Tysabri, the MS Drug Haunted by Deadly Side Effect, Doesn’t Look So Deadly Anymore
Few doctors knew much about a rare brain infection called PML back in 2005, when two patients on a hot new multiple sclerosis drug from Biogen Idec and Elan died from the side effect. The infection, at the time, was generally considered a death sentence. But now with three years of data from more than 60,000 patients worldwide who have taken natalizumab (Tysabri) under strict monitoring by physicians, a new picture is emerging that shows PML is still very much a serious threat, but that it isn’t nearly as deadly as first feared.
While each and every confirmed case of PML, known formally as progressive multifocal leukoencephalopathy, scares investors in Cambridge, MA-based Biogen (NASDAQ: BIIB) and Ireland-based Elan (NYSE: ELN), I sought to assemble a big picture view of exactly how deadly PML really is when I interviewed Al Sandrock last week. He’s the senior vice president of neurology R&D at Biogen, and an assistant clinical professor of neurology at Harvard Medical School.
Before diving too far into the numbers about the risk of Tysabri, a little background is required. This drug, an antibody treatment designed to block certain white blood cells that cause MS when they attack nerves, has a history of also making patients vulnerable to infection. Biogen and Elan yanked it off the market in February 2005 after two cases of the brain disease were confirmed among patients taking the drug; a month later, a third case was confirmed. But legions of patients still demanded the drug, considered to be the most effective medicine on the market at reducing the disabling nerve damage from multiple sclerosis flare-ups. The FDA allowed the drug to return to the market in July 2006 after determining its benefits outweighed the risks, but it also forced doctors into a strict monitoring program to keep an eye out for the early signs of PML.
This matters not just for doctors and patients, but for Biogen’s and Elan’s financial futures. The drug, Biogen’s fastest-growing product, generated $560 million in sales in the first nine months of this year. (The importance of this drug is one reason why investors get so ticked at Biogen when it isn’t exactly forthcoming about every newly diagnosed case, but that’s a bone to pick another day.)
When the drug came back on the market, its FDA-approved prescribing information contained a prominent warning that about 1 out of every 1,000 patients on the drug were likely to get PML. But that was really just a forecast, and the actual risk-benefit balance for this drug is really a moving target that shifts over time when a new case is confirmed. So I sought to build a simple chart when I spoke to Sandrock that provides a snapshot of PML cases in February 2005, when the drug was pulled off the market because of the PML risk, versus those confirmed as of yesterday. Here’s what I gathered:
Number of patients
|Nov. 18, 2009||63,000||27||5|
The February 2005 figures came from clinical trial data and formed the foundation for the FDA-required warning of the 1-in-1,000 chance of getting PML. The more recent figures include all the experience of patients who have gotten the drug since it was returned to the market in July 2006. The thing that jumped out at me was the fact that only five of the 27 confirmed patients with PML have died—meaning that the current survival rate stands at over 80 percent.
That curious fact has been buried under a rash of scary headlines this fall, in which the number of cases of PML appears to have gone through a sharp increase. There were 13 cases confirmed in an FDA notice on September 17, and the number climbed to 23 when European regulators issued an update on October 23. The latest tally now stands at 27, Sandrock said in a Nov. 11 interview (and which was confirmed again yesterday by company spokeswoman Naomi Aoki). Based on this evolving body of evidence, the FDA updated the Tysabri prescribing information earlier this month to say that the risk appears to increase as patients stay on the drug for longer periods of time.
But even as more cases get diagnosed, I wanted to know why there hasn’t been a corresponding increase in the number of deaths. The chart shows more than 80 percent of patients diagnosed with PML since the drug came back on the market in July 2006 are still alive. Lots of doctors also want to know what’s happening to those patients who get PML, but still live. Biogen isn’t releasing a patient-by-patient breakdown of what is happening to them all, but its medical affairs staff is answering those questions from physicians, Aoki says.
Some of the patients are severely disabled, Sandrock says, but a few—he wouldn’t say exactly how many of the 27 cases—have recovered and even feel strong enough to return to work. The New England Journal of Medicine recently published an article about one case of PML in which a patient became critically ill with PML, but recovered.
“Back when I trained as resident, we didn’t know much about it, but it was assumed that PML was almost always fatal,” Sandrock says. “But the survival rates appear to have changed. The outcome from PML is not certain death. Some have done well, and others, not so well.”
Of course, this isn’t the only way to look at the PML risk. The drug’s makers and regulators are looking at databases that factor in different variables, like the number of patients who have been on the drug for more than a year, more than two years, more than three years, and so forth. Those figures are what led regulators to add new language to the label that says the risk increases over time. Sandrock emphasized that even though the risk does increase, it’s still within the 1-in-1,000 rate that patients were warned about back in July 2006, even for patients who have been on the drug since shortly after it was brought back to the market. The degree of statistical confidence in the PML risk figure declines over time, because there are fewer patients who have been on the drug for two and three years, Aoki says.
So while the incidence appears to be rising, what has changed that’s keeping PML patients alive? One of the big reasons is the strict monitoring program insisted upon by regulators, Sandrock says.
Now that every physician who treats MS has heard a lot about PML, they are trained to look for telltale signs like changes in cognition, changes in personality, or seizures that are not typical symptoms of an MS patient, Sandrock says. Doctors watch for those signs every time a patient comes in for their monthly IV infusion of the drug. If the side effect is suspected, the doctor will withhold the drug and run an MRI test to look for signs that the cause of PML, the JC virus, has entered the brain. If the MRI shows the virus is there, the doctor then does a spinal tap that can confirm the presence of the virus.
“We think we’re catching the virus early now,” Sandrock says, noting that some cases have been detected with extremely low concentrates of virus in the spinal fluid that only a specialty lab at the National Institutes of Health can detect.
Once the virus is detected, doctors do what is called a “plasma exchange,” which basically means that whatever amount of drug is still circulating in the bloodstream gets washed out through a filtering procedure. Doctors then reconstitute the patients’ immune system, which “clears the virus out,” Sandrock says.
So while early detection has helped lower the risk of death, I wanted to know what else Biogen and Elan think holds promise to bring the fatality rate down even further. Analysts suggested earlier this month that regulators might impose a drug “holiday” for patients who stay on Tysabri for long-periods of time. Sandrock didn’t like the sound of this idea, saying there isn’t data to support the notion that such a step will help lower the risk, while it essentially guarantees that a patient’s disabling MS symptoms will come back. But he did note that the company is looking to conduct a clinical trial to test the idea.
Biogen says it is more interested in developing a blood test that might offer doctors clues as to which individual patients have an elevated risk of getting PML from the start, and which ones are less likely to become infected. The key to determining this could lie in a relatively common lab technology, known as Elisa, that can detect whether a patient has antibodies in the blood made to fight the JC virus that can awaken from a dormant state to cause PML in Tysabri patients. About half of adults are estimated to have these antibodies, which suggest that JC virus is lurking in the body with the potential to become an opportunistic infection in at least some Tysabri patients. The remaining patients are thought to have no JC virus in their systems, and therefore would be considered less likely to get PML, Sandrock says.
Biogen also has a lot of other strategies for minimizing the risk of PML. It is forming a consortium with the makers of other drugs that have been linked to the side effect, such as rituximab (Rituxan), efalizumab (Raptiva), alemtuzumab (Campath), and mycophenolate (Cellcept), Sandrock says. The idea is that accumulating more data in a global database will help each company deal with the issue, Sandrock says.
There are other ideas in the works as well, such as a vaccine to protect against PML—”we’re very excited about that,” Sandrock says—and a potential anti-PML treatment called mephaquin, which is commonly used to treat malaria. A Biogen collaboration with Cambridge, MA-based Alnylam Pharmaceuticals to develop an RNA interference-based treatment for PML has stopped, Sandrock says.
Pressure to show progress with innovative new treatments, vaccines, or drug holidays will surely intensify if the incidence of PML keeps climbing past that 1-in-1,000 benchmark rate as patients stay on the treatment for years. The pressure to figure out how to best manage PML will come from investors, regulators, and from the patients themselves, who want to make sure that they can still get the drug.
“Not a week goes by around here that we don’t hear a story from a patient who has had their life transformed by this treatment,” Sandrock says. “The benefit has been confirmed, by real life stories. And we’ve made a lot of progress at mitigating the risk.”