Low-Dose Metformin and Profibrotic Signature in Central Centrifugal Cicatricial Alopecia

Key Points

Question

Can low-dose oral metformin lead to clinical and molecular signs of improvement in patients with treatment-refractory central centrifugal cicatricial alopecia (CCCA)?

Findings

In this case series of 12 patients with CCCA treated with daily metformin, 500 mg, 9 patients experienced improvement and 6 had clinical evidence of hair regrowth after 6 months. Transcriptomic analysis of scalp biopsies from 4 patients revealed upregulation of hair growth pathways and downregulation of fibrotic pathways following treatment.

Meaning

Low-dose oral metformin may reverse the fibrotic transcriptional signature in CCCA and promote hair regrowth, suggesting its potential as a targeted therapy for this scarring alopecia.

This case series investigates the associations of low-dose oral metformin with clinical symptoms and scalp gene expression patterns in patients with central centrifugal cicatricial alopecia.

Abstract

Importance

Central centrifugal cicatricial alopecia (CCCA) is a scarring alopecia predominantly affecting Black female individuals. Current conventional treatments target inflammation but not the underlying fibrotic processes, often leading to permanent hair loss.

Objective

To investigate the associations of low-dose oral metformin, an antidiabetic medication with antifibrotic properties, with clinical symptoms and scalp gene expression patterns in patients with CCCA.

Design, Setting, and Participants

This retrospective clinical case series and transcriptomic analysis included patients treated at a single tertiary academic medical center between January 2023 and March 2024. All patients had biopsy-confirmed CCCA refractory to standard treatments. Transcriptomic analysis was performed on patients with previously banked, paired scalp biopsies before and after treatment with adjuvant metformin for at least 6 weeks.

Exposure

Extended-release metformin, 500 mg, once daily was added to participants’ baseline CCCA treatment regimens.

Main Outcomes and Measures

Clinical assessments included pruritus, inflammation, scalp resistance, and hair regrowth. Gene expression profiling via bulk RNA sequencing analysis evaluated differential gene expression and pathway enrichment.

Results

A total of 12 Black female participants were included in the study, and transcriptomic analysis was performed in 4 participants. After at least 6 months of metformin treatment, 9 participants experienced improvement in disease, including scalp pain, inflammation, and/or pruritus, and 6 demonstrated clinical evidence of hair regrowth. The addition of metformin led to reversal of many prominent gene pathways previously identified in CCCA. Transcriptomic analysis revealed upregulation of pathways and genes (keratin-associated proteins [KRTAPs]) involved in keratinization, epidermis development, and the hair cycle (absolute log₂-fold change > 4), with concomitant downregulation of fibrosis-related pathways and genes (eg, MMP7, COL6A1) (fold change >1.5; all false discovery rate <.05). Gene set analysis showed reduced expression of helper T cell 17 and epithelial-mesenchymal transition pathways and elevated adenosine monophosphate kinase signaling and KRTAPs after metformin treatment.

Conclusions and Relevance

In this case series of patients with treatment-refractory CCCA, low-dose oral metformin was associated with symptomatic improvement and dual modulation of gene expression, stimulating hair growth pathways while suppressing fibrosis and inflammation markers. These findings provide a rationale for future clinical trials studying metformin as a targeted therapy for CCCA and other cicatricial alopecias.

Introduction

Cicatricial or scarring alopecias are a group of chronic inflammatory hair disorders characterized by permanent hair loss resulting from destruction of the hair follicle and replacement with fibrous scar tissue. These disfiguring conditions substantially impair patients’ quality of life. Central centrifugal cicatricial alopecia (CCCA) is the most common form of primary lymphocytic cicatricial alopecia and predominantly affects Black female individuals. Patients with CCCA have an increased risk of systemic comorbidities, particularly type 2 diabetes (T2D).¹

Current treatments for CCCA, including topical glucocorticoids, intralesional triamcinolone injections, and oral tetracycline antibiotics, primarily target inflammatory aspects of the disease but do not address the underlying processes of fibrosis and scarring that ultimately lead to hair loss.² Gene expression profiling of CCCA-affected scalps has revealed upregulation of profibrotic transcripts,³ sharing molecular signatures with abnormal scarring disorders, such as idiopathic pulmonary fibrosis and systemic sclerosis. In cases where fibrosis persists amid low-grade inflammation, decreased vascularity of fibrotic tissues may further impair drug delivery, diminishing the effectiveness of anti-inflammatory therapies. While medications like topical and oral minoxidil can be used concurrently if there is suspicion for concomitant female pattern hair loss, these agents do not directly target the core pathogenesis of CCCA and have limited efficacy.⁴ Thus, a strategic focus on interventions targeted at reversing fibrosis is warranted.⁵

Metformin is an antidiabetic drug that upregulates adenosine monophosphate kinase (AMPK), enhancing insulin sensitivity. Preclinical studies have demonstrated metformin’s antifibrotic effects via IL1β-mediated inhibition of transforming growth factor β signaling and collagen production.^6,7 Furthermore, metformin blocks the angiotensin II pathway leading to fibroblast-to-myofibroblast transition.^8,9 With its approval for treating T2D, a comorbidity enriched in patients with CCCA, metformin represents a promising repurposed therapeutic candidate.^4,10 A preliminary case series has demonstrated potential clinical utility of topical metformin for CCCA, but the role of systemic oral metformin in cicatricial alopecias remains unexplored.^11,12

We present a case series evaluating the clinical associations of low-dose extended-release oral metformin, 500 mg, daily in 12 patients with treatment-refractory CCCA. To evaluate the impact of metformin on gene expression, bulk RNA sequencing was performed on a subset of 4 patients who had existing scalp biopsies both before and after treatment with metformin.

Methods

Study Participants

This is a retrospective case series of 12 participants seen at the Johns Hopkins alopecia clinic with biopsy-confirmed CCCA that had been clinically refractory to conventional therapies. Given the prevalence of CCCA among Black patients, demographic data, including sex and race, were collected through patient self-reporting. All participants had remained taking their baseline CCCA treatment regimen for at least 6 months with stagnant or worsening clinical symptoms prior to adding oral extended-release metformin, 500 mg, once daily as an adjunct treatment. Duration between pretreatment biopsy and initiation of metformin was within 1 week. Medical records were reviewed for changes in clinical symptoms (pruritus, inflammation, pain, scalp resistance, and hair regrowth) following metformin initiation. Scalp resistance is a subjective measurement reflecting the amount of force required to inject the scalp, with 0 being normal (as would be expected in nonscarring alopecia, like alopecia areata), 1 being mild resistance, and 2 being significant resistance, oftentimes leading to bending of the needle on insertion. A total of 10 of 12 patients received baseline hemoglobin A_1c screening prior to starting metformin, of whom 3 were in the prediabetic range. One patient (patient 12), who was treated with low-dose metformin by her primary care physician 1 week prior to evaluation, had been diagnosed with T2D and was taking low-dose metformin as monotherapy. This study was reviewed and approved by the Johns Hopkins Institutional Review Board. Oral informed consent was obtained from all participants. and written informed consent was obtained from all patients who received biopsies. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Tissue Sampling

A total of 4 of 12 patients had existing scalp biopsy samples in the alopecia tissue repository prior to and after at least 6 weeks of sustained oral metformin treatment (range, 6 to 31 weeks after treatment start). Biopsy samples were originally obtained from areas of active disease on the scalp vertex.

Sample Processing and Sequencing

Fresh tissue samples were immediately submerged in a 2-mL tube containing RNAlater (QIAGEN) stabilization reagent. Tissue was stored at 4 °C for 24 hours and transferred afterward to a freezer for long-term storage at −80 °C. Total RNA was later extracted using the RNA extraction RNeasy Kit (QIAGEN) according to vendor protocols. Pretreatment and posttreatment tissue samples were subsequently processed in a single batch. RNA purity and concentration were measured using NanoDrop (Thermo Fisher Scientific). The RNA library preparation and subsequent transcriptome sequencing were outsourced to Novogene. mRNA was extracted and purified using polyT oligo-attached magnetic beads. Following fragmentation, the first strand of cDNA was synthesized with random primers, while the second strand of cDNA was generated using deoxyuridine triphosphate for directional and deoxythymosine triphosphate for nondirectional libraries. Library integrity and size distributions were assessed through Qubit (Thermo Fisher Scientific) and real-time polymerase chain reaction. The quantified libraries were pooled and sequenced on the HiSeq 2500 platform (Illumina). Percentage error rate, guanine-cytosine distribution, and sequence read filtering was performed for quality control (eTable 1 in Supplement 1). Genes with an average cohort count less than 10 were systematically excluded from subsequent analyses to enhance result validity and statistical reliability.

Differential Gene Expression Analysis and Statistical Analysis

Differential gene expression analysis comparing premetformin and postmetformin samples was performed using the DESeq2 R package, version 1.40.2 (R Project for Statistical Computing), which uses a Wald test. The Benjamini-Hochberg procedure was used to control for false discovery rate (FDR). Genes with an adjusted FDR less than 0.05 and an absolute log₂-fold change of 0.5 or greater were considered differentially expressed.

Gene Set and Pathway Analysis

Gene set enrichment analysis (GSEA) was conducted using the local version of the GSEA analysis tool (Broad Institute) to evaluate enrichment of Gene Ontology molecular function, cellular component, and biological process pathways in the set of differentially expressed genes. Pathways with an adjusted 2-tailed P < .05 were considered significantly enriched.

In addition, gene set variation analysis (GSVA) using the GSVA package version 1.50.0 generated pathway enrichment scores for each sample. GSVA gene sets were curated from the literature of relevant genes associated with known pathways. Prior to GSVA, raw counts underwent variance-stabilizing transformation using DESeq2 version 1.40.2. The limma package version 3.58.1 was used to perform moderated 2-tailed t tests to assess statistical differences in GSVA scores between the premetformin and postmetformin samples.

Results

Clinical Improvements and Hair Regrowth

The study included 12 Black female patients with treatment-refractory CCCA. Eight patients experienced noticeable improvement in clinical symptoms after metformin treatment, including symptoms of scalp pain, scalp resistance, pruritus, and inflammation (Table 1), and 2 patients noted continued worsening of symptoms since metformin initiation. Six patients demonstrated clinical evidence of hair regrowth after sustained treatment for at least 6 months, although 1 patient experienced subsequent regression 3 months following metformin discontinuation (Figure 1).

Table 1. Patient Cohort Characteristics at Baseline and Following Therapy With Daily Extended-Release Metformin, 500 mg.

Patient, No.; sex; age, y	Treatments prior to metformin treatment	Pretreatment symptomsa	Posttreatment symptoms	Improvement noted	Time to symptom improvement, mo	Notes
1; Female; 50sb	Clobetasol, 0.05%, ointment Compounded minoxidil, 7%, and RA, 0.025% Fluocinolone, 0.01%, oil ILK Minoxidil, 2%, solution	None	None	No	NA	Had discontinued all treatments prior to initial biopsy; no clinical regrowth
2; Female; 30sb	Compounded metformin, 10%; clobetasol, 0.05% Doxycycline, 100 mg, twice daily Minoxidil, 2.5 mg, daily ILK	Scalp resistance	None	Yes	5	Was being treated with oral minoxidil at time of initial biopsy but discontinued on day of metformin initiation; clinical appearance of regrowth at 6 mo
3; Female; 30sb	Compounded metformin, 10%, in lipoderm cream Fluocinolone, 0.01%, oil ILK	None	None	No	NA	Had discontinued all treatments at time of initial biopsy; no clinical regrowth noted
4; Female; 50sb	Clobetasol, 0.05%, ointmentc Doxycycline, 100 mg, twice daily ILKc Oral minoxidil, 2.5 mgc	Pruritus	None	Yes	1	Was taking oral minoxidil, clobetasol, and ILK at time of initial biopsy; clinical appearance of regrowth at 6 mo
5; Female; 40s	Clobetasol, 0.05%, ointment Compounded RA, 0.025%, minoxidil, 7%, and fluocinolone, 0.01%, oil ILKc Oral minoxidil, 2.5 mg	Pruritus	Pruritus	Yes	3	No clinical regrowth
6; Female; 50s	Doxycycline, 100 mg, twice daily Compounded minoxidil, 8%, and RA, 0.025%c Fluocinolone, 0.01%, oil ILKc	Pruritus, scalp pain, and scalp resistance	Pruritus (decreased) and scalp pain	Yes	6	All symptoms noted in first follow-up after metformin started, which resolved after 3 mo more of metformin use; no clinical regrowth
7; Female; 50s	Clobetasol, 0.05%, ointment Compounded metformin, 10%, in lipoderm cream ILKc Oral minoxidil, 2.5 mg	Scalp resistance	Scalp resistance	No	NA	Slight worsening of scalp resistance noted between visits; no clinical regrowth
8; Female; 50s	Clobetasol, 0.05%, ointmentc Finasteride, 2.5 mg ILKc Oral minoxidil, 2.5 mgc Plasma-rich platelet injection	Pruritus	None	Yes	3	Clinical appearance of regrowth at 6 mo
9; Female; 30s	Compounded metformin, 10%, and clobetasol, 0.05%c ILKc Plasma-rich platelet injection	None	Pruritus and scalp resistance	No	NA	Noted clinical worsening of symptoms after starting oral metformin
10; Female; 30s	Clindamycin Clobetasol, 0.05%, ointment Compounded RA, 0.025%, minoxidil, 8%, and TAC Fluocinolone, 0.01%, oilc ILKc	Pruritus and erythema	None	Yes	6	Clinical appearance of regrowth at 6 mo
11; Female; 40s	Clobetasol, 0.05%, ointment Compounded metformin, 20%, and clobetasol, 0.05%, ointmentc Doxycycline, 100 mg, twice daily ILKc Minoxidil, 5%, solution Minoxidil, 2.5 mg, daily	Scalp resistance	None	Yes	4	Clinical appearance of regrowth at 6 mo but subsequent regression noted 3 mo after discontinuation of metformin
12; Female; 50sd	Fluocinolone, 0.01%, oil Compounded minoxidil, 8%, RA, 0.025%, and TAC ointmentc Minoxidil, 2.5 mg, dailyc	None	None	Yes	NA	Clinical appearance of regrowth at 6 mo

Abbreviations: ILK, intralesional triamcinolone acetonide; NA, not applicable; RA, retinoic acid; TAC, triamcinolone acetonide.

^a Pruritus and scalp pain were patient reported. Scalp resistance is a subjective measurement reflecting the amount of force required to inject the scalp during ILK administration.

^b RNA sequencing performed on banked samples before and after addition of metformin to treatment regimen. Baseline samples taken after patients had been taking treatment regimens for at least 6 months.

^c Simultaneous usage during metformin treatment period.

^d Patient was given immediate-release metformin, 500 mg, tablets twice daily by primary care physician.

Figure 1. Results Before and After Treatment With Metformin, 500 mg, Daily for 6 Months or More.

Patients at baseline with evidence of decreased density in areas of follicular dropout on the scalp, characteristic of central centrifugal cicatricial alopecia with improvement after metformin was added to treatment. Each patient was taking metformin treatment for at least 6 months between images without other changes to their regimen.

Differential Gene Expression Analysis

In the transcriptomic analysis of 16 655 genes, we identified 34 genes that were upregulated and 8 genes that were downregulated posttreatment (FDR less than 0.05; absolute log₂-fold change greater than 0.5). Gene set analysis revealed enriched upregulated pathways related to keratinization, epidermis development, and hair cycle (FDR less than 0.001). Upregulated genes were associated with insulin regulation (SGK1: fold change, 0.51; FDR = 0.0288), immune processes (IGHG1: fold change, 0.66; FDR = 0.0064; IGHG2: fold change, 0.83; FDR = 0.0007; NFIL3: fold change, 0.60; FDR = 0.0400), and calcium signaling (S100P: fold change, 0.94; FDR = 0.0424). Notably, 23 different hair keratin-associated proteins (KRTAPs) were upregulated after treatment, encompassing members of the KRTAP9, KRTAP4, KRTAP3, KRTAP2, and KRTAP1 families (all fold change >4) (Figure 2A-C; eTable 2 in Supplement 1).

Figure 2. Differential Gene Analysis of Metformin in Central Centrifugal Cicatricial Alopecia .

A, Heatmap of entire gene signature (n = 16 655) of patients before and after metformin treatment. Distinct gene differentiation patterns emerged in scalp before and after treatment. B, Heatmap subset of all differentially expressed genes, clustered using hierarchical clustering. C, Volcano plot displaying fold change and −log₁₀ P value for all genes. Differentially expressed genes are labeled.

Enriched downregulated pathways included those associated with extracellular matrix organization, collagen fibril organization, and collagen metabolism (all fold change >1.5). Prominently downregulated genes included matrix metalloproteinase 7 (MMP7; fold change, −1.76; FDR = 0.0018), collagen VI alpha 1 (COL6A1; fold change, −0.55; FDR = 0.0179), and dermcidin (DCD; fold change, −1.1; FDR = 9.78 × 10⁶) (Figure 3A).

Figure 3. Gene Set and Pathway Analysis of Metformin in Central Centrifugal Cicatricial Alopecia .

A, Top upregulated and downregulated pathways enriched in gene set enrichment analysis. The blue line indicates P = .05. B, Gene set variation analysis (GSVA) scores before and after metformin treatment using 8 distinct gene sets representing helper T cell 1 (T_H1), T_H2, T_H17, T_H22, epithelial-mesenchymal transition (EMT), keratin-associated protein (KRTAP), lipid metabolism, and adenosine monophosphate kinase (AMPK) signaling pathways. Differential expression was noted in pathways representing AMPK signaling, EMT, and T_H17. The dots represent individual values for each patient before and after metformin. The midline indicates the median value, the box indicates the lower quartile to upper quartile, and the whiskers reflect the highest and lowest values.

^aP < .05.

GSVA

GSVA quantified posttreatment expression within established pathways. Notably, there was a reduction in the mean posttreatment expression of gene sets associated with helper T cell 17 (T_H17; mean [SD] GSVA score: pretreatment, 0.17 [0.11]; posttreatment, −0.11 [0.23]; P = .04) and epithelial-mesenchymal transition (mean [SD] GSVA score: pretreatment, 0.22 [0.21]; posttreatment, −0.20 [0.24]; P = .03). Conversely, an AMPK signaling gene set (mean [SD] GSVA score: pretreatment, −0.10 [0.13]; posttreatment, 0.11 [0.08]; P = .02) and a gene set comprising all KRTAPs (mean [SD] GSVA score: pretreatment, −0.49 [0.14]; posttreatment, 0.45 [0.28]; P < .001) exhibited elevated expression following metformin treatment (Figure 3B).

Discussion

Our study found that, in patients with treatment-refractory CCCA, low-dose oral metformin was associated with downregulation of fibrotic processes and concurrent upregulation of keratinocyte proliferation and hair cycle–related processes. Notably, despite patients already receiving standard therapies, baseline sequencing still showed expression patterns similar to those in untreated affected scalps,⁴ a pathogenic state that was largely reversed with metformin (Table 2). Additionally, most patients experienced symptomatic improvement after initiating metformin treatment. Of the 6 patients who showed clinical signs of regrowth, half had not taken minoxidil during 6 months of metformin usage, suggesting that metformin may help with regrowth in select patients. One patient noted disease worsening within 3 months of treatment discontinuation, which may reflect the need for long-term treatment in patients.

Table 2. Summary of Previously Characterized Pathways Dysregulated in Central Centrifugal Cicatricial Alopecia (CCCA) Pathogenesis and Metformin Treatment-Associated Pathway Changes.

Gene ontology pathway	CCCA-defined dysregulationa	Posttreatment with metforminb
Collagen fibril organization	Upregulated	Downregulated
Extracellular organization	Upregulated	Downregulated
Collagen catabolic process	Upregulated	Downregulated
Hair cycling	Downregulated	Upregulated
Keratinization	Downregulated	Upregulated
Unsaturated fatty acid biosynthetic process	Downregulated	Upregulated

^a Previously defined gene pathways characteristic of affected scalp in patients with CCCA by Aguh et al.³

^b Relative to pretreatment values.

Antifibrotic Effects of Metformin

Beyond its effects on glycemic control, metformin exhibits antifibrotic activity, particularly through proliferator–activated receptor α–dependent upregulation of AMPK and subsequent downregulation of transforming growth factor β expression.^6,7 While these patients were treated with low-dose metformin, 500 mg, AMPK signaling was still upregulated in scalp tissue, consistent with prior translational studies demonstrating that low-dose metformin is enough to activate AMPK and lower blood glucose in patients.^13,14 However, while low-dose metformin may be sufficient to target fibrosis in patients without diabetes, it is likely inadequate to improve insulin resistance in susceptible individuals. Baseline hemoglobin A_1c levels could help identify those requiring additional monitoring and higher-dose titration. Clinically, metformin has shown benefits in treating fibrotic conditions, like scleroderma and metabolic dysfunction–associated steatotic liver disease.^15,16 In a murine model of idiopathic pulmonary fibrosis, metformin reversed fibrosis by deactivating myofibroblast activity and inducing apoptosis.¹⁷

Metformin as a Therapeutic Agent for CCCA

The potential use of metformin as a therapeutic agent for CCCA has been previously explored. The current findings build on a prior case series demonstrating hair regrowth with topical metformin in patients with CCCA.¹² A retrospective analysis also found increased odds of clinical improvement, as assessed by a central scalp alopecia photographic severity scale, in patients with CCCA who were incidentally taking metformin.¹⁸ Furthermore, transcriptomic profiling has revealed unique differential expression of the AMPK signaling pathway in CCCA-affected scalps compared with other scarring alopecias, like frontal fibrosing alopecia and lichen planopilaris.¹⁹ Lastly, epidemiologic studies have identified an association between CCCA and T2D, with CCCA conferring an increased diabetes risk (hazard ratio, 1.68; 95% CI, 1.38-2.06) in African American women, although the underlying mechanisms remain unclear.²⁰

Effects of Metformin on the Immune System

Metformin’s observed impact on immune pathways, such as the T_H17 axis, may be mediated through AMPK-dependent inhibition of the mammalian target of rapamycin pathway, which plays a crucial role in suppressing T_H1, T_H2, and T_H17 cell differentiation and mitigating production of inflammatory cytokines, such as interferon-γ and IL-17.²¹ Metformin’s anti-inflammatory effects may also stem from its enhancing effect on mitochondrial metabolism and mitophagy, and treatment in patients with T2D was associated with reduced levels of IL-6, tumor necrosis factor–α, and reactive oxygen species.²²

While the precise role of the T_H17 axis in CCCA pathogenesis remains unclear, immunohistochemistry studies demonstrate that IL-17–positive cells are markedly elevated in scalp tissue of patients with lichen planopilaris or discoid lupus erythematosus, 2 of the most frequent causes of primary cicatricial alopecia.²³ Furthermore, the highest cellular infiltration was noted at perifollicular and perivascular regions. Given these observations, T_H17 may be explored as a potential treatment target in cicatricial alopecias.

Other dermatologic conditions, such as hidradenitis suppurativa and scleroderma, are canonically characterized by T_H17 dysregulation and IL-17–mediated inflammation.^24,25 In both conditions, metformin treatment has demonstrated anti-inflammatory effects, immune-metabolic reprogramming, and suppression of T_H17 signaling.^9,26,27 These observations establish therapeutic precedents and carry implications for other dermatological diseases that could potentially benefit from metformin treatment.

Transcriptomic Trends Associated With Metformin Treatment

The observed downregulation of fibrotic pathways was reflected in decreased expression of key genes previously implicated in CCCA, including MMP7 and COL6A1. MMP7 is a small protease involved in extracellular matrix degradation and a prospective marker of systemic fibrotic diseases. Elevated plasma MMP7 levels correlate with declines in lung function in idiopathic pulmonary fibrosis and with liver stiffness in metabolic dysfunction–associated steatotic liver disease.^28,29 Moreover, previous microarray analyses identified aberrant MMP7 upregulation, among other MMPs, in CCCA-affected scalp tissue compared with nonlesional tissue, supporting its relevance as a potential biomarker of CCCA activity.^4,30

COL6A1, an integral component of type VI collagen, serves as an abundant extracellular matrix component with pronounced deposition in hair follicles. In vitro studies have elucidated that cellular knockout of collagen VI results in diminished cellular focal adhesion, heightened activation of the phosphoinositide 3-kinase/protein kinase B signaling pathway, and, ultimately, fibroblast apoptosis.³¹ Genetic variants in collagen VI manifest in diverse keratin-related disorders, including abnormal scarring, follicular hyperkeratosis, and, more recently, an alopecia scalp phenotype.^32,33,34 Murine studies have demonstrated that lack of COL6A1 promotes increased wound-induced hair regrowth.³⁵

Lastly, DCD, an antibiotic peptide secreted by sweat glands,³⁶ was also downregulated in our cohort. Notably, in mouse keratinocytes, DCD-derived peptides induce mast-cell activation,³⁷ a process that has been histologically and translationally characterized in all forms of cicatricial alopecia.^19,38,39

In contrast, we also observed prominent upregulation of multiple KRTAPs, which play a key structural role in conferring rigidity to the hair shaft. Prior studies have substantiated the relevance of KRTAP dysregulation in diverse alopecic conditions. Downregulation of KRTAPs has been observed in the hair cortex of bald harlequin mice.⁴⁰ Furthermore, in transcriptomic studies, KRTAPs were markedly downregulated in patients with androgenetic alopecia,⁴¹ as well as in those with alopecia areata who carry the CCHCR risk allele.⁴² Consistent with our findings, KRTAP downregulation was previously identified in active CCCA scalp tissue.⁴³

In prior studies comparing lesional with nonlesional tissue in CCCA scalps, fibrotic processes were enriched among upregulated genes, whereas hair-related processes were enriched among downregulated genes in lesional scalps (Table 2).^3,9 Notably, our findings reveal that oral metformin appears to reverse these aberrant baseline expression pathways typically observed in untreated lesions, with metformin treatment being associated with downregulated fibrotic processes and upregulated hair growth and cycling pathways. The magnitude of differential gene expression between nonlesional and lesional scalps (approximately 800 genes with nominal P < .05) is comparable with that between baseline lesional and metformin-treated lesional scalps (approximately 530 genes with nominal P < .05),³ suggesting metformin elicits transcriptomic changes of similar scale to the inherent biological distinction between nonlesional and lesional states.

Clinical Considerations and Dosing Strategies

Metformin should be considered in patients who have not responded to standard therapies or are not candidates for adjunct therapies, such as oral minoxidil. Often, younger women will not display signs of concomitant androgenetic alopecia, rendering minoxidil less effective. While results in this case series are preliminary, metformin is already frequently used in those with prediabetes or harbingers of insulin resistance, such as polycystic ovarian syndrome or acanthosis nigricans, and should be similarly considered for patients with CCCA experiencing the aforementioned comorbidities.⁴⁴ However, this case series suggests possibly even patients with CCCA without prediabetes can benefit from treatment. Given its association with insulin resistance, we recommend routine testing of patients with CCCA for insulin resistance via hemoglobin A_1c levels (normal, <5.7%) or homeostatic model assessment of insulin resistance (HOMA-IR; normal, <2).²

In this study, patients benefited from low-dose metformin, 500 mg, once daily, although 1 patient received immediate-release tablets of metformin, 500 mg, twice daily. Extended-release formulation is thought to lead to fewer gastrointestinal adverse effects than immediate-release tablets and, combined with its less frequent dosing, may improve compliance.⁴⁴ Potential adverse effects of metformin include gastrointestinal upset, hypoglycemia lactic acidosis in patients with kidney impairment, or congestive heart failure. Weight loss and enhanced fertility are also reported as unintended effects but, in our experience, patients have either a neutral or positive reaction to these effects and, in some cases, are more eager to try the medication as a result. Overall, the medication was well-tolerated in our cohort.

Limitations

This study has limitations, including the small sample size, retrospective design, lack of placebo control group, and single-center setting, which limit the generalizability of our findings. Additionally, while we implemented a graded systematic assessment of clinical symptoms, no validated activity or severity scale currently exists for CCCA. Our transcriptomic analysis is constrained by single posttreatment sampling without parallel sampling of nonlesional tissue for comparison. Longer-term repeated biopsies at multiple time points and at various dosing strategies, along with biopsies from healthy scalp regions, may clarify evolving transcriptional trends across treatment duration and elucidate metformin’s effects relative to nonlesional baseline states. Larger prospective, multicenter controlled trials are needed to validate these results and further delineate metformin’s mechanism of action and clinical applications in cicatricial alopecias and other dermatologic disorders.

Conclusions

In this retrospective case series of patients with treatment-refractory CCCA, adjuvant low-dose oral metformin was associated with symptomatic improvement, clinical evidence of hair regrowth, and modulation of gene expression profiles. Transcriptomic analyses revealed metformin upregulated hair growth pathways while downregulating fibrotic and T_H17 inflammatory pathways at a low daily dose of 500 mg. The appearance of hair regrowth seen in some patients challenges the notion that cicatricial alopecias inexorably lead to permanent hair loss. Metformin’s ability to concomitantly target fibrosis and inflammation provides a plausible mechanism for its therapeutic effects in CCCA and other fibrosing alopecia disorders. However, larger prospective, placebo-controlled randomized clinical trials are needed to rigorously evaluate metformin’s efficacy and optimal dosing for treatment of cicatricial alopecias.

Supplement 1.

eTable 1. RNA Sequencing Quality Control Measures

eTable 2. Differentially Expressed Genes Following Metformin Treatment

Supplement 2.

Data Sharing Statement