CNS Drugs DOI 10.1007/s40263-014-0168-0
REVIEW ARTICLE
Natalizumab: Risk Stratification of Individual Patients with Multiple Sclerosis Carmen Tur • Xavier Montalban
Ó Springer International Publishing Switzerland 2014
Abstract At present, three risk factors for the development of progressive multifocal leukoencephalopathy (PML) in natalizumab-treated patients have been identified: the presence of antibodies against JC virus (JCV); the duration of natalizumab treatment, especially if longer than 2 years; and the use of immunosuppressants prior to receiving natalizumab. The most commonly used strategy to assess the individual PML risk includes serum anti-JCV antibody testing. Based on the knowledge on all known risk factors, an algorithm for PML risk stratification has been proposed, where patients with the highest PML risk are those with positive anti-JCV antibodies, treatment duration longer than 2 years, with or without prior history of immunosuppression. These patients would have an approximate incidence of PML of 11.1 (with prior immunosuppression) or 4.6 (without prior immunosuppression) cases per 1,000 patients treated with natalizumab (and treatment duration longer than 2 years). In this review, new data on PML risk factors and possible new strategies for PML risk stratification are discussed.
C. Tur (&) X. Montalban Multiple Sclerosis Centre of Catalonia (Cemcat), Department of Neurology-Neuroimmunology, Vall d’Hebron University Hospital, Barcelona, Spain e-mail: [email protected] X. Montalban e-mail: [email protected]
Key Points The presence of serum antibodies against JC virus (JCV), duration of natalizumab treatment longer than 2 years, and the use of immunosuppressants prior to natalizumab treatment are the three risk factors that have been associated with the development of progressive multifocal leukoencephalopathy (PML) in natalizumab-treated patients. At present, serum anti-JCV antibody testing is the most commonly used strategy to assess the individual PML risk. Current recommendations include serum anti-JCV antibody testing before starting natalizumab treatment and then every 6 months. According to the latest algorithm for PML risk stratification, the estimated incidence (per 1,000 natalizumab-treated patients) of PML for JCV seropositive patients with treatment duration longer than 2 years is 11.1 (95 % CI 8.3–14.5) or 4.6 (95 % CI 3.7–5.6), for people with or without prior history of immunosuppression, respectively. Throughout the whole natalizumab period, but especially in JCV seropositive patients who have completed the first 2 years of treatment, frequent MRI scanning (every 3–6 months) is advisable. Whenever clinical or MRI signs suggestive of PML are encountered, natalizumab should be immediately stopped. Should initial serum anti-JCV testing be negative, it must be repeated, and, if suspicion of PML is still high, the current diagnostic algorithm for PML diagnosis should be followed. As soon as PML diagnosis is confirmed, it should be treated, considering the possibility of carrying out plasma exchange.
C. Tur, X. Montalban
1 Introduction Natalizumab (TysabriÒ; Biogen-Idec; Elan Pharmaceuticals, Cambridge, MA, USA), a humanised antibody directed against the a4 subunit of the a4b1 adhesion molecule, was approved for the treatment of multiple sclerosis (MS) in the US in 2004. Its efficacy had been proven in the setting of phase II and phase III clinical trials [1–3]. Since its approval, it has been extensively used and its effectiveness has been proven in day-to-day practice [4]. However, in 2005, natalizumab was voluntarily withdrawn from the market after the appearance of two cases of progressive multifocal leukoencephalopathy (PML) over the follow-up extension of the SENTINEL (Safety and Efficacy of Natalizumab in Combination with Avonex in Patients with Relapsing-Remitting MS) phase III trial, and a third case of PML in a patient with Crohn’s disease who was also participating in a trial with natalizumab [5]. PML is a rare infection of the CNS caused by the John Cunningham virus (JC virus, JCV) [6], which may have fatal consequences in 20 % of those affected, or lead to serious disability in 40 % of survivors [7]. In 2006 it was reintroduced to the market (reintroduced in the United States and released in Europe) together with a Global Risk Management Plan, to be carried out in the US (TOUCHÒ) and the rest of the world (TYGRISÒ). This plan provided updated data on PML cases associated with natalizumab treatment and the risk factors for the development of PML. Up to September 2013, with more than 118,100 patients treated with natalizumab, 401 cases of PML have been reported around the world [8]. Over the last years, a great effort has been made to understand the reasons leading to the development of PML in some patients and not in others and to quantify the risk of PML at the individual level. A more accurate knowledge of this individualised risk will help clinicians to design protocols to minimise the appearance of PML cases amongst natalizumab-treated patients. In this paper, we aim to review all available PML risk data, discuss whether the use of an accurate and updated risk algorithm may reduce the risk of PML, and suggest some recommendations based on such a risk algorithm.
2 Search Strategy For this review, different sources of information were used. Data available in PubMed [9] up to October 2013 were potentially considered to be included in this review, after including the terms ‘multiple sclerosis’, ‘natalizumab’,
‘progressive multifocal leukoencephalopathy’, ‘risk stratification’, and ‘risk management’. Only those papers written in English were selected. Data reported in international conferences and meetings were also considered for inclusion in this review. Finally, the official website of Biogen-IdecÒ was consulted as well to obtain the most recently updated data on natalizumab-treated patients and the total number of PML cases [8].
3 Diagnosis of Progressive Multifocal Leukoencephalopathy (PML) in NatalizumabTreated Patients The diagnosis of PML is based on clinical, imaging, pathological and virological evidence of JCV, as stated in a recently published consensus paper [10]. Importantly, given that JCV can also be found in the normal brain, the sole demonstration of the presence of JCV in the CNS, either in cerebrospinal fluid (CSF) or in brain tissue samples, without clinical, imaging and pathological evidence of PML, is not sufficient for the diagnosis of this infectious complication of natalizumab-treated patients [11]. From the clinical point of view, there is a high heterogeneity of signs and symptoms that a person with PML may have, given that virtually any area of the brain could be involved. Amongst the most common clinical features, there is the presence of cognitive and behavioural symptoms, which may appear in more than 50 % of patients with PML. Also, motor weakness (especially relevant if it denotes supratentorial damage), gait disturbances and visual field deficits may frequently appear, in more than 40 % of cases [10]. Regarding the neuroimaging features of natalizumabassociated PML, lesions are generally large ([3 cm), subcortical, hyperintense in T2/FLAIR (fluid-attenuated inversion recovery) sequences and hypointense in T1 ones. In most cases, natalizumab-related PML lesions are monofocal and localised in the frontal lobes, and exhibit a sharp border toward the cortex and an illdefined border toward the white matter. Importantly, up to 40 % of these lesions may show gadolinium enhancement [12]. As regards the different MRI sequences that can be used for the diagnosis of PML, probably FLAIR images offer the highest sensitivity, although the distinction from MS lesions may still be challenging since both appear hyperintense in FLAIR sequences. Diffusion weighted imaging (DWI) is also used and can help to detect early PML lesions, which are typically also hyperintense in around
Risk Stratification in Natalizumab-Treated MS Patients
100 % of cases (and can therefore be mistakenly considered as acute vascular lesions). As time goes along and the PML lesion increases, a hyperintense rim at the lesion’s edge may be observed instead [10, 12]. From the pathological point of view, PML is characterised by the presence of extensive demyelination, which results from the cytocidal effect of the virus on oligodendroglia, hyperchromatic, enlarged oligodendroglial nuclei, and enlarged bizarre astrocytes with lobulated hyperchromatic nuclei [6, 10]. As regards the virological evidence of PML, JCV can be directly seen through electron microscopic examination or immunohistochemistry. Normally, JCV particles (virions) are observed within oligodendroglial cells, although they can also be seen within reactive astrocytes or macrophages. Alternatively, JCV can be detected through in situ hybridisation and in situ polymerase chain reaction (PCR) Fig. 1 Algorithm for the diagnosis of PML (reproduced with permission from Berger et al. [10]). CNS central nervous system, CSF cerebrospinal fluid, FLAIR fluid-attenuated inversion recovery, JCV JC virus, PCR polymerase chain reaction, PML progressive multifocal leukoencephalopathy, PRES posterior reversible encephalopathy syndrome, VZV varicella-zoster virus
for JCV. Finally, the presence of JCV within the CNS can also be demonstrated through the analysis of CSF using quantitative PCR techniques. Since the sensitivity of the newest quantitative techniques is greater than 95 % [13], in many cases, the demonstration of JCV DNA in the CSF together with clinical and radiological features typical of PML is acceptable to diagnose this disease without the need of a brain biopsy [14]. Importantly, when the clinical picture is highly suggestive of PML, even if the MRI does not suggest PML or the CSF JCV DNA is negative, PML has to be suspected, natalizumab treatment must be suspended, and MRI scanning and CSF analysis need to be repeated in a few days’ time [15]. Therefore, based on all available data and after the development of the consensus statement, an algorithm for PML diagnosis was proposed (Fig. 1) [10].
C. Tur, X. Montalban
4 PML Among Natalizumab-Treated Patients: Updated Data Up to now, 401 cases of PML have been reported [8]. Considering that up to September 2013, around 118,100 patients had received at least one dose of natalizumab worldwide, the estimated overall incidence of PML is 3.3/ 1,000 patients (95 % CI 2.99–3.64 per 1,000 patients). Taking into account the time of exposure to natalizumab treatment, since up to 31 March 2013, there were approximately 279,196 patient-years of natalizumab exposure (excluding 5,100 patients exposed in clinical trials) and 343 PML cases had been reported, the estimated rate of incidence (up to March 2013) would be 1.23/1,000 patientyears. Figure 2 shows the most recent estimated PML incidence by treatment epoch (i.e., the PML incidence depending on the amount of natalizumab doses received). As regards the doses of natalizumab treatment before the diagnosis of PML, they ranged from 8 to 91, with the mean duration of natalizumab treatment being around 40 months (as of September 2013). Most of the PML cases (84 %) occurred in patients who had received more than 24 doses, whereas 16 % occurred in patients with less than 24 doses. From the clinical point of view, approximately 7 % of PML cases were found to be asymptomatic at the time of PML diagnosis, whereas 93 % had clinical symptoms of PML when they were diagnosed [8]. Survival percentages of PML patients as of January 2013 were 100 % and 76.5 % for, respectively, asymptomatic and symptomatic patients. Regarding the disability outcomes of the patients diagnosed with PML, those who were asymptomatic at the
Fig. 2 Natalizumab progressive multifocal leukoencephalopathy (PML) risk estimates by treatment epoch (extracted from Biogen IdecÒ website [8], reproduced with permission). Incidence estimates
time of diagnosis showed lower EDSS (expanded disability status scale) scores [16] both at diagnosis (EDSS 3.6 vs 5.3) and after 6 months of follow-up (EDSS 4.7 vs 6.9). Finally, although most of the PML cases were diagnosed while on natalizumab treatment, a few cases have been diagnosed after the discontinuation of natalizumab. For this reason, it is recommended that patients who discontinue natalizumab treatment be closely monitored for signs and symptoms suggestive of PML for at least 6 months after natalizumab withdrawal [8].
5 Risk Factors for the Development of PML Up to now, three risk factors have been identified. These are the presence of antibodies against JCV, meaning that the person has been in contact with the virus; the duration of natalizumab treatment, with this risk being higher specifically after 24 monthly doses of treatment; and the use of immunosuppressants prior to receiving natalizumab [17]. As regards the JCV, it is a polyomavirus that was first observed through electron microscopy in the brain tissue obtained from a deceased patient who had PML. The infection by JCV is widespread throughout the world and at least 50–60 % of the general population seem to have been in contact with it, affecting men slightly more frequently than women [7, 18]. After primary infection, which is generally asymptomatic, JCV can remain in the renal tissue of healthy individuals with no apparent harm to the host [18]. In around 20–40 % of infected patients, JCV can be detected in urine [19, 20]. Importantly, the form of JCV that has been detected in urine of healthy individuals
by treatment epoch are calculated based on natalizumab exposure through March 31, 2014 and confirmed PML cases as of April 3, 2014
Risk Stratification in Natalizumab-Treated MS Patients
(archetype) differs from the pathogenic form that has been isolated in the brain and CSF of PML cases [21]. Regarding the different factors that may play a part in the development of PML in some patients but not in others, recent studies suggest that archetype JCV can mutate within the lymphoid organs and consequently enter the brain, where, in predisposed (i.e., immunosuppressed) patients, it may cause PML [21]. The JCV genome is divided into three segments: the early region genes, involved in replication and transcription of late region genes; the late region genes, which encode capsid proteins (VP1, VP2 and VP3); and the noncoding control region (NCCR), involved in viral replication and the transcription of early and late genes regions. Thus, among the mutations that have been most commonly found in the pathogenic (neurotrophic) form of JCV of PML cases, there are genetic rearrangements within the NCCR [21–23], which can be seen in up to 100 % of PML cases, and point mutations in VP1, which have been found in 81 % of them. All these mutations seem to alter the tropism ability of JCV, making it more prone to binding with brain cells [24]. Importantly, no genetic mutations have been found in the archetype form of JCV (excreted through urine) of those patients with PML, even after the diagnosis of PML. In addition, no mutations within the VP1-encoding region have been seen without the presence of NCCR rearrangements, whereas NCCR mutations have indeed been observed without VP1gene mutations, maybe suggesting a temporal sequence [22]. Besides, there is some evidence that other factors apart from those directly related to JCV may be involved in the development of PML. For instance, a Swedish study showed that in patients who were to develop PML, an increase in serum anti-JCV antibody levels could be detected prior to the diagnosis of PML [25]. In agreement with these results, another study also suggested that natalizumab-associated PML patients might well have an absent or altered anti-JCV-specific T-cell response [26], maybe implying an altered immune surveillance, which would therefore increase the risk for PML. Along these lines, other authors recently found that patients on natalizumab therapy for long periods of time showed changes in the T-cell receptor repertoire not only in CSF but also in blood [27]. These findings would therefore imply not only a weakening of CNS immune surveillance, thus leading to a higher risk of PML, but also probably an altered immune surveillance peripherally, a condition that might also be involved in the development of PML [27]. Finally, a multicentre European study also found that natalizumabtreated patients who afterwards developed PML had markedly low levels of L-selectin (CD62L)-expressing CD4? T cells prior to PML diagnosis. In this study, where the samples were gathered between 4 and 26 months
before the diagnosis of PML, the authors found that low percentages of L-selectin (CD62L)-expressing CD4? T cells (especially below 15 %) were strongly correlated with the risk of developing PML [28]. Should all these results be confirmed in future studies, new strategies for PML risk stratification can probably be designed. Although other circumstances such as environmental factors might also have an influence on the individual risk of developing PML, no data are currently available supporting this hypothesis. More studies are therefore warranted.
6 PML Risk Stratification At present, the strategy to quantify the individual PML risk includes serum anti-JCV antibody testing, which classifies patients as JCV seropositive or seronegative, together with taking into account length of natalizumab treatment and use of immunosuppressants in the past. As regards anti-JCV antibody testing, in 2011, a robust and reproducible method consisting of a 2-step assay in serum was published [29, 30]. In 2013, the validation of a new, second-generation anti-JCV antibody detection method (enzyme-linked immunosorbent assay [ELISA] kit) was also reported [31]. Before these methods were validated, some other methods, mainly one-step ELISA with lower specificity, were available [29]. They had lower specificity than current 2-step methods. Besides, the discrepancies in JCV seropositivity prevalence also reflected important underlying differences between them [32]. Although some authors found that serum detection of anti-JCV antibodies might still show a certain percentage of false-negative results (around 25 % [33] or 35 % [34]) and other authors proposed the complementary use of urine JCV detection to minimise the rate of such false-negative results [35], a recent study reported that the latest antibody detection technique offered an acceptable, reproducible and precise method to detect the presence of anti-JCV antibodies [31]. Importantly, in this study, which was carried out with patients from three clinical settings, there was good agreement with the 2-step technique previously used, with the overall seroprevalence, as measured through both techniques, being around 50–60 %, as already reported for the adult population [36]. In fact, according to a very recently published update (September 2013), available data suggest that the latest anti-JCV antibody assay is 99 % accurate in stratifying PML risk [8]. Yet, it is to be acknowledged that, for some patients in whom the production of serum anti-JCV antibodies is relatively low despite having been infected by JCV, the risk-stratification algorithms that are mainly based on the detection of such antibodies may not be sufficient to estimate the risk of
C. Tur, X. Montalban
PML. In those cases, the detection of CSF anti-JCV antibodies [37] or the demonstration of JCV viraemia [34] would be more appropriate. These alternative approaches to determine JCV infection become especially relevant when there is suspicion of PML and serum anti-JCV antibodies are repeatedly absent. According to the results of the anti-JCV antibody detection, patients can therefore be classified as JCV seropositive or seronegative. Based on this, taking into account that only JCV seropositive patients have developed PML, and also considering the other two known PML risk factors, an algorithm for PML risk stratification based on available data was first presented at the Annual Meeting of the American Academy of Neurology in May 2011. After this, a number of updates of this very algorithm have taken place, with the last one published by Bloomgren et al. [17]. It was drawn based on the data from 5,896 patients with MS who had not developed PML by the time of the study (and for whom blood samples for anti-JCV antibody testing had been taken), and from 54 patients with MS who, after having started on natalizumab, had developed PML [17]. Table 1 illustrates the current algorithm that should be followed in order to obtain an individual PML risk, according to the presence or absence of the known risk factors [17]. However, it must be acknowledged that risk estimates for PML are only based on the already known risk factors, and this may have important implications from a practical point of view. That is, the outline of this algorithm might drastically change if an up-to-now unknown— but relevant—risk factor were to be unveiled. In addition, although preliminary data published online [8] suggest that PML risk among JCV seropositive patients with no prior history of immunosuppression and more than 4 years of natalizumab treatment is higher than the risk when treatment duration is between 2 and 4 years, data beyond these 4 years of treatment, let alone 6 years, are limited, so that it is difficult to predict the risk of PML beyond that time. More recently, during the last congress of the European Neurological Societies, the possibility was raised of using the index of anti-JCV antibodies, which reflects the titres of antibodies but expressed in a logarithmic scale, as a marker of PML risk among natalizumab-treated patients [38]. In this study, for which blood samples from 71 natalizumabtreated PML patients at least 6 months prior to PML diagnosis and from 2,522 non-PML JCV seropositive patients were analysed, the index of anti-JCV antibodies was significantly associated with the risk of PML among patients with no prior use of immunosuppressant drugs. More specifically, the authors suggested a cut-off point for antibodies’ index of 1.5, over which the risk was significantly greater than below this value, provided that patients had received natalizumab for more than 2 years and had no prior history of immunosuppression. The authors therefore
suggested that this new information could be integrated within the current algorithm to assess individual PML risk [38].
7 Recommendations For PML risk stratification, current recommendations include serum anti-JCV antibody testing before starting natalizumab treatment. Afterwards, it is recommended to perform serum antibody testing every 6 months. Finally, if possible, it would be advisable to assess the serum antiJCV antibody index, especially if patients have received more than 24 monthly infusions of natalizumab and have no prior history of immunosuppression. If, for any reason, patients have undergone plasma exchange, anti-JCV antibody testing should not be performed during that procedure, or for at least 2 weeks following it. For PML risk management, no known interventions have proved effective in preventing PML. Additionally, no specific treatment has proved effective in treating PML if it occurs. Therefore, after 2 years of natalizumab treatment, if a given patient is JCV seropositive, especially if the index is high or there is a prior history of immunosuppression, the clinician must evaluate the risks and benefits of natalizumab treatment carefully, and discuss them with the patient. Throughout the whole natalizumab period, but especially after having completed the first 2 years of treatment, clinical and MRI vigilance are mandatory and remain crucial in the early diagnosis of PML. As regards MRI vigilance, for patients on the upper-risk strata, frequent MRI scanning (every 3–6 months) is advisable. Whenever clinical or MRI signs suggestive of PML are encountered, natalizumab should be immediately stopped. Should initial serum anti-JCV testing be negative, it must be repeated, and, if suspicion of PML is still high, the current diagnostic algorithm for PML diagnosis should be followed (Fig. 1). As soon as PML diagnosis is confirmed, it should be treated, considering the possibility of carrying out plasma exchange. Early diagnosis and aggressive clinical management are associated with a better prognosis, with better survival rates, according to post-marketing data [8].
8 Conclusions Up to now, three risk factors for the development of PML in natalizumab-treated patients have been identified: presence of antibodies against JCV, long duration of natalizumab treatment (especially if longer than 2 years), and use of immunosuppressants prior to receiving natalizumab. The most commonly used strategy to assess the individual
Risk Stratification in Natalizumab-Treated MS Patients Table 1 Estimated PML incidence (per 100 patients treated) according to the latest PML risk algorithma Positive JCV serological statusb
No prior history of IS
Prior history of IS
1–24 natalizumab dosesc PML cases (n)
25
16
Patients treated (n)
44,721
10,043
Estimated incidence (95 % CI)
0.56 (0.36–0.83)
1.6 (0.91–2.6)
25–48 natalizumab doses PML cases (n)
94
52
Patients treated (n)
20,362
4,681
Estimated incidence (95 % CI)
4.6 (3.7–5.6)
11.1 (8.3–14.5)
Negative JCV serological status
Either prior history of IS or not
1–48 natalizumab doses PML cases (n)
1 (hypothetical)
Patients treated (n)
11,625
Estimated incidence (95 % CI)
B0.09 (0–0.48)
This table shows the approximated incidence (per 1,000 patients treated) of PML according to the presence of the three known risk factors for the development of PML: treatment duration, prior immunosuppressant use and presence of serum anti-JCV antibodies 95 % CI 95 % confidence interval, IS immunosuppression, JCV JC virus, PML progressive multifocal leukoencephalopathy a
This table was created according to the latest PML risk algorithm in patients on natalizumab treatment by Bloomgren et al. [17]
b
According to the currently available 2-step ELISA method
c
Monthly intravenously infusion of natalizumab 300 mg
PML risk includes serum anti-JCV antibody testing, together with allowing for the other two risk factors. According to that, an algorithm for PML risk stratification has been defined, where patients with highest PML risk are those with positive anti-JCV antibodies, treatment duration longer than 2 years, with or without prior history of immunosuppression. These patients would have, respectively, an approximate incidence of PML of 11.1 or 4.6 cases per 1,000 patients treated with natalizumab (and treatment duration longer than 2 years). In the near future, it is likely that new and more precise strategies to quantify individual PML risk among natalizumab-treated patients will be available. They will increase the accuracy of PML risk algorithms, which will help clinicians and patients in the decision-making process. This might be especially relevant when considering the possibility of treating patients with natalizumab as a first-line drug; for example, in particularly active forms of the disease. In the meanwhile, the current PML risk algorithm has to be taken as a reference. Therefore, when a patient who is receiving natalizumab has two or three risk factors for developing PML, caution is advised. The decision whether to continue or interrupt natalizumab treatment has to be carefully assessed and the patient and physician should finally come to terms, after discussing all available options.
Disclosures Carmen Tur has received honoraria and support for travel from Merk-Serono, Serono Foundation, Sanofi-Aventis, BayerSchering, Teva Pharmaceuticals, and Novartis. Xavier Montalban has received speaking honoraria and travel expenses for scientific meetings, has been a steering committee member of clinical trials or participated in advisory boards of clinical trials in the past years with Bayer Schering Pharma, Biogen Idec, EMD Merck Serono, Genentech, Genzyme, Novartis, Sanofi-Aventis, Teva Pharmaceuticals and Almirall. The authors did not receive any specific funding for writing this article.
References 1. Miller DH, Khan OA, Sheremata WA, et al. A controlled trial of natalizumab for relapsing multiple sclerosis. N Engl J Med. 2003;348(1):15–23. 2. Polman CH, O’Connor PW, Havrdova E, et al. A randomized, placebo-controlled trial of natalizumab for relapsing multiple sclerosis. N Engl J Med. 2006;354(9):899–910. 3. Rudick RA, Stuart WH, Calabresi PA, et al. Natalizumab plus interferon beta-1a for relapsing multiple sclerosis. N Engl J Med. 2006;354(9):911–23. 4. Horga A, Castillo J, Rio J, et al. An observational study of the effectiveness and safety of natalizumab in the treatment of multiple sclerosis. Rev Neurol. 2011;52(6):321–30. 5. Langer-Gould A, Atlas SW, Green AJ, Bollen AW, Pelletier D. Progressive multifocal leukoencephalopathy in a patient treated with natalizumab. N Engl J Med. 2005;353(4):375–81.
C. Tur, X. Montalban 6. Zurhein G, Chou SM. Particles resembling papova viruses in human cerebral demyelinating disease. Science. 1965;148(3676):1477–9. 7. Sorensen PS, Bertolotto A, Edan G, et al. Risk stratification for progressive multifocal leukoencephalopathy in patients treated with natalizumab. Mult Scler. 2012;18(2):143–52. 8. Biogen IdecÒ Official Site. http://www.biogenidec-international. com/tysabri_fact_sheet.aspx?ID=4698. Accessed Oct 2013. 9. PubMedÒ Official Site. http://www.ncbi.nlm.nih.gov/pubmed. Accessed Oct 2013. 10. Berger JR, Aksamit AJ, Clifford DB, et al. PML diagnostic criteria: consensus statement from the AAN Neuroinfectious Disease Section. Neurology. 2013;80(15):1430–8. 11. White MK, Khalili K. Pathogenesis of progressive multifocal leukoencephalopathy—revisited. J Infect Dis. 2011;203(5):578–86. 12. Yousry TA, Pelletier D, Cadavid D, et al. Magnetic resonance imaging pattern in natalizumab-associated progressive multifocal leukoencephalopathy. Ann Neurol. 2012;72(5):779–87. 13. Iacobaeus E, Ryschkewitsch C, Gravell M, et al. Analysis of cerebrospinal fluid and cerebrospinal fluid cells from patients with multiple sclerosis for detection of JC virus DNA. Mult Scler. 2009;15(1):28–35. 14. Cinque P, Koralnik IJ, Clifford DB. The evolving face of human immunodeficiency virus-related progressive multifocal leukoencephalopathy: defining a consensus terminology. J Neurovirol. 2003;9(Suppl 1):88–92. 15. Boster AL, Nicholas JA, Topalli I, et al. Lessons learned from fatal progressive multifocal leukoencephalopathy in a patient with multiple sclerosis treated with natalizumab. JAMA Neurol. 2013;70(3):398–402. 16. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983;33(11):1444–52. 17. Bloomgren G, Richman S, Hotermans C, et al. Risk of natalizumab-associated progressive multifocal leukoencephalopathy. N Engl J Med. 2012;366(20):1870–80. 18. Chesters PM, Heritage J, McCance DJ. Persistence of DNA sequences of BK virus and JC virus in normal human tissues and in diseased tissues. J Infect Dis. 1983;147(4):676–84. 19. Egli A, Infanti L, Dumoulin A, et al. Prevalence of polyomavirus BK and JC infection and replication in 400 healthy blood donors. J Infect Dis. 2009;199(6):837–46. 20. Zanotta N, Delbue S, Rossi T, et al. Molecular epidemiology of JCV genotypes in patients and healthy subjects from Northern Italy. J Med Virol. 2013;85(7):1286–92. 21. Havla J, Berthele A, Kumpfel T, et al. Co-occurrence of two cases of progressive multifocal leukoencephalopathy in a natalizumab ‘‘infusion group’’. Mult Scler. 2013;19(9):1213–5. 22. Reid CE, Li H, Sur G, et al. Sequencing and analysis of JC virus DNA from natalizumab-treated PML patients. J Infect Dis. 2011;204(2):237–44. 23. Bellizzi A, Anzivino E, Rodio DM, et al. Human Polyomavirus JC monitoring and noncoding control region analysis in dynamic cohorts of individuals affected by immune-mediated diseases
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36. 37.
38.
under treatment with biologics: an observational study. Virol J. 2013;10(1):298. Gorelik L, Reid C, Testa M, et al. Progressive multifocal leukoencephalopathy (PML) development is associated with mutations in JC virus capsid protein VP1 that change its receptor specificity. J Infect Dis. 2011;204(1):103–14. Warnke C, Ramanujam R, Plavina T, et al. Changes to anti-JCV antibody levels in a Swedish national MS cohort. J Neurol Neurosurg Psychiatry. 2013;84(11):1199–205. Perkins MR, Ryschkewitsch C, Liebner JC, et al. Changes in JC virus-specific T cell responses during natalizumab treatment and in natalizumab-associated progressive multifocal leukoencephalopathy. PLoS pathog. 2012;8(11):e1003014. Warnke C, Mausberg AK, Stettner M, et al. Natalizumab affects the T-cell receptor repertoire in patients with multiple sclerosis. Neurology. 2013;81(16):1400–8. Schwab N, Schneider-Hohendorf T, Posevitz V, et al. L-Selectin is a possible biomarker for individual PML risk in natalizumabtreated MS patients. Neurology. 2013;81(10):865–71. Plavina T, Berman M, Njenga M, et al. Multi-site analytical validation of an assay to detect anti-JCV antibodies in human serum and plasma. J Clin Virol. 2012;53(1):65–71. Bozic C, Richman S, Plavina T, et al. Anti-John Cunnigham virus antibody prevalence in multiple sclerosis patients: baseline results of STRATIFY-1. Ann Neurol. 2011;70(5):742–50. Lee P, Plavina T, Castro A, et al. A second-generation ELISA (STRATIFY JCV DxSelect) for detection of JC virus antibodies in human serum and plasma to support progressive multifocal leukoencephalopathy risk stratification. J Clin Virol. 2013;57(2):141–6. Gorelik L, Lerner M, Bixler S, et al. Anti-JC virus antibodies: implications for PML risk stratification. Ann Neurol. 2010;68(3):295–303. Berger JR, Houff SA, Gurwell J, Vega N, Miller CS, Danaher RJ. JC virus antibody status underestimates infection rates. Ann Neurol. 2013;74(1):84–90. Major EO, Frohman E, Douek D. JC viremia in natalizumabtreated patients with multiple sclerosis. N Engl J Med. 2013;368(23):2240–1. Laroni A, Giacomazzi CG, Grimaldi L, et al. Urinary JCV-DNA testing during natalizumab treatment may increase accuracy of PML risk stratification. J Neuroimmune Pharmacol. 2012;7(3):665–72. Kean JM, Rao S, Wang M, Garcea RL. Seroepidemiology of human polyomaviruses. PLoS Pathog. 2009;5(3):e1000363. Lin J, Bettin P, Lee JK, Ho JK, Sadiq SA. Cerebrospinal fluid and serum JC virus antibody detection in multiple sclerosis patients treated with natalizumab. J Neuroimmunol. 2013;261(1–2):123–8. Plavina T, Subramanyam M, Bloomgren G, Richman S, Pace A, Lee S, Schlain B, Campagnolo D, Ticho B. Use of JC virus antibody index to stratify risk of progressive multifocal leukoencephalopathy in natalizumab-treated patients with multiple sclerosis [oral communication]. ENS Congress, Barcelona, June 2013.
REVIEW ARTICLE
Natalizumab: Risk Stratification of Individual Patients with Multiple Sclerosis Carmen Tur • Xavier Montalban
Ó Springer International Publishing Switzerland 2014
Abstract At present, three risk factors for the development of progressive multifocal leukoencephalopathy (PML) in natalizumab-treated patients have been identified: the presence of antibodies against JC virus (JCV); the duration of natalizumab treatment, especially if longer than 2 years; and the use of immunosuppressants prior to receiving natalizumab. The most commonly used strategy to assess the individual PML risk includes serum anti-JCV antibody testing. Based on the knowledge on all known risk factors, an algorithm for PML risk stratification has been proposed, where patients with the highest PML risk are those with positive anti-JCV antibodies, treatment duration longer than 2 years, with or without prior history of immunosuppression. These patients would have an approximate incidence of PML of 11.1 (with prior immunosuppression) or 4.6 (without prior immunosuppression) cases per 1,000 patients treated with natalizumab (and treatment duration longer than 2 years). In this review, new data on PML risk factors and possible new strategies for PML risk stratification are discussed.
C. Tur (&) X. Montalban Multiple Sclerosis Centre of Catalonia (Cemcat), Department of Neurology-Neuroimmunology, Vall d’Hebron University Hospital, Barcelona, Spain e-mail: [email protected] X. Montalban e-mail: [email protected]
Key Points The presence of serum antibodies against JC virus (JCV), duration of natalizumab treatment longer than 2 years, and the use of immunosuppressants prior to natalizumab treatment are the three risk factors that have been associated with the development of progressive multifocal leukoencephalopathy (PML) in natalizumab-treated patients. At present, serum anti-JCV antibody testing is the most commonly used strategy to assess the individual PML risk. Current recommendations include serum anti-JCV antibody testing before starting natalizumab treatment and then every 6 months. According to the latest algorithm for PML risk stratification, the estimated incidence (per 1,000 natalizumab-treated patients) of PML for JCV seropositive patients with treatment duration longer than 2 years is 11.1 (95 % CI 8.3–14.5) or 4.6 (95 % CI 3.7–5.6), for people with or without prior history of immunosuppression, respectively. Throughout the whole natalizumab period, but especially in JCV seropositive patients who have completed the first 2 years of treatment, frequent MRI scanning (every 3–6 months) is advisable. Whenever clinical or MRI signs suggestive of PML are encountered, natalizumab should be immediately stopped. Should initial serum anti-JCV testing be negative, it must be repeated, and, if suspicion of PML is still high, the current diagnostic algorithm for PML diagnosis should be followed. As soon as PML diagnosis is confirmed, it should be treated, considering the possibility of carrying out plasma exchange.
C. Tur, X. Montalban
1 Introduction Natalizumab (TysabriÒ; Biogen-Idec; Elan Pharmaceuticals, Cambridge, MA, USA), a humanised antibody directed against the a4 subunit of the a4b1 adhesion molecule, was approved for the treatment of multiple sclerosis (MS) in the US in 2004. Its efficacy had been proven in the setting of phase II and phase III clinical trials [1–3]. Since its approval, it has been extensively used and its effectiveness has been proven in day-to-day practice [4]. However, in 2005, natalizumab was voluntarily withdrawn from the market after the appearance of two cases of progressive multifocal leukoencephalopathy (PML) over the follow-up extension of the SENTINEL (Safety and Efficacy of Natalizumab in Combination with Avonex in Patients with Relapsing-Remitting MS) phase III trial, and a third case of PML in a patient with Crohn’s disease who was also participating in a trial with natalizumab [5]. PML is a rare infection of the CNS caused by the John Cunningham virus (JC virus, JCV) [6], which may have fatal consequences in 20 % of those affected, or lead to serious disability in 40 % of survivors [7]. In 2006 it was reintroduced to the market (reintroduced in the United States and released in Europe) together with a Global Risk Management Plan, to be carried out in the US (TOUCHÒ) and the rest of the world (TYGRISÒ). This plan provided updated data on PML cases associated with natalizumab treatment and the risk factors for the development of PML. Up to September 2013, with more than 118,100 patients treated with natalizumab, 401 cases of PML have been reported around the world [8]. Over the last years, a great effort has been made to understand the reasons leading to the development of PML in some patients and not in others and to quantify the risk of PML at the individual level. A more accurate knowledge of this individualised risk will help clinicians to design protocols to minimise the appearance of PML cases amongst natalizumab-treated patients. In this paper, we aim to review all available PML risk data, discuss whether the use of an accurate and updated risk algorithm may reduce the risk of PML, and suggest some recommendations based on such a risk algorithm.
2 Search Strategy For this review, different sources of information were used. Data available in PubMed [9] up to October 2013 were potentially considered to be included in this review, after including the terms ‘multiple sclerosis’, ‘natalizumab’,
‘progressive multifocal leukoencephalopathy’, ‘risk stratification’, and ‘risk management’. Only those papers written in English were selected. Data reported in international conferences and meetings were also considered for inclusion in this review. Finally, the official website of Biogen-IdecÒ was consulted as well to obtain the most recently updated data on natalizumab-treated patients and the total number of PML cases [8].
3 Diagnosis of Progressive Multifocal Leukoencephalopathy (PML) in NatalizumabTreated Patients The diagnosis of PML is based on clinical, imaging, pathological and virological evidence of JCV, as stated in a recently published consensus paper [10]. Importantly, given that JCV can also be found in the normal brain, the sole demonstration of the presence of JCV in the CNS, either in cerebrospinal fluid (CSF) or in brain tissue samples, without clinical, imaging and pathological evidence of PML, is not sufficient for the diagnosis of this infectious complication of natalizumab-treated patients [11]. From the clinical point of view, there is a high heterogeneity of signs and symptoms that a person with PML may have, given that virtually any area of the brain could be involved. Amongst the most common clinical features, there is the presence of cognitive and behavioural symptoms, which may appear in more than 50 % of patients with PML. Also, motor weakness (especially relevant if it denotes supratentorial damage), gait disturbances and visual field deficits may frequently appear, in more than 40 % of cases [10]. Regarding the neuroimaging features of natalizumabassociated PML, lesions are generally large ([3 cm), subcortical, hyperintense in T2/FLAIR (fluid-attenuated inversion recovery) sequences and hypointense in T1 ones. In most cases, natalizumab-related PML lesions are monofocal and localised in the frontal lobes, and exhibit a sharp border toward the cortex and an illdefined border toward the white matter. Importantly, up to 40 % of these lesions may show gadolinium enhancement [12]. As regards the different MRI sequences that can be used for the diagnosis of PML, probably FLAIR images offer the highest sensitivity, although the distinction from MS lesions may still be challenging since both appear hyperintense in FLAIR sequences. Diffusion weighted imaging (DWI) is also used and can help to detect early PML lesions, which are typically also hyperintense in around
Risk Stratification in Natalizumab-Treated MS Patients
100 % of cases (and can therefore be mistakenly considered as acute vascular lesions). As time goes along and the PML lesion increases, a hyperintense rim at the lesion’s edge may be observed instead [10, 12]. From the pathological point of view, PML is characterised by the presence of extensive demyelination, which results from the cytocidal effect of the virus on oligodendroglia, hyperchromatic, enlarged oligodendroglial nuclei, and enlarged bizarre astrocytes with lobulated hyperchromatic nuclei [6, 10]. As regards the virological evidence of PML, JCV can be directly seen through electron microscopic examination or immunohistochemistry. Normally, JCV particles (virions) are observed within oligodendroglial cells, although they can also be seen within reactive astrocytes or macrophages. Alternatively, JCV can be detected through in situ hybridisation and in situ polymerase chain reaction (PCR) Fig. 1 Algorithm for the diagnosis of PML (reproduced with permission from Berger et al. [10]). CNS central nervous system, CSF cerebrospinal fluid, FLAIR fluid-attenuated inversion recovery, JCV JC virus, PCR polymerase chain reaction, PML progressive multifocal leukoencephalopathy, PRES posterior reversible encephalopathy syndrome, VZV varicella-zoster virus
for JCV. Finally, the presence of JCV within the CNS can also be demonstrated through the analysis of CSF using quantitative PCR techniques. Since the sensitivity of the newest quantitative techniques is greater than 95 % [13], in many cases, the demonstration of JCV DNA in the CSF together with clinical and radiological features typical of PML is acceptable to diagnose this disease without the need of a brain biopsy [14]. Importantly, when the clinical picture is highly suggestive of PML, even if the MRI does not suggest PML or the CSF JCV DNA is negative, PML has to be suspected, natalizumab treatment must be suspended, and MRI scanning and CSF analysis need to be repeated in a few days’ time [15]. Therefore, based on all available data and after the development of the consensus statement, an algorithm for PML diagnosis was proposed (Fig. 1) [10].
C. Tur, X. Montalban
4 PML Among Natalizumab-Treated Patients: Updated Data Up to now, 401 cases of PML have been reported [8]. Considering that up to September 2013, around 118,100 patients had received at least one dose of natalizumab worldwide, the estimated overall incidence of PML is 3.3/ 1,000 patients (95 % CI 2.99–3.64 per 1,000 patients). Taking into account the time of exposure to natalizumab treatment, since up to 31 March 2013, there were approximately 279,196 patient-years of natalizumab exposure (excluding 5,100 patients exposed in clinical trials) and 343 PML cases had been reported, the estimated rate of incidence (up to March 2013) would be 1.23/1,000 patientyears. Figure 2 shows the most recent estimated PML incidence by treatment epoch (i.e., the PML incidence depending on the amount of natalizumab doses received). As regards the doses of natalizumab treatment before the diagnosis of PML, they ranged from 8 to 91, with the mean duration of natalizumab treatment being around 40 months (as of September 2013). Most of the PML cases (84 %) occurred in patients who had received more than 24 doses, whereas 16 % occurred in patients with less than 24 doses. From the clinical point of view, approximately 7 % of PML cases were found to be asymptomatic at the time of PML diagnosis, whereas 93 % had clinical symptoms of PML when they were diagnosed [8]. Survival percentages of PML patients as of January 2013 were 100 % and 76.5 % for, respectively, asymptomatic and symptomatic patients. Regarding the disability outcomes of the patients diagnosed with PML, those who were asymptomatic at the
Fig. 2 Natalizumab progressive multifocal leukoencephalopathy (PML) risk estimates by treatment epoch (extracted from Biogen IdecÒ website [8], reproduced with permission). Incidence estimates
time of diagnosis showed lower EDSS (expanded disability status scale) scores [16] both at diagnosis (EDSS 3.6 vs 5.3) and after 6 months of follow-up (EDSS 4.7 vs 6.9). Finally, although most of the PML cases were diagnosed while on natalizumab treatment, a few cases have been diagnosed after the discontinuation of natalizumab. For this reason, it is recommended that patients who discontinue natalizumab treatment be closely monitored for signs and symptoms suggestive of PML for at least 6 months after natalizumab withdrawal [8].
5 Risk Factors for the Development of PML Up to now, three risk factors have been identified. These are the presence of antibodies against JCV, meaning that the person has been in contact with the virus; the duration of natalizumab treatment, with this risk being higher specifically after 24 monthly doses of treatment; and the use of immunosuppressants prior to receiving natalizumab [17]. As regards the JCV, it is a polyomavirus that was first observed through electron microscopy in the brain tissue obtained from a deceased patient who had PML. The infection by JCV is widespread throughout the world and at least 50–60 % of the general population seem to have been in contact with it, affecting men slightly more frequently than women [7, 18]. After primary infection, which is generally asymptomatic, JCV can remain in the renal tissue of healthy individuals with no apparent harm to the host [18]. In around 20–40 % of infected patients, JCV can be detected in urine [19, 20]. Importantly, the form of JCV that has been detected in urine of healthy individuals
by treatment epoch are calculated based on natalizumab exposure through March 31, 2014 and confirmed PML cases as of April 3, 2014
Risk Stratification in Natalizumab-Treated MS Patients
(archetype) differs from the pathogenic form that has been isolated in the brain and CSF of PML cases [21]. Regarding the different factors that may play a part in the development of PML in some patients but not in others, recent studies suggest that archetype JCV can mutate within the lymphoid organs and consequently enter the brain, where, in predisposed (i.e., immunosuppressed) patients, it may cause PML [21]. The JCV genome is divided into three segments: the early region genes, involved in replication and transcription of late region genes; the late region genes, which encode capsid proteins (VP1, VP2 and VP3); and the noncoding control region (NCCR), involved in viral replication and the transcription of early and late genes regions. Thus, among the mutations that have been most commonly found in the pathogenic (neurotrophic) form of JCV of PML cases, there are genetic rearrangements within the NCCR [21–23], which can be seen in up to 100 % of PML cases, and point mutations in VP1, which have been found in 81 % of them. All these mutations seem to alter the tropism ability of JCV, making it more prone to binding with brain cells [24]. Importantly, no genetic mutations have been found in the archetype form of JCV (excreted through urine) of those patients with PML, even after the diagnosis of PML. In addition, no mutations within the VP1-encoding region have been seen without the presence of NCCR rearrangements, whereas NCCR mutations have indeed been observed without VP1gene mutations, maybe suggesting a temporal sequence [22]. Besides, there is some evidence that other factors apart from those directly related to JCV may be involved in the development of PML. For instance, a Swedish study showed that in patients who were to develop PML, an increase in serum anti-JCV antibody levels could be detected prior to the diagnosis of PML [25]. In agreement with these results, another study also suggested that natalizumab-associated PML patients might well have an absent or altered anti-JCV-specific T-cell response [26], maybe implying an altered immune surveillance, which would therefore increase the risk for PML. Along these lines, other authors recently found that patients on natalizumab therapy for long periods of time showed changes in the T-cell receptor repertoire not only in CSF but also in blood [27]. These findings would therefore imply not only a weakening of CNS immune surveillance, thus leading to a higher risk of PML, but also probably an altered immune surveillance peripherally, a condition that might also be involved in the development of PML [27]. Finally, a multicentre European study also found that natalizumabtreated patients who afterwards developed PML had markedly low levels of L-selectin (CD62L)-expressing CD4? T cells prior to PML diagnosis. In this study, where the samples were gathered between 4 and 26 months
before the diagnosis of PML, the authors found that low percentages of L-selectin (CD62L)-expressing CD4? T cells (especially below 15 %) were strongly correlated with the risk of developing PML [28]. Should all these results be confirmed in future studies, new strategies for PML risk stratification can probably be designed. Although other circumstances such as environmental factors might also have an influence on the individual risk of developing PML, no data are currently available supporting this hypothesis. More studies are therefore warranted.
6 PML Risk Stratification At present, the strategy to quantify the individual PML risk includes serum anti-JCV antibody testing, which classifies patients as JCV seropositive or seronegative, together with taking into account length of natalizumab treatment and use of immunosuppressants in the past. As regards anti-JCV antibody testing, in 2011, a robust and reproducible method consisting of a 2-step assay in serum was published [29, 30]. In 2013, the validation of a new, second-generation anti-JCV antibody detection method (enzyme-linked immunosorbent assay [ELISA] kit) was also reported [31]. Before these methods were validated, some other methods, mainly one-step ELISA with lower specificity, were available [29]. They had lower specificity than current 2-step methods. Besides, the discrepancies in JCV seropositivity prevalence also reflected important underlying differences between them [32]. Although some authors found that serum detection of anti-JCV antibodies might still show a certain percentage of false-negative results (around 25 % [33] or 35 % [34]) and other authors proposed the complementary use of urine JCV detection to minimise the rate of such false-negative results [35], a recent study reported that the latest antibody detection technique offered an acceptable, reproducible and precise method to detect the presence of anti-JCV antibodies [31]. Importantly, in this study, which was carried out with patients from three clinical settings, there was good agreement with the 2-step technique previously used, with the overall seroprevalence, as measured through both techniques, being around 50–60 %, as already reported for the adult population [36]. In fact, according to a very recently published update (September 2013), available data suggest that the latest anti-JCV antibody assay is 99 % accurate in stratifying PML risk [8]. Yet, it is to be acknowledged that, for some patients in whom the production of serum anti-JCV antibodies is relatively low despite having been infected by JCV, the risk-stratification algorithms that are mainly based on the detection of such antibodies may not be sufficient to estimate the risk of
C. Tur, X. Montalban
PML. In those cases, the detection of CSF anti-JCV antibodies [37] or the demonstration of JCV viraemia [34] would be more appropriate. These alternative approaches to determine JCV infection become especially relevant when there is suspicion of PML and serum anti-JCV antibodies are repeatedly absent. According to the results of the anti-JCV antibody detection, patients can therefore be classified as JCV seropositive or seronegative. Based on this, taking into account that only JCV seropositive patients have developed PML, and also considering the other two known PML risk factors, an algorithm for PML risk stratification based on available data was first presented at the Annual Meeting of the American Academy of Neurology in May 2011. After this, a number of updates of this very algorithm have taken place, with the last one published by Bloomgren et al. [17]. It was drawn based on the data from 5,896 patients with MS who had not developed PML by the time of the study (and for whom blood samples for anti-JCV antibody testing had been taken), and from 54 patients with MS who, after having started on natalizumab, had developed PML [17]. Table 1 illustrates the current algorithm that should be followed in order to obtain an individual PML risk, according to the presence or absence of the known risk factors [17]. However, it must be acknowledged that risk estimates for PML are only based on the already known risk factors, and this may have important implications from a practical point of view. That is, the outline of this algorithm might drastically change if an up-to-now unknown— but relevant—risk factor were to be unveiled. In addition, although preliminary data published online [8] suggest that PML risk among JCV seropositive patients with no prior history of immunosuppression and more than 4 years of natalizumab treatment is higher than the risk when treatment duration is between 2 and 4 years, data beyond these 4 years of treatment, let alone 6 years, are limited, so that it is difficult to predict the risk of PML beyond that time. More recently, during the last congress of the European Neurological Societies, the possibility was raised of using the index of anti-JCV antibodies, which reflects the titres of antibodies but expressed in a logarithmic scale, as a marker of PML risk among natalizumab-treated patients [38]. In this study, for which blood samples from 71 natalizumabtreated PML patients at least 6 months prior to PML diagnosis and from 2,522 non-PML JCV seropositive patients were analysed, the index of anti-JCV antibodies was significantly associated with the risk of PML among patients with no prior use of immunosuppressant drugs. More specifically, the authors suggested a cut-off point for antibodies’ index of 1.5, over which the risk was significantly greater than below this value, provided that patients had received natalizumab for more than 2 years and had no prior history of immunosuppression. The authors therefore
suggested that this new information could be integrated within the current algorithm to assess individual PML risk [38].
7 Recommendations For PML risk stratification, current recommendations include serum anti-JCV antibody testing before starting natalizumab treatment. Afterwards, it is recommended to perform serum antibody testing every 6 months. Finally, if possible, it would be advisable to assess the serum antiJCV antibody index, especially if patients have received more than 24 monthly infusions of natalizumab and have no prior history of immunosuppression. If, for any reason, patients have undergone plasma exchange, anti-JCV antibody testing should not be performed during that procedure, or for at least 2 weeks following it. For PML risk management, no known interventions have proved effective in preventing PML. Additionally, no specific treatment has proved effective in treating PML if it occurs. Therefore, after 2 years of natalizumab treatment, if a given patient is JCV seropositive, especially if the index is high or there is a prior history of immunosuppression, the clinician must evaluate the risks and benefits of natalizumab treatment carefully, and discuss them with the patient. Throughout the whole natalizumab period, but especially after having completed the first 2 years of treatment, clinical and MRI vigilance are mandatory and remain crucial in the early diagnosis of PML. As regards MRI vigilance, for patients on the upper-risk strata, frequent MRI scanning (every 3–6 months) is advisable. Whenever clinical or MRI signs suggestive of PML are encountered, natalizumab should be immediately stopped. Should initial serum anti-JCV testing be negative, it must be repeated, and, if suspicion of PML is still high, the current diagnostic algorithm for PML diagnosis should be followed (Fig. 1). As soon as PML diagnosis is confirmed, it should be treated, considering the possibility of carrying out plasma exchange. Early diagnosis and aggressive clinical management are associated with a better prognosis, with better survival rates, according to post-marketing data [8].
8 Conclusions Up to now, three risk factors for the development of PML in natalizumab-treated patients have been identified: presence of antibodies against JCV, long duration of natalizumab treatment (especially if longer than 2 years), and use of immunosuppressants prior to receiving natalizumab. The most commonly used strategy to assess the individual
Risk Stratification in Natalizumab-Treated MS Patients Table 1 Estimated PML incidence (per 100 patients treated) according to the latest PML risk algorithma Positive JCV serological statusb
No prior history of IS
Prior history of IS
1–24 natalizumab dosesc PML cases (n)
25
16
Patients treated (n)
44,721
10,043
Estimated incidence (95 % CI)
0.56 (0.36–0.83)
1.6 (0.91–2.6)
25–48 natalizumab doses PML cases (n)
94
52
Patients treated (n)
20,362
4,681
Estimated incidence (95 % CI)
4.6 (3.7–5.6)
11.1 (8.3–14.5)
Negative JCV serological status
Either prior history of IS or not
1–48 natalizumab doses PML cases (n)
1 (hypothetical)
Patients treated (n)
11,625
Estimated incidence (95 % CI)
B0.09 (0–0.48)
This table shows the approximated incidence (per 1,000 patients treated) of PML according to the presence of the three known risk factors for the development of PML: treatment duration, prior immunosuppressant use and presence of serum anti-JCV antibodies 95 % CI 95 % confidence interval, IS immunosuppression, JCV JC virus, PML progressive multifocal leukoencephalopathy a
This table was created according to the latest PML risk algorithm in patients on natalizumab treatment by Bloomgren et al. [17]
b
According to the currently available 2-step ELISA method
c
Monthly intravenously infusion of natalizumab 300 mg
PML risk includes serum anti-JCV antibody testing, together with allowing for the other two risk factors. According to that, an algorithm for PML risk stratification has been defined, where patients with highest PML risk are those with positive anti-JCV antibodies, treatment duration longer than 2 years, with or without prior history of immunosuppression. These patients would have, respectively, an approximate incidence of PML of 11.1 or 4.6 cases per 1,000 patients treated with natalizumab (and treatment duration longer than 2 years). In the near future, it is likely that new and more precise strategies to quantify individual PML risk among natalizumab-treated patients will be available. They will increase the accuracy of PML risk algorithms, which will help clinicians and patients in the decision-making process. This might be especially relevant when considering the possibility of treating patients with natalizumab as a first-line drug; for example, in particularly active forms of the disease. In the meanwhile, the current PML risk algorithm has to be taken as a reference. Therefore, when a patient who is receiving natalizumab has two or three risk factors for developing PML, caution is advised. The decision whether to continue or interrupt natalizumab treatment has to be carefully assessed and the patient and physician should finally come to terms, after discussing all available options.
Disclosures Carmen Tur has received honoraria and support for travel from Merk-Serono, Serono Foundation, Sanofi-Aventis, BayerSchering, Teva Pharmaceuticals, and Novartis. Xavier Montalban has received speaking honoraria and travel expenses for scientific meetings, has been a steering committee member of clinical trials or participated in advisory boards of clinical trials in the past years with Bayer Schering Pharma, Biogen Idec, EMD Merck Serono, Genentech, Genzyme, Novartis, Sanofi-Aventis, Teva Pharmaceuticals and Almirall. The authors did not receive any specific funding for writing this article.
References 1. Miller DH, Khan OA, Sheremata WA, et al. A controlled trial of natalizumab for relapsing multiple sclerosis. N Engl J Med. 2003;348(1):15–23. 2. Polman CH, O’Connor PW, Havrdova E, et al. A randomized, placebo-controlled trial of natalizumab for relapsing multiple sclerosis. N Engl J Med. 2006;354(9):899–910. 3. Rudick RA, Stuart WH, Calabresi PA, et al. Natalizumab plus interferon beta-1a for relapsing multiple sclerosis. N Engl J Med. 2006;354(9):911–23. 4. Horga A, Castillo J, Rio J, et al. An observational study of the effectiveness and safety of natalizumab in the treatment of multiple sclerosis. Rev Neurol. 2011;52(6):321–30. 5. Langer-Gould A, Atlas SW, Green AJ, Bollen AW, Pelletier D. Progressive multifocal leukoencephalopathy in a patient treated with natalizumab. N Engl J Med. 2005;353(4):375–81.
C. Tur, X. Montalban 6. Zurhein G, Chou SM. Particles resembling papova viruses in human cerebral demyelinating disease. Science. 1965;148(3676):1477–9. 7. Sorensen PS, Bertolotto A, Edan G, et al. Risk stratification for progressive multifocal leukoencephalopathy in patients treated with natalizumab. Mult Scler. 2012;18(2):143–52. 8. Biogen IdecÒ Official Site. http://www.biogenidec-international. com/tysabri_fact_sheet.aspx?ID=4698. Accessed Oct 2013. 9. PubMedÒ Official Site. http://www.ncbi.nlm.nih.gov/pubmed. Accessed Oct 2013. 10. Berger JR, Aksamit AJ, Clifford DB, et al. PML diagnostic criteria: consensus statement from the AAN Neuroinfectious Disease Section. Neurology. 2013;80(15):1430–8. 11. White MK, Khalili K. Pathogenesis of progressive multifocal leukoencephalopathy—revisited. J Infect Dis. 2011;203(5):578–86. 12. Yousry TA, Pelletier D, Cadavid D, et al. Magnetic resonance imaging pattern in natalizumab-associated progressive multifocal leukoencephalopathy. Ann Neurol. 2012;72(5):779–87. 13. Iacobaeus E, Ryschkewitsch C, Gravell M, et al. Analysis of cerebrospinal fluid and cerebrospinal fluid cells from patients with multiple sclerosis for detection of JC virus DNA. Mult Scler. 2009;15(1):28–35. 14. Cinque P, Koralnik IJ, Clifford DB. The evolving face of human immunodeficiency virus-related progressive multifocal leukoencephalopathy: defining a consensus terminology. J Neurovirol. 2003;9(Suppl 1):88–92. 15. Boster AL, Nicholas JA, Topalli I, et al. Lessons learned from fatal progressive multifocal leukoencephalopathy in a patient with multiple sclerosis treated with natalizumab. JAMA Neurol. 2013;70(3):398–402. 16. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983;33(11):1444–52. 17. Bloomgren G, Richman S, Hotermans C, et al. Risk of natalizumab-associated progressive multifocal leukoencephalopathy. N Engl J Med. 2012;366(20):1870–80. 18. Chesters PM, Heritage J, McCance DJ. Persistence of DNA sequences of BK virus and JC virus in normal human tissues and in diseased tissues. J Infect Dis. 1983;147(4):676–84. 19. Egli A, Infanti L, Dumoulin A, et al. Prevalence of polyomavirus BK and JC infection and replication in 400 healthy blood donors. J Infect Dis. 2009;199(6):837–46. 20. Zanotta N, Delbue S, Rossi T, et al. Molecular epidemiology of JCV genotypes in patients and healthy subjects from Northern Italy. J Med Virol. 2013;85(7):1286–92. 21. Havla J, Berthele A, Kumpfel T, et al. Co-occurrence of two cases of progressive multifocal leukoencephalopathy in a natalizumab ‘‘infusion group’’. Mult Scler. 2013;19(9):1213–5. 22. Reid CE, Li H, Sur G, et al. Sequencing and analysis of JC virus DNA from natalizumab-treated PML patients. J Infect Dis. 2011;204(2):237–44. 23. Bellizzi A, Anzivino E, Rodio DM, et al. Human Polyomavirus JC monitoring and noncoding control region analysis in dynamic cohorts of individuals affected by immune-mediated diseases
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36. 37.
38.
under treatment with biologics: an observational study. Virol J. 2013;10(1):298. Gorelik L, Reid C, Testa M, et al. Progressive multifocal leukoencephalopathy (PML) development is associated with mutations in JC virus capsid protein VP1 that change its receptor specificity. J Infect Dis. 2011;204(1):103–14. Warnke C, Ramanujam R, Plavina T, et al. Changes to anti-JCV antibody levels in a Swedish national MS cohort. J Neurol Neurosurg Psychiatry. 2013;84(11):1199–205. Perkins MR, Ryschkewitsch C, Liebner JC, et al. Changes in JC virus-specific T cell responses during natalizumab treatment and in natalizumab-associated progressive multifocal leukoencephalopathy. PLoS pathog. 2012;8(11):e1003014. Warnke C, Mausberg AK, Stettner M, et al. Natalizumab affects the T-cell receptor repertoire in patients with multiple sclerosis. Neurology. 2013;81(16):1400–8. Schwab N, Schneider-Hohendorf T, Posevitz V, et al. L-Selectin is a possible biomarker for individual PML risk in natalizumabtreated MS patients. Neurology. 2013;81(10):865–71. Plavina T, Berman M, Njenga M, et al. Multi-site analytical validation of an assay to detect anti-JCV antibodies in human serum and plasma. J Clin Virol. 2012;53(1):65–71. Bozic C, Richman S, Plavina T, et al. Anti-John Cunnigham virus antibody prevalence in multiple sclerosis patients: baseline results of STRATIFY-1. Ann Neurol. 2011;70(5):742–50. Lee P, Plavina T, Castro A, et al. A second-generation ELISA (STRATIFY JCV DxSelect) for detection of JC virus antibodies in human serum and plasma to support progressive multifocal leukoencephalopathy risk stratification. J Clin Virol. 2013;57(2):141–6. Gorelik L, Lerner M, Bixler S, et al. Anti-JC virus antibodies: implications for PML risk stratification. Ann Neurol. 2010;68(3):295–303. Berger JR, Houff SA, Gurwell J, Vega N, Miller CS, Danaher RJ. JC virus antibody status underestimates infection rates. Ann Neurol. 2013;74(1):84–90. Major EO, Frohman E, Douek D. JC viremia in natalizumabtreated patients with multiple sclerosis. N Engl J Med. 2013;368(23):2240–1. Laroni A, Giacomazzi CG, Grimaldi L, et al. Urinary JCV-DNA testing during natalizumab treatment may increase accuracy of PML risk stratification. J Neuroimmune Pharmacol. 2012;7(3):665–72. Kean JM, Rao S, Wang M, Garcea RL. Seroepidemiology of human polyomaviruses. PLoS Pathog. 2009;5(3):e1000363. Lin J, Bettin P, Lee JK, Ho JK, Sadiq SA. Cerebrospinal fluid and serum JC virus antibody detection in multiple sclerosis patients treated with natalizumab. J Neuroimmunol. 2013;261(1–2):123–8. Plavina T, Subramanyam M, Bloomgren G, Richman S, Pace A, Lee S, Schlain B, Campagnolo D, Ticho B. Use of JC virus antibody index to stratify risk of progressive multifocal leukoencephalopathy in natalizumab-treated patients with multiple sclerosis [oral communication]. ENS Congress, Barcelona, June 2013.
