Thomas Morrow, MD

FDA officials said the agency "remains very concerned" that patients who take natalizumab (Tysabri, Biogen Idec/Elan) may develop a rare, potentially fatal brain infection called progressive multifocal leukoencephalopathy (PML). So starts a report from Reuters published on February 15, 2006.¹ Should managed care also be concerned? If so, how should managed care formulary decision-makers respond? What is their role in managing patients who are clamoring for even more products to treat their diseases in an era in which it seems that miracles are a daily occurrence in the world of biotechnology? What role should managed care play in refereeing potentially devastating side effects for a disease that frightens patients as much as multiple sclerosis (MS)?

Under an accelerated approval process based upon a single year of data, FDA approved natalizumab on November 23, 2004, for the treatment of patients with relapsing forms of MS to reduce the frequency of clinical exacerbations.² Natalizumab is a recombinant alpha₄-integrin antibody derived from a monoclonal antibody (mAb). The agent is the first alpha₄-integrin antagonist in the class of selective adhesion molecule (SAM) inhibitors. Natalizumab binds to alpha₄-integrin on the surface of activated T-cells and other mononuclear leukocytes, which prevent cellular adhesion between T-cells and the endothelial cell. The disruption of cell adhesion molecule interactions results in the prevention of leukocyte migration across the endothelium and into the parenchyma, with a subsequent reduction in proinflammation cytokines.³

On February 28, 2005, just 3 months after FDA approval of natalizumab, Biogen Idec, the manufacturer, and Elan, the distributor, voluntarily suspended marketing of natalizumab due to 2 adverse events in patients taking a combination of natalizumab and interferon beta-1a (Avonex, Biogen Idec). Both adverse events involved PML and resulted in the death of 1 of the patients.⁴

About a month later, another fatal case involving PML was discovered in a patient enrolled in a clinical trial studying the effects of natalizumab in Crohn's disease.⁵ In June 2005, published reports said that Biogen Idec had notified FDA that it had identified a fourth patient taking natalizumab who may have developed a rare neurologic disorder.^6,7 Further information on this reported event does not appear to be available.

On September 26, 2005, Biogen Idec submitted an sBLA requesting priority review of natalizumab for approval to begin marketing the product again. Biogen Idec included a "risk management plan" consisting of a patient registry and patient monitoring program as part of their request.⁸ FDA accepted the application and agreed to the priority review, which meant that FDA would take action within 6 months, rather than the standard 10-month time frame.⁹

In early March 2006, FDA's Peripheral & Central Nervous System Drugs Advisory Committee met publicly to discuss their "concern" about the potential PML risk and ultimately recommended that natalizumab should be allowed to return to the US market for the treatment of patients with MS who have not tried any of the other available first-line therapies.¹⁰ FDA is scheduled to act on the advisory committee recommendations on or before June 28, 2006. Recent articles have attempted to quantify the PML risk, but all of the analysis is hampered by the rather short-term and limited patient exposure.

PROGRESSIVE MULTIFOCAL LEUKOENCEPHALOPATHY

PML is a rare, clinically relevant infection that affects the oligodendrocytes, resulting in demylination and lytic destruction of brain tissue. It is caused by the Jakob-Creutzfeldt (JC) virus, a double-stranded DNA virus held in a latent form in the majority of humans. Most humans demonstrate exposure to this virus, typically during childhood, and appear to harbor it for prolonged periods of time with seemingly no ill effects. The JC virus has been shown to persist in the kidneys and is shed in the urine. More than 80% of the human adult population is seropositive for JC virus-specific antibodies, and there is currently no known way to prevent infection by the JC virus.¹¹ By some unknown mechanism, the use of natalizumab with perhaps a contribution by other immunomodulators appears to activate a destructive infective process.

Christy Krames, MA, CMI Under an accelerated approval process based upon a single year of data, FDA approved natalizumab (Tysabri, Biogen Idec/Elan) on November 23, 2004, for the treatment of patients with relapsing forms of multiple sclerosis to reduce the frequency of clinical exacerbations. Natalizumab is a recombinant alpha4-integrin antibody derived from a monoclonal antibody (mAb). The agent is the first alpha4-integrin antagonist in the class of selective adhesion molecule (SAM) inhibitors.

The original route of entry for the primary infection in humans is not known, but is suspected to be oral or inhaled. It is known to appear in the stromal cells of the tonsils and in the tonsilar B lymphocytes. It is suspected that it then spreads hematogenously to other organs and is kept at bay in virtually all people by the immune system unless the immune system fails.

PML was first identified in 1958. In most individuals, the JC virus remains in a latent state after primary infection but can be reactivated when the immune system is impaired. PML occurs most commonly in immunocompromised patients such as those with acquired immune deficiency syndrome (AIDS), as well as medication-induced immunosuppressed patients such as transplant recipients. It became a much more common disorder after the advent of the AIDS pandemic. Interestingly, PML had not previously been reported in persons with MS.¹²

It is not known if PML is a reactivation of the JC virus dormant within the brain or if it results from the migration of infected cells from the periphery. It is assumed that the B cells deliver the virus to the oligodendrocytes, but active replication of the virus does not appear to occur in the bloodstream.¹³

The hallmark of PML can be summed up as unpredictable and difficult. Diagnosis of PML is difficult because its symptoms are similar to other neur-odegenerative diseases. Although PML typically presents as a multifocal, asymmetric disease, it may present as a solo lesion, making distinction with a new MS lesion problematic. PML has a predilection for the frontal and parietal-occipital regions but may occur in virtually any area. The lesions can occur in both white and gray matter. PML is rapidly progressive and fatal in 80% of patients within 9 months of onset. Symptoms and clinical findings occur in a variety of places in the neurologic history and exam. Cognitive findings include impairment, behavioral changes, confusion, and over time, dementia. Visual symptoms include blurring, diplopia, cortical blindness, and hemianopia. Other symptoms include coordination impairment, focal motor weakness, headache, extrapyramidal symptoms, and seizures.¹¹

Radiologic findings are variable. There is an absence of edema and mass effect around lesions. The lesions demonstrate no enhancement with contrast, but do show an increased T2 signal. The lesions may be singular, but are often confluent and asymmetric, occurring within the white matter in the periventricular and subcortical frontal area and parietal-occipital lobes. PML lesions can appear indistinguishable from MS lesions on magnetic resonance imaging (MRI).¹²

The gold standard for diagnosis of PML is a tissue biopsy with demonstration of the classic histopathologic findings and presence of JC virus DNA. Large, atypical astrocytes are common on biopsy, leading to occasional misdiagnosis of the lesions as malignant. An alternative diagnostic procedure is a spinal tap analyzed for the presence of JC virus DNA when accompanied by a suspicious clinical presentation and imaging findings.^14,15

There is no known effective treatment for the JC virus, though due to the high mortality rate associated with PML, numerous compounds have been tested. Reversing immune system suppression may slow or arrest progression of the disease.¹⁶ Although cytosine arabinoside has demonstrated activity in decreasing JC virus in vitro, it has not been shown to be effective in HIV-positive patients. In AIDS patients, highly active anti-retroviral therapy has increased median survival, but it appears that this may be the result of improved immune function due to inhibition of the HIV, not a primary effect on the JC virus. The use of retroviral therapy in 1 of the MS patients who developed PML resulted in immune-reconstitution inflammatory syndrome, which nearly killed the patient.¹⁷

OVERVIEW OF PML CASES ASSOCIATED WITH NATALIZUMAB

Crohn's patient. A man aged 60 years with Crohn's disease received 300 mg of natalizumab intravenously every 4 weeks as part of a clinical trial initiated in March 2002. He received 3 monthly infusions, followed by treatment with placebo for 9 months. Treatment with natalizumab resumed in February 2003. The patient received 5 doses before being admitted to the hospital with neurologic problems. After discharge from the hospital with temporary improvement, his condition deteriorated rapidly and the patient died in December 2003. Although the patient was not initially diagnosed with PML, the announcement in March 2005 of the withdrawal of natalizumab due to incidences of PML caused the physicians who treated the patient to re-examine that patient's course.

Notably, post-mortem analysis of the patient determined that no JC virus was detected in serum obtained at multiple times from April 1999 through February 2003, as well as intestinal samples obtained during surgery in 2000 while the patient was on several immunosuppressive drugs. "However, JC virus DNA was detected in a serum sample obtained 2 months before admission, while the patient was receiving monthly infusions of 300 mg of natalizumab, and had increased by a factor of 12 by the time the patient was admitted [to the hospital in July 2003]," according to an article in the New England Journal of Medicine (NEJM).¹⁸

MS patient #1. In April 2002, a woman aged 44 years was enrolled in the Biogen Idec clinical trial combining natalizumab therapy with interferon beta-1a for treatment of relapsing-remitting MS (RRMS) (the SENTINEL study). During the study, she received 30 doses of Tysabri (300 mg by intravenous infusion) at 4-week intervals between April 12, 2002, and July 9, 2004. She also was treated with tizanidine, donepezil, and briefly, galantamine. Starting in July 2004, she received an additional 7 doses of natalizumab (300-mg infusion every 4 wk), with the last dose on January 18, 2005. In November, December, and January, the patient reported new neurologic difficulties and was prescribed methylprednisolone. On February 12, 2005, her neurologic status continued to decline, and she was hospitalized. Shortly before the patient died on February 24, 2005, a cerebrospinal fluid sample confirmed the patient had PML.¹⁹

MS patient #2. In October 2002, a man aged 42 years enrolled in the SENTINEL study for treatment with natalizumab and interferon beta-1a. In November and December 2004, the MS patient exhibited neurologic difficulties. In mid-December 2004, the patient received the last of 28 infusions of natalizumab. The patient was not identified as immunocompromised upon presentation, but the JC virus was detected by PCR in the serum, peripheral-blood mononuclear cells, and cerebrospinal fluid. Over 8 weeks, the patient's condition continued to deteriorate, with severe cognitive impairment and a fluctuating level of alertness. The patient was rendered bedridden, mute, and almost completely noncommunicative despite administration of 3 infusions of cidofovir and a 5-day course of intravenous immune globulin. In early April 2005, treatment with cytarabine was initiated. The patient demonstrated some progress after the first course of cytarabine, and a second course was administered. By May 2005, the patient showed continued improvement, starting to walk and to have meaningful conversation. The patient continued to have "disabling ataxia, cognitive impairment, mild neglect, and mild left hemiparesis."¹²

SUMMARY

As these cases demonstrate, PML appears to be a rare but devastating natalizumab-associated adverse drug event. It is possible that natalizumab will be re-released after FDA finalizes its review of Biogen Idec's request.

How should managed care respond? A review of the facts will aid formulary decision-makers in their considerations:

• MS is a lifelong disease requiring decades of therapy with immune-modulating drugs.
• PML is caused by the reactivation, via an unknown mechanism, of the JC virus.
• Childhood exposure to the JC virus is almost ubiquitous.
• The JC virus is likely to be stored in a latent form in the vast majority of the US population.
• PML has been associated with treatment with natalizumab after a relatively short time, compared with the natural history of MS.
• PML is either fatal or will lead to significant disability.
• The exact mechanism of JC virus reactivation upon exposure to natalizumab is unknown.
• There is insufficient knowledge presently available for a physician to predict the chances of onset of PML in a patient receiving natalizumab treatment.
• There is no easy way to diagnose this disease by either serologic or imaging studies.
• The neurologic and imaging findings of a patient with PML cannot easily be distinguished from a routine relapse of MS.
• Even with the best of therapy for MS, most patients will have a relapse every year or so.
• Either a brain biopsy or a spinal tap in conjunction with extensive imaging would be required each time a patient with MS who is receiving natalizumab therapy develops a new neurologic symptom, however minor it may be.
• The therapy for a relapse of MS is to increase the immune system suppression.
• The suggested therapy for a patient with PML is to withdraw immune suppression and perhaps use antiviral drugs.
• Treating a suspected relapse with immunosuppressive therapy such as high-dose steroids may theoretically worsen the activated JC virus if there is a misdiagnosis.