Most drugs withdrawn from the market because of serious toxicities never make it back. But Biogen Idec succeeded in getting its multiple sclerosis (MS) drug Tysabri reintroduced in less than 16 months.
On the morning of February 18, 2005, Biogen Idec of Cambridge, Massachusetts, learned that two patients in a clinical trial of its recently approved MS drug Tysabri (natalizumab) were suspected of contracting progressive multifocal leukoencephalopathy (PML), a brain infection caused by the JC virus. One patient had already died; the second was critically ill. (The drug is partnered with Elan, of Dublin, Ireland.)
In the next few hours, Biogen informed the US Food and Drug Administration (FDA) of the problem. After confirming PML in each case, on February 25 the company went public with the findings and then pulled the drug. The reaction was immediate: patients demanded access, industry observers clucked their tongues (Tysabri was approved on the basis of one-year phase 3 trial data) and investors recoiled—Biogen's stock fell 43%. With the FDA still under fire for the messy withdrawal in late 2004 of the painkiller drug Vioxx (rofecoxib), many predicted the world had seen the last of Tysabri.
If Tysabri were to get back on the market, Biogen needed to center attention on the drug's efficacy against a disease in desperate need of effective treatments and balance that against the relatively small risk of fatal brain infection.
A comprehensive safety evaluation of all Tysabri patients was quickly undertaken, uncovering a PML death in a Crohn's trial from 2003 but nothing new in MS. Biogen then put together a package containing final two-year data from a monotherapy trial, which backed up the response rates of the one-year interim data1, and additional information from a trial investigating Tysabri in combination with Avonex (interferon-β-1a)2. And it drew up the TOUCH (Tysabri outreach unified commitment to health) prescribing program, which limited patients, prescribers, pharmacies and infusion sites to those authorized by the program; assigned each patient a case manager; and monitored patients at each monthly infusion and by means of doctor visits every six months. The idea was to catch PML cases early to decrease fatalities.
With all this in hand, at the end of September 2005 Biogen submitted a supplemental biologics license application for Tysabri in MS. The FDA advisory committee meeting—held March 2006—turned out to be quite an event: a standing-room-only crowd spilled into the hallway with a line of patients waiting to relate their positive experiences on the drug. The next day, the panel voted unanimously to allow Tysabri's reintroduction to the market and the FDA gave official clearance in early June 2006.
It has not been all smooth sailing since, and in fact the European Medicines Agency said in October it began a risk/benefit review of the drug in light of the 23 PML cases reported after Tysabri's return. But as of September's end, more than 46,000 patients were taking the drug in trials or commercially, and sales have grown yearly (Table 1).
In the 25 years before Tysabri's reintroduction, only two drugs have raised serious safety issues for the FDA in the post-approval setting and still ultimately found a market. The first of these, New York–based Pfizer's Trovan (trovafloxacin mesylate), was an anti-infection product approved for 14 different types of bacterial infections. Two years after its clearance in 1997, reports of severe adverse liver reactions emerged, prompting the European Medicines Agency to recommend a one-year suspension (which was eventually made permanent) and the FDA to limit the product's use to only serious infections in institutionalized patients. Trovan sales were halved that year, and disappeared thereafter.
Another drug, GlaxoWellcome's Lotronex (alosetron) for irritable bowel syndrome, was recalled in 2000 after it was linked to several cases of ischemic colitis and at least five deaths. In 2002, the FDA allowed a highly restricted patient population access to Lotronex, but sales were disappointing owing to the poor enrollment of physicians in the drug's risk-management program. Ultimately, the drug was sold to San Diego–based specialty pharmaceutical company Prometheus in 2007.
For the majority of products, therefore, there is no way back from a product's market withdrawal. Those few drugs that are relaunched are so hamstrung by regulatory constraints that their business proposition is either severely compromised or no longer tenable.
Tysabri succeeded because its efficacy is head and shoulders above any other therapy available for MS—a factor appreciated by physicians and patients alike. The initial, one-year interim trial data on 2,100 MS patients indicated that the reduction in MS relapse rates was ∼35% higher than the existing standard-of-care, interferon-β. And given the lack of efficacious treatment options, the remaining years of patent market exclusivity and the size of the patient population, the regulatory restrictions placed on Tysabri have not compromised profitability. Indeed, in the second quarter this year, drug sales reached a blockbuster rate: $254 million.
Contrast this with the psoriasis drug Raptiva (efalizumab), which has also been associated with PML infections. Approved in 2003, Raptiva was given to some 46,000 patients worldwide and was associated with three confirmed cases of PML—an incidence rate lower than what is seen with Tysabri in MS. Yet Raptiva's maker, Genentech, of S. San Francisco, California, killed the product in April. Why? Because the psoriasis market is glutted with products of similar efficacy (e.g., Enbrel (etanercept), Remicade (infliximab) and Humira (adalimumab)), Raptiva's patent exclusivity was easily circumnavigated and worldwide sales for the drug were unlikely to grow more from the ∼$125 million in 2008.
The swift and decisive actions of Biogen's board after the emergence of the PML deaths helped ensure that Tysabri made it back to market. But that was possible only because of the drug's effectiveness, the lack of competition from other therapies and the existence of a physician and patient community in MS willing to embrace Tysabri's risk management program.
Polman, C.H. et al. N. Engl. J. Med. 354, 899–910 (2006).
Rudick, R.A. et al. N. Engl. J. Med. 354, 911–923 (2006).
The author is grateful to John Richert, Geoff Meacham, Amy Zeisler and Yaron Werber for insights in drafting this article.
About this article
Delays in the post-marketing withdrawal of drugs to which deaths have been attributed: a systematic investigation and analysis
BMC Medicine (2015)
Nature Biotechnology (2015)
Autoimmune Diseases (2012)
International Journal of Cell Biology (2012)