Info & Opinion
July 21, 2019
With news about big data, Aetna, Covance, GNS, Scrip, Quintiles, PPD, Icon, BioClinica, Merge, Medidata and GSK.
With news about FDA, CTMS, PMG Research, Inclinix, EMA, Hemofarm, Parexel and the Korea Drug Development Fund
David Underwood of Quanticate says some firms are giving short shrift to the basics of clinical trials
With news about Roche, Quintiles, Allscripts, Janssen, SGS, Oracle, TriReme, OpenClinica and FDA
Last week’s 2012 Post-Approval Summit had one theme in common with the events of prior years. It’s the flip side to the rising importance of patient registries and observational Phase IV studies. Like it or not, there is diminishing faith in randomized clinical trials (RCT) on the part of the industry and regulatory agencies.
So what greets the majority of untold tens of billions of dollars in industry-sponsored R&D is basically a ... shrug. A sigh.
At least at the Harvard-located event last week, there was a triumphant trumpet blast to meet the small slice of science that happens after drugs are on the market. All of the speakers at the 2012 Summit were in agreement. They concurred that industry will be reliant on some combination of traditional RCT, registries and healthcare databases.
Quintiles supplied a textbook example of the value of a good database. Quintiles acquired the conference's original organizer, Outcome, in late 2011, and was the only contract research organization (CRO) with speakers on the agenda this year.
Back in 2007, readers will recall, the big North Carolina firm developed a website to enable patients to share information with each other. Lo and behold, that effort has morphed into a clinical trial recruitment service, as Elisa Cascade explained. She is the VP of global operations for the Quintiles division called MediGuard. (ClinPage last covered the organization in 2008, when it was called iGuard). MediGuard is transparent about its funding; none of its users are likely to be shocked to learn that industry supports the site.
The Quintiles/MediGuard operation now has 2.6 million members, all of whom are required to supply a valid email address to identify themselves. Participants also send in paper consent forms signed in the traditional Dickensian manner.
With MediGuard deeply versed in the details of research, it has streamlined the identification of patients that meet any inclusion and exclusion criteria. The firm is fine-tuning the process of getting information from patients online, not to mention giving them financial payments in return for their participation. “We have found that incentivizing does produce a better response rate,” said Cascade. She observed that institutional review boards (IRB) were initially surprised by the company’s digital approach to patient recruitment, but have since come to believe that it is as ethically appropriate as any traditional strategy.
The new method is cheaper. In one case, Cascade presented data that in a traditional approach would have cost $1.7 million, versus a MediGuard-enabled plan costing only $800,000. The firm is exploring international expansion. It will be interesting to see if any other CROs attempt to replicate MediGuard and how energetically they do so.
MediGuard has begun efforts to verify that patients truly have the conditions they claim, and are not idle teenagers or middle-aged pranksters. One project to compare traditional hospital charts and self-reported medical facts on the Quintiles site showed extremely high levels of accurate self-reporting by patients.
Cascade related what can only be termed astonishing turnaround times to conduct feasibility assessments and post-approval studies. She made finding 1,000 patients with rheumatoid arthritis sound as easy as Googling the disease.
In another case, for a trial of 500 patients with chronic obstructive pulmonary disorder (COPD), the staff were amazed to find their first patient online in six minutes. Some reports are returned by MediGuard, wholly completed, in a week. Other more complex projects may finish in half the time of a conventional registry. The reason? MediGuard appears to have an unusual number of engaged, cooperative subpopulations of subjects. “Patients are out there,” Cascade said. “They are looking for data online.”
No one can say how quickly Quintiles will exhaust its available, willing subjects in any particular therapeutic area. But the company appears to be able to offer a more unified view of the patient’s experience during research. The company is able to use email and SMS text messages for more immediate, intimate check-ins to the patient than some trials can provide. “You get this clear line of sight throughout the study,” says Cascade.
Michael Rosenblatt is Merck’s executive VP and chief medical officer. His career includes time at Tufts and Harvard in several lofty administrative roles. Rosenblatt left no ambiguity about the emphasis his employer is placing on both registry and database-driven research.
“RCT were in many senses idealized. If we look at the direction of post-approval studies, they are really beginning to transform healthcare,” he said. “Rapid answers can be obtained.”
He also sounded a clarion call to peers who may still not be allocating enough of their R&D budgets to post-approval research. In Merck’s case, post-approval ideas are filtering all the way back to the discovery phase of research and being anticipated in RCT design in pre-approval studies.
“We see a clear increasing importance of the use of observational data and comparative effectiveness research (CER),” Rosenblatt said. “We are now baking it into every one of our drug candidate and vaccine development programs. We are preparing for the types of questions we might receive from health authorities around the world.”
The governance and geographical constraints in a multinational corporation, he implied, have not yet enabled all of Merck's top brains in observational and CER research to be assembled into a unified team, as they might be in an academic department. But the big firm will at least try to align such people informally so that they are working in concert.
As insurance payors use more observational research, Rosenblatt predicted, they will be drawing less on the artificial, constrained, manicured data from RCT—and more on typical, average patients with multiple diseases taking multiple medications.
That could mean a more stark, unforgiving formulary gauntlet for some drug and device makers that cannot prove their products work against competitors outside the simplified boundaries of a traditional RCT with a placebo. Said Rosenblatt: “A drug will be either in or out. Some decisions will be binary.”
A brave new world of additional non-industry databases, he said, does pose dangers. “My worry is that clinical research can be done by one person with a laptop,” Rosenblatt said. “That person can rummage around for a signal. There is no control group.”
Once a specious signal is discovered, it may be etched online before any rigorous statistical assessment can occur. “It can be viral,” he added, referencing public confusion about vaccines and the un-erasable nature of online medical myths. “That's one of the reasons we need to be very careful in how we report this data. The media coverage is not exactly on the side of strict science.”
In a twist, some big hospital groups and data repositories are borrowing a page from industry’s playbook. They’re keeping some data off-limits for a variety of reasons. To be sure, the availability of some national-level drug safety data repositories is still possible.
But reading between the lines, Merck seems to be exploring ways to handle circumstances when it is locked out of access to a particular dataset—and what sorts of alternatives might need to be bartered into existence or constructed from scratch. “If those databases are not available,” Rosenblatt went on to say, “we will have to have access to our own databases that are on the same scale. We cannot be in the position of being completely reliant on others.”