Metabolic Associations of Central Centrifugal Cicatricial Alopecia: A Case-Control Study Open Access

Central centrifugal cicatricial alopecia (CCCA) is a primary scarring alopecia that predominantly affects black women with unclear etiology [1]. There is some evidence to suggest an association with metabolic dysregulation. For example, proteomic analyses have identified decreased levels of proteins involved in metabolic pathways within CCCA-affected scalp tissue [2], and clinical studies have noted a higher prevalence of insulin resistance among CCCA patients [3]. Therefore, we explored for potential associations between CCCA and metabolic dysfunction utilizing a large, multicenter database.

On April 07, 2025, TriNetX Research Network was searched for black women with CCCA (International Classification of Diseases [ICD]-10: L66.81; excluding L66.0–L66.4) and other primary scarring alopecias (controls) (ICD-10: L66.0–L66.4; excluding L66.8–L66.9) from 2004 to 2024. These criteria ensured that no patients were included in both cohorts. We selected other primary scarring alopecias as the comparator group to account for shared features of inflammation and follicular destruction that are common across scarring alopecia subtypes. This approach allows for a more biologically relevant comparison than using nonscarring alopecias or patients without alopecia, who may differ in inflammatory pathways, psychosocial burden, or treatment exposures. Cohorts were propensity score matched by current age, age at diagnosis, and ethnicity. All eligible patients meeting inclusion criteria were included. We analyzed incidence of metabolic disorders and laboratory results 1 day to 10 years after alopecia diagnosis. Benjamini-Hochberg was applied to correct for multiple hypothesis testing with significance at adjusted p value = 0.0042.

We included 2,482 CCCA patients and 2,482 matched controls (Table 1). Demographics were similar between cohorts (standardized mean differences <0.08). Dissecting cellulitis (56.4%) and lichen planopilaris/frontal fibrosing alopecia (30.1%) were the most common control diagnoses.

CCCA patients versus controls had similar incidence of type 2 diabetes mellitus (T2DM; 19.7% vs. 19.5%, respectively, p = 0.591), prediabetes (17.8% vs. 21.0%, respectively, p = 0.265), hyperlipidemia (32.9% vs. 35.5%, respectively, p = 0.195), obesity (30.9% vs. 38.5%, respectively, p = 0.054), and metabolic syndrome (1.4% vs. 1.2%, respectively, p = 0.404) diagnoses (Table 2). CCCA patients versus controls had lower incidence of essential hypertension (39.7% vs. 49.5%, respectively, p < 0.0001).

CCCA patients versus controls had similar body mass index (BMI; 33.3 vs. 32.8 kg/m², respectively, p = 0.069), hemoglobin A1c (6.11% vs. 6.20%, respectively, p = 0.088), glucose (106.1 vs. 107.6 mg/dL, respectively, p = 0.347), and low-density lipoprotein (106.4 vs. 106.0 mg/dL, respectively, p = 0.803) values. CCCA patients versus controls had higher high-density lipoprotein (HDL; 58.0 vs. 55.5 mg/dL, respectively, p = 0.0016) and lower triglycerides (96.4 vs. 106.3 mg/dL, respectively, p = 0.0001) values.

We found no significant differences between cohorts in the incidences of T2DM, prediabetes, obesity, or metabolic syndrome, nor in BMI, hemoglobin A1c, or glucose levels. Similarly, a retrospective cohort study of 153 CCCA patients (% biopsy-proven unknown) and 153 nonscarring alopecia controls found no differences in T2DM (50.3% vs. 47.1%, respectively, p = 0.647) or obesity (28.8% vs. 28.8%, respectively, p = 1.000) prevalence [4]. Likewise, another retrospective study of 53 CCCA patients (79% biopsy-proven) and 212 age-matched controls without CCCA reported no differences in T2DM (25% vs. 20%, respectively, p = 0.4983) or obesity prevalence (64% vs. 57%, respectively, p = 0.3192) [5]. In contrast, a retrospective study of 201 biopsy-proven CCCA patients and 39,280 controls without CCCA found increased T2DM risk (OR = 4.13 [95% CI = 2.76–61.8]) after adjusting for obesity [3]. Differing findings between studies may stem from methodological differences. Notably, our study represents the first comparison of CCCA patients to those with other primary scarring alopecias, thereby controlling for potentially shared mechanisms of scarring hair loss.

We also found decreased incidence of hypertension in CCCA patients versus controls, which differs from previous studies. For example, a retrospective study of 153 CCCA patients (% biopsy-proven unknown) and 153 nonscarring alopecia controls reported no difference in hypertension prevalence (84.3% vs. 82.4%, respectively, p = 0.759) [4], and a retrospective study of 53 CCCA patients (79% biopsy-proven) and 212 age-matched controls without CCCA found no differences in hypertension prevalence (70% vs. 58%, respectively, p = 0.1164) [5]. In contrast, a retrospective study of 22 CCCA patients (% biopsy-proven unknown) and 19 age-, race-, gender-, and BMI-matched controls without hair loss reported higher systolic (134 vs. 118 mm Hg, respectively, p = 0.002) versus and diastolic pressures (89 vs. 73 mm Hg, respectively, p = 0.092) [6], but small sample size limits power. Collectively, these studies show inconsistent associations between CCCA and hypertension. Large, prospective studies of biopsy-proven CCCA patients are needed to clarify whether a true relationship exists.

While hyperlipidemia incidence did not differ significantly between CCCA patients and controls, our CCCA cohort paradoxically exhibited higher HDL and lower triglycerides. In contrast, a retrospective cohort study of 153 CCCA patients (% biopsy-proven unknown) and 153 nonscarring alopecia controls found no difference in dyslipidemia prevalence (48.4% vs. 46.4%, respectively, p = 0.819) [4]. Furthermore, a retrospective cohort study of 427 biopsy-proven CCCA patients and 1,281 age- and sex-matched controls without alopecia reported no difference in the incidence of atherosclerosis (20.8% vs. 19.7%, p = 0.55), for which hyperlipidemia is risk factor [7]. In contrast, a retrospective study of 53 CCCA patients (79% biopsy-proven) and 212 age-matched controls without CCCA found increased prevalence of hyperlipidemia (66% vs. 17%, respectively, p < 0.0001) [5]. The relationship between CCCA and lipid abnormalities remains unclear, complicated by paradoxical findings of favorable lipid profiles (higher HDL, lower triglycerides) in our CCCA cohort. Prospective studies utilizing standardized lipid assessments to evaluate contribution of lipid metabolism in CCCA pathogenesis are needed.

Limitations include retrospective database design and potential miscoding. We used mutually exclusive ICD-10 codes to define the CCCA and comparator groups and avoided overlap by excluding L66.81 from the comparator group and L66.0–L66.4 from the CCCA group. Of note, the CCCA-specific code (L66.81) was introduced in 2021. Therefore, some earlier CCCA cases may have been coded under other scarring alopecia diagnoses, which could result in misclassification. Additionally, lack of histopathological confirmation and unmeasured confounders (e.g., socioeconomic status, lifestyle) may persist despite demographic matching. While using other scarring alopecia patients as controls offers biological relevance, their metabolic profiles are not well characterized, potentially affecting results.

In summary, we found no significant association between CCCA and most metabolic disorders, including T2DM, obesity, and metabolic syndrome, which aligns with findings from several prior studies. Since some studies show discrepant results, a metabolic relationship may be more nuanced than currently understood. The available evidence suggests that while systemic metabolic dysfunction likely does not drive CCCA pathogenesis, it may influence disease susceptibility or progression. Therefore, dermatologists could consider metabolic screening based for CCCA based on individual risk factors (e.g., central adiposity or family history of metabolic disease). Directions for future studies include investigating whether metabolic disorders interact with genetic, inflammatory, or autoimmune pathways in CCCA.

This retrospective review of patient data is exempt from informed consent and did not require ethical approval in accordance with local/national guidelines. The data reviewed are a secondary analysis of existing data, do not involve intervention or interaction with human subjects, and are de-identified per the de-identification standard defined in Section 164.514(a) of the HIPAA Privacy Rule. The process by which the data are de-identified is attested to through a formal determination by a qualified expert as defined in Section 164.514(b)(1) of the HIPAA Privacy Rule. This formal determination by a qualified expert refreshed on December 2020. Written informed consent from the participants was not required for this retrospective study in accordance with local/national guidelines.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Conceptualization: M.M.O. and S.R.L.; methodology, M.M.O., M.H.Z., A.S., and S.R.L.; data curation and formal analysis: M.H.Z.; investigation: M.M.O., M.H.Z., and A.S.; supervision: S.R.L.; writing – original draft: M.M.O.; writing – review and editing: M.H.Z., A.S., and S.R.L.

The data that support the findings of this study are not publicly available due to the inclusion of protected health information and restrictions imposed by the data provider, TriNetX, which prohibit public sharing of raw data. However, aggregate-level data or analyses supporting this study’s findings are available from the corresponding author S.R.L. upon reasonable request and with permission from TriNetX.