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.
When you started Susan G. Komen, there was a lot of stigma surrounding the subject of breast cancer. How did you bring the conversation about breast cancer into the open?
Well, we really brought it out into the open because it was there, it was growing. People just didn’t have the place to talk about it or share their feelings or share their insights or survival or their treatment—or their lack of survival. It was really waiting to happen. Because I had lost my only sister to the disease in 1980 and she asked me, I promised her that I would help cure breast cancer.
So, it’s been a long ride, a long time. It’s been almost 40 years of my life—she was diagnosed in 1978—and I realized that we had to do something, and it was going to require a movement. It wasn’t just going to be a one-[time] fundraiser, it was a movement. And of course, this pre-dated computers and cellphones and fax machines and everything else. We really only had telephones and each other to work with.
So, we named it Susan G. Komen in her honor, her memory, and set about working for many, many years. We’ve managed to bring up the awareness, the sharing, and we’ve created a large community—a worldwide community—focused on breast cancer.
What do you see as the future of the Susan G. Komen Foundation?
What I think is that the organization, every organization goes through major changes and rebuilding and new ideas and innovation. It’s time for major innovation.
We have such strength in our affiliate communities, who are really people on the ground that have learned so much. It’s not so much what we do, it’s how we do it that’s going to be very important.
People are very sensitive to high overheads, they’re very sensitive to any- thing spent. They want to see money that they give go directly to a mission. So we’re going to try very hard to make sure that we have collaborations with other people already doing the same thing, that we do things in a way—as we’ve always tried to do—as cost-effective as possible. Events become very ex- pensive to do, so we have to figure out other ways [to use] technology to raise money and make sure that we connect with people.
[We also need to] get people of wealth in the country to understand that it’s as important to fund a program as it is a building. In fact, sometimes the program goes farther to train people, to bring people in the system. So those are the sorts of things we’ll work on.What has your research shown about different methods to prevent ovarian cancer?
[My] presentation [was] about ovary cancer and the current updates. It is the case that ovary cancer happens in about 1 [in] 70 women in their lifetime. [There are] people that are at higher genetic risk for that, but really, it’s been a disappointing number of decades trying to identify a screening test. So part of the presentation is just kind of presenting the data [that are] out there, showing that it’s not all that helpful.
There are some methods of preventing ovarian cancer, such as removal of the ovary. It’s [known as] prophylactic surgery, [and it’s meant] to reduce the risk. However, that comes with menopausal, early menopause, and other health consequences. What has been more recently shown is that many of these “ovarian cancers” actually start in the tube, and the tube has no function other than allowing pregnancy.
So, what has tilted in the last couple of years—and the ACOG [American College of Obstetrics and Gynecology] practice bulletins sort of magnify—[is] the recommendation to remove the tube at the time of GYN surgery or instead of tubal ligation. We think that that is one of the more effective ways, and easy ways, of decreasing ovarian cancer in the United States.
What were the findings of your study that investigated the delivery of neoadjuvant chemotherapy to women with advanced ovarian cancer?
One of the conundrums, and controversies, in the field is people [who] present with ovarian cancer. Two-thirds of the time it’s fairly advanced. Whether to start with surgery first or chemotherapy first has been the dilemma. Historically, people have started with surgery first, and yet over the last decade or so, there’s been a lot more evidence that starting with chemotherapy— such as neoadjuvant chemotherapy, which means chemo first—seems to work out just as well in the long run in many cases.
Part of our work was looking at that trend and then determining that [to be] an interval operation, which means halfway through the chemotherapy. If you’re able to do a minimally invasive operation, like a standard laparoscopy, people have similar outcomes to open surgery and a lot fewer side effects and a lot less complication risk.
In your retrospective analysis, how were palliative care services utilized in ovarian cancer and was the utilization in line with national guidelines?
Yes, we did look at palliative care practices specifically for ovarian cancer. It is a relevant topic in that field in that even though most ovarian cancer patients will go into remission, 80% to 90% will relapse. And when a woman has relapsed, it is at some point a palliative care discussion.
Yet, the logistics of care in the country means that many people never hear the word “hospice.” If they do, it’s at the very end—like the last days or hours of their life—and yet there’s a tremendous amount of expense that goes into the last 6 or 12 months on a fairly predictable end result, specifically for ovarian cancer.
So, what we were trying to do is just identify where the deficiencies were so that more attention could be brought to that and I think that more people would be thinking about that earlier.
What sort of disparities did your research uncover about the receipt of care for high-grade endometrial cancer?
High-grade endometrial cancer is sort of the atypical version of endometrial cancer. Endometrial cancer has doubled in incidence in the United States in the last 20 years. The reason for that is the obesity epidemic, by and large, but most of those people are cured. It’s the people with the high-grade, or so-called type 2, endometrial cancers [who] require something other than surgery and have a much higher relapse rate.
So, even though it’s 5% of people being diagnosed, it accounts for more than half of people that die from endometrial cancer. It seems to be shifted in the minority and nonclassic populations.
For example, black women are more at risk for fibroids. They’re also more at risk for some of these type 2 or high-grade endometrial cancers. So, it’s looking at the different health disparities, and you would think that if there [are]
100 people with the same diagnosis that they would be treated the same way, [patients with] one of these more aggressive types of tumors. But that’s not the case. And so, there are certain ethnic groups or race groups that tend to be undertreated for some reason or another.
Sometimes it’s related to their insurance status, sometimes it’s related to different philosophy, but that’s what we were trying to drill down into.Why is genetic counseling so important after a patient is diagnosed with ovarian cancer?
Thank you for asking this question. I think this is really, really important. A lot of patients and providers don’t realize that ovarian cancer has a huge genetic component as far as the causes of it. About 20% of women who are diagnosed with an ovarian cancer actually have a gene mutation that caused it. When I say “gene mutation,” I mean something that they inherited, something that they could potentially pass on to their children or be sharing with a sibling.
Ovarian cancer is one of the most lethal cancers that a woman can be diagnosed with. In fact, 75% of the time it’s diagnosed at an advanced stage. We don’t have a screening test for it, so picking it up early is really by luck or by chance. In women who are diagnosed with advanced stage, the likelihood of them being alive at 5 years is around 60%. Again, that is because this is a highly lethal disease.
Women who are diagnosed with an ovarian cancer should be seen by a genetic counselor. Basically, what will happen when they’re seen by a genetic counselor is that the genetic counselor will go through their family history, personal history, and look for other signs that may indicate a genetic mutation that caused the cancer.
So, [because] ovarian cancer tends to travel with breast cancer, it’s important for women to know that a strong family history of breast and/ or ovarian cancer should prompt genetic counseling. The National Comprehensive Cancer Network [NCCN] basically has guidelines for when women should be referred for genetic counseling. A personal diagnosis of ovarian cancer is all you need to be recommended for genetic counseling.
The Society of Gynecologic Oncology also has a statement supporting the fact that women should be referred for genetic counseling if they have this diagnosis. At Mayo Clinic, in 2015, we looked at our genetic counseling referral rates for women with [an] ovarian cancer diagnosis. It was a quality improvement project, actually, and we defined what our current rate was, and it was 20%. Which really, we weren’t compliant, then, essentially with NCCN guidelines. And so, we implemented a bundled approach that included patient education, order sets. and referral guideline implementation into the electronic medical record for each woman diagnosed with ovarian cancer.
Patient education [was] not just through check lists of risks factors but also education in the form of a short video that they watched in the hospital after their surgery. What we were able to do is we were able to increase our rates of referral from our pre- or historic, referral rate, which by the time we implemented the project, we actually moved the needle a little bit already. We were at 40% referral. But with our quality improvement project, we actually went up to 77% referral rate.
So, we’re continuing to try to refine that because, ideally, we do want to chase 100% because all of these women really should be seen for genetic counseling and offered genetic testing.
What sort of novel approaches did you research to assist in the early detection of endometrial cancer?
One of the projects that I’ve been working on from the research side of things, since I came on staff at Mayo Clinic, is the development of an early detection test for endometrial cancer.
In gynecology, we’ve had huge success when it comes to screening for cervical cancer with the Pap test and Pap + HPV [human papillomavirus], but we do not have a screening test for endometrial cancer. Endometrial cancer is now the most common gynecologic malignancy that we care for in the United States. One in 50 women will develop an endometrial cancer, and when they come in to see me, oftentimes the question that I get asked is, “My Pap smear was normal. How could I have cancer and have a normal Pap smear?”
Well, the Pap smear is not a test that is designed to pick up endometrial cancer. In fact, the sensitivity of a Pap test to pick up abnormal cells that indicate endometrial cancer up inside the uterus is only about 30%. But we know that there are molecular markers that are shed from cancers into other biospecimens, whether it’s peripheral blood or other downstream biospecimens.
So, the biospecimen that we have been focusing our research on is that of the vaginal pool. The vaginal pool is defined as everything that comes from up in the female reproductive tract and flows down into the vagina. This could
be fluid and material that comes from the fallopian tubes, maybe even the ovaries. [There are] actually some data out there that suggest that mutations that arise in ovarian cancer can actually be detected in samples from the vagina and cervix, so we know that there is some downstream targeting of that. Also, fluid from the endometrium cavity, the cervix and the vagina. So, all of that is what’s in the vaginal pool.
What we’ve been doing is trying to design an early detection test using a tampon as the tool for collecting that biospecimen. In that biospecimen, we look for molecular markers such as methylation or mutations. Similar to the most recent paradigm shift in colorectal cancer screening, which is that of the stool-based test called Cologuard, we aim to develop a multipanel/ multimarker test that will pick up endometrial cancer or endometrial cancer precursors. [This would] allow women to have the diagnosis earlier, ideally, and allow better access to care, because this would be a test or a specimen that could be collected from the comfort of our patient’s own bathroom.What role does precision medicine currently play in the community setting?
I just got done giving a talk with Lora Jane Black, [RN, MPH, OCN, CCRP, Sanford Research], and Edward S. Kim, [MD, Levine Cancer Institute, Atrium Health], about community oncology and precision medicine. As anywhere, most patients are treated in the community and understanding precision medicine is important.
There are some unique barriers, including geographic access, infrastructure, and things like that. But every physician needs to know about precision medicine, as it’s becoming both standard of care as well as [an] emerging target that we can try to work on and [is] being discussed widely at ASCO. So, this is important information for every oncologist, but it’s increasingly becoming important in the community setting.
Does this role differ from that in an academic medical center?
Many of the issues are the same, whether you’re at an academic medical center or at the community sites. Some community sites may have people [who] treat multiple tumor types and aren’t specializing in 1 area, where the information is exploding in every area of cancer. There may be a greater need for a systemized, centralized way to approach all [of] this molecular information versus if you’re only doing 1 cancer. If you’re only doing brain cancers, you may know the mutations that are most important and the information more than someone who’s treating every type of cancer available. That’s why we have been talking about how do you set up a whole system to track that.
But even in the university settings, that’s still a need. One of the limitations is actually genetic counselors. There’s not enough to go around, and that’s one of the huge areas of need, including doing telemedicine or having clinical trials that bring in genetic counseling for around the whole country.
What are some novel therapies being used in the treatment of hematologic malignancies?
So, there are many examples of this. BCR-ABL is the prototypic example of precision medicine used for chronic myeloid leukemia [CML] with tyrosine kinase inhibitors, or TKIs. Acute myeloid leukemia [AML], which had not seen a lot of new therapies for decades, now has targeted therapies versus Flt3, versus IDH2, and we’re developing therapies for acute lymphoblastic leukemia [ALL], including Philadelphia-positive, just like CML.
And then in multiple myeloma, there have been limited reports of BRAF inhibition in myeloma. T(11;14) appears to be a marker for BCL-2 inhibition with venetoclax [Venclexta].
What biomarkers have you identified for deciding on a treatment approach for hematologic malignancies?
The biomarkers for treatment can either be prognostic, which we have a lot of biomarkers, or they can be predictive, which actually tells us that a drug is more likely to work in that area. My colleague has studied t(11;14) as a predictive biomarker in myeloma for effective venetoclax, and then we talked about BCR-ABL for Philadelphia- positive ALL for CML, and there are other emerging ones, so IDH2, Flt3 in AML.
Increasingly, people are trying to find other markers, which may be targets, and it’s somewhat theoretical sometimes, so FGFR in various hematologic malignancies. There’s JAK2 inhibition, but other JAK-related pathways, so there’s a lot of emerging information. Some of these are in standard of practice. Some are almost becoming standard. And then others are still in the research category of trying to discover and implement new targeted therapies.What challenges are associated with trying to pursue precision medicine?
There are multiple challenges. We’ll start with the clinician. So, when the physician comes in to see the patient, [there’re] so much data [that are] out there, and depending on the environment the physician is seeing the patient in, [that] also dictates some of this. A physician [who] has to know every single tumor type [is not] going to be able to keep up with the massive data and the studies and everything that’s coming out for every single tumor type. So, keeping up with the data, integrating it, being able to actually teach the patient about it, and implement[ing] it is one of the great challenges that is out there.
Other challenges are who should we be testing with this, what test should we be doing, and keeping the physician up-to-date with that. [On] the patient’s end of things, the willingness to do some of the precision techniques, because they do take time, and also their expectations may be somewhat unrealistic. Things that you have to worry about [are] that the patient says, “Oh great, I have this mutation. I should be responding to X therapy,” or perhaps, “My tumor proportion score of PD-L1 is 80. I should have a great response to one of the checkpoint inhibitors,” and in fact, they don’t.
Studies have shown, with most of these agents, response rates are somewhere in about [the] 50th percentile, maybe 54th percentile. A couple actually go up into the 60 to 70th percentile. But not everybody who should respond is responding, so the expectations can be unrealistic on both the clinicians and the patient standpoint.
As far as testing, that’s a whole other ball of wax. You know, we start out with, how do we identify the patients [who] should be getting testing and how do we identify, of those patients, who should be getting treatment and who shouldn’t. One of the early examples of this is the Oncotype DX, which is used in breast cancer to determine patients, who have undergone surgery with estrogen- receptor positivity, who should be getting adjuvant chemotherapy. Well, that’s only 1 example. How do we pick out other patients who should be being treated and who shouldn’t be being treated?
Another example would be: How do we pick out who is going to respond. What test should we be doing? Should they be immunohistochemistry? Should they be fluorescence in situ hybridization—type testing? Should it be something along the lines of next-generation sequencing? Nobody really knows the answer to that, and there are standardization issues. Other issues that occur are: What toxicity tests should we be doing?
Other things could include who should be being tested for germline mutations. We usually [decide on] that if a patient comes in and has many of their family members affected by cancer. But those are the only patients that, right now, we’re recommending have germline testing. Are there are other families that we are missing? Other germline mutations we really don’t know about yet?
So, I think, overall, it’s a very complicated situation, and really determining which patients should get what tests and as to what patients should get what tests is complicated.
Now, when we get to treatment, what are the actual treatments we should be having? Some of it has been well worked out. A lot of it hasn’t been. The other thing is, we need to have well-developed molecular marker-driven clinical studies for patients to be enrolled in.
The other things are the master protocols and getting patients to participate in that. And then what about those that are nonresponders? What do we do about them? How do we figure out what the heck is going on with them? So, overall, it’s a very complicated and complex problem.
How have novel therapies changed the cancer treatment landscape?
The premise of precision medicine, first and foremost, is to match the right patient with the right treatment at the right time. Immune therapies, or checkpoint inhibitors, have been very successful in a subset of patient with multiple different types of malignancies and has changed their trajectory tremendously.
For example, melanoma has had a very high success rate with the immunologic therapies. As we all know, it’s where we started out in. President Jimmy Carter had actually gotten an experimental checkpoint inhibitor for his melanoma, which had traveled to his brain. He has been disease free for a number of years, potentially even cured, which doesn’t happen often. But it can happen with these therapies.
Additionally, there are [patients with] lung cancer who have also benefitted tremendously form these therapies. Case-in-point: [There] are patients who have had [a] greater than 50% tumor proportion score of PD-L1. These particular patients tend to have a very strong and lasting response to these drugs. However, it’s only about half of these patients who have these parameters.
I think [there are] a lot of questions that still remain unanswered with them. I think they’re great therapies, they’re here to stay, and they do benefit some of our patients. But my questions still remain: What is the proper sequencing of many of these agents? What should we be doubling them up with? Should we be doubling them up with anything? And why are there patients who, in all rightfulness, should be responding to them and are not? Is it because of their innate immune systems? Is it because of other underlying comorbidities? Is it due to other problems within the microenvironment of these tumors? Even their microbiome, that remains pretty unclear at this time, and I think there’s a lot of work to be done.How has precision medicine changed the information needs for oncologists and tumor boards?
So, I talked about 5 “rights” of precision medicine today: the right diagnosis, the right test that needed to be made to make a diagnosis, the right targets, the right treatment, and the right monitoring. That all goes along with this. But along these rights for the right treatment for precision medicine are the data sets that underlie it. So, you need prognostic biomarkers, diagnostic biomarkers, monitoring biomarkers.
You also need guidelines in terms of what are the appropriate tests to run in the first place. Precision medicine has really opened up the world of biomarkers to oncologists, and that is, I think, one of the big paradigm shifts.
How does health information technology play a role in obtaining the information needed to make an informed decision for a patient?
In order for precision oncology to be fruitful and to be effective, we need interoperability and we need to be able to share patient data, because the more data that we have that we can aggregate together, the better the better the quality of the predictions we can make. So, predicting for a very small set of patients is going to be prone to error, but if we have a very large set of patients, we’re going to be able to make better predictions for who might respond to therapy.
So, in the era of electronic health records, I don’t think we’re quite there yet, but there’s a lot of work that needs to be done from a harmonization standpoint, from an interoperability standpoint, as well as [simple] data standardization. One of the issues that came up today during the talk was that we don’t have a common nomenclature on how to capture genes or gene alterations. These standards are being developed and they’re starting to be implemented. But they’re starting to be implemented. We’re not quite there yet from an interoperability point of view.
How has the shift to precision medicine changed the way oncologists think about cancer treatment?
We still need to see the patient, diagnose the patient, figure out what’s targetable, and find treatments for the patients. So, that part is still the same and that will probably not change. What has really changed is that precision medicine adds a layer of data to each of those steps, and each of those steps now require a lot more data synthesis. So, in effect, what we have is a data problem. [It’s] less about a clinical problem. But how do you manage all this data to help treat your patient?
What is the difference between personalized medicine and precision medicine?
There is a shift from personalized medicine to precision medicine, because if you really think about it, we’re really trying to customize medications for a group of patients with a similar feature. For example, if we think about BRCA1 or BRCA2 alterations, we think about PARP inhibitors. We’re not talking about a particular patient’s cancer with that alteration. We’re talking about a group of patients who all have that particular gene loss or gene alteration.
I like to think of this as: If you were to try to start a clothing store, for example, if you tried to customize clothing for every single person who came into the store, that’s not really feasible or possible with what we have today. Instead, what we’re trying to do is find what are the most common alterations, what are the most targetable ones, and create treatments or therapies that really do fit these groups of people.
So, when we talk about precision medicine, we’re moving to trying to treat similar groups of patients.How have health technologies helped enhance high-quality cancer care and improve outcomes for patients?
So, the great challenge we’ve always faced is how do we keep pace with the new technologies that are coming down the pipe? Because we’re getting so good at developing new technologies, whether it’s informatics or it’s precision medicine or new drug therapies or diagnostics. Our ability to create these new tools is much faster than our ability to figure out what to do with it, because when you want to measure outcomes improvement, it requires acquisition of the technology, which has certain expense to it. It requires learning how to use it properly, adapting it to its use, and then measuring the outcomes.
There may be many intermediary steps between the technology and actually the outcomes measures, so it’s a real difficult question to answer: How have these improved our outcomes? It’s really that lack of ability to measure adequately that’s slowing us down more than anything else at this time.
Having said that, I think we are seeing real progress with the use of information technology. I think, without question, if you ask most physicians would they rather go back to paper charts, the answer would probably be “no.” If you asked patients, would you like to go back to the age where you could not get a patient portal? Find out your lab tests? I mean you always want more, but would you want to go back to the point where you had nothing? People would say, “No, that’s not a good idea either.”
So, I think that we are frustrated with how slow it goes, but that tends to make us not realize how far we’ve gone and where we used to be just a short while ago.
How is next-generation sequencing changing the landscape for cancer care?
Next-generation sequencings tests are just starting to change the landscape for [patients with] cancer. We’ve had quite a bit of FDA and CMS regulations just in the first half of this year, which have begun to lay down the road map for how the FDA will look at laboratory diagnostic tests based on genomics and how [the agency] will go about ascertaining the quality of [those tests] and how CMS is starting to think about how they will decide when they will pay for these tests. So, the fact that we have the 2 major regulatory agencies in the United States starting to set policy is a big step forward.
However, the payment policy issued by the National Coverage Decision by CMS a couple of months ago [March 16, 2018] is extremely narrow. It essentially, as I interpret it, limits it to when there is a specific genomic biomarker match to a specific drug that’s been FDA approved. In that circumstance, combined with a patient with advanced cancer, there will be payment for it. But it’s still a very, very limited role.
The alternative is that we rely on traditional biomarker testing, which is either a series of sequential tests that consume a lot of sample—and very often you may run out of sample before you’ve done all your testing—or some limited panel, which is increasingly being looked at as both an inefficient and expensive way of doing genomic testing.
So, I think that it’s good news that the FDA and CMS are recognizing that requiring a specific companion diagnostic test for every single drug that is precision medicine-based is no longer the way to go. But we’re not quite there yet.