Glass bottles are filled with liquid pharmaceuticals at the Fresenius Kabi plant. The company is a leading supplier of I.V. generic drugs. (Fresenius SE & Co. KGaA via AP Images) (Fresenius SE & Co. KGaA/Fresenius SE & Co. KGaA via AP Images)

A decade ago, physicians who treat epilepsy got what seemed like a piece of good news: Eight companies had received federal approval to sell a generic version of an injectable lifesaving drug.

Doctors liked the brand-name drug Cerebyx because it was safer and easier to use than a previous medicine that stopped seizures but could cause terrible skin reactions. The only problem was that it was too expensive for many hospital pharmacy budgets. A widely available and cheaper generic version would remove those cost barriers — or so doctors thought.

But the introduction of Cerebyx’s generic form, fosphenytoin sodium, in 2007 — called “fospheny” for short by some doctors — did not over the long term produce the robust marketplace competition that is one of the main arguments for the generic industry.

The trajectory of generic Cerebyx shows the limitations of relying on the market alone to bring down drug prices and ensure supply — particularly for hard-to-make injectables or low-volume drugs, which are critical to the people who need them.

It’s the flip side of the low prices in the generic industry: Thin profit margins can turn temporary challenges such as active ingredient shortages or manufacturing delays into decisions to discontinue a drug altogether.

Companies stopped making fosphenytoin for several reasons, including technical challenges, production problems and decisions to concentrate on producing other drugs.

And Cerebyx was not an outlier. Recent studies of the generic drug industry show that low levels of competition may be far more normal than the public appreciates.

“More than 50 percent of generic drugs are supplied by one to two manufacturers; that really turns on its head the idea this is a competitive, commodity-like market, like widgets or milk or things that people use every day,” said Rena Conti, a health economist at the University of Chicago. “While there may be robust entry into these markets, over time these markets degenerate into monopoly or duopoly supply.”

Concerns about high drug prices often focus on the lack of generic competition, a problem the Trump administration has decided to tackle head on. The Food and Drug Administration last month hosted a day-long meeting devoted to understanding how companies may be gaming the approval process to delay or prevent competition from emerging.

Overall, generic drugs have been a major source of savings to the American health-care system, accounting for nearly nine out of 10 of all prescriptions filled but only about a quarter of the total spending on prescription drugs, according to the Association for Accessible Medicines, a trade group for the industry. But concerns that the market isn’t functioning as intended have been sparked by drug shortages and old drugs that have risen in price.

When it came to Cerebyx, generic competition blossomed — at first. Then, it withered. The drug’s price plummeted initially, but makers, facing production and commercial problems, left the market and the drug’s price crept back up. It has cycled through various shortages over the years.

The fading of competition has meant that years after it went generic, the lifesaving emergency drug wasn’t always easily available.

“We would have loved to use the fospheny and have it available,” said Tricia Ting, regional director of epilepsy at MedStar Georgetown University Hospital. “We were always told it’s so expensive . . . it wasn’t in our control as clinicians. I think we would have used it like gangbusters.”

Only two of the 13 approved generic versions of Cerebyx are being sold, according to Erin Fox, director of drug information at University of Utah Health. The list price, which plunged from $65 for a 10-milliliter vial of the brand name in 2007 down to a few dollars for the generic, according to Truven Health Analytics, has swung back up. Today, the generic price is $48 and Cerebyx is nearly $100 a vial.

Most people think about the drug price problem in simple terms: new cancer drugs or other specialty medicines with eye-popping price tags. For years, huge savings have been wrung from producing generic versions of old drugs.

But over the past few years, drug shortages and a few extreme cases in which companies have raised old drug prices to an outrageous level have fueled concern that this spending cliff isn’t working as well as it should. A Government Accountability Office report last year found that more than 300 of 1,441 generic drugs had at least one “extraordinary price increase” over a five-year period.

Most of the focus centers on the tactics companies use to actively delay or quash competitors, and regulatory obstacles that make it harder or less attractive for drugs to get approved in the first place.

But according to the FDA, more than half of its approved generics are not being sold and another 14 percent are of “unknown marketing status.” The agency said that less than a third of the generics it approved are being actively marketed. For some old drugs that have been eclipsed by newer and better drugs, that drop-off makes sense. But it also highlights a less obvious barrier to robust competition in the generic market: keeping competition going.

A consolidated market

The modern generic drug industry was kick-started by the Hatch-Waxman Act of 1984, considered one of the most important pieces of legislation in American health care. The reward for pharmaceutical companies investing in risky research and development was the power to set prices and reap the rewards of government-granted monopolies for a limited amount of time. Once their patents expired, competition could rush in.

Over the years, that set off a cottage industry of firms trying to outmaneuver one another. Drug companies use the legal system to try to discourage or delay generic competitors. They file “citizen petitions” with the FDA to raise questions about competitors’ products. They limit the distribution of their drugs so that other companies can’t obtain enough of the product to prove to regulators that their version is equivalent. They merge with competitors. Some took advantage of well-meaning policy incentives, finding old drugs that had never gone through the approval process, collecting the data to have them approved — and then raising the price.

“In the 1980s, when we passed the Hatch-Waxman Act, we thought we were getting a good balance between incentives for innovation and development of new drugs with the ability to have competition, which would have the advantage of giving us market forces to lower drug prices,” said Henry Waxman, a former congressman who co-sponsored the law. “I think that balance worked for a number of years, but it’s clearly not in balance today.”

Many of those problems would be amenable to policy or legislative fixes. A bill proposed in the Senate would expedite the review of drugs that are facing a shortage or that lack competition. But there is growing evidence that the generic market has become extremely consolidated, and it is far less clear what the FDA can do to change that.

A recent study in the Annals of Internal Medicine found that of 1,120 generic drugs, half were sold by just two companies. Low levels of competition were typical for drugs that were used by smaller numbers of people — and those drugs with the least competition experienced the biggest increases in price.

“It was surprising to me that the generic market looked as consolidated as it did,” said Aaron Kesselheim, an associate professor of medicine at Harvard Medical School who co-wrote the study. “That’s been building over the last 10 to 20 years, and I think it’s something that policymakers are going to have to look at now, given the fact the system relies so much on generic drugs to save money. This seems like a trend that should raise a red flag.”

What happens when the industry consolidates? A clue may lie in the injectable generics industry, where mergers, production problems and manufacturing delays have led to shortages.

Fox, at the University of Utah, said that when she is researching a drug shortage, she almost inevitably finds herself tracking down a drug that is made by just one or two, or even three producers — one of which has had some kind of a glitch. Maybe the active ingredient has had a shortage, or the producer suffered a problem with its manufacturing process.

“The same characteristics that make a product susceptible to shortages are the same characteristics that make a product susceptible to high price spikes,” Fox said.

In the case of a generic version of Cerebyx, fosphenytoin shortages were caused at various times by companies discontinuing products , voluntary recalls, manufacturing delays and production lines that went offline for various reasons.

The price didn’t spike, but after the price of the generic plunged, to as low as $3.45 for a 10-milliliter vial in 2011, it crept back up.

Where did the competition go?

The low price may have been too low for some companies to make money, or the thin margins may have led companies to make a business decision to devote their infrastructure to making a more lucrative drug.

Apotex stopped selling the drug in 2009 when the company ceased making injectable products due to “market dynamics,” the company said in a statement. But a spokeswoman said the drug was not profitable when the company discontinued it and was selling at less than $7 a vial.

Sun Pharmaceutical Industries received approval from the FDA in 2008, but didn’t launch the generic drug because of “technical challenges,” a company spokesman said.

Teva Pharmaceutical Industries put the production facility where it made the drug on a voluntary manufacturing hold in 2010. Fosphenytoin, which last sold for $6.60 for a 10-milliliter vial, was one of the drugs that the company did not bring back online. A Teva spokeswoman did not explain why.

Fresenius Kabi in 2014 recalled lots of the drug because of potential “glass delamination” which can leave glass particles in the vial.

The company has been working to change the vial to make it safer and is waiting for the FDA’s response, spokesman Matt Kuhn said in an email.

Pfizer had discontinued Cerebyx, but brought it back in 2013 “in response to a public need identified by regulators,” spokeswoman Rachel Hooper said in an email.

“Given the complexity of the production process, significant investment was required to produce reliable supply while ensuring the highest quality and safety standards,” Hooper said.

Mylan Pharmaceuticals and Amneal Biosciences launched their versions of the drug last year, at $48 a vial.

Mylan’s FDA approval for the drug, however, isn’t new — it dates back to 2010 and came with a company it acquired in 2013. A spokeswoman said that it launched the drug in December “upon operational readiness.”

Cerebyx and its generic aren’t a big, top-selling drug. Total annual sales of the drug were about $36 million, according to data disclosed by Mylan when it launched its generic in December. In part, Ting said, the early pricing challenges meant fosphenytoin wasn’t as widely used as it might have been, and she said many physicians may not even be aware that it is available in generic form. Meanwhile, other drugs have also gained popularity in treating seizures.

In that way, fosphenytoin is representative of the ups and downs of the generic industry — a big one that is made up primarily of small markets. A new analysis by Conti found that the median market size for any particular form of a drug is about $10 million in annual sales revenue — a far cry from the billion-dollar, blockbuster brand- name drugs. Keeping that industry vibrant will not only require more companies to be able to enter, but also keeping them in the game.

“The big thing is to prevent more erosion of competition in these markets,” Conti said. “The second piece, about what to do about markets that already have very limited competition, is a harder nut to crack.”