Tysabri safety and pml risk stratification april 2013

Natalizumab Safety Update &

PML Risk Stratification

April 2013

Gavin Giovannoni

Barts and The London

Benefit / Risk

Benefit

Risk

Natalizumab

81% reduction in

relapse rate1

64% reduction in

disability

progression1

>1 in 3 patients free

of disease activity2

PML risk ≈

2.97 in 1,000 3

Other Adverse

Events Per

Labelling

1. Hutchinson M, et al. J Neurol. 2009;256:405-415.

2. Havrdova E, et al. Lancet Neurol. 2009;8(3):254-60.

3. Biogen Idec, data on file.

VP1=viviparous 1; RR=regulatory region; CNS=central nervous system.

1. Presence of

asymptomatic JCV

2. Viral factors:

VP1 mutations,

RR permutations

3. Host factors:

peripheral immune function,

genetics

4. Drug effects that reduce

CNS immune surveillance

45 nm

JC virion

What causes PML?

• PML is uncommon and likely caused by interplay between multiple factors

Key Learnings: PML Risk

• Duration of natalizumab dosing prior to PML diagnosis ranged from 8 to 76

doses1

– Mean duration of natalizumab dosing at time of PML diagnosis was approximately

39.0 months1

• Overall incidence: 2.97 per 1000 patients (95% CI: 2.66 to 3.29 per 1000

patients)1

– Currently the average post-marketing natalizumab exposure worldwide is

approximately 2 or more years of natalizumab exposure

• Factors that increase the risk of PML have been identified2 – JCV exposure indicated by anti-JCV antibody positive status

– Receiving an immunosuppressant prior to receiving Natalizumab

– Natalizumab treatment duration, especially >2 years


2. Natalizumab Summary of Product Characteristics

Natalizumab PML Incidence Estimates by

Treatment Duration

Biogen Idec, data on file.

Calculations based on exposure through 31 March 2013 and 347 confirmed cases as of 2 April 2013

Inc

ide

nc

e p

er

10

00

pa

tie

nts

3.29

4.56

5.18

5.73 5.49

5.31

4.99

4.20

3.89

3.51

2.66

3.68

4.16

4.57

4.25

3.98

3.58

2.77

2.35

1.79

2.97

4.10

4.65

5.13

4.84 4.61

4.25

3.43

3.05

2.54

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Natalizumab PML Incidence Estimates by

Treatment Epoch In

cid

en

ce p

er

10

00

pa

tie

nts

Calculations based on exposure through 31 March 2013 and 347 confirmed cases as of 2 April 2013


3.29

0.11

0.84

2.23

2.83 2.88 2.93

2.66

0.02

0.48

1.52

1.86 1.69

1.33

2.97

0.05

0.64

1.85

2.31 2.23 2.02

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Use of Natalizumab in the Post-Marketing Setting*

Patients


112,200

79,100

67,100

56,100

46,700

38,700

31,200

24,300

OverallExposure

≥12 Months

≥18 Months

≥24 Months

≥30 Months

≥36 Months

≥42 Months

≥48 Months

Patients

Worldwide post-marketing data from 23 Nov 2004 to 31 December 2012

261,990 Patient-Years

of natalizumab exposure

*Post-marketing data includes patients exposed since 23 November 2004. This excludes approximately 5,100 patients exposed

in clinical trials; 2,200 exposed for ≥12 months; 1,900 exposed for ≥18 months; 1,700 exposed for ≥24 months; 1,300 were

exposed ≥30 months; 1,000 were exposed ≥36 months; 700 were exposed ≥42 months; and 700 were

exposed for ≥48 months. Exposures are estimates and may not fully reflect treatment interruptions that

are used in certain patients.

• The 2-step anti-JCV antibody assay has been developed to help identify

patients who have been exposed to the JC virus. It is designed to detect the

presence of antibodies to JCV in the serum or plasma.

• Anti-JCV antibody testing may be considered for: – All MS patients with disease activity who are contemplating a start/change in MS

therapy where antibody status could be a factor in assessing therapeutic choice;

– Patients treated with natalizumab who have not had their serostatus assessed.

Anti-JCV Antibody Testing

• Current data on the assay as a risk stratification tool from STRATIFY-1 (n=1096) – 46% anti-JCV antibody negative and 54% anti-JCV antibody positive at baseline1

– False negative rate: 2.2% at baseline2 and 2.4% over 18 months3

– Over 18 months, with testing every 6 months:1

• 38% of subjects remained consistently anti-JCV antibody negative

• 52% remained consistently anti-JCV antibody positive

• 10% changed serostatus

– In patients who tested anti-JCV antibody negative at baseline:1

• 84% (274/328) remained negative at 18 months

• 16% (54/328) changed serostatus

– In the subjects who changed serostatus:1

• 31% (17/54) had intermittent positive results

• 69% (37/54) changed to anti-JCV antibody positive and remained positive for the duration of the study

• The probability of consistently testing anti-JCV antibody negative over time

increases with the number of sequential anti-JCV antibody negative test results. – In subjects who tested anti-JCV antibody negative at baseline (n=328), the probability of remaining

anti-JCV antibody negative after 3 tests performed at 6-monthly intervals was 93.5% over 18 months1

– In subjects who tested anti-JCV antibody negative at baseline in AFFIRM (n=241), the probability of

remaining anti-JCV antibody negative after 3 tests performed at 6-monthly intervals was 95%3

1. Plavina et al. Presented at AAN: March 18-23, 2013, San Diego, CA. S30.001

2. Lee et al. J Clin Virol. 2013 [Epub ahead of print]




• As of 2nd April 2013,139 natalizumab-treated MS PML patients with known pre-PML anti-JCV

antibody status who had samples tested for anti-JCV antibodies, all of which were collected

at least 6 months prior to PML diagnosis (range 6-187 months). Of these 139 patients:

– 137 (99%) patients tested anti-JCV antibody positive at all time points where samples were available.

– 1 patient (<1%) tested anti-JCV antibody negative 9 months prior to PML diagnosis and anti-JCV

antibody positive 6.5 months prior to PML diagnosis. The patient had >5 years of natalizumab

therapy + prior IS use.

– 1 patient (<1%) tested anti-JCV antibody negative 9 months prior to PML diagnosis; no additional

pre-PML samples were available. The patient had received ~ 3-4 years of natalizumab therapy + no

prior IS use.

• A sample obtained from one CD clinical trial patient prior to PML diagnosis tested positive.

• In addition, one MS patient tested anti-JCV antibody negative 15 months prior to PML

diagnosis and tested positive two months before PML diagnosis. The patient had received

>3 years of natalizumab + no prior IS use. At the time of testing positive, the patient had

detectable IgM and IgG antibodies. The patient changed antibody status at some point, but

the time of serochange is unknown. The patient’s anti-JCV antibody status 6 months prior to

PML diagnosis is unknown.





• Re-testing of anti-JCV antibody negative patients every 6 months is

recommended.1

• The anti-JCV antibody assay should not be used to diagnose PML.1

• Data from a Biogen Idec study of plasma exchange (PLEX) in natalizumab-

treated MS patients demonstrated that anti-JCV antibody levels are

decreased by 2-5 fold after PLEX and thus may lead to an anti-JCV antibody

negative result in some patients with a relatively low titer before PLEX. Anti-

JCV antibody testing should not be performed during or for at least two weeks

following plasma exchange due to the removal of antibodies from the serum.1

• One sample, collected from a patient at the time of PML diagnosis following a

cycle of PLEX tested negative for anti-JCV antibodies. Because this sample

was collected immediately following PLEX, and PLEX removes antibodies

from the circulation, the information obtained from this sample is unreliable.2

Geographic Distribution of PML Cases

• Of the 347 cases reported through 2nd April 2013:

– 123 are from the United States

– 203 are from the European Economic Area

– 21 are from the rest of the world


Status of PML Cases

• As of 2nd April 2013:

– 79 patients have died (23%)

– 268 patients are alive (77%)

• It is too early to draw conclusions about the outcomes of patients who develop PML while on natalizumab treatment

• PML may be fatal or result in severe disability1

The median time to death was 2.2 months (range, 0.1 to 15.2 months) for 44 deaths as of 29th February, 2012.



• As of 4 March 2011, 39 of 93 patients (42%) with PML and known prior IS

status had been treated with IS therapy before initiating natalizumab. Of the

total 102 confirmed PML patients as of 4 March 2011, prior IS status was

unknown for 9 patients and they were excluded from the analysis.

• As of 23 November 2010, the proportion of all patients treated with

natalizumab in the TYGRIS Observational Study* who had been treated with

an IS prior to receiving natalizumab was 20.3% (13.9% in US and 23.6% in

EU/ROW).

• Compared to patients who have never been treated with a prior IS therapy,

patients with prior IS use had a ~3-4-fold greater risk of PML.

• In patients with PML, there was no specific pattern in: – type of prior IS therapy

– duration of prior IS therapy

– time from last dose of IS to initiation of natalizumab therapy

Estimated PML Risk Associated with Prior IS Use

*http://clinicaltrials.gov/ct2/show/NCT00477113 and http://clinicaltrials.gov/ct2/show/NCT00483847


No Specific Pattern in Type of Prior IS Use

Identified in Patients with PML

• Type of prior IS use varied

• Some patients had received more than one type of IS therapy

• Types of prior IS use included:

– Mitoxantrone (n=38)

– Azathioprine (n=11)

– Methotrexate (n=9)

– Cyclophosphamide (n=14)

– Mycophenolate (n=6)

– Other (n=8)

Data based on prior IS agents reported in 68 out of 197 patients with PML as of 29th February 2012

(Prior IS status was unknown for 15 patients and they were excluded from the analysis).


No Specific Pattern in Duration of Prior IS Use or

Time from Last Dose of IS in Patients with PML

• Duration of prior IS use varied:

– Mean 19.9 months, median 12.5 months (minimum 0.03 month,

maximum 204 months)

• Time from last dose of IS until start of natalizumab varied:

– Mean 25.8 months, median 17.2 months (minimum 0.5 months and

maximum 95.4 months)

Data based on prior IS agents reported in 68 out of 197 patients with PML as of 29th February 2012

(Prior IS status was unknown for 15 patients and they were excluded from the analysis).


Risk Stratification Tool: The Presence of Anti-JCV Antibodies,

Prior Immunosuppressant Use, Treatment Duration*

0.07/1000 95% CI 0-0.38

No Yes

Anti-JCV

Antibody Status

Negative Positive

Prior IS Use


Natalizumab

Exposure No Prior IS Use Prior IS Use

1–24 months 0.6/1000 95% CI 0.4-0.9

1.8/1000 95% CI 1.1-2.8

25–48 months 5.2/1000 95% CI 4.3-6.2

10.6/1000 95% CI 8.1-13.8

Data beyond 4 years of treatment are limited. *Based on natalizumab exposure and 285 confirmed PML cases as of September 5, 2012. Prior IS data in overall natalizumab-treated

patients based on proportion of patients with IS use prior to natalizumab therapy in TYGRIS as of May 2011; and prior IS data in PML

patients as of September 5, 2012. The analysis assumes that 55% of natalizumab-treated MS patients were anti-JCV antibody positive

and that all PML patients test positive for anti-JCV antibodies prior to the onset and diagnosis of PML. The estimate of PML incidence in

anti-JCV antibody negative patients is based on the assumption that all patients received at least 1 dose of

natalizumab . Assuming that all patients received at least 18 doses of natalizumab , the estimate of PML

incidence in anti-JCV antibody negative patients was generally consistent (0.1/1000; 95% CI 0.00–0.62).

Risk Stratification Tool: The Presence of Anti-JCV Antibodies,

Prior Immunosuppressant Use, Treatment Duration*

Data beyond 4 years of treatment are limited. *Based on natalizumab exposure and 285 confirmed PML cases as of September 5, 2012. Prior IS data in overall natalizumab-treated

patients based on proportion of patients with IS use prior to natalizumab therapy in TYGRIS as of May 2011; and prior IS data in PML

patients as of September 5, 2012. The analysis assumes that 55% of natalizumab-treated MS patients were anti-JCV antibody positive

and that all PML patients test positive for anti-JCV antibodies prior to the onset and diagnosis of PML. The estimate of PML incidence in

anti-JCV antibody negative patients is based on the assumption that all patients received at least 1 dose of

natalizumab . Assuming that all patients received at least 18 doses of natalizumab , the estimate of PML

incidence in anti-JCV antibody negative patients was generally consistent (0.1/1000; 95% CI 0.00–0.62).

0.07/1000 95% CI 0-0.38

No Yes

Anti-JCV

Antibody Status

Negative Positive

Prior IS Use


Natalizumab

Exposure No Prior IS Use Prior IS Use

1–24 months 0.6/1000 95% CI 0.4-0.9

1.8/1000 95% CI 1.1-2.8

25–48 months 5.2/1000 95% CI 4.3-6.2

10.6/1000 95% CI 8.1-13.8

1 in 14,286

1 in 192

1 in 1,667

1 in 94

1 in 556

1. Clifford DB, et al. Lancet Neurol. 2010;9:438–446


• Heightened clinical vigilance led to prompt natalizumab discontinuation

upon first signs or symptoms suggestive of PML – Median duration from symptom onset to PML diagnosis is approximately 1

month 1

• The majority of patients who developed PML in the post-marketing

setting received plasma exchange (PLEX) and/or immunoadsorption

(IA) to accelerate removal of natalizumab

• In the majority of natalizumab-treated patients with PML, Immune

Reconstitution Inflammatory Syndrome (IRIS) has occurred after

discontinuation or removal (by PLEX) of natalizumab

• In patients who have undergone PLEX, IRIS has occurred within days

to several weeks2

Key Learnings: PML Management

At this time, there are insufficient data to predict survival outcomes in patients treated with natalizumab who develop PML.

Longer-term data are required in order to more accurately predict such outcomes.

Factors that may affect survival in patients with PML

Vermersch P, et al. Neurology. 2011;76:1697-1704.


• Gender

• Prior immunosuppressant

therapy

• MS duration

• Natalizumab exposure at

PML diagnosis

• JCV DNA load in CSF at PML

diagnosis

• Gd enhancement on MRI at

diagnosis

Factors that appear to be

associated with decreased

survival Factors that do not appear to affect

survival

Factors that appear to be

associated with improved

survival

• Younger age at PML

diagnosis

• Lower pre-PML EDSS

• Shorter time from first

symptoms of PML to

diagnosis

• Localized PML extension

on MRI at diagnosis

PML Presenting Symptoms

PML symptom % PML cases with symptom

Cognitive/behavioral 49%

Motor (eg: hemiparesis) 37%

Speech (eg: dysarthria, aphasia) 31%

Visual (eg: hemianopsia) 26%

Cerebellar (eg: ataxia) 17%

Seizure (eg: focal motor, generalized) 17%

Sensory (eg: paresthesia) 3%

Neurologic deficits evolved over several weeks

Individual PML cases frequently presented with symptoms in multiple categories

Based on the first 35 PML cases.


IRIS Presents as Clinical Decline

IRIS symptom % PML cases with symptom

Motor (eg; hemiparesis) 66%

Speech (eg; dysarthria, aphasia) 38%

Cognitive/behavioral 34%

Seizure 19%

Visual (eg; hemianopsia) 13%

Cerebellar (eg; ataxia) 13%

Fever 6%

May be rapid, can lead to serious neurological complications or death

Individual cases frequently presented with IRIS symptoms in multiple categories.

Based on the first 35 PML cases. At the time of the analysis, 32 of 35 PML cases reported the occurrence of IRIS.


• Monitoring for development of IRIS and appropriate treatment should be undertaken

• Treatment of IRIS with IV corticosteroids:

– Experts uniformly recommend corticosteroids at onset after PLEX1

– Majority of patients have been treated with IV corticosteroids for IRIS, typically 1 gram IV methylprednisolone daily for 5 days, followed by tapered dose of oral steroids1

– Duration and timing of corticosteroid use remains to be refined. In many cases, repeated courses of IV corticosteroids were given1

– Does not appear to be associated with increased mortality

• Prophylaxis of IRIS with corticosteroids has not been systematically evaluated

– Given the severe nature of IRIS and its consistent presentation in most patients, pre-emptive treatment starting immediately after PLEX might be justified. A randomized comparison of these alternatives would be helpful to refine IRIS management1

1. Clifford et al. Lancet Neurol. 2010;9:438–46.

Key Learnings: Treatment of IRIS

Clinical Status of PML Cases: Karnofsky scores

Karnofsky Performance Scale (KPS) scores for 80 PML survivors for whom data were available. Each point represents

the score of an individual patient at the indicated interval relative to PML diagnosis; only those patients for whom KPS

scores were available for a given interval are shown.

pre-PML PML diagnosis 6-9 months 10-13 months ≥14 months

Mean 81.1 49.4 53.1 49.6 52.6

Median 80 50 50 50 50

n 33 32 45 27 25

Dong-Si et al. ECTRIMS 2012; P1098.

0

10

20

30

40

50

60

70

80

90

100

Ka

rno

fsk

y S

co

res pre-PML

PML diagnosis

6-9 Months

10-13 Months

14+ Months

Mean

Median

Pre PML PML diagnosis 6-9 months 10-13 months ≥14 months

On average, Karnofsky scores decrease at PML diagnosis but remain stable through ≥14

months of follow-up

Clinical Status of PML Cases: EDSS scores

EDSS scores for 118 PML survivors for whom data were available. Each point represents the score of an individual patient

at the indicated interval relative to PML diagnosis; only those patients for whom EDSS scores were available for a given

interval are shown.

Dong-Si et al. ECTRIMS 2012; P1098.

On average, EDSS scores increase at PML diagnosis but stabilize at 6-9 months and

remain stable through ≥14 months of follow-up

0

1

2

3

4

5

6

7

8

9

10

ED

SS

Sc

ore

s

pre-PML

PML diagnosis

6-9 Months

10-13 Months

14+ Months

Mean

Median

pre-PML PML diagnosis 6-9 months 10-13 months ≥14 months

Mean 3.7 5.2 6.3 6.4 6.6

Median 3.5 5.5 6.4 6.8 7

n 90 75 28 16 21

Pre PML PML diagnosis 6-9 months 10-13 months ≥14 months


Summary

• As of 31st December 2012, approximately 112,200 patients received

natalizumab in post-marketing setting worldwide

• Factors that increase the risk of PML have been identified1 – JCV exposure indicated by anti-JCV antibody positive status

– Receiving an immunosuppressant prior to receiving Natalizumab

– Natalizumab treatment duration, especially >2 years

• Clinical vigilance remains key in the early diagnosis of PML – Early diagnosis and aggressive clinical management appears to be associated

with improved survival rates observed in post-marketing cases

– PML may be fatal or result in severe disability1

• The following factors appear to be associated with improved survival after

PML: – Shorter duration between symptom onset and PML diagnosis

– Localized PML on MRI at diagnosis

– Younger age

– Lower pre-PML EDSS

Tysabri safety and pml risk stratification april 2013

Health & Medicine

antijcv antibody testing

antijcv antibodies

step antijcv antibody

identified2 jcv exposure

risk of pml

ms patients

certain patients

current data