Transcript: The Value and Opportunity in Individualized Cancer Care: Part 1

Gordon: Hi, my name is Gordon Kuntz. I've been working in the business of oncology and oncology clinical pathways since about 2004. In my consulting practice, I'm fortunate to work with several clients on pathways, value-based care, and practice economics.

Welcome to the first installment of this video series. In these interviews, we will explore a wide variety of topics in oncology and meet some interesting people doing some very interesting things to advance quality and value for the stakeholders throughout the cancer care ecosystem.

Before we get started, I want to thank the Journal of Clinical Pathways for giving us an opportunity to bring this video series to you. For today, we're fortunate to have two good friends of mine, both of whom have great experience but different perspectives on our topic, the value and opportunity in individualized cancer care.

Our first guest is Dr Bryan Loy. Bryan and I first worked together about 10 years ago when we were developing some of the early Oncology Medical Home projects. Bryan has been with Humana for 18 years and is currently the Physician Lead for Oncology, Laboratory, and Personalized Medicine at Humana.

He's been a great collaborator and innovator in bringing value to cancer care. Welcome, Bryan.

Also joining us is Tushar Pandey. Tushar and I met about three years ago when he became CEO of SimBioSys, a precision medicine company using a novel simulation technology to enable individualized treatment planning.

Tushar's background includes several years as VP of Client Services at Strata Decision Technology, a financial analytics, planning and performance company. Welcome to you both. I know that both Tushar and Bryan are absolutely passionate about improving quality and increasing value in cancer care. I can't wait to get our conversation started.

Bryan, I'm going to kick it off with you. To start off, please take a couple minutes and share a bit about the work you're doing now as it relates to personalized medicine in cancer care.

Dr Loy: Gordon, it's a great question to start off on because when you say personalized medicine it means a lot of different things to a lot of different people.

If you go to the NCI website, for example, you'll get a definition that feels like it's largely around molecular testing and tumor biology, whereas if you go to some of the professional organizations, you'll get learning about the whole patient, what their patient's values are, and I'll steal your word, individualizing treatment to align with your values.

I work with a health plan, as you said, with Humana, 18 years and 20-plus years in the payer business. When you're managing populations, you get the luxury of doing a little bit of both, to working down at the micro level around the testing and also to think about what can we do as a health plan to improve the member journey.

In normal health plan—I'll call it operations—there are functions like P&T, so pharmacy and therapeutics, where we write policy that are informed by personalized testing and medicine decisions.

There are policies that are developed in the medical side of the house, where we look at new technologies for testing, biomarkers, and next-generation sequencing molecular panels, for example, that inform many of those medical decisions that would help one to personalize and craft a medical management path forward that's specific to the patient.

I spend a lot of my time in oncology, but I also would say I live with both definitions. I too am interested in using that type of information, that type of technology to improve upon the member experience.

I also would be very careful to say that, even though we manage populations, we're interested in the deliverers of care, or our collaborators, vendor partnerships, and many of our clinical programs, getting to know our patients and making sure whatever technologies we're bringing to bear align with patient values.

Personalized, to me, represents a spectrum of definitions to which all are applicable.

I do a lot of that work, everywhere from the policy work that I've talked about to try to optimize our operations, both the request for the test itself, adoption to new technologies and into our operations, where folks are insisting on learning about the patient's desires, as well as molecular test results, biomarker results, when it comes time to look at requests for targeted therapies.

Gordon: Great, thank you. Tushar, tell us a little bit about SimBioSys and your approach to personalized medicine.

Tushar Pandey:  Absolutely, Gordon. Bryan, such a pleasure to see you again. I appreciate your insight, and I appreciate your guidance on the precision medicine side and being a leader in this space. SimBioSys is an early-stage startup out of the Midwest. We’re about 30 people now. We're taking a pretty unique approach towards precision medicine.

You're absolutely right. Bryan has been defined as molecular diagnostician. Precision medicine is more than that. Precision medicine is, what is the right treatment for that particular patient on that particular day based on their tumor?

What we've developed is a software application called TumorScope that can essentially create a virtual model of a patient's tumor in three-dimensional space, from standard of care imaging, and then empowers the clinician to be able to test the treatment options that are indicated for that patient, all prior to initiation.

We hear this a lot from clinicians, we can't take back toxicity. Once the patient has been given a drug, it's hard to take that away. Getting that right upfront is incredibly important to be precise and individualized care for the patient. We're serving really with the patient at heart. We're serving clinicians who want to assess what is the right therapy for my patient.

We're also serving the drug development space in order to be able to bridge the gap between preclinical success and failure in human clinical trials. There is a right drug for patients, whether it passes a clinical trial or not. Where we're helping pharma companies with is being able to find that right subpopulation for those patients so that novel therapies are getting to the right people, and faster.

Gordon: Great. Thank you. Bryan, back to you with another question here. Precision medicine and genomic testing have altered the treatment for cancer over the last several years. What do you see is the role for even more precise, personalized, or even individualized medicine for cancer care? How precise would you want those technologies to be?

Dr Loy: It's an interesting question because there is a technology looking for clinical utility.

Again, I'll build off from some of my previous comments. That would be I see this as one of many tools to personalize medicine, and to the extent that we can take these types of capabilities that Tushar was talking about and bring them to bear on the experiences and the end points that are important to patients.

What does that mean? If I'm living with a cancer diagnosis, and I have a chance of cure, I want what Tushar described. I want minimum amount of toxicity. I want the information that will give me the assurance that I'm doing everything I can to be cured and get on with my life.

Whereas if I'm in the metastatic setting or in the incurable setting, I'm looking for show me the things that are going to give me overall survival or quality of life, both, and reduce toxicity, and any lever that I can bring to bear on that clinical experience that would help me not to get drug that's not going to benefit me, which may in fact harm me, or get me on the right drug at the very first time. You'll have improved upon at least my care and, ultimately, you may have improved on the standard of care, which would have otherwise been offered to me.

To get to your second question, how precise does that have to be? I recognize that we don't have any perfect tests. Somewhere between a coin toss and a perfect test, we're going to have to come to grips on where do you set the needle.

I'm not here to assign a number. Somewhere in our discussions, we got to figure out what the range of predictive values need to look like and how accurate a test needs to be.

At the end of the day, we have an obligation to make sure that the clinicians that are using the tool and the end users, the receivers of the consequences of those tools, they have a right to understand what the limitations are, what the risks and the benefits are, ultimately, before we even go down the road of testing.

We have an obligation to share with the end user (ie the patient), what they can expect out of it, because just because you have a druggable mutation, does that align with your values?

If you're going to get 14 extra days on average of therapy as a result of doing the tests, probably you need to have that conversation upfront with both the provider, as well as the patient, to decide if that aligns with their goals and their values.

There's a lot here other than looking for an expectation of "Here's a test result, and that should guide our therapy." It's a much more complex conversation, in my view.

Gordon: Fair enough.

Tushar: Gordon, if I could add to what Bryan was saying.

Gordon: Please do!

Tushar: It's a tough conversation because the risk-benefit calculus in oncology is so complex. The additional month of therapy could add you two additional months of overall survival, but the toxicity that you live with during that period is unbearable for some patients.

Reducing 1% risk of a recurrence could give you secondary leukemia for a lifetime, or cardiotoxicity for a lifetime. That risk-benefit is incredibly complex.

With precision medicine today, we really haven't empowered clinicians to be able to share and collaborate with patients on that risk benefit-calculus in an effective manner.

Dr Loy: Building off of what Tushar just said, not only is it complex, sometimes it's unknown. Having tools that help us to get to near equivalence or the effectiveness we're looking for while reducing toxicity, especially early on in life, where you've got a lot of months of potential life left, becomes extremely important.

In our geriatric populations which we serve, my employer, Humana, we're looking for what are those indicators, those markers that can help folks that may not otherwise have been represented well in clinical trials to at least understand how those drugs are going to perform in the geriatric population.

There's a lot of unmet need here that we're talking about. Bringing these tools to bear will only help us answer these questions more intelligently.

Gordon:  Tushar, with the advent of personalized medicine, clinical pathways are still the gold standard in cancer treatment selection, certainly for earlier stages disease. They direct as much as 60% to 80% of treatment decisions today.

If I understand directly, the goals of SimBioSys and clinical pathways are actually rather similar. That's to help clinicians identify the most appropriate treatment for patients and their disease presentation. How does SimBioSys approach that question differently from existing paradigms, such as pathways built on clinical trials or even the precision medicine we've been talking about?

Tushar: Excellent question, Gordon. Clinical pathways, what I think of them are trailblazers in this concept of promoting standardization, promoting the use of data and data-driven decision-making. They've been incredibly important in this field to help clinicians level the playing field.

Just like any industry, there is a natural evolution that needs to happen. For far too long, in oncology, we've been treating what's best for a population. That's what pathways encourages. What is best for this population of patients? We look at the evolution from a treatment planning perspective.

We used to do best for the indication, then we started stratifying on biomarkers of commonly known pathological markers. We're heading towards the genomic era, where we're becoming more precise. Pathways bridges some of the gap between indication and biomarker. There's that next evolution that needs to happen.

We at SimBioSys, we rally around the analogy of navigation. When you and I leave home, we have this tendency of using Google Maps, and knowing exactly where we're going, how long it's going to take us to get there, is there a better route for me on that particular day, on that particular moment, what is the traffic going to look like, other roadblocks.

We need to know all of this before we leave home now. If you look at that, maybe a few decades ago, we used to use maps, or this tool called MapQuest, where we would type in the address, would print out the direction, then we would go there.

We think of the parallel in pathways more being towards, "Here's a map of potentially how you get there." Where we need to get to is the true individualized nature of the patient's cancer and the patient's treatment, so they can assess with their physician, "What are the various paths for me, for my particular tumor, to be able to get there?"

That evolution is incredibly, incredibly important. What that evolution is going to be triggered by, very similarly with the navigation industry, was that it was a convergence of technologies that happen. Satellites had to be up. We needed to know coordinates. We needed standardized data. Applications and tools had to be built like Google Maps to be able to empower people to use it.

We think of precision medicine very similarly. There's technologies like genomics, single-cell, liquid biopsies, even simpler technologies, like imaging, that are part and parcel. Those needed to be foundationally present and used for technologies like us to be able to converge and empower larger group of clinicians over time.

Gordon: That's the difference between you and I. I work from home, so I don't have to use Google Maps to get down the hall to my office. You work in Downtown Chicago, whole different issue.

Tushar: Hoping the best for the world. It's only a matter of time where we open things up again.

Gordon: There we go. Hey, a follow-up question on that. In your experience working with oncologists, what do you see and what do they see as the biggest gaps in treatment planning today?

Tushar: A few areas. One is transparency in understanding the why. Why would a patient need this therapy versus another? We're empowering them with data today. It still doesn't answer the question why, or why a patient will respond or won't respond, or why a patient did well to a particular disease.

Answering that why is incredibly important. You can't answer that just with an expression of a gene. You need to answer that with an insight, like does my patient have a drug delivery issue? Does my patient have a particular receptor that the drug would work better on?

The reason for it is two-fold. One is for them to be more confident on their decision-making. The other is patients are consumers now. They're asking questions. They want to be part of their care. They have to be part of their treatment decision plan.

Physicians really don't have the tools today to be able to collaborate with their patients to be able to educate their patients on why particular decisions will do certain things to them, which is what we were talking about earlier today.

Gordon: Yeah, okay. I’ve got one that I want to hear both your perspectives on. What do you believe are some of the gaps in precision medicine today? Is what we have today good enough? Bryan, I'll start with you. Thoughts?

Dr Loy: Nice piece of work The Cory put out recently. They did an assessment on the information that's available. The overall conclusion, if I may paraphrase, was either we still have a lot of opportunity, because we really don't have randomized clinical trials to answer the question of how much do patients benefit versus nonmanaged care using molecular testing, for example.

We're not there yet, which begs this question of, "Well, so what do the evidentiary requirements need to look like?" Some would say, "Randomized clinical trial or we're going to have to live with a lot of uncertainty." I don't know what the exact answer is or isn't. I don't know what trade-offs and compromises we'll have to live with.

We've entered that arena, where folks are beginning to press for those questions. We need to do that. I, for one, I don't want to see us wait a decade before we get the answers to those questions. I also don't want to see us live with the regret of depending on a technology that we find out later on really wasn't serving those needs very well.

We've got plenty of examples that I can point out today where we said yes to something and we've learned our way into it, said, "You know what? It probably did not serve this population as well as or with as much confidence as we thought that it would have."

I'd say, similarly, we've experienced the same thing from randomized clinical trials for populations that really aren't representative of real world. Somewhere in all of this, we've got to take an approach to learn our way into and lead a path of nonregret to get where we can get comfortable either living with a certain amount of uncertainty or acknowledging that those gaps are the price we're going to have to pay.

Either way, I would say we've got to have pretty strong evidence that we're a work in progress. We're putting a measurement system in here that ultimately leads to things that are important to patients, living better or living longer.

Gordon: Tushar, what's your perspective on this? Is what we have today in precision medicine good enough?

Tushar: Couldn't agree more with Bryan. There is so much opportunity. I said this to somebody earlier. The rate of innovation in this space is so inspiring. There are new things happening every day, but we haven't been able to measure the impact yet. There is that uncertainty that we live with. Is it good for the patient? Is it good for the physician? Is it not? That's hard to judge.

Where I slightly differ where the opportunity, as we see it, exists and why precision medicine hasn't lived up to its promises. A few reasons, one, the availability and the impact has been for very limited populations.

If you think about precision medicine today, it has been, and rightfully so, primarily geared towards advanced stages of the disease with less focus on earlier stages of the disease. Any physician or scientist will tell you, if you get things right in the earlier stage, fewer patients will progress to the advanced stage of the disease.

The other is around access. Whether it's a technical barrier, whether it's a financial barrier or a geographical barrier, precision medicine today is not precision medicine for all. Precision medicine, for it to be impactful, needs to be for all. You need a broader usage of precision medicine to be able to find those cases that are truly benefiting from them.

Secondly, as we think about precision medicine, and Bryan called this out earlier, the orientation has been purely on the molecular side. Everybody thinks precision medicine is a genomic testing or sequencing testing. Precision medicine needs to be a little bit more comprehensive.

You're treating a three-dimensional disease, and you're trying to be precise by understanding only at a molecular level what's happening. Precision medicine and individualized medicine needs to be broader and needs to be understood multidimensionally, just the way the disease is.

The last thing I would say, precision medicine has been focused primarily on physicians today and researchers or scientists. We've missed the orientation towards the patient.

There are nuanced ways you do that. How do you make precision medicine understandable to a patient? Then how do you take that understanding and ultimately lead to a shared decision between the patient and the physician? Immense amount of opportunity, but just like Bryan is, we're all excited about what the future holds there.

Dr Loy: I don't feel like that's false optimism. We do have some good examples, imatinib in CML, for example. In my career, my lifetime, that's a great story. Where we've gotten into other areas, we've seen some marginal increases and some surrogate end points.

I would just tell you, rather than trying to solve all problems all at once, if we can figure out ways to pose the right questions, where we're exposing patients to either too much drug, too much toxicity, and reduce that it either improves the quality of life or to, in fact, enhance quality of life, that would be a major accomplishment in a population today that, in my view, remains unserved.

Tushar: Totally agree.

Gordon: Another one for the both of you. Breast cancer has among the highest incidence rates of cancer in the US. In general, the survival rate is very good these days, thanks to the many new therapies introduced since the late 1990s. Perhaps because of that, the standard of care is fairly well-known for most subtypes.

We still see more tests and therapies emerging in this area. Because of the volume of patients and the rapid advances and broad range of treatment options, do you think breast cancer will always—always being a very long time—but, for the foreseeable future, be a major area of focus for the oncology community? What are the challenges you see for patients and physicians in this area? Bryan, we'll start with you again.

Dr Loy: I can hardly wait to hear what Tushar says. That's such a great example to react to. Breast cancer is so many diseases, and there's breast cancer-specific problems.

If I were to anchor any of my thoughts to one concept, it would be overtreatment, whether it be screen-detected breast cancer, trying to make sure that folks are going to benefit from therapy, whether it be radiation therapy or chemotherapy, for recurrence, overall survival, or assurance of cure. There's still work yet to do there. We've got tools there.

Early-stage breast cancers, for example, DCIS, ductal carcinoma in situ, plenty of opportunity there to figure out which ones are really going to benefit from treatment, aggressive treatment versus not.

As you get into different phenotypic dispersions from the tissue diagnosis and start asking, who's going to benefit from what therapy, and what does that sequence need to look like, and at what point can we stop there and let the patient enjoy a drug holiday? I think we've got a lifetime of opportunity in front of us just answering some of those questions.

We haven't talked about dosing. We haven't talked about scheduling. We haven't talked about where we've gotten into recurrence, etc, and trying to maximize that. Quite frankly, figuring out where we need to step back and ask, "Is this patient likely to benefit from this therapy at all?" All very important questions. They play up quite nicely in the world of breast cancer.

Tushar: Great points there, Bryan. If I could jump in, Gordon?

Gordon: Please.

Tushar: Breast cancer was an area that a lot of people were doing research in, but there wasn't a lot of clinical evolution happening till in the late 90s, early 2000, Herceptin, trastuzumab, was approved. It completely transformed how breast cancer treatment was done, especially for a subpopulation, the HER2-positive population of patients.

Since then, the one part I disagree with you on, there's this assumption that there's a really strong standard of care in breast cancer. There is, but there are multiple standards of care in breast cancer.

For the same patient, you could have up to 13 different regimens that may consist of similar drugs, with 13 different regimens that exist with different schedules, with different doses that may have completely different responses. Physicians know the standard of care, but physicians also see significant variability or uncertainty in what's the right decision for a particular patient.

You compound that with the amount of drugs that are getting approved in this space, everything from immunotherapy. Merck's Keytruda just recently got approved by the FDA. That's going to completely disrupt, in a really great way, the standard of care for triple-negative breast cancer patients.

There are antibody-drug conjugates like T-DM1 that are showing some amazing efficacy in adjuvant populations. There are drugs like tucatinib that increase the survival in metastatic patients. All of these come with their own risks, alongside the benefit.

Doctors are finding it really difficult to be able to keep up with this evolving standard of care in breast cancer, and continues to be an area of unmet need, not just from a population perspective, but because of the changing landscape as well.

You take the evolving treatment paradigm, and what you end up seeing is anything that gets approved in breast cancer essentially becomes additive, especially in the curative setting. What used to be a standard of care with two therapies, now may be a standard of care with five therapies. We still clearly don't know which are the patients who are benefiting from this or not.

We understand it at a population level, and we see the benefit. The benefit in breast cancer is now, because the survival, fortunately, is so high, the benefit is very minimal. You see a 1% increase in survival.

For that one patient that benefits, that's incredible. For the 100 patients that may receive it for that single patient to benefit, there is toxicity, there's costs associated with it that need to be weighed. There's a significant unmet need there. That's why at SimBioSys, we made breast cancer our first indication or first priority, because of what we were hearing from physicians and patients.

Dr Loy: Tushar, I don't look for that to get solved anytime soon by our existing, I'll call it, research framework. We're going to live with five, as you said, preferred regimens plus as we continue to see the pipeline issue new opportunities, even for breast cancer.

Tushar: What we really need, as you think about precision medicine applied to breast cancer, is for physicians to know, "When can I safely de-escalate? When do I need to escalate?" Being able to assess the incremental benefit of what they had versus the risk is going to be incredibly important from an outcome perspective.