[Update: Zafgen announced on Dec. 2 that a second patient in its clinical trial has died, see more details here.] Whether or not a new drug is approved boils down to one basic question: How does the risk of taking the treatment balance against the benefit it provides? And the weight that regulators give to each of those elements can be different for different diseases—for a rare one with no approved treatments, the bar for a drug’s safety and effectiveness might be lower than for, say, yet another drug that reduces blood pressure or soothes aches and pains.
That calibration is going to play a big part in determining the future of a Cambridge, MA-based biotech called Zafgen (NASDAQ: ZFGN). Zafgen delivered some bad news a month ago when a 23-year-old patient taking its lead drug in a clinical trial died after suffering a pulmonary embolism—a blood clot that breaks loose from a leg or other part of the body and gets lodged in an artery in the lungs.
A death of a young volunteer in a clinical trial of a “me too” drug for a relatively minor condition might be enough to kill the drug and perhaps even sink the company, but Zafgen’s drug, beloranib, isn’t for blood pressure or tennis elbow. It’s for severe, life-threatening forms of obesity—including a rare genetic condition called Prader-Willi Syndrome that causes insatiable hunger that, if not managed properly, can lead to extreme obesity and in some cases, death. The only approved treatment for Prader-Willi is human growth hormone, and while it helps with some of the complex condition’s myriad symptoms, it doesn’t do anything about one of the disease’s core problems, the never-ending drive to eat.
The possibility of addressing that problem and changing the lives of patients who have virtually no other treatment options represents a powerful potential counterweight to the safety questions raised by the trial volunteer’s death—and gives Zafgen a fighting chance to still get its drug to market. Data from the trial, to be revealed early next year, will show how well the drug lives up that promise, and will largely determine whether the scales tip toward risk or benefit for beloranib.
The story of how Zafgen and its drug got to this point and what might be in store next is a fascinating example of the complicated interplay between drug companies, researchers, patients, and regulators, and of the lengths patients and their families are willing to go in pursuit of new treatments for rare conditions. And it illustrates one of the reasons—alongside premium pricing power and market exclusivity—why biotech companies often target rare diseases before tackling common ones.
Zafgen’s Prader-Willi trial was nearly complete when the company found out about the patient’s death—sometime in early October, according to CEO Tom Hughes—and based on subsequent discussions with the FDA, it expects the agency to still view the study as pivotal (ie, substantial enough to base a drug-approval decision on), the company has said in a statement.
Still, the whole episode has been a mess for Zafgen. Reports on October 12 from Bloomberg and TheStreet.com that it had cancelled a few investor meetings led to speculation that bad news about beloranib was coming, and Zafgen shares went into freefall as the company stayed radio silent. Zafgen finally broke that silence on Oct. 14, revealing that someone in the study had died, but not what the cause of death was or whether the patient had been taking beloranib or placebo. Then, two days later, the company announced that patient had indeed been taking beloranib, and that the FDA had stepped in with a so-called partial clinical hold. Under its terms, Zafgen could continue the trial, but it must screen each patient for possible blood clots. Six days after that, on Oct. 22, Zafgen revealed that the patient’s cause of death was respiratory failure from a pulmonary embolism.
Investors, meanwhile, were clearly livid. Zafgen has already been named in several class action suits that allege it withheld some old data potentially tying beloranib to blood clots, and that the company revealed the information only after its share price was decimated and the FDA hold began.
And questions about beloranib’s safety don’t just jeopardize the drug’s future as a treatment for Prader-Willi. Zafgen hopes to eventually get it approved to treat a more common type of severe obesity as well. When a drug is potentially meant to be taken over the long term by a large group of patients, safety concerns become particularly critical. “Overall, I think you have to lower the probability of [the drug’s] success and the market penetration it might have,” says RBC Capital Markets analyst Simos Simeonidis.
But before Zafgen can reach even the first small market segment that it’s targeting it will need, among other things, to address the question of whether beloranib causes clots and embolisms of the sort that killed the patient in the Prader-Willi trial. There’s not a lot of data to go on: So far, seven out of about 400 patients dosed with the drug have had serious blood clots (including the patient who died), compared to none out of 150 on placebo, according to Zafgen. Simeonidis says he thinks there is a correlation between the drug and the clots, but “you can’t really say that with certainty” given the data so far.
Zafgen’s Hughes, on the other hand, says that patients with Prader-Willi may be predisposed to clots and embolisms, and so the fact that a patient taking beloranib suffered one is not necessarily an indictment of the drug’s safety.
“It’s a very sad thing, and we’re very touched by it, and we feel horrible for the family who had to experience this as well as for the investigator involved, but this is not an uncommon form of death or cause of death in this population,” Hughes says. “This is by no means an unusual situation, nor is it one that would keep us from having a commercially viable product. It’s just something that requires some discussion.”
Involved in that discussion are the FDA and the company, of course, but also patient groups, researchers, and the families of Prader-Willi patients. Jim Loker, a pediatric cardiologist at Bronson Hospital in Kalamazoo, MI, belongs to all of the latter three groups. The father of a daughter with Prader-Willi, Loker has been analyzing the causes of death for patients with Prader-Willi using data collected by the Prader-Willi Syndrome Association since 1973. The study is meant to help better understand Prader-Willi, the types of problems it causes for patients, and how to prevent the sudden deaths that are sometimes associated with the disease. In a study presented two weeks ago at the annual meeting of the PWSA in Florida, Loker found that of the 310 patients in the PWSA’s database whose cause of death was known, 19 had been killed by a pulmonary embolism. An additional four died of something else but had a pulmonary embolism as well.
Interpreting this number without knowing, for example, how it compares to pulmonary embolism deaths among obese patients without Prader-Willi, is tricky. Still, Zafgen views it as a piece of the beloranib puzzle. The company included Loker’s findings as a slide in its latest earnings call with analysts last week, with Hughes stating that “it’s against this backdrop of severe morbidity and increased mortality that our drug should be evaluated.” Via e-mail, Simeonidis said the findings are “helpful and relevant…definitely helps their case.” No surprise, then, that Zafgen has asked Loker to talk to the FDA to help make its case about beloranib.
Loker had only sporadically heard of beloranib until recently—his 21-year-old daughter doesn’t need it, he says. But he and his wife have been active members of the PWSA for years, and much of that time he’s been on the nonprofit’s “mortality review committee.” After a Prader-Willi patient dies, the agency talks to the family, and collects information about the patient and his or her cause of death.
“The best way to prevent future deaths is to know what’s been causing the deaths that we have,” Loker says.
Indeed, there’s much that’s still not known about Prader-Willi. The disease is rare, currently affecting one in every 12,000 to 15,000 babies born, according to the PWSA. Much of the underlying biology remains a mystery, but two things are certain. First, it’s genetic, caused when certain genes on chromosome 15 don’t work properly. Second, patients with Prader-Willi are essentially born with a malfunctioning hypothalamus, the part of the brain that controls the production of hormones.
As a result, people with Prader-Willi have very low muscle tone, a super slow metabolism, and an insatiable hunger. If left alone, they might raid the cabinets, fridge, or trash to steal food, and it isn’t unheard of for them to eat so fast they choke to death, or rupture their esophagus or stomach. Loker says choking and gastrointestinal perforations were responsible for 5 percent and 6 percent, respectively, of the Prader-Willi deaths he’s tracked.
“It’s a wicked combination of everything that could go wrong has gone wrong to create obesity,” says Janalee Heinemann, the research director of the PWSA, whose 42-year-old son Matt (pictured above, left, with Heinemann and her husband) has Prader-Willi, and is in the clinical trial for beloranib.
Appetite suppressants don’t work for people with Prader-Willi, and gastric bypass and other types of surgery are dangerous—they only make the stomach smaller and more likely to be overwhelmed by a food binge. Loker and his wife instituted a strict food regimen to help their daughter, Anna, who was diagnosed with Prader-Willi before the age of two.
“She’d always had the same schedule—we never varied. Always the same thing for breakfast, morning snack, lunch, afternoon snack, and we just fed her what we were eating but a smaller portion at dinner,” he says. “That’s what she got used to, so if she ever said she was getting hungry, we could say, ‘Ok look at the time, this is when you eat.’ And that was enough for her.”
This is an example of the sort of thing that most families and caregivers of Prader-Willi patients have to do to protect their loved ones. It could mean bolting the refrigerator, monitoring exactly what a patient eats and when, and having someone around at all times to make sure they don’t ultimately harm themselves. Most, though not all patients can’t live independently; Loker’s daughter, for instance, is in a group home.
What would make beloranib potentially a game changer here is that, in a 17-patient, four week (the small scale and brevity should be noted), mid-stage trial, it helped patients not only to lose weight, but also to feel less hungry. This was measured via a composite score based on patients’ behaviors (logged by caregivers) such as stealing food, or becoming “unmanageable” because of a preoccupation with food, according to Hughes. The official number was a 52.4 percent reduction in these types of behaviors at the highest dose of beloranib (1.8 mg) over the course of the study.
“To be quite frank, this has been the most exciting potential drug [for Prader-Willi] to come along,” Heinemann says.
Beloranib is an injectable small molecule that inhibits the activity of an enzyme called methionine aminopeptidase 2 which, in turn, is supposed to cause the body to burn more fat. Zafgen’s strategy has been to test beloranib against very severe obesity-related conditions, including Prader-Willi and a type of obesity caused by injury to the hypothalamus. It’s also testing the drug in a more general population of severely obese patients with type 2 diabetes—a larger market opportunity—but moved Prader-Willi to the forefront so it could run smaller, quicker trials and potentially get to market faster. [An earlier version of this story incorrectly referred to beloranib as an antibody.]
In October of last year, the company began enrolling patients in its key test, a 108-patient, placebo-controlled, six-month study of beloranib in Prader-Willi patients. Given the type of rapid and substantial weight loss Zafgen has shown in earlier studies and the apparent lack of severe safety problems—the most common side effects reported were nausea, vomiting, diarrhea, and sleep trouble—expectations have been high. Zafgen priced an initial public offering in June 2014 at $16 per share, the top of its projected range, and shares climbed following the IPO. They were worth as much as $50 apiece in March 2015 and $46 apiece in mid-September, in the heights of the biotech boom. Zafgen raised money through secondary offerings during the good times, and had $204 million in the bank at the end of September.
“We were as surprised as everybody else that our stock rose to the level that it did on the absence of any news,” Hughes says.
Those numbers have only made Zafgen’s subsequent fall that much more dramatic. Hughes won’t say the exact date, but towards the beginning of this October the Prader-Willi study was “essentially more than two thirds done,” he says, when Zafgen learned that a participant had died. FDA protocol mandates that this information has to be submitted to the agency within seven days, and Zafgen did so, notified the trial’s safety oversight committee, and began talks with the FDA.
Hughes says when Zafgen got the news, it didn’t have a death certificate or the autopsy results. A scramble ensued over the Columbus Day weekend to get more information so Zafgen could begin to figure out what happened. “As the sponsor of a study, we don’t know who these patients are, we don’t know where they live, they’re blinded to us for patient confidentiality reasons, and so the study center itself is what was really charged with working with the family to get this information, to get consent, and get the documents transferred,” Hughes says. “The family is grieving, there’s a holiday weekend involved, and it was just very, very hard for us to get information that would be very useful.”
From the outside, no one had a clue about any of this until reports surfaced on Columbus Day Monday, October 12 that Zafgen cancelled investor meetings in New York slated for the 13th and 14th. Word spread quickly through social media, and speculation ran rampant that something was up. Zafgen stonewalled requests for comment from reporters (Xconomy included), analysts, and shareholders, and its stock began dropping faster and faster. In two days, without any news, shares lost more than half their value, from $34.76 at the open on the 12th to $15.75 at the close on the 13th.
Zafgen finally made its initial disclosure about the patient’s death on Oct. 14, and its shares temporarily rebounded. Hughes says Zafgen worked with the FDA to get all the information, and unblinded the trial and agreed on the clinical hold the following day. Then, on the morning of Oct. 16, it released another announcement and held a conference call, during which it gave details on both the patient who died and the six other instances of blood clots in beloranib trial participants. Zafgen’s shares plummeted more than 50 percent that day and closed at $10.36 apiece. Lawsuits soon started popping up, with accusations made that Zafgen hadn’t previously disclosed some of the prior instances of blood clots in its trials. Given all that, does Hughes have any regrets? “There’s nothing we would’ve done differently,” he says.
Winning back the trust of investors rattled by the trial participant’s death and Zafgen’s handling of it is one of many challenges that company now faces. But, interestingly, members of the Prader-Willi community itself have reacted with relative calm to the incident. The PWSA’s Heinemann says that there’s been “surprisingly little concern” among PWSA members about the death—a fact that illustrates just how heavily the potential benefit of a new drug can weigh against its potential risks for families coping with serious diseases and few treatment options. “Obviously everybody is sad about a death, everybody gets disturbed about a death,” Heinemann says matter-of-factly. “But this is not the only death they’ve been disturbed about.”
Heinemann plans to have her son, who she says has lost weight and become less obsessed with food on beloranib, participate in a so-called open-label (ie, unblinded) extension of the trial. Zafgen is conducting the study to accumulate more safety data and give patients—at least those who pass the blood-clot screening—continued access to the drug.
“The drug has a lot of potential, and a lot of potential to take away some of the other life threatening issues from Prader-Willi and make his life easier. That has to be taken into consideration,” Heinemann says. “The risk factor needs to be dealt with, but he’s getting the blood tests and the ultrasound and if everything’s ok, I have no problem putting him back on the drug.”
For his part, Hughes took me through each instance of pulmonary embolism or deep vein thrombosis in Zafgen’s trials, and made a case as to why each could’ve simply been part of the patient’s underlying condition. One patient in a Phase 2 study had gout, her leg became inflamed and had to be elevated and immobilized for days, and she developed a blood clot that led to a pulmonary embolism. Another patient had a “major inflammatory dysfunction” in the lungs.
“This is one of the underlying problems of studying an ill population that is at elevated risk of these types of events in the first place,” Hughes says, saying that, in “the absence of any specific knowledge,” all Zafgen can conclude is that beloranib might exacerbate problems these patients already have. If so, as is the case with certain drugs for other diseases, patients taking beloranib might just have to be monitored for particular risks, he says. “In diseases like multiple sclerosis, there’s a screening that has to be done to make sure that the drug doesn’t cause a lethal brain virus to re-emerge,” he says.
In a similar vein, Heinemann points to a past “crisis” with growth hormone, which was eventually linked to potentially severe respiratory problems and even deaths in Prader-Willi patients, and now requires certain safeguards. While people died, the Prader-Willi community learned more about the disease and preventative care from it, she says, and now reaps the benefit of an important drug. “I think that’s what we have to do here,” she says.
And Loker is hoping that the measures Zafgen is now taking will help everyone get a better read on how big a problem blood clots are for Prader-Willi patients, and who is at risk for them. Maybe a trend will be found, and someday as standard procedure, a child with certain characteristics will get an anticoagulant drug that saves his life. That’s beneficial regardless of what happens with beloranib, or Zafgen’s share price.
“Whether it’s this clinical trial or others, we need to know the answer of, is this a risk factor?” Heinemann says. “Because doing nothing and just not doing clinical trials isn’t going to save our kids’ lives.”