AJMC®tv Interviews

Sara Tolaney, MD, MPH, instructor of medicine, Harvard Medical School, attending physician of medical oncology, Dana-Farber Cancer Institute

What challenges are there in the management of patients with HER2-positive breast cancer?

I think 1 particular challenge for patients withHER2-positive disease, unfortunately, is brain metas- tases. Approximately half of all patients who develop metastatic HER2-positive disease will die from progression in the brain, and so, there have been much efforts made to try to improve outcomes for these patients. One class of drugs that we’ve focused on for this patient population has been the use of tyrosine kinase inhibitors. These agents have a lower molecular weight than monoclonal antibodies and are able to penetrate through the blood—brain barrier.

There have been several trials done, looking at these agents, and [results from] a few trials would suggest that when you combine tyrosine kinase inhibitors with capecitabine [Xeloda], their response rates do seem to be much greater, and maybe as high as even 60%. So, now actually, the National Comprehensive Cancer Network has added it to their guidelines to consider the use of capecitabine and neratinib [Nerlynx] for patients who have progressive disease in the brain.

How do clinical trials provide another treatment approach for these patients?

Tim Gronniger, MPP, MHSA, senior vice president of development and strategy at Caravan Health

I think clinical trials are a great opportunity for patients, because [they allow] them to get novel therapies before they get approved, and so it really gives them an opportunity to get a new agent that may have very promising activity earlier than they would normally have access to it. So, usually when I’m seeing a patient I like to always present what the standard option is so that they’re very well aware of what they could get outside of a clinical trial. But [I] also discuss what the clinical trial option would be at that time and see what makes more sense at that particular moment, because it’s not always that the clinical trial is the right decision for them, and so it’s important to weigh the pros and cons of that approach.What have been some of the biggest barriers for organizations trying to implement new payment models?

Relationships within an organization and with physicians in the community are [some] of the first things that you have to figure out, and a lot of that comes up under the rubric of governance structures. Who’s going to be making the decisions about the ACO’s [accountable care organization’s] performance, and who’s going to be making decisions about who’s doing what about who the leaders of it are? And, then, technology ends up dominating a lot of discussion time as well. How are we getting all of our information in 1 data warehouse if we’re building an ACO with our independent community network of physicians? Then there might be 50 EMO products in that network, so we have to find a way where can at least pull data from that set of vendors, if not push and pull that data. So, getting a good handle on IT and technology and analytics takes a lot of time early on in an ACO.

What changes do you think need to be made to improve the move to value-based payment models?

The changes that we need to see are really consistent policy direction from Washington around what they want the health system to do first. A way to enable providers who are working together on improving quality in trackable form ensure, that is going to continue into the future and avoid unforced errors, such as cancellation of a mandatory bundle payment programs last year, anything that would upset the apple cart in terms of the Medicare Shared Savings Program, where we could see a huge withdrawal if the agency moves to promote risk too aggressively. Now, I said earlier, we all know that risk-bearing models are the future and Medicare Shared Savings Program, but trying to make the future happen in 6 months is a recipe for confusion and turmoil, not a recipe for success. I’m all for the agency being aggressive on risk-bearing models, but it needs to be done in a stepwise fashion.

When working with hospital leaders, how do you discuss the future of ACOs in the United States?

That’s a big question. We talk about a lot of different things with our clients, and the reading we’re getting from Washington, very clearly, is that Medicare wants ACOs to take risk. So, a lot of our discussions lately have been around, how do you get ready for that world? What sort of systems and competencies do you need to put in place? What does your staffing model need to look like? What do you need to do to work more effectively with your physicians? And like we talked about a little bit earlier this morning, what do you need to do to make sure you’re large enough to be able to take risk effectively?

What concerns are top of mind for the providers and hospital systems that you work with?

Sally Okun, RN, MMHS, vice president, Policy and Ethics, PatientsLikeMe

There’s the joke that every health system CEO has a to-do list that’s about 100 items long and they only get to the first 10 on the list. So, at any given day in my world, they’re worrying about what’s going to be happening with CMS policy. They’re worrying about what’s happening with their staff. They’re worrying about what their performance is over the last month and over the last quarter on met- rics that we track. And they’re always planning for the next year as well. So, the initial work in our model is geared around getting ready to put in place practices, build up primary care capacity, and get hired and train the right kind of nursing support to deliver effective preventive and primary care, and then to track performance on metrics that will determine success in the ACO over time. How well informed do patients usually feel when they start treatment after a new diagnosis?

I’m not so sure I can point to any one data point that would say, “Well, this is how well informed they are.” What I would say is that over the course of the last number of years, as we’ve created a framework for what we call “the patient and caregiver journey,” we’ve identified different points in time across that journey and the kinds of questions that patients have, and the things that they are experiencing during those different stages. And so, we have again distilled that down to about 6 common questions that most anybody has. Regardless of what their condition is, they seem to pretty much all fall around the same 6 questions, and one of them that I thought was interesting, in light of your question, was, “What will this treatment do to me?” or “What will this diagnosis do to me?” or “What will this prescription drug that you’re going to prescribe for me do to me?”

So, I think what we’re finding is that when patients come into an environment like PatientsLikeMe and tell us about that journey they’ve been on, what we’ve learned is that they didn’t have a good understanding in the beginning, and these are the kinds of questions that they needed to ask, and they didn’t necessarily get the answers that they needed. So, when they come into an environment like ours and they start asking other patients, they’re starting to get some information that they wouldn’t previously have gotten maybe from the clinician. It’s not that the clinician isn’t giving it to them; they might be giving it to them in a form that isn’t necessarily digestible for the patient. It’s not as easily understood as another person telling them about what that treatment had done for them, not necessarily saying, “It’s going to do the same for you, but these are the things that I experienced.”

Your question made me pause, and I almost asked you not to ask me because I didn’t have a data point I could point to, but I think the interviews that we do continuously reinforce that patients have many, many, many questions and they often are the kinds of things that don’t necessarily get addressed. But that one question of, “What will this do to me?” is, I think, what gets the closest to what you’re asking. And so, it’s 1 of our 6 common questions that all patients ask, regardless, again, of the conditions they have, and they don’t necessarily get it answered all the time.

What benefits does technology offer to gather and use real-world evidence? And what are the barriers preventing the use of real-world evidence in meaningful ways?

I think what we have ahead of us right now is a monumental amount of data. So, I think the real-world data [that’re] being generated, from all different sources, whether that’s from patients themselves and devices they may be using or electronic health records and claims data, it almost doesn’t matter. The volume is just growing so exponentially that we, actually, are ahead of ourselves in thinking about how we can turn that into evidence.

I think we’re a little ways from that, yet, so I think the benefits actually are that we should take a step back and recognize these digitally native data sets that we’re collecting from many situations that never existed before. So we need to be thinking what are the ways that we can start to harness the power of the data, first [by] empowering the data, actually, to start answering questions that could be better understood once we know what the evidence is we’re trying to generate.

So, I think we need to be careful not to suggest that real-world data [are] going to translate into real-world evidence without some fair amount of work in between. I think between the benefits of having more access, greater integration of a variety of different data sources coming together to really form a more holistic picture of a person is incredibly wonderful and I think people will welcome that. It’s the other side of that to say, then, “How can all of that information become meaningful to me when I’m trying to make a healthcare decision or when my clinician is trying to help me with a healthcare decision?”

How has the understanding of the need for patient input changed over the years?

I think in light of where we are, at ISPOR [the Professional Society for Health Economics and Outcomes research], and the fact that there’s been such huge advances in the last number of years in recognizing that people have preferenc- es around [their] care—and not only the care, but also the research they want to participate in—I think what we’re seeing here and recognizing here this year is that it’s finally starting to resonate with the research and clinical community in a way that people are embracing it. They want that kind of partnership.

I was on a panel [at ISPOR] that asked whether or not we’re at an inflection point, particularly around transforming digital health into something that can actually inform real-world evidence. And I do think we are. I think we’ve gotten to the point where patients and people are more engaged in their health. They’ve got more tools that can help them have that kind of experience with it.

Ejim E. Mark, MD, MPH, MBA, Chief Executive Officer and founder of AccessHealthcare Foundation

Now, we really need to empower the data so that data become more meaningful in their daily lives. And I think that would be something that ISPOR as an organization can help advance in the kinds of work that it does and the research it supports.With increasing amounts of health data available as a result of new digital technologies, what challenges exist for providers and researchers to use that data in meaningful ways?

The data [are] not being collected in a similar manner, so you have a lot of data sets that are not connected. Because of that, analyzing such data can be challenging for researchers. But looking into the future, with things like artificial intelligence, or looking into things that analytics can actually do, we hope that we keep gathering data in anticipation that one day we can integrate all that data and be able to get better analysis around it.

Are consumers’ personal data being adequately protected as more digital technologies collect and share health information?

We try as much as we can. The reality is that there are breaches all the time, and it’s not about just single breaches. There can be breaches from the provider side and the user side as well. So, being able to stay on top of protecting or securing this data is something we should always think about. We shouldn’t put it at the back of our mind, but we should be thinking about that, because I’ve heard about 2000 attacks in hospitals each day.

As systems are becoming more interconnected, how can they help protect that data while still allowing it to flow freely?

Michael Thompson, MD, PhD, FASCO, Aurora Advanced Care

You have to determine who owns the data and where the data resides, because that’s important. If the data resides in the cloud or if it’s in servers that are within the hospital systems, then those are different protections that are needed. And that’s why it’s imperative to know who owns the data and where it resides and provide adequate security around that. What challenges are associated with the mass adoption of electronic health records (EHRs) and precision medicine?

There are a number of issues around EHRs that are separate from just precision medicine. But for precision medicine, it’s how do you get this 20-page PDF that has a dashboard summary, that has all this information? How do you get that into an electronic format with discrete variables that you can search and do something with? That is one of the challenges that we were recently discussing: How do you make all of that happen? And there are different companies trying approaches.

We have purchased, and are trying to implement, Syapse to do that, but within 1 molecular testing vendor, you can go to individual portals they have to find out some of that information, and sometimes they will give you downloads of electron- ic information. But I think the big idea is how do you get all of that information from 1 site, like our place, and share it across the country? And that’s a part of the Biden Moonshot initiative, to try to break down the silos so we gain information faster, so we aren’t making the same mistakes over years. We’re learning in weeks and months what people are doing. So, I’d say that there are a variety of vendors and approaches trying to do this, but it’s still very much a work in progress and it’s still very early on.

I think we know a number of things [about precision medicine]: We know that precision medicine works and there’re a lot of examples where it’s standard of care. We know precision medicine doesn’t work, including BRAF inhibition in colorectal cancer. We have to combine it with other therapies, and we know that there are a lot of problems with precision medicine and lots of areas that we need to define better. But that shouldn’t stop us from moving forward and trying to optimize the care for our patients. And we realized that sometimes we don’t have enough data, but we’re trying to move as fast as we can to help people.

Download PDF: AJMC^®tv Interviews