The most expensive drug in history is a money loser that’s not reaching patients. In fact, it’s only been paid for and used commercially once since being approved in 2012.

The medication in question is alipogene tiparvovec, better known as Glybera, a medicine widely heralded as the “first gene therapy” in the Western world and whose approval helped ignite an explosion of investment and excitement around treatments that correct DNA. 

But when the Berlin physician Elisabeth Steinhagen-Thiessen wanted to give a patient Glybera last fall, it wasn't so easy. She says she had to prepare a submission as thick as “a thesis” for German regulators and then personally call the CEO of DAK, one of Germany’s large sickness funds, or insurers, to ask him to pay the $1 million price tag.

Last September, she gave 40 injections to the muscles of a 43-year-old woman with an ultra-rare disease called lipoprotein lipase deficiency. Such patients don’t process fat correctly. “You draw blood and you are astonished, there is no red blood, it's cream,” Steinhagen-Thiessen says. One symptom is debilitating abdominal pain. Her patient had been hospitalized more than 40 times.

A dose of Glybera contains trillions of viruses harboring correct copies of the lipoprotein lipase gene. And Steinhagen-Thiessen says the treatment, at Charite Hospital in Berlin, was a success. The woman hasn’t been back to the emergency room since the treatment and is now “living like you and me.” 

But this single use of the drug just proves that Glybera is a flop. The problem is its staggering million-dollar price tag, too few patients, and questions about how effective it is.

The company that developed Glybera, UniQure, based in Amsterdam and Lexington, Mass., last fall dropped plans to get it approved in the U.S. and has turned over European sales to the Italian drug maker Chiesi Farmaceutici, which calls selling the drug "challenging."

“I think we learned a tremendous amount about what to do and what not to do, but commercially it has not been a success. It still drains a lot from the company,” says Dan Soland, UniQure's CEO. UniQure is now focused on developing other gene therapies, including one to treat hemophilia.

How Glybera turned into a money loser is a cautionary tale for gene therapy, a resurgent technology that has been drawing investor interest because of its promise to cure rare, inherited diseases with one-time repairs to a person’s DNA. A single dose of gene therapy can change the genetic instructions inside a person's cells in ways that last many years, or even a lifetime.

In addition to Glybera, there is at least one form of gene therapy approved in China to treat cancer by adding a gene to tumors, and late last year Amgen won U.S. approval for Imlygic, which uses the herpes virus to shrink skin cancers.

But reversing inherited genetic disease remains gene therapy's great promise. And more treatments will reach the market soon. In April, European authorities gave a preliminary green light for a gene therapy for severe combined immune deficiency, to be sold by GlaxoSmithKline. And by 2017, handicappers expect, a Philadelphia company called Spark Therapeutics could win approval in the U.S. for a gene fix that partly reverses one form of blindness. Like the metabolic condition Glybera treats, both these diseases are incredibly rare. Glaxo estimates that only 14 cases of severe combined immune deficiency come to light each year in Europe.

The combination of rare diseases with a costly new technology that needs to be used only once is what could lead to exorbitant prices. Analysts have said Spark’s treatment might cost $500,000 per eye. And with nearly 670 gene or cell-therapy trials under way, and 68 in the later stages, known as Phase III, it’s becoming “urgent” to understand how these therapies will be paid for, says Morrie Ruffin, managing director of the Alliance for Regenerative Medicine, a trade and advocacy group. The current system “was not established with these types of products in mind,” he says.

Executives at Glaxo say they don’t expect to charge anything near a million dollars. But with $35 billion a year in revenue, they don’t need to rely on gene therapy for profits. “From a pure business, money-making perspective, it’s a challenge,” says Sven Kili, head of gene-therapy development at Glaxo. “It’s a single treatment that lasts indefinitely and patients don’t come back. And there aren’t many of them. It is not something that would make a venture capitalist jump with joy.” He says Glaxo is thinking of a price “way below” what people might expect for once-in-a-lifetime therapy and “nowhere near” that of Glybera.

There isn’t much doubt that Glybera’s approval in 2012 was important. It showed that gene therapy, once branded as too risky, could be safe and was ready to be commercialized. Since then, a large number of new companies have been formed and more pharmaceutical firms have shown interest. Piper Jaffray, the investment bank, says 2015 was the most “momentous” year yet for gene therapy. About $2 billion was raised by 10 public gene-therapy companies in the U.S. last year.

“No one knew whether this was going to be a viable strategy,” says David Schaffer, an expert on gene-therapy viruses at the University of California, Berkeley, who is also on UniQure’s board.  “It was just a huge boost for the field to know someone got something over the finish line. “

UniQure started out as Amsterdam Molecular Therapeutics, a 1998 spin-off from the University of Amsterdam. Doctors there had zeroed in on an enzyme, lipoprotein lipase, that muscles produce and which digests fat in the blood. In some people, both copies of the gene to make the enzyme are mutated, and don’t work correctly.

The Amsterdam company spent more than $100 million testing the drug and carving a path through Europe’s medical rules and regulations, which weren’t geared to consider a new technology like gene therapy. Initially, for instance, regulators said they expected a clinical trial of 342 patients. Executives wryly noted that there were only 250 people with the disease in all of Europe. “The filing was a nightmare experience,” says Sander van Deventer, a biotech investor who was once the company’s chief medical officer. “They didn’t have the know-how to approve an advanced therapy.”

But the data for Glybera wasn’t rock-solid either. The drug was given to just 27 people in three “open-label” experimental studies, meaning no patient got a placebo. Those tests never showed a lasting change in fat levels in their blood—though the company argued that people who got the drug had fewer episodes of pancreatitis, the painful complication of the disease.

By 2012, the company had failed twice to convince European regulators. It decided to reorganize, incorporating as UniQure, and mounted a last, successful attempt to get the drug approved. “We didn’t want to give it up even though the commercial outlook was not good,” says Van Deventer.  

A year after the approval by Europe’s top medicines' body, UniQure was able to go public on the Nasdaq, raising $82 million. Although it called the commercialization of Glybera a top priority, by last fall UniQure instead chose to scrap its plans to sell the drug in the U.S. after the Food and Drug Administration said new, expensive trials would be needed.  

Nor has Glybera convinced the national regulators in Europe who decide what drugs get reimbursed. Last year, French authorities said they would not pay for the drug. Germany judged Glybera’s benefits “non-quantifiable.” It leaves doctors and insurers to make decisions on a case-by-case basis. In the Netherlands, where the technology was invented and supported by R&D grants, the health minister complained that Glybera was part of a pattern of price gouging by drug companies.

The decision to charge around $1 million for Glybera (the exact amount depends on the patient’s weight) does raise questions. Some people think gene therapies should be paid for in yearly installments, and only so long as they keep working. But that wasn’t possible with Glybera because there was no clear-cut way to track the effects of the drug. “There is pressure to pay all at once—the payers said they wanted to pay up front,” says Van Deventer. “But that created a lot of anger in Europe. It’s a shame. It became the 'one-million' therapy. There is an anti-innovation climate and people don’t want to pay for it at all.” 

Even modest expectations for Glybera’s sales now look much too rosy. (Early last year, one analyst projected peak revenues of $57 million a year.) Part of the reason is that the disease is so rare, just one in a million people, that it’s often not diagnosed correctly, and finding customers is hard work. Following its reorganization, UniQure sold European marketing rights to Chiesi, a small Italian drugmaker with little experience selling such a complex therapy. 

That means even the few patients who want the West’s first gene therapy can’t easily get it. Steinhagen-Thiessen says she knows of a man in Luxembourg and a family in Czechoslovakia who are interested, but they don’t yet have a way to pay for it. A Chiesi spokesperson says it is working to create a registry of patients in Europe.

Eventually, the German insurer DAK did pay 900,000 euros, or about $1 million, to cover the cost of treating Steinhagen-Thiessen’s patient. Most of that went to pay for Glybera. “I think the price is absolutely too high. I am not sure that other insurance companies are willing to pay that,” says the doctor. “From an ethical point of view we should have a lower price. I don’t think the companies can make a profit by it in any case.”

Schaffer, the Berkeley professor, says gene-therapy companies still have work to do demonstrating their concepts. It may be too soon to turn a profit. “I think that the field should focus on disease targets which are well known, where there is a big chance for medical success, whether or not there is commercial success,” he says. “People who are investing in early stage biotechs are doing it because of the promise of the future, not immediate returns tomorrow.”