1. ¹Aderans Research Institute, Philadelphia, Pennsylvania, USA
2. ²Department of Dermatology, University of Pennsylvania Medical School, Philadelphia, Pennsylvania, USA
3. ³Department of Dermatology, University of California, San Francisco, San Francisco, California, USA

Correspondence: Kurt S. Stenn, Aderans Research Institute, 3401 Market Street, Philadelphia, Pennsylvania 19104–3318, USA. Email: kstenn@aderansresearch.com

On 17 October 2005, a research colloquium^* was held on the National Institutes of Health campus to review the primary cicatricial alopecias (CICALs) — their etiology, their pathogenesis, and directions for future research. In this rare, but important, group of inflammatory hair disorders, the hair follicle and its sebaceous gland are replaced by fibrous tissue, leading to permanent destruction of the pilosebaceous unit. The cause and pathogenesis of these diseases are unknown. Although treatments may arrest signs and symptoms of the CICALs, they do not usually influence the underlying disease process and may not arrest progression of the disease.

Participants came from around the world. The meeting was organized into major topics: Clinical and Histological Presentation of CICAL in Humans and Animal Models; Pilosebaceous Apparatus Growth and Structure; Mechanisms; and, finally, Hypotheses and Research Directions.

S. Katz, director of the National Institute of Arthritis and Musculoskeletal and Skin Diseases (Bethesda, MD), opened the meeting with comments underscoring the paucity of epidemiological, clinical, and laboratory studies of these follicular diseases and the therapeutic challenge they present. P. Mirmirani (Case Western Reserve University, Cleveland, OH) the meeting organizer, put a challenge to the attendees to consider, in light of the colloquium's presentations and discussions, (1) what mechanistic directions might be indicated at this time, and (2) what steps should be taken to improve our understanding and therapy.

The clinical and histological characteristics of the CICALs were reviewed by V. Price (University of California, San Francisco). The CICALs have been divided into lymphocytic and neutrophilic subtypes based on pathological changes. Price emphasized that the nosology is difficult because the histological findings do not distinguish the various clinical forms beyond separating the predominantly lymphocytic group from the predominantly neutrophilic group. As a rough guide to therapy, the former is treated with immunomodulating agents and the latter with antimicrobials; however, even the response to treatment is not always predicted by the histology. The concept that the CICALs may be a final common pathway for a number of different processes was raised. Not only is it unclear whether the follicle injury is due to intrinsic or extrinsic factors, but it is also unclear whether the fibrosing events are primary or secondary to inflammation. Knowledge of these rare disorders is further confounded by geographic and racial factors that influence epidemiology. The importance of estimating the true incidence of CICAL was noted. Barriers to achieving this include a lack of the expertise necessary to make an accurate diagnosis and the poor clinicopathological correlation. Price stressed the need to distinguish clinical variants with specific chemokines, and molecular markers gleaned from microarray-based approaches.

J.T. Headington (University of Michigan, Ann Arbor, retired) presented an overview of mechanistic, nosological, and clinical aspects of these diseases. He indicated that pathogenic mechanisms in CICALs involve a common pathway that leads to irreversible damage of both the hair matrix and the follicular stem cells, resulting in generic "cirrhosis" of the scalp. He questioned how and why lymphocytes or neutrophils effect destruction of follicular units and whether these infiltrates are just late, secondary responses to some other event.

A comprehensive review of animal models — spontaneous and induced — that manifest CICAL was presented by J. Sundberg (The Jackson Laboratory, Bar Harbor, ME). He pointed out the numerous examples of correlation between mouse and human hair disorders. Two mouse models with mutations of genes expressed in the sebaceous gland, the stearyl-coenzyme A desaturase gene and gasdermin 3 (a transcription factor), exhibit histological changes mirroring those in human lymphocytic CICALs. These studies implicate the sebaceous gland as the first injury site in the pathogenesis of mouse CICAL. CICAL in C57BL/6 substrains shows progressive hair loss over the dorsum with changes reminiscent of central centrifugal CICAL. This appears to be a complex polygenic disorder influenced by diet and environment. An unresolved issue is how to manipulate mouse skin to mimic potential extrinsic factors that may be related to development of CICAL (for example, heat and chemical injury).

The role of cutaneous stem cells in the evolution of CICALs was presented by G. Cotsarelis (University of Pennsylvania, Philadelphia). Reviewing over a decade of work on the follicular stem cell, Cotsarelis pointed out that epidermal stem cells and hair follicle stem cells are distinct populations. Recent studies using lineage analysis have demonstrated that follicular stem cells contribute to epidermal growth only after injury. The role of hair follicle stem cells in human CICALs remains to be determined, but the fact that inflammation in these disorders generally involves the bulge area and, in some cases, such as graft-versus-host disease, specifically targets epithelial stem cells raises the possibility that bulge-cell death may be one pathway leading to permanent loss of the follicle. Experimental evidence supporting this concept was presented. When follicular stem cells are destroyed by targeting with a thymidine kinase suicide gene under the control of the keratin 15 promoter, a complete loss of the follicle and sebaceous gland ensues within a relatively short period of time (1 week). The epidermis persists after destruction of the bulge cells. When the epidermal stem cell population is altered in another mouse model (the Rac-1 knockout mouse; Science 309:933–5, 2005), total-organ growth reduction (follicle loss and epidermal atrophy) is found. That epidermal atrophy is not seen in CICAL suggests that hair follicle stem cells, but not epidermal stem cells, are injured in these disorders. A major question remaining is whether the stem-cell injury is a primary specific event or a bystander effect of destruction or injury of other portions of the follicle.

J. Schweizer (German Cancer Research Center, Heidelberg) presented a review of the keratins expressed in the hair follicle. Specific hair keratins are expressed in each of the cell lineages of the follicle, including the three layers of the hair shaft (cuticle, cortex, and medulla) and the four layers of the inner root sheath (companion layer, Henle's layer, Huxley's layer, and cuticle). These precise hair follicle keratin expression patterns imply defined control mechanisms affecting follicle structure, and even follicle hormone sensitivity, as the Ha7 promoter contains androgen-responsive elements and is expressed in the medulla of the hair shaft. Schweizer thought it unlikely that CICALs are due to a problem with keratin structure per se, but polymorphisms in specific keratins or differences in their expression might theoretically compromise structural integrity of the follicle and predispose one to the breakdown of the follicle with rupture of epithelial components into the dermis, causing inflammation and foreign body reaction.

C. Zouboulis (Dessau Medical Center, Dessau/Freie Universität, Berlin) reviewed the physiology and morphogenesis of the sebaceous gland and its functions. He pointed out that sebaceous glands may communicate with the hair follicle and affect function, as suggested by the fact that (1) active androgens produced by the sebaceous gland influence dermal papilla cell proliferation and (2) sebaceous-gland cell apoptosis, and thus destruction, as seen in chemotherapy-induced hair loss (e.g. with doxorubicin) precedes the hair loss. Recent work indicates that the sebaceous gland can synthesize and release corticotropin-releasing hormone; moreover, it makes large amounts of free fatty acids, is steroidogenic, and plays a role in skin hydration, antimicrobial activity, and surface antioxidant properties. Whether microorganisms or alterations of the sebaceous gland-regulated innate immunity of the pilosebaceous unit trigger or contribute to CICAL remains a question.

As the immune privilege of the hair follicle bulb is compromised in alopecia areata, the possibility arises that such a disruption could also occur in the CICALs. B. Nickoloff (Loyola University, Maywood, IL) reviewed the elements of the follicular immune system as well as the basis for immune privilege in general. Nickoloff suggested that generating a lichen planopilaris mouse model could be very useful, because such a model would more accurately approximate the human condition. He also questioned whether the loss of immune privilege is a primary or a secondary event in the pathogenesis of the CICALs.

J. James (Oklahoma Medical Research Foundation, Oklahoma City) sleuthed out the possibility of an infectious etiology or an immune process secondary to exposure to an infectious organism (molecular mimicry). Using lupus erythematosus as her experimental model, James showed that a subclass of lupus erythematosus autoantibodies react with epitopes from the Epstein-Barr virus. In presenting these data, James suggested very specific and practical approaches to testing these pathways in the pathogenesis of CICAL by using large-scale humoral immune autoantigen testing, high-throughput lymphocyte antigen recognition screening, and scanning for environmental exposures. She also suggested a global approach that would test genetic linkage with any of the clinically specific CICAL phenotypes by using single-nucleotide polymorphism genome association studies. She emphasized that the immune dysregulation found in CICAL needs to be defined.

The role of the macrophage in CICAL was presented by P. Mirmirani (Case Western Reserve University, Cleveland, OH). In studying a group of patients with lichen planopilaris, Mirmirani and her group, using the CD68 antibody, found that there were a large number of macrophages around the hair follicle and in the perifollicular sheath in patients with CICAL. To describe the changes molecularly, the group performed gene array analyses and found modules of expressed genes that are characteristic of an immediate macrophage immune response. Eighty percent of the identified genes were modulated or dependent on p53, implicating this pathway in the pathogenesis of the CICALs. This research raises the possibility that other aspects of the immune system partake in the pathogenesis of CICALs.

K. Stenn (Aderans Research Institue, Philadelphia, PA) presented the sebotrophic hypothesis as a pathomechanism of the CICALs. The notion is that the CICALs result from obstruction of hair follicle outflow; in this process the sebaceous gland plays a central role, but the problem could be proximal or distal to this gland. Etiologies include toxin, immune injury, infection, and so on. The CICALs might fall into a group of obstructive folliculopathies that would include acne vulgaris (sebaceous follicle) and the overlap syndrome follicular occlusion triad, in which sebaceous and terminal follicles are diseased. The hypothesis implicates a need to understand the mechanism of follicle traffic outflow, hair shaft egress, and the slippage plane.

At the end of the meeting, session chairpersons summarized ideas developed during the meeting in lecture or discussion. L. Sperling (Uniformed Services University of the Health Sciences, Bethesda, MD) emphasized that though the differential histology is difficult, it still merits incorporation in the nosology. He suggested that the nature of the scarring response could be mechanistically important and emphasized the role of genetic and environmental factors in the etiology. M. Philpott (St Bartholomews and Royal London School for Medicine and Dentistry) stressed the importance of considering the sebaceous gland and follicle as a single unit, and the potential role of the defensins in CICAL pathogenesis. At the same time, understanding differences between cell types within follicle compartments is critical to following the dysregulation of the follicle in CICAL. Philpott suggested that a better definition of the interaction between the follicle stem cells and the connective tissue sheath, and its niche, is needed. K. McElwee (University of British Columbia, Vancouver, Canada) questioned which comes first in the CICALs: an immune response or the injury. He expressed the need for better animal models. He suggested that tissues besides the involved scalp skin, such as the draining lymph nodes, should be studied in the human disease. M. Rosenbaum (University of Wisconsin, Madison) commented that expression of the immune-suppressing molecule CD200 may also play a role in CICAL. J. Shapiro (University of British Columbia, Vancouver, Canada) raised the issues of the origin of the cells that overproduce collagen and are involved in fibrosis, the role of matrix metalloproteinases and chemokines, and the different susceptibilities to apoptosis of different cell types within the follicle.

At one point in the discussion, reference was made to the work of M. Hardy (University of Guelph, Canada), who described the CICAL phenotype of the asebia mouse — a condition we now know is caused by the mutation of a single sebaceous gland-expressing gene. She concluded that that condition was due to a dermal-cell or a systemic effect. In retrospect, Hardy's approach and conclusion should be a lesson and a caveat for us. Early in the colloquium the parable of the elephant and the six blind men was invoked to describe our current state of confusion regarding where we should start in defining the CICAL disorders. Although this colloquium did not remove our blindfolds, it did help define directions that need exploration, as well as some tools and approaches that might be productive. If the spirit expressed was any indicator — the camaraderie, collegial interaction, constructive scientific interchange, and research anticipation seen within and outside the hall — then the colloquium was an unusual success. Building on this comity of purpose, the scientists who attended cannot but quickly outperform the blind men of the parable and, in due course, bring this group of formidable disorders into focus.