Molecular genetic exploration of frontal fibrosing alopecia and uncharacterised hair disorders

Student thesis: Doctoral ThesisDoctor of Philosophy


Frontal fibrosing alopecia (FFA) is an increasingly prevalent sub-type of lichen planopilaris (follicular lichen planus), associated with inflammation and scarring hair loss mainly affecting eyebrows and frontotemporal scalp. In 2019, a genome-wide association study including 1,016 female FFA cases has characterised this disorder as an immune-mediated inflammatory trait with a strong genetic basis, identifying four susceptibility loci at 2p22.2, 6p21.1, 8q24.22, and 15q26.1. FFA mainly affects postmenopausal women, but cases in younger females and males have also been reported. The genetic aetiology underlying male FFA remains obscure, as it only represents 3-5% of FFA cases and remains exceedingly rare. Establishing a better understanding of these disorders is expected to provide new insight into physiological and pathological molecular mechanisms in hair development and disease.

Phenotypic data from a new case series of male FFA subjects suggests that male and female disease have overlapping clinical features, albeit with a younger onset in males and greater facial hair involvement. A total of 92 male FFA cases from three different cohorts (Northern European, Spain, and Greece) were genotyped and data were aligned with unselected, ancestry-matched male controls. Data demonstrated that male FFA is associated with risk locus 6p21.1 (HLA-B) at genome-wide significant level (ORMETA:3.95, 95% CI: 2.48-6.29, P-value: 6.9 × 10-9). Significant association was also observed for the missense variant in CYP1B1 (rs1800440, p.Asn453Ser) at 2p22.2 (ORMETA: 2.36, CI: 1.40-3.98, P-value: 1.2×10-3). Genetic risk score calculation comprising four significant female FFA risk loci was higher in male FFA than controls (p<0.05) in all three cohorts.

To establish additional genetic risk loci associated with female FFA, additional female participants of European ancestry were ascertained from the UK and Spain and added to the previously established cohort of 1,016 female FFA patients to conduct an expanded genome-wide meta-analysis of FFA in females. The new meta-analysis (Ncases:1,585, Ncontrols: 5,083) confirmed the association at 6p21.1 (HLA-B, ORMETA: 4.62, 95% CI: 4.15 - 5.20, P-value: 1.5 × 10-154), as the strongest locus in female FFA, and at 2p22.2 (CYP1B1, ORMETA:1.55, 95% CI: 1.37 - 1.75, P-value: 2.4 × 10-12) and 15q26.1 (SEMA4B, ORMETA:1.45, 95% CI: 1.29 – 1.64, P-value: 4.0 × 10-10). Step-wise conditional analysis of classical HLA-imputed alleles reveal four independent HLA-alleles associated with FFA, including HLA-B* 07:02, HLA-A*33:01, HLA-A*11:01 and HLA-B*35:01. Further, a novel genomic locus was identified at 5p15 which corresponds to ERAP1 (ORMETA: 1.3, 95%CI: 1.19 - 1.43, P-value: 3.6 × 10-8). Gene-gene interaction analysis showed that the risk locus in ERAP1 (rs10045403) is in epistasis with the HLA-B*07:02 allele (P-value = 0.013).

To evaluate a possible interaction between the CYP1B1 risk allele at 2p22.2 and the oral contraceptive pill (OCP) as a potential environmental trigger in the pathogenesis of FFA, a gene-environment interaction (GxE) study with a case-control design was performed using genotype and retrospective clinical data of female FFA subjects (N=493) from the UK and selected controls from UK-Biobank, matched based on age and OCP status (N=2,006). A formal GxE analysis showed interaction between risk locus in CYP1B1 (rs1800440) and OCP status (ORADD×OCP: 1.63, 95%CI: 1.02 - 2.61, P-value: 0.04). Further stratified analysis indicated that rs1800440 showed significant association with FFA only in the OCP user group, but not in the non-OCP user group.

Aside from FFA, this study also aimed to identify rare causal genetic variants in several Mendelian hair disorders that are still poorly characterised. Whole-exome sequencing based analysis of a clinical series of in-house uncharacterised hair disorders identified rarepathogenic variants in genes already linked to monogenic hair-related disorders, including CDSN, RPL21, and WNT10A, thus improving the genotype-phenotype correlation of these conditions.

In summary, this thesis provides the novel evidence of genetic risk
background in male FFA at 6p21.1 and 2p22.2; confirms the association with the previously established FFA risk loci at 6p21.1, 2p22.2, and 15q26.1; and identifies a novel genomic risk locus at 5q15. These findings further support the notion that FFA is underpinned by robust genetic aetiology related to antigen presentation, inflammatory response, and xenobiotic metabolism. This study also demonstrates a gene-environment interaction between OCP and FFA risk allele in CYP1B1. In addition, this study also extends the genotype-phenotype paradigm for some poorly characterised Mendelian hair disorders.

Date of Award1 Dec 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorJohn McGrath (Supervisor), Christos Tziotzios (Supervisor) & Nick Dand (Supervisor)

Cite this