The riddle of multinucleated “floret-like” giant cells and their detection in an extensive gluteal neurofibroma: a case report
© Stanger et al.; licensee BioMed Central Ltd. 2013
Received: 19 December 2012
Accepted: 7 May 2013
Published: 26 July 2013
The neurofibromatoses are inherited tumor predisposition syndromes involving two major clinical phenotypes: neurofibromatosis type 1 (von Recklinghausen's disease) is linked to chromosome 17q, and tends to occur seven times more frequently than neurofibromatosis type 2. Neurofibromatosis type 1 entails a distinctive cutaneous manifestation prevailed upon by benign neurofibromas, which may vary in size, number and distribution. On the histological level, neurofibromas are composed of an admixture of neurilemmal cells, including Schwann cells, fibroblasts, and – to a lesser extent – perineurial cells.
The case of a 39-year-old Caucasian man with a voluminous recurrent neurofibroma of 27×15cm extending from the left gluteal region to thoraco-lumbar levels Th6 through L4 is reported. Within the soft tissue tumor a pseudocyst of 7.3×9.3cm was found preoperatively.
Histopathological study of the excised mass was conspicuous for revealing a large number of multinucleated floret-like giant cells within an otherwise classical soft tissue neurofibroma.
Previous reports on neurofibromas with multinucleated floret-like giant cells are distinctly scant. Available evidence from the literature does not suggest any consistent correlation of multinucleated floret-like giant cells in neurofibromas with gender, age, traumatic antecedents, size of the lesion, recurrence, or malignant transformation. Furthermore, the presence of such cells may not be specific for neurofibromatosis type 1, as they occasionally are encountered in some unrelated mesenchymal neoplasms as well.
KeywordsFloret-like multinucleated giant cells Neurofibroma Neurofibromatosis type 1
Neurofibromatosis type 1 (NF1) is a dysgenetic complex which is due to a loss of function of the tumor-suppressor gene product neurofibromin located on chromosome 17q . Mutations therefor exert a pleiotropic effect on derivatives of the neural crest including embryonic misplacement and deregulated proliferation of Schwann cells, perineural cells, and melanocytes . The condition is autosomal dominant with variable penetrance and exhibits a high degree of de novo mutations . NF1 exhibits a conspicuous cutaneous phenotype including benign neurofibromas, hyperpigmented macules ("café au lait" spots), axillary and inguinal freckling, as well as pigmented hamartomas of the iris (Lisch nodules), and distinctive osseous lesions [4, 5]. Variable in number and distribution, neurofibromas are painless, slow-growing tumors. Their transformation into malignant peripheral nerve sheath tumors will occur in approximately 7 to 13% of the affected individuals [6, 7]. Most of these malignant tumors tend to evolve from plexiform lesions, males being at a higher risk for such evolution . Early diagnosis based on the clinical phenotype, genetic counseling, treatment of symptoms, and appropriate surgeries are the tools available for management of patients with NF1 . Surgical procedures generally involve resection, combined with reconstructive strategies aimed at functional and aesthetic improvement . Although the majority of neurofibromas can be excised completely by a subcutaneous approach, and closed primarily, surgery of extensive neurofibromas (massive soft tissue neurofibromas) can be challenging due to their high vascularity [3, 8–10].
Although the histological identification of the NF1-associated neurofibromas is rather straightforward in practical terms, these tumors are notorious for exhibiting microscopic details that are apt to confound correct morphological diagnosis in an individual. The presence of so-called floret-like multinucleated giant cells (FMGCs) is one such detail .
We describe an intriguing case of an extensive lumbar and gluteal neurofibroma with FMGCs in a patient with NF1 and review the literature to investigate the association of FMGCs with NF1.
The patient underwent wide partial excision (20×12cm) of the neurofibroma under general anesthesia. Because of the presence of irregular frail and plexus-like veins, subtle hemostasis was mandatory throughout the procedure. Direct wound closure could be achieved by advancement and rotation of the adjacent skin. Seven weeks following surgery, there was significant reduction of the left paravertebral and gluteal bulging, and an improved symmetry of the lumbar and gluteal regions (Figure 1B). The patient also reported a reduction in the localized pain he had experienced prior to resection and remained satisfied at 1 year follow-up, with no recurrence of the neurofibromatous mass.
Following overnight fixation in 10% buffered formalin, the resection specimen was routinely processed to paraffin, and 3μm-thick slides were stained with hematoxylin and eosin for microscopic study. Immunohistochemistry was performed with the following panel of antibodies: vimentin (clone Vim 3B4; Dako, Glostrup, Denmark), CD34 (clone QBend/10; Dako), S100 protein (polyclonal; Dako), smooth muscle actin (clone 1A4; Sigma, St. Louis, MO, United States of America), epithelial membrane antigen (EMA; clone E29; Dako), glucose transporter 1 (GLUT-1; polyclonal; Abcam, Cambridge, United Kingdom), according to established protocols in our laboratory. Slides were developed with polymer-bound horseradish peroxidase (Envision+; Dako) and 3,3'-diaminobenzidine as chromogen.
The etiology of FMGCs is unknown [11, 12]. We are aware of only four previous case reports on FMGCs in neurofibromas arising in the setting of NF1 [5, 13–15], as well as a recently published single retrospective clinicopathologic study of 22 patients with NF1 and FMGCs . Magro et al. raise the question whether this phenomenon is merely an uncommon incidental finding or if it is pathognomonic for NF1 [11, 12, 14]. These authors suggest that FMGCs in an otherwise typical neurofibroma may not be indicators of NF1, but more probably represent reactive change of the indigenous dermal or endoneurial fibroblasts or dendritic cells in response to some unknown microenvironmental stimulus . In their series of 94 neurofibromas, only 23% of the cases contained FMGCs, in 17 of these cases FMGCs cells were only identified after meticulous search. Only in 5% of the series were FMGCs found to be numerous and readily identifiable. With regard to clinical features, no correlation was found between FMGCs and gender (p=0.59) or age (<45 versus ≥45; p=0.43) .
Interestingly, FMGCs have also been detected in other tissues, including breast tissue in gynecomastia [13, 16]. They were further described in uncommon soft tissue tumors including pleomorphic lipoma, giant cell collagenoma, giant cell fibroblastoma and giant cell angiofibroma, pseudoangiomatous stromal hyperplasia of the male breast in NF1, as well as in a breast hamartoma from a 6-year-old boy with NF1 and in a 50-year-old woman without any signs of NF1 [12, 14, 16, 17].
Swick  concluded that FMGCs in neurofibromas should not be misinterpreted as harbingers of malignant transformation. Instead, they may represent a subpopulation of non–Schwannian S100 protein- and CD34+ positive cells similar to CD34-positive endoneurial dendritic cells found in normal nerves and neurofibromas. Negativity of these cells for S100 protein combined with the absence of either cytologic atypia or mitoses assist in ruling out actual malignant transformation. FMGCs are also seen in other soft tissue tumors “not” associated with NF1: the presence of FMGCs in these neoplasms support the hypothesis that FMGCs are not specific to neurofibromas or to hyperplastic lesions associated with NF1 but, instead, represent non-neoplastic supporting cells in some tumors and true neoplastic cells in other tumors .
Shaktawat and Golka recommend these cells to be interpreted carefully, keeping in mind the rare malignant change in neurofibromas . According to Shaktawat and Golka these are positive with vimentin and CD34 and negative with S100 and CD68 . Taungjaruwinai and Goldberg, Swick and Magro et al. described in their cases the S100 negative FMGCs [5, 13, 15]. In the study by Shaktawat and Golka the FMGCs were also negative for the macrophage marker CD68 . Multinucleated cells have also been known as part of the spectrum of “degenerative atypias” in neurofibromas.
The issue of possible gender predilection for FMGCs to appear in NF1-associated neurofibromas remains unclear as well.
In the case presented here, FMGCs were found in a man with NF1. Satter observed that such a constellation may be specific for males, pointing out that all recorded cases involved male patients [5, 12–15]. Conversely, 53% of the study cohort in the series by Magro et al. was female . It is not clear, however, to what respective extent male and female patients were represented among those 8% of patients in whom FMGCs actually were seen .
In one case report, the hypothesis has been put forward that trauma may trigger the emergence of FMGCs . Hypoxia and/or reparative changes were hypothesized to have contributed to the induction of FMGCs [12, 15]. Therefore it remains unclear whether recurrence, cellular stress or injury play a role in the development of FMGCs [12, 13, 16].
Furthermore, the size of the neurofibroma might also be a stimulating factor for the appearance of FMGCs. In our case we found a neurofibroma of 27×15cm. The literature generally describes these cells in much smaller lesions. Taungjaruwinai and Goldberg describe FMGCs in neurofibromas with an average size of 0.6cm . Swick et al. mentioned a neurofibroma inferior to 0.8cm . Magro et al. saw a neurofibroma with 10cm diameter where they reportedly found FMGCs; whereas Shaktawat and Golka found FMGCs in a neurofibroma of 1.2 cm [11, 13, 14].
Having discussed earlier that FMGCs are not specific for NF1, and may appear also in other tissue, further studies are warranted to clarify the etiopathogenesis of FMGCs together with possible association of long-term risk of neoplastic changes in NF1 [5, 11–13]. Complete or partial surgical removal of the lesion is indicated for symptom alleviation and histopathological diagnosis of the tumor [9, 18].
Written informed consent was obtained from the patient for publication of this manuscript and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.
- Viskochil D: Genetics of neurofibromatosis 1 and the NF1 gene. J Child Neurol. 2002, 17: 56.View ArticleGoogle Scholar
- Ball N, Kho G: Melanocytic nevi are associated with neurofibromas in neurofibromatosis type 1, but not sporadic neurofibromas: a study of 226 cases. J Cutan Pathol. 2005, 32: 523. 10.1111/j.0303-6987.2005.00376.x.View ArticlePubMedGoogle Scholar
- Renshaw A, Borsetti M, Nelson R, Orlando A: Massive plexiform neurofibroma with associated meningo-encephalocoele and occipital bone defect presenting as a cervical mass. Br J Plast Surg. 2003, 56: 514-517. 10.1016/S0007-1226(03)00151-6.View ArticlePubMedGoogle Scholar
- Gerber P, Antal A, Neumann N, Homey B, Matuschek C, Peiper M, Budach W, Bölke E: Neurofibromatosis. Eur J Med Res. 2009, 14: 102-105.PubMedPubMed CentralGoogle Scholar
- Taungjaruwinai W, Goldberg L: Multinucleate giant cells in neurofibromas: a clue to the diagnosis of neurofibromatosis. J Cutan Pathol. 2009, 36: 1164-1167. 10.1111/j.1600-0560.2009.01249.x.View ArticlePubMedGoogle Scholar
- Lin B, Weiss L, Medeiros L: Neurofibroma and cellular neurofibroma with atypia: a report of 14 tumors. Am J Surg Pathol. 1997, 21: 1443-10.1097/00000478-199712000-00006.View ArticlePubMedGoogle Scholar
- Shack R, Reilley A, Lynch J: Neurofibromas of the head and neck. South Med J. 1985, 78: 801-804. 10.1097/00007611-198507000-00008.View ArticlePubMedGoogle Scholar
- Spira A, Riccardi V: Neurofibromatosis. Clin Plast Surg. 1987, 14 (2): 315-325.PubMedGoogle Scholar
- Bloem J, Van der Meulen J: Neurofibromatosis in plastic surgery. Br J Plast Surg. 1978, 31 (1): 50-53. 10.1016/0007-1226(78)90015-2.View ArticlePubMedGoogle Scholar
- McClatchey A: Neurofibromatosis. Annu Rev Pathol Mech Dis. 2007, 2: 191-216. 10.1146/annurev.pathol.2.010506.091940.View ArticleGoogle Scholar
- Magro G, Amico P, Vecchio G, Caltabiano R, Castaing M, Kacerovska D, Kazakoy D, Michal M: Multinucleated floret-like giant cells in sporadic and NF-1 associated neurofibromas: a clinicopathologic study of 94 cases. Virchows Arch. 2010, 456: 71-76. 10.1007/s00428-009-0859-y.View ArticlePubMedGoogle Scholar
- Satter E: Floret-like multinucleated giant cells in a neurofibroma outside the context of neurofibromatosis type 1. Am J Dermatopathol. 2009, 31: 724-725. 10.1097/DAD.0b013e3181a58275.View ArticlePubMedGoogle Scholar
- Magro G, Scavo S, Ruggieri M: Floretlike multinucleated giant cells in a neurofibroma from a patient with NF-1: an unusual finding for such a tumor. Virchows Arch. 2002, 441: 525-526. 10.1007/s00428-002-0669-y.View ArticlePubMedGoogle Scholar
- Shaktawat S, Golka D: Floret-like multinucleated giant cells in neurofibroma. Diagn Pathol. 2007, 2: 47. 10.1186/1746-1596-2-47.View ArticlePubMedPubMed CentralGoogle Scholar
- Swick B: Floret-like multinucleated giant cells in a neurofibromatosis type 1- associated neurofibroma. Am J Dermatopathol. 2008, 30: 632-634. 10.1097/DAD.0b013e318174e73f.View ArticlePubMedGoogle Scholar
- Lipper S, Wilson C, Copeland K: Pseudogynecomastia due to neurofibromatosis – a light microscopic and ultrastructural study. Hum Pathol. 1981, 12: 755-759. 10.1016/S0046-8177(81)80180-3.View ArticlePubMedGoogle Scholar
- Agabiti S, Gurrera A, Amico P, Vasquez E, Magro G: Mammary hamartoma with atypical stromal cells: a potential diagnostic dilemma. Pathologica. 2007, 99: 434-437.PubMedGoogle Scholar
- Prokopakis E, Raissaki M, Bourolias C, Karatzanis A, Velegrakis G: Massive plexiform neurofibroma and spinal deformity presenting as dysphagia. Am J Otolaryngol Head Neck Med Surg. 2007, 28: 280-283.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.