Escalation to Anti-CD20 Treatment for Multiple Sclerosis Following Natalizumab-Associated Progressive Multifocal Leukoencephalopathy

Abstract

Objectives

Disease-modifying therapy (DMT) for multiple sclerosis (MS) after natalizumab-associated progressive multifocal leukoencephalopathy (PML) is controversial due to concern for recurrent PML. We describe DMT utilization for over a decade in a patient with MS who survived PML.

Methods

Case report.

Results

A 36-year-old woman was diagnosed with MS in 2002 and treated with interferon beta-1a until 2006, when she transitioned to natalizumab due to relapses. She presented in 2012 with 2 months of progressive cognitive and gait concerns and was diagnosed with PML by positive CSF JC virus testing with concordant clinical and MRI findings. She was treated with plasma exchange and then corticosteroids for PML immune reconstitution inflammatory syndrome before starting glatiramer acetate for DMT. She transitioned to dimethyl fumarate in 2013 after MS activity on MRI with negative CSF JC virus testing. Owing to worsening footdrop consistent with progression, she transitioned to ocrelizumab in 2017 and then to ofatumumab in 2020 due to logistics of medication administration. There has been no clinicoradiographic or CSF evidence of recurrent PML.

Discussion

DMT selection is challenging for patients with MS who survive PML. We used an escalation approach extending to ocrelizumab and ofatumumab due to MS progression. Anti-CD20 DMTs are a high-efficacy option post-PML.

Classification of Evidence

This provides Class IV evidence. It is a single observational study without controls.

PRACTICAL IMPLICATIONS

Consider anti-CD20 disease-modifying therapies as high-efficacy options in patients with MS who survive progressive multifocal leukoencephalopathy.

Introduction

Multiple immunosuppressive disease-modifying therapies (DMTs) can place patients with multiple sclerosis (MS) at risk for progressive multifocal leukoencephalopathy (PML) caused by JC virus reactivation (Table). Natalizumab is the most common cause of DMT-associated PML, with risk associated with longer duration of treatment (especially >24 months), higher JC virus antibody titer (index >0.9 on STRATIFY), and previous use of immunosuppressive therapies.^8,9 Natalizumab-associated PML results in death in ∼25% of patients.¹⁰ For patients who survive natalizumab-associated PML, there is limited experience on how to retreat with DMT. As many patients with natalizumab-associated PML have highly active MS, balancing effective treatment for MS with risk of PML recurrence is paramount. We present the case of a patient with MS who developed natalizumab-associated PML and was subsequently treated with glatiramer acetate and dimethyl fumarate before safe long-term escalation to anti-CD20 DMT.

Table.

Progressive Multifocal Leukoencephalopathy Cases Reported With Disease-Modifying Therapies FDA-Approved for Multiple Sclerosis

DMT	FDA-approval date	PML cases for PwMS	Estimated PML incidence for PwMS
Alpha-4-integrin inhibitor
Natalizumab1	2004	857	310/100,000 patient y
S1PR modulators
Fingolimod2	2010	61	5.88/100,000 patient y
Ozanimod3	2020	1
Ponesimod	2021	None
Siponimod4	2019	2
Fumarates
Dimethyl fumarate5	2013	9	1.07/100,000 patient y
Diroximel fumarate	2019	None
Monomethyl fumarate	2020	None
Anti-CD20
Ocrelizumab6	2017	3	0.37/100,00 patient y
Ofatumumab	2020	None
Ublituximab	2022	None
Other DMTs
Alemtuzumab7	2014	1
Cladribine	2019	None
Glatiramer acetate	1997	None
Interferon beta-1a	1996	None
Interferon beta-1b	1993	None
Teriflunomide	2013	None

Abbreviations: DMT = disease-modifying therapy; MS = multiple sclerosis, PML = progressive multifocal leukoencephalopathy, PwMS = patients with MS.

Reported PML cases for nonnatalizumab DMTs are without known preceding natalizumab exposure. Estimated incidence of PML as available for DMTs discussed herein. PML cases when DMT is used for indications other than MS are not included and risk may be different in these circumstances. Rituximab is used off-label for MS and is not included in the above.

Case Presentation

An otherwise healthy 36-year-old woman was diagnosed with MS in 2002 based on brain MRI and CSF evaluations after optic neuritis. She was started on interferon beta-1a (Figure 1). After 4 relapses over 4 years, she started natalizumab in 2006. She did not have relapses for five years. Over 2 months in 2012, she developed gait difficulty requiring a cane along with memory deficits. Retrospectively captured EDSS was 6.0.¹¹ Brain MRI demonstrated confluent right frontal white matter and diffuse infratentorial T2 hyperintensities, both without associated gadolinium enhancement and on background of prior multifocal demyelinating lesions (Figure 2). Serum JC virus antibody testing was positive (no index available), and CSF JC virus PCR was positive at 15,912 copies/mL (NIH, Bethesda, MD) and 3600 copies/mL (University of Colorado Hospital, Aurora, CO) confirming diagnosis of PML in setting of concordant clinical and radiographic features. Natalizumab was discontinued, and 5 sessions of plasma exchange were completed followed by IV methylprednisolone and oral prednisone for immune inflammatory reconstitution syndrome (IRIS).

Figure 1. Timeline of Clinicoradiographic Features, JC Virus Testing, and Disease-Modifying Therapy Use for the Patient Case.

MS = multiple sclerosis; PML = progressive multifocal leukoencephalopathy; RRMS = relapsing-remitting multiple sclerosis; SPMS = secondary progressive multiple sclerosis.

Figure 2. Sequential Brain MRIs for the Patient Case.

Selected T2 FLAIR MRI slices from the level of the middle cerebellar peduncle (A–D) and centrum semiovale (E–F) demonstrating chronology of PML-associated signal changes from presentation to most recent follow-up study. MS = multiple sclerosis; PML = progressive multifocal leukoencephalopathy.

In February 2012 with CSF JC virus, PCR remained positive at 2598 copies/mL (University of Colorado Hospital). In the context of highly active MS before natalizumab, she restarted DMT in February 2012 with glatiramer acetate chosen due to no known association with PML.¹² In May 2012, repeat CSF JC virus PCR was undetectable using an assay with sensitivity of 250 copies/mL (University of Colorado Hospital), but a more sensitive assay detected 783 copies/mL (Focus Diagnostics, Lyndhurst, NJ). Monthly brain MRIs demonstrated evolution consistent with PML-IRIS for 5 months until July 2012. Brain MRIs completed every 6 months were then stable for 12 months (17 months total on glatiramer acetate) before development of 3 T2 hyperintense lesions in the right temporal and right parietotemporal lobes with associated gadolinium enhancement in July 2013. Repeat CSF JC virus PCR in September 2013 was negative (University of Colorado Hospital). These new lesions were consistent with MS activity, and she transitioned to dimethyl fumarate in December 2013 due to need for higher-efficacy treatment in context of breakthrough disease, as well as fewer reported cases of PML with dimethyl fumarate than fingolimod.¹² Notably, there were no anti-CD20 DMTs FDA-approved for MS in 2013. Retrospectively captured EDSS was 6.0. Surveillance CSF JC virus PCRs remained negative for 5 tests using an ultra-sensitive PCR (NIH, Bethesda, MD) between June 2014 and October 2016. Serum JCV antibody index was positive at 3.49 in July 2014 (STRATIFY, Focus Diagnostics, Cypress, CA). Absolute lymphocyte counts (ALCs) were >0.6 × 10⁹/L. Brain MRIs every 6 months had no evidence of recurrent PML or MS activity.

At 46 months of treatment with dimethyl fumarate, the patient reported worsening leg weakness over the prior 12 months. Retrospectively captured EDSS was 6.5. CSF JCV PCR was negative in April 2017 (University of Colorado Hospital), and brain MRI was stable. With secondary progressive MS, she was transitioned to ocrelizumab in October 2017. Surveillance CSF JC virus PCR was negative in September 2018 (NIH, Bethesda, MD). She had slowed progression while on ocrelizumab with stable brain MRIs every 12 months. Owing to infusion logistics, she transitioned to subcutaneous ofatumumab after 36 months on ocrelizumab. She has been on ofatumumab for 30 months with stable brain MRIs every 12 months. Although on B-cell depleting DMT, the ALC remained >1.0 × 10⁹/L with CD19 counts <17 cells/µL. The IgG nadir was 621 mg/dL with the last value 680 mg/dL. No additional CSF JC virus PCRs have been completed. She was last seen April 2023 with mild progression of foot drop. Retrospectively captured EDSS was 6.5.

Discussion

Treating MS while limiting risk of recurrent PML makes DMT selection after natalizumab-associated PML challenging. For most patients, DMT should be considered because of concern for ongoing disease activity. DMT de-escalation and/or discontinuation should be considered for older patients with prolonged stability before PML. Discussion of potential harms and benefits in the context of patient risk tolerance is fundamental. Here, we used an escalation strategy for management of MS following natalizumab-associated PML and demonstrated that anti-CD20 DMT can be used safely in this context. For patients with ongoing disease activity, use of anti-CD20 DMT provides a high-efficacy alternative to natalizumab. Although approved for primary progressive MS, ocrelizumab may also have utility for inactive SPMS from our experience.

The largest case series on DMT use following natalizumab-associated PML reports 23 patients using glatiramer acetate, interferons, fingolimod, or dimethyl fumarate as first-line or second-line DMT.¹³ With mean 24-month follow-up, 6 patients had relapse or MS activity on MRI, with one breakthrough on fingolimod and the other on dimethyl fumarate. Follow-up was limited to 3.3 months for dimethyl fumarate and 17 months for fingolimod because they were typically used second-line. Additional reports include use of teriflunomide, rituximab, and alemtuzumab.^14-16 No patient developed recurrent PML, but this has been reported in patients with HIV despite immune recovery.¹⁷ DMT may be safe after natalizumab-associated PML, although conclusions are limited by selection bias and follow-up duration.

Anti-CD20 DMTs have rarely been associated with PML in patients with MS who have not been treated with natalizumab. Three cases of PML in patients with MS have been reported in the setting of ocrelizumab use without preceding fingolimod or natalizumab exposure; no cases have been reported with ofatumumab or ublituximab.⁹ How to extrapolate this to risk for recurrence after natalizumab-associated PML is uncertain. Anti-CD20 DMTs cause sustained depletion of peripheral B-cells and potential development of hypogammaglobulinemia.¹⁸ However, anti-CD20 DMTs have had low rates of PML due to mechanisms that contribute to immunosurveillance for JC virus being relatively spared compared with reductions in humoral immunity.¹⁹ However, lymphopenia (including T-cell reductions) occurred in 26% of patients in the ocrelizumab phase III studies.²⁰ T-cell deficiencies are a risk factor for PML.²¹

For any natalizumab-treated patient with MS, discontinuation is associated with risk of rebound disease and transition to anti-CD20 DMT with ocrelizumab may result in lower risk.²² Whether early use of anti-CD20 DMT in patients with natalizumab-associated PML may limit risk of PML-IRIS because natalizumab is discontinued and the immune system is reconstituted is not known.

As natalizumab-associated PML outcomes are better with presymptomatic detection, patients who restart immunosuppressive DMT should be carefully monitored. New lesion(s) on MRI should be followed-up by CSF JC virus PCR with DMT cessation and consideration of experimental PML-specific treatment if positive. Caution is advised on selection of a CSF JC virus PCR; some assays may be insensitive because natalizumab-associated PML has been reported to have JC virus <100 copies/mL in the CSF in up to 30% of cases.²³

This case highlights considerations for managing MS after natalizumab-associated PML and presents an extended period of follow-up with safe use of anti-CD20 DMTs; however, clinical experience remains limited.

Appendix. Authors

Name	Location	Contribution
Andrew B. Wolf, MD	Department of Neurology and Rocky Mountain Multiple Sclerosis Center, University of Colorado School of Medicine, Aurora	Drafting/revision of the manuscript for content, including medical writing for content; study concept or design; analysis or interpretation of data
John R. Corboy, MD	Department of Neurology and Rocky Mountain Multiple Sclerosis Center, University of Colorado School of Medicine, Aurora	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data

Study Funding

The authors report no targeted funding.

Disclosure

A.B. Wolf: research from Genentech and Rocky Mountain MS Center; honoraria from MedLink Neurology and Horizon Therapeutics. J.R. Corboy: research from National Multiple Sclerosis Society, Patient Centered Outcomes Research Institute, Rocky Mountain MS Center, EMD Serono; medical legal work; advisory board for Bristol Myers Squibb editing for Annals of Neurology. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.