This is part two of Johanna Ryan’s posts on Ghosts in the Clinical Trial Machine or Clinical Trial Fraud. Last weeks post reported on how doctors in three states have been convicted of handing in fake clinical-trial results from fake patients. This week we’ll examine how small-time crooks like these ended up doing world-class medical research for companies like GSK.
When they hear about Medical Research, most people picture gleaming, high-tech labs where new drugs are developed, and prestige university hospitals where expert doctors test them on patients. They don’t picture doctors like Sami Anwar but maybe they should.
Take the VESTRI Study of GlaxoSmithKline’s Advair for asthma. It was published in the New England Journal of Medicine, the dean of the peer-reviewed journals, with multiple authors. Drs. Szefler and Liu were from the University of Colorado-Aurora. Dr. Zeiger was from Kaiser Permanente in San Diego, and Dr. Pederson from the Lillebault Center in Denmark. The lead author, Dr. Stempel, and others worked for GlaxoSmithKline in Durham, NC.
Did they do the actual research? It was carried out at 532 locations in 32 countries. Denmark was not on that list, although Bulgaria, South Korea and Peru were. Just over half the sites were in the USA, from Wenatchee, Washington to Scottsboro, Alabama and from Fargo, North Dakota to Gun Barrel City, Texas. But only one was in Aurora, Colorado and none were in San Diego or Durham.
In other words, while the study’s authority relied on its academic authors, it’s unlikely any of them actually gave an inhaler to a patient and watched what happened. They simply signed off on the results of research done by 530 or so local doctors: some skilled, some unskilled; some extremely ethical, some not exactly.
These local doctors included Dr. Bencosme, Dr. Panuto and lots of others we know far less about.
Why do they do it? For money, mostly. While American doctors make good money, some do better than others. Specialists out-earn primary care physicians (GPs), while surgeons do better still. Commercial insurance pays more than Medicare; Medicare pays more than Medicaid; and uninsured patients might end up paying you nothing at all. For doctors taking care of poor people in primary-care clinics, the income gap is huge, and hard to ignore. A side-line in clinical research can be an attractive way to narrow that gap.
Most are like the defendants: treating physicians who do clinical trials on the side. A large, diverse patient pool with plenty of chronic ailments like diabetes or asthma can let them run a “research center” out of their existing clinic without much extra investment. Some doctors do take the opposite tack, treating patients as a side-line to their primary research business. They may recruit by advertising, or build networks of local doctors who refer their patients for studies.
In either case, clinical trials increasingly rely on U.S. patients who need free medical care. Some are completely uninsured; others can’t afford their insurance copays, or can’t find a specialist who takes Medicaid. The research clinic may be their only source of care. Others find their treating doctor increasingly intent on signing them up for a study. It’s hard to even talk about “informed consent” for patients in jams like this.
So how do multi-national companies like GSK find local clinics like Zain Medical Center or Healing Touch C&C to do clinical trials? Most don’t do so directly. Instead, they rely on Contract Research Organizations (CROs): large commercial brokers that recruit and manage the hundreds of local sites and doctors in a “gig economy” of medical research.
The VESTRI trial was run by a mega-CRO called Parexel, which signed up two of our doctor-defendants: Dr. Bencosme of Unlimited, and Dr. Panuto of CRS.
Medpace, another CRO, managed the trials of Braeburn’s buprenorphine opioid implant, enlisting both Tellus and Mid-Columbia Research.
IQVIA signed up Unlimited and Tellus for that CARMELINA diabetes trial.
The doctors are independent contractors in these arrangements, much like a driver who transports passengers for Uber one day and pizza for DoorDash the next. If the pizza arrives cold or the ride is downright dangerous, both Uber and the pizza parlor will tell you they’re not to blame. The driver doesn’t work for them!
Likewise, when Dr. Bencosme was arrested, the system allowed GSK to position themselves as victims not suspects. “As soon as we became aware of possible clinical trial fraud we conducted an internal investigation; excluded the data from the studies and reports, and reported the potential fraud to the FDA,” they told reporters. Parexel, for its part, sniffed that it had “no affiliation” with Unlimited Medical Research.
The bond between drug companies and the CRO’s that serve them, however, is a much more strategic partnership. Firms like Parexel and Iqvia offer help with everything from designing the studies to writing up the research papers and pitching the newly-approved drug to doctors. As Iqvia asks its clients, “What can we do for you today?”
King & Spalding’s bulletin didn’t mention Parexel; it cast Unlimited itself as a “CRO.” It didn’t name any doctors either. In fact, the only name mentioned was Ms. Raventos, the clinical coordinator. It did advise its clients to make sure their researchers and CRO’s had policies and procedures in place to monitor for problems. That way, when trouble did surface, GSK and others could paint themselves as victims not suspects. Their own hands would be clean.
Some of Sami Anwar’s patients suffered serious harm in these trials. The “ghost patients” at other clinics were in some sense lucky that they didn’t actually take part. Millions more of us, however, have our medical treatment determined by the results of trials that can’t be trusted. Ultimately, we all get hurt.
A lot was at stake in the VESTRI trial. One of the two drugs in GSK’s inhaler could trigger worse asthma attacks in some patients – including fatal ones. The resulting research paper helped convince FDA that Advair’s second ingredient made it just safe enough to remove its Black Box Warning. The move and the paper were controversial – but the critics had no way of knowing that some of the patients in VESTRI were fake.
Testing an asthma remedy whose worst side effect is asthma attacks is obviously tricky. (A sane health-care system might take its asthma research in a different direction.) A randomized controlled trial could get this issue wrong, even if carried out by 100% honest scientists. But the current system of Ghost Research makes errors almost inevitable.
The same is true for studies of opioid addiction, liver disease, mania and others that involved any largely bogus “research centers.”
Most of the doctors who make this system run are not fraudsters like the defendants. Nonetheless, the system itself creates pressures for subtler misconduct. Like giving your patient the diagnosis that will qualify him for the study, even if it doesn’t quite fit, or massaging incomplete data to match the sponsor’s requirements.
Even patients, desperate to qualify for free medical care, can play along. If this Uber-style research system has trouble smoking out obvious crooks like Sami Anwar, how will it monitor the thousands of other (mostly) honest gig researchers?
A final note: The hottest new trend in drug research, it seems, is the “decentralized trial” in which subjects may never see the doctor face-to-face. They can be recruited online, get meds delivered to their homes and report their results via Zoom calls, online surveys or wearable sensors.
In the wake of the Covid-19 crisis, CRO’s and trial centers have jumped on this bandwagon in record numbers. Generating fake results (not to mention fake subjects) could become as easy as adding 5,000 fake “followers” to your Twitter account.
We may need a bunch of new words to cover these new realities particularly if the legal system cannot keep up with the pace of change and hold people responsible – the idea of responsibility is starting to feel a touch twentieth century.
This clinical trial fraud has been going on for decades. In 1996 Richard Borison and Bruce Diamond were found to be inventing fake patients and ended up in jail. These university and highly regarded researchers had a clinical trials company in Georgia and participated in trials for many of the antidepressants and antipsychotics and related drugs that readers of these posts may have had. All of us certainly know people who have had at least one of the drugs these guys were linked to.
The licensing of Prozac looks like it depended on a dodgy clinical trial in Texas – see Let Them Eat Prozac.
Things have gotten worse – not better – both within the United States and by virtue of these trials being extended to corners of the earth where folk have never heard of FDA and look like they might be about to get worse again if trials become virtual.
It says a lot, given some of the trials of antidepressants for children have taken place in Colombia, the Russian Federation, Mexico, along with US foster homes, that companies still haven’t been able to show these drugs are safe or effective, despite claiming they have.
There are huge overlaps between Johanna’s two posts on Clinical Trial Ghosts and Fraud and current posts on the Politics of Care Forum.