought to be 0.1% to 0.2% and approximately 2%, respectively. AA occurs even in young age groups, and in particular, AT or AU markedly reduces the quality of life of patients by affecting their appearance10. Although treatment with steroids or local immunotherapy improves AA in many cases, there are some refractory patients, and others develop repeated recurrences, with no adequately effective treatment having been developed to date. The autoimmune response against the anagen hair bulb is thought to be the major mechanism underlying the development of AA. As the background for such predisposition, many genes and other factors are thought to be involved in the development of AA, and research to identify the precise causative genes for AA is underway around the world.

We recently identified the CCHCR1 gene as a susceptibility gene by association analysis and next-generation sequencing of the HLA region. We further revealed that CCHCR1-variant genome-edited mice and CCHCR1 knockout mice develop AA-like lesions, indicating that the CCHCR1 variant or defect is a risk factor for AA6, 7.

Among AA patients, while the clinical symptoms look similar, the clinical course is often variable. For example, in cases of AT, steroid pulse therapy is often used, but the response to therapy can differ. Therefore, it remains difficult to predict the clinical course and determine who will relapse before treatment. It would be favorable to know which factors (i.e., age at onset, clinical types, family history of AA, complications of atopic dermatitis) could predict the risk of recurrence of AA11, 12, 13, 14, 15, 16, 17.

In this study, we compared the sex, age at onset, clinical types, family history, association of atopic dermatitis, response to steroid therapy, and recurrence rate between patients with and without the variant. We also evaluated the structure of hair obtained from patients with or without the variant by electron microscopy.

The results showed that the mean age at disease onset in patients with the variant was 33.45±10.52 years and tended to be higher than that in the group without the variant. In addition, in the group without the variant, AA was experienced by women of all ages, while in the group with the variant, there were no cases in women younger than 30 years of age (Fig. 1B). The relationship between CCHCR1 variants and age of onset of AA is currently unknown.

The CCHCR1 gene has been reported to be involved in the development of the hair shaft structure6, 7. Based on electron micrographs of hair samples from patients with the variant, the hair epithelium was disordered, with cross-sectional deformation. In general, immune privilege collapse around hair follicles is thought to be the mechanism underlying the development of AA18, 19, 20. Our previous and present studies imply that other pathological conditions, such as hair fragility, may be involved in the development of AA in a subset of patients with the CCHCR1 variant6, 7. Our basic research reveals that rack of CCHCR1 shows the reduction of the expression of keratinization-related proteins and the expression of hair-related genes. Fuentes-Duculan et al.21 showed downregulation of several hair keratins in lesions of AA patients and demonstrated concurrent activation of Th1 and Th2 immune axes in AA. Similarly, Th1/Th2 may act together in CCHCR1-related AA, and we also need to analyze cytokine expression in the future.

AA is a multifactorial disease involving many triggers. Variant of CCHCR1 gene is thought to be one of many causative factors. For example, it has been reported that the combination of certain drugs and certain genetic mutations can cause generalized pustular psoriasis22. Similarly, this type of AA may be a multifactorial disease triggered by stress load and other factors.

Regarding the function of the CCHCR1 protein, it has been reported to be involved in multiple different biological processes, including not only keratinization of the skin and hair but also cytoskeletal regulation and muscle differentiation23, 24, 25, 26. Furthermore, a CCHCR1 gene variant (CCHCR1*WWCC) has been reported to reduce the amount of pregnenolone in steroidogenic cells and may have some effect on the intrinsic immune privilege of the anagen hair bulb27. The finding that all patients with the variant responded to treatment but had a high rate of recurrence might support this possibility.

The treatment of AA has entered a new era, and not only steroids but also JAK inhibitors have been added to the treatment options. CCHCR1 may be used as a prognostic marker for choosing short-term steroid administration in deciding treatment options, and further research is needed to investigate the relationship between AA and CCHCR1.

Our goal for the future is tailor-made medical treatment, such as whether to perform systemic steroid therapy according to diagnosis based on the presence or absence of the CCHCR1 variant. This approach may reduce the frequency of unnecessary systemic steroid therapy, and further studies are needed to increase the number of patients.

This work was supported by the Department of Epidemiology and Environmental Health, Biomedical Research Core Facilities, Juntendo University Graduate School of Medicine.