High incidence of demodicidosis in eyelid basal cell carcinomas

Oxford, IJD International 0011-9059 Blackwell 45 UK Publishing Journal Ltd. Ltd, of Dermatology 2003 Dermatologic surgery Demodicidosis Erbagci, Dermatologic Erbagci, surgery in and eyelid Erkiliç basal cell carcinomas High incidence of demodicidosis in eyelid basal cell carcinomas Zülal Erbagci, MD, Ibrahim Erbagci, MD, and Suna Erkiliç, MD From the Departments of Dermatology, Ophthalmology, and Pathology, Gaziantep University Medical Faculty Gaziantep, Turkey Correspondence Zülal Erbagci, MD GazimuhtarpaÍa Bulvarı Geçit 1. no. 1/5 27090 Gaziantep Turkey E-mail: zerbagci@yahoo.com Abstract Background Although UV radiation is the major cause of basal cell carcinoma (BCC), local factors, such as chronic trauma, irritation, or inflammation, may also have some role in its etiopathogenesis. The pilosebaceous follicle mites, Demodex folliculorum and D. brevis, inhabit most commonly and densely certain facial skin areas, including the nose and periorbital regions, where BCC also develops most frequently. Aim To investigate, in a retrospective histopathologic study, whether a possible etiopathogenetic relationship exists between demodicidosis and eyelid BCCs. Methods We examined 32 eyelid BCC specimens that contained at least five eyelashes or five hair follicles with respect to the presence and density of Demodex mites. As controls, we evaluated 34 matched specimens consisting of benign eyelid skin lesions. Results Twenty-one of 32 BCC cases (65.6%) and eight of 34 control cases (23.33%) had demodicidosis. Mean mite counts were 1.31 ± 1.57 and 0.47 ± 0.99 in BCC cases and controls, respectively. The differences were significant for both prevalence (P < 0.001) and density (P = 0.0052). Although there was a significant positive correlation between increasing mite number and patient age in the control group (r = 0.47, P < 0.05), no significant correlation was found between these two factors in BCC cases (r = 0.102, P > 0.05). Conclusions Demodicidosis may be one of the triggering factors of carcinogenesis in eyelid BCCs in otherwise predisposed people due to its traumatic/irritating effect or chronic inflammation. Introduction Basal cell carcinoma (BCC) is the most common malignant eyelid tumor, which invades most frequently the lower eyelids and medial canthi. 1 Although UV radiation is the major risk factor for the development of BCC, other factors, including genetic predisposition, immunosuppression, carcinogenic stimuli such as chronic trauma, irritation, and inflammation, exposure to inorganic arsenic, X-irradiation, and scars, have also been implicated in its etiopathogenesis. Nearly one-third of all BCCs appear on relatively light-protected skin areas, including the medial canthi and lower eyelids, whereas the tumor arises very rarely on the back of the hand. 1,2 A previous trauma history was found in 7.3% of 1774 BCCs, 3 and some cases of BCC have been reported to develop on the sites of venipuncture or scars resulting from various causes. 4,5 The hair follicle parasites, Demodex folliculorum and D. brevis, are the only metazoan organisms found within the pilosebaceous components of eyelids, including the meibomian glands, eyelashes, and small hair follicles. 6,7 Although usually considered a nonpathogenic parasite, Demodex has been implicated as a causative agent for some dermatologic conditions, such as papulopustular rosacea, rosacea-like eruptions of the face, and pityriasis folliculorum, and ocular diseases, including some types of blepharitis, madarosis (scarcity or loss of the eyelashes), meibomian dysfunction, and chalazion. 8 13 If the mites are found in large numbers or extrafollicularly in the dermis, they usually evoke an obvious inflammatory response. Vollmer 14 has shown that Demodex mites, even if found in small numbers, are commonly associated with histologic folliculitis. Interestingly, these mites most frequently and densely inhabit certain facial areas, including the nose, lower eyelids, forehead, and periauricular region, where BCC also arises most frequently. 15,16 In a recent histopathologic study, we found a significantly higher incidence of demodicidosis and mean mite number (44.68% and 0.872 ± 1.193, respectively) in 94 BCCs located in various facial areas, compared with those (25% and 0.458 ± 1.009, respectively) in benign skin lesions found at similar locations and taken from age- and sex-matched individuals. 17 In this study, another notable finding was that the mite frequency was significantly higher in eyelid BCCs than in those located in other facial regions. Therefore, we decided to investigate further the possible role 567 International Journal of Dermatology 2003, 42, 567 571

568 Dermatologic surgery Demodicidosis in eyelid basal cell carcinomas Erbagci, Erbagci, and Erkiliç of demodicidosis in the etiopathogenesis of eyelid BCCs in a retrospective histopathologic study. Patients and methods A total of 32 BCC specimens, either biopsied or excised from eyelids, containing at least five eyelashes or five pilosebaceous units, were retrieved from the archives of the Pathology Department of Gaziantep University Medical School. Hematoxylin and eosin-stained slides were examined under light microscopy to determine the presence and density of Demodex. As controls, 34 biopsy specimens belonging to location-matched benign eyelid skin lesions, also containing at least five eyelashes or five hair follicles, taken from age- and sex-matched healthy subjects, were examined. Demographic data, including patient age, sex, and any past chronic inflammatory eyelid skin diseases which may be related to Demodex, such as chronic blepharitis, recurrent chalazion, and madarosis, were obtained from patient records. The degree of inflammatory infiltrate surrounding infested pilosebaceous components was evaluated in the study groups. Sparse lymphohistiocytic infiltration was graded as mild, numerous inflammatory cell accumulation as severe and the rest as moderate. The frequencies of mite-positive cases in the tumor subtypes and in the tumors with mild, moderate, or severe peritumoral inflammation were also compared in the BCC group. Chi-squared test (Yates corrected or noncorrected) was used to compare frequencies. Mean values were compared by Student s t-test. Pearson s correlation analysis was used to reveal possible correlations between mite numbers and patient ages in each study group. All statistical tests were two-tailed, and a P value of less than 0.05 was considered to be statistically significant. BCC group (r = 0.102, P > 0.05). Twenty-one cases (65.6%) with BCC (Figs 1 3) and eight control cases (23.53%) were infested with Demodex mites, a significant difference (χ2 = 11.97, P < 0.001). The mean mite number per pilosebaceous unit or eyelash was also significantly higher in BCCs (1.31 ± 1.57) than in controls (0.47 ± 0.99) (P = 0.0052). The results are summarized in Table 1. No significant differences in mite prevalence were found between BCC subtypes or between tumors with mild, moderate, or severe peritumoral infiltration. In the BCC group, a severe lymphohistiocytic inflammation surrounded the infested pilosebaceous units or eyelashes in 15 cases, whereas mild or moderate perifollicular inflammatory infiltrates were observed in the remaining cases (Table 2). In the control group, there was usually a mild mononuclear cellular inflammation around the infested pilosebaceous components. Results Discussion The specimens were collected from 17 men and 15 women (age range, 36 71 years; mean ± SD, 53 ± 10.82 years) with BCC, and 18 men and 16 women (age range, 31 70 years; mean ± SD, 48.4 ± 12.55 years) with benign eyelid skin lesions, including nevi (n = 8), xanthelasmas (n = 4), epidermal inclusion cysts (n = 8), pyogenic granulomas (n = 4), eccrine hidrocystomas (n = 2), trichilemmal cysts (n = 3), and squamous papillomas (n = 5). There was no significant difference between the mean patient ages in the BCC and control cases (P > 0.05). The incidences of possible Demodex-related past eyelid diseases were also equivalent in the study groups [14/32 (43.75%) and 13/34 (38.22%) in BCC and controls, respectively) (P > 0.05). Although there was a significant correlation between Demodex number and increasing patient age in the control group (r = 0.47, P < 0.05), no correlation was found between mite number and increasing age in the BCC can cause considerable morbidity when located on the eyelids and periocular skin.1 It has been suggested that some as yet unidentified mesodermal factor(s) may work as an intrinsic promoter in the pathogenesis of BCC, because it is strongly dependent on its surrounding stroma, metastasizes very rarely, and BCC transplants can survive only if transplantation is performed with the surrounding stroma.2 Although UV radiation is accepted as the main cause of BCC, sun exposure only partly explains the etiology of periorbital BCC. In a recent study, the doses of UV radiation were found to be similar on the upper and lower eyelids, but the number of lower eyelid BCCs was 225, much higher than that (17) on the upper eyelid.18 Although the mechanism is unknown, trauma, chronic irritation, inflammation, and scars due to burns or other causes have been implicated as etiologic factors in some BCCs.1,2,4,5 International Journal of Dermatology 2003, 42, 567 571 Figure 1 Histologic section of eyelid basal cell carcinoma. Note the Demodex mite (black arrowhead) within a small hair duct (hematoxylin and eosin, 40)

Erbagci, Erbagci, and Erkiliç Demodicidosis in eyelid basal cell carcinomas Dermatologic surgery Figure 3 Abundant mites within an eyelash canal in close proximity to an island of basal cell carcinoma (hematoxylin and eosin, 100) Figure 2 Multiple Demodex mites (black arrowhead) within a hair follicle duct located adjacent to basal cell carcinoma lobules (hematoxylin and eosin, 100) Demodex mites ( sebaceous worm ) may represent an important cofactor in some dermatologic and ophthalmologic diseases. D. folliculorum, the long form of the mite, inhabits eyelashes and hair follicle ducts, whereas D. brevis, the short form, usually lives in small hair and eyelash sebaceous glands and in the lobules of the meibomian glands.15,16 They obtain their nutritive requirements from cellular proteins through epithelial and sebaceous gland cell destruction by their own enzymes, including immunoreactive lipase. Therefore, they cause follicular distension, keratinization, hyperplasia, and occasional mild chronic perifolliculitis.6,19,20 Scanning electron microscopic examinations have revealed that D. folliculorum can move freely on the skin surface, implying a vector potential for the mite.16 Some mites have also been shown to contain bacteria inside their gut and on their skin.12 Increased mite numbers, possibly in conjunction with microorganisms, may lead to tissue damage, resulting in eyelid and facial skin diseases.8 The frequency of D. folliculorum was found to be very high in 100 patients with blepharitis compared with normal controls, implying a mediator role for the mite in chronic blepharitis.10,11 Mites have been found on the palpebral margins in 68% of 568 patients with chronic marginal blepharitis, more abundantly in older persons and in persons with diabetes.21 Persons whose lid margins contained Staphylococcus aureus had more mites than those without S. aureus.22 The presence of the mite is usually associated with madarosis of the lower lid, independent of age. Examination of 388 follicles revealed that mites were found in 42% of follicles with inflammation, and 83% of follicles with Demodex were surrounded by a lymphohistiocytic inflammation, suggesting that Demodex is usually associated with histologic folliculitis.14 Recently, Millikan23 has emphasized the role of progressive inflammation associated with Demodex infestation in androgenetic alopecia, and has suggested that chronic inflammation induced by the mite can progress to alopecia, with or without scarring. Hair follicle infestation by Demodex may be associated with an intense perifollicular infiltrate of predominantly CD4 helper/inducer T cells and increased numbers of macrophages and Langerhans cells. Immunohistochemical findings have implied a delayed-type hypersensitivity reaction, possibly triggered by antigens of follicular origin which may be related to the mite.24 A successful host parasite relationship is a balance between the limitation of the parasite by host defenses and the ability of the parasite to modulate, evade, or even restrict the host response.25 Host immune responsiveness may be altered as a result of infestation and the impact of these alterations on vector-borne pathogen transmission is unknown.26 It has been suggested that generalized demodicidosis in dogs is caused by a mite-specific immunosuppression induced by D. canis, especially in cases with T-cell defects.27 Although UV exposure is the predominant cause of squamous cell carcinoma, it is not an obligatory factor in BCC. International Journal of Dermatology 2003, 42, 567 571 569

570 Dermatologic surgery Demodicidosis in eyelid basal cell carcinomas Erbagci, Erbagci, and Erkiliç Table 1 Demographic data, Demodex positivity, and mean mite densities in the eyelid basal cell carcinoma (BCC) cases and controls BCC cases (n = 32) Controls (n = 34) P Mean patient age ± SD (years) 53 ± 10.82 48.4 ± 12.55 > 0.05 Patient sex Men 17 18 > 0.05 Women 15 16 Correlation of mite numbers with No Yes increasing patient age (r = 0.102, P > 0.05) (r = 0.47, P < 0.05) Frequency of possible Demodexrelated past ocular diseases 14 (43.75%) 13 (38.23%) > 0.05 Incidence of demodicidosis 21 (65.6%) 8 (23.33%) < 0.001 Mean mite density ± SD 1.31 ± 1.57 0.47 ± 0.99 0.0052 SD, standard deviation. Mite-positive cases (n = 21) Mite-negative cases (n = 11) BCC subtypes Solid (n = 25) 16 9 Morpheaform (n = 5) 3 2 > 0.05 Cystic (n = 2) 2 Severity of peritumoral inflammation Severe (n = 15) 9 6 Moderate (n = 13) 10 3 Mild (n = 4) 2 2 > 0.05 P Table 2 Patient characteristics in Demodex-positive basal cell carcinomas (BCCs) and Demodex-negative BCCs Environmental and hereditary factors, as well as local conditions, are suspected to play an important role in the development of BCC as it prefers certain locations. Local factors, other than sun exposure, have been proposed to be responsible for the relatively high incidence of eyelid BCC in the lower lid and medial canthi. 18 A well-established relationship exists between BCC and hair follicles, as BCC develops most frequently on the seborrheic and hair-bearing skin areas, especially on the face, whilst palmar skin is extremely rarely involved, except in nevoid BCC syndrome. 2 The origin of BCC has been shown to be a cell with pluripotential capacity which is located particularly in the follicular outer root sheath. 28 In our study, we observed a significantly higher incidence of demodicidosis and increased mite densities in cases with eyelid BCCs compared with healthy controls. As the incidence of possible Demodex-related past eyelid diseases was equivalent in BCC and control cases, increased numbers of mites in BCC did not seem to be relevant to such ocular diseases. Mite positivity was not correlated with the severity of peritumoral inflammation or the type of tumor. Although denser inflammatory infiltrates were observed around the infested eyelashes or pilosebaceous units in BCC cases compared with controls, this phenomenon was likely to be related to the peritumoral host immune response. Our results may indicate only a coincidental association between the high incidence of demodicidosis and eyelid BCC. Alternatively, the increased numbers of Demodex could be secondary to stromal changes, e.g. increased vascularization, produced by BCC rather than vice versa. Nevertheless, we cannot exclude a hypothetical etiopathogenetic role for Demodex infestation. The chronic perifollicular inflammation or presumptive immunosuppression caused by the parasite may favor BCC development in older individuals whose immune responses have already decreased. On the basis of the relevant literature data and our findings, one may further speculate that persistent immunologic stimulation by Demodex-specific antigens, biochemical tissue changes occurring during chronic irritation/traumatization due to the presence of the mite, or chronic inflammatory diseases caused by microorganisms vectored by the parasite may act as mesodermal promoters for BCC development in genetically and environmentally predisposed persons. Further prospective histologic, immunologic, epidemiologic, and animal studies with larger case numbers are needed to clarify the pathogenetic significance of chronic Demodex infestation of the eyelid in BCC pathogenesis. International Journal of Dermatology 2003, 42, 567 571

Erbagci, Erbagci, and Erkiliç Demodicidosis in eyelid basal cell carcinomas Dermatologic surgery 571 References 1 Cook BE Jr, Bartley GB. Epidemiologic characteristics and clinical course of patients with malignant eyelid tumors in an incidence cohort in Olmstead County, Minnesota. Ophthalmology 1999; 106: 746 750. 2 Patterson JW. Basal cell carcinoma. In: Demis DJ, Thiers BH, Burgdorf WHC, et al., eds. Clinical Dermatology, 19th revision. Philadelphia: JB Lippincott, 1992: unit 21-18, 1 26. 3 Noodleman FR, Pollack SV. Trauma as a possible etiologic factor in basal cell carcinoma. J Dermatol Surg Oncol 1986; 12: 841 846. 4 Ozyazgan I, Kontacs O. Basal cell carcinoma arising from surgical scars: a case and review of the literature. Dermatol Surg 1999; 25: 965 968. 5 Wilkin JK, Strobel D. Basal cell carcinoma arising at the site of venipuncture. Cutis 1985; 36: 297 298. 6 Bardach HG, Raff M, Poitschek C. [Nosologic position of demodicidosis in humans.] Hautarzt 1981; 32: 512 518. 7 English FP, Nutting WB. Demodicidosis of ophthalmic concern. Am J Ophthalmol 1981; 91: 362 372. 8 Pena GP, Andrade Filho JS. Is Demodex really nonpathogenic? Rev Inst Med Trop Sao Paulo 2000; 42: 171 173. 9 Erbagci Z, Ozgoztasi O. The significance of Demodex folliculorum density in rosacea. Int J Dermatol 1998; 37: 421 425. 10 Kamoun B, Fourati M, Feki J, et al. [Blepharitis due to Demodex: myth or reality?] J Fr Ophthalmol 1999; 22: 525 527. 11 Demmler M, de Kaspar HM, Mohring C, et al. Demodex folliculorum, associated pathogen spectrum and specific therapy. Ophthalmologie 1997; 94: 191 196. 12 Abd-El-Al AM, Bayoumy AM, Abou Salem EA. A study on Demodex folliculorum in rosacea. J Egypt Soc Parasitol 1997; 27: 183 195. 13 Roihu T, Kariniemi AL. Demodex mites in acne rosacea. J Cutan Pathol 1998; 25: 550 552. 14 Vollmer RT. Demodex-associated folliculitis. Am J Dermatopathol 1996; 18: 589 591. 15 Norn MS. Incidence of Demodex folliculorum on skin of lids and nose. Acta Ophthalmol (Copenh) 1982; 60: 575 583. 16 English FP, Zhang GW, McManus DP, et al. The presence of the parasite Demodex folliculorum on the skin surface of the eyelid. Aust N Z J Ophthalmol 1991; 19: 229 234. 17 Erbagci Z, Erkiliç S. Basal cell carcinoma and demodicidosis. Is there an etiologic or coincidental relationship? Turkish J Cancer 2000; 30: 111 118. 18 Lindgren G, Diffey BL, Larko O. Basal cell carcinoma of the eyelids and solar ultraviolet radiation exposure. Br J Ophthalmol 1998; 82: 1412 1415. 19 Jimenez-Acosta F, Planas L, Penneys N. Demodex mites contain immunoreactive lipase. Arch Dermatol 1989; 125: 1436 1437. 20 Roth AM. Demodex folliculorum in hair follicles of eyelid skin. Ann Ophthalmol 1979; 11: 37 40. 21 Humiczewska M. [Demodex folliculorum and Demodex brevis (Acarida) as the factors of chronic marginal blepharitis]. Wiad Parazytol 1991; 37: 127 130. 22 Clifford CW, Fulk GW. Association of diabetes, lash loss, and Staphylococcus aureus with infestation of eyelids by Demodex folliculorum (Acari: Demodicidae). J Med Entomol 1990; 27: 467 470. 23 Millikan LE. Androgenetic alopecia: the role of inflammation and Demodex. Int J Dermatol 2001; 40: 475 476. 24 Georgala S, Katoulis AC, Kylafis GD, et al. Increased density of Demodex folliculorum and evidence of delayed hypersensitivity reaction in subjects with papulopustular rosacea. J Eur Acad Dermatol Venereol 2001; 15: 441 444. 25 Wikel SK, Ramachandra RN, Bergman DK. Tick-induced modulation of the host immune response. Int J Parasitol 1994; 24: 59 66. 26 Wikel SK. Immunomodulation of host responses to ectoparasite infestation an overview. Vet Parasitol 1984; 14: 321 339. 27 Barriga OO, al-khalidi NW, Martin S, et al. Evidence of immunosuppression by Demodex canis. Vet Immunol Immunopathol 1992; 32: 37 46. 28 Zackheim HS. Origin of the human basal cell epithelioma. J Invest Dermatol 1963; 40: 283 297. International Journal of Dermatology 2003, 42, 567 571