 I'd like to congratulate everybody for their excellent talks. Dr. Thompson asked me if I would contextualize what they're saying from a big philosophy point of view and then present some cases to the group. So here we go. We have a cancer problem in this country. I don't like the way this started, but let's try that again. We have a cancer problem in this country and by that I mean that we are better at finding these cancers early than we are at fixing these cancers early. This is a very nice graph looking at thyroid cancer, prostate cancer, melanoma and breast cancer. We find them more frequently. We don't fix them over the years, much better. The same is true for kidney cancer. New diagnoses up, deaths from kidney cancer about the same. The reason for this is clear. We have this find it mentality. We're a better with the technology of finding cancer than we are with understanding the biology of fixing the tumor. So we scan people more often. We find it more often, but do we fix it any better? That's the question. So the problem here is that the cost of care for the caring of these cancer cases is increased. This is some data from Neurology Diseases Online, Mark Litwin's data, to look at how much we've changed the cost of caring for these patients over the last four years between 2006. And as you can see, we've almost doubled the cost of care in these patients. So at the same time, while heart disease has decreased in terms of death rates and cerebrovascular disease as well as infectious disease, cancer death rates have really been pretty much the same. So what we now have is a situation where we're costing or paying more to give equal outcomes. So if you define value in healthcare as the best possible outcomes at the lowest necessary cost, if you increase the cost and you have the same outcomes, you decrease the value. And that's the problem that can play out very well with a lot of early cancers, kidney cancer, no exception. So if you think about it, this is a very nice article looking at how you think about these different types of cancers. There's a type of cancer that grows very fast from inception to death is quick. We will call that the hawk. That is the most aggressive type of cancer. That's the one where you're at risk of under-trading that patient. Then we have the slow progressors. We'll call that the tortoise. That's the one where no matter what you do, you're probably going to win. And then somewhere in the middle is the rabbit. So what this article really talks about is the simple concept of risk stratification. Risk stratification with localized kidney cancers. So if you'd like, when a patient is called with a diagnosis of a small renal mass, their first thought might be, I'm going to die of cancer. The doctor's first thought, however, might be, I can cure this with a robot. Because we're managing cancer. We have this find it, fix it mentality. And I propose to you that really what we should have is managing their risk. We shouldn't be managing their cancer. We should be better at managing their risk. Should we find it? And if we do, should we fix it? And the same can be said for early stage prostate cancer or breast cancer. So to start off and contextualize that, let me just sort of talk to you about the different types of risk in a very general sense. There's a voidable risk, right? There are not a lot of avoidable risks in kidney cancer, smoking perhaps, perhaps obesity or hypertension. And there are some undefined genetic and epigenetic risks which may or may not be avoidable. But the first type of risk is avoidable risk. The second is the difference between relative risk and absolute risk. So this guy just jumped out of an airplane. He has a risk of dying from what he just did. Relatively speaking, this gentleman who's about to jump out of an airplane has a risk of dying that is not much different, maybe a little bit lower because he didn't take the leap. But from an absolute standpoint, the guys on the ground that never get in the airplane have an absolute risk of dying from skydiving to be very, very low. So you need to communicate absolute, not relative risks. The third type of risk is calculated risk. And this is a calculated risk. You have to know when to accept those risks. And the fourth type of risk is the unexpected one. How do you predict the unexpected risk? How do you account for it? And how do you manage it? So medical risks are, you have to educate for avoidable risks, communicate absolute risks, accept calculated risks, and study unexpected risks. That's how you manage risk. Well, this is a fundamental human endeavor. Drive home, listen to the radio on any given minute. You'll see that we as humans try to manage risk all the time. Why? Because we try to predict the future. Weathermen try to predict hurricanes. Stockbrokers try to predict the stock market. Real estate or all state agents try to predict crashes. Ben Bernanke or his successor try to predict inflation. And actuaries do a good job of trying to predict the future as well. Well, how do we manage risk? We try to predict the future. That's what this discussion's all about. Well, here's a guy that can predict the future. That's Carl Frederick Gauss. And Carl Frederick Gauss came up with this equation. It's very simple. He said, if you want to predict the future, it's the bell shaped curve. There's probable and there's possible. And we use this all the time. We all flew to this meeting, or most of us did, because we believed the probable outcome was going to be we were going to get here safe. The possible outcomes where we were never going to leave the airport or where the airplane was going to crash. But we predicted the future because we managed what's probable, not what's possible. Well, in this elderly man here with that small renal mass, what's probable? It's possible he can get metastatic disease. It's possible, which we don't really usually admit, that we can cause his demise. But what's probable is he's going to die from something else. And in fact, we know that from lots of studies. ERTC said, ERTC 30904 says 10% of people will die of kidney cancer deaths, and 90% will die of other causes. Dr. Lane published data that says 4% of elderly people die from kidney cancer, and 30% from cardiovascular disease. Dr. Kudovkoff and myself, the group at Michigan, again, competing risks appear to be the biggest risk. So we tried to do this study now with Brian Eggleston, one of our PhD statisticians, looking at latent class analysis using perpensity score matching, a very sophisticated modeling. And if you ask me, there's just three or four options. If you've got a long life expectancy and you treat the person, that's a good thing. If you've got a short life expectancy and you don't treat the person, that's probably also a good thing. But if you've got a long life expectancy and that person goes untreated, that's probably bad. And if you've got a short life expectancy and you treat that patient, that's probably bad. Can we estimate each of these blocks? What winds up, we've tried. And we're trying to, we're submitting this publication now, it winds up then in the majority of cases, surgery appears to be helpful, and the outcomes appear to be relatively beneficial. That's what's probable. But what's possible is right here. Surgery can be very highly beneficial in a small, maybe 10 or 11% of patients, where you really affect the natural history, or it could be harmful, where your intervention actually causes the patient harm, causes them to live less, not longer. So that's really risk stratification, right? How do you take this tumor and predict if it's a hawk, if it's a turtle, or if it's a rabbit? Now a lot of times in medicine, what we're doing is we're managing the case, not what's probable, but rather what's possible because that's in fact how we get paid, right? We get paid for fee for service. And so this type of thinking sometimes threatens physicians because how do you make money in risk stratification? I'm gonna submit to you that you can make a lot of money in risk stratification. Here's my example. If nobody knows who this is, that's John Bogle. John Bogle found index funding in Vanguard. He's a billionaire. Why? Because he realized that there's not a lot of Bernie Madoffs and there's not a lot of Warren Buffets that most of us sort of regressed to the mean just invest in the middle, managing what's probable, not what's possible. It's possible to win in those kind of scenarios. So if you have a patient you're trying to do if you should excise, ablate, or observe, we can argue that there are some factors that we could use, but probably the number one factor that makes that decision is the physician. The physician who that person goes to will decide how they interpret the patient's risk for them and therefore how that patient will go have their treatment. And so instead of managing cancer, I say manage risk. And here's my risk model. There's four buckets. You have to understand tumor risk, which is the inherent biology of the tumor. We don't do that very well. You have to understand patient risks, which is comorbidity. We don't do that very well either. You have to understand the physician risk, which means what's my individual skill set. And you know, we've got a lot of dogma and what Stephen Hawking's term's illusion of knowledge, that we don't really do quite as well. We don't like to report on our own outcomes quite as much as we should. And of course there's hospital risk, which is process. And the goal in my risk model is to understand and manage variability. And if the young people in the room don't think that they could make a career out of this, in just one of these boxes, Atul Gawande has made an entire career reporting on our inability to manage hospital risk and process. There's lots of opportunity here, guys. So with that, I've put a little context around, I hope the discussion, I'd like to present the case. I apologize to the panel because they have not seen this case. So anything they say about this is truly honest. So with that, I'm going to start. This is a 73-year-old female with an incidental left renal mass. She's got hypertension, some diabetes, and had a stent in her heart a while ago. She had an extended left colactomy and colostomy with a delayed renalastomosis. She's got a midline score and a GFR of 48. That's her tumor. What you can see better here, it's an afrometer score of 10AH, meaning it's a high complexity tumor. Here's the tumor. You can see it in multiple views. And I'm going to ask Dr. Kudakov first, is a biopsy necessary in a 73-year-old woman with that tumor? And if so, what is it likely to tell you that you don't already know? So I think you can make a good argument for biopsy here. And I think you can make a good argument against biopsy. I think it's a discussion with the patient. I, as I said in my talk, there's basically no downside to a biopsy here. The upside, you biopsy this is an oncocytoma. This woman avoids an operation. And so I would have a very low threshold to biopsy a 75-year-old person because she looks better than her age, but I can make her look her age very quickly if I take it for an operation. All right, well, then let me ask Dr. Lane, would you forego the biopsy because it's not in an exactly great location and what have you gotten a biopsy if it might have been easier? I think there's, can I answer a different question first, which is I'd love to know the natural history on that renal mass. I'm suspicious that she's got multiple prior CTs since she's already had a couple operations for cancer. And I'd like to know if it's grown from undetectable to 3.8 centimeters over three years or whether it's been 3.8 centimeters for three years. So good question. She's had no prior, she's had no prior CTs. Okay, so it was for diverticulitis. So again, I think if this were a stable lesion, I'd feel much more comfortable watching it. At this point, I'm fine with doing a biopsy for that. This is a high risk patient who has, high surgical risk patient who has multiple options that would be affected by biopsy. Well, I'm gonna tell you that I think in most cases, as Dr. Lane described before, you could probably tell more often than not what it is without the biopsy. It's heterogeneous, there's brisk uptake. If you talk to your radiologist, they can tell you about the uptake patterns. If you've got an MRI, you can use ADC mapping to tell that this is probably a clear cell carcinoma. I grant you that, it might be an oncocytoma, but it's certainly not a papillary and unlikely a chromophobe. So I'm not entirely sure that biopsy completely helps there. I would also say that most of these are low grade. We've got very good data. And we've got nomograms that will predict as well in some cases as a biopsy, as Dr. Lane told us, that this is probably unlikely a man eater. So, Dr. Jewett, say you biopsyed the patient and it came back a clear cell. Now you tell the patient that she comes back to you. Can you estimate for her, her biological risk of this metastasizing in the next three, five or 10 years? How would you counsel? I would have biopsied this patient as I said earlier. And if it was a clear cell, I would still manage this initially by active surveillance given those comorbidities and the patient's desire. She's right on the line of age, but she has significant comorbidities. So I wouldn't object to treating her, which would probably be a partial nephrectomy, but I certainly wouldn't rush into it. And if it wasn't growing and she had other issues, we'd be quite happy to watch it for the short term. So you know now it's a low grade clear cell. What she asks you, if I watch it, what's the likelihood I will die in the next three or five years of metastatic kidney cancer? Can you give her an answer? So it'd be very low, and we would put that in probably one to 2%. But again, by watching it for three and six months, we probably can refine that. Okay. All right, let me ask Dr. Atwell early. Is this an ablatable lesion? It's anterior, it's difficult perhaps. You gotta go through the spleen or maybe entirely through the kidney. Let's start with Dr. Lee, can you microwave this? Well, I just wanna go back and comment on one question that was put out earlier about the difficulty of certain biopsies. And I would say that if you're looking at just a CT scan, you can get fooled into thinking that something is incredibly easy or incredibly difficult based on location. As someone that does hundreds, I guess, or thousands of these biopsies myself, we use almost exclusively ultrasound for this purpose. And the major advantage of that is, well, besides being real-time, is that you have multi-planar capability. So for example, for this particular tumor, it's likely if you come from below, you'll have a really nice window to pop into this thing easily. So I'd just be careful about estimating the difficulty of a biopsy or procedure based on CT. Now to then follow on with about ablatable, I showed a case almost identical to this in my series just a minute ago. The anterior lesions used to be a problem before we understood how to get some of the intervening structures like the pancreas and the bowel out of the way. I would say, and I'd be interested in Tom's comments on this, I would say that that is almost never a factor anymore, except for patients that have undergone multiple retroperitoneal surgeries or for some reason the bowel is actually stuck on the tumor. So assuming that this bowel was able to be moved with hydro dissection, sure, this is pretty easy tumor, I think. So before Dr. Atwell comments, can you tell us that ablation does a better job than Dr. Jewett would do watching it? No, not necessarily. What I can tell you though, is that our chances of destroying the tumor completely with minimal or no risk is very high. How that translates into a long-term outcome we're not quite sure yet. Dr. Atwell? I think the easy answer is yes, it is ablatable. The risk of bleeding, the central collecting system isn't an issue. So technically the answer is yes, whether it provides a better outcome for the patient. I guess it depends on how you define an outcome. Will she live longer maybe? Will she sleep better maybe? Yeah. Will you sleep better? Things go well, yes. All right, so the biology of the tumor, I'm not gonna go over much of this. Dr. Jewett and our group and others have talked about the biological risk, right? A 2% three to five year metastasis progression rate. There are also very good nomograms to look at what is the success rate, a 98% five year cancer specific success rate in this patient if you excise it. So anytime you win 98% of the time, I think Dr. Jewett and I and many of us would agree that you have to question the fortitude of the competition. Okay, what about comorbidities? Now let's quantitate patient risk. Charleston Comorbidity Index, competing risks of death. There are life expectancy tools. This is a good one from the NCCN. You can put in a 73 year old female in the 50th percentile of health. She's got a 13 year life expectancy. I mean, what resonates here with me is that if I go to buy life insurance, this is what my agent's gonna do and this is how he's gonna price out my policy. Similarly, you can look at competing risks and Dr. Kudakov and I and others have worked on how do you actually measure competing risks? Her competing risks of death, 3% from kidney cancer, but 15% from other causes. So trying to objectify these risks in this patient becomes important. Dr. Stifleman, what do you tell the patient the likelihood is of a major or minor complication? If you do a robotic portion of the front of me, in this case at NYU. So I'll answer that question first, but I think this patient, and you showed this earlier, has a higher risk of having a high grade disease based on that nephrometry score. So you show this, we've looked at this as well, using the nephrometry score to help assess the risk of being a high grade tumor and that is a high grade nephrometry score and I think that's gonna be a high grade tumor and this could be end up being a shark. So Ursula, I would offer her definitive therapy at 75 years old with a 13 year life expectancy. In terms of the likelihood of a minor complication, I would quote her probably a 20% chance of a minor complication for this. And in terms of a major complication, which would be a urine leak or a pseudo aneurysm or a bleed requiring a transfusion, I would quote her less than 5% chance. So my point exactly is that these are the risks the patients need to make an informed decision. Let me ask you to follow up. Would all the surgeons at NYU use fluorescence and does its use and associated cost lead to improved patient-centered outcomes in your estimation? So two of the three of the surgeons would use fluorescence for this. I would absolutely use fluorescence for this. This is a high grade tumor. Though it's a 10AH, if you look at it, you're gonna basically enucleate this out of the sinus. This is not gonna, you're not gonna have to remove a lot of parankula for this. So because of her comorbidities, I would definitely use fluorescence. I would try to do this with selective clamping and I think she's got an excellent chance of being home in two days from the hospital and a week going back and doing her errands. And you believe the cost is justified? Well, I think in this case, the person has diabetes, she's got hypertension, she already has a GFR, CKD of 30. And in our earlier experience, we've shown that the delta change in GFR initially is lower when you can do selective clamping. So yes, the answer would be, I do think it would be worth a $250 cost. Okay. Well, it's $250 for the diet, but more for the more for the hardware, right? Yeah, depends how you, you know. All right, so here is how, this is the physician risk box, right? How do you know, how do you know what your own physician set is, skill set is? If someone's to ask me where my skill set was on this bell shaped curve to fix a bladder extra fee, I'd put it down here. But most physicians believe that their skill set is quite high. I just remind everybody that it is a bell shaped curve and most patients believe their physician skill set is extremely high. So the question is how do you know what your skill set is? Whether it's with the knife, the scope or the robot and that is, you gotta measure it. And we've published ours. You know, we've got relatively robust data to suggest in a patient like this, she's got a 20% major complication rate, mostly leak and a 30% minor complication rate. We could also tell you based upon what system she is likely to have the complication in. So understanding your own data in your own institution with your own processes becomes an important part of counseling this patient. So again, how are the processes in your hospital are you improving value by either decreasing cost or increasing patient centered outcomes? Because most of the costs are buried in variability and it gets back to my concept of risk modeling. How do you use your system to minimize physician variation and avoid unnecessary risk with these patients? It requires a village as they say, of people, physicians to own this and to move it forward locally in their own institution. So what I would tell the patient is, there's a 93% chance that if you come to our institution and you have a surgery, it's gonna be a portion effect, it's gonna be a portion effect. I mean, it's a 79% chance for a high nephrometer score like yours, it will be robotic. I will tell her that based on our data, there's a 22% chance of a Klavian 3, 4, or 5 complication and I will give her the other statistics because you can't manage what you don't measure. So in the end, managing risk is what we are supposed to be doing, I believe, and managing that variability is important. So move away from managing kidney cancer to managing risk. And as was said by William Osler, science, medicine is a science of uncertainty and an art of probability. And with that, I have another case, but we only got about two minutes. I could run through the panel very quickly. We want me to stop here, Houston. Okay, very quickly. Okay, 63-year-old male left radical nephrectomy for a chromophobe carcinoma in 2007. He's got high-grade prostate cancer T3A, Gleason 8. He had a prostatectomy and has a PSA recurrence for which he was radiated. That was seven or eight years ago. A smoker with some coroner disease. I will promise to make this short. He's got a solitary kidney. He had chromophobe on the other side. Let's go down the list. Dr. Steifman, would you biopsy this patient? Solitary kidney chromophobe on the other side. Dr. Lane. Yes. Would anybody not biopsy this patient? You got chromophobe on the other side. Dr. Kudakov mentioned something about hybrid tumors. Dr. Kudakov, would you like to sort of elaborate, is this likely to be chromophobe or do you think it's likely to be something else? It's a good question. The concordance rates for, there's not large areas for chromophobe concordance rates, but this, I mean, the biopsy will certainly be able to distinguish if this is, you know, if this is an oncocytoma versus another chromophobe versus a clear cell. I mean, I think biopsy is very informative here and will absolutely change management, especially in this patient with chromorbidities and with solitary kidney. So a young man, solitary kidney. He had kidney cancer on the other side. He's very worried. He was referred for a portion of rectomy. And here I am gonna say you need a biopsy because I don't know what it is even though there was cancer on the other side. But we've all agreed that's what you should do. And in fact, he had a biopsy and in fact it was an oncocytoma and in fact, he's now avoided the morbidity of that operation, managing risk. So again, I'm not gonna go through that again. I've already beat that to death. The risk model, managing variability in tumor, risk, patient risk, physician risk and hospital risk. There's a very good article. I've got 45 seconds left by Fagerland and JNCI about how to communicate risk to patients. I invite you to read it. It is very, very good. I think that if nothing else, if it doesn't change the outcome of your decision, it certainly engages the patient in that decision-making process. We're moving toward accountable care. We've got to quantitate and estimate risk. It forces us to give better care. And I'll give you three quotes. I said it before, medicine is a science of uncertainty but an art of probability. Let's manage what's probable, not what's possible. The second one is it's always difficult to get a man to understand something when his salary depends on his not understanding it. And as we move more towards accountable care, I think we're all gonna have to deal with that one. And finally, in the midst of every challenge, ACA, Obamacare, call it what you want. There's opportunity and the doctors have to lead it. We cannot be put upon, this is ours to own. Thank you very much.