Multiple Sclerosis and Related Disorders (2014) 3, 203–210
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/msard
Natalizumab-related progressive multifocal leukoencephalopathy in Greece Dimos D. Mitsikostasa,n, Vasileios Mastorodemosb, Minas Tsagournizakisb, Antonios Kodounisc, Antonios Tsagkaropoulosc, Spyridon Konitsiotisd, Panagiotis Toulase, Alexandros Papadimitriouf, Dimitra Papadimitriouf, Antonios Tavernarakisg, Dimitrios Papadopoulosa,g a
Neurology Department, Athens Naval Hospital, 40 Deinokratous Street, 77A Vasilissis Soﬁas Avenue, 11521 Athens, Greece b Neurology Department, Heraklion University Hospital, Voutes, 71110 Heraklion, Crete, Greece c Neurology Department, 251 General Airforce Hospital, Kanellopoulou 3, 11525 Athens, Greece d Neurology Department, University of Ioannina Medical School, Ioannina, Greece e Radiology Department, Encephalos-EUROMEDICA, Rizariou 3, Halandri, 15233 Athens, Greece f Neurology Department, Henry Dunant Hospital, Mesogeion 107, 11526 Athens, Greece g Neurology Department, Evangelismos General Hospital, Ipsilantou 45-47, 10676 Athens, Greece Received 23 April 2013; received in revised form 12 July 2013; accepted 24 August 2013
KEYWORDS Natalizumab; Progressive multifocal leukoencephalopathy; Immune reconstitution inﬂammatory syndrome; Multiple sclerosis; Treatment; Adverse events
Abstract Background & objectives: Progressive multifocal leukoencephalopathy (PML) may complicate natalizumab treatment in multiple sclerosis patients. We sought to characterize the clinical and laboratory features of natalizumab-related PML (NR-PML) cases from Greece. Methods: Pharmaceutical industry, national drug authorities and all neurology departments within the Greek territory were asked to provide data for cases of NR-PML until October 2012. Collected cases were classiﬁed according to their level of diagnostic certainty using the ﬁvelevel system introduced by Mentzer et al. (2012). Results: Thirteen NR-PML cases were identiﬁed by the neurology departments. Data were provided for only 9 cases. PML manifestations appeared after a median number of 40 (21–52) natalizumab infusions. All but two patients were treated with plasma exchange and some were treated adjunctively with mirtazapine while the others were treated with meﬂoquine. IRIS developed in 6 cases after a median time of 6 (2–10) weeks from PML presentation and were treated with different regimens of corticosteroids. PML was fatal in 3 cases. The median EDSS after a median follow-up time of 12 (8–23) months in the surviving cases was 4.75 (2–8.5). Conclusions: Outcomes for collected NR-PML cases varied from death to returning to baseline. Close surveillance is essential for early diagnosis and treatment of NR-PML patients. & 2013 Elsevier B.V. All rights reserved.
Corresponding author. Tel.: +30 210 7261148, +30 210 7254316; fax: +30 210 7261369. E-mail addresses: [email protected]
, [email protected]
2211-0348/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msard.2013.08.006
D.D. Mitsikostas et al.
Natalizumab is a humanized IgG4 monoclonal antibody that works against the α4 subunit of the VLA4 integrin; licensed for the treatment of highly active relapsing–remitting multiple sclerosis (MS). Its primary mechanism of action is thought to be the prevention of leukocyte migration from the lumen of the cerebral vessels into the brain parenchyma. Natalizumab binds to the VLA-4 molecules on the surface of the leukocytes and prevents interaction of the VLA-4 molecules with their ligands (VCAM1 and ﬁbronectin) on the endothelial surface, which is essential for leukocyte adhesion to the endothelial wall (Engelhardt and Briskin, 2005). Two phase III clinical trials (AFFIRM and SENTINEL) and a post-hoc analysis of the AFFIRM trial have provided evidence of natalizumab's remarkable efﬁcacy as an immunomodulator (Polman et al., 2006; Rudick et al., 2006; Havrdova et al., 2009). Unfortunately, despite its very satisfactory overall safety proﬁle, natalizumab treatment in MS has been complicated with the development of progressive multifocal leukoencephalopathy (PML), which is an opportunistic infection of the central nervous system (CNS) associated with severe morbidity and mortality due to re-activation of latent John Cunningham virus (JCV). PML occurs in the context of HIV infection, immunosuppressive treatments, chemotherapy and organ and bone marrow transplantation. Natalizumab-related PML (NR-PML) was reported for the ﬁrst time in 2005 in two multiple sclerosis patients receiving natalizumab in combination with interferon beta-1a in the SENTINEL trial and shortly after in a third patient participating in a natalizumab trial for Crohn's disease (ENACT-1 & 2) (Kleinschmidt-DeMasters and Tyler, 2005; Van Assche et al., 2005). Nevertheless, the ﬁrst postmarketing case occurred in July 2008 and 323 PML cases have been reported among 110,000 patients treated with natalizumab worldwide as of the end of 2012. PML has also been reported in association with other monoclonal antibodies including rituximab, alemtuzumab, efalizumab, adalimumab, bevacizumab, cetuximab and inﬂiximab, used to treat conditions other than MS (Keene et al., 2011). Interestingly, NR-PML in MS patients exhibits several clinical and radiological differences compared to PML seen under different circumstances (Hellwig and Gold, 2011; Yousry et al., 2012). This study presents 9 cases of NR-PML from Greece focusing on their clinical and radiological features, treatments and outcomes.
Biogen Idec, Genesis Pharma SA (the representative of Biogen for natalizumab distribution in Greece), the Greek National Organization for Medicines, and all Neurology Departments within the Greek territory were asked to provide data for cases of NR-PML up to October 2012. Contacts were made via email (Biogen Idec), personal and ofﬁcial letters (Genesis Pharma SA) or ofﬁcial letters only (National Organization for Medicines). Neurology Departments were approached via telephone. Units performing plasma exchange were also contacted. Treating neurologists had to ﬁll a datasheet with demographic, clinical, MRI,
laboratory and treatment data prior, during and after PML presentation for each NR-PML case. Demographic and clinical data included age, gender, disease duration to PML presentation, comorbid conditions, disease-modifying treatments (DMTs) prior to natalizumab, other medications, EDSS at natalizumab treatment onset, number of natazizumab courses at PML presentation, PML presenting manifestations, presence of seizures, EDSS at PML diagnosis and at latest follow-up and time to PML diagnosis. MRI data sought included lesion localization, presence of Gadolinium enhancement, DWI signal abnormalities and presence of edema. All MRI scans were reviewed by a neuroradiology specialist (PT). Laboratory data including CSF analysis (cell counts, protein levels, JCV PCR), brain biopsy and anti-JCV ELISA antibody status were collected. Information on plasma exchange (PLEX), treatment of IRIS and its manifestations, other treatments used for PML (meﬂoquine, mirtazapine or cidofovir), intensive care requirement, the disease-modifying treatments used after PML–IRIS and MS disease activity after PML–IRIS were collected. IRIS was deﬁned as the clinical deterioration following immune reconstitution with or without MRI changes (Hellwig and Gold, 2011). Given the heterogeneous clinical presentations and the range of paraclinical investigations (MRI, PCR, CSF vs. serum anti-JCV antibody levels, histopathology) that complicate PML diagnosis, a standard 5-level system for deﬁning diagnostic certainty of monoclonal antibody treatmentassociated PML was recently proposed based on the clinical, imaging and laboratory ﬁndings suggestive of the diagnosis (Mentzer et al., 2012). Our group of NR-PML cases were categorized according to the above system for future reference and data comparability. Correlations between EDSS at follow-up and a number of variables (number of natalizumab courses, patient age, time from PML presentation to diagnosis, time from PML presentation to PLEX, duration of DMTs prior to natalizumab, MS disease duration, EDSS at onset, JCV CSF PCR DNA copies, CSF cell counts, CSF protein levels) was calculated with the Spearman's rank correlation coefﬁcient. The GraphPad Prism version 5 software was used for statistical analysis.
Biogen Idec, Genesis Pharma SA and the National Organization for Medicines did not provide data. Only treating neurologists provided data. Thirteen NR-PML cases were identiﬁed by October 2012. Relevant data were collected for 9 NR-PML cases and are summarized in Tables 1–3. Four more NR-PML cases from Greece have been reported before and will not be dealt with here in detail (Travasarou et al., 2012). All data in the text are presented as a median (range) unless otherwise stated.
Demographic and epidemiological data
The median age at PML presentation was 42 years and ranged between 31 and 64 years. Seven of the 9 patients were female. The median disease duration from MS diagnosis to PML was 11 (8–24) years. All but one patient had
Natalizumab-related progressive multifocal leukoencephalopathy in Greece
Table 1 Demographic and clinical data. DMT: disease-modifying treatment, NLZ: natalizumab, GAD: generalized anxiety disorder, MMSE: mini mental state examination.
previously been treated with beta-interferons for 77.5 (27– 96) months. None of the patients had been treated with immunosuppressants prior to the onset of natalizumab. Median EDSS at the onset of treatment with natalizumab was 3.5 (1–6.5). The median number of natalizumab infusions to PML presentation was 40 and ranged between 21 and 52 infusions. Along with the 4 other cases previously published (Travasarou et al., 2012), a total of 13 cases of NR-PML until October 2012 have been diagnosed in Greece from natalizumab becoming commercially available.
The clinical features of the 9 cases reported are summarized in Table 1. PML presented with motor, cognitive, visual and somatosensory deﬁcits with the cognitive functions being most commonly affected in 7 of the 9 patients
followed by the motor deﬁcits (6 of 9 cases). PML–IRIS deﬁned as the clinical deterioration with or without MRI changes was evident in 6 of the 9 cases and became evident after 6 (2–10) weeks from PML presentation. Three PML cases exhibited seizures in the course of their disease that persisted even after PML and IRIS subsided.
PML investigation and diagnosis
In all the cases lesions presented with a subcortical U-ﬁber pattern on T2/FLAIR MRI and involved both the hemispheres in 6 of the 9 cases (Table 2; Fig. 1). The frontal and parietal lobes were the most commonly affected brain regions. All but one case showed restricted diffusion of lesions on DWI– MRI at some point in the disease course. Also, 4 of the 9 cases exhibited Gd enhancement at some point during their disease course. Evidence of mass effect was present in
D.D. Mitsikostas et al.
Table 2 PML investigation. B: bilateral; U: unilateral; O: occipital; P: parietal; F: frontal; rT: right thalamus; ND: not done, NA: not available.
2 cases. JCV PCR was positive in 7 of the 9 cases examined and median JCV DNA copies were 133/mL, ranging from 50 to 6499/mL. The CSF cell count ranged from 0 to 12/μL (median: 5 cells/μL) and protein concentration from 21.3 to 64 mg/dL (median: 45 mg/dL). CSF protein concentration was increased in 3 patients. Cell counts exceeded 5 cells/μL in 4 cases. Data on JCV antibody status was available only from 4 cases and were all positive for JCV antibodies. Brain biopsy was only performed in one of our NR-PML cases and was diagnostic (Table 2). The median time from PML onset to diagnosis was 40 days and ranged between 18 and 180 days (Table 1).
According to the ﬁve-level system recently introduced by Mentzer et al. (2012), 5 cases were diagnosed with level 3 and 4 cases with level 1 diagnostic certainty (Table 2).
PML treatment and outcome
Seven of the 9 PML cases were treated with PLEX within 1–11 weeks from PML presentation. Mirtazapine was used in 7 of the 9 patients in doses ranging from 30 to 90 mg per day for varied time periods (Table 3). Meﬂoquine was used in 3 cases whereas cidofovir was not used in our group of NR-PML cases. All cases of PML–IRIS were treated prophylactically or
Treatment. ICU: intensive care unit, M-P: methyl-prednisolone, DMT: disease-modifying treatment, NK: not known.
Natalizumab-related progressive multifocal leukoencephalopathy in Greece
208 therapeutically with different regimens of corticosteroids for varied time periods, including the 3 that did not exhibit signs or symptoms of IRIS (Table 3). After a follow-up of 12 (5–26) months, 3 cases were fatal. In case #1 death occurred 13 months from presentation after initial stabilization from PML and IRIS and was associated with poor control of seizures as a complication of PML (Papadopoulos et al., 2012). In the second patient (case #2), death occurred due to sepsis 26 months after PML onset, following prolonged coma in intensive care with respiratory support. In the third patient (case #4), death also resulted from sepsis. Along with the 2 fatal cases 4 more NR-PML cases that were described elsewhere (Travasarou et al., 2012), 5 from a total of 13 cases that occurred in Greece were fatal. The EDSS score of the 6 nonfatal cases was 4.75 (2.0–8.5) and 4 patients remained stable after a follow-up period of 12 months (8–23). However, in one case, EDSS exceeded a score of 8.0 (Table 1). Correlation analysis was used to examine the relationship between the EDSS at follow-up and a number of variables (number of natalizumab courses, patient age, time from PML presentation to diagnosis, time from PML presentation to PLEX, duration of DMTs prior to natalizumab, MS disease duration, EDSS at onset, JCV CSF PCR DNA copies, CSF cell counts, CSF protein levels). No statistically signiﬁcant association was found between EDSS at follow-up and any of the above parameters. Following stabilization from PML–IRIS, the majority of the cases (5 of 9) were treated with glatiramer acetate to suppress MS disease activity. Seven of the 8 patients who
D.D. Mitsikostas et al. overcame PML exhibited no clinical or MRI evidence of disease activity during follow-up (Table 3).
PML is a rare yet potentially fatal complication of natalizumab treatment. In this retrospective study we have characterized the clinical and laboratory features of 9 NR-PML cases from Greece To our knowledge this is the ﬁrst pharmaceutical industry-independent report of NR-PML cases. The risk of PML in natalizumab-treated patients was originally calculated to be approximately 0.1% from the data obtained from clinical trials (Yousry et al., 2006). Nevertheless, the risk of PML from current “real-life” data is thought to be approximately 2.4 per 1000 cases (Sørensen et al., 2012). Together with 4 more PML cases from Greece published elsewhere, a total of 13 NR-PML cases have been diagnosed in Greece so far (Travasarou et al., 2012). According to recent data, an equal number of NR-PML cases have been diagnosed in Spain, a country whose population is at least 6-fold higher than Greece (Castillo-Trivino et al., 2012). Differences in the predisposing factors including duration of natalizumab treatment and JCV seropositivity may be responsible for the discrepancies in the incidence of NR-PML between countries. Prior use of immunosuppressants cannot explain the higher incidence of NR-PML in Greece as no patient from our cohort was previously exposed to immunosuppressants. Nevertheless, no valid data on the number of MS patients exposed to natalizumab
Fig. 1 MRI changes in case 1: High signal intensity lesion in the white matter of the right occipital lobe in axial FLAIR- sequence – MRI (B), with mild Gd-enhancement on T1 (C), two weeks from onset of visual deterioration. This lesion is absent in an MRI scan done 9 months earlier (FLAIR) (A). Bilaterally increased FLAIR signal (D) with low signal on T1 sequences (E) involving the occipital lobes bilaterally, seen a month after disease presentation. Anterior parieto-temporal extension of the right occipital lesion seen two months after presentation, with increased signal axial (F) and coronal FLAIR (H) and DWI-MRI (G), coinciding with the development of left hemiparesis.
Natalizumab-related progressive multifocal leukoencephalopathy in Greece is available from Greece and Spain in order to appropriately examine and compare the incidence of NR-PML in the two countries. The clinical manifestations of our cases did not differ from the previously described manifestations of PML. Cognitive dysfunction, behavioral changes, motor symptoms, speech disturbances, visual disturbance, ataxia and seizures are the most frequently encountered ﬁrst symptoms of NRPML appearing insidiously or subacutely (Clifford et al., 2010). Mild cognitive and behavioral changes were the most common presenting symptom in our cases, often underestimated by both the patients and treating physicians. Restoration of the immune response following withdrawal of natalizumab and its removal from the circulation commonly lead to the exaggerated inﬂammatory response of the immune reconstitution inﬂammatory syndrome (IRIS). Although IRIS is thought to occur universally in patients with NR-PML by some workers (Tan et al., 2011) three of our cases did not appear to have developed IRIS. However, despite the lack of clinical deterioration it is possible that the gadolinium enhancement of lesions seen in two of these cases might have been the MRI equivalent of IRIS reﬂecting the inﬂammatory changes associated with it. Additionally, preventive treatment with high dose corticosteroids may have altered the clinical and MRI features of IRIS. PML has been shown to also involve the gray matter leading to the formation of leukocortical and intracortical inﬂammatory demyelinated lesions (Moll et al., 2008). PML lesions involving the cortex although they may not be readily visible with routine MRI techniques could explain the development of seizures in three of our cases. Seizures have been reported to occur in 18% of the PML cases worldwide (Lima et al., 2006). Although MRI has a key role in NR-PML surveillance and diagnosis, identifying PML lesions against the background of pre-existing MS remains challenging. The MRI features of our cases were largely consistent with those described in larger NR-PML cohorts (Yousry et al., 2012). Although MRI surveillance at 6-montlhly intervals is a standard practice for MS patients on natalizumab in Greece none of the NR-PML cases were identiﬁed in the presymptomatic phase via routine MRI. In addition to the typical clinical picture and imaging ﬁndings suggestive of the disease (Bag et al., 2010), PML diagnosis requires conﬁrmation by PCR for JCV DNA detection in CSF. As reported in other NR-PML cases, JCV DNA detection by CSF PCR in some of our cases was delayed or even persistently negative (Clifford et al., 2010; Kuhle et al., 2011). In one of these JCV PCR negative cases deﬁnitive diagnosis was made by brain biopsy. The essential role of anti-JCV antibody index in the risk of PML could not be explored in our cases since this technique was not available at the time of examination. Currently, there is no speciﬁc antiviral treatment with established efﬁcacy against JC virus. Evidence suggests that early withdrawal followed by the removal of circulating natalizumab improves survival and residual disability (Clifford et al., 2010). In our group of PML cases, PLEX was not performed in two cases of which one was severely disabled, whereas the other returned to baseline. Despite the relatively delayed deﬁnitive diagnosis (median = 40 days), PLEX in our cohort was not delayed and initiated on
clinical suspicion. Meﬂoquine, mirtazapine and cidofovir are sometimes used as complementary measures (Clifford et al., 2010). In our group of NR-PML cases only meﬂoquine or mirtazapine were used but the small sample size does not allow comparison of their efﬁcacy. Prevention and treatment of IRIS is an important aim in the management of PML. Intravenous corticosteroids are commonly used to prevent and minimize the clinical deterioration that accompanies IRIS, with some evidence of efﬁcacy (Clifford et al., 2010; Tan et al., 2011). Although treatment of PML, including prevention and treatment of IRIS in Greece was along the lines previously described, the treatment regimens were highly variable. Glatiramer acetate was used as a disease modifying treatment following PML in the majority of the cases with an apparently good response in all but one case, mainly because in most patients it had not been used prior to natalizumab. As most patients following PML were treated with disease modifying drugs it is not possible to draw conclusions regarding any potential changes in activity of MS after PML. Fatality due to PML is estimated to be 18% worldwide (Sørensen et al., 2012). Five of the 13 NR-PML cases from Greece were fatal in total (38.5%), after a long follow-up of 14.5 months. Although the sample size of the Greek NR-PML cases is small it is possible that the proportion of the fatal cases worldwide may have been underestimated, as up to now only short follow-up information was available. Identiﬁed factors associated with increased survival in NR-PML include younger age at presentation, less disability before PML presentation, more localized disease on MRI and shorter time from presentation to diagnosis (Vermersch et al., 2011). The increasing number of cases with a favorable outcome may indicate the increased awareness amongst MS neurologists about the condition (Clifford et al., 2010; Schröder et al., 2010; Kuhle et al., 2011). Nevertheless, non-fatal cases frequently exhibit moderate to severe PMLassociated residual disability although less commonly mild disability or even return to pre-PML baseline have been reported (Lima et al., 2006; Schröder et al., 2010). Lack of disability progression in four of our patients after a long follow-up, indicates that NR-PML is a manageable condition when early and appropriate treatment is offered (Blair et al., 2012). The full range of PML outcomes, from fatal to returning to baseline is reﬂected in our group of PML cases from Greece. Although correlation analysis suggests that disability after PML exhibits no signiﬁcant association with any of the parameters examined the small sample size of our cohort (n= 9) necessitates our ﬁndings be interpreted with caution. Data from the all-increasing number of cases is an invaluable resource for studying the pathogenesis and efﬁcacy of treatments for NR-PML. The authors upon refusal of the pharmaceutical industry to provide the relevant data collected information provided by the treating physicians. Thus, the establishment of an independent NR-PML data registry regulated by an international professional body in order to make the growing experience in this issue readily accessible to MS neurologists is required. In addition, adequate PML risk assessment and surveillance requires MS specialist centers with neuroradiological support in order to minimize the risks of the otherwise largely efﬁcacious natalizumab treatment.
Conﬂict of interest Dr. Mitsikostas has been serving as a member of Medical Advisory Boards for Allergan, Bayer Hellas, Genesis Pharma, Genzyme Hellas, Merck-Serono Hellas, Novartis Hellas and Teva Hellas for the last two years. This study was not funded.
References Bag AK, Cure JK, Chapman PR, Roberson GH, Shah R. JC virus infection of the brain. American Journal of Neuroradiology 2010;31:1564–76. Blair NF, Brew BJ, Halpern JP. Natalizumab-associated PML identiﬁed in the presymptomatic phase using MRI surveillance. Neurology 2012;78:507–8. Castillo-Trivino T, Arroyo R, de Andres C, et al. Progressive multifocal leukoencephalopathy in natalizumab-treated multiple sclerosis patients in Spain: more than expected? Multiple Sclerosis Journal 2012;18:279–508. Clifford DB, De Luca A, Simpson DM, Arendt G, Giovannoni G, Nath A. Natalizumab associated progressive multifocal leukoencephalopathy in patients with multiple sclerosis: lessons from 28 cases. Lancet Neurology 2010;9:438–46. Engelhardt B, Briskin MJ. Therapeutic targeting of alpha 4-integrins in chronic inﬂammatory diseases: tipping the scales of risk towards beneﬁt? European Journal of Immunology 2005;35:2268–73. Havrdova E, Galetta S, Hutchinson M, Stefoski D, Bates D, Polman CH, et al. Effect of natalizumab on clinical and radiological disease activity in multiple sclerosis: a retrospective analysis of the Natalizumab Safety and Efﬁcacy in Relapsing-Remitting Multiple Sclerosis (AFFIRM) study. Lancet Neurology 2009;8:254–60. Hellwig K, Gold R. Progressive multifocal leukoencephalopathy and natalizumab. Journal of Neurology 2011;258:1920–8. Keene DL, Legare C, Taylor E, Gallivan J, Cawthorn GM, Vu D. Monoclonal antibodies and progressive multifocal leukoencephalopathy. Canadian Journal of Neurological Sciences 2011;38:565–71. Kleinschmidt-DeMasters BK, Tyler KL. Progressive multifocal leukoencephalopathy complicating treatment with natalizumab and interferon beta-1a for multiple sclerosis. New England Journal of Medicine 2005;353:369–74. Kuhle J, Gosert R, Bühler R, Derfuss T, Sutter R, Yaldizli O, et al. Management and outcome of CSF-JC virus PCR negative PML in a natalizumab-treated patient with MS. Neurology 2011;77:2010–6. Lima MA, Drislane FW, Koralnik IJ. Seizures and their outcome in progressive multifocal leukoencephalopathy. Neurology 2006; 66:262–4.
D.D. Mitsikostas et al. Mentzer D, Prestel J, Adams O, Gold R, Hartung HP, Hengel H, et al. Case deﬁnition for progressive multifocal leukoencephalopathy following treatment with monoclonal antibodies. Journal of Neurology, Psychiatry and Neurosurgery 2012;83:927–33. Moll NM, Rietsch AM, Ransohoff AJ, Cossoy MB, Huang D, Eichler FS, et al. Cortical demyelination in PML and MS: similarities and differences. Neurology 2008;70:336–43. Papadopoulos D, Plafadeli A, Zafeiropoulou E, et al. Natalizumabrelated PML in multiple sclerosis: a case report. Journal of Neurology 2012;259:S109–10. ([Abstract]). Polman CH, O'Connor PW, Havrdova E, Hutchinson M, Kappos L, Miller DH, et al. AFFIRM investigators. A randomized, placebo-controlled trial of natalizumab for relapsing multiple sclerosis. New England Journal of Medicine 2006;354:889–910. Rudick RA, Stuart WH, Calabresi PA, et al. Natalizumab plus interferon beta-1a for relapsing multiple sclerosis. New England Journal of Medicine 2006;354:911–23. Schröder A, Lee DH, Hellwig K, Lukas C, Linker RA, Gold R. Successful management of natalizumab-associated progressive multifocal leukoencephalopathy and immune reconstitution syndrome in a patient with multiple sclerosis. Archives of Neurology 2010;67:1391–4. Sørensen PS, Bertolotto A, Edan G, Giovannoni G, Gold R, Havrdova E, et al. Risk stratiﬁcation for progressive multifocal leukoencephalopathy in patients treated with natalizumab. Multiple Sclerosis 2012;18:143–52. Tan IL, McArthur JC, Clifford DB, Major EO, Nath A. Immune reconstitution inﬂammatory syndrome in natalizumabassociated PML. Neurology 2011;77:1061–7. Travasarou M, Giannouli E, Kalamatas T, et al. Natalizumab-associated progressive multifocal leucoencephalopathy: lessons from four different cases. Neurology 2012;78:P07.053. ([Abstract]). Van Assche G, Van Ranst M, Sciot R, Dubois B, Vermeire S, Noman M, et al. Progressive multifocal leukoencephalopathy after natalizumab therapy for Crohn's disease. New England Journal of Medicine 2005;353:362–8. Vermersch P, Kappos L, Gold R, Foley JF, Olsson T, Cadavid D, et al. Clinical outcomes of natalizumab-associated progressive multifocal leukoencephalopathy. Neurology 2011;76:1697–704. Yousry TA, Major EO, Ryschkewitsch C, Fahle G, Fischer S, Hou J, et al. Evaluation of patients treated with natalizumab for progressive multifocal leukoencephalopathy. New England Journal of Medicine 2006;354:924–33. Yousry TA, Pelletier D, Cadavid D, Gass A, Richert ND, Radue EW, et al. Magnetic resonance imaging pattern in natalizumabassociated progressive multifocal leukoencephalopathy. Annals of Neurology 2012;72:779–87.