AJMC®TV interviews let you catch up on what’s new and important about changes in healthcare, with insights from key decision makers—from the clinician, to the health plan leader, to the regulator. When every minute in your day matters, AJMC®TV interviews keep you informed. Access the video clips at ajmc.com/interviews.
What are some policy priorities for the Community Oncology Alliance (COA) in 2019?
I think the top priorities when you look at COA, and as we move into 2019, are going to be surrounding these middlemen. These pharmacy benefit managers have gotten so brazen, and we have more examples of where cancer patients cannot get their medications on time or at all, or get the wrong dose or too much of the medication.
I saw one case recently where a patient received $172,000 of drugs that were totally wasted. So, we’re going to be looking at this and we’re going to be looking at these middlemen getting in the way of the patient and the physician making a decision about their therapy.
We’re seeing too many preauthorizations being held up, so this is a top priority. We’re very concerned about the federal government now giving these middlemen more power, especially in Part B, which is chemotherapy delivered in the physician’s office.
We will be all over that, and any proposals at all that get in the way of the physician and the patient making informed decisions and going ahead with timely cancer treatment.What are some key learnings in the Oncology Care Model (OCM) and how can they be applied to other reform models?
There have been a lot of key learnings from the OCM and how it would relate to other payment models. One is that attribution is key and making that an easy process for the practices. They have to know that the patients they enroll in the model are in the model and have that collaboration with the payer to make sure that the attribution is done well and done quickly for the practice.
Looking at total cost of care is difficult, and figuring out what the practice is and is not responsible for can be very difficult to do. Case mix is a huge issue as well; with breast cancer, for example, there can be quite a lot of variation in the appropriate price of care for different patient populations. Especially for small practices with small populations, just a little bit of a case-mix difference can cause huge variations in the total cost of care.
I think the big key learning is that whatever payment model you’re in requires significant collaboration between the practice and the payer. A lot of trust has to go into that relationship—trust that the payer will give you timely information, and trust that the practice will do what they say they will do and really look at improving the value that they provide to the patients.
What is the struggle for reimbursement for new innovative cancer therapies?
The cost of cancer care is rising dramatically, and the cost of new innovative cancer therapies is quite high for many reasons. It’s difficult to address that in a practice because we don’t control the cost of care. You just want to provide the appropriate care for the appropriate patient at the appropriate time— that’s the mantra for taking good care of patients.
It’s a difficult question to figure out how we can address that in the care that we provide. I think pathways is one way to do that, where we can look at what an appropriate, new, innovative therapy is and which patient population it should be given to. Doing innovative payment adjustments is also important so the practices aren’t discouraged from using novel therapies where appropriate. I think the challenge is figuring out which novel therapies are appropriate for which patients and making sure you’re utilizing them appropriately for the patients that you have.
What has been Texas Oncology’s experience with the OCM thus far?
The OCM has been challenging for Texas Oncology—as it has been for all practices—but it did make us stop and think about how to take care of patients and be more patient-centric than we ever have been before.
A lot of good has come out of the program for us: having shared decision making with patients, really thinking about the cost of care, and having access for our patients [to go to] our clinics to avoid hospitalizations and emergency departments; all of which has been very good for our patients and the patient care that we provide.
The things we’ve succeeded in are simply those—making sure that we have access to our patients, making sure that we keep them out of the hospital and the emergency department—and we have done really well in those areas and have shown improvement over time in those areas for patient care.
We’ve done some very simple things for drug utilization—drugs are a big cost of cancer care—and we’ve done some very simple things for managing drugs and utilization. For example, using pathways and looking at other drugs, like antiemetics and growth factors, and making sure we’re using those appropriately.
What we’re challenged with right now is that next step, that next hurdle where things get a little bit more complicated. I feel like we’ve done the easy things, and the things that are easy to rally around and do well with for the practice, but now we’ve got to start making some very tough decisions to make that next hurdle and really succeed in the program.How does the Community Oncology Alliance (COA) view “value” in oncology, and how is that distinction determined?
That’s a great question, and it’s a question that we often get asked. COA’s perspective on value is: let’s ask all of the different stakeholders, and let’s try to come up with a model that can meet that statement or meet the value guide for everyone whether they’re a patient or an employer. A patient has their own set of values, providers have their own set of values, and the people that pay for it have their own set of values.
So, let’s come together and make sure we understand that, especially when we present the model and all the options for whatever your definition [of value] may be. To us, there’s no simple definition. But the model, the payment system, and the delivery system need to consider all of the different aspects of value.
Where do you think the future of the Oncology Care Model (OCM) is headed? How will the design model change?
I think the OCM will live its life for the next 2.5 years. They have been fairly vocal about living out the life of the model. So, it’ll end 2.5 years after this December, and then they’ll have to read reports, and maybe 3 years or 2.5 years after that they’ll come to a conclusion, “OK, well what did we learn?” Now, that’s a lifetime to me as far as reform goes.
We’re trying to learn from what we’ve seen so far, what they’ve done well, and what they struggled with, and then create sort of a baseline or a template for changing the way cancer care is to be reviewed and rewarded for everyone, not just Medicare. So, that means making it simpler—hopefully—and communicating and communicating and communicating again, because you’re going to have to. Introducing different participants, maybe national payers, is something we haven’t done in the past; maybe some big employers as well, to give them the tools to get started.
We’ve learned a lot from the complexity of this one and the communication (or sometimes the lack thereof). Sometimes they’re a little too rigid when making changes—and at times they need to be—but sometimes they need to talk through some solutions and try to make it better for everyone. It needs to change; they’re doing a great job trying to get that process started, but we need to finish that process.Do you think there should be greater adoption of risk-based models, and if so, why?
In a study I had read, there was, I believe, a 23% in- crease over the past 2 years, so there is a growth rate relative to adoption of alternative payment models. But relative to risk, the question is preparedness. I’m not quite certain that the system is equipped to provide a comfort level for clinicians so that they’re ready to take on 2-sided risk.
For example, something that was published in your journal, I think it was in April, David Nash from Jefferson talked about why providers were not willing to take on 2-sided risk. It was that they felt they did not have control over variation and quality within the system, and also that their lack of control was overestimated. I believe that has to do with a lack of infrastructure, so the technology is not there; the data and the feedback around medical information and claims information to allow a practice or a health system to prepare may not be there in the broader scheme for there to be wider adoption at more than the pace we’re moving at now.
How is the National Comprehensive Cancer Network (NCCN) working to define and assess “value” in cancer care?
So, in 2015, NCCN introduced the NCCN Evidence Blocks, which is a tool that allowed for clinicians and patients to have a conversation about what is important to a patient’s individual value system. We’d consider the efficacy of a therapy, safety, quality of the evidence, consistency of the evidence, and affordability.
The categories of preference are another step; it’s another tool to help clinicians understand what optimal care is. We believe that value truly is defined by an individual patient’s need, and so the categories of preference allow for our panel members to signal to clinicians, of all the recommendations in the NCCN guidelines, which are the preferred ones.
What is the struggle with reimbursement for new innovative cancer therapies?
You know, it’s interesting, because my feeling is that doctors want to do what’s right by patients, and doctors are evidence-driven. And so, truly the greatest thing that can help a provider feel comfortable providing an innovative therapy to a patient is the evidence behind that therapy.
So whether they’re in the [Oncology Care Model] or any type of alternative payment model, the strength of the evidence is the strength of the efficacy. The outcomes will drive the use of that innovation.How are mutational data helping inform clinical prognosis and treatment protocol?
I think we’re beginning to understand that mutational profiles can tell us something about the character of the leukemia. Historically, we [defined] patients as either fit for induction or unfit, and we made therapy decision largely based on those clinical factors. Using mutational profiles, we can actually identify patients whose disease is likely to be intransigent to conventional chemotherapies, and with that approach, we can actually slate patients to receive therapies that are more likely to benefit them.
So, instead of giving them traditional cytotoxic therapies, which tend to have a very low rate of remission induction, in say, patients with p53-mutant disease or complex karyotype. We might offer them an alternative induction strategy, potentially with a novel therapeutic combination, like one of the venetoclax combinations. I think that recognition is likely to change the way we approach patients in the future.
Right now, a fit patient would still likely be offered a conventional induction strategy or Vyxeos, the CPX-351 liposomal cytarabine— daunorubicin combination. But perhaps in the future, we might be able to change our approach.
I think the molecular profiling has also unveiled a variety of targets that we can use in combination with traditional chemotherapeutics. So, as highlighted by Keith W. Pratz, MD, associate professor of oncology, John Hopkins University, and Eytan Stein, MD, hematologic oncologist, Memorial Sloan Kettering Cancer Center, up-front combinations of both novel Flt-3 inhibitors and IDH [isocitrate dehydrogenase] inhibitors with conventional chemotherapy like 7+3 can provide really substantial rates of remission induction success and potentially deep molecular emissions, which may translate into better long- term survival for patients.How has the Oncology Care Model (OCM) evolved? What do you think it does well, and what are some of the pain points that participating practices continue to face?
The OCM has evolved in a couple of interesting ways. Data are now a very regular kind of term with the OCM practices. Almost all of them have done something with the access to claims data that they have, so that’s actually progress.
The second thing that they’ve all noticed is that these transformation activities—care transformation and quality improvement—are all actually working. Not necessarily in the way they might have thought, but they’re making patients feel like they’re getting better care and, in some cases, they’re delivering on improvements in quality measures.
The thing that’s still not working as well is this kind of clunkiness with things like attribution. It takes a long time—there’s a lag in how CMS processes the data about attribution and what we call reconciliation. It might take doctors up to about 18 months to ultimately know if the patient they thought they were taking care of is actually acknowledged by Medicare as their patient, and then vice versa. People that they thought were their patients are not neces- sarily theirs, and they don’t find out until about a year later. So, there are still some clunky things that have to do with how Medicare just processes claims under the model.
Can you discuss some of the key trends from the OCM’s second performance period results?
Practices overall are doing better, but it’s not like it’s a massive shift. There were improvements from the first to second performance period, but we ultimately haven’t seen a large majority of practices with improvements. Now, what I don’t know about is the financial; CMS doesn’t publish the financial improvements, so you’re hearing kind of fits and spurts about practices saving money that did not save it before.
So overall, I would say if I were sitting in CMS’s shoes, I’d be like, “Yes, this program is still working and it’s working well.” I don’t know if it’s necessarily “meeting the expectations” of what people thought they would be when they started the OCM, but I spend a lot of time looking at all of CMS’ models and this is a model that’s doing pretty well compared to [accountable care organizations and some others. So even I would step back from performance period 2 and say that this is a model that’s working.How represented are adolescents and young adults in clinical trials?
Overall, AYAs, which are adolescents and young adults—and we typically define that as 15 to 39 years of age—are less represented in clinical trials than children. So, back in 2006, we reported that 14% of AYAs participated in clinical trials. This was using population-based data in the Surveillance, Epidemiology, and End Results Program. This is in contrast
to children, where approximately 90% are treated at institutions with NCI [National Cancer Institute]—sponsored clinical trials and as many as two-thirds [are] participating in clinical trials. So, there’s pretty dramatic differences by age, in terms of clinical trial participation.
Has representation changed over recent years?
So, not really, actually. Part of our goal was to look at changes over time. There’s been substantial efforts to increase both access to and participation for AYAs in clinical trials, and this is really because we’ve noticed less improvement in AYAs in survival over the past 30 years, and this has been attributed to lower participation in clinical trials as well as a number of other factors.
So, our goal was to look to see if we have seen an increase in participation over time. We were able to do some work ourselves in population-based data and found that in 2012 and 2013, we saw a modest increase from 15% to 18%, and these were in patients with acute lymphoblastic leukemia [ALL], Hodgkin, non-Hodgkin lymphoma patients, and sarcoma. But really, no significant increases over time.
There was a suggestion that there is an increased participation in ALL trials, and this has been noted by others. So, that may be the one exception; that for those patients that there is some increase in trial accrual. And there has been a lot of attention given to adolescent and young adult patients with ALL. But again, overall and in general, we don’t see increased clinical trial participation.How does the standard of care for CLL differ based on treatment setting?
Chronic lymphocytic leukemia [CLL] is a slow-growing disease. It takes a long time for many patients to have their disease grow to a point where it needs treatment, and when patients need their first-line therapy, there are a lot of choices out there, and they’re broadly divided into a chemoimmunotherapy approach or a targeted-agent approach.
We’re going to see information in the plenary session that that exact question is being addressed in a study where patients get either ibrutinib, ibrutinib and rituximab, or bendamustine and rituximab. The findings of the study, and this is kind of somewhat an older population, is that the progression-free survival would appear to favor those patients treated with ibrutinib compared with chemoimmunotherapy.
But if you dive a little bit deeper and look at the different types of subgroups—and as we begin to talk about personalized medicine, there are some markers where the benefit of ibrutinib over chemoimmunotherapy is very clear—there are other groups where you can make an argument for a fixed-duration chemoimmunotherapy-based approach.
When you get to the relapsed setting, it really does tend to favor more of the novel targeted therapies, such as ibrutinib or venetoclax, compared with chemotherapy-based treatment.
How has the era of personalized medicine changed the way you think about treatment for patients with CLL?
Personalized medicine is a complex topic because what it generally refers to is a notion that you might find some feature or marker that’s unique to a patient and then select therapy on the basis of that marker. In the case of chronic lymphocytic leukemia, the disease is divided primarily into 2 groups of patients: those who have what’s called a mutated B-cell receptor and those who have an unmutated B-cell receptor.
In this case, mutation is good. It means you tend to have a slower-growing disease, fewer high-risk genetic markers for chemotherapy resistance, and so forth. So, by looking at the IGHV [immunoglobulin heavy chain variable] mutation or the B-cell receptor mutation status, those patients who have mutated disease are generally those patients who are going to benefit from a chemoimmunotherapy approach, whereas the patients who are unmutated are going to clearly benefit more from the novel targeted agent approach.
I think that there is debate, frank debate within the field, even amongst those who know the disease best, as to whether or not those patients with the mutated disease should get ibrutinib [Imbruvica] or chemoimmunotherapy, and I think that in a lot of cases, that would be subject to patient preference.What are the most recent treatment advances in multiple myeloma?
This year’s [American Society of Hematology Annual Meeting & Exposition had] many significant advances in multiple myeloma. Everybody will hear a lot about CAR [chimeric antigen receptor] T-cell therapy, but unfortunately, we’re not curing anybody with that and the responses are lasting sometimes long, sometimes short. Another big focus will be the bispecific antibodies, or the BiTE molecules. Several of these are in development, early clinical trials.
Probably the hottest target of the whole meeting will be the BCMA target in multiple myeloma, which is the B-cell maturation antigen. So, there will also be many presentations on agents directed for BCMA, some monoclonal antibodies, some antibody-drug conjugates. So, it’s exciting, in terms of the advances that we’ve made.
In terms of new drug approvals, it’s not quite the banner year it was a year ago, but [there’re] still many new agents entering the market and many exciting things to come.
How are these advances impacting clinical outcomes?
So, a lot of these are really new. In most of the trials, we barely know progression-free survival. There will be some early overall survival data being presented with new chemotherapy regimens, but we really have to wait and see the impact. I think it will also be extraordinarily important to control cost as all of these new agents come out.
There was one oral presentation on a quadruple combination of therapy, which works very, very nicely, but there was little, if any, mention of the financial impact on the patient and the healthcare system. So, in multiple myeloma, we’re going to have to get more responsible as all of these new agents march quickly to the forefront.How important is it to engage patients with Hodgkin lymphoma so that they understand their diagnosis and their treatment options?
I think it’s incredibly important that patients understand their diagnosis and what the treatment options are and what their prognosis is. I think that the treatment options that we have to offer patients are often not straightforward and not black-and-white where there’s one right or wrong answer. Some of it really has to depend upon the patient’s own values and their own choices.
One example of that is when we are offering patients with early-stage Hodgkin lymphoma a treatment approach where we’re going to consider either combined modality therapy, where we include radiation therapy, or chemotherapy alone. That is an important discussion to have with the patients, because the cure rate with chemotherapy alone for early-stage Hodgkin lymphoma is not as high as when we use combined modality therapy. But by eliminating the radiation, we are reducing long-term toxicity. But there’s a small group of patients who are going to relapse and then going to need much more intense chemotherapy and a stem cell transplant in order to cure their disease at that point.
So, even though the majority of patients still will be cured with chemotherapy alone, it’s an important discussion to have with the patients to make sure they understand what they’re potentially giving up if they’re not getting radiation earlier on in their treatment.
I think it’s perfectly reasonable to avoid radiation for many, many patients, but I think we have to be informing patients why we’re making this choice and have them be a part of that choice as well.
Are there any disparities in survival among patients with Hodkgin lymphoma?
There are disparities with regard to survival. Patients who are over 60 [who have] Hodgkin lymphoma have much lower survival than patients who are under 60, and that has been found partly because the biology of the disease may be a little different. But the major reason is because of reduced tolerability to treatment and more toxicity related to treatment and likely also due to more comorbidities.
So, our treatment for patients who are over 60 tends to be modified a little bit. One of the things we really try to be careful with [in] those patients is exposing them to bleomycin because they do have a higher chance of having bleomycin lung toxicity. So, one approach could be to try to still expose them to bleomycin but try and reduce it as much as possible, which is possible if we use the approach where patients receive 2 cycles of regular ABVD [doxorubicin (Adriamycin), bleomycin, vinblastine, and dacarbazine] chemotherapy. And if they have a PET [positron emission tomography]—negative response, then we drop bleomycin for further treatment. For some patients, we really want to try to avoid bleomycin all together. It’s not unreasonable to give them just AVD [doxorubicin, vinblastine, and dacarbazine] chemotherapy without the bleomycin. It’s probably associated with a less favorable outcome but still could be curable.How at risk are patients with Down syndrome for developing leukemia compared with the general population?
That very much depends on the age. So, for example, in a lifetime, say up to age 60, there’s about a 12-fold increased risk of acute myeloid leukemia compared with individuals without Down syndrome and a 13-fold increase of acute lymphoblastic leukemia. But if you’re a young child, the statistics are a lot different. So, the risk of a myeloid leukemia in children under 4, for example, climbs to 114-fold compared with other young children who don’t have Down syndrome.
What’s the reason behind this risk?
That’s a very good question, and we don’t fully know the answer, and a lot of people are researching on that. The assumption is that it has something to do with the extra chromosome in the blood cells. So, having an extra chromosome, 21, in some way affects the behavior of the blood cell, which makes it more likely to transform. But which genes do it, we still don’t know.Does genomic sequencing play a role in determining prognosis and treatment for patients with chronic lymphocytic leukemia (CLL)?
Certainly, certain genomic prognostic markers are quite important in CLL. Historically, we’ve used chromosome abnormalities as determined by FISH [fluorescence in situ hybridization], with the highest-risk abnormality being the deletion of 17p. Now, that’s often accompanied by mutation of the TP53 gene, which we now recognize [as] being similarly adverse to 17p deletion. One thing that was remarkable about venetoclax [Venclexta] and rituximab [Rituxan] is that there was no difference in progression-free survival [PFS] at 2 or 3 years based on 17p deletion.
That’s actually even better than the BTK [Bruton tyrosine kinase] inhibitors, where we actually do see that PFS is a bit shorter in the patients with that very high-risk marker. There are other gene mutations in CLL, as well. NOTCH1 is one that we worry about associated with Richter transformation, SF3B1, and ATM. And then [there is] a long list of less frequent mutations. We don’t really know yet what the impact of those are in the context of novel agent therapy, and that’s something that we’re all very interested still in studying.