An Old Man With a Stroke and Multiple Bumps on His Head

HISTORY

A 79-year-old man admitted for mental status changes, with neuroimaging that reveals lacunar brain infarct.

Cannot report whether other family members have had changes on scalp resembling his own striking ones, nor whether anybody has had brain tumors or tumors of nerves. Denies that his tumors hurt or itch.

PHYSICAL EXAMINATION

Elderly, confused man with exceptionally numerous rounded pink nodules at anterior hairline and farther back on scalp. All feel smooth and rubbery; do not indent when pressed by a palpating finger. None are fixed to underlying fascia or galea aponeurotica.

No skin lesions elsewhere on body. No parotid enlargement. No focal neurological deficit.

WHAT'S YOUR DIAGNOSIS?

ANSWER: TURBAN TUMOR (ECCRINE CYLINDROMA)

The nodules are much more prominent than they might be, because no hair grows from them. Two similar bumps on the forehead show a more ordinary flesh-color than the pink-to-light-purple of the others. No lesion is ulcerated.

Such numerous nodules form a pattern that is more distinctive than the individual lesion morphology: they suggest some kind of predisposition to multiple neoplasms, or a syndrome. The masses are clearly not hematomas from falls, whether fresh or old. They are far too numerous to raise serious consideration of multiple epidermal inclusion cysts (“sebaceous cysts” or “wens”), and they stand out too sharply from the surrounding skin. Nor can they be multiple keloidal scars. Multiple long-neglected basal cell carcinomas¹ offer an additional prospect, but are likely to include ulceration and to occur much more in areas where the skin is not shaded by hair.

An image of the Elephant Man of literature² may come to mind, raising the spectre of neurofibromatosis.³ But the sparing of the rest of the body, and the first clinical manifestation in old age, speak to a generally bettertolerated genodermatosis, namely turban tumor, the multiple eccrine cylindromas that constitute Brooke- Spiegler syndrome.^4-7

Cylindromas are benign neoplasms of sweat duct cells, by far most commonly originating from scalp or nearby head and neck. Under the microscope, this patient’s lesion is typical (Figure): at first glance it suggests basal cell carcinoma. However, extensive basement membrane material deposited by the growing cells forms a matrix that separates polygonal islands of tumor cells that resemble pieces of a puzzle fitted together. Non-connection with the overlying epidermis both underscores the origin in a skin adnexa rather than from surface epithelium, and corresponds to clinical behavior: ulceration occurs late, and when it does, it reflects upgrowth from below with or without pressure necrosis. Like most patients treated for cylindromas, this man did well; he died several years later of unrelated causes.

ADVERSE OUTCOMES

Almost every patient with turban tumors has an undesirably public disfigurement, and the cosmetic issues are formidable. Even in a society where headgear is consistently worn in public, the adverse cosmetic effect is only too devastating in one’s private life in the home. When ulceration does take place, it adds a further layer to the misery: malodor. The severity of depression that tails turban tumors was emphasized 54 years ago and by a surgeon—testament to its intensity.⁸

The tumors, unsightly enough at best, can coalesce and enwrap the entire head in a fashion that bespeaks their common name, which is not entirely fanciful. Clinical images in Rosai’s text⁴ or the reports by Poblete Gutierrez and colleagues,⁵ Freedman and Woods,⁹ or Fabian and Shugar¹⁰ show extreme forms that can arise after years of neglect or failed treatment.

Infection of ulcerated masses can occur, and unless the underlying problem is addressed, is likely to recur after every successful course of antibiotic therapy.

Erosion of bone can follow expansive growth, and the external auditory canal can be destroyed.^6,8

The development of carcinoma is dreaded but mercifully quite rare. However, the aggregate complications clearly justify considering cylindroma a biologically benign tumor that can pursue a malignant course.

GENETICS

For nearly 2 centuries it has been known that turban tumors run in families,^4-6,8,11 and now we have a stunning amount of information about them.¹¹ There is a CYLD (short for cylindroma) gene on chromosome 16, which produces a tumor suppressor protein. Affected individuals lack the normal heterozygosity at this site, and as a result, some normal modulation of the growth of various skin adnexal cells is disturbed. Although the inheritance is autosomal dominant, variable expression is prominent: Women are affected more often than men. The distribution of resultant tumors does not follow the pattern of sweat glands but that of (capital, not body) hair. Some genetically indistinguishably affected patients develop histologically related benign skin tumors called trichoepithelioma either in association with cylindromas, or exclusively; yet others show transition from one histology to another within a single tumor. A plethora of histologic variants recalls glands that produce pure salt-and-water sweat (eccrine), sebum and hair (pilosebaceous), or sweat that includes bits of pinched-off luminal cytoplasm (apocrine).⁴

Among the many humbling things I learned in trying to dig out of my decades-outdated knowledge of genetics was that one element has been mapped, using extraordinary tools for base sequencing, as a frameshift mutation.⁵ This is a region of DNA in which a base has been added or deleted, fouling the coding for a sequence of amino acids.

To illustrate, here's an analogy in English letters and words:

pea-tea-ran-din-kit-zoo

Let’s use these letters to create words that are always exactly three letters long (just as each set of three base pairs in DNA or RNA encodes for one amino acid of protein). If we delete all three of the letters “pea,” the remaining words still make sense:

tea-ran-din-kit-zoo

But if we drop out the “p” in “pea,” we shift the frame that holds each three-letter word, and wind up with letter groups that are totally different:

pea-tea-ran-din-kit-zoo becomes

eat-ear-and-ink-itz-oo

By good luck, the first four of these new products are actual words; the last two are gibberish. So too with the coding error that results from a genetic frameshift mutation. One interim outcome can be creation of a new protein that no longer serves its function, or to cut short synthesis altogether if the new sequence constitutes an instruction to terminate synthesis. The end result can be a lifetime of head and face tumors starting with small papules in adolescence.

FEATURES AND LIMITS OF CLINICAL IDENTIFICATION

Cylindromas are usually described as plum-colored to violaceous, but can be any shade of pink or tan as well. Multiple coexistent scalp tumors carry a limited differential diagnosis, and all the more so when reported as slowly growing over multiple decades. Fixation, induration, or regional lymphadenopathy will raise the question of malignant transformation in one or more. Lymphadenopathy can be misleading, however, in that ulcerated tumors can lead to dermatopathic lymphadenopathy, a striking abnormality in which the cut surface of the excised node is often black but the histology is purely reactive, not metastatic (and not melanomatous). Some patients with Brooke-Spiegler syndrome have parotid or other salivary gland neoplasms, typically benign,¹¹ and so a putative “node” may prove to be salivary gland instead.

Even those dermatologists who specialize in genetic skin tumor syndromes^5,11-13 agree that one cannot be certain of which adnexal tumor is which by clinical examination characteristics. Hence excision for histologic typing, as well as to head off later enlargement, is favored.

THERAPIES

Ablative surgery is the standard when technically feasible. After resection of numerous individual lesions, in some cases new tumors arise in the remaining scalp. Surgical scalping—the term is used by authors who have published on it^8-10—and skin grafting has produced excellent cosmetic and long-term results including in challenging locales such as the external ear.⁹ The role of Mohs microsurgery is not yet definitively settled.

Radiotherapy tends to be ineffective.^7-12 Laser removal has been employed with some success.¹³ Investigation of some topical salicylates relies on laboratory inhibition of a biochemical intermediary expression in vitro.¹²

Genetic investigation is warranted, and counseling becomes vital. The benefit to those who are shown to be free of the abnormal gene is tremendous: they no longer fear that every tiny papule represents a potential large mass on the head.

For those who carry the gene, regular skin surveillance should prevent the development of the kind of scalp displayed by the patient in this report. Such assurance might help affected family members feel what is only too easy for us to say: that this is not the worst familial affliction in the world, because it can be wrestled down to the level of a colossal nuisance rather than a destroyer of life and of quality of life.

Schneiderman H. Turban tumor (eccrine cylindroma, Brooke-Spiegler syndrome): genetic facets and dreadful potential of a multiple benign neoplasm. CONSULTANT. 2010;50:21-24.