Comprehensive Geriatric Assessments in Urologic Cancer Care: Identifying Modifiable Risk Factors for Personalized Treatment - Sarah Psutka
August 8, 2023
Sarah Psutka, MD, MSc, Urologic Oncologist, Associate Professor of Urology, Department of Urology, University of Washington, Seattle, WA
Sam S. Chang, MD, MBA, Patricia and Rodes Hart Endowed Chair of Urologic Surgery Professor Department of Urology at Vanderbilt University Medical Center
Sam Chang: Hello, everyone. My name is Sam Chang. I'm a urologist in Nashville, Tennessee. And we have the great honor of having Dr. Sarah Psutka, who's an Associate Professor of Urology at the University of Washington and the Fred Hutchinson Cancer Research Institute. I think I'm close to that, the Fred Hutch, or the Hutch, as they commonly say amongst those that are in the know.
I've known Sarah ever since she was applying for an oncology fellowship. And she has had a meteor like rise in terms of urologic oncology, in multiple different areas in urologic oncology. I wax poetic because I just want to make sure that people understand her presence and her importance in our field at such an early stage in her career. But she is going to actually be focusing on an area that I know has been one of her passions, actually for years, looking at evaluation of patients, their determination of frailty and what we can do perhaps to better prepare patients for their treatments as they go through urologic oncology, major type procedures. And so Sarah, thank you so much for joining us and look forward to your presentation and a discussion afterwards.
Sarah Psutka: Thanks so much, Sam. I am thrilled to be here. And indeed this is something I'm so excited about. It's great to have the opportunity to talk with you a little bit about it. When I was thinking about this, I really want to just in the next couple of minutes highlight what I think are some key concepts in preoperative risk stratification and taking that from a very subjective practice that I think we all strive to be really good at. But a lot of times we sort of lack tools to do it reproducibly and in a fashion that's validated and objective. And then think a little bit about how we can quantify frailty and then ultimately move towards understanding once we know about a patient's innate frailty across different domains, how we can use that to define targets for prehabilitation or rehabilitation to optimize outcomes.
When I think about the timelines of risks and windows of opportunity for outcome improvement in our field, patients are diagnosed with a cancer, they may receive some treatment before they come to us for surgery. Certainly, there are things that we can do around the time of surgery to improve outcomes such as utilize enhanced recovery after surgery protocols. There's certainly opportunities in the survivorship period where we can really improve patient's quality of life as they're recovering from treatment and living their life after treatment.
But where I get really excited about thinking about the things that we can do that will ultimately move the needle and change how patients do, I think there's tremendous opportunity in the realm of prehabilitation and then also in how we risk stratify patients and ultimately elect treatments appropriately. And if you think about it, inaccurate risk stratification has pretty substantial ramifications. If we don't do a good job of risk stratifying patients, we could either under treat patients and basically patients who need treatment don't get the surgery or they get the surgery that's not the optimal surgery they should have gotten. They might get ineffectual or inappropriate or no systemic therapy when they really would benefit from it.
And ultimately, we end up having the risk of excess cancer specific morbidity and mortality. Now the converse, of course, is if we over treat patients and we put patients through operations that perhaps they're not great candidates for, there's the risk of excess morbidity and mortality, potentially little survival benefit because of the potential mortality risks associated with our treatment. And really importantly, there's the potential for loss of independence and functional status and decrements to quality of life.
When I think about novel tools that can really help us in this risk stratification process, I've been spending a lot of time over the last several years now, since pretty much since I was in my fellowship, studying frailty. And frailty goes and can be sort of defined in multiple different contexts. But what I really want to sort of highlight with frailties is this is a medical syndrome. It's got lots of different things that ultimately can contribute to it. It can be reflective of cancer and oncologic processes, metabolic changes, patients' genetics, their other competing comorbidities, their body composition. But how it manifests is its reductions in strength, reductions in endurance and reduction in physiologic function. And it increases vulnerability for bad outcomes.
And we know that in urologic oncology, there's a lot of data across cancers that shows that frailties associated with adverse outcomes such as complications, extended length of stay, not being able to get home, high cost of care, cancer specific and oncologic mortality. It's highly, highly predictive of adverse outcomes if a patient is frail. There's a lot of different ways to measure it. And one thing that's really exciting about frailty is that it is a dynamic risk factor that we can measure at different times and we can watch it, actually patients if they improve, potentially become more frail, but then we can get patients back to their pretreatment status.
And it's something that we can measure in a validated way across time points in the continuum of care for patients. Something I've been really excited about and have spent quite a bit of time working on since about 2013 has been an anatomic assessment of that frailty. And one of the terms that we used to describe this is sarcopenia, which is a severe paucity in muscle mass. I think this term has really entered the urologic lexicon maybe in the past eight to 10 years largely due to sort of transmission over from the medical oncology literature and the geriatrics literature where we realize that muscle wasting is a really bad prognostic factor. It's associated with all kinds of adverse outcomes in different medical conditions, surgical conditions. There's great data in burns and trauma that sarcopenia is associated with adverse outcome and it's really prognostic in the geriatric syndrome.
We all lose muscle mass as we age. The problem is in many of these medical conditions and especially in cancer, that muscle mass loss is accelerated. And interestingly, a lot of our treatments can accelerate it further. And the less muscle mass we have, it's clearly a manifestation of that frailty syndrome. And it's strongly associated with complications. And here's some data that we developed at Mayo Clinic, which shows that it's associated with a nearly twofold increase in risk of death, both overall death or all-cause mortality as well as cancer specific death in patients undergoing radical cystectomy. And interestingly, this is a highly prevalent risk factor in many of our cancer populations. For example, in bladder cancer, nearly 70% of patients undergoing radical cystectomy are sarcopenic. These are important factors. This is just a summary table that comes from a review that we did a few years ago. That shows that not only is sarcopenia, however you measure it, associated with oncologic outcomes, but also with treatment specific outcomes such as complications.
Interestingly, muscle mass and body composition is also a dynamic risk factor and it's something that we can measure across the continuum of care. And this is some early data, some pilot work, in the black figure, you see up in the upper right hand corner. And then, the figures on the bottom show a much more robust series where we looked at patients undergoing neoadjuvant chemotherapy before radical cystectomy. And we've measured how their muscle mass, fat mass, both visceral fat and subcutaneous mass change before they get chemotherapy and after cisplatin based chemotherapy. The important thing to know here, is we're talking about a time difference of only about an average of between 90 and 110 days. In a relatively short period of time, usually around three months, the average patient with bladder cancer is going to lose about 6.4% of their muscle mass.
Now, if you think about the fact that these patients are already muscularly deplete in nearly 70% and meet these criteria for severe muscle deficiency, we're increasing their risk profile potentially even more substantially. And we noted that especially in our University of Washington cohort, those pre and post chemotherapy skeletal muscle index changes are associated with increased risk of cancer specific death and all-cause death, on multi-variable analysis. There's an independent association in that loss of muscle mass. We have this window of time before surgery where we actually may be further increasing risk. And it's important that we acknowledge that.
One other way that I've been getting interested in studying frailty is using comprehensive geriatric assessments. Now, this is not a novel concept. CGAs are well studied, validated assessments that basically look at these vulnerabilities that are exemplified by frailty and sarcopenia across multiple domains. They're widely used in geriatrics and they're actually part of the guidelines and in terms of how we care for patients with urologic malignancies. But these are multidisciplinary and multidimensional evaluations of a patient's health that ultimately identify potentially modifiable or subtle risk factors that can improve outcomes.
We know that a lot of the ways that we traditionally, conventionally measure risk, chronologic age, comorbidity sort of indices or lists of comorbidities, just don't give us enough information about a patient's innate risk. Using a CGA, we can assess a patient across the domains of function and mobility, the degree to which comorbidities affect their daily living, their mental health and cognition and their social supports and networks, in ways that are validated and objective, that directly identifies specific vulnerabilities that then we can target, potentially improve and improve outcomes. And this gives us a much better sense of things like physiologic age.
And what's really important about CGAs is they're extremely specific and sensitive in patients who have a normal ECCOG performance status or a normal ASA score. If we look at older patients with cancer who have an ECCOG score of 0 to 1, up to half of those patients actually need assistance with their activities of daily living, but that's not captured in these traditional scores that we use. These CGAs are much more sensitive for picking up these vulnerabilities that are really important for us as surgeons to know about. And they're very strongly associated with adverse outcomes after cancer surgery, including things like ICU admission and mortality and importantly discharge to places other than home.
When we use them in practice, we can reduce these adverse outcomes. And importantly, they're associated with outcomes that are really important to patients, like being able to stay in their home after surgery and return to their home and return to the type of life they were living before they underwent care.
We started studying these at the University of Washington two years ago. We ran a prospective study in our multidisciplinary bladder cancer clinic where we enacted a CGA in our clinic using validated measures across these different domains and then collected a whole battery of outcomes, not only looking at clinical outcomes, but also looking at things like quality of life, decisional regret and decisional conflict. We're working on through the analysis on this now, but one thing that was really important about this study is that we demonstrated that you can actually do this in a busy urology practice. We were able to gather all of this data in a relatively short period of time. On average, it took about 18 minutes to collect this data in a urology multidisciplinary bladder cancer clinic. And what's important also is that it wasn't that burdensome for patients. And we were able to do this in the setting of the COVID pandemic and sort of had to pivot midway through our study and actually do a lot of these over Zoom and we were still able to get most of the data.
The other thing that's important is we actually identified a number of potentially actionable vulnerabilities, including a relatively high burden of mental health concerns, malnutrition and low functionality that ultimately allowed us to get people to physical medicine and rehab, to psychiatric services and also to nutritional services before they underwent chemo and surgery. We're building this out now with a NIA-sponsored study that we're just about to take off to launch in the coming weeks. And what we're going to perform and over the next two years, we're going to undertake a longitudinal prospective assessment where we're going to use an even more, easy to utilize, GA or geriatric assessment in our clinic. We're going to use a self-reported assessment that's widely validated in the medical oncology literature called the Cancer and Aging Resilience Evaluation. We're going to augment this with studies of body composition and we work with a lab at Harvard that uses a machine learning based platform, so it's artificial intelligence that allows us to use CT scans that we get as oncologists in normal cancer care.
And we're going to be able to measure patient's body composition and their frailty, basically across the episodes of care as they're cared for their muscle invasive bladder cancer. And importantly, we're going to be investigating not only things like the adverse events, systemic therapy of associated adverse events, surgical complications and oncologic outcomes, but we're going to start to quantify the degree to which the treatments that we use in muscle invasive bladder cancer, are associated with functional decline. And try to identify predictors, using things like frailty, using these measures of frailty and measures of muscle mass, to try to understand who we can predict might suffer greater functional decline with these different treatments to ultimately help patients select treatments that are more in line with their personal priorities and their goals of care.
And then of course, the most important thing is once we identify these vulnerabilities, these risks with our CGAs, with our body composition measurements, we can then enact really specific, personalized prescriptions, that can be used for pre rehabilitation or rehabilitation in concert with our therapies. And these are just some examples. If we identify vulnerabilities in physical function, we can get patients into PMNR and we have these wonderful colleagues who work at our cancer center who are cancer rehab specialists, who can give personalized prescriptions for example, prior to cystectomy, to strengthen core muscles, build balance and stability when ambulating to help patients undergoing abdominal surgery get back on their feet faster.
Prehabilitation is an area of active interest. I was looking on PubMed, you can see the number of publications that have just skyrocketed, that are looking at prehabilitation interventions in especially cancer care over the past 20 years. And prehab really focuses on these core domains of improving physical function, mental function and nutritional sort of robustness in patients who are undergoing treatment for other major medical issues.
This is a concept that we're really getting into, but of course right now what we're focusing on, at least in my lab, is building the foundational understanding of what are the key targets for prehab that are important for our patients with urologic malignancies. I just gave you some examples here of a couple of the many different prehab programs that are out there, that are really these multi-domain, personalized programs, that are meant to mitigate surgical risk by doing things like kicking bad habits such as smoking and alcohol use, optimizing nutrition, optimizing physical function in the time leading up to surgery. And especially now that we know, for example in bladder cancer patients that are getting chemotherapy, if we know we're going to lose muscle during that time period, this is a great time period.
This is a huge win of opportunity we think, to be thinking about what nutritional supplementation or nutritional programs and exercise programs we can be asking patients to use that are pragmatic, scalable, low cost and easy to use. They can hopefully mitigate some of those losses and hopefully bring people into surgery a little bit less frail than they might've been, had we not done those things.
Just in summary, this is obviously an area that I'm excited about and an area that a lot of people are really focusing on is an active area of research right now, which is really exciting. But I think the goal of all of this is really to take the traditional sort of eyeball test, this assessment of fitness for treatment and really put some numbers on it, hard numbers, that are validated, that allow us to measure these really complex constructs of frailty and low function and vulnerabilities across the domains of things like cognitive function and mental health, as well as physical function and nutrition, to then guide further testing, prehab interventions and ultimately personalized treatment decision making, so we can help patients optimize their outcomes.
And I think the important thing here is also optimizing outcomes that are directly congruent with patients priorities for what matters to them. For a lot of our older patients, it's important to define what success looks like to them, whether it's length of life and being cancer free or whether it's getting back home and living independently. Quantity of life and quality of life are two things that need to be sort of balanced. But ultimately I think that having objective data that is reproducible and reliable, that can guide our decision making, is going to be really helpful in helping patients sort of achieve that optimal outcome, whatever that means to them. I hope that was helpful, Sam. I'd love to chat a little bit more about this with you. I'll stop-
Sam Chang: Sure. That's just fantastic work. This is an area that we have used and I'll be the first to say I'm guilty of it, of using the eyeball test. And so many points you've brought up, really rang true in terms of attempting to codify, objectify, setting standards to show improvement. Along those lines, how do you think we can best balance, we have the patient situation in terms of his or her physical strength, psychosocial wellbeing, a disease burden, all the things that you've mentioned. How do we balance that with their environmental support or perhaps lack of. To me, some of the healthiest people may have some of the worst support and vice versa, and that can really impact on how a patient may ultimately do. Are there ways that you all are starting to look at measuring the external environmental role on frailty, on mental health capabilities, et cetera? I was wondering if you guys are starting to look at that as well.
Sarah Psutka: That's such an important question. It's really a patient's access to or sort of social support network, their caregivers. That contribution to frailty I think, is really understudied, understanding a patient's own resources. Something else that we've been working on is obviously learning more about financial toxicity and of the financial resources that our patients have, which are critical in terms of, especially given the expense of a lot of the therapies that we enact. And I just even this week had a patient who messaged me the night before a cystectomy and said, "I'm here, I've traveled across the state for my surgery tomorrow, I'm going to sleep in the garage overnight. "And I said, "What?" And I was like, "No, we're front door admitting you." But we didn't even know that this was an issue because it hadn't come up that he did not anticipate having a place to stay the night before surgery. I, as someone who's focused on these 360 degree evaluations, should have known that, but it hadn't come up in our preoperative conversations.
We do need to do a good job of assessing social networks support and definitely understanding who's at home and how involved they are in care. I would say that hasn't been something that I've been... This work that we've done so far has really focused on, but we're really trying to do now, a much better job of understanding those risk factors. And one of the things I really like about this CARE GA assessment, which I would love to of see disseminated because I think, here's the thing, we can't go with this risk assessment process alone as urologists. As urologists and urologic oncologists, we've got a lot to worry about when we're taking care of these patients. A limited amount of time to get through a ton of data.
And when I was actually pitching this protocol to our cancer center research team yesterday that we're going to be starting up, I said, "But one of the nice things about the CARE GA is it's a 10 minute, validated, self-reported survey that your patients can fill out before they come to see you. It's pretty straightforward. It's 42 questions, it's less than seven pages. It takes the average older adult, over 65, less than 10 minutes to fill out. Over 90% of these individuals can fill it out by themselves. They don't need help, which is good. It's not a high tech confusing survey, but it gets at those questions of things like social support that you might not pick up when you're just trying to run through all the questions you need to ask them about their hematuria and their pain and the weight loss or wasting and all the other things that you're thinking about as you're bringing them into surgery."
The way that I use these, is I have patients fill it out and I quickly review it before I walk in the room to identify any things that I realize I need to actually make sure I pay attention to that vulnerability up front.
Sam Chang: For sure, absolutely. Along those lines, as you counsel patients, I think that whole social support regarding who's their loved ones, family, et cetera, play a real role. How do you help prepare both the family as well as the patient of what's going to be happening down the road? Because people respond variably. Clearly, those that have sarcopenia, those that already have had issues with previous delirium, they're more likely to have those. What are some of the techniques that you use with your patients and their families, to prepare them for what's actually going to be happening down the road?
Sarah Psutka: One, obviously this requires a fairly high degree of emotional intelligence because you have to read the room and you have to get a pretty quick sense, both through your direct questions but also through observing interactions between patients and family members for the degree to which everyone's relying on each other-
Sam Chang: Okay. I want everybody to hear that, because what Sarah just said is basically to me, the combination of both science and the art of medicine, we're really attempting to increase the scientific knowledge of this. But at the same time, you really need those things that Sarah just said about reading the room. It's something that I try to talk with our residents and fellows about all the time and my partners who sometimes I give them some grief about, "Hey, you got to at least pick up on some of these signals." I hate to interrupt but I'm so glad you said that.
Sarah Psutka: It's so important, because there are these subtle signals that will tell you the degree to which your patient is looking to their family members for either support or there potentially is some friction and some problems and trying to navigate that is hard, but it's really important. First of all, I start this conversation by saying, "I'm not saying what I'm about to say to scare you, but if I don't do my duty and tell you what to expect and have you have realistic expectations, we're all going to be in for a rough ride." And so I say, "I want us all to go into this with open eyes, so I want you to know everything I know about what I'm expecting for you." And then I paint a pretty realistic picture.
If I know that I've got a patient, the patient that I did a cystectomy on yesterday, had lost 34 pounds in the past three months and he was not a big person to begin with. He and I talked at length beforehand about my concerns about somebody going into a cystectomy cachectic, what that looks like. And I said, "Here's how we're going to support you and here's what I'm anticipating." I do use the NIS Quip Risk Calculator that gives you personalized predictions using a lot of easily available patient specific characteristics for things such as length of stay, the likelihood that they're going to require a discharge to a place other than home. I love the geriatric build out there for the risk for things like needing a new assistive device, the risk of delirium, things like that.
I do think it underestimates some risks and overestimates others, so it's not perfect but of course it's a statistical model so there's confidence intervals and so you have to couch it. But I say here, if I had an entire population of patients like you, this would be kind of the average outcomes. And I say, "We could be in a better situation or could be worse. Here's what we're going to do to try to optimize these outcomes." But I definitely try to be pretty upfront about what I'm expecting and I do, especially in bladder cancer care and I'm sure you do this too, I tell them, I said, "We're going to have good days and we're going to have bad days. That is the nature of this beast. It doesn't matter which of the fantastic hospitals across the country you go to where people are doing hundreds of these a year, there's so many moving parts to this operation. Something goes wrong in almost everybody at some point during the recovery. And that's okay, we're going to be watching for it and we're going to deal with it when it happens."
And I think that having kind of a high level of vigilance for those adverse outcomes, whether it's an early postoperative ileus or whatever else, so you know you need to be thinking about early parenteral nutrition or getting people.. You have your rest protocols that are so well kind of in place. There's just a lot of things you need to be... I think you have to maintain a little bit of hyper vigilance with all of that. But I do think that setting the expectation early, that one, it's okay if things don't go great, because patients oftentimes also get really disappointed if they have a rough day. I think setting the expectations that this is normal, three steps forward, one step back, is kind of the way that I think a lot of these recoveries go. And also just letting them know that we're going to support them through the whole darn thing and being at their side as they go through it,
Sam Chang: Just nodding in agreement because I've got plenty of gray hair and I'm nodding and I'm thinking your wisdom is way beyond your years in terms of what you've discussed, in terms of what happens with these patients as we work together. And I love the idea, the framework regarding, they're going to be some bumps in the road and the fact that you don't want these patients to be disappointed, that they're not meeting the pathway day of when they're supposed... A lot do, but a lot don't. And so I so appreciate the steps forward that you're making with you and your team at University of Washington and thank you so much for spending some time with us and I'll look forward to seeing the results and I look forward to your being able to hopefully go ahead and spread that throughout your cancer center. Because I think those types of initiatives will only serve us well for all our patients that have different types of cancers and different types of treatments, both on the medicine side and the surgical side. Sarah, thanks again and we really appreciate the time.
Sarah Psutka: Thanks Sam. This was really fun.