Clinical Reference / Dermatopathology: Practical & Conceptual / Jan – Mar 2001 | Vol. 7, No. 1 / Keratoacanthomas: A new classification based on morphologic findings and on anatomic site

Keratoacanthomas: A new classification based on morphologic findings and on anatomic site

Jan – Mar 2001 | Vol. 7, No. 1
Ackerman, A. Bernard; Choonhakarn, Charoen

Historical Perspective

In 1889, Hutchinson1 described for the first time what now is known as keratoacanthoma and he designated it “crateriform ulcer of the face.” Because the neoplasm he studied grew rapidly and because of particular findings viewed by conventional microscopy, he concluded that it was an epithelial cancer (Fig. 1). In 1934, Ferguson-Smith2 told of a young man with multiple primary squamous-cell carcinomas of the skin that healed without treatment. His was the first report of multiple keratoacanthomas. In 1950, Grzybowski3 shared his experience with a patient with widespread peculiar epithelial tumors of the skin analogous to those in the patient of Ferguson-Smith, but the lesions in his patient were far more numerous and involved the oral mucosa, too. In 1961, Fisher4 reported on one patient who had painful nodule with a central depression in the sulcus of the paronychial region of a finger, which was diagnosed histopathologically as a “senile acanthoma with intraepidermal prickle-cell epithelioma of very low-grade malignancy.” The x-ray findings were those of early osteomyelitis of the distal phalanx. Fisher dubbed the phenomenon “distinctive, destructive, digital disease.” His seems to have been the first recorded example of subungual keratoacanthoma. In 1962, Miedzinski and Kozakiewicz5 called attention to a tumor they named keratoacanthoma centrifugum because it grew outward steadily to reach mammoth proportions. Subsequently, Belisario6 provided details about seven other examples of the same neoplasm and coined the term keratoacanthoma centrifugum marginatum based on the irrepressible extension of it outward.

Fig. 1

Three illustrations of ‘The crateriform ulcer of the face” (from Hutchinson J. The crateriform ulcer of the face: a form of acute epithelial cancer. Trans Pathol Soc London 1889;40:275-81).

In times past, keratoacanthoma was deemed to be diagnosable, clinically and histopathologically, by virtue of a peculiar crateriform pattern. It went under a variety of appellations besides those already mentioned, among them, verrucome,7 kyste sebace atypique,8 self-healing primary squamous-cell carcinoma,9 molluscum sebaceum,10 tumor-like keratosis,11 molluscum pseudocarcinomatosum,12 idiopathic cutaneous pseudoepitheliomatous hyperplasia,13 button epithelioma,14 and keratocarcinoma.15 Today, the name keratoacanthoma, which was introduced by Freudenthal in the 1936, is employed by dermatologists and pathologists universally.14,16

Theories of Pathogenesis

Many workers have sought to elucidate the mechanisms responsible for evolution and involution of keratoacanthoma of the type presented first by Hutchinson. On the basis of study by conventional microscopy of sections of tissue of keratoacanthomas from both humans and animals, the neoplasm was thought initially to derive from hair follicles, a proposition illustrated diagrammatically by Ghadially in 1958 and again in 1961.17,18 He categorized keratoacanthomas according to three morphologic types as follows: type 1 or bud-shaped; type 2 or dome-shaped; and type 3 or berry-shaped. The domed and berry-shaped types were thought by him to arise from follicles below the site at which they were tethered to a muscle of hair erection, whereas the bud-shaped type was believed by him to originate from follicles above the site where the muscle was attached. The idea of keratoacanthoma emerging from the lower part of a follicle has not been accepted generally.19,20-24 Reed25 proposed that “the normal epidermis, the epithelium forming the acral portions of the skin appendages, and an abnormal clone of epidermal cells participate as a unit in the formation of an actinic keratoacanthoma.” The initial rapid growth and the eventual regression of keratoacanthomas were interpreted by some workers to represent recapitulation by the neoplasm of the follicular cycle, to wit, growth (anagen), involution (catagen), and rest (telogen).18,26 Expulsion of the plug of cornified cells from the crater of the neoplasm in conjunction with destruction of viable cells in what remained of it were claimed to be a consequence of a phase of the neoplasm analagous to telogen. Some students of the subject have proposed that regression of keratoacanthoma results from progressive differentiation toward cornification,27 whereas others conceive of it as being consequent to immunologic mechanisms.17,28,29 After extensive review of the literature and performance of studies on “spontaneously-generated” lesions in humans and on experimentally-produced lesions in animals, Remselaar and van der Meer30,31 came to the conclusion that regression of keratoacanthoma was a function of “nonimmune mechanisms.” In short, the mechanisms responsible for regression of the type of keratoacanthoma that occurs on sites which bear hair and described originally by Hutchinson have not yet been unraveled.

Sites: Skin, Subungual, Mucosa

Because keratoacanthomas occur at times on mucous membranes and in the subungual region, as well as on hair-bearing skin, it is obvious that hair follicles cannot be essential to development of them on all anatomic sites. The occurrence of keratoacanthomas on mucous membranes is believed by some authors to be mere extension of a neoplasm that actually began on contiguous hairy skin.32 Some keratoacanthomas, however, appear to arise directly on a mucous membrane.14 Belisario33 raised the possibility that such lesions could derive from ectopic hair follicles; Miles34 even claimed to have found a hair follicle in the mucosa of a cheek, a supposition that seems to us to lack credence. Svirsky et al.35 have postulated that keratoacanthomas on mucous membranes may evolve from sebaceous glands that are present normally on the buccal mucosa and ectopically on other mucous membranes (“Fordyce spots”). That speculation also boggles, as does the assertion that appearance of keratoacanthomas in association with sebaceous neoplasms in the Muir-Torre syndrome is analogous to the situation that obtains for keratoacanthomas on a mucous membrane, that is, origin of them from sebaceous units.27 All of these speculations are reminiscent of the appellation “molluscum sebaceum” by MacCormac and Scarff, who thought that condition was due to hypertrophic and inflammatory changes in a sebaceous cyst, a proposal that is not compelling. Eversole et al.36 averred more believably that keratoacanthomas on mucous membranes originated from surface epithelium and not from sebaceous glands. Until now, the basic character of keratocanthomas on non-hair-bearing sites remains as uncertain as that of keratoacanthomas that develop on skin that harbors hair follicles, but the nature of them seems to be different.

Keratoacanthomas Are Squamous-cell Carcinomas

All keratoacanthomas, irrespective of site of origin, qualify as being squamous-cell carcinomas, not only for reasons that relate to architectural features as assessed by silhouette and those that pertain to cytopathologic findings, but because of biologic behavior, in particular, capability for metastasis and for destruction of important structures locally.37-41 The most common type of keratoacanthoma occurs on skin that bears hair and is typified by crateriform appearance (as a consequence of dilation of contiguous infundibula that are filled with corneocytes), rapid growth, and involution usually, but not always, in the absence of treatment. An uncommon kind of keratoacanthoma develops beneath a nail plate or on a mucous membrane and is characterized by a crateriform appearance that is wholly unrelated to infundibula (which are not present in the nail unit or on a mucous membrane), slow growth, and no tendency to involution, but rather for evolution persistently. In the past, no classification of keratoacanthoma took these profound differences into account.

Six Types of Keratoacanthoma

Diagnosis of keratoacanthomas has been based mostly on clinical features (Table 1). At least 6 different clinical types of keratoacanthoma have been recognized in textbooks, among them the (1) solitary (Fig. 2), (2) multiple self-healing (Ferguson-Smith), (3) generalized eruptive (Grzybowski) (Fig. 3), (4) combined Ferguson-Smith and Grzybowski (Witten and Zak), (5) giant or confluent (Fig. 4), and (6) centrifugum marginatum (Fig. 5).42,43 Keratoacanthomas are acknowledged to occur on mucous membranes (Fig. 6),35,36,44-55 on palms and soles,19,56,57 and in nail units (Fig. 7).4,58-61 Patients with generalized eruptive keratoacanthomas may have the neoplasms in the oral cavity.3,62-64

Table 1 Reputed Variants of Keratoacanthomas

Solitary19
Dyskeratoticum and segregans66
Giant67
Centrifugum marginatum5,6
Agressive68
Verrucous68
Subungual4,58-61
Mucous membrane35,36,44-55
Pseudorecidive69
Reactive70
Multiple
– Ferguson-Smith type2
– Grzybowski type3,62-64
– Witten and Zak type71
– Marshall-Pepler type12
– Multiple persistant type56
Chemical-induced (tar)19
In nevus sebaceus of Jadassohn72
In Muir-Torre syndrome19
In xeroderma pigmentosum73
In an immunosuppressed patient74

Fig. 2

Solitary (Hutchinson) type (from Kopf AW, Bart RS, Andrade R. Atlas of Tumors of the Skin. Philadelphia: W.B. Saunders Company, 1978:134).

Fig. 3

Eruptive type (from Grzybowski M. A case of peculiar generalized epithelial tumors of the skin. Br J Dermatol 1950;62:310-13).

Fig. 4

Giant type (from Kopf AW, Bart RS, Andrade R. Atlas of Tumors of the Skin Philadelphia: W.B. Saunders Company, 1978:149).

Fig. 5

Centrifugum marginatum type (from Weedon D, Barnett L. Keratoacanthoma centrifugum marginatum. Arch Dermatol 1975;111:1024-26).

Fig. 6

Mucous membrane type, oral (from Scofield HH, Werning JT, Shukes RC. Solitary intraoral keratoacanthoma. Oral Surg Oral Med Oral Pathol 1974;37:889-98).

Fig. 7

Subungual type (from Kopf AW, Bart RS, Andrade R. Atlas of Tumors of the Skin. Philadelphia: W.B. Saunders Company, 1978:145).

Silhouette the Denominator in Common

Regardless of type, number of lesions (solitary or multiple), or site (hair-bearing or not), the denominator in common for diagnosis of keratoacanthoma is histopathologic appearance of a crateriform squamous-cell carcinoma that consists of a central core of corneocytes and a proliferation of squamous cells that display an abnormal nucleus and abundant acidophilic cytoplasm (Fig. 8). In the type of keratoacanthoma that originates from follicles, abscesses punctuate the squamous epithelium, a mixed infiltrate of inflammatory cells (lymphocytes, plasma cells, and eosinophils) envelops it during the period of evolution, and granulomatous inflammation and fibroplasia come into being at the base of it during the period of involution. The granulomatous inflammation is of the foreign-body type in response to destruction of neoplastic cells secondary to the effects of inflammatory cells on them, and the fibrosis is responsible for pushing the neoplasm upward, the cornified contents of the crater outward, and, eventually, the entire residuum of the neoplasm out of the skin, leaving behind a scar (Fig. 9). In keratoacanthomas that do not originate from follicles, such as the subungual and mucous membrane types, histopathologic findings differ. For example, in subungual keratoacanthoma, there is a tendency to vertical orientation, more dyskeratotic cells within the neoplastic epithelium, few, if any, neutrophils within that epithelium, and absence of fibrosis at the base (Figs. 10A, B).38,59,65 Keratoacanthoma on mucous membrane differs histopathologically from keratoacanthomas on hair-bearing skin mostly by absence of fibrosis at the base of it. No evidence of involution has been described in this or the subungual type (Figs. 11A, B).35,36,44-55 Some reputed keratoacanthomas on mucous membrane reported on lack morphologic signs that enable a diagnosis to be made with confidence.

Fig. 8

Histopathologic findings at scanning magnification (out of focus) of a 6-week-old, fully-developed keratoacanthoma on the thumb of a 75-year-old man (from Rook A, Whimster I. Le Kerato-acanthoma. Arch Belg Dermatol Syphyligr 1950;6:137-46).

Fig. 9

Histopathologic findings at scanning magnification (out of focus) of a 15-week-old keratoacanthoma largely involuted on a preauricular site of a 58-year-old man (from Rook A, Whimster I. Le Kerato-acanthoma. Arch Belg Dermatol Syphiligr 1950;6:137-46).

Fig. 10A-B

Histopathologic findings at (A) scanning magnification and (B) higher magnification of a subungual keratoacanthoma (from Stoll DM, Ackerman AB. Subungual keratoacanthoma. Am J Dermatopathol 1980; 2:256-71).

Fig. 11A-B

Histopathologic findings at (A) scanning magnification and (B) higher magnification of a keratoacanthoma on mucous membrane: conjunctiva (from Schellini SA, Marques MEA, Milanezi MFG, Bacchi CE. Conjunctival keratoacanthoma. Acta Ophthalmol Scan 1997;75:335-37).

Multiple self-healing keratoacanthoma, generalized eruptive keratoacanthoma, giant keratoacanthoma, and keratoacanthoma centrifugum marginatum seem to be variations on a theme of keratoacanthoma that derives from hair follicles. Giant keratoacanthoma, for example, develops by fusion of closely-set individual keratoacanthomas of the classic Hutchinson type, and keratoacanthoma centrifugum marginatum seems to be the same type that merely grows outward with a scalloped border formed of closely-set individual nodules, the center, over time, coming to consist of a scar.42

Classification: Follicular and Non-follicular

In sum, in the context of these observations, keratoacanthomas can be classified by morphological findings and anatomic site into two basic types (Table 2): (1) follicular and (2) non-follicular, the former occuring on skin that bears hair and the latter on sites devoid of hair. The number of lesions on hair-bearing skin may vary from one to hundreds and those in nail units from one to several, whereas on mucous membrane the condition usually is solitary, being multiple only in patients who sport generalized eruptive keratoacanthomas. All of the clinical variants of the follicular type of keratoacanthoma tend to involute in the absence of therapy after months and even years, and leave as residuum a scar.

Table 2 Our Classification of Keratoacanthomas by Morphologic Findings and Anatomic Site

1. Follicular
Hutchinson’s (classic “solitary” type)
Eruptive
Giant
Centrifugum marginatum
2. Non-follicular
Subungual/volar
Mucous membrane

Because the follicular type, particularly the classic (solitary, Hutchinson) expression, has capability for metastasis and sometimes exercises that potential, especially in persons who are immunocompromised, it must be regarded as a kind of squamous-cell carcinoma. Although Hutchinson regarded keratoacanthoma as a squamous-cell carcinoma, Rook and Whimster, half a century later, made a distinction between keratoacanthoma and squamous-cell carcinoma. It was Kwittken in 1975 who stated directly that ” I have come to the firm conclusion that all of these lesions [keratoacanthomas] are malignancies and that the formerly accepted concept of a self-healing squamous cell carcinoma of the skin [by Ferguson-Smith] is a correct one.” Subsequent reports of metastasis of keratoacanthomas, especially in persons who were immunosuppressed, confirmed the assessments of Hutchinson, Ferguson-Smith, and Kwittken.

Keratoacanthomas that develop on sites devoid of hair follicles, especially subungual ones, are fundamentally different from those of the follicular type, both morphologically and biologically. They differ histopathologically from keratoacanthomas of follicular nature by exhibiting many more dyskeratotic cells, but far fewer inflammatory cells. The subungual type may be painful, destructive of bone, persist for many years, and grow in slow but unrestrained fashion. Keratoacanthomas that are situated on mucous membrane, particularly on conjunctiva, grow rapidly, but those in the oral cavity sometimes behave like the subungual type, to wit, they grow slowly and persist for many months or years. The type of keratoacanthoma that appears on a mucous membrane shares histopathologic findings in common with keratoacanthomas that develop in skin that bears hairs, except for what seems to be a mostly lymphoplasmacytic infiltrate and absence of fibrosis. Neither regression nor metastasis have been observed in keratoacanthomas that are positioned in nail units and on mucous membranes (Tables 3, 4). Despite the fact that non-follicular keratoacanthomas rarely kill by either destruction locally or metastasis widely, they nonetheless fulfill criteria histopathologically for squamous-cell carcinoma.

Table 3 Similarities Between Follicular and Non-follicular Keratoacanthomas

Non-follicular
Follicular (Subungual and Mucosal)
Clinical features
Number One or more, but sometimes hundreds Single usually, sometimes multiple
Growth Rapid Rapid for mucosal, but not subungual
Histopathologic findings
Crateriform + +
Squamous cells with abundant acidophilic cytoplasm + +
Nuclear atypia + +
Mitotic figures + +

Table 4 Differences Between Follicular and Non-follicular Keratoacanthomas

Non-follicular
Follicular (Subungual and Mucosal)
Clinical features
Site Hair-bearing skin Non hair-bearing skin
Lesion Crateriform Not always crateriform
Symptoms None Painful at times (especially subungual)
Histopathologic Findings
Infundibular continuity +
Inflammatory cells Many Few or absent
Dyskeratotic cells Few Many (especially subungual)
Fibrosis at base (involution) +
Biologic Aspects
Destruction of bone + (subungual)
Regression +
Metastasis + (rarely)

Keratoacanthomas Represent More than One Type of Squamous-cell Carcinoma

In conclusion, keratoacanthomas are not a single neoplasm, but represent at least two different types of squamous-cell carcinoma that have clinical and histopathologic features in common. Despite those shared features, they seem to be unrelated to one another. Not only does one type begin from follicles and the other does not, but they behave differently. Those distinctions are the basis for the classification that we advocate (Table 2).

From the Ackerman Academy of Dermatopathology in New York City where Dr. Choonhakarn did this work during a Visiting Fellowhip. He is now at Division of Dermatology, Faculty of Medicine, Khon Kaen University in Khon Kaen, Thailand. Dr. Ackerman is director of the Academy.

References

1. Hutchinson J. The crateriform ulcer of the face: A form of acute epithelial cancer. Trans Pathol Soc London 1889;40: 275-81.

2. Ferguson-Smith J. A case of multiple primary squamous-celled carcinomata of the skin in a young man, with spontaneous healing. Br J Dermatol 1934;46:267-72.

3. Grzybowski M. A case of peculiar generalized epithelial tumors of the skin. Br J Dermatol 1950;62:310-13.

4. Fisher AA. A distinctive destructive digital disease. Arch Dermatol 1961;83:1030-1.

5. Miedzinski F, Kozakiewicz J. Das Keratoakanthoma centrifugum – eine besodere Varietät des Keratoakanthoms. Hautarzt 1962;13:348-52.

6. Belisario JC. Brief review of keratoacanthomas and description of keratoacanthoma centrifugum marginatum, another variety of keratoacanthoma. Aust J Dermatol 1965;8:65-72.

7. Gougerot H. Verrucome avec adenite, a structure epitheliomatiforme, curable par le 914. Arch Dermatol Syphiligr Clin Saint-Louis 1929;1:374-85.

8. Dupont A. Kyste sebace atypique. Bull Soc Belge Dermatol Syphiligr 1930:177-79.

9. Dunn JS, Ferguson-Smith J. Self-healing primary squamous cell carcinoma of the skin. Br J Dermatol Syphilol 1934;46: 519-23.

10. MacCormac H, Scarff RW. Molluscum sebaceum. Br J Dermatol 1936;48:624-27.

11. Poth DO. Tumor-like keratoses: Report of a case. Arch Dermatol Syphilol 1939;39:228-38.

12. Marshall J, Pepler WJ. Mollusca pseudocarcinomatosa: Discussion of a case of the Ferguson-Smith type, of unilateral distribution. Br J Cancer 1954;8:251-54.

13. Grinspan D, Abulafia J. Idiopathic cutaneous pseudoepitheliomatous hyperplasia. Cancer 1955;8:1047-56.

14. Rook A, Whimster I. Keratoacanthoma-a thirty year retrospect. Br J Dermatol 1979;100:41-47.

15. Kwittken J. A histologic chronology of the clinical course of the keratocarcinoma (so-called keratoacanthoma). Mt. Sinai J Med 1975;42:127-35.

16. Rook A, Whimster I, Le Kerato-acanthoma. Arch Belg Dermatol Syphyligr 1950;6:137-46.

17. Ghadially FN. A comparative morphological study of keratoacanthoma of man and similar experimentally produced lesions in the rabbit. J Pathol Bacteriol 1958;75:441-53.

18. Ghadially FN. The role of the hair follicle in the origin and evolution of some cutaneous neoplasm of man and experimental animals. Cancer 1961;14:801-16.

19. Schwartz RA. Keratoacanthoma. J Am Acad Dermatol 1994; 30:1-19.

20. Calnan CD, Haber H. Molluscum sebaceum. J Pathol Bacteriol 1955;69:61-66.

21. Rook A, Champion RH. Keratoacanthoma. In a conference on Biology of Cutaneous Cancer, Bethesda. National Cancer Institute Monograph No. 10, 1963: 257-73.

22. Yang JS, Lavker RM, Sun TT. Upper human follicle contains a subpopulation of keratinocytes with superior in vitro proliferative potential. J Invest Dermatol 1993;101:652-59.

23. Lavker RM, Miller SJ, Sun TT. Hair cycle and skin carcinogenesis; potential involvement of epithelial stem cells. In: Mukhtar H, ed. Skin cancer: Mechanisms and human relevance. Boca Raton: CRC Press, Inc., 1995:71-78.

24. Kirkham N. Tumors and cysts of the epidermis. In: Elder D, Elenitsas R, Jaworsky C, Johnson Jr B, eds. Lever’s Histopathology of the Skin. 8th ed. Philadelphia: Lippincott-Raven, 1997:731-34.

25. Reed RJ. Actinic keratoacanthoma. Arch Dermatol 1972;106: 858-64.

26. Whiteley HJ. The effect of the hair growth cycle on experimental skin carcinogenesis in the rabbit. Br J Cancer 1957; 11:196-205.

27. Schwartz RA. The keratoacanthoma: A review. J Surg Oncol 1979;12:305-17.

28. Brown FC, Tan EM. Immunological reaction in keratoacanthoma, a spontaneously resolving skin tumor. Cancer Res 1973;33:2030-33.

29. Walder BK, Robertson MR, Jeremy D. Skin cancer and immunosuppression. Lancet 1971;2(7737):1282-3.

30. Ramselaar CG, van der Meer JB. The spontaneous regression of keratoacanthoma in man. Acta Derm Venereol (Stockh) 1976;56:245-51.

31. Ramselaar CG, van der Meer JB. Non-immunological regression of dimethylbenz (a)anthracene-induced experimental keratoacanthomas in the rabbit. Dermatologica 1979;158: 142-51.

32. Kopf AW. Keratoacanthoma: Clinical aspects. In: Andrade R, Gumport SL, Popkin GL, Rees TD, eds. Cancer of the Skin. Vol. 1. Philadelphia: W.B. Saunders Company, 1976:755-81.

33. Belisario JC. Keratoacanthomas: Review and therapy. Cutis 1975;16:527-48.

34. Miles AEW. A hair follicle in human cheek mucosa. Proc Roy Soc Med 1960;53:527-30.

35. Svirsky JA, Freedman PD, Lumerman H. Solitary intraoral keratoacanthoma. Oral Surg Oral Med Oral Pathol 1977; 43:116-22.

36. Eversole LR, Leider AS, Alexander G. Intraoral and labial keratoacanthoma. Oral Surg Oral Med Oral Pathol 1982;54: 663-67.

37. Ng P, Ackerman AB. The major types of squamous cell carcinoma. Dermatopathol: Pract & Conc 1999;5(3):250-52.

38. Hodak E, Jones RE, Ackerman AB. Solitary keratoacanthoma is a squamous-cell carcinoma: Three examples with metastasis. Am J Dermatopathol 1993;15:332-42.

39. Ackerman AB, Kerl H, Sanchez J, Guo Y, Hofer A, Kelly P, et al. Squamous-cell carcinoma. In: A Clinical Atlas of 101 Common Skin Disease with Histopathologic Correlation. New York City: Ardor Scribendi, 2000:572-86.

40. Schnur PL, Bozzo P. Metastasizing keratoacanthomas? Plas Reconst Surg 1978;62:258-62.

41. Piscioli F, Boi S, Zumiani G, Cristofolini M. A gigantic metastasizing keratoacanthoma. Am J Dermatopathol 1984; 6:123-29.

42. Ghadially R, Ghadially FN. Keratoacanthoma. In: Freedberg IM, Eisen AZ, Wolff K, Austen KF, et al., eds. Dermatology in general medicine, 5th ed. Vol. 1. New York: Mc Graw-Hill, 1999:865-72.

43. Koh HK, Bhawan J. Tumors of the skin. In: Moschella SL, Hurley HJ, eds. Dermatology. 3rd ed. Vol. 2. Philadelphia: W.B. Saunders Company, 1992:1728-9.

44. Helsham RW, Buchanan G. Keratoacanthoma of the oral cavity. Oral Surg Oral Med Oral Pathol 1960;13:844-49.

45. Freeman RG, Cloud TM, Knox JM. Keratoacanthoma of the conjunctiva. Arch Ophthalmol 1961;65:817-19.

46. Bellamy ED, Allen JH, Hart NL. Keratoacanthoma of the bulbar conjunctiva. Arch Ophthalmol 1963;70:512-14.

47. Friedman RP, Morales A, Burnham TK. Multiple cutaneous and conjunctival keratoacanthoma. Arch Dermatol 1965;92: 162-65.

48. Johnson HA. Query: Keratoacanthoma of mucous membrane? Plast Reconstr Surg 1968;41:373-75.

49. Zegarelli DJ. Solitary intraoral keratoacanthoma. Oral Surg Oral Med Oral Pathol 1975;40:785-88.

50. Roth AM. Solitary keratoacanthoma of the conjunctiva. Am J Ophthalmol 1978;85:647-50.

51. Heidenreich RK, Gongloff RK, Wescott WB. A solitary, exophytic, crateriform lesion on the mandibular retromolar gingiva. JADA 1986;112:377-79.

52. Hamed LM, Wilson II FM, Grayson M. Keratoacanthoma of the limbus. Ophthalmic Surg 1988;19:267-70.

53. Grossniklaus HE, Martin DF, Solomon AR. Invasive conjunctival tumor with keratoacanthoma features. Am J Ophthalmol 1990;109:736-38.

54. Munro S, Brownstein S, Liddy B. Conjunctival keratoacanthoma. Am J Ophthalmol 1993;116:654-5.

55. Schellini SA, Marques MEA, Milanezi MFG, Bacchi CE. Conjunctival keratoacanthoma. Acta Ophthalmol Scan 1997; 75:335-37.

56. Schwartz RA. Mutiple persistant keratoacanthomas. Oncology (Basel) 1979;36:281-85.

57. Hales DS, Dockery GL. Giant keratoacanthoma of the plantar foot: A report of two cases. J Foot Ankle Surg 1993;32:75-84.

58. Fisher AA. Subungual keratoacanthoma: possible relationship to exposure to steel wool. Cutis 1990;46:26-28.

59. Stoll DM, Ackerman AB. Subungual keratoacanthoma. Am J Dermatopathol 1980; 2:256-71.

60. Oliwiecki S, Peachey RDG, Bradfield JWB, Ellis J, Lovell CR. Subungual keratoacanthoma-a report of four cases and review of the literature. Clin Exp Dermatol 1994;19:230-35.

61. Baran R, Goettmann S. Distal digital keratoacanthoma: A report of 12 cases and a review of the literature. Br J Dermatol 1998;139:512-15.

62. Young SK, Larsen PE, Markowitz NR. Generalized eruptive keratoacanthoma. Oral Surg Oral Med Oral Pathol 1986; 62:422-26.

63. Street ML, White Jr JW, Gibson LE. Multiple keratoacanthomas treated with oral retinoids. J Am Acad Dermatol 1990;23:862-66.

64. Jaber PW, Cooper PH, Greer KE. Generalized eruptive keratoacanthoma of Grzybowski. J Am Acad Dermatol 1993;29: 299-304.

65. Ackerman AB, Ragaz A. The Lives of Lesions: Chronology in Dermatopathology. New York: Masson Publishing, 1984:102-09.

66. Stevanovic DV. Keratoacanthoma dyskeratoticum and segregans. Arch Dermatol 1965;92:666-69.

67. Rapaport J. Giant keratoacanthoma of the nose. Arch Dermatol 1975;111:73-75.

68. Straka BF, Grant-kels JM. Keratoacanthoma. In: Friedman RJ, Rigel DS, Kopf AW, Harris MN, Baker D, eds. Cancer of the Skin. Philadelphia: W.B. Saunders Company, 1991:390-407.

69. Shaw JC, Storrs FJ, Everts E. Multiple keratoacanthomas after megavoltage radiation therapy. J Am Acad Dermatol 1990;23:1009-11.

70. Allen JV, Callen JP. Keratoacanthomas arising in hypertrophic lichen planus. Arch Dermatol 1981;117:519-21.

71. Witten VH, Zak FG. Mutiple, primary, self-healing prickle-cell epithelioma of the skin. Cancer 1952;5:539-50.

72. Wilkinson SM, Tan CY, Smith AG. Keratoacanthoma arising within organoid naevi. Clin Exp Dermatol 1991;16:58-60.

73. Stevanovic DV. Keratoacanthoma in xeroderma pigmentosum. Arch Dermatol 1961;84:53-4.

74. Washington CV Jr, Mikhail GR. Eruptive keratoacanthoma en plaque in an immunosuppressed patient. J Dermatol Surg Oncol 1987;13:1357-60.