Tricholemmoma and “Catagen-like” Patterns:
One approach to the common follicular hamartomas stresses the separateness of inverted follicular keratosis and tricholemmoma; this approach is a convenient compromise. To exercise this option, many overlaps in patterns must be ignored.
The complicated patterns, that are encountered in a study of the stages of normal follicular development and involution, tend to be rigidly structured; the fluidity with which follicular epithelium reverts from a trichogenic phase to a regressive phase is hardly expressed in the concepts of anagen, catagen, and telogen phases. These categories are little more than a histologic documentation of a clinical evaluation of the condition of the hairs. The classic histologic classification of a growing hair gives recognition to cellular sheaths, a germinative bulb, “cuticles,” and Henle’s, and Huxley’s “layers.” Perhaps, the problems reflect a general lack of interest in images, for which the descriptions have remained static for so many years. The patterns, in the general category of follicular hamartomas of the butterfly area of the face, are manifold. Many of the optional displays, as manifested in the regular cycling of normal hair follicles, are represented in distorted patterns in the follicular hamartomas.
A somewhat altered characterization of the phases of follicular growth is available in Chapters 9 through 13a. In this altered characterization, it is difficult to utilize the classic features of the structure of an anagen follicle in evaluating the histologic features of a regressing follicle. In regression, as a portion of the cellular component of the altered bulb moves upward, components such as Henle’s layer, Huxley’s layer (i.e., the main component of the inner root sheath), and the outer root sheath are altered; they eventually lose their cytologic identities (i.e., the distinctive phenotypes, as seen in an anagen phase, give way to a population more restricted in its phenotypic expressions); new images and new designations are required. In the chapters that are “children” of Chapter 9 (Chapter 9 is a parent, at tier 2, for nine chapters along tier 3; some of the 9 “children” along tier 3 are parents for eleven chapters along tier 4.), new images and designations are offered.
Ideally, the reader, having worked through Chapters 9-14b, will come away with a new appreciation for the complexity of follicular processes during normal cycling. He should come away with a recognition that the expression of phenotypes in the common follicular hamartomas of the butterfly area of the face recapitulate, in distorted fashions, some of the phenotypes that are expressed during cyclic follicular processes. The varied patterns of these follicular neoplasms are difficult to integrate in a usable classification. In the expression of patterns in the follicular neoplasms of the butterfly area of the face, the most glaring deficiency is a striking omission of the trichogenic component, the progenitors of inner root sheath, and an inner root sheath. It is possible to find similarities beween some of the patterns of “inverted follicular keratosis,” and the constant portion of a hair follicle; in addition, and for the same shared images, there are overlaps with the patterns of the common seborrheic keratosis (Ch25). It is, however, incorrect to accept the notion that “inverted follicular keratosis” is nothing more than an “irritated seborrheic keratosis.”
Classic tricholemmoma finds its distinctiveness in the character of the component cells (Chs24 & 40c). These cells often are prominently vacuolated; they appear as clear cells with well-defined cell membranes. In addition, a palisade of distinctive columnar basal cells is often a feature; this palisade of cells is associated with a hyaline (“glassy membrane”). In a classic example, nuclei are small and uniform. On the other hand, for many examples, varying degrees of nuclear atypia are common. Cysts occasionally are a feature; they contain a pale mucinous material. A classic tricholemmoma, at the histologic level, manifests a basophilia that contrasts with the acidophilia of the regional squamous epithelium.
Two basic patterns (i.e., inverted follicular keratosis and tricholemmoma) are so commonly combined in a single lesion that one might question the utility of maintaining a taxonomic separation between the two. In practice, vacuolated cells, once identified, are assigned primacy; priority then is given to the tricholemmal category. “Inverted follicular keratosis” is a tumor of acidophilic squamous cells; a component of vacuolated cells is a regular feature, but the cytologic features of the vacuolated cells is more in keeping with intracellular edema than with the clear cell quality of the cells of a classic tricholemmoma. With these two basic components (clear cells, and vacuolated squamous cells) as the standard by which divisions are to be attempted, any comparisons with normal follicular phenomena might then foster the notion that the pattern of inverted follicular keratosis is representative of the constant portion of a follicle (the portion above the upper level of the outer root sheath), and that the tricholemmal component is representative of the inconstant portion (that component below the isthmus). Even this advance is compromised.
There are lesions, or multiple foci in lesions that are rich in vacuolated cells, but the cells are clearly keratinocyte-like; they basically look like, and have the tinctorially qualities of, adjacent, non-vacuolated keratinocytes. The vacuoles, in contrast to the clear cytoplasm of the cells of a classic tricholemmoma, tend to be rounded and peri-nuclear; the vacuolization is of a type that is more in keeping with intracellular edema. For some lesions, these vacuolated cells are prominently represented. Cytologic differences are likely to be ignored; such lesions are likely to be dismissed as classic tricholemmoma. In many lesions, that otherwise resemble classic inverted follicular keratosis, these acidophilic, vacuolated cells are represented in scattered clusters. For some lesions, in which clusters of vacuolated keratinocytes are a feature, cytolytic and dyskeratotic changes occur almost exclusively in the areas with the vacuolated cells. It would seem that the patterns, although having some resemblance to those of the outer root sheath (as seen in an anagen follicle), are indicative of a process that serves some purpose other than that of an outer root sheath of an anagen follicle. In the process of dyskeratosis and cytolysis, the whorls of keratinocytes tend to be preferentially spared, even if some of the cells of the whorls are vacuolated. As an intermediate phase, lytic clefts, that result from the cytolytic process, are spaced among islands and cords of plump, acidophilic keratinocytes. The lytic process finds its correspondent in the patterns of a catagen follicle in which dyskeratosis and cytolysis trail the follicular lumen as the catagen bulge retracts upward (Fs4 & 5Ch11).
The cytolytic process in the follicular tumors often extends to the periphery of a nest to include the basal layer; in this event, a path is opened for the inlay of activated connective tissue in the lytic defects. In the process, the surviving whorls and cords become entrapped in the newly formed connective tissue. It must be emphasized that the entrapped epithelium usually lacks a basal layer. The epithelium, mostly representing what remains of the whorls, is immunologically exposed; being, by its very nature, committed to terminal differentiation, it undergoes keratinization. The resulting patterns are easily mistaken for those of an invasive carcinoma; the process is sclerosing entrapment.
It would be nice if this were the end of the complicating patterns, but such is not the case.
The lesion, as represented in chapters 25 & 26, has many of the features of a follicular hamartoma of catagen-like type. As such, it qualifies, in part, as a catagen-like, or catagen-committed tricholemmoma. In addition, there are areas in which the cells, of what first appears to be more of the tricholemmal component, have plump, round nuclei, each with a prominent, central nucleolus; the nuclear changes can be compared to the nuclear features of the germinative layer of a sebaceous gland. Small aggregates of sebaceous cells are scattered about in the squamous cell component.
The lesion in chapter 27 has well-defined follicular patterns; it is a vertically oriented column. The hyperplastic epidermis of the lesion shows papillomatosis. Many of the cells forming the dermal component have vacuolated cytoplasm. On the basis of these two features, it would appear that the lesion is best classified as variant of tricholemmoma; catagen-like patterns (i.e., dyskeratosis and cytolysis) are not a feature. Cytologically, the vacuolated cells have dense, acidophilic cytoplasm; the qualities deviate from the cytologic features of tricholemmoma of the classic type - a lesion in which the most characteristic vacuolated cells have thin cell walls with no significant acidophilia. It would be difficult to find a more deserving category, to which the lesion can be assigned, than tricholemmoma. Sebaceous gland lobules are closely spaced in the dermis near the extremity of the lesion; they are present in a regularly spaced pattern that is not carried over into the reticular dermis beyond the limits of stroma of the tricholemmoma. It is tempting to propose that the sebaceous gland hyperplasia is a nevoid feature; it would appear to be part and parcel of a single lesion.
The lesion in chapter 28 shows focal, solid squamous cell patterns in which some of the keratinocytes have prominent perinuclear vacuoles. The cytologic features are not those that would be emphasized in the diagnosis of a classic tricholemmoma, but are of a type commonly seen as a regional variation in tricholemmomas. A second common pattern is characterized by cords and nests of squamous cells in a fibrous matrix that is uniformly cellular; some of the cells forming isolated nests have brightly acidophilic cytoplasm (a quality identifying the cells as being late in the process of keratinization). It should be noted that these nests and cords of cells lack a defined basal layer, and that keratinizing cells tend to die after completing the process of keratinization. Peculiarly, a foreign body cellular response is rarely a feature of these follicular lesions showing patterns of sclerosing entrapment. The pattern in this lesion is representative of a late stage in the process of sclerosing entrapment; the lytic defects that characterize an earlier stage have all been inlaid with fibrous tissue. Some observers might characterize the lesion as a variant of a “inverted follicular keratosis” (i.e., desmoplasmic variant). Other observers might assign primacy to the population of vacuolated cells, and then assign the lesion to the category of tricholemmoma. The combination of features can be embraced in the concept of catagen-like tricholemmoma (with patterns of sclerosing entrapment).
In chapters 29b & 29c, a multinodular tumor shows some features in keeping with tricholemmal differentiation. Cytologically, the tumor cells are atypical, and have basosquamous qualities. Coarse, membranous deposits of hyalin are a feature. In areas, the tumor shows degenerative changes with cytolysis. There are scattered areas in which the changes are compatible with sweat duct differentiation. Finally, the stroma focally is myxoid and expansile; some of the tumor cells have lost their attachment to other epithelial cells; they are isolated in the myxoid matrix. The tumor is a low-grade tricholemmal basosquamous carcinoma, but may additionally represented a tricholemmal variant of a mixed tumor showing areas of divergent differentiation.
The lesion, that is represented in chapters 29a & 29, is interpreted as a basosquamous carcinoma. A list of its features might include immature follicle-like patterns (mostly near a margin of the lesion), prominent hyaline membranes, columns of squamous cells that extend upward from the main mass of tumor to attach the lesion to the epidermis, confluence of nests and fascicles in the deep portion of the lesion, distinctive sclerosing patterns, and focal clear cell patterns. A comparison of the items of this list with a list of features of tricholemmoma (including all the variant expressions) might lead to a diagnosis of a variant of tricholemmoma. The differences, however, seem to identify the lesion as a tricholemmal basosquamous carcinoma.
Irritated seborrheic keratosis is a second lesion of chapter 29 (F5-Ch29); the discussion of this lesion is continued at the beginning of chapter 30.
At the beginning of chapter 30, a discussion of irritated seborrheic keratosis is continued from chapter 29. The remainder of the chapter (beginning with (F3Ch30) is a presentation of a lesion with patterns of follicular differentiation. Columns of squamous cells extend from a hyperplastic epidermis. The epithelial component of the columns becomes hyperplastic and confluent in the deeper portion of the lesion. Numerous whorls, many of which show central areas of keratinization, are a feature in the deeper portions of the lesion. Superficially, there are two follicular shaped cavities that contain keratinized debris. There are mild infiltrates of inflammatory cells in the adjacent dermis. The patterns satisfy the criteria for the diagnosis of “inverted follicular keratosis.” There are scattered, small lytic defects; the patterns are similar to those seen as a regional variation in both catagen-like tricholemmomas, and mixed or combined tricholemmomas. The lesion can be assigned to a category of catagen-like tricholemmoma.
The lesion of chapter 31 satisfies the criteria for inclusion in the category of follicular hamartoma and, by current criteria, qualifies as an inverted follicular keratosis; the basic pattern sets the lesion apart from “irritated seborrheic keratosis.” By the criteria herein, the lesion would qualify as a catagen-like tricholemmoma. Patterns of dyskeratosis, and early (mild) cytolysis are evident.
The patterns of the lesion represented in chapters 32a & 32 are compatible with an atypical tricholemmoma; there is mild to moderate cytologic atypia. Some of the patterns might be characterized as a variant expression of sclerosing entrapment, but the character of the stromal response is unusual. The stromal response sets the lesion apart; the stroma is hyalinized; in areas, spindle and stellate shaped tumor cells are individually entrapped in the specialized stroma. The lesion has mixed qualities of the type sometimes characterized as pleomorphic. The tumor cells in fascicles, and as individual cells in the hyalin matrix, tend to outlined by straight, brightly acidophilic fibers. These epithelial cells might be peculiar myoid (myoepithelial) cells, or the fibers might be of basement membrane type. The pattens might be compared to the relationships between fibers and cells in the deep portion (root) of an anagen hair shaft; on this basis, the fibers could be interpreted as a peculiar expression of a distinctive form of keratinization (i.e., trichogenic keratinization).
The lesion of chapter 33 is a small carcinoma showing an expansile pattern of growth with scattered, small areas of destructive micro-invasion. The cytologic features offer support for an interpretation that the lesion is a tricholemmal carcinoma. It is confined to the dermis (i.e., has not invaded the subcutaneous tissue).