Clinical, neuropathological and radiological evidence for a rare complication of rituximab therapy

Accepted Manuscript Title: Clinical, neuropathological and radiological evidence for a rare complication of rituximab therapy. Author: E.G. Healy, R. Phadke, Kidd M, M.M. Reilly, M.P. Lunn PII: DOI: Reference:

S0960-8966(15)00129-7 http://dx.doi.org/doi:10.1016/j.nmd.2015.04.004 NMD 3032

To appear in:

Neuromuscular Disorders

Received date: Revised date: Accepted date:

3-2-2015 5-4-2015 7-4-2015

Please cite this article as: E.G. Healy, R. Phadke, Kidd M, M.M. Reilly, M.P. Lunn, Clinical, neuropathological and radiological evidence for a rare complication of rituximab therapy., Neuromuscular Disorders (2015), http://dx.doi.org/doi:10.1016/j.nmd.2015.04.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

TITLE Clinical, neuropathological and radiological evidence for a rare complication of rituximab therapy.

AUTHORS EG Healy1, R Phadke2, Kidd M 3, MM Reilly1, 4, MP Lunn1, 4 CORRESPONDENCE TO Dr Michael P Lunn 1 MRC Centre for Neuromuscular Diseases, Department of Molecular Neurosciences, UCL Institute of Neurology and 4National Hospital for Neurology and Neurosurgery London, United Kingdom Box 76, Queen Square, London WCN1 3BG United Kingdom E-mail: [email protected] Telephone: 020 3448 3267 Fax: 020 3448 3633

CO-AUTHORS Dr Estelle G Healy 1 MRC Centre for Neuromuscular Diseases, Department of Molecular Neurosciences, UCL Institute of Neurology, London, United Kingdom Dr Rahul Phadke 2 Division of Neuropathology National Hospital for Neurology and Neurosurgery London, United Kingdom Dr Michael Kidd 3 Department of Virology University College London Hospitals NHS Foundation Trust London, United Kingdom Professor Mary M Reilly 1 MRC Centre for Neuromuscular Diseases, Department of Molecular Neurosciences, UCL Institute of Neurology and 4National Hospital for Neurology and Neurosurgery London, United Kingdom KEY WORDS Rituximab, echovirus, hypogammaglobulinaemia, myofasciitis, meningitis. WORD COUNT: 1,487 1

Page 1 of 13

Highlights for Review     

Anti CD 20 therapy may result in unusual infections. We report enterovirus myofasciitis with documented pathology for the first time. Increased susceptibility to rare viruses should be considered in treated patients. Echovirus related muscle disease manifests as myofasciitis not dermatomyositis. Successful treatment can be considered as discussed in this report.

ABSTRACT We report a rare case of myofasciitis and meningitis with deafness caused by systemic enterovirus infection in the setting of hypogammaglobulinaemia induced by rituximab. Whilst effective and generally safe, anti- CD 20 antibody therapy is increasingly recognised to result in unusual infectious complications to be considered in a treated patient presenting with neurological symptoms. These cases may pose diagnostic difficulties and can have atypical presentations. We present this rare complication of rituximab therapy, with histopathological confirmation of myofasciitis. In the older literature, enterovirus associated myofasciitis may have erroneously been termed dermatomyositis and we review the literature to demonstrate this important nosological point.

INTRODUCTION

Anti CD 20 therapy may result in unusual infections with atypical presentations, posing diagnostic difficulties. We highlight the long term immunomodulatory effects of rituximab by reporting the rare complication of myofasciitis, with clear documentation of the pathology of enterovirus myofasciitis for the first time. 2

Page 2 of 13

CASE REPORT A 36 year old man was treated four years into his diagnosis of a grade 1 follicular lymphoma, with six cycles of R-CHOP chemotherapy. He achieved a complete metabolic remission on re-staging positron emission tomography (PET) scanning. Chemotherapy was complicated by two episodes of neutropenic sepsis and mild iatrogenic deafness with vestibular failure from aminoglycoside use. One year following treatment, whilst on maintenance therapy with rituximab three monthly, he developed myalgia, shoulder pain, fatigue, weight loss and progressive hearing and balance impairment. There was no history of foreign travel or vaccination. General examination was unremarkable; in particular there was no skin rash, lymphadenopathy or splenomegaly. Neurological examination revealed subtle sensorineural hearing loss, upper limb fasciculations and present but reduced reflexes in the upper and lower limbs. A recurrence of lymphoma was excluded. Investigations showed panhypogammaglobulinaemia; IgG 6.91 g/L (normal: 7-16), IgA 0.54 g/L (0.7-4.0) and IgM 0.33 g/L (0.44-2.3). Creatine kinase was normal, 64 IU/L (38-204) and LDH was elevated 929 IU/L (240-480). CSF chemistry showed an elevated protein 2.55 g/L (0.13-0.40) and reduced glucose 2.0/5.7 mmol/L (3.9-5.8). There were 105 white blood cells per cubic mm of CSF, mainly lymphocytes, and examination by flow cytometry and PCR confirmed a reactive population with no clonality. Neurophysiological examination demonstrated a mild, predominantly sensory, sensorimotor neuropathy consistent with previous chemotherapy with mild myopathic changes only by electromyography with small polyphasic motor units with short duration and no spontaneous activity. The muscles were hard and ‘woody’ on needle insertion. Audiometry confirmed sensorineural deafness with vestibular and cochlear failure. MRI of the brain, spine and internal auditory meati was normal. MRI axial imaging of the 3

Page 3 of 13

thighs showed florid myofascial high signal on STIR sequences, consistent with fasciitis (Figure 1: A). A biopsy of the left vastus lateralis showed macrophages infiltrating the epimysium and perimysial septa, with accompanying perivascular lymphocytic cuffs (CD3+, CD8+ T cells), fewer CD4+ T cells and no CD20+ B cells. Minimal extension of the fascial inflammation was seen, limited to adjacent muscle fibres at the periphery of the fascicles. Perifascicular atrophy, endomysial capillary loss and capillary complement activation were absent. The pathology was of myofasciitis (Figure 1 C-J). PCR was performed on the serum and CSF, demonstrating high copy numbers of enterovirus, sequenced as echovirus type 9. Matching high copy number PCR was also demonstrated in blood. Additional infections were ruled out, including herpes viruses and fungal species. No virus was detected on PCR of the muscle and fascia sample. No viral particles or aluminium deposits were visualised. The diagnosis of systemic enterovirus infection due to rituximab induced panhypogammaglobulinaemia was made, with increasing deafness due to inflammatory meningitis and accompanying myofasciitis. The mild neuropathy was consistent with previous CHOP chemotherapy. Treatment with IVIG was commenced with an initial dose of 2g/kg over 5 days, reduced to a standard dose of 0.4g/kg every 4 weeks, resulting in a sustained rise in IgG levels with peak dose levels in the high normal range for IgG. Serum and subsequently CSF became negative for enterovirus by PCR at 6 and 10 weeks, however, the CSF remained active (70 lymphocytes/mm3, protein 1.94g/L, glucose normal). The muscle pain and fasciculations resolved, permitting gym exercise. Repeat MRI imaging of the thighs showed complete resolution of the fascial inflammation (Figure 1: B) at 16 weeks. Additional steroid treatment was initiated, in addition to the IVIG in an attempt to mitigate deafness at week 18. However, a sudden and rapidly progressive deterioration to total deafness occurred at week 20. Repeat CSF examination indicated an increasingly active CSF (200 WCC and protein 2g/l) with 4

Page 4 of 13

reactivation of enterovirus and subsequent complete hearing loss. The dose of IVIG was increased to 1g/kg every 4 weeks, with cycling of plasma pools and trough levels of IgG in the normal range and peaks >25g/l. This has resulted in complete clearance of the CSF by PCR testing, a reduction in CSF inflammation (9 WCC, 1.2 g/l protein), and maintenance of systemic recovery at 40 months. The deafness has not improved and a cochlear implant has returned speech discrimination.

DISCUSSION: Chronic ‘myositis’ and meningitis due to echovirus infection is documented in patients with primary immunoglobulin deficiencies.[1-3] A number of enteroviral

strains

have been

isolated from the CSF but not in muscle, as in this case. It is hypothesised that the role of enteroviruses in the pathogenesis of inflammatory muscle disease may be related to pathways of autoimmunity following acute infection, without persisting viral replication. Many early cases have been described to have an unfavourable prognosis, with death in ten out of thirteen patients in one series.[4] Rituximab is generally a well-tolerated and safe drug and induced persistent hypogammaglobulinaemia is not uncommon. Recent findings suggest that rituximab combined chemotherapy may induce persistent differentiation arrest and apoptosis of B cell lineage with alteration of T lymphocyte homeostasis resulting in panhypogammaglobulinaemia.[5]

Although

neurological involvement is

rare with

infectious

complications

the exception

are

recognised,

of progressive multifocal

leucoencephalopathy due to JC polyoma virus re-activation.[6] Six cases of enteroviral meningoencephalitis following rituximab therapy are reported in the literature.[7, 8] The concomitant presentation with unusual muscle pathology is unique to this case, but the interval from initial treatment to presentation with infectious complications is typical; Servais

5

Page 5 of 13

et al.[9] reported the time between rituximab treatment and neurological manifestations up to 11 months after rituximab. IVIG contains anti-enterovirus antibodies, because of almost ubiquitous commensal infection in the population and components from over 1000 donors per IVIG infusion. Enteroviruses are cytolytic and hence cleared predominantly by humoral mechanisms. In patients with hypogammaglobulinaemia, titres of anti-enteroviral antibodies may be low or lacking. Supplementation by IVIG may be effective in treating the virus in the systemic compartment; this patient who had a good response with viral clearance, resolution of muscle symptoms and restoration of exercise tolerance. Furthermore, evidence of complete resolution of the fascial inflammation was obtained by repeat MRI imaging (Figure 1: B – a repeat biopsy was not performed). Penetration of the blood brain barrier with immunoglobulins is more of a challenge but with higher doses of IVIG, the CSF remains clear. Since its first description by Janeway et al. in 1956,[10] a ‘’dermatomyositis-like syndrome’’ comprising varying combinations of erythema, oedema and induration of the skin, muscle weakness, wasting and flexion contractures of the extremities has been recognised in patients with primary immunoglobulin deficiencies harbouring systemic echovirus infection. Differences from classical dermatomyositis were recognised; the skin rash was transient and non-violaceous, calcinosis was never described and the myositis was relatively mild.[11] With the exception of the case reported by Mease et al.[3] the clinical and laboratory evidence for true muscle involvement was poor or lacking (Table 1).[1-4, 12] In many patients, significant CNS involvement was a confounding factor in the assessment of new onset muscle weakness. No biopsy confirmation is available in the largest series described.[12]

In

other

instances,

descriptions

are

limited

to

‘’perivascular

lymphocytic/mononuclear infiltrates’’ and precise characterisation of the cellular infiltrates and their extent is absent. Most of the descriptions pre-date the wider use of 6

Page 6 of 13

immunohistochemistry in muscle biopsies and the recognition of capillary microvascular pathology as integral to classical dermatomyositis. The pathology in our case is of myofasciitis, limited to the periphery of the fascicles with minimal muscle involvement and dominated by macrophages. Dalakas [13] also reported two patients with echovirus-related muscle disease, showing identical features. Perivascular CD3+ and CD8+ T cells are present in all three cases. Dermal perivascular infiltrates are also documented in previously reported cases; however, a skin biopsy was not performed by us as there was no skin involvement. Finally, the myofascial inflammation in our case has features in common with inflammatory myopathy with abundant macrophages (IMAM), a condition first described in patients clinically diagnosed with dermatomyositis.[14] More recently, distinct immunophenotypic differences have been demonstrated between the infiltrates in IMAM and classical dermatomyositis, but a precise relationship is as of yet, unresolved.[15] As in IMAM, in our case macrophages predominate, B cells are absent and CD4+ T cells are few. A T cell mediated, cytokine driven, macrophage hyperactivation may drive the pathological process and the resultant musculocutaneous manifestations. In this case report we highlight the long term immunomodulatory effects of rituximab therapy, resulting in increased susceptibility to rare humorally cleared viral infections such as echovirus, which should be considered in patients with a history of therapy who present with unusual symptoms. The rare complication of myofasciitis is reported and clear documentation of the pathology of enterovirus myofasciitis complicating rituximab therapy is presented for the first time. We believe that echovirus-related muscle disease manifests in the form of myofasciitis rather than true dermatomyositis. We recommend vigilance and investigation by the modalities employed to achieve a diagnosis and prompt treatment with IVIG.

7

Page 7 of 13

ACKNOWLEDGMENTS

We thank Dr Tim Diss, Clinical Scientist at University College Hospital London for PCR analysis and Dr Zane Jaunmuktane, Specialty Registrar in Neuropathology, Division of Neuropathology, National Hospital for Neurology and Neurosurgery for slide scanning and image preparation. We thank the UCLH/UCL CBRC for their support.

CONFLICTS OF INTEREST: None.

FUNDING: None.

CONTRIBUTOR STATEMENT

ML and RP contributed to the concept of the study. All authors worked on analysis and interpretation of the data. EH drafted the report. ML, RP and MMR revised it critically for the intellectual content.

8

Page 8 of 13

REFERENCES

1. J.A. Bardelas, J.A.Winkelstein, D.S.Y. Seto, et al. Fatal ECHO 24 infection in a patient with hypogammaglobulinaemia: Relationship to dermatomyositis-like syndrome. The Journal of Pediatrics 90 (1997) 396-399.

2. A.D.Webster , J.H. Tripp, A.R. Hayward, et al. Echovirus encephalitis and myositis in primary immunoglobulin deficiency. Arch Dis Child 53 (1978) 33-37.

3. P.J. Mease, H.D. Ochs, R.J. Wedgewood. Successful treatment of echovirus meningoencephalitisand myositis-fasciitis with intravenous immunoglobulin therapy in a patient with X-linked agammaglobulinaemia. N Engl J Med 304 (1981) 1278.

4. P. Rudge, A.D. Webster, T. Revesz T, et al. Encephalomyelitis in primary hypogammaglobulinaemia. Brain 119 (1996) 1-15.

9

Page 9 of 13

5. E. Irie, Y. Shirota, C. Suzuki, et al. Severe hypogammaglobulinemia persisting for 6 years after treatment with rituximab combined chemotherapy due to arrest of B lymphocyte differentiation together with alteration of T lymphocyte homeostasis. Int J Haematol 91 (2010) 501-8.

6. K.R. Carson, A.M. Evens, E.A. Richey, et al. Progressive multifocal leukoencephalopathy after rituximab therapy in HIV-negative patients: a report of 57 cases from the Research on Adverse Drug Events and Reports project. Blood 113 (2009) 4834–4840.

7. M.F. Cheng, B.C. Chen, T.S. Huang, et al. Clinical application of reverse-transcription polymerase chain reaction and intravenous immunoglobulin for enterovirus encephalitis. Jpn J Infect Dis 61 (2008)18–24.

8. C. Schilthuizen, H.W. Berenschot, M.D. Levin. Enteroviral encephalitis in a patient with a marginal zone lymphoma treated with rituximab. Neth J Med 68 (2010) 221-3.

9. S. Servais, J. Caers, O. Warling, et al. Enteroviral meningoencephalitis as complication of Rituximab therapy in a patient treated for diffuse large B-cell lymphoma. Br J Haematol 150 (2010) 379-381.

10. C.A. Janeway, D. Giltin, J.M. Craig, et al. ‘Collagen diseases’ in patients with congenital agammaglobulinaemia. Trans Assoc Am Physicians 69 (1956) 93-97.

11. A.D.B.Webster. Echovirus disease in hypogammaglobulinaemic patients. Clin Rheum Dis 10 (1984) 189-203. 10

Page 10 of 13

12.H.M. Lederman, J.A.Winkelstein. X-linked agammaglobulinaemia: an analysis of 96 patients. Medicine 64 (1985) 145-156.

13. M. Dalakas, Virus-related muscle diseases, in: A.G. Engel, C. Franzini-Armstrong (Eds.), Myology: Basic and Clinical, McGraw-Hill, New York, 2004, pp.1389-1417.

14. G. Bassez, F.J. Authier, E. Lechapt-Zalcman, et al. Inflammatory myopathy with abundant macrophages (IMAM): a condition sharing similarities with cytophagic histiocytic panniculitis and distinct from macrophagic myofasciitis. J Neuropathol Exp Neurol 62 (2003) 464-473.

15. A. Brunn, V.J. Hans, S. Vogelgesang. Inflammatory myopathy with abundant macrophages anddermatomyositis: two stages of one disorder or two distinct entities? Acta Neuropathol 118 (2009) 793-801.

LEGENDS

Figure 1: MRI and muscle biopsy findings. Axial STIR imaging of the thighs showing diffuse inflammation within the inter-muscular spaces in the anterior, adductor and posterior compartments. The muscle signal is normal. There is fascial inflammation on the superficial enveloping fascia lata, fat and dermal tissue, overlying the anterior and posterior aspects of the thighs (A). Further axial STIR imaging after an interval of 3 months and IVIG therapy (same scanner and imaging protocol) shows complete resolution of the inflammation (B). Haematoxylin and eosin stained sections from a muscle biopsy of the left vastus lateralis (CF) show marked inflammation centered on the epimysium and perimysial septa (C,D & F) with perivascular lymphocyte cuffing in the fascia (blue arrows in C and in D, at higher power). 11

Page 11 of 13

The inflammatory infiltrate extends minimally in to the peripheral of the fascicles, with scattered foci of lymphocytes which do not invade the fibres (black arrow in E). There is no perifascicular atrophy (F). Immunotyping shows strong positive staining with CD68 which highlights the presence of macrophages (G) and CD8+ T lymphocytes (H). HLA up regulation ( antibody to MHC Class I) is strongest within the fascia and in fibres at the edge of fascicles (I) and labelling for complement (antibody to C5b-9) shows no abnormal deposits on the endomysial capillaries (J). The pathology is consistent with myofasciitis and there are no features of classical dermatomyositis. Scale bar: 200 µm (C, G, H & I); 100 µm (E &J); 50 µm (D).

Table 1: Echovirus associated ‘Dermatomyositis-like syndrome’ in primary immunoglobulin deficiencies: review of clinical and pathological features.

12

Page 12 of 13

Year

Author

Reference

Number of patients affected

Description of skin changes

1977

Muscle weakness

Laboratory parameters

Bardelas

[1]

1/1

Brawny oedema, violaceous telangiectatic rash on lower limbs

Not documented

1978

Webster

[2]

1/2

Subcutaneous oedema, transient erythematous rash (right biceps)

History of Not documented paralytic poliomyelitis

1980

Jenn

[16]

1/1

Absent

1981

Mease

[3]

1/1

Woody oedema of extremities, face and sacrum

Myalgia and tender thighs, no weakness Marked

CK 19,600 IU, myoglobinaemia and myoglobinuria CK increased, EMG: generalised myopathy

1985

Lederman

[17]

5/96

Present

Not documented

1996

Rudge

[4]

3/13

Rash, subcutaneous oedema Erythema and oedema of legs and arms in 2/3, not documented in 1/3 patients.

Normal CK

History of Not documented flaccid poliomyelitis in 1/3, not documented in 2/3

Biopsy findings

Perivascular lymphocytic infiltration and small areas of muscle degeneration Mononuclear infiltrate in muscle and in connective tissue around small blood vessels, some necrosis Biopsy obtained from unaffected muscle Perivascular lymphocytic infiltrates of fascia and adjacent muscle Not documented

Perivascular infiltrate in skin and muscle in two, myositis (not further characterised) in one

13

Page 13 of 13