Article Text

Download PDFPDF

Treatment of Erdheim-Chester disease with cladribine: a rational approach
  1. C Myra1,
  2. L Sloper1,
  3. P J Tighe1,2,
  4. R S McIntosh2,3,
  5. S E Stevens2,
  6. R H S Gregson4,
  7. M Sokal5,
  8. A P Haynes6,
  9. R J Powell7
  1. 1Division of Ophthalmology, University of Nottingham School of Clinical Laboratory Sciences, Queen’s Medical Centre, Nottingham NG7 2UH, UK
  2. 2Division of Immunology, University of Nottingham School of Clinical Laboratory Sciences, Queen’s Medical Centre, Nottingham NG7 2UH, UK
  3. 3Division of Ophthalmology, University of Nottingham School of Clinical Laboratory Sciences, Queen’s Medical Centre, Nottingham NG7 2UH, UK
  4. 4Department of Diagnostic Imaging, Queen’s Medical Centre, Nottingham NG7 2UH, UK
  5. 5Department of Oncology, City Hospital, Nottingham NG5 1PB, UK
  6. 6Department of Haematology, City Hospital, Nottingham NG5 1PB, UK
  7. 7Clinical Immunology Unit, Queen’s Medical Centre, Nottingham NG7 2UH, UK
  1. Correspondence to: C Myra L Sloper Neuro-Ophthalmology Department, The National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Erdheim-Chester disease is a rare, life threatening lipoid granulomatosis1 with fewer than 100 cases described in the world literature. The disease typically affects the long bones and symmetrical sclerosis of the diaphyseal and metaphyseal regions is pathognomonic. Extraskeletal manifestations may affect the lungs, pericardium, aorta, retroperitoneum, skin, and orbits and diabetes insipidus occurs in approximately 30% of cases. Erdheim-Chester disease is characterised microscopically by an infiltrate of lipid laden foamy macrophages (histiocytes), scattered Touton giant cells, chronic inflammatory cells, and fibrosis. The foamy macrophages can be distinguished from Langerhans cells on the basis of negative results on staining for S-100 protein and CD1a. Treatment of the disease has been on an ad hoc basis and no treatment regimen has been shown to be clearly superior.

This study documents the clinical findings in a patient with Erdheim–Chester disease, investigates the pathogenesis, and provides a rational basis for effective treatment.

Case report

This white man, aged 45, developed aching in his legs, night sweats, lethargy, and impotence in October 1988, for which no cause was found. His night sweats resolved by July 1989 and he was discharged. He presented in November 1990 with reduced vision (6/9) in the left eye, bilateral proptosis of 12 months’ duration, chemosis, ophthalmoplegia, and optic disc oedema. He still had sexual dysfunction and lethargy and now also had leg oedema and thrombocythaemia. At that time his thyroid function was normal, but erythrocyte sedimentation rate (ESR) and C reactive protein (CRP) were moderately elevated. A computed tomography (CT) scan of the orbits showed bilaterally enhancing masses lying predominantly within the muscle cone and encasing both optic nerves. An orbital biopsy in November 1990 showed an inflammatory picture. There was no evidence of vasculitis on muscle biopsy and a clinical diagnosis of orbital pseudotumour was made. He was initially treated with prednisolone 80 mg daily then reducing, in conjunction with azathioprine 150 mg daily increasing to 250 mg daily, to which he made only a partial response. After treatment was stopped, he failed to attend in 1995 and was lost to follow up.

He presented again in January 1999 with bilateral proptosis, gross chemosis, and infiltration of the conjunctiva and ophthalmoplegia (fig 1A). At this time, the ESR was raised at 74 mm in the first hour and the CRP was elevated at 95 mg/l. A CT scan showed a marked increase in size of the orbital masses. A left anterior orbitotomy and biopsy were performed in February 1999 and histology suggested a diagnosis of fibrous histiocytoma. He was treated with intravenous methyl prednisolone and intravenous cyclophosphamide (total 11 g) but failed to respond. He was referred for consideration of radiotherapy to the orbits. A CT scan was performed which showed soft tissue shadowing around the aorta, pericardium, and kidneys and in addition small pericardial and pleural effusions. The clinical findings of proptosis and widespread soft tissue infiltration suggested Erdheim-Chester disease. Paravertebral tissue was biopsied under CT control. A diagnosis of Erdheim-Chester disease was made on the basis of non-Langerhans histiocytosis, negative for S-100 protein and without intracytoplasmic (Birbeck) granules in this and the previous biopsy specimens (fig 2). X ray of the long bones showed the sclerosis and increased trabecular markings typical of this disease (fig 3).

Figure 1

(A) The degree of bilateral proptosis and right chemosis in July 1999 in the patient with Erdheim-Chester disease, following treatment with cyclophosphamide. (B) The degree of proptosis and chemosis following treatment with cladribine.

Figure 2

Biopsy of paravertebral tissue showing fibrocollagenous tissue, epithelioid cells, and Touton multinucleate giant cells. The biopsy was negative for S-100 protein.

Figure 3

X ray of long bones showing sclerosis and increased trabecular markings.

He was treated with etoposide 50 mg daily and a reducing course of prednisolone, starting at 30 mg daily, without clinical improvement. Etoposide was increased to 100 mg daily with some improvement in the degree of proptosis and chemosis. However, his haemoglobin fell rapidly by 3 g to 8 g/dl and etoposide was discontinued, although a blood film suggested haemolysis or bleeding rather than etoposide induced myelosuppression. On prednisolone 30 mg daily, the haemoglobin rose and there was limited improvement in the proptosis. However, he developed back pain and long term high dose steroids were considered inappropriate.

In November 1999, his visual acuity was 6/6 in the right eye and 6/5 in the left. Colour vision and pupil reactions were normal. He had very restricted eye movements but was binocular with no diplopia. There was bilateral proptosis, right chemosis, and a soft tissue swelling at the left inner canthus. New clinical signs were increased swelling of both optic discs and the presence of choroidal folds bilaterally. Surgical debulking of the orbits was not possible as the xanthogranulomatous tissue surrounded both optic nerves and the external ocular muscles. Bony decompression was considered, but the patient was reluctant to have surgery and also he was a bad anaesthetic risk in view of his cardiac and pulmonary involvement.

In January 2000, hot spots were seen on a bone scan. Spiral CT of the chest showed pulmonary infiltration, but no fibrosis. Abdominal CT showed infiltration around the aorta; there was also enlargement of the seminal vesicles and infiltration of the testes with a resultant low testosterone level, hence the impotence noted previously. Renal infiltration was present and the serum creatinine was elevated. He was treated with cyclosporine 250 mg twice daily (weight 85 kg) reducing to 200 mg twice daily and had some reduction in the degree of proptosis, but he experienced adverse effects of hypertension and renal toxicity, so cyclosporine was discontinued.

Serial peripheral blood samples showed a monocytosis, which responded to treatment with cyclophosphamide and etoposide, but not cyclosporine (fig 4). Analysis of cytokine and activation marker expression was carried out using quantitative RT-PCR. A highly distinctive pattern of cytokine activation was found in the peripheral blood. Interleukin 1α (IL-1α), IL-1β, IL-2, and IL-8 all had raised expression compared with controls (fig 5), consistent with monocyte activation.

Figure 4

Serial monocyte counts in the patient with Erdheim-Chester disease (day 0 equivalent to 14 January 1999). The shaded area denotes the laboratory normal range (0.2–0.8×109/l). Drug treatment periods are indicated by horizontal arrows (CPh = cyclophosphamide, Etop = etoposide, CyA = cyclosporine, and six courses of cladribine). Peripheral blood sampling for PCR based immunological analysis was perfomed on day 195.

Figure 5

Quantitative RT-PCR analysis of cytokine expression on this patient with Erdheim-Chester disease (solid bars) compared to seven normal controls (open bars). A highly distinctive pattern of cytokine activation was found in the peripheral blood. Interleukin-1α (IL-1α), IL-1β, IL-2, and IL-8 all had raised expression compared with controls, consistent with monocyte activation. Smaller rises were seen in IL-6 and tumour necrosis factor α (TNF-α).

He was treated with cladribine, a purine analogue toxic to monocytes, starting in March 2000 at a dose of 0.14 mg/kg/day (given via a Hickman line) for 5 consecutive days every 4 weeks. After two courses, there was clinical improvement in the proptosis and chemosis. After six courses of cladribine, which were well tolerated, there was considerable clinical improvement and his monocyte count normalised (fig 4). Bone scintigraphy in October 2000 showed a great reduction in the abnormal activity. Lung function initially improved, then stabilised. He has now been off treatment for more than 2 years and remains well. His exercise tolerance increased and a CT scan of the thorax in April 2001 showed a decrease in the interlobular septal thickening throughout the lungs. His visual acuity is currently 6/6 bilaterally; the proptosis has resolved, but he has residual, although much reduced, right chemosis (fig 1B). The external ocular movements are now full with no diplopia.


Erdheim-Chester disease is characterised by slow progression of multiple organ system dysfunction with a high mortality. In the largest review, of 59 cases, reported by Veyssier-Belot et al,1 common causes of death included pulmonary fibrosis and cardiac failure. Treatment of patients has been on an individual basis and no randomised controlled trials have been possible as the condition is so rare. Treatments have included systemic steroids,2,3 cytotoxic agents such as vinblastine,4,5 cyclophosphamide, doxorubicin and adriamycin, and also interferon alfa.1,6 Local radiotherapy to the orbits has been used.7,8 The results of treatment have been generally disappointing. In the review by Veyssier-Belot et al, follow up data were available on 37 patients with a mean follow up of 2.7 years. Twenty two out of 37 (59%) patients died within the follow up period, eight within 6 months of diagnosis.1

Very little is known regarding the pathogenesis of this disease. The monocytosis and the highly distinctive pattern of cytokine activation detected in the peripheral blood of this patient with Erdheim-Chester disease suggested monocyte activation as a significant part of the pathophysiology.

Cladribine is a purine analogue that is toxic to monocytes.9 Cladribine also destroys both resting and dividing lymphocytes,10 and causes T cell depletion.11 In 1999, Saven and Burian12 described encouraging responses to cladribine in 13 patients with adult Langerhans cell histiocytosis. This information, together with our new evidence of increased monocyte activation in this patient, made clabribine, an agent toxic to monocytes, a rational choice. There has been one previous report of treatment of Erdheim-Chester disease with cladribine.13 That patient had orbital involvement and unfortunately developed bilateral blindness. It was postulated that cladribine might have caused toxic injury to the optic nerves which predisposed them to ischaemic injury. However, the clinical signs suggested progression of the Erdheim-Chester disease as the cause of the blindness. There has also been a case of transient blindness occurring during therapy with clabribine.14 If treatment with cladribine is instituted for orbital disease, careful monitoring of the degree of proptosis and optic nerve compression is mandatory.

This patient with Erdheim-Chester disease showed evidence of increased monocyte activation. He has shown a significant recovery and maintained clinical improvement following treatment with cladribine, an agent toxic to monocytes. Although the long term durability of its effect is not yet known, this patient has had a good quality of life for 2 years after stopping treatment. This is the first report to correlate the clinical findings and response to treatment with the laboratory results in peripheral blood and provides a rational basis for treatment of this life threatening condition.


We gratefully acknowledge the assistance given by Professor J Lowe in the pathological analysis of the biopsy specimens.



  • The authors have no proprietary interest in any aspect of this work.