CareTalk Podcast @ HLTH 2021 – Dr. Amy Abernethy. From FDA to Verily and Back Again… Maybe?

Former FDA Deputy Commissioner Amy Abernethy has been rumored as the next FDA Commissioner. And why not? She shares her experience and vision for clinical trials and explains why data quality keeps her up at night.

David Williams:

Welcome to Care Talk, America’s home for incisive debate about healthcare business and policy. I’m David Williams, President of Health Business Group.

John Driscoll:

And I’m John Driscoll, the CEO of CareCentrix. David, Health 2021. When will it end? How many more of these do we have to do?

David Williams:

It ends. I think it goes straight till next year. I see it’s going to be in Las Vegas. So count your blessings, John. You’re in Boston.

John Driscoll:

We’ve got one of our most esteemed guests yet, David. Do you want to introduce our guest, because you’re the only one who can figure out that long title?

David Williams:

I wrote the title down. I thought I was trying to make it into an acronym, but it didn’t really spell out right. So this is Dr. Amy Abernethy, and she is President, Clinical Studies Platforms, at Verily Life Sciences. Welcome to the show.

Dr. Amy Abernethy:

Terrific to see you guys.

John Driscoll:

So Amy, how did you get involved in healthcare and medicine? What’s your origin story? Before we get into the FDA and Verily and Google and money and data and drugs and all that cool stuff, how did you get interested in medicine? What got you started?

Dr. Amy Abernethy:

Well, I went to medical school actually thinking I was going to be a tumor immunologist, work in the lab. And I was basically running studies, injecting people with tumor vaccines, and realized I really liked taking care of patients. Became an oncologist, and as I was really doing the work of taking care of patients with melanoma, I realized that ultimately, I wanted to try and figure out: How do we get treatments that work to patients who need them, as quickly as possible? And that’s really been not only my origin story, but the whole purpose of what I do every day.

David Williams:

Clinical trials. You’ve had a big role there, and I heard you talking about data in clinical trials a couple of weeks ago. And you said data quality keeps you up at night, which is how I know you really are interested in clinical trials. What should we know about what you’re worried about? What should we be worried about?

Dr. Amy Abernethy:

So it’s interesting. Data quality does keep me up at night. Ultimately, I think about clinical trials, and in the clinical research space, really, we all aim to want to leverage all possible data and information available to answer the question at hand. But it turns out that there is a huge data quality mismatch or gap between traditional clinical trials data (data that are basically collected according to a very well-scripted study protocol), versus data, for example, that lives in the electronic health record or claims data.

And so we’re going to leverage all available data to answer questions confidently, like, “Is this vaccine safe?” Or “Should we get a booster?” Then we’re going to need to make sure that we understand the characteristics of the data, and when we can reliably use them, which is essentially understanding and solving for data quality.

John Driscoll:

One of the things, Amy, that I worked on at Medco many years ago, is it didn’t make any sense to me that we had information on 60 or 70 million people, and on 20 million, we actually had full medical claims and the prescription data. And yet FDA was not taking advantage of real-world evidence, as it’s now the acronym. That’s RWE for you. David didn’t like the acronyms.

David Williams:

Oh, can I pronounce that one?

John Driscoll:

But what’s happened with that? Because it always struck me that that was one of the richest datasets of how drugs, and actually people, interact over time with chronic disease.

Dr. Amy Abernethy:

So first, if we go back and we think about how traditional clinical research has been done, it’s really been by very prescribed protocols that get to very high-quality data sets. Or we can trust all of the background of the information. We understand where it came from, who collected it. We know exactly: If it got changed, who changed it, and why? And can we make a confident result from those datasets?

Meanwhile, as we think about all that information, like the claims data and the pharmacy data and the RWE, being able to make sense of it can be hard, because of the data quality mismatch we just talked about. In 2016, the 21st Century Cures Act compelled FDA to really start to figure out: How can regulators and others make confident decisions based on these types of real-world data sets? It didn’t say, “FDA, you must,” but “FDA, help the whole landscape figure out: What does good look like?” And really point in that direction.

And what you’ve seen since that time is first in 2018, FDA put out a framework of, “Here’s how we think about the problem.” And subsequently, we’ve seen a number of use cases be brought before FDA, where they can now make judgements based on a particular data set and a particular question, and say, “Okay, here’s the kind of decision we can make today, and we could make even better decisions in the future if you improve the quality of the data. Develop more standards to how information is being collected. Improve the way we analyze the data sets.”

And so that’s where the space of real-world data and real-world evidence is going.

David Williams:

So Amy, you were saying FDA. Of course, you were at FDA and had what many considered to be a very successful stint there. Now, why did you leave FDA, and what would it take to get you back?

Dr. Amy Abernethy:

Oh, interesting. So in the context of the pandemic, I was Principal Deputy Commissioner of FDA during really, the beginning, and hopefully, this will be towards the end of the pandemic. Left in April of 2021, and what I could see was that our discovery engines are working really, really well. The ability to develop vaccines, for example. And I had been working at FDA to how do we scale the regulatory apparatus, so that the FDA can be ready to receive and review applications?

But when you look at the middle, clinical development, like clinical trials and using real-world data and real-world evidence, it’s circa 1995. And I really left FDA to try and solve for that problem.

John Driscoll:

And so now that you’ve raised the FDA question, what qualifications would you look for in a commissioner, if you were president?

Dr. Amy Abernethy:

Qualifications in an FDA commissioner? I think the first thing I would want, especially right now, in the pandemic and late in the cycle, is a person who could hit the ground running and make sure they know how to do the job of being commissioner. And also have the confidence of 18,000 FDA employees and Congress.

The second is I would want a person who understands what it means to develop drugs and devices and how we can do that thoughtfully and confidently.

And then the third thing I would be looking for is a person who understands that the FDA’s task is to be a science-based public health agency, so that person needs to understand science and public health.

David Williams:

So what did you learn from Flatiron? Because you were there before FDA, as I recall, and now it’s a place certainly working right in the trenches on real-world data and trying to do something practical with it.

John Driscoll:

David, why don’t you explain what Flatiron is? I mean, throwing you that inside place in the conversation. [crosstalk 00:06:41].

David Williams:

I think it’s a building in New York City, I thought, and it has a whole district named after it. But I think that Flatiron is basically electronic medical record for community oncology practices that then has also expanded to be able to use that data for pharmaceutical development. It’s had a lot of interest in investment from the pharmaceutical industry. But Amy worked there, so she might be able to correct that.

Dr. Amy Abernethy:

There you go. Yes, I was the Chief Medical and Chief Scientific Officer at Flatiron, and really, what we were doing at the time was learning how to take the electronic health record data and clean it up. Curate it, so that then the data sets were of improved quality and could now be used to reliably answer questions like, “Did this cancer treatment help men with breast cancer be able to have resolution of their disease, not just women with breast cancer?” So we were using data sets to answer critical questions like that.

You asked, really: What did I learn? And the first thing that I learned is that we can indeed improve the quality of data sets of all types and bring them to the table to answer important questions in clinical research.

The second thing I learned is that importantly, this is not a solve for any one particular group. It’s not a tech company solve. It’s not a research company solve. It’s not an academia solve. Actually, you have to bring all the groups together to solve a hard problem.

The third thing I learned is that as we’re trying to improve data sets to answer questions, you actually need the other stakeholders at the table, as well, like regulators and payers, people who need to use the answers to those questions. So they can say, “Yes, this is what good looks like. Here’s where I am concerned. Fix this, and here’s how we move forward.”

And importantly, when I got to FDA, what I could see now is, “Wait a second. We need to continue to drive that story forward, and as FDA, FDA needs to be at that table.” And then people who understand what the solve looks like, maybe somebody like myself, needs to now think about how do we inform the tech industry to keep pushing forward?

John Driscoll:

Would you be qualified for the FDA Commissioner job, Amy? Would that be a good fit? I mean, just speaking purely hypothetically.

Dr. Amy Abernethy:

So first of all, serving for our country is really, I think, something that all of us should do, if we have the opportunity in life. And I will always serve as asked. I have, certainly, a background that I think I can bring to FDA. I also think I can serve FDA in other ways, as well, by being a good steward outside of FDA, too.

David Williams:

So you were talking about bringing the different stakeholders together, and there has been more discussion, I’ve heard lately, about the healthcare ecosystem, particularly as it relates to data. Is the ecosystem evolving in such a way that you think you will eventually be able to sleep better at nights, with data quality being richer and better?

Dr. Amy Abernethy:

A huge and hearty yes. So I really do believe that first, we can measure data quality. We can understand the characteristics of data. And because we can measure it, then we’re going to be able to systematically improve it. On the 30th of September, FDA put out a draft guidance that describes how they think about real-world data from the electronic health record. And also, what are the quality characteristics that are looking for?

So they’ve given a blueprint for describing data quality, and now we can use, for example, sensors in our watch. Or better data curation through a combination of machine learning and in-person abstraction to clean up data sets. And we can also think about: How do we combine real-world data sets with clinical trial datasets to constantly improve the quality?

John Driscoll:

So does all of that shrink the amount of time that it takes to actually develop a drug that’s effective and potentially lower the price of drugs? Because these CAR-Ts, or a lot of these curative therapies that are based on biologics, are going to be titanically expensive if we don’t find a way to reduce their costs.

David Williams:

Even John might not be able to afford the deductible on that one. A co-pay.

Dr. Amy Abernethy:

I think we do three things. So first, I think we do make the process of conducting clinical research and getting to the answers that we need more efficient. So we shorten the clinical development time. That’s number one.

Number two is I think not only do we make it more efficient; we bring some of the costs out of the system by improving data quality and our ability to use all available information.

But three, we actually also equip ourselves with the ability to continue to evaluate medical products across the entire life cycle of the medical product. And that means that we can not only make confident decisions up front, but to continue to hone where a drug should be potentially prescribed. Better understanding about populations, personalization. And that actually happens as we continue to use a drug or device across time, and I think we can solve for that, too.

John Driscoll:

That whole Phase IV real-world evidence. So you think that what’s been promised for 20 years is actually we’re on the cusp of being able to crack it?

Dr. Amy Abernethy:

If I have anything to do with this, I am going to push that story forward and get to a yes.

David Williams:

Well, that sounds like a great place to wrap things up with Dr. Amy Abernethy, President, Clinical Studies Platforms at Verily Life Sciences and many other things in the past and possibly in the future. I’m David Williams, President of Health Business Group.

John Driscoll:

And I’m John Driscoll, the CEO of CareCentrix. Thanks for listening, and please subscribe. And thank you so much, Amy, for joining us. Good luck with your public and your private career.