Holding the Government Accountable

Drug Money: FDA Depends on Industry Funding; Money Comes with “Strings Attached”

This is part 1 in a series; read the second part, “DRUG MONEY: In FDA Meetings, ‘Voice’ of the Patient Often Funded by Drug Companies.

It is often said that some government regulator or another is in the pocket of industry.

In the case of the Food and Drug Administration, it’s a fact prescribed by federal law.

The bottom line: The FDA is addicted to drug money.

To understand how, consider the legally mandated process that has been playing out over the past year between regulators at the FDA and representatives of the drug industry they oversee.

From September 2015 to January 2016, teams of FDA officials held 70 meetings with drug company executives and lobbyists to set goals that could have far-reaching consequences for the pharmaceutical industry, the FDA, and anyone who uses or pays for prescription drugs.

At issue: How the Food and Drug Administration should go about approving new drugs.

The meetings, closed to the public, weren’t just talks. They were negotiations. And if the two sides didn’t reach an agreement, the FDA could find itself in a world of hurt.

The agency, whose responsibilities include making sure that prescription drugs sold in the United States are safe and effective, receives almost three-quarters of its funding for that work from drug makers. In Fiscal Year 2015, the agency spent about $1.1 billion on prescription drug oversight—more specifically, those activities the agency categorizes as part of its “process for the review of human drug applications.” Congress delivered 29 percent of that money—$331.6 million. Drug companies provided the rest—$796.1 million.

Over the past two decades, “user fees” paid by industry have climbed from 35 percent of the FDA’s spending on such oversight to 71 percent.

The bottom line: The FDA is addicted to drug money.

If drug companies merely wrote the checks, there might be less cause for concern. Indeed, many federal agencies collect user fees. However, at the FDA, as a study by the Institute of Medicine has observed, the corporate money comes with “strings that are attached.”

Negotiating With Industry

Every five years, the law requiring drug companies to pay the fees—known as the Prescription Drug User Fee Act, or “PDUFA”—expires. Before Congress considers renewing it, the FDA must negotiate with industry to keep the money coming. That’s what the law demands: “negotiations with the regulated industry.” In practice, the FDA has interpreted that to mean negotiate over how it spends the money and assesses drugs. The negotiations produce recommendations for Congress and an FDA “commitment letter” laying out goals for the agency. It’s unclear what would happen to the user fees if the FDA and industry didn’t agree on terms.

The FDA fee negotiations have placed too much emphasis on getting drugs approved quickly and not enough on making sure they actually work—and work safely.

This arrangement gives the pharmaceutical industry extraordinary influence over its government overseer. It leaves the regulator beholden to the regulated. It spares taxpayers some of the burden of paying for consumer protection—unless those costs are passed along to consumers in the form of higher drug prices. And it has the potential to compromise consumer protection.

The system at the FDA is “unique in the degree to which industry sets the terms of the agenda,” said Daniel Carpenter, a Harvard professor of government who has published work on the FDA and on “regulatory capture,” a process by which special interests gain influence over their regulators.

As some critics see it, the FDA fee negotiations have placed too much emphasis on getting drugs approved quickly and not enough on making sure they actually work—and work safely. They say the FDA should be focusing more on issues that get little, if any, attention in the negotiations, such as preventing misleading ads from reaching the public.

Paul Brown of National Center for Health Research, a think tank that receives no money from the drug industry, framed the issue at an August 2016 FDA forum.

“Have user fees changed FDA's priorities?” Brown asked. “Is FDA now treating industry as a customer that it needs to please instead of acting as a regulator to ensure the public health?”

For consumer advocate Michael Carome, director of Public Citizen’s Health Research Group, the answer is an emphatic “yes.”

The system “basically has tilted the agency more towards protecting the interests of industry rather than the interests of public health,” Carome said in an interview.

Former FDA Commissioner David Kessler, who helped establish the user fees, said in an interview that, “on balance, the FDA is stronger” because of them. He also offered a warning: “We have to be mindful that subtle biases and effects could creep into the system.”

The months-long negotiation that played out as part of the latest five-year PDUFA cycle called for increases in FDA hiring and user fee funding over the coming years. It also entailed an extensive review and rethinking of how the FDA does its job. Tentatively, it has set the FDA on a course to increasingly look beyond data from controlled clinical trials—the costly mainstay of medical research—when assessing drugs. In the commitment letter, in buzzwords that track industry’s wish list, the FDA has pledged to pursue the use of “patient-reported outcomes” and “real-world evidence.” The agency also promised to explore “novel clinical trial designs,” including “highly innovative” ones for which “simulations are necessary.”

Some observers fear that such approaches may not be scientifically sound, that they could be more susceptible to manipulation, and that they could lead to lower standards.

Unless “real-world evidence” follows principles of well-designed clinical trials, such as comparing the experimental drug to a placebo and committing to a study design before observing the results, “then it should be regarded as simply a deregulatory exercise and a departure from rigorous science,” Harvard's Carpenter said by email.

The drug industry has been pursuing similar policy objectives in Congress, and some of these themes run through legislation passed by the House in 2015 with a motherhood-and-apple-pie name: the 21st Century Cures Act. The bill is one of lawmakers’ top priorities in the current lame duck session of Congress. Even if that legislation goes nowhere, industry could advance part of its agenda through the deal it made with the FDA.

If the results of the latest user fee negotiations benefit the public, and not just the drug companies, that would be a bonus: The system is explicitly set up to make sure industry gets its money’s worth out of the FDA.

“User fees pay for services that directly benefit fee payers,” the FDA said in a recent slide presentation.

“And because it’s a fee, it’s really intended to benefit directly . . . the fee payers and benefit them in a way that exceeds the benefit to the general public,” Theresa Mullin, who led the FDA’s negotiating team, explained at the August FDA forum.

More specifically, the fees are intended “to speed and enhance the [drug] review process,” the FDA presentation said.

Speeding Drug Approvals

Drug testing at the FDA
(Photo: Food and Drug Administration / Flickr)

Since the prescription drug user fees were enacted in 1992, the review process has gotten much faster.

The median time from manufacturer application to approval for drugs and biologic products involving a new medicine—a “new molecular entity” in FDA jargon—was 8.5 months in 2015, down from 19.1 months in 1993, according to an FDA analysis from December 2015.

That means that, typically, new drugs could start generating revenue in less than half the time.

Meanwhile, approval rates have climbed sharply, to the point that the FDA approves almost all new medicines submitted for review. Among those submitted during fiscal year 2015 and brought to a decision, 95 percent were approved on the first try, according to the FDA analysis. That was up from 36 percent in fiscal year 1993.

Former FDA drug reviewer Ron Kavanagh told POGO that, when he was at the agency from 1998 to 2008, PDUFA’s target dates for FDA action left too little time to review drug company submissions, which could total 160,000 pages not counting supporting data. Reviewers were told not to worry about studying all of the material, Kavanagh said.

“There’s a lot of things I simply didn’t look at,” Kavanagh said. “And even without looking at things I barely made the deadlines.”

Kavanagh shared an internal FDA email from 2007 in which he gave this account: “I finally had to stand up and say that I would take being written up for insubordination and would risk a poor performance evaluation, but that I would not curtail my evaluation of a potential safety concern simply to meet a PDUFA goal date.”

He was later fired.

The FDA says speedier approvals have not come at the expense of consumer protection.

“Since the implementation of PDUFA I in 1992, FDA has used PDUFA resources to significantly reduce the time it takes to evaluate new drugs and biologics without compromising its rigorous standards for demonstration of safety, efficacy, and quality of new drugs and biologics before approval,” the agency said in an undated report to Congress for the 2015 fiscal year. “The efficiency gains under PDUFA have revolutionized the drug review process in the United States and enabled FDA to ensure more timely access to innovative and important new therapies for patients.”

Patient-Reported Outcomes Clash with Evidence

Prescription drugs

The FDA’s recent approval of a drug for Duchenne muscular dystrophy shows how far from rigorous the FDA’s standards can be—and what “patient-reported outcomes,” a focus of the latest negotiations, can mean.

The FDA approved the muscular dystrophy drug in September 2016 over the vociferous objections of FDA scientists who essentially argued that there was no sound evidence it offered any benefit.

“By allowing the marketing of an ineffective drug, essentially a scientifically elegant placebo, thousands of patients and their families would be given false hope in exchange for hardship and risk,” Ellis F. Unger, director of the FDA’s “Office of Drug Evaluation – I,” wrote in a July memo.

“I argue that this would be unethical and counterproductive,” Unger wrote. “There could also be significant and unjustified financial costs – if not to patients, to society.” The drug is priced at $300,000 per year.

Duchenne muscular dystrophy is a fatal degenerative disease that afflicts boys, and until the agency’s action in September, there was no FDA-approved treatment. Patients and parents urged the FDA to approve the drug, called eteplirsen. The research that drug-maker Sarepta Therapeutics used in support of its application was deeply flawed, the FDA found, and “included analyzable data from only 11 patients,” Unger wrote.

When an FDA advisory committee met in April to consider the drug, the manufacturer arranged for a presentation of “Patient and Caregiver Reported Outcomes,” and the committee heard testimony from patients who had been subjects of the clinical research.

“The testimonies of these patients were quite consistent and remarkably positive: all were convinced that eteplirsen had made a substantial positive impact on their physical performance, improving numerous aspects of their lives,” Unger recounted. “Many of the Committee members seemed obviously moved and deeply affected by these testimonies and videos, seemingly convinced that there was a treatment effect.”

But there was a catch. FDA reviewers and advisory committee members were unable to reconcile the patient testimonies with the data the drug maker had gathered. The “testimonies spoke of improvement; the data showed progressive worsening,” Unger wrote, using italics for emphasis.

The FDA’s decision ultimately rested with Robert Califf, the head of the agency, who formerly ran clinical trials for drug companies. Califf concluded that the drug’s “clinical benefit has not yet been established.”

He approved it anyway.

(Past reporting by POGO shows how low the FDA has set the bar for approval of other drugs and what the FDA has glossed over or overlooked along the way. See highlights of the “Drug Problems” series.)

Industry Agenda

When the latest round of PDUFA negotiations ended and the FDA unveiled a draft of the new commitment letter, the main industry lobbies endorsed the outcome and lavished praise on the process.

The user fee system “is, in my mind, a shining example of a program that has produced positive and tangible results that matter not only to patients, but they drive innovation,” Sascha Haverfield of Pharmaceutical Research and Manufacturers of America (PhRMA) said at an August 15, 2016, FDA forum.

The FDA user fee negotiations illustrate a broader phenomenon: the pharmaceutical industry’s influence over many institutions that are supposed to serve as checks on its interests.

In theory, there was at least a modest counterweight to industry’s negotiating power. While the FDA’s negotiations with industry were under way, the agency held six other meetings with what it called “stakeholder groups.” Those sessions were intended to fulfill a legal requirement, which the law frames as an obligation of the FDA’s bureaucratic parent, the Department of Health and Human Services.

“Not less frequently than once every month during negotiations with the regulated industry, the Secretary [of Health and Human Services] shall hold discussions with representatives of patient and consumer advocacy groups,” the law says.

However, POGO’s investigation found that of the 42 “stakeholder groups” listed as participants in the meetings, the vast majority had connections to pharmaceutical companies. More than 90 percent of them received funding from drug companies. More than 35 percent of them had drug company personnel on their governing boards. (See “In FDA Meetings, ‘Voice’ of the Patient Often Funded by Drug Companies” for more on this.)

The FDA user fee negotiations illustrate a broader phenomenon: the pharmaceutical industry’s influence over many institutions that are supposed to serve as checks on its interests, from universities and research labs to doctors’ offices, from the pages of medical journals to the halls of Congress, and even to the FDA’s sprawling campus in Silver Spring, Maryland. It is model of how an industry can mold its ecosystem.

It is symptomatic of a system in which the chairman of a key Congressional committee goes from championing a new prescription drug benefit to leading the pharmaceutical industry’s main lobbying organization—a job that paid millions. A system in which FDA drug reviewers move to more lucrative positions at drug companies, former FDA commissioners become directors of pharmaceutical firms, and a university researcher goes from building an institute that runs clinical trials for drug companies to heading the FDA.

It’s possible that drug companies approached the latest fee negotiations with the best of public-spirited intentions. But any impulse to give them the benefit of the doubt should be tempered by the often controversial prices drug companies charge to generate profits, consistent with their objectives as for-profit businesses—and by the tens of billions of dollars such companies have paid to settle allegations of hiding dangerous side effects, promoting drugs for uses not approved by the FDA, defrauding Medicare and Medicaid, and otherwise overreaching in the push to sell their products. Such cases, including alleged repeat offenses, have placed drug firms near the top of the Project On Government Oversight’s rankings of federal contractors charged with civil or criminal wrongdoing. As of November 2016, when ranked by the amount paid in fines and settlements since 1995, drug firms took three of the top four spots (trailing only BP, whose Deepwater Horizon oil spill fouled the Gulf of Mexico).

At other federal agencies that rely heavily on user fees, paying a fee buys companies the right to conduct business—but not to negotiate with their regulator. For example, at the Securities and Exchange Commission, user fees are a cut-and-dried affair. If a firm wants to sell stocks or bonds, it must pay the SEC a registration fee, much like a tax. The SEC adjusts the rate each year to meet budgetary targets set by Congress. Corporate lobbyists can lean on Congress to change the way the SEC operates or to cut its funding, but they don’t get to pull directly on the agency’s purse strings.

The Nuclear Regulatory Commission is required to recoup about 90 percent of its annual appropriation through various fees, some of which are levied on operators of nuclear power plants. When the NRC proposes the fee amounts each year as part of a standard rulemaking process, the companies responsible for paying the fees can submit public comments—just like everyone else.

At the FDA, makers of brand-name prescription drugs aren’t the only ones who pay fees and get a seat at the negotiating table. In addition to PDUFA, which covers brand-name drugs, there are parallel programs—and parallel negotiations—for the medical device and generic drug industries. The fees those programs, known as MDUFA and GDUFA, generate for the FDA are separate from the numbers discussed above. Altogether, in fiscal year 2016, user fees accounted for about $2 billion of a $4.7 billion budget for “Medical Product Safety and Availability,” according to an FDA budget summary.

POGO asked the FDA how the user fees and the negotiations affect the agency and whether it would prefer to be funded through appropriations or user fees. The FDA ducked the questions.

“The FDA follows the statutory requirements for developing recommendations to present to Congress on the reauthorization of PDUFA, including by holding public meetings to allow the public to present its views on reauthorization, conducting negotiations with regulated industry, and holding monthly stakeholder meetings during those negotiations, and . . . conducting the other activities specified by statute,” FDA spokeswoman Tara Goodin said in an emailed response.

PhRMA spokesman Andrew Powaleny said the PDUFA VI agreement “was carefully negotiated between the biopharmaceutical industry and FDA with regular input from the public, patient advocacy and professional groups.” It will, he said by email, “play a critical role as we continue working to help patients live longer, healthier lives.”

The User Fee Era

From its inception in 1992, the prescription drug user fee system was based on a bargain: Drug companies impatient to move their products through a backlogged FDA review pipeline would provide funds for the FDA to increase its staffing, and the FDA would aim to complete reviews by new target dates.

The AIDS crisis had provided momentum. Desperate, dying patients were willing to take greater risks on new treatments, and they were imploring the FDA to give them that chance, as recounted in an FDA history. Their plight highlighted a basic tension for regulators. Approving dangerous or ineffective drugs can hurt people. So can delaying treatments that help.

The stakes are also high financially—for drug makers, consumers, and taxpayers. Once drugs are approved, private health insurance plans and government programs such as Medicaid are called upon to pay for them. Though the FDA is not supposed to consider the price of a drug or whether it represents a good value for money, the agency’s decisions affect the nation’s health care spending.

User fees may appear to save the taxpayers money, but at an unacceptable cost to public health.

Kessler, who was FDA commissioner in the 1990s and helped create the user fee system, told POGO that getting the money the FDA needed from Congressional appropriations “was undoubtedly our first, second, and third choice,” but “that was not realistic.”

The goal of the user fee law was “solely to accelerate the process for reviewing” new drug applications, a report by the Congressional Research Service said.

In the years after fees were introduced, a series of FDA drug approvals came under fire as highly promoted medicines were later associated with dangerous side effects. For example, in 2004, Merck withdrew the arthritis drug Vioxx from the market, saying it increased the risk of heart attacks and strokes. FDA safety official David Graham and other researchers estimated that the drug caused as many as 139,000 heart attacks and sudden deaths, but the agency had withheld permission for them to publish their findings in a medical journal, Graham testified at a Senate hearing.

“I was ordered to withdraw the paper or 'face severe consequences'—which I interpreted to mean that I would be fired," The Los Angeles Times quoted Graham as saying when the research was ultimately published.

The culture of the FDA “views the pharmaceutical industry it is supposed to regulate as its client. It overvalues the benefits of the drugs it approves, and it seriously undervalues, disregards and disrespects drug safety,” Graham said at the 2004 Senate hearing.

Despite the user-fee money, pressures to review drugs expeditiously have been a strain on the agency.

An internal FDA email chain from 2007 obtained by POGO provided a glimpse of the time pressures. Following up on a staff meeting at which employees reportedly voiced concern about PDUFA deadlines, John K. Jenkins, director of the Office of New Drugs, thanked employees for “your candor in expressing your concerns about the pressures you are facing every day.”

“I agree that the competing pressures, workload, and goals we face today are daunting and that we don’t have nearly enough staffing and resources to do our work to the level we strive for as stewards of the public health,” Jenkins wrote. “We clearly cannot do all that we are being asked to do,” he added.

Meanwhile, the Institute of Medicine conducted a sweeping assessment of the FDA at the agency’s request and issued its findings in a watershed 2006 report.

“For some staff and policy analysts, user-fee funding, combined with industry’s considerable role in shaping PDUFA-associated goals and expectations, further reinforces the perception that the industry has become a primary driver of the agency’s priorities and performance,” the Institute of Medicine reported. “The notion of ‘regulatory capture’ has been employed to describe the state of affairs created or, more likely, exacerbated by the user fee system, namely that powerful industry interests control or strongly influence the regulatory agency’s decision making.”

Criticizing the FDA’s “overdependence on PDUFA funding with the strings that are attached,” the Institute of Medicine recommended that new FDA safety functions be funded through Congressional appropriations rather than user fees.

The following year, at a George Washington University panel discussion, former FDA commissioners unanimously agreed that appropriations would be better than user fees, the FDA history recounts. “Appropriations, no question,” Kessler said.

“User fees may appear to save the taxpayers money, but at an unacceptable cost to public health,” 22 academic experts, former FDA officials, and others said in a 2007 open letter to the FDA’s Congressional overseers.

In a 2008 study published in The New England Journal of Medicine, Harvard scholar Carpenter and his co-authors found that PDUFA deadlines appeared to be causing safety problems. Under PDUFA, drug approval decisions tended to take place in the two months before their deadlines, and drugs approved just before the deadlines tended to have more safety problems after they were approved. For example, they were more likely than other drugs to be withdrawn from the market or to have major warnings added to their labels, the study found.

For much of the user-fee era, the FDA was allowed to spend the user fee revenue only on activities leading up to drug approval—thereby hastening approvals but limiting the FDA’s ability to monitor the safety of drugs once they were in widespread use. In time, that changed, and the focus on postmarket surveillance increased. Postmarket surveillance is important in part because even the largest clinical trials may not reveal what will happen when drugs are used over the long term by much larger numbers of people.

Nonetheless, tracking the performance of drugs once they are on the market continued to be a weakness at the FDA.

“FDA’s lack of reliable, readily accessible postmarket safety data has prevented the agency from publishing required reports in a timely manner and has restricted its ability to conduct systematic oversight,” the Government Accountability Office said in a December 2015 report. As of October 2015, the FDA had not published required annual reports containing data on postmarket studies for fiscal years 2013 or 2014, the GAO said.

The FDA lagged in tracking postmarket safety even as it was approving many new drugs on an expedited basis under programs for drugs deemed worthy of “fast track” and other special consideration, the GAO said.

Whether drugs approved on an expedited basis have deserved the shortcuts is a separate question.

At an FDA meeting last year, Aaron Kesselheim, who runs a program on regulation, therapeutics, and law at Harvard Medical School, noted that a program created in 2012 for so-called “breakthrough therapies”—drugs that, based on preliminary data, have the potential to substantially improve the treatment of serious or life-threatening diseases—was the fifth pathway created for drug companies to get products approved on an expedited basis. Others bear such names as “accelerated approval” and “priority review.”

“I think what these data show is over time, an increase and creep of designations towards drugs that may not be truly the most transformative products,” Kesselheim said.

In a study published last year in The BMJ, formerly known as British Medical Journal, Kesselheim and co-authors said special pathways were meant as exceptions to the standard FDA review process. As it turned out, “the exceptions had become more common than the rule.”

Despite the user-fee money, pressures to review drugs expeditiously have been a strain on the agency.

For example, when FDA officials met with representatives of the pharmaceutical and biotech industries last year to negotiate the renewal of PDUFA, the agency said that the breakthrough therapy program was turning out to be more work than expected. “FDA stated that because the program was not funded with new resources at its inception, the demand is currently met through either uncompensated staff overtime or by rebalancing existing priorities,” according to minutes of the negotiating session.

The FDA has pointed a finger at Congress.

When the user fees expire every five years, the bills renewing them are essentially must-pass legislation. That makes them convenient vehicles for other provisions—giving lobbyists and lawmakers an additional opening to influence the FDA’s workings.

When Congress takes up the reauthorization of PDUFA, “additional requirements are often placed on the agency without being properly resourced,” FDA spokeswoman Goodin said in an email to POGO. “This phenomenon may lead to the notion of the agency not being adequately resourced to take care of all the work that society asks us to do.”

Currently, the fees drug companies pay for the FDA to review new drugs run about $2 million each for applications that involve clinical trials. The fee doesn’t guarantee that the FDA will approve the drug. However, under PDUFA, the FDA has committed to dozens of performance goals, from how quickly it acts on different types of meeting requests from drug companies to how much time it takes to review drugs. The agency is required to measure and report its compliance with those goals.

The FDA’s latest annual “Performance Report to Congress” under PDUFA totals 64 pages. It notes that in fiscal year 2014, the FDA was supposed to review at least 90 percent of certain expedited applications within six months of their filing dates. The FDA exceeded that goal, reviewing 96 percent of them within the six-month time frame. As the document explains, activities “not captured by PDUFA goals and therefore not presented in this report include . . . the ongoing monitoring of drug safety in the postmarket setting.”

The target dates for FDA drug review decisions, known as “PDUFA dates,” focus the minds of people inside the agency and out. Investors hang on them, knowing that company fortunes can rise or fall depending on the outcomes. Hence, a constant stream of headlines such as “Egalet (EGLT) Says FDA Will Not Meet October 14 PDUFA Date” and “DVAX Vaccine Faces Crucial PDUFA Date.”

The FDA history of PDUFA—co-authored by Janet Woodcock, director of the agency’s Center for Drug Evaluation and Research—warned that drug industry money has a price:

“The provision of industry fees to FDA has undermined public trust in the agency, which is perceived by some as having lost independence and credibility as a result of accepting industry money.”

PDUFA VI: The Latest Round of Negotiations

The latest round of PDUFA negotiations—the sixth, known as “PDUFA VI”—charted goals for the FDA through fiscal year 2022. In drafting the goals, the industry and FDA negotiators got under the agency’s hood and examined its inner-workings.

The 70 negotiating sessions were almost double the 36 sessions held five years earlier. In a change since last time, there were subcommittees to discuss each of several facets of the agency, from finances to information technology, and from pre-market oversight of drugs to postmarket oversight. To guide it all, there was a steering committee whose participants included the global head of regulatory affairs at Novartis, the senior vice president for global regulatory affairs and clinical safety at Merck, the global head of regulatory policy and intelligence at Alexion Pharmaceuticals, and lobbyists for Pharmaceutical Research and Manufacturers of America (the main trade group for drug companies) and Biotechnology Industry Organization (the main trade group for biotech firms, renamed this year as “Biotechnology Innovation Organization”).

POGO sought to observe negotiating sessions, but the FDA said they were closed to the public.

Federal law required the FDA to release minutes of the meetings to the public. The minutes the FDA posted were written in vague, general terms—descriptions such as, “Industry proposed that PDUFA VI build on the PDUFA V performance goals for further implementation of a structured benefit-risk framework to inform regulatory decision-making.” The agency gave the public meager real-time insight into the discussions.

The commitments the negotiations produced establish broad mandates for the FDA while leaving many details to be filled in later.

For example, the FDA agreed to hold early consultations with drug companies about the feasibility of approving individual drugs on the basis of measures never used before—new types of “surrogate endpoints.” Surrogate endpoints are proxies for what the drugs are supposed to achieve. If a drug is supposed to prevent heart attacks, a surrogate might be lowering cholesterol. If a drug is supposed to save people from dying of cancer, a surrogate might be shrinking the size of a tumor.

Designing clinical trials around surrogate endpoints can be quicker and easier than waiting to track actual heart attacks or survival rates, and it can hasten the availability of new treatments. But researchers have found that surrogate endpoints are not always good predictors of the desired outcomes.

“Most trial-level meta-analyses in oncology found low correlation between a surrogate end point and overall survival,” researcher Vinay Prasad of the National Cancer Institute and his co-authors wrote in a study published last year.

“Our findings call into question the widespread use of surrogate end points in oncology as a basis for treatment decisions,” they wrote, warning that the practice “has led to use of toxic drugs that do not improve survival.”

Approving drugs on the basis of “weak surrogates” can set back the cause of helping cancer patients because it “may encourage the pharmaceutical industry to pursue even more cancer drugs with marginal to no benefits,” they said.

In another major PDUFA VI commitment, the FDA said it “recognizes the potential value of utilizing ‘real-world’ evidence in evaluating not only the safety of medications but also their effectiveness.” The term “real-world evidence” was not defined, leaving possibilities open-ended. Real-world evidence evidently differs from the kind of data that has long been the bedrock of clinical research and FDA reviews. It could include, for instance, medical data gathered from millions of health insurance claims. In a similar vein, the FDA said it will beef up its ability to consider “patient-reported outcomes,” and it said it will foster the development of “tools to collect meaningful patient input that can be incorporated into regulatory review.” Those weren’t described in any detail, either.

Kay Holcombe of BIO, the biotech industry group, provided this perspective at the August FDA forum:

“We cannot put anecdotes on the drug label. So this is all a way for us to convert what always has been an anecdote into real data that can be on the label and help people use this therapy in the way that is the safest and most effective for each individual patient.”

At the same event, Peter Pitts, president of Center for Medicine and the Public Interest and a former FDA associate commissioner, offered a cautionary word.

“When it comes to the patient voice, or any voice, the plural of anecdote isn't data,” he said.

“To evaluate the reliability of data, FDA must assess how they were collected, their adequacy for answering relevant questions, and whether they were collected in a manner that minimizes bias,” he said.

In a third commitment, the FDA said it would take steps to “facilitate the advancement and use of complex adaptive, Bayesian, and other novel clinical trial designs.” Those terms weren’t defined, either. However, to focus on the first, the term “adaptive” has been used to describe clinical trials whose procedures are changed while the trial is in progress, as information is gathered. At their best, such approaches can reduce wasted effort and increase efficiency. At their worst, they can open the door to bias and manipulation.

The FDA told POGO that the use of highly innovative trial designs should not increase the risk of bias or manipulation. In a written response to questions, the FDA gave as its reason something not promised in the PDUFA VI commitment letter: It said adaptations would be specified before trials begin.

At the August FDA forum, BIO’s Holcombe, an invited panelist, explained that “innovative trial design” was a prime topic for BIO member companies.

“I want to say a few words about innovative trial design, not because I understand what it means,” Holcombe said.

The PDUFA VI agreement calls for a “pilot program” for “highly innovative trial designs” for which “simulations are necessary.”

“It's going to allow companies to bring in creative ways of doing trials,” Holcombe said. In an apparent nod to an FDA official at the meeting named Lisa, Holcombe added, “By ‘creative’—Lisa is getting nervous about the word ‘creative’—I mean like statistically okay ones.”

Paul Brown of National Center for Health Research saw a downside. The “faster, less thorough reviews” contemplated under the agreement “will cost patients and taxpayers billions of dollars, but many will later be found to offer risks that far outweigh the benefits,” he said.

Lowering the Bar for Drug Approval

Drug testing
(Photo: Food and Drug Administration / Flickr)

The FDA Commissioner’s September decision to approve the muscular dystrophy drug eteplirsen over the objections of FDA staff illustrates, for better or worse, where some of these commitments could lead. It also sheds light on the agency’s posture toward industry in the age of user fees.

The FDA’s review of the drug included extensive input from patients. At the direction of Janet Woodcock, head of the FDA’s Center for Drug Evaluation and Research, members of the FDA team reviewing eteplirsen joined her for meetings with patient advocacy groups “anywhere from six to twelve times,” according to an internal FDA account. The meetings frequently included boys with Duchenne muscular dystrophy and their parents. One member of the review team described the sessions as “’intense,’ ‘personal,’ and ‘intimidating,’” the internal account said.

When an FDA advisory committee held a hearing on the drug in April 2016, one of the witnesses was Pat Furlong, who lost two sons to Duchenne muscular dystrophy and founded the advocacy group Parent Project Muscular Dystrophy. Furlong recounted that, before her son Patrick died at age 15, “he couldn’t lift his hand to his mouth.”

“Your goal, the FDA, is to improve how an individual feels, functions, and survives,” she said, urging the committee to “exercise maximal flexibility” in assessing the drug.

In pursuit of a cure, Parent Project Muscular Dystrophy has joined forces with drug makers. Its website lists several pharmaceutical companies as sponsors—including Sarepta Therapeutics, maker and sponsor of eteplirsen.

At the advisory committee hearing, speakers from the public were asked “to advise the committee of any financial relationship that you have with the sponsor, its product, and if known, its direct competitors.”

When it was Furlong’s turn to speak, she began: “My name is Pat Furlong, and I am president and CEO of Parent Project Muscular Dystrophy. I have nothing financial to disclose.”

Furlong’s organization did not respond to inquiries for this story.

The scientific research that drug-maker Sarepta submitted to the FDA did not show that eteplirsen cured the disease. Rather, the case for eteplirsen hinged on an unproven surrogate endpoint: production of dystrophin, a protein that is lacking in muscular dystrophy patients. One FDA official said the increase in test subjects’ dystrophin levels was “trivial.” Another described it as “miniscule.”

Unger, director of an FDA drug evaluation office, said the case for approval was weak.

“The approval of this NDA”—new drug application—“in its present form would have far reaching negative consequences for the public health,” he wrote in a July memo.

Unger was overruled by Woodcock. He appealed to an internal FDA board responsible for resolving scientific disputes.

In a presentation to the board, Woodcock talked about the drug maker’s financial condition and stock price. “Dr. Woodcock cautioned that, if Sarepta did not receive accelerated approval for eteplirsen, it would have insufficient funding to continue to study eteplirsen and the other similar drugs in its pipeline,” Luciana Borio, the FDA’s acting chief scientist, wrote in an account of the appeal.

Woodcock also “emphasized her view that the agency needs to accept more uncertainty when granting accelerated approval,” Borio wrote.

Borio sided with Unger. “[A]pproving products based on hope, on subjective clinical judgment, or on theoretical constructs that are not anchored in data leads to irreparable damage to patients,” she wrote.

The decision fell to the FDA Commissioner.

“Major flaws in both the design and conduct of the clinical trials using eteplirsen have made it impossible to use much of the resulting trial data as reliable evidence,” Califf wrote. He also wrote that he was “troubled” to read that Woodcock’s decision “may have been inappropriately motivated by concerns over” the drug maker’s “financial well-being.”

But Califf said he spoke to Woodcock and was “satisfied that her decision is indeed based on her scientific evaluation of the evidence.”

Saying he would defer to Woodcock, he approved the drug.

“Patients with a particular type of Duchenne muscular dystrophy will now have access to an approved treatment for this rare and devastating disease,” Woodcock said in an FDA news release.

Charles R. Babcock, Halle Zander, and Lydia Dennett contributed to this report. Editing by Danni Downing.