By Mark D. Uehling
June 14, 2005 | Symlin was stalled. Almost two decades in development, it was a replica of a natural hormone that slows the passage of food through the stomach and curbs the urge to eat after meals. But several clinical trials sponsored by Symlin’s developer, San Diego’s Amylin Pharmaceuticals, had shown the injections might be responsible for hypoglycemia. FDA reviewers were impressed but hesitant.
So Amylin launched one last study. In this trial, only diabetic patients (and their Palm hand-helds, all called Ed, for “electronic diary”) would be gathering data. The role of doctors and nurses was to monitor secured Web-based reports on the patients’ conditions. That study, with just 300 patients, did the trick. The FDA approved Symlin in March 2005, saying it was the first new drug for type I diabetes since 1921, when Frederick Banting and Charles Best isolated canine insulin at the University of Toronto.
“Symlin provides a new option for many patients who, despite their best efforts with insulin therapy, continue to struggle to achieve their glucose control targets,” says Robert Ratner of Georgetown University. “The science behind Symlin has improved our understanding of the physiology of diabetes and has provided a welcome new tool for insulin users.”
Patient diaries are increasingly common throughout the industry. In the only independent survey of technology usage throughout pharma, the Clinical Data Interchange Standard Consortium (CDISC) received replies from almost 400 managers. CDISC volunteer David Iberson-Hurst, a consultant and clinical software developer, tabulated the results of the 2004 poll, which has details on the technologies used in 1,552 clinical trials run by sponsors.
Roughly a quarter of all sponsor studies, according to the CDISC survey, use a diary of some kind. In general, 20 percent have a paper-based diary; 5 percent have an electronic component. The survey did not delineate which electronic diary studies used the plain-old telephone or interactive voice response (IVR), the cheapest way to get data directly from patients. From 2003 to 2004, there was a 44-percent increase in the number of electronic patient-reported outcome technologies. The bad news is that the absolute number of such trials in 2004 was just 107.
“E-diaries are gaining acceptance, no doubt about it,” Iberson-Hurst says. “But there are still issues, such as logistics, the expense compared to paper, and the support and training given to sites and subjects.” He notes that regulators have made general statements from the podiums of conferences supporting diaries. The absence of an official written guidance or regulation covering e-diaries, however, is just one element in a larger discussion on “source” documents in clinical trials that are no longer written on dead trees.
So while Iberson-Hurst is a fan of the technology, and of the scientists in the vendor community, he also believes it could be good to have a written document in which regulators, sponsors, and technology vendors could agree on how the technology and the process should be united. “It’s not as though there aren’t regulations,” he says. “There are. It’s how we apply them in the e-world. That’s the thing. I’m a great believer in the technology. It is good stuff. It is the way to go.”
In the end, Iberson-Hurst says, written guidance from regulators could protect all interested parties from a hypothetical, less-than-honorable vendor that skirted the rules. “The technology can be compliant, it can be noncompliant,” he says. “The technology and the process — that is the crucial factor. Technology itself you can use or abuse.” He speculates that a written clarification of the FDA position on e-diaries and e-source might appear as soon as the end of 2005 — or, if the discussion becomes contentious, never.
The industry is not waiting for Washington. One Boston electronic diary company alone, PHT, sold 20,000 Palm hand-helds for clinical trials in 2004. The entire industry is migrating away from paper diaries, fraught as they are with patients’ well-meaning but well-documented procrastinations and fabrications. Electronic gadgets, priced from $500 to $1,500, are not inexpensive, even for rich sponsors of clinical trials. But the quality of diary data, cleanly and quickly delivered, is letting the largest companies spend money on hand-helds without the pangs of doubt that cling to deployments of other major technology programs such as electronic data capture (EDC).
To be fair, the words “diary,” “Palm,” and “patient-reported” are nowhere to be found in the Smylin label approved by the FDA. But make no mistake: It was data from a 29-week trial chronicling patients’ ordinary, humdrum lives that finally won the confidence of the FDA reviewers. Patients took their own glucose measurements before and after each meal. Some patients reported that the device was “beeping all the time,” according to a conference presentation by Amylin’s Gary Bloomgren. Within 24 hours, clinical personnel at research sites could see if problems were arising in individual cases. And hypoglycemia was an issue for some patients. But on balance, the risk was sufficiently contained to get the drug to market.
For Amylin, there were several efficiencies. No other data needed to be recorded, largely because 4,500 patients in previous trials had already been studied. Patient compliance with the study regimen was 74 percent — and that figure represents data that (unlike paper diaries) were impossible to forward- or backdate or otherwise misplace in time. But Bloomgren estimated that compliance with a three-times-a-day regimen might be 20 percent with a paper diary. The electronic diary, crucially, provided a real-time monitoring opportunity for safety and compliance.
“You are able to give the researchers a better tool, a sharper scalpel, for doing their work. The doctors and the patients commented on how it tightened their personal relationships,” says Doug Engfer, president and CEO of invivodata. “The patients felt they were getting better care.”
Engfer says his company will not be branching out to other parts of the clinical landscape. “Our focus is and will remain on the patient — on delivering the data from the patient. It’s too important a problem to dilute it.” He notes that a major EDC vendor, with which he now partners, had attempted to do diaries but eventually abandoned the effort. invivodata has always prided itself on especially close attention to its relations with the FDA, which is not to say its competitors neglect that. The Pittsburgh company recently launched a consulting arm, PRO Consulting, including academic experts who have spent their careers working on the subtleties of user interface and device design in clinical trials.
Other technology suppliers are also reporting robust demand for diaries. The North Carolina-based etrials (specializing in EDC) says that every customer has been using a patient-reported modality since April 2004, when etrials acquired a telephone-based IVR company. The universal appeal of a patient-reported component is surprising, for etrials’ client roster includes smaller and midsized companies where money is presumably not growing on trees. Uniquely in the industry, etrials rolls up EDC, clinical trial management, clinical data management, and patient-reported solutions in one platform.
“The integration story is resonating,” etrials CEO John Cline says. He notes that just as Microsoft Office allows users to paste text from a spreadsheet, etrials easily unites a variety of data streams. “We want to add as many technologies as we can under one platform. For me, to be a single-technology vendor puts you at extreme risk. That makes you less attractive to the sponsor because you’re so dependent on one technology.”
At PHT, founded in the mid-1990s, there is plenty of focus not only on diaries but also on related peripherals to capture patients’ cardiac and pulmonary function through wireless links. And growth is hot. “My estimate is that this market is growing at a 40-to-50-percent clip,” says Phil Lee, PHT’s CEO. “We are seeing tremendous market momentum. We’re seeing the rate of adoption grow quite dramatically. I’ve been running PHT for two years, and the degree of enthusiasm from the market has grown remarkably.”
Part of the appeal is the gentle nudging of the industry by regulators. “The FDA and the EMEA have been extremely supportive,” Lee says. “They used to be somewhat discouraging. Since the electronic patient-reported outcome conference [in October 2003], they’ve been just wonderful to work with. They’ve been supportive not just of our technology, but electronic approaches in general.”
PHT sold 50 trials in 2004, he says, and will do significantly better this year. In part, PHT echoes its competitors in bragging that higher-quality data mean that fewer patients need to be recruited. The patient shortage is such a perennial crisis for clinical trials that it seems a bit far-fetched to think that a diary is a panacea. In one case, Lee says, Merck (which is also an investor in PHT) needed 56 percent fewer patients. “The last 20 to 30 percent of the patients are the hardest to get,” Lee says. “That’s real return on investment. It has a huge impact on the market.”
And like his competitors, Lee is withering on the failings of paper patient diaries. “Paper diaries just don’t work. People just don’t fill them out on time. They basically just make it up.”
Having worked in the clinical technologies for a while, Lee has no difficulty analyzing why diaries are so warmly received. “The problem with EDC, in my opinion, is that they did not reengineer the process. At the end of the day, the paper case report form and the source document verification [are] very similar to the way it was in paper-based clinical data management.” Diaries, however, can save major amounts of labor and tedium for oppressed and overburdened clinical sites. “We’ve taken the sites out of the tactical data collection part of the system,” Lee says. “We just give the site access to the data.”
The Finnish Factor
With trials starting in dozens of countries simultaneously, it’s probably no surprise that a European company, CRF, is a major force. The Finnish concern says more than 92,000 patients at 2,000 sites in 55 countries have used its technology. In what CRF calls the largest diary trial ever, it sent devices to 5,000 patients in 23 countries, locking the data five days after the last patient’s last visit — an unattainable feat in the paper realm.
CRF also claims the industry’s highest patient compliance rates — an average of 95-percent compliance in phase I to IV trials. The company believes its diary setup and trial design application, TrialMax, is superior to anything in the industry. “TrialMax is the world’s most configurable platform to do mobile healthcare applications,” says Timo Ahopelto, CRF’s worldwide vice president of operations. “We never have to do a single line of manual programming. You can configure the whole study with that tool, including workflow and edit checks.”
He notes that other tools, from Microsoft, Nokia, AvantGo, and other firms, lack CRF’s WYSIWYG interface, or multilanguage libraries. So-called “branching,” guiding patients to one screen or another depending on their answers, is easy to drag and drop into place.
‘Comfortable with Technology’
With a headquarters in the Boston area, CRF’s revenues are split evenly between the United States and Europe, the company reports, and may grow at a 125-percent clip in 2005. “The pharmaceutical industry has learned how to deploy e-diaries,” says Ahopelto, who started the company without knowing a soul in the industry. “They have very realistic assumptions about the technology and a good view of how the technology helps them. They are comfortable with the technology.”
His system can handle up to 70,000 patients without significant performance penalties. CRF is already experimenting with clinical trial participants using their own cellular phones. Ahopelto notes that one study in the United States, Finland, Sweden, Germany, Taiwan, and the Philippines will use the personal cell phones of 18,000 patients who have received a vaccine. After agreeing to participate in studies of the safety of the vaccine, those patients get text messages inquiring about their health.
The savings are considerable. A follow-up telephone call costs about $30; a short message service (SMS) message costs a penny. For clinical technologists, Ahopelto says, “The mobile phone is a whole new universe. You could recruit a vast number of patients. You don’t have training, you don’t have hardware concerns.”
With CRF’s open architecture, Ahopelto notes, integration of diary data with other clinical data streams is a piece of cake because the metadata can be easily mapped and exported. “That’s pretty important when you talk about EDC integration,” he says, noting several projects with a few unnamed vendors in that space. The technology is also easy for the oldest, most technophobic patients to get the hang of, as Ahopelto recounts in one study of elderly, alcoholic cancer patients.
Other companies are taking patient-reported data to another level. VivoMetrics makes a garment that gathers pulmonary and cardiac information in real time. The LifeShirt weighs just half a pound and is used by rescue workers and military personnel. Children have worn the shirts, as has a 450-pound individual.
When working with conditions such as sleep apnea, overbooked hospital labs can be augmented by a device worn at home. Surprisingly, almost everyone can use a LifeShirt without training. “Being able to collect the data in the person’s own bed is a much more realistic way of looking at what’s going on,” says Alexander Derchak, VivoMetrics’ principal scientist and director of clinical development. “How real is the data you get in a sleep lab when you don’t sleep there?”
Data from Your Chest
In one study sponsored by GlaxoSmithKline, the shirts measured coughing with 99-percent accuracy; the gold standard in understanding coughing using other approaches is a video of the patient. But patients who cough are notoriously poor at recording the condition — they tune out the cough after a few minutes. “We had very, very few false positives,” Derchak says. “If we said it was a cough, you can be really sure it was a cough.”
In general, Derchak says, it can be arbitrary for human observers to agree on what constitutes a cough. The shirt makes that determination automatically. “This gives more relevance, and we get data more continuously. We get hundreds of data points.”
The shirt comes with software that allows scientists to pinpoint the intervals where the variability in the data changes dramatically. In some cases, as with the hand-helds, the technology is about to blur the line between research and ordinary medical care. “We can see the day when we may be alerted about a child who has asthma — and have a parent or a physician intervene,” Derchak says.