By Maxine Bookbinder 

 

April 11, 2014 | When Jeri Burtchell entered a 2007 trial testing a first-ever MS pill, she feared the unknown, didn’t understand the medical jargon, and couldn’t tell if the informed consent was written in English or Klingon. She did what any well-intentioned, apprehensive patient would do: she searched the Internet.   

 

Burtchell consulted online friends with medical backgrounds who translated the technical verbiage into English, but still couldn’t find any clinical veterans.  So, she blogged her experiences to share the patient perspective of clinical trials. 

 

Then a strange thing happened. “For me, it was just a place to journal,” says Burtchell. “But, people started to find me.  MS patients eager for a pill were following along.” (The other therapies were injectables.) Some joined the trial. Online, they shared illness symptoms, family issues, and daily life. “We were looking for others to understand what we were going through. A trial is an isolating experience. We formed a bond.”

 

Burtchell also made Internet history as its first patient to blog a clinical trial from start to finish.     

 

Craig Lipset, head of Clinical Innovation at Pfizer, read her blogs and took note.  Even today, “there is not much organized research on social conversation during clinical trials.” He says the medical community not only needs data, but an “informed voice at the table on health care decisions. This includes patients, caregivers, and others to provide information. It is insightful to understand the patient’s perceptions and feedback on investigators and study protocols.”  

 

Wanting to connect to others with shared experiences is natural. Twitter and Facebook make it easy; smartphones make it immediate. Asking trial participants to not talk online is the equivalent of expecting jury members to not discuss witnesses while at lunch break. Social media provides a quick and efficient recruiting tool and allows immediate communication, comprehension, and feedback. In their excitement of finding others online with common concerns, fears, and illnesses, patients may unintentionally, and perhaps carelessly, skew scientific integrity of blinded, randomized trials by sharing symptoms, comparing medications, uploading photos and videos, and coaching prospective participants on the “right” answers to win a spot in a trial. 

 

Change must be scientific and innovation must be carefully planned—and begin now.  

 

“The Internet has outpaced our ability to keep up,” says Paul Wicks, Ph.D., vice-president of innovations at PatientsLikeMe, a health-data sharing site for patients worldwide.  “Social media is a venue for patients to give emotional and tactical support to each other. Patients who are isolated and depressed might well stay sicker longer,” says Dr. Wicks. 

 

The medical community must accept the reality of social media, develop patient and researcher guidelines to prevent inadvertent un-blinding, and help patients understand where the proverbial line is and what is at stake.  For every study, millions of dollars and years of work depend on successful trial outcomes. “But, as patients,” says Burtchell, “we of all people want this to succeed.  We stand the most to lose or gain.” 

 

Off Limits 

 

Patients, caregivers, friends, and strangers can log in to social media sites, talk, and share. Researchers and regulators, however, cannot; if an investigator reads something, then he must report it. 

 

For example, a patient who identifies himself as Hockeystud36 complains of a monstrous headache following his last trial dosage. Is this patient in the current study? Or is this report from a previous trial? Is the trial randomized, and did the patient take the study medication or placebo? Did he take it as prescribed? Even though the investigator does not know who Hockeystud36 is, he still has an obligation to report the headache. But, what if the headache was already reported? Was it, in fact, due to the medication? Every infinitesimal safety event must be reported, despite the ability to verify its authenticity.  

 

Craig Lipset, Pfizer

In a trial of, say, 200 participants, one or two safety reports, verified or not, can skew results. After Hockeystud36 reports his headache, Honeybun_01 and three others may respond that they, too, had headaches. The research sponsors don’t know the patients’ identities, if they’re all in the same trial, or whether their headaches were real or suggestive, psychosomatic responses. 

 

Anecdotes abound of patients comparing symptoms or cutting meds to determine who has what drug. “They can go on Twitter, identify themselves as a trial participant and say, ‘Hey, that’s funny, since starting this trial, my pee is green,’” notes Dr. Wicks. “Someone else in the study might then think, “Gee, mine is normal, so I must have the placebo.”  

 

A person can search online for trials testing new medication for Alzheimer’s and, through online sharing, contact another caregiver who already has a loved one in a trial. “A caregiver whose parent has Alzheimer’s might coach Mom or Dad how to answer… the questions in a certain way,” says Lipset.  It’s hard to blame someone for doing this, he says. “If I had a parent with Alzheimer’s, I’d want him in a study, too.” 

 

Burtchell, now a patient advocate, MS blogger, and freelance writer, recalls an online trial patient on her MS blog who tried to figure out who was assigned the study drug. “I nipped it in the bud,” says Burtchell. “I reminded her of the protocol and told her not to do that. I learned about the possibility of unblinding from my lead investigator. The lesson stuck.”  

 

According to Burtchell, most clinical trial patients seek support, not sabotage. “We have too much to lose. The last thing we want to do is compromise or unblind data and ruin the trial.  We are risking our lives hoping the study succeeds.” 

 

In 2007, patients with the incurable and rapidly progressive disease ALS initiated their own impromptu trial after the Proceedings of the National Academy of Sciences published an Italian study of 16 ALS patients taking lithium carbonate and 28 control patients, suggesting the widely-

prescribed mood stabilizer might dramatically slow ALS. Within six months, 149 patients who obtained lithium off-label worldwide attempted to recreate the study. “This was a motivated population,” says Dr. Wicks, “that was willing to take a risk.” PatientsLikeMe graphed algorithms, tracked data, monitored side effects, and identified 447 patients who had not taken lithium to act as historical controls. In six months, this patient-led study found the same conclusion as five worldwide controlled and randomized doctor-led trials over the next four years—that the lithium was ineffective. 

 

Paul Wicks, PatientsLikeMe

This may have been the first time patients created their own study; the website responded to them and published the results. “The tracking tools available on the Internet today dramatically increase the accuracy with which patients can track progress relative to mere anecdotes,” notes Dr. Wicks.     

 

Lipset says researchers must appeal to patients who want to contribute to medicine and help them understand how talking online can jeopardize that. Investigators must work with patients’ natural desire to talk, to share, and sometimes, to make mistakes. 

 

“For every one patient who posts, nine others repost, 90 others read the post.” 

 

Patients who are online before a trial begins will likely remain online throughout. While some will naturally want to share symptoms or side effects, others will self-regulate. In one discussion forum involving a trial hepatitis C, patients shared tips to determine who was assigned which drug; some said they would drop out if taking the placebo. Several patients replied that they did not want to unblind themselves, believing that doing so would harm the study—and the potential development of a new medicine. “This was very impressive,” says Lipset. “They wanted to contribute to the study and to science.” 

 

“The challenge in trying to monitor to social media,” says Lipset, “is that when you close the door on one conversation, another will open somewhere else.” There’s the risk that disgruntled patients will move from private room to public tweeting. 

 

The industry needs to develop tools, preferably with the FDA, to maintain clinical integrity and still allow patients to meet online. “The genie is out the bottle,” says Dr. Wicks. “Now we have to ask patients to help us collaborate to deal with it.” 

 

“The key, says Lipset, “is being transparent but not attributing everything to the drug by default.” 

 

Some possibilities include: 

• Establish private, monitored discussion boards in which patients can vent, offer support, and give feedback.  Assign user names and passwords so only trial participants can enter.  When Burtchell set up her blog, participants not only needed specific user names and passwords, but also had to show their trial med bottles as proof they were in that trial.  
• Replace immediate posts with pending posts, similar to blogs, rather than running live discussion boards, giving monitors a chance to delete inappropriate discussions.  
• Give specific conversation do’s and don’t’s, with examples, prior to beginning the trial.  
• Clearly delineate how breaches, even if inadvertent, can impact data and trials.  
• Insist that participants sign a user agreement prior to trial initiation. If they stray into questionable conversation, the moderator gives one warning. After that, they are banned from the discussion board.  Of course, they could then go on Facebook or underground.  “Although an idea like this (mandating signed user agreements) might sound appealing, it is short-sighted and bound to backfire,” says Dr. Wicks. “Is pharma really going to sue a participant volunteering to test a drug because she talked about her side-effects on Facebook?”   
• Establish private discussion posts in which patients can ask questions and give feedback that only the moderator sees. “Misinformation spreads like wildfire,” says Burtchell. “This way, we can go directly to the source with questions.”   
• Develop an easy method allowing patients to track their data, contact their study center, and make appointments. Burtchell suggests an app for that.  

 

If there is any doubt about the significance of camaraderie, consider this: after her year-long trial ended, Burtchell’s blog, which started with a handful of followers, moved to a private Facebook group. After the drug won FDA approval, the group opened to the public and continues today—with 2000 members. 

 

“The e-patient today is the e-participant of tomorrow. This is the right time to address the problem,” says Lipset. Regulators must start to think the same about clinical responsibility and R & D as they do about marketing.”