Learning From Clinical Mistakes: Craig Lipset, Head Of Clinical Innovation At Pfizer

By Ed Miseta, Chief Editor, Clinical Leader

Years ago when I started my career in banking, I was assigned to a branch office where I learned to open accounts and approve personal loans for customers. At that time the loan decision was still made by people rather than computers. Information was gathered and reviewed, a few calculations were made, and the loan request would be approved or declined. A year into the job realized I did not have a single borrower in default, which made me think I was doing a great job.   

At my annual review I was told otherwise. The branch manager informed me that every loan officer will have defaults, regardless of how well they perform their research. There will always be unforeseen events that can change a borrower’s situation overnight. But more importantly, if I had no defaults in my portfolio, then I was likely not only rejecting loans from bad risk customers, but also those who would likely have repaid the loan in full. “There are certain risks that are inherent in lending, and eliminating all of those risks also means eliminating many potential opportunities,” he said.    

That same philosophy holds true in drug development. Craig Lipset, head of clinical innovation for Pfizer, summed it up best with this comment: “If everything we do succeeds, then we probably weren’t trying hard enough, and we probably were not being very innovative.”

Patient Recruitment Can Be A Challenge

Craig Lipset, Head of Clinical Innovation, Pfizer

Speaking during the Clinical Dirty Laundry session at Disruptive Innovations 2014, Lipset notes that if you’re only going after things that work, there are likely a lot of opportunities you are probably missing out on. To make his point, Lipset references a study his company worked on a few years ago. The goal behind it was to enable patients to be identified online and participate entirely from home in a blind and randomized trial with an investigational drug. It forced the company to tackle quite a few challenges.

“Some of the biggest challenges were finding patients, centralizing an investigator, and getting the medicine into patient’s hands,” says Lipset. “At the end of the day, we had to stop the program because of enrollment. We simply did not get the patients for that particular protocol. Recruitment just did not go as we had planned.”

Many of the headlines that came out of this would lead you to believe the entire effort was a complete failure, and Lipset will not argue that the headlines were fair. But at the same time he tells the story of the DVD rental business Redbox.

“Redbox was originally a kiosk launched by McDonalds to get grocery items – designed as a way to drive traffic to the restaurant,” he states. “The vending machine contained milk, eggs, and other essentials a parent might need, including DVDs. “What the operators soon found out is that customers did not want milk, eggs, or any of the other food products contained in the machine. But they did get DVDs. At that point no one threw out the entire box and called it a failure. Instead they took what did work and ran with that.”

Is There A Lesson To Be Learned?

Eventually Redbox replaced Blockbuster and there is likely a Redbox within five miles of where you are right now. This was one of the first lessons learned by Pfizer. If something doesn’t work, you can take a modular approach and keep the things that are working while getting rid of the rest. “Recruitment didn’t work, but that doesn’t mean we didn’t learn some new things, like how to ship products to patients in their homes,” says Lipset. “Our recruitment for remote monitoring really required patients to do a lot on their own. We went directly to patients and when we did they responded and came to the website. We found them and they found us.” 

But then Pfizer asked them to put a lot of faith into the trial, but didn’t provide them with the same interaction with a clinician that the patient may have wanted or expected. Lipset believes many of those patients may have gone back to their health care provider and their own doctor had they been given the option, since each one was already under some existing treatment for the given indication.

That of course presents another problem. If each of those patients went back to their treating doctor to say they were considering taking part in a clinical study, the doctor may not have been supportive of it.   

“We should have invested more energy in giving the patient tools to better engage their treating physicians,” adds Lipset. “It has taught me the need to double down on efforts to try to find ways to gain referrals from those treating physicians for patients to take part in studies. That is probably one of the biggest barriers for us in trial recruitment.”