Bob Garfield: Over the past few months, as we've watched trial results, roll in about Pfizer and Moderna, Johnson and Johnson, and AstraZeneca, we've become accustomed to seeing those results as soon as they are computed.
Speaker 1: Pfizer, announcing early trial results in teens showing its vaccine it's 100% effective.
Speaker 2 : Stunning early results from this trial from US firm Moderna, have a clips the strong ones just a week ago from Pfizer.
Speaker 3: Johnson & Johnson has announced trial results for the first single-dose coronavirus vaccine with 66% efficacy, the American company Novavax said that its vaccine was 89% effective in UK trials.
Bob: This may seem like common sense. It is not, however, necessarily common practice. Pharma researchers are often negligent about timely reporting of trial outcomes. As recently as five years ago, one prominent politician scolded the industry over its transparency failure.
President Biden: There's a report in Stat News that although at age funding for cancer trials requires the funding. When you get the funding, requires the results of that data will be reported instantly. The number of institutions that received the most funding from NIH funding don't report back to NIH in a timely fashion, sometimes a year or more. NIH scientists themselves, 75% of the time are late or don't report their results. Doc, I'm going to find out whether it's true, and if it's true, I'm going to cut funding. That's a promise. That's a promise.
Bob: The Stat News report Biden referred to was the work of Charles Pillar, an investigative journalist for Science Magazine, who was calling on now President Joe Biden to make good his threat. Charles, welcome to the show.
Charles Pillar: Thanks very much, Bob. I'm glad to be here.
Bob: Let's start with what the law requires. There is a time limit for reporting clinical results. What is it and what statute mandates it?
Charles: This is governed by the food and drug amendments act. What it requires is that clinical trial results, and now these are the results of experiments using volunteer human subjects, be reported within one year of when the trial is completed. It requires reporting by both pharma companies and also by non-profits universities, anybody who's doing clinical research. These reports must go to something called clinicaltrials.gov, which is the federal repository for clinical experimentation.
Bob: Now, it's not just arbitrary big government red tape. Transparency matters. It can be a life-and-death issue. Can you tell me please about the experimental monoclonal antibody TGN1412?
Charles: This was a drug that was tested on six men in London in 2006. All of them immediately fell gravely ill and this was a terrible, terrible outcome, very unusual, even for an early experimental drug. One of the subjects suffered heart failure. Others lost his fingers and toes in a reaction that was very much like gangrene. The problem, aside from the horrific result, was that a very close analog to this drug had been found potentially dangerous in those ways years earlier, but the results had never been published or posted anywhere by the company that developed that drug.
The researchers in this case didn't have the warning they should have had to realize they were doing something that could be very hazardous. Evidence is the basis of evidence-based medicine. If you don't have it, you cannot make the best medical decision. Doctors rely on it. They rely on it to see how effective a drug or device performs, and to know the frequency of the side effects that might come up. It's also crucial for researchers who are trying to devise their own treatments or to understand how a particular drug might be applied to new conditions. Without that data, they're unable to do that effectively.
I should say, it's also a matter of honoring volunteers who sometimes are sacrificing or risking their lives or their health for the common good of testing a drug or a device to see how well it works on their particular condition.
Bob: You've listed a number of compelling reasons for deadlines, yet the chronic tardiness or, worse, suppression, persists. Back in 2015, you wrote that non-compliance rates at NIH were around 75% and Biden quoted that. I gather across the board, not much better today?
Charles: Well, things have gotten somewhat better. They're still pretty poor results. Let me give you a few examples. Some of the leaders in the academic world are still reporting the results of their trials either late or not at all more than half the time. These are institutions that are the forefront of medicine and yet for some reason they just can't seem to comply with the law, even though they've got plenty of money and plenty of staff.
Bob: I actually have a personal connection to this issue. In the late 90s, one of my young teenage kids was being treated for an anxiety disorder. She was prescribed a new drug called Paxil and she was on it for a while. Then in 2004, it came to light that in testing the drug, GlaxoSmithKline had suppressed damning clinical trial data linking Paxil to suicidal ideation among other things. That was a horrifying discovery for us, but it was also a turning point in drug trial reporting.
Charles: This antidepressant was found to do the opposite of what it was intended to do. First of all, it didn't work very well for depression. Second, it did cause teenagers to think about suicide. GlaxoSmithKline had aggressively marketed Paxil to doctors and parents for use by teens, and ultimately the company pleaded guilty to criminal charges associated with that drug and others and paid billions of dollars in penalties. Now it was a big case, in part because the results were so terrifying to so many people, but it was representative of so many of these circumstances where for years and years and years and years and years companies would cherry pick data.
They would withhold data and they would misrepresent data because there was no law that required them to show their work.
Bob: When you first started writing on this subject, you quoted Ben Goldacre, a doctor in the UK who had broken his leg while running to treat a cardiac arrest patient. He observed ruefully, nobody has ever broken a leg running down a carter to try to fix the problem of publication bias. He characterized this as a failure of empathy, but surely there are other grim realities in the overall ecosystem of pharma research that reward flouting the law?
Charles: Yes. What you've got are companies that would really like to have only good news about their drugs being publicized. Often if a trial shows something less than good results, they're really not eager to have that information in the news or in the scientific record. Now, I have to say that things have gotten better. Many large drug companies have been complying with the letter of the law, meaning that they're reporting their results according to the one year timeline. This is excellent. Reporting the results, of course, isn't necessarily enough. There are all kinds of ways to report yet not be fully disclosing what went on.
I think we all saw that disastrously in the recent AstraZeneca episode where they reported results on their COVID vaccine, but later an independent monitoring board for the data said that they had exaggerated the effectiveness of the vaccine. It's not just reporting, it's other forms of transparency, but reporting is the first critical step. It's just not going on as much as it should. Researchers, particularly in academic institutions, say, "Look, we're doing everything that's really necessary. We're publishing journals." There are several different problems though, with the logic of that position.
One is journals have paywalls and other payment schemes that make the accessibility to that data very low for many patients and doctors. Second, the journals don't publish everything, particularly negative results. What I mean by that is if you've got a drug you're testing and you have a hypothesis that it might have a beneficial effect on, say, skin cancer and then it turns out that it does not have any beneficial effect and skin cancer. Well, that result is critical for researchers, doctors, scientists as a whole, to understand in order to gauge treatments in the future and future research projects.
Journals are very, very reluctant to publish negative results. They have more papers than they can publish. Consequently, a vast number of clinical trials would never be published in journals. It's not a replacement for clinicaltrials.gov. It's merely another way of communicating, but it should be a supplement.
Bob: At the time of Joe Biden's threat, the director of the NIH, Dr. Francis Collins, said that NIH is serious about enforcing reporting compliance and that it would create a so-called wall of shame to finger violators. Is there a wall of shame? Who's on it?
Charles: The agency has yet to place a single name on that wall of shame despite 1,000s of violators to choose from, in general, NIH believes in tried to cooperate with trial sponsors. their grantees are their scientific and cultural colleagues, people with whom they socialize and that they admire and that they rely on. They're really not into being a tough cop, and I might add that this extends to FDA as well. Both of these agencies believe in quote-unquote, voluntary compliance. even though they threaten to actually enforce the law they never do.
The Food and Drug Administration has the ability to impose more than $10,000 a day in fines on companies it regulates if they don't comply with the law to report their results to clinicaltrials.gov. In the time since the rule was clarified in 2017, the FDA could have collected $18 billion in fines for blown deadlines, or cases where companies never reported a single result. How much have they collected so far? Zero.
Bob: Which is, I'm just doing the math in my head, less than $18 billion.
Charles: A bit less than 18 billion. Collins who has the ability, to turn a phrase I remember so vividly back when he was announcing this new rule before it was implemented in 2017, he said, "It's hard to herd cats, but you can move their food or take their food away." Since then, a lot of cats have gotten fat on NIH grants without submitting their trial results to clinicaltrials.gov.
Bob: You refer to the AstraZeneca debacle in which they came out with very timely reporting, but the reporting was proven to be wrong and mischaracterized the actual result and has been described as exaggerating them. There is the benefit of speed, but I guess speed isn't everything. What do you believe? In the absence of the cops on the beat being a little bit more scrupulous, what do you think the solution is for timely, relevant reporting?
Charles: Look, let's not dwell too much on compliance as the end game. What we're talking about is honoring the sacrifices of these trial volunteers, of supporting evidence-based medicine, of creating the evidence base that these institutions exist to create, in that they say they are completely dedicated to it, so let's see it. I can tell you from my own experience of covering this subject since 2015, that in the absence of real enforcement of the regulations, which has yet to occur, there's really only one thing that seems to work, which is public embarrassment. What we found at Stat when I was working there, and since in work that I've done for Science, and now for the New York Times, is that naming names gets people to do a better job.
Bob: What an opportunity this is for you. You mentioned that something between half and two-thirds of the grantees of cancer research from NIH have been non-compliant with the deadline. Who are the worst offenders?
Charles: I looked at Brigham and Women's Hospital, Massachusetts General Hospital, Dana Farber Cancer Center, the three largest Harvard affiliated institutions that do large numbers of clinical trials. In each case, less than half the time do they comply with the law, less than half the time do they submit their results as required within a year. Another one that I think is worth mentioning is MD Anderson Cancer Center. MD Anderson is a University of Texas affiliated institution that is one of the leading clinical trial institutions in the world in sheer numbers and in influence, and half the time they don't submit the results of their experiments on time or at all.
These are institutions that are accepting hundreds of millions of dollars from the federal government, sometimes billions, and they are conducting groundbreaking and vital research and yet about half the time they don't care enough about that to support the system that exists to make sure that people can use that result of their research in a timely fashion.
Bob: You mentioned the Harvard triad and MD Anderson, probably the two most prestigious centers of oncology research. The third very likely is NIH itself. How are NIH grantees doing at reporting under NIH's own requirements?
Charles: If you look at trials sponsored by NIH, and when I say sponsored by NIH, what I mean is that they're the party that's responsible for reporting the results, even if it's sometimes done by academic scientists and other institutions. They are violating the law, their own law, almost 60% of the time. In about six out of 10 trials, they're not reporting the results on time, or at all.
Bob: Now, dude, that there is an ender. Charles, thank you very much.
Charles: Bob, my pleasure. Thank you for having me.
Bob: Charles Pillar is an investigative journalist for Science Magazine.
New York Public Radio transcripts are created on a rush deadline, often by contractors. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of New York Public Radio’s programming is the audio record.