—  SPECIALTY CONFERENCE  —

Neuropathology

Case 2 - Cylindroma

Bette K. Kleinschmidt-DeMasters
University of Colorado School of Medicine
Denver, CO





Virtual Slides as well as Still Images are displayed below.
For the fastest viewing of virtual slides, click:



under each thumbnail image below. You must have Aperio ImageScope installed on your PC.
If you do not already have Aperio ImageScope, Windows users with administrator privileges may download and install a free version in order to view USCAP Virtual Slides. Click the icon on the right to get your free copy:  
Or, click on slide thumbnail images to view each slide
in a Web-based slide viewer, which is somewhat slower.

If you have any difficulties viewing these slides, email or call George Clay at +1.724.449.1137.



Click here to download handout in 1-up pdf format for the current section (4.19 MB)

Click here to download handout in 6-up pdf format for the current section (1.01 MB)





Discussion

This boney invasive neoplasm on gross examination showed a white firm tumor mass traversing an identifiable strip of dura and the full thickness of the skull specimen. Low power microscopic examination revealed a hypercellular, nested tumor invading through bone and dura, leaving few intact boney trabeculae (Case 2-Figures 2, 3). No individual tumor cells were identified infiltrating the bone or dura; rather the tumor invaded as cohesive nests of cells (Case2-Figure 2). Microscopic brain invasion was not identified.


Case 2 - Slide 1
Click to view with ImageScope
Click to view with a Web-Based Viewer


Case 2 - Figure 1 – Sagittal magnetic resonance imaging (MRI) scan, with gadolinium, shows invasion of the tumor deep into the left parietal lobe. Note the overlying scalp portion is nearly equal in volume to the intracranial parts of the tumor.

Case 2 - Figure 2 – Low power photomicrograph shows the large and small basophilic nests of tumor cells invading through the skull, leaving only a few widely dispersed eosinophilic bone spicules and causing extensive fibrosis. Hematoxylin and eosin (H&E), 200 X.

Case 2 - Figure 3 – Low power photomicrograph of the tumor shows the characteristic islands of basaloid cells with peripheral palisading and surrounding hyaline bands. H&E, 200X.

Case 2 - Figure 4 – High power photomicrograph best highlights the fact that histological features of malignancy, such as loss of the eosinophilic hyaline sheath at the perimeter of the nests, nuclear pleomorphism, and loss of peripheral palisading, are NOT present in this skull-invasive tumor. H&E, 600X.


On higher power examination, the tumor was characterized by multiple, irregularly-shaped islands of basaloid cells surrounded by a distinct eosinophilic hyaline band (Case 2-Figure 4). Hyaline material was also identified within the cell nests (Case2-Figures 3, 5). Two cell types were identified within the cell nests: a peripheral cell with a dark staining nucleus that manifested a palisading pattern and a larger cell with a vesicular nucleus which was centrally located in the tumor nests (Case2-Figure 4). In a few areas, the tumor assumed a spiradenoma pattern with duct formation (Case2-Figure 4), a known variation of pattern in this tumor type that is without additional prognostic implications. [1]

Immunohistochemical staining for pancytokeratin demonstrated 3+ membranous staining. EMA (m, Dako) and S-100 (polyclonal, Dako) immunohistochemical staining of tumor cells was negative. Mitoses were exceedingly rare, but MIB-1 (monoclonal (m), Dako, Carpinteria, Ca.) cell cycle labeling index was slightly elevated, with positive nuclear staining in 4% of the basaloid tumor cells (1000 cell count). TP53 (m, Dako) immunostaining showed strong, diffuse 3+ immunostaining but was comparable to the staining intensity and distribution seen in five sporadic cases of cutaneous, non-bony invasive cylindromas which showed near equal, strong, diffuse, 3-4+ staining for TP53.

1. What is the diagnosis?
The diagnosis is cylindroma; no malignant features were identified.

Cylindromas are benign cutaneous adnexal neoplasms that usually occur sporadically and only rarely show transformation into cylindrocarcinoma. Sporadic cylindromas are most common in middle-aged to elderly women. Grossly, the tumors occur as tan-pink nodules with a predilection for the head and neck. [1] The key microscopic features of benign cylindroma are illustrated in this case and include irregularly shaped islands of basaloid cells arranged in a "jigsaw puzzle" pattern, separated a hyaline sheath. Two cell types can be identified: a small cell with a dark nucleus that tends to palisade around the periphery of the islands and a cell centrally located within the islands that shows a pale staining, vesicular nucleus. [1] Malignant transformation of cylindromas is defined by anaplasia and pleomorphism of the nuclei with increased mitotic figures, loss of the hyaline sheath, loss of peripheral palisading, and stromal invasion. [2] These features were conspicuously absent in this case, although the MIB-1 cell cycle labeling rate was slightly elevated. The pathological features seen in this case negate the diagnosis of meningioma, metastasis, invasive basal cell carcinoma, invasive squamous cell carcinoma, or invasive adenoid cystic carcinoma.

2. Is it common for cutaneous tumors to invade the skull?!?
Definitely not; cutaneous tumors of the scalp, even highly malignant variants such as basal cell carcinoma or squamous cell carcinoma, seldom invade the underlying skull and dura and seldom enter diagnostic consideration for clinicians, pathologists, or especially neuropathologists! This is a very unusual case of a scalp cylindroma without features of malignancy invading through the skull and dura and producing massive intracranial extension.

3. This patient sounds as if he has an autosomal dominant phakomatosis associated with benign cutaneous tumors-- and probably not one of the neurofibromatosis syndromes. Could this be Gorlin syndrome or Cowden syndrome?
Cowden syndrome is an autosomal dominant condition frequently caused by germline mutation of the PTEN/MMAC1 gene. [3] Patients manifest combinations of multiple trichilemmomas (85% of patients), oral papillomatosis, cutaneous keratoses, gastrointestinal polyps (40%), hamartomatous soft tissue lesions, thyroid tumors, benign breast tumors, breast cancers (30%), and craniomegaly. The most characteristic CNS manifestation is the dysplastic gangliocytoma of the cerebellum (Lhermitte-Duclos disease) but occasional meningiomas have been documented; these have not been reported to be aggressive, nor have the trichilemmomas. [3]

Gorlin syndrome (nevoid basal cell carcinoma syndrome) is also autosomal dominant caused by germline mutation of the PTCH gene and manifesting multiple basal cell carcinomas and jaw keratocysts; both features are found in more than 90% of affected patients by the age of 40 years. [4] Increased incidence of calcification of the falx cerebri is also seen (79% in patients >20 years). The most common CNS manifestation is medulloblastoma; most medulloblastomas develop in the first 3 years of life and most are the desmoplastic variant. It is estimated that up to 1-2% of all medulloblastomas are associated with Gorlin syndrome. Several examples of meningiomas (5% of patients [5]) have been documented but are not specifically aggressive. A few aggressive basal cell carcinomas in Gorlin syndrome have however been documented. [6, 7]

This patient had none of the multiple other features listed above for either Cowden or Gorlin syndrome; instead he had a significant past medical history of a basal cell adenoma/carcinoma of the parotid. His family history clearly suggested autosomal dominant inheritance pattern but they had had no other problems except for the multiple, histologically similar benign skin lesions. The histological features illustrated for this extra/intracranial mass excluded diagnoses of several of the likely tumors seen with Cowden or Gorlin syndromes, ie., trichilemmoma, basal cell carcinoma, and meningioma.

4. This patient perhaps had some even rarer syndrome—what is it?
This patient had familial cylindromatosis (Brooke-Spiegler syndrome, OMIM 132700, 605041), a rare autosomal dominant tumor syndrome characterized by cylindromas, trichoepitheliomas, spiradenomas, and basal cell adenomas and adenocarcinomas of the parotid glands and minor salivary glands . [8, 9, 10, 11]

He had had removal of a salivary gland tumor 12 years previously and had undergone numerous previous surgical excisions of his scalp "turban tumors" over a 30 year time period. In patients affected by familial cylindromatosis, the numerous cylindromas are typically on the head and neck and the trichoepitheliomas are generally located on the face, especially in the paranasal region. Scalp cylindromas may coalesce and become quite bulky, hence the name "turban tumor". The tumors generally begin to manifest during the first two decades of life, earlier than for sporadic cylindromas, and accumulate throughout life, often resulting in disfiguration.

5. Have other cases of intracranial invasion been reported for cutaneous cylindromas?
Yes; while an uncommon occurrence, both benign and malignant cylindromas have rarely been reported to invade bone. A literature review revealed several examples of malignant transformation with invasion of underlying skull/dura and intracranial extension, some of which were also definitely in patients with familial cylindromatosis [12, 13, 14] (with two more examples in the older German-language literature cited by reference 13).We were able to identify only one other case in the literature of a benign cylindroma invading the underlying skull and producing intracranial extension and this patient also suffered from familial cyndromatosis. [15]

6. What is known about the genetics and biology of familial cylindromatosis?
The syndrome is autosomal dominant with complete penetrance, but expression is variable, as is illustrated in this patient's family history. [10] His paternal grandmother was severely affected, with his father and brother less severely afflicted; thus far his 21 year old daughter does not manifest the features. The CYLD gene involved with the syndrome has been mapped to 16q12-q13. A clinically similar but apparently distinct disorder, multiple familial trichoepitheliomas (MFT), has been mapped to 9p21. Recently, however, a family presenting predominantly with trichoepitheliomas (and resembling the MFT phenotype) was found to have a novel missense mutation in the CYLD gene and was recognized to be a variant of the Brooke-Spiegler syndrome. [9] This, and other studies, suggests that the phenotypic range of expression in Brooke-Spiegler syndrome is broad, with considerable inter- and intra-familial heterogeneity. [8, 9, 11]

Loss of heterozygosity at the CYLD locus has also been identified in sporadic skin adnexal tumors such as apocrine hydrocystomas, eccrine spiradenomas, and sebaceous adenoma. [15] Cylindromas and spiradenomas are tumors of sweat gland origin, whereas trichoepithelioma is a tumor of hair follicle origin. This observation has lead many to believe that CYLD gene mutations result in dysregulation of the folliculo-sebaceous-apocrine stem cells. [8]

7. If this tumor was so benign, why did it invade his skull?
Cylindromas in familial cylindromatosis tumor syndrome may have a propensity to invade underlying skull due to their long-standing nature and multiplicity. There may also be an element of "patient neglect" in these few patients where the tumor has been skull invasive. [13]

In addition, the multiple previous surgical resections down to the level of the pericranium that these patients undergo to reduce the disfiguring effect of these innumerable tumors encircling their scalp may lead also to local thinness in the tissue and predispose to skull invasion. In our patient, there was a site of breakdown in one of his previous scalp skin grafts with focal boney skull destruction that was near to the site of tumor invasion. We suspect that previously damaged bone was particularly vulnerable to invasion by the tumor. Another plausible, but less likely, explanation for the presence of tumor cells within the bone of the skull is that ectopic rests of folliculo-sebaceous-apocrine stem cells were trapped during development. These ectopic rests might be at increased risk to transform into a benign cylindroma because of the patient's mutation of the CYLD gene.

8. Are there biological markers present in these tumors that might explain its aggressive behavior other than mitotic rate?
Normally the dura and skull serve as effective barriers to tumor penetration, either from tumors originating in the intracranial space (ie., meningiomas) or from cutaneous tumors originating in the overlying scalp. While basal cell carcinomas and squamous cell carcinomas of the head seldom are boney-invasive, some examples can be quite aggressive. A recent study demonstrated increased TP53 immunohistochemical expression in basal cell carcinomas that were invasive of skull base, compared to controls. [17] In our case of cylindroma invading the parasagittal regions of the skull, the relative rarity of the condition precludes a controlled study, but TP53 immunostaining was noted to be equally strong and diffuse in both our case and five cutaneous, non-invasive cylindromas. Hence, TP53 immunostaining does not appear to correlate with invasiveness in cylindromas, at least in our experience.

9. This is a pretty rare case—all there other types of cutaneous or head and neck tumors that the pathologist is more likely to see invading the skull?
This is a rare case and the additional feature that makes this case exceptional is the fact that the invasion occurred over the convexity of the skull.

The most common site for non-meningioma/non-neurogenic tumors to invade the skull is of course at the skull base. Tendency for boney invasiveness in some, but not all, tumor types can be related to histological subtype, differentiation, depth of invasion, and/or perineural invasion. Basal cell and squamous cell carcinomas of face and ear; paragangliomas of head and neck; olfactory neuroblastomas, nasopharyngeal carcinomas, and sinonasal undifferentiated carcinomas of the frontal sinuses; and mucoepidermoid carcinomas, adenocarcinomas, and adenoid cystic carcinomas of parotid or minor salivary gland all come to mind as tumor types that can manifest skull invasion.

Of all these various tumor types, adenoid cystic carcinomas in particular are important for the pathologist to know about; they are also interesting and challenging for neurosurgeons. [18] Although ACC constitutes only about 1% of head and neck malignancies, the reported incidence of cranial base invasion is 4-22%. Skull-invasive ACCs most commonly involve the infratemporal fossa, cavernous sinus, and middle fossa. [18] Many (69%) of patients with skull-base invasive ACCs already have evidence of their cranial base invasion by the time they first come to medical attention. [13] Remaining patients later develop skull base involvement months after the head and neck primary was diagnosed and treated with surgery, radiation, chemotherapy. Patients with involvement of the skull base by their ACC manifest with neurological symptoms including trigeminal neuralgia, diplopia, visual deficits, nasal symptoms, hearing alterations, and swallowing difficulties. Recurrence in ACCs has been associated with solid growth pattern (as opposed to cribriform and tubular) and positive surgical margin. Intracranial involvement by ACC may be due to either direct extension or hematogenous spread, although skull-base invasive tendency in ACCs is perhaps best linked to histological finding of perineural invasion. Rare patients show intracranial ACCs without a known extracranial primary.

Take Home Bullet Points:
  1. Cutaneous tumors seldom are skull-invasive and invasion is especially rare over the skull convexity. Basal cell carcinomas and squamous cell carcinomas are two cutaneous tumor types that can, however, manifest skull base invasion and this invasive tendency may be associated with increased expression of TP53.

  2. Adenoid cystic carcinomas of major and minor salivary glands show relatively frequently intracranial involvement. CNS involvement may occur via direct extension/perineural invasion or hematogenous spread.

  3. Patients with ACC may manifest neurological symptoms at the time of initial presentation, reflecting the ability of these tumors to develop skull-base invasion early in the clinical course. Rare patients show intracranial ACC without a documentable extracranial primary.

References
  1. Weedon D. Skin Pathology, 1997, Churchill Livingstone, Edinburgh, 713-758.

  2. Lin PY, Fatteh SM, Lloyd KM. Malignant transformation in a solitary dermal cylindroma. Arch Pathol Lab Med. 1987;111(8):765-7

  3. Wiestler OD, Padberg GW, Steck PA. Cowden disease and dysplastic gangliocytoma of the cerebellum/Lhermitte-Duclos disease. In: Kleihues P, Cavenee WK, editors. World Health Organization Classification of Tumours: Pathology & Genetics of Tumours of the Nervous System. New York, IARC Press: 2000

  4. Reifengerger G, Wiestler OD, Chenevix-Trench G. Naevoid basal cell carcinoma syndrome. In: Kleihues P, Cavenee WK, editors. World Health Organization Classification of Tumours: Pathology & Genetics of Tumours of the Nervous System. Lyon, France, IARC Press: 2000

  5. Kimonis VE, Mehta SG, Digiovanna JJ, Bale SJ, Pastakia B. Radiological features in 82 patients with nevoid basal cell carcinoma (NBCC or Gorlin) syndrome. Genetics in Medicine. 6(6):495-502, 2004

  6. Lasso JM, Garcia-Tutor E, Bazan A. Aggressive basal cell carcinoma of the temporal region in a patient with Gorlin-Goltz syndrome. Annals of Plastic Surgery. 44(4):429-434, 2000

  7. Nasir S, Kayikcioglu A, Aydin MA, Tuncbilek G. Coverage of the cranium with latissimus dorsi in the recurrent basal cell carcinoma of Gorlin syndrome: for protection against clinical invasion. Journal of Craniofacial Surgery. 17(3):599-602, 2006

  8. Bowen S. Gill M. Lee DA. Fisher G. Geronemus RG. Vazquez ME. Celebi JT. Mutations in the CYLD gene in Brooke-Spiegler syndrome, familial cylindromatosis, and multiple familial trichoepithelioma: lack of genotype-phenotype correlation. Journal of Investigative Dermatology. 124(5):919-20, 2005

  9. Hu G, Onder M, Gill M, Aksakal B, Oztas M, Gurer MA, Celebi JT. A novel missense mutation in CYLD in a family with Brooke-Spiegler syndrome. Journal of Investigative Dermatology. 121(4):732-4, 2003

  10. Poblete Gutierrez P, Eggermann T, Holler D, Jugert FK, Beermann T, Grussendorf-Conen EI, Zerres K, Merk HF, Frank J. Phenotype diversity in familial cylindromatosis: a frameshift mutation in the tumor suppressor gene CYLD underlies different tumors of skin appendages. Journal of Investigative Dermatology. 119(2):527-31, 2002

  11. Puig L, Nadal C, Fernandez-Figueras MT. Alegre M. de Moragas JM. Brooke-Spiegler syndrome variant: segregation of tumor types with mixed differentiation in two generations. American Journal of Dermatopathology. 20(1):56-60, 1998

  12. Durani BK. Kurzen H. Jaeckel A. Kuner N. Naeher H. Hartschuh W. Malignant transformation of multiple dermal cylindromas. British Journal of Dermatology. 145(4):653-6, 2001

  13. Urbanski SJ. From L. Abramowicz A. Joaquin A. Luk SC. Metamorphosis of dermal cylindroma : possible relation to malignant transformation. Case report of cutaneous cylindroma with direct intracranial invasion . Journal of the American Academy of Dermatology. 12(1 Pt 2):188-95, 1985

  14. Volter C, Baier G, Schwager K, Muller JG, Rose C. Cylindrocarcinoma in a patient with Brooke-Spiegler syndrome. Layryngo-Rhino-Otologie 81(3):243-6, 2002.

  15. Wyld L. Bullen S. Browning FS. Transcranial erosion of a benign dermal cylindroma . Annals of Plastic Surgery. 36(2):194-6, 1996

  16. Leonard N, Chaggar R, Jones C, Takahaski M, Nikitopoulou A, Lakhani SR. Loss of heterozygosity and cylindromatosis gene locus, CYLD, in sporadic skin adnexal tumours. J Clin Pathol 2001. 54: 689-92.

  17. Cernea CR. Ferraz AR. de Castro IV. Sotto MN. Logullo AF. Bacchi CE. Potenza AS. p53 and skin carcinomas with skull base invasion: a case-control study. Otolaryngology - Head & Neck Surgery. 134(3):471-5, 2006.

  18. Gormley WB, Sekhar LN, Wright DC, Olding M, Janecka IP, Snyderman CH, Richardson R. Management and long-term outcome of adenoid cystic carcinoma with intracranial extension: a neurosurgical perspective. Neurosurgery 38(6):1105-1113, 1996.