1757

□ CASE REPORT □

Spontaneously Regressed Langerhans Cell Histiocytosisof Lymph Nodes in an Elderly Patient

Tohru Takahashi 1, Mitsuru Yoshimoto 2 and Nobuo Kondoh 3

Abstract

Langerhans cell tumors are neoplastic proliferation of Langerhans cells and are classified into Langerhanscell histiocytosis (LCH) and Langerhans cell sarcoma (LCS). We report a case of LCH in an 89-year-old-woman with left axillary lymphadenopathy. A histologic examination demonstrated a proliferation of histio-cytoid cells which were positive for CD1a, S-100 protein, and Lagerin (CD207). Initial diagnosis was LCSbased on morphologic features, high MIB-1 index, and multi-system involvement detected by FDG-PET.However, the disease disappeared spontaneously without specific treatment in six months. The disease wasconsidered to be spontaneously regressed LCH with multi-system involvement rather than LCS.

Key words: Langerhans, regression, histiocytosis

(DOI: 10.2169/internalmedicine.46.0356)

Introduction

Langerhans cell tumors include diseases previously desig-nated histiocytosis X, eosinophilic granuloma, Letterer-Siwedisease, Hand-Schuller-Christian syndrome, Hashimoto-Pritzker syndrome (1). They may involve one or more bodysystems or tissue, such as bone, lung, hypothalamus, skin,etc. These diseases usually are considered to be childhooddiseases. However, adults may frequently be affected as ob-served in eosinophilic granuloma of the bone and pulmonaryLangerhans cell histiocytosis (2). The clinical course variesfrom generalized and fulminant disease to localized and cur-able disease. The prognosis of patients with multi-system in-volvement is generally poor. Langerhans cell tumors are cur-rently classified by World Health Organization (WHO) intoLangerhans cell histiocytosis (LCH) and Langerhans cellsarcoma (LCS) (3). LCS shows overtly malignant cytologi-cal features and is considered to be a higher-grade variant ofLCH. We report a case of LCH which developed in thelymph nodes of an elderly patient. Spontaneous regressionof the tumor was observed in this case.

Case Report

An 89-year-old woman was referred to our hospital in

September 2006 because of left axillary lymphadenopathy.She had no remarkable occupational history. There was nofamily history of hematological disease or malignant tumors.She had a history of mild cerebral infarction several yearsago. On physical examination, a firm and smooth surfacedenlarged axillary lymph node of 3 cm in diameter was pal-pated. No other superficial lymph nodes, liver, or spleenwere palpated.The laboratory findings were as follows; white blood cellcount was 6,900/mm3, with a normal differential count. Thered blood cell count was 427×104/mm3 and hemoglobin con-centration 12.7 g/dl. Her platelet count was 190,000/mm3.Biochemistry tests were normal. C-reactive protein was 0.42mg/dl (reference range, less than 0.3). Soluble IL-2 receptorwas 565 U/ml (reference range, 135-483).18F-fluorodeoxyglucose (FDG)-positron-emission tomogra-

phy (PET) with simultaneous whole body CT scan was per-formed for detecting tumors. As shown in Fig. 1A, the studyshowed increased uptake of FDG in the left lower neck, su-pra clavicular, and axillary lymph nodes, as well as lymphnodes close to the porta hepatis, the left iliac bone, and theleft knee (data not shown).A biopsy of the left axillary lymph node was performed.There was massive infiltration of histiocytoid cells, somemultinuclear giant cells, and areas of necrosis (Fig. 2). Im-munohistochemical analysis disclosed that the histiocytoid

１Department of hematology, Teshi hospital, ２Department of gastroenterology, Teshi hospital, Sapporo and ３GLab pathology center, SapporoReceived for publication May 30, 2007; Accepted for publication July 12, 2007Correspondence to Dr. Tohru Takahashi, tohrut@cocoa.ocn.ne.jp

DOI: 10.2169/internalmedicine.46.0356

1758

Figure 1. FDG-PET in September, 2006 showed an increased uptake of FDG in the left lower neck, supraclavicular, and axillary lymph nodes, as well as in a lymphnode close to the porta hepa-tis, and the left iliac bone (A). Lack of specific uptake of FDG indicated absence of the tumors in March, 2007 (B). Subtle uptake at the right hilum and the mediastinum were considered to be non-specific.

Figure 2. Lymphnode biopsy specimen showed massive infiltration of histiocytoid cells, multinuclear giant cells (arrows), and areas of necrosis.

cells and the giant cells were positive for CD1a, CD4,CD68, and S-100, and negative for CD3, CD20, CD21, orCD56 (Fig. 3). Simultaneous expression of both CD1a, S-100, and Langerin (CD207) indicated that the tumor cellswere of Langerhans cell origin. EBV integration status wasnot studied because it became apparent that the tumor cellswere not derived from lymphocyte lineage. More than 50%of the tumor cells were positive for MIB-1 staining. Basedon malignant cytologic features and high MIB-1 index, LCSwas diagnosed initially. Chromosome analysis of the lymphnode showed 46, XX. Bone marrow aspiration and biopsy

were not performed.The patient had been given no chemotherapy or radiother-apy considering her age. She had been followed up once amonth in out patient clinic. She had no symptoms throughthe follow-up period. The left axillary lymph node had beenpalpated up until two months after the biopsy though re-duced in size. However, it became unpalpable thereafter. Sixmonths later, FDG-PET was performed again. To our sur-prise, no abnormal uptake indicating existence of the tumorwas observed anymore (Fig. 1B). A whole body CT scanperformed at the same time also disclosed no abnormal find-

DOI: 10.2169/internalmedicine.46.0356

1759

Figure 3. Immunohistochemical analyses of the lymph node specimen showed that the tumor cells were positive for S-100 protein, Langerin (CD207), and CD1a. More than 50% of the tumor cells were positive for MIB-1 staining.

ings. The tumors were considered to have regressed sponta-neously. Thus, the disease was considered to be spontane-ously regressed LCH with multi-system involvement ratherthan LCS. No recurrence has been observed (11 months af-ter the presentation).

Discussion

Although neoplastic proliferations of Langerhans cells arecurrently classified by WHO into LCH and LCS (3), it isoften difficult to draw a definite morphologically-based dis-tinction between the two (4). The present case was initiallydiagnosed as LCS based on morphologic feature, high MIB-1 index, and multi-system involvement. However, the dis-ease regressed spontaneously in six months. Therefore, thiscase should be considered as LCH rather than LCS. Kawaseet al reported that tumor cells of 4 LCS cases were all posi-tive for CD56 positive while tumor cells of 8 LCH caseswere negative for CD56 (5). They suggested that CD56might be a clinically relevant biologic marker for predictingan intractable course of Langerhans cell neoplasms. The tu-mor cells in the present case were negative for CD56 (datanot shown). This may support the predictive value of CD56for aggressive Langerhans cell neoplasms. However, Bohn etal recently reported a case of cutaneous LCS with a poorclinical course in which CD56 expression was not found (6).Specificity of CD56 expression on LCS should be investi-gated.FDG-PET is now being used in the evaluation of a vari-

ety of neoplasms such as lymphomas and breast, esopha-geal, stomach, colorectal, head and neck, and non-small-celllung cancers (7). There is a rapid and significant decline in18F-FDG uptake in tumors that respond to therapy by clini-cal, radiographic, or histopathological assessment, whereasno such decline is observed in non-responding tumors. Italso may predict response earlier than conventional tech-niques. In the present case, clear uptake of the tracer notonly in the lymph nodes but in the presumable osseous le-sions was observed. For Langerhans cell tumors, FDG-PETappears to have greater sensitivity and specificity than bonescintigraphy, radiography, and MR imaging for the identifi-cation of active osseous lesions in LCH (8, 9). Other reportshave also suggested that FDG-PET might be a useful mo-dality for the monitoring of central nervous system diseaseactivity in LCH (10, 11).It is to be emphasized that the tumor cells disappeared inthe present case. The mechanism of the spontaneous regres-sion remains to be elucidated. Scattered CD3 positive T-cellswere observed among the tumor cells in our case in thehistopathological study (data not shown). One can presumethat cytotoxic T-lymphocytes against the tumor cells mighthave been induced in our case. Alternatively, some infec-tious agents might have promoted proliferation of the dis-ease even though LCH is recognized as a clonal neoplasticdisorder (12, 13). Clearance of an unknown infectious agentmight have resulted in regression of tumors as recognized incases of gastric MALT lymphoma with H. pyroli.Spontaneous regression of LCH has been reported in the

DOI: 10.2169/internalmedicine.46.0356

1760

literature. Congenital self-healing Langerhans cell histiocyto-sis (CSHLCH) is a rare variant of LCH, presenting at birthor in the neonatal period with cutaneous lesions that invo-lute spontaneously (14). Aside from CSHLCH, only a fewcases of the spontaneous regression of LCH in other bodysites, have been reported. Kamimura et al reported 2 casesof spontaneous disappearance of eosinophilic granuloma inthe vertebral bones (15). A large series by Howarth et al de-scribed that some osseous and pulmonary lesions regressedspontaneously (1). Yamaguchi et al reported a case of LCHin the CNS with spontaneous regression (16). Broadbent et

al reported two cases of spontaneous remission of multi-system histiocytosis X (17). Iwafuchi et al reported a caseof histiocytosis X of the stomach showing spontaneous re-mission (18). There were no common features betweenthose cases to predict spontaneous regression of tumors.It is difficult to predict the prognosis of Langerhans tumoronly based on the histologic features or multi-system in-volvement. Although CD56 expression in the tumors may bea relevant biologic marker to predict an intractable course ofLangerhans cell tumors, careful clinical observation is neces-sary to follow cases of LCS and LCH.

References

1. Howarth DM, Gilchrist GS, Mullan BP, Wiseman GA, EdmonsonJH, Schomberg PJ. Langerhans cell histiocyotsis: diagnosis, natu-ral history, management, and outcome Cancer 85: 2278-2290,1999.

2. Vassallo R, Ryu JH, Schroeder DR, Decker PA, Limper AH.Clinical outcome of pulmonary Langerhans’-cell histiocytosis inadults. N Engl J Med 346: 484-490, 2002.

3. Weiss LM, Grogan TM, Pileri SA, et al. Langerhans cell sarcoma.In: World Health Organization Classification of Tumors: Pathologyand Genetics of Tumours of Haematopoietic and Lymphoid Tis-sues. Jaffe ES, Harris NL, Stein H, Vardiman JW, Eds. IARCPress, Lyon, France, 2001: 283.

4. Ben-Ezra J, Bailey A, Azuma N, et al. Malignant histiocytosis X.A distinct clinicopathologic entity. Cancer 68: 1050-1060, 1991.

5. Kawase T, Hamazaki M, Ogura M, et al. CD56/NCAM-positiveLangerhans cell sarcoma: a clinicopathologica study of 4 cases.Int J Hematol 81: 323-329, 2005.

6. Bohn OL, Ruiz-Arguelles G, Navarro L, Saldivar J, Sanchez-SosaS. Cutaneous Langerhans cell sarcoma: A case report and reviewof literature. Int J Hematol 85: 116-120, 2007.

7. Juweid ME, Cheson BD. Positron-emission tomography and as-sessment of cancer therapy. N Engl J Med 354: 496-507, 2006.

8. Daldrup-Link HE, Franzius C, Link TM, et al. Whole-body MRimaging for detection of bone metastases in children and youngadults: comparison with skeletal scintigraphy and FDG PET. AJRAm J Roentgenol 177: 229-236, 2001.

9. Binkovitz LA, Olshefski RS, Adler BH. Coincidence FDG-PET inthe evaluation of Langerhans’ cell histiocytosis: preliminary find-ings. Pediatr Radiol 33: 598-602, 2003.

10. Büchler T, Cervinek L, Belohlavek O, et al. Langerhans cell histi-ocytosis with central nervous system involvement: follow-up byFDG-PET during treatment with cladribine. Pediatr Blood Cancer44: 286-288, 2005.

11. Steiner M, Prayer D, Asenbaum S, et al. Modern imaging methodsfor the assessment of Langerhans’ cell histiocytosis-associatedneurodegenerative syndrome: case report. J Child Neurol 20: 253-257, 2005.

12. Willman CL, Busque L, Griffith BB, et al. Langerhans’ cell histio-cytosis (histiocytosis X)-a clonal proliferative disease. N Engl JMed 331: 154-160, 1994.

13. Yu RC, Chu C, Buluwela L, Chu AC. Clonal proliferation ofLangerhans cells in Langerhans cell histiocytosis. Lancet 343:767-768, 1994.

14. Larralde M, Rositto A, Giardelli M, et al. Congenital self-healinghistiocytosis (Hashimoto-Pritzker). Int J Dermatol 38: 693-696,1999.

15. Kamimura M, Kinoshita T, Itoh H, Yuzawa Y, Takahashi J, Oh-tsuka K. Eosinophilic granuloma of the spine: early spontaneousdisappearance of tumor detected on magnetic resonance imaging.Case report. J Neurosurg 93: 312-316, 2000.

16. Yamaguchi S, Oki S, Kurisu K. Spontaneous regression ofLangerhans cell histiocytosis: A case report. Surg Neurol 62: 136-140, 2004.

17. Broadbent V, Pritchard J, Davies EG, et al. Spontaneous remissionof multi-system histiocytosis X. Lancet 1: 253-254, 1984.

18. Iwafuchi M, Watanabe H, Shiratsuka M. Primary benign histiocy-tosis X of the stomach. A report of a case showing spontaneousremission after 51/2 years. Am J Surg Pathol 14: 489-496, 1990.

Ⓒ 2007 The Japanese Society of Internal Medicinehttp://www.naika.or.jp/imindex.html