As the head of clinical innovation for Eli Lilly and Company, Jeff Kasher, Ph.D. likes to spend time thinking about different parts of the clinical business and how they can be improved. Speaking during the Dirty Laundry session at the 2014 Disruptive Innovations conference, he noted a recent experiment where he toyed with the idea of eliminating central labs from the clinical trial process.
Kasher’s experience in trials led him to believe the labs were expensive and an area where the company invested a significant amount of money. The process of going through a central lab can also be incredibly slow, thereby slowing the pace of the entire trial. “When a patient comes in for an initial visit, I oftentimes will need a lab value before I can proceed to the next step,” he says. “If I cannot get an immediate turnaround on the data, as a consequence I may have to schedule another visit with the patient. So in terms of the patient perspective, as well as the time and cost of central labs, I started to question how I could get rid of them.”
The experiment he tried was to place point of care devices into the investigator sites. He started out with clinical chemistry machines, which were tried in a couple of different phase 1 settings (one with a single site and one with multiple sites) to see how they would work. The hope was to later try the same experiment in a phase II or phase III trial.
|Jeff Kasher, Ph.D., head of clinical innovation, Eli Lilly and Company|
“From this experiment we learned several things,” says Kasher. “The main thing I learned was the constraints imposed by the economics of the situation. The manufacturers of these diagnostic point-of-care machines have a model in place whereby they rent them to you. The payment plan is based on rental fees and the specialized cartridges they put in every machine. Unfortunately, when you start doing the math, you realize you are running into many thousands of dollars to perform a relatively simple assay that can be done a lot more cost effectively via a central lab.”
Long story short, it was the economics of the situation that halted the experiment. Unfortunately, Kasher also came away with the realization that incentives for those companies to make changes to their business models are simply not there. Although many sites have these machines, the manufacturers are not used to using them in a clinical trial setting.
When conducting an experiment of this type, oftentimes what is most important is not the end result, but what you were able to learn along the way. Kasher shares a few success stories. “We learned that when doing the trials, the sites liked having the machines,” he notes. “They were simple to use, they produced very quick results, and that allowed personnel at the sites to make decisions quickly and move on to the next step of their trial. We learned about the capabilities of the machines. For example, they were really designed to process a limited number of samples. So if we were looking at a large-scale trial where we would need to process multiple samples, we would be limited in doing that. There is also a fair amount of overhead required to maintaining the machines and we had to provide training to the people who would be using them.”
In retrospect, a key learning point for Kasher had to do with site personnel. One significant downside to the experiment was the shifting of additional work to those working at the sites. Personnel were asked to perform additional work they neither had the staff nor the infrastructure to accommodate. Instead of just drawing the blood, putting it in a box and shipping it off, they were asked to do the assay. In addition, they had to alter the flow of patients coming through the site. They now had to wait around 30 to 40 minutes to get the result, whereas normally they would have drawn the blood, done the other procedures, and moved on. “In the future, we really need to think about what we are going to do to provide additional support to the site and the patients,” he adds.
“Going forward, our learning is if we have a device that can be used at a site that is inexpensive, that has high throughput, and that was easy to use, there would be tremendous value to the site. They could take a rapid measurement, look at the data, and then be able to act based on that finding. I think that is something we still see opportunity in and are pursuing. This also has us thinking about other possibilities as well, such as being able to go someplace like a pharmacy, which is closer to where a patient lives, and having blood drawn. That data can then be waiting for them at the investigator site when the patient goes in for their visit. I also think there is a lot of potential in wearable devices, which will allow the investigator site to get data from patients prior to visiting the site, whether they are at work or home. These technologies truly have the potential to be disruptive.”