Brain Tumor Pathol DOI 10.1007/s10014-014-0182-4

CASE REPORT

Comparison of magnetic resonance imaging with invasive histological findings of Langerhans cell histiocytosis Shinya Watanabe • Tetsuya Yamamoto • Kaishi Satomi • Masahide Matsuda • Hiroyoshi Akutsu • Eiichi Ishikawa • Akira Matsumura

Received: 17 September 2013 / Accepted: 14 February 2014 Ó The Japan Society of Brain Tumor Pathology 2014

Abstract Due to the rarity of skull Langerhans cell histiocytosis (LCH), correlations between abnormalities on magnetic resonance imaging (MRI) and pathological findings have not been fully evaluated. Only a few reports have as yet compared MRI of LCH infiltration and histological diagnostic features. A 22-year-old man with an unremarkable past medical history presented with headache followed by a nodular mass in the right parietal area. Gadolinium (Gd)-enhanced T1-weighted images showed a ‘‘mushroom-shaped’’ extension, extending from the subcutaneous to the intracranial space. The galeal tissue, the dura mater, and the cranium adjacent to the lesion showed enhancement on Gd-enhanced T1-weighted MRI. Histologically, diagnosis was LCH, with immunohistochemical positivity for S-100, CD68, and CD1a, and showed a MIB1 labeling index of 29.5 %. By immunostaining, we confirmed LCH cells to be present in the galea with Gd enhancement and the skull without Gd enhancement on MRI. LCH cells might exist in the marginal cranium, if MRI shows only slight intensity change without Gd enhancement. Further evaluation to determine pathologically proven extent of LCH and its relation with neuroradiological diagnosis as well as clinical outcome is needed

S. Watanabe  T. Yamamoto (&)  M. Matsuda  H. Akutsu  E. Ishikawa  A. Matsumura Department of Neurosurgery, Faculty of Medicine, Graduate School of Comprehensive Human Sciences, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-8575, Japan e-mail: [email protected] K. Satomi Department of Diagnostic Pathology, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8575, Japan

to understand heterogeneous response to surgical and/or nonsurgical treatment in LCH. Keywords Langerhans cell histiocytosis  Skull  Invasion  MRI  Immunostaining  MIB-1

Introduction Langerhans cell histiocytosis (LCH) is characterized by clonal proliferation of histiocytes (Langerhans cells) [1] with CD1a and/or Langerin (CD207)-positive Birbeck granules [2]. LCH is a rare disease, with an estimated incidence of 3–5 pediatric cases per million/year and 1–2 adult cases per million/year [3, 4]. The incidence appears to be higher in whites of northern European descent than in blacks [5], but is unknown in Asians. The clinical presentation of patients with LCH varies depending on the sites and extent of involvement. The disease is limited to one organ system in approximately 55 percent of patients, while the remainder present with multisystem diseases [6]. Unifocal LCH was reported in bone, skin, lymph nodes, liver, spleen, oral mucosa, lungs, and the central nerve system [6]. The skull is the most frequent site of unifocal LCH. With skull LCH, adjacent tissue sometimes shows intensity change or gadolinium (Gd) enhancement on magnetic resonance imaging (MRI). However, due to the rarity of this pathology, the correlations between abnormalities on MRI and pathological findings have not been fully evaluated. Only a few reports have as yet compared MRI of LCH infiltration with histological diagnostic features [7–9]. Thus, in this report, we present a patient who underwent removal of a skull LCH together with the adjacent tissue (galea, cranium, subcutaneous tissue) which

123

Brain Tumor Pathol

Fig. 1 Initial and preoperative magnetic resonance imaging or computed tomography, and intraoperative findings. Computed tomography (CT) (a) demonstrates an osteolytic lesion 28 mm in diameter occupying the space between the outer and inner plate of the right parietal bone expanding into the intracranial space. T1-weighted (b) images show a tumor with low intensity and T2-weighted (c) images with mixed low/high intensity, while the adjacent cranium shows low intensity on T1-weighted images. Gadolinium (Gd)enhanced T1-weighted images with fat suppression (d) show a

‘‘mushroom-shaped’’ tumor extension with a heterogeneous enhancement effect, from the subcutaneous to the intracranial space involving galeal tissue, with marginal skull and dural enhancement. One month later, the skull tumor showed shrinkage and marginal skull loss enhancement with an intensity change on Gd-enhanced fat suppression T1-weighted images (e). Intraoperatively, dull-red tumor was exposed at the time of skin flap reversal (f), xanthomatous dura mater was exposed after removal of the skull tumor (g), and two pieces of cranium were removed (h)

showed intensity change or Gd enhancement on MRI. The lesion and adjacent tissue were examined pathologically and compared with the abnormal findings on MRI.

no evidence of periosteal reaction or bone sclerosis. Thoracoabdominal CT demonstrated no abnormal findings in other organs, including the lungs, pancreas, and spleen. There was no abnormal uptake on gallium scintigraphy. On MRI, the lesion showed low intensity on T1weighted images and mixed low/high intensity on T2weighted images with a heterogeneous enhancement effect on Gd-enhanced T1-weighted images (Fig. 1b–d), representing the ‘‘mushroom-shaped’’ extension into the subcutaneous and intracranial space. The galeal tissue, the dura mater, and the cranium adjacent to the tumor showed enhancement on Gd-enhanced T1-weighted MRI. The skull lesion subsequently showed shrinkage with loss of enhancement in the marginal cranium but an intensity change on Gd-enhanced T1-weighted images with fat suppression (Fig. 1e). Based on the findings, LCH with invasion to adjacent tissues was suspected. The differential diagnosis included metastatic tumor and myeloma. For the purpose of tissue diagnosis and removal of the suspected invasive lesion, the extent of surgery was planned to cover all of the tissue

Case report Clinical summary A 22-year-old man with an unremarkable past medical history presented to a hospital with headache in the right parietal area. The initial head MRI revealed no abnormalities. One month later, he noticed a nodular mass lesion in the same painful region. He was then referred to our hospital and was diagnosed with a right parietal skull tumor. Examinations revealed a painful right parietal rubber mass but no neurological deficits or blood testing abnormalities. Computed tomography (CT) demonstrated an osteolytic lesion 28 mm in diameter occupying the space between the outer and inner plate of the right parietal bone extending into the intracranial space (Fig. 1a). There was

123

Brain Tumor Pathol

Fig. 2 Pathological findings of the skull lesion showing numerous Langerhans cells with scattered eosinophils and small lymphocytes in hematoxylin-eosin stain; inset is high magnification of Langerhans cells with folded nucleus (a). Langerhans cells are positive for CD68 (b) and CD1a (c). Most part of the fragile inner cranium and the stiff outer cranium did not contain Langerhans cells but only macrophages and lymphocytes infiltrated into them. But region around surgical

defect, the cranium was infiltrated by Langerhans cells (d) which were positive for CD68 (e) and CD1a (i) using immunohistochemistry. The dura shows no infiltration of Langerhans cells (g) and CD68-positive cells (h), although a few cells were positive for CD1a (i). But we regard those CD1a-positivity cells were not tumor cells, and the staining was nonspecific reactivity. Scale bar 100 lm

showing intensity change or Gd enhancement on MRI. Intraoperative observation using neuronavigation confirmed a dull-red mass, overlying thickened discolored galeal tissue (Fig. 1f), fragile cranium which showed altered intensity on MRI, and Gd-enhanced dura mater attached to the tumor on MRI (Fig. 1g). En bloc resection was performed for the mass with the adjacent tissue showing abnormal MRI findings (Fig. 1h). Postoperative MRI revealed no residual tumor. Postoperative course was uneventful, and he was discharged from the hospital 9 days after the surgery. No recurrence has been found for 18 months. Furthermore, CT scans have confirmed new bone formation.

Pathological findings The surgical specimens were categorized into 5 groups for further pathological diagnosis and to identify correlations with MRI findings using neuronavigation: skull lesion, galeal tissue, fragile inner cranium, stiff outer cranium, and dura. The skull lesion consisted of diffusely infiltration of histiocyte-like cells with a distinct cell margin, eosinophilic cytoplasms, and nuclear grooves and intranuclear inclusions. Also, various inflammatory cells such as lymphocytes, neutrophils, macrophages, and eosinophils were also infiltrated (Fig. 2a). Furthermore, the histiocyte-like cells were immunohistochemically positive for S-100,

123

Brain Tumor Pathol

CD68 (Fig. 2b), and CD1a (Fig. 2c). The final histological diagnosis was LCH. The MIB-1 index was 29.5 %. In addition to hematoxylin-eosin (HE) staining, Langerhans cell infiltrated into galeal tissue and was positive for CD68 and CD1a using immunohistochemistry. Histiocyte-like cells or multinucleated cells were also observed, and lymphocytes, neutrophils, and eosinophils were infiltrated. Most part of the fragile inner cranium and the stiff outer cranium did not contain Langerhans cells. Some macrophages and lymphocytes infiltrated into them. But region around surgical defect, the cranium was infiltrated by Langerhans cells (Fig. 2d), and they were positive for CD68 (Fig. 2e) and CD1a (Fig. 2i) using immunohistochemistry. The dura showed mild infiltration of lymphocytes but no infiltration of Langerhans cells (Fig. 2g) and CD68-positive cells using immunohistochemistry (Fig. 2h). Although a few cells seemed positive immunoreaction for Cdna (Fig. 2i), they were negative for neither CD68 (Fig. 2h) nor S-100 (data not shown). Immunohistochemistry for Langerin was not performed.

Discussion Clinical presentations of skull LCH are highly variable, ranging from one lesion as single-system involvement, generally benign, to one or more skull lesions constituting multisystem life-threatening disease [10]. Due to difficulties in understanding the pathogenesis of rare and varying LCH, optimal therapy remains controversial. Patients with skull LCH reportedly have good outcomes after total surgical removal [11, 12]. With a unifocal skull lesion, good outcomes after partial removal have also been reported [13], and spontaneous remission is occasionally seen [14– 16]. On the other hand, 0–30 % of postoperative skull LCH reportedly relapses on long-term follow-up [11, 17]. Kawamoto et al. [18] stressed the need for tumor removal with a sufficient margin because tumor cells often invade the adjacent cranium. However, preoperative prediction of tumor invasion in skull LCH using MRI and/or CT has not been established, and diagnosis consequently depends on intraoperative macroscopic observation and rapid pathology. Beltran et al. reported, based on 16 cases with pathologically proven eosinophilic granuloma, that the most common MR appearance was a focal lesion, surrounded by an extensive, ill-defined bone marrow and soft tissue reaction with low signal intensity on T2-weighted images [19]. In addition, they speculated that this intensity change on MRI represents bone marrow and soft tissue edema [19]. As for the adjacent tissue of our case, the galeal tissue and the cranium were positive for CD1a and CD68. The dura showed no infiltration of Langerhans cells, but a few CD1a-positive

123

Table 1 Summary of reports comparing MR imaging findings with histological tumor invasion Author

Tissue

MR imaging

Tumor invasion

Yunoki et al. [7]

Dura

CE (?)

Invasion (-), IC (?)

Takeuchi et al. [8]

Kamoshima et al. [9] Our case 2013

Dura

CE (?)

Invasion (?)

Galea

CE (?)

Invasion (-), IC (?)

Muscle

CE (?)

Invasion (?)

Dura

CE (?)

Galea

CE (?)

Invasion (-), IC (?)

Skull

Intensity change (?)

Invasion (?) Invasion (?)

CE contrast enhancement, IC inflammatory cells

cells were observed. But we regard those CD1a-positivity cells were not tumor cells, and the staining was nonspecific reactivity, because the cell shape unlike Langerhans cell or histiocyte and other Langerhans cell markers, such as S-100 and CD68, were negative. On the basis of these HE staining and immunostaining findings, the dural enhancement on MRI did not represent Langerhans cell invasion, while galeal enhancement on MRI did indicate actual Langerhans cell invasion. Furthermore, the marginal cranium showing a slight intensity change on MRI without Gd enhancement indicated Langerhans cell invasion of the inner, but not the outer, cranium. There are only three previous reports referring to the correlation, between MRI findings and pathology in skull LCH, and even these are inconsistent, especially as regards adjacent tissues. Table 1 summarizes reports including our case comparing MRI with histological Langerhans cell invasion findings. Yunoki et al. [7] reported an operative case with LCH of the skull. In their report, there was no dural enhancement on MRI indicative of Langerhans cell invasion, but inflammatory cells were present. Takeuchi et al. reported [8] an operative case with LCH of the skull with dural invasion. In their report, dural enhancement on MRI indicated Langerhans cell invasion, but there was no galeal enhancement on MRI. Kamoshima et al. [9] reported LCH of the temporal bones with infiltration of temporal muscle. In their report, temporal muscle enhancement on MRI demonstrated Langerhans cell invasion. The Gd enhancement on MRI is generally considered to represent tumor invasion in the majority of intra- and extraaxial tumors. Nontumoral Gd enhancement of soft tissues adjacent to a tumor is a well-known phenomenon. For example, the ‘‘dural tail sign’’ is one of the nontumoral enhancement on MRI of meningioma. In some reports on LCH, in the case of Gd-enhanced tissue, inflammatory cells were present, but tumor cells were not (Table 1).

Brain Tumor Pathol

Furthermore, to the best of our knowledge, this is the first report to compare the MRI changes of the skull with pathological findings. In our case, slight intensity change on T1-weighted images showed LCH invasion, despite the absence of Gd enhancement. It is noteworthy that Langerhans cells were present in the skull with only slight intensity change but no Gd enhancement on MRI. The correlation between MRI and pathological findings might be important in some LCH cases with relatively rapid and aggressive clinical courses. Reznik et al. [20] reported a skull LCH showing remarkable brain invasion. In these cases, high Ki-67 LI (6.2–20 %) may correlate with a poor prognosis [21, 22].

Conclusion In conclusion, we have described a young patient with LCH who presented with a right parietal nodular lesion. We were able to compare MRI with the histological findings of skull LCH in terms of invasion in this case and found that LCH cells might exist in the marginal cranium, if MRI shows only slight intensity change without Gd enhancement. Further evaluation to determine pathologically proven extent of LCH and its relation with neuroradiological diagnosis as well as clinical outcome is needed to understand heterogeneous response to surgical and/or nonsurgical treatment in LCH. Acknowledgments The authors thank Bierta E Barfod, Katsuta Hospital Mito Gammahouse, for the revision and translation, and Tsukuba Central Hospital, for the follow-up treatment of this patient. The authors have no sources of financial and material support to declare. Conflict of interest exists. Ethical standards patient.

No actual or potential conflict of interests

Written informed consent was obtained from the

References 1. Favara BE, Steele A (1997) Langerhans cell histiocytosis of lymph nodes: a morphological assessment of 43 biopsies. Pediatr Pathol Lab Med 17:769–787 2. Beverley PC, Egeler RM, Arceci RJ, Pritchard J (2005) The Nikolas Symposia and histiocytosis. Nat Rev Cancer 5:488–494 3. Carstensen H, Ornvold K (1993) The epidemiology of Langerhans cell histiocytosis in children in Denmark, 1975-89. Med Pediatr Oncol 21:387 4. Baumgartner I, von Hochstetter A, Baumert B et al (1997) Langerhans’-cell histiocytosis in adult. Med Pediatr Oncol 28:9 5. Jaffe R, Weiss LM, Fachetti F (2008) Tumours derived from Langerhans cell. In: Wwerdlow SH, Campo E, Harris NL (eds) World health Organization classification of tumours of haematopoietic and lymphoid tissues. IARC Press, Lyon, p 358

6. Grois N, Po¨tschger U, Prosch H, Minkov M, Arico M, Braier J, Henter JI, Janka-Schaub G, Ladisch S, Ritter J, Steiner M, Unger E, Gadner H, DALHX- and LCH I and II Study Committee (2006) Risk factors for diabetes insipidus in langerhans cell histiocytosis. Pediatr Blood Cancer 46:228–233 7. Yunoki M, Hirashita K, Gohda Y, Yoshino K, Fujimoto S (2007) An operative case of eosinophilic granuloma of the skull: case report—special emphasis on surgical procedure. Curr Pract Neurosurg 17:358–364 (in Japanese with English abstract) 8. Takeuchi S, Takasato Y, Masaoka H, Hayakawa T, Otani N, Yoshino Y, Yatsushige H (2008) An operative case of eosinophilic granuloma of the skull with dural invasion. No Shinkei Geka 36:239–243 (in Japanese with English abstract) 9. Kamoshima Y, Sawamura Y, Sugiyama T, Yamaguchi S, Iwasaki Y, Kubota K (2008) Langerhans cell histiocytosis of the temporal bones with infiltration of temporal muscle: a case report. No Shinkei Geka 36:435–439 (in Japanese with English abstract) 10. Abla O, Egeler RM, Weitzman S (2010) Langerhans cell histiocytosis: current concepts and treatments. Cancer Treat Rev 36:354–359 11. Rawlings CE 3rd, Wilkins RH (1984) Solitary eosinophilic granuloma of the skull. Neurosurgery 15:155–161 12. Sampson JH, Rossitch E Jr, Young JN, Lane KL, Friedman AH (1992) Solitary eosinophilic granuloma invading the clivus of an adult: case report. Neurosurgery 31:755–757 13. Key SJ, O’Brien CJ, Silvester KC, Crean SJ (2004) Eosinophilic granuloma: resolution of maxillofacial bony lesions following minimal intervention. Report of three cases and a review of the literature. J Craniomaxillofac Surg 32:170–175 14. Ladisch S, Gadner H (1994) Treatment of Langerhans cell histiocytosis—evolution and current approaches. Br J Cancer Suppl 23:41–46 15. Muscolo DL, Slullitel G, Ranalletta M, Aponte-Tinao LA, Ayerza MA (2003) Spontaneous remission of massive solitary eosinophilic granuloma of the femur. J Pediatr Orthop 23:763–765 16. Oliverira M, Steinbok P, Wu J, Heran N, Cochrane D (2003) Spontaneous resolution of calvarial eosinophilic granuloma in children. Pediatr Neurosurg 38:247–252 17. Carangelo B, Peri G, Tacchini D, Mariottini A, Palma L (2012) Operative case of Langerhans’ cell histiocytosis of the skull with dural invasion. An immunohistochemical study of ki-67 expression of eosinophilic granuloma: case report and review of the literature. J Neurosurg Sci 56:67–72 18. Kawamoto K, Azuma K, Ohishi T, Uesaka T, Seno T, Kawaguchi T, Nakashima Y (2004) Clinico-pathology of small skull tumor: giant cell tumor, myofibroma, myofibromatosis, congenital fibromatosis, infantile myofibromatosis, Paget disease of bone. No Shinkei Geka 32:1085–1090 (in Japanese with English abstract) 19. Beltran J, Aparisi F, Bonmati LM, Rosenberg ZS, Present D, Steiner GC (1993) Eosinophilic granuloma: MRI manifestations. Skeletal Radiol 22:157–161 20. Reznik M, Stevenaert A, Bex V, Kratzenberg E (1993) Focal brain invasion as the first manifestation of Langerhans cell histiocytosis in an adult. Case report. Clin Neuropathol 12:179–183 21. Fujimura M, Nishijima M, Umezawa K, Hayashi T, Kaimori M (2002) A case of calvarial eosinophilic granuloma with rapid expansion and wide skull invasion: immunohistochemical analysis of Ki-67. J Clin Neurosci 9:72–76 22. Postovsky S, Daitzchman M, Dale A, Elhasid R, Ben Arush MWB (2001) Unusual presentation of mastoid eosinophilic granuloma in a young patient. Pediatr Hematol Oncol 18:283–289

123

Comparison of magnetic resonance imaging with invasive histological findings of Langerhans cell histiocytosis.

Due to the rarity of skull Langerhans cell histiocytosis (LCH), correlations between abnormalities on magnetic resonance imaging (MRI) and pathologica...
1MB Sizes 0 Downloads 3 Views