Incysting on understanding pilar cysts

By Warren R. Heymann, MD
July 15, 2020
Vol. 2, No. 28

Chances are you will see patients with at least one pilar cyst today. You will reassure them of their benignity, stating that the risk of their becoming cancerous is infinitesimal, and perhaps you will excise some lesions. Inquisitive patients will ask you how they got them, and you will likely respond that we really do not know, but pilar cysts may be inherited in many families.

Paraphrasing the National Enquirer — inquiring minds need to know. That time has arrived.

What we already know — Pilar (PC, trichilemmal cysts) arise from the outer root sheath of the follicle at the isthmus. PCs frequently appear sporadically, but may be inherited as an autosomal dominant trait. Patients with familial PCs are usually younger, often presenting with multiple lesions. They are most commonly distributed on the scalp and grow slowly. PCs are usually asymptomatic, although they may calcify, become inflamed, and rupture. (1) Proliferating trichilemmal tumors (PTT) may develop from PCs. PTTs are characteristically seen in elderly women on the head and neck. Although most PTTs are benign, potential for malignant transformation exists, classifying those lesions as a malignant proliferating trichilemmal tumor. (2)

A step forward in understanding the pathogenesis of familial PCs occurred in 2005 when Eiberg et al performed a genome wide scan on a Danish family of 38 people, 11 of whom had PCs. Linkage analysis with 580 DNA micro-satellite markers identified a locus for a gene, which they termed TRICY1 (for trichilemmal cysts), located on chromosome band 3p24‐p21.2. (3)

Dr. Alfred Knudson, for whom the “two-hit hypothesis” explaining tumor suppression genes in familial tumor-predisposing syndromes is named, passed away in 2016. (4) Originally described for patients with retinoblastomas, the hypothesis states that two “hits” to DNA are necessary to cause the cancer [or tumor], the first being inherited and the second acquired. This hypothesis led to conceptualization of tumor suppressor genes, loss‐of‐heterozygosity (LOH) as being relevant to carcinogenesis, and the possibility that “delayed mutation” may correspond to germline‐mosaic mutations. (4) Could the two-hit hypothesis be relevant in the pathogenesis of PCs?

What’s new — Hörer et al described the pathogenetic mechanism for the development of hereditary PCs. By whole-exome sequencing of DNA from the blood samples of 5 affected individuals and subsequent Sanger sequencing of a family cohort including 35 affected individuals from 12 Tunisian families and 14 unrelated affected patients of North African and Caucasian ethnicity, the authors identified a combination of the Phospholipase C Delta 1 germline variants as a high-risk factor for PC development. Allele-specific PCRs and cloning experiments demonstrated that these two variants are present on the same allele. The analysis of tissue from several cysts revealed that an additional somatic Phospholipase C Delta 1 mutation on the same allele is required for cyst formation. This study showed that the protein function of the cyst-specific 1-phosphatidylinositol 4, 5-bisphosphate phosphodiesterase delta-1 protein variant is modified. Phospholipase C delta 1 (PLCD1) lies within the TRICY1 locus and encodes 1-phosphatidylinositol 4, 5-bisphosphate phosphodiesterase delta-1 (PLCδ1), a member of the phospholipase C family. PLCδ1 plays an important role in intracellular calcium signaling via the hydrolysis of phosphatidylinositol 4, 5-bisphosphate (PIP2) into the second messengers diacylglycerol (DAG) and inositol 1,4,5-trisphosphate [IP3]. This pathologic mechanism defines a monoallelic model of the two-hit mechanism proposed for tumor development and other hereditary cyst diseases. (5) In an accompanying editorial, Shimomura et al note that DAG further mediates the activation of protein kinase C (PKC), and IP3 releases calcium from intracellular stores playing a role in a variety of physiological functions. In addition to the hair follicle, PLCδ1 is also expressed in the nail matrix. (6)

The intrigue continues. Fu et al, by using whole-exome sequencing, identified a heterozygous variant within the BPIFC gene that cosegregated with the phenotypes of a Chinese Han PC family (3 generations with 12 affected individuals) with an autosomal‐dominant inheritance. The BPIFC gene is located on chromosome 22q12.3. The gene encodes bactericidal/permeability‐increasing protein fold-containing family member C protein (BPFIC) belonging to the BPI fold‐containing (BPIF) superfamily. The BPIF family mainly involves in the innate immune system and lipoprotein metabolism. This implies that the BPIFC protein plays a role in inflammation involved in the formation of PCs. This hypothesis mandates validation by further studies. (7)

And there’s more! I have rarely seen patients with total leukonychia (who have mentioned that this was a familial trait), but I was unaware of Bauer syndrome, an autosomal dominant disorder characterized by leukonychia, koilonychia, and PCs (8, 9) Although PLCD1 variants have been previously associated with the inherited leukonychia totalis, the novel association of PLCD1 variants with cyst formation reveal diverse functional roles of PLCδ1 in skin homeostasis. (6)

When my patients now ask me why they get PCs, I will not go through the detail you just read (and kudos to those of you who have done so!), but still explain that this may be a familial trait. I will take solace in knowing that the molecular mechanisms, while not completely understood, are on the verge of being deciphered. That is the first step on the path of ultimate medical treatments for surgical disorders.

Point to Remember: The two-hit hypothesis appears to be at play for familial pilar (trichilemmal) cysts related to the PLCD1 gene. In other families, the BPFIC gene affecting innate immunity could be pathogenic. Developing an understanding of the molecular basis of pilar cysts may ultimately lead to a non-surgical approach to these common, but frequently bothersome lesions.

Our expert's viewpoint

Rhonda E. Schnur, MD
Professor of Pediatrics
Head, Division of Genetics
Cooper Medical School of Rowan University

Tumor biology is well recognized to be associated with acquired “second hits” that occur in trans with a predisposing constitutional variant, with the second hit typically occurring on the normal allele on the other paired copy of the chromosome. Such second hits often result in complete loss of function of a critical protein, typically a protein involved in regulating cell growth and division, allowing for tumor progression. Pilar cysts have now been linked in some patents to acquired (somatic) pathogenic variants in PLCD1 (reference 5). However, in contrast to traditional molecular tumor biology, the somatic changes in pilar cysts were noted to occur in cis (on the same chromosome copy) with germline alleles that confer a predisposing risk of tumors. The study could not completely exclude the possibility of loss of heterozygosity (LOH) of the wild type allele due to the mix of normal and cystic cells in analyzed tissue, but all of the detected somatic PLCD1 variants in appeared to be heterozygous, suggesting that LOH was unlikely.

This report and the commentary by Shiomura et al (reference 6) also reviewed a similar mechanism of acquired somatic variants in JAK2-related myeloproliferative tumors that occur in cis with a predisposing germline sequence variant. Pilar cysts are typically benign, but the implications for tumor biology of acquired pathogenic variants occurring in cis with a predisposing constitutional variant may be huge. We should also bear in mind that acquired variants occurring in cis with an underlying germline variant may be protective against disease. For example, constitutional heterozygous gain of function variants, usually de novo, in SAMD9, a growth suppression gene, are associated with the severe and often fatal MIRAGE syndrome (Myelodysplasia, Infection, Restriction of growth, Adrenal hypoplasia, Genital phenotypes, Enteropathy). Recently Roucher-Boulez F et al (Front Endocrinol [Lausanne] 2019 Sept 11; 10: 625) reported two different, independently acquired “protective reversion variants,” nonsense variants, in a mother and child who both carried a known pathogenic germline variant in SAMD9. The acquired loss of function variants both occurred in cis with the pathogenic germline variant. The mother was clinically unaffected; the child had neonatal symptoms of MIRAGE syndrome, but many of the child’s symptoms resolved. The authors proposed that the timing of acquisition of each acquired nonsense variant, expected to result in loss of function of the mutant SAMD9 protein, likely impacted the different phenotypes in parent and child. Presumably, the acquired variant in the unaffected mother arose very early in embryogenesis, protecting her from any clinical symptoms. However, in her child with neonatal MIRAGE syndrome, serial genetic assessment confirmed that the acquired nonsense variant was not present at birth, but was acquired later, consistent with the improvement in symptoms. These observations of acquired second hits occurring on the same allele as a predisposing allele (PLCD1 or JAK2), or with an outright pathogenic germline variant (SAMD9), affirm that genetic analysis may merit repetition in different tissue samples over time, to optimally diagnose disease and/or understand an evolving, or resolving, disease course.

The paucity of published precedent for somatic “second hit” events occurring in cis with germline predisposing/pathogenic alleles is impressive. However, if we search more rigorously, through the direct study of phenotypically abnormal versus germline cells, repeating genomic studies “over time and space” as phenotypic change is observed, we will likely find other similar “second hit” genetic events, in cis as well as in trans, with underlying constitutional pathogenic variants. The big question, of course, is what really induces these second event changes to occur and how we can thwart or optimize the mechanism of these acquired changes to prevent or treat disease.

1. Al Aboud DM, Patel BC. Pilar cyst. StatPearls [Internet]. Treasure Island (FL); StatPearls Publishing 2019 Jan.

2. Demirdag HG, Serel S, Akay BN, Kirmizi A, Okcu Heper A. Malignant proliferating trichilemmal tumour arising on a trichilemmal cyst with dermoscopic and histopathologic findings. J Eur Acad Dermatol Venereol 2019; 33: e322-e324.

3. Eilberg H, Hansen L, Hansen C, Mohr J, Teglbiaerg PS, Kiaer KW. Mapping of hereditary trichilemmal cyst (TRICY1) to chromosome 3p24-p21.2 and exclusion of beta-CATENIN and MLH1. Am J Med Genet A 2005; 15: 133A: 44-47.

4. Hino O, Kobayashi T. Mourning Dr. Alfred G. Knudson: The two-hit hypothesis, tumor suppression genes, and the tuberous sclerosis complex. Cancer Sci 2017; 108: 5-11.

5. Hörer S, Marrakchi S, Radner APW, Zolles G, et al. A monallelic two-hit mechanism in PLCD1 explains the genetic pathogenesis of hereditary trichilemmal cyst formation. J Invest Dermatol 2019; 139: 2154-2163.

6. Shimomura Y, O’Shaughnessy R, Rajan N. PLCD1 and pilar cysts. J Invest Dermatol 2019; 139: 2075-2077.

7. Fu XG, Huang Z, Zhou SJ, Yang J, et al. Novel heterozygous BPIFC variant in a Chinese pedigree with hereditary trichilemmal cysts. Mol Genet Genomic Med 2019; 7: e697.

8. Bushkell LL, Gorlin RJ. Leukonychia totalis, multiple sebaceous cysts, and renal calculi. A syndrome. Arch Dermatol 1975; 111: 899-901.

9. Mutob M, Niiyama S, Nishikawa S, Oharaseki T, Mukai H. A syndrome of leukonychia, koilonychia and multiple pilar cysts. Acta Derm Venereol 2015; 95: 249-250.

---