About This Episode

The inaugural interview for the new Global Medical Office Dialogues video series is with William Kaelin Jr., MD, Sydney Farber Professor at Harvard Medical School, Dana Farber, and Brigham and Women's Hospital in the United States, and winner of the 2019 Nobel Prize for his research with two others regarding how cells sense and adapt to oxygen availability with implications for anemia and chronic kidney disease care. 

 

 

Video player requires JavaScript enabled. You can watch this video here: https://youtu.be/eGgLQFUBCus



 

Featured Guest: William Kaelin, Jr., MD

William G. Kaelin, Jr., MD, is the 2019 Nobel Prize recipient in medicine or physiology. Dr. Kaelin received his MD from Duke University in 1982 and was a house officer and chief resident in internal medicine at Johns Hopkins Hospital. He was a medical oncology clinical fellow at Dana-Farber and a postdoctoral fellow in the laboratory of Dr. David Livingston, where he began his studies of tumor suppressor proteins. He became an independent investigator at Dana-Farber in 1992, and a Howard Hughes Medical Institute Investigator and Professor of Medicine at Harvard Medical School in 2002.

The 2019 Nobel was awarded jointly to Kaelin, Sir Peter J. Ratcliffe and Gregg L. Semenza for their discoveries of how cells sense and adapt to oxygen availability.


 

Listen to this episode
 

Episode Transcript:

Dr. Maddux:
Bill thanks for joining me today. 

Dr. Kaelin:
My pleasure. 

Dr. Maddux:
Let's start our conversation by talking a little bit about your perspectives on oxygen sensing and its role as one of the human adaptive physiologic processes.  

Dr. Kaelin:
Yes. So when I was younger I trained as a clinician in internal Medicine and later in medical oncology but then later acquired the tools to become a laboratory based investigator, and both from my clinical work and my laboratory work I became very impressed with the power of human genetics and genetic variants to teach us about not only human disease but human health and shortly after I started my own laboratory in the early 90s, the gene that scientists referred to as the Von Hippel Lindau or VHL gene was isolated and unfortunate people who are born with a defective version of this gene, develop a variety of tumors, where those tumors constantly produced the distress signals, if you will, that would normally be produced. If our cells and tissues weren't getting enough oxygen. And so that was the clue that if we could understand the normal functions of the VHL protein that is to say the protein that is made using the directions in the DHL gene. We could potentially understand at the molecular level for the first time how cells and tissues sensed and responded to oxygen. And of course I knew from my clinical training that there were many diseases, where oxygen and abnormal oxygen delivery plays an important role, such as anemia, heart attack, and stroke. And Eventually I along with others in the field were able to use this experiment of nature, if you will, as well as other clues to sort of lay out in molecular detail how our cells sense and respond to oxygen. And once we understood that pathway, we saw there were several places where we could pharmacologically intervene, either to mimic the effects of low oxygen or in some cases to block the ability of certain cancers to co op this pathway so we could basically play both sides of the street. 

Dr. Maddux:
If we look back at the gene mutations. Is there a single type of mutation or can you describe whether there are several variants and how the mutations occur? 

Dr. Kaelin:
That's an excellent question. So it turns out there are a variety of mutations that have been described in the VHL gene. Some are very subtle actually and others are very deleterious. And so that's also helped us to begin to understand further this system not as a digital system that's completely on or completely off, but also to see it more like an analog system, where it sort of rheostat where we can tune the system up and down. And of course, that becomes very important when you start to think about drug development because most drugs are successful, because you can titrate the dose of the drug to the desired effect. If it was a simple digital system that was on or off it would be much more difficult to exploit this for therapeutic purposes. 

Dr. Maddux:
We’ve seen some gene editing techniques with sickle cell disease. Any opportunities for gene editing techniques that would impact the system of this oxygen sensing capacity and VHL? 

Dr. Kaelin:
I could imagine that happening in the future at least with respect to cancers that perhaps have a mutation in the VHL gene. Of course there the challenge would  be to correct the abnormal gene, and all of the millions and frequently billions of cells making up an individual's tumor so that at the moment that's a daunting challenge but maybe eventually we'll be able to do it. But even now laboratory studies suggests that there may be opportunities to for example alter the function of the VHL protein or at least the VHL pathway, an organ such as the liver for example where that might be to increase production of erythropoietin. So there are all sorts of opportunities that we never dreamt up now that we understand how this pathway works. I should point out the pathway is ubiquitous. It's present in every cell in our body. And we've also learned that some other components of the pathway are also occasionally altered in the human population. And for example, in some cases, give rise to excess red blood cell production in families. 

Dr. Maddux:
You had the chance to tell me a story about how you had one of your first interactions with the folks at Fibergen related to understanding that this, these cells, and this activity might actually impact anemia. And can you just describe that story I thought it was quite interesting. 

Dr. Kaelin:
At the heart of this oxygen sensing mechanism is the addition of a little chemical flag that scientists refer to as hydroxyprolene to another protein that scientists refer to as HIF or HIF, and HIF is a master regulator of genes that help promote adaptation to oxygen. So once we knew this was the critical signal the addition of this little chemical flag this  pro-hydroxylation mark. We discovered that a little biotech company called Fibrogen was already developing chemicals or compounds that could inhibit the addition of Bismarck to collagen, it turns out has been known for years that collagen which we usually think of as being part of our connective tissue undergoes this mark as well but for very different reasons unrelated to oxygen sensing. So after doing a little detective work I reached out to one of their scientists, with the simplistic idea that perhaps some of the drugs they had been developing to block the pro player hydroxylation of collagen might fortuitously also block the pro level hydroxylation of HIF. And if that was true that would be a good starting point these chemicals or compounds would be a good starting points to make more specific inhibitors. And I must say initially the scientist I spoke to was quite polite. But then sort of dismissed me and said well you have to be wrong because this modification has never been seen largely outside of collagen and it's really unimaginable that it is modifying this protein called HIF which we know lives in the nucleus, where this modification’s again, never been seen. And then about 15 minutes after hanging up by me he called me back and said please start from the beginning and tell me the story all over again. And what had happened was the CEO of Fibrogen, the late Tom Neff had seen one of his scientists walking down the hall by laughing about this crazy scientist from Boston, who thought Provo hydroxylation might be involved in oxygen sensing and Tom despite not being formally trained as a scientist had remembered that some of the early compounds when given to rabbits caused the rabbits to make an increased  number of red blood cells. So it was really Tom who said call them back immediately because this Boston scientist is trying to tell you why the rabbits were making too many red blood cells. b

Dr. Maddux:
Well, certainly, what led to Roxadustat is certainly changed the, the nature of bringing HIF as a class of drug to nephrologists, and I would say that it's one of the changes that we're seeing in anemia management that's got great interest. I'd like to shift gears just a little bit and talk about one of the things that I think you've spoken about before and that is that there are pleiotropic effects from these types of drugs and one of them is elaboration of the VEGF and angiogenesis  which certainly played a role in your oncology look at the substances  I'd love to hear your thoughts on what the risks are potentially of this class of drugs and how great or low that might be.  

Dr. Kaelin:
This is a very important question and one that frankly we've been thinking about for over 20 years. So, 20 years ago, I think most of us, including myself, would have imagined that as you turned on HIF every gene that was under the control of HIF would be activated at the same time and to the same degree. We've subsequently learned that that's not true that some genes need a lot of HIF to be activated and some genes need much less HIF to be activated so that's part of the puzzle. The next part of the puzzle is as follows. You know, for pretty much that same reason when we saw the early data with some of the HIF pro hydroxylase inhibitors, it was clear that some of them induced erythropoietin, which we wanted to see, but they did not induce that VEGF which was frankly something we had been concerned about for the reasons you sort of alluded to, and that was a bit of a head scratcher because again the paradigm was all HIF changed to be activated at the same time. But we've subsequently learned that they said not only is that not true. But we've begun to understand why it's not true. And now we're going to get a little more nuanced because scientists are a little bit lazy when they refer to HIF, because there's a HIF one, and there's a HIF two, and there's even a less studied HIF three. So it turns out, HIF one which is usually the more famous HIF the one people are thinking about is in most tissues the HIF that could turn on VEGF if it was left to its own devices. But we know that. Fortunately, HIF one has a second failsafe mechanism that keeps it from becoming active if oxygen is present. So even when you inhibit the hydroxyprolaces, and you turn on HIF one. HIF one is largely in feeble because of this backup mechanism that involves a protein called FIH1. It turns out, HIF two isn't subject to this, this other failsafe mechanism or if it is it's much less sensitive to this failsafe mechanism. So when you treat with a HIF  prolyl hydroxylase inhibitor HIF two does become stabilized and it is capable of turning on erythropoietin so we frankly got a little lucky there that you could start to dissect the induction of erythropoietin from the induction of VEGF. So that's part of the answer. The other important part of the answer is even though we know that inactivation of the VHL gene is an important step in many kidney cancers. It's not sufficient to cause kidney cancer. You need to have mutations in many other genes that sort of cooperate with the loss of the VHL gene so I use the analogy of tumblers on a lock where to develop a kidney cancer maybe on the order of four or five tumblers have to be in place. VHL mutations can constitute one of those tumblers, but we know from studying VHL patients, it can take years or decades to acquire those other mutations so that's part of the answer and the other thing I would say is, we know from at least laboratory models. And it's highly unlikely that HIF by itself can cause cancer with very few rare exceptions, and in fact there are quite a few individuals around the world who have conditions where they're having chronic problems with oxygen delivery. They might be people living in high altitude. They might be people with certain chronic heart or lung conditions, and some of these patients we can be sure have chronically but mildly elevated HIF, and are making more red blood cells, but to a first approximation, they certainly don't look like they're excessively cancer prone. So, you know, we can, we can go on with, you know, indirect arguments, but I think there are a number of indirect arguments that are suggestive there could be a so called therapeutic window. It actually also comes back to a discussion we were having a few moments ago that most drugs are useful because we can titrate them to the desired effect. So we're not talking about, you know, altering oxygen sensing maximally 24/7. You know, we're talking about episodic sub maximal activation of the head pathway so that the body can make just enough erythropoietin to make more of its own red blood cells. I guess the final genetic experiment of nature is there are some individuals around the world who have mutations in the actual enzyme. That is the target of roxadustat. And we know from following those patients that they make an excess number of red blood cells, and I should also point out that those individuals are making many more red blood cells than we would ever hope to achieve in the kidney cancer population. And again, they don't seem to be at an excessively increased risk of kidney cancer, they may have been a few rare case reports, but nothing like the patients we had earlier studied, who had mutations in the VHL gene. 

Dr. Maddux: 
Well the experience we're gaining in our patients with end stage kidney disease, and advanced kidney disease not on dialysis with these oral drugs is going to be actually quite interesting. I think there's quite a bit of both clinical interest, and interest around just validating the safety that's been tested in the phase three programs, quite a bit over the coming months. I'd like to shift gears on something that you mentioned just a minute ago and that is, I think the tumblers on the lock is a great analogy to the fact that today at least in kidney disease we look at classification of kidney disease, almost strictly through pathologic basis. So it's what's a kidney look like when it's been injured or has been injured. In oncology you have moved well beyond that to look at what these multivariant genetic changes that can occur or mechanisms of injury that get elicited that generate tumors. What are your thoughts about the opportunities for precision nephrology to involve genetic and genomic information recognizing that it isn't necessarily about how the kidney looks but it's how the actual physiology and mechanistic injury is occurring. I'd be interested in your perspectives on whether that's a pipe dream or whether that's something that you think has real merit. 

Dr. Kaelin: 
I don't think it's a pipe dream I think it's a very good question. You know I think genetics. Broadly, is very powerful in predicting. For example, the consequences of various therapeutic interventions and so just to give you one famous example outside of oncology, you know, there are certain individuals who were found to have inactivating mutations and PCSK9 and these individuals had low cholesterol so that actually de risked I think quite a bit the efforts to make drugs that inhibit PCSK9 because the definitive genetic experiment had been done we talked about genetic validation. We're lucky that in oncology, although cancer’s a very complicated disease, at least we do have genetic information once again we can see which genes are altered in specific cancers and try to tailor the drugs to those mutations, so called precision medicine. I alluded a moment ago to the fact that there are some individuals who have a loss of function or inactivating mutations in the HIF prohydroxylase so that again gives me great confidence that we can begin to make some predictions about what it would be like to inhibit that enzyme with the caveat that there again the model is really a model of 24/7 sort of overdosing if you will, because those are very deleterious mutations in some cases. So coming back to kidney cancer, I don't know whether kidney cancer, and kidney disease broadly, I mean let's, let's take away kidney cancer where we do have some genetics let's go to your question of more broadly kidney diseases. I don't know how often we're going to find those genetic guideposts whether deeper inspection will find out that there are polymorphisms or genetic variants and the human population that all along were determining the development and severity of various kidney conditions. My fear as a non expert in this area is that it could turn out to be like some of the therapeutic areas such as psychiatry where those genetic guideposts have really been few and far between, but certainly in many other areas of medicine, including gastrointestinal disease and autoimmune disease, genetics have been very powerful so I would hope eventually we can apply them more to nephrology. 

Dr. Maddux: 
One of the things that I've observed is in clinical practice of oncology, almost every oncologist offers available research studies to the patient whose particular cancer has ongoing clinical trials. In nephrology there's actually relatively little of that occurring and, and one of the hopes I would have and I'd be interested in your philosophy on how physicians approach clinical research is could we get to a point where all nephrologists are at least making aware and offering to their patients the opportunities that clinical research might bring. Any perspectives on that from the standpoint of having both seen clinical oncology develop over the years and also being part of the basic science that has developed those opportunities for additional research? 

Dr. Kaelin: 
Well, I think that would be wonderful as you say, hey, I'm a physician scientist, and I think physician scientists are in such a good position to look at clinical challenges, clinical conundrums, shortcomings in our therapy, and then to try to dovetail that with, hopefully, increasing knowledge of the underlying pathophysiology of the diseases that, that one is encountering. So I certainly hope we get there, I think unfortunately many physician scientists are feeling a little bit under threat because they're getting pulled in multiple directions and so people worry sometimes that this could become a dying breed and I think that's exactly the wrong direction to go for reasons your sort of hinting at so, unless you're completely satisfied with the management of a particular disease, I think there's room for investigation, and hopefully room for hypothesis based clinical trials.  

Dr. Maddux: 
In the midst of this coronavirus pandemic it's been fascinating to look at not only how we've responded or haven't been prepared to respond but also the variations of the learning curve that we've had around coronavirus, it's been a learning curve that for everyone has been exceedingly steep, and certainly in understanding the genetics of it came very early in the course of this as well as these sort of wide rapid mutations that are occurring. One of the things that I'd be interested in your perspective on is the impact of societies and social determinants on how we prepare our healthcare systems for a variety of diseases. In this case we saw it through a wide infectious disease, but we're also seeing other social determinants as we're having the discussions of some of the racial inequities that are recognized and have occurred. I would just be interested in your perspective on social determinants, society's response. And where does healthcare stand in the sort of importance to a society being vibrant and healthy? 

Dr. Kaelin:
Boy, what a terrific question. I do think, first of all, as a physician scientist, I mean, you're absolutely correct we benefited tremendously from decades of investment in basic virology, as well as immunology because when the virus came along. It was quickly sequenced we start to identify the likely genes we can start to predict which proteins were going to be on the cells, or on the surface of the virus and hence, potentially targets for vaccine development, etc etc. So the good news was we weren't really so much knowledge limited as resource limited there was clear if you then marshaled the resources you could start to quickly think about making vaccines, neutralizing antibodies, repurposing drugs,  etc etc so that's all the good news and I think that's why you've seen some really some developments that are happening really unprecedented speed. The bad news is I think we're also paying for decades of sort of calling convenient science bashing when it served people's purposes to bash the scientific community, in the context of discussions such as climate change and others, because I think we've gone that our country from maybe having a healthy degree of skepticism when it came to dogma and centralized authority to in some quarters, you know, frank cynicism and disregard for people's expertise and credentials and I think that's one of the reasons why you're seeing very variable uptake when it comes to mitigation strategies across the country. I think we have to get back to you know making decisions based on science and data and not simply expedience and wishful thinking. And then you bring up the health disparity and the racial disparities, which of course is is heartbreaking and I fear that this has been one giant stress test that has revealed some of the weaknesses in the American healthcare system, one of which being that I think arguably it's not a system a capital S, it’s a patchwork of lots of things. And I think for some people it provides the best care on the planet but too many people can fall through the cracks. I think that has to be readdressed.  

Dr. Maddux:
We’ve been talking about the not small anti-vaccine movement that exists, certainly in the United States and some other places. And I'm curious your perspective as a scientist on sort of how do we as medical professionals try to actually begin to recognize that a true resolution to coronavirus will be highly influenced by the degree to which we can all generate a population based immunity? 

Dr. Kaelin: 
Well unfortunately I think the anti-vaxxer movement is another symptom of what I was just referring to earlier where unfortunately, a number of people have grown to disregard true expertise and credentials and are too prone to perhaps get their medical advice if I can call it that, from less trustworthy sources and sometimes involving social medias, so I don't know all of the solutions to this problem. I think it would help to start our political leaders modeled good behavior and science driven decision making, I think. Unfortunately, we don't always get that and I think that is harming us tremendously. But I think we all have a stake in this trying to educate the public. I had, perhaps naively hoped that the coronavirus pandemic would remind people why we use vaccines when they see the suffering that this disease that was caused, and perhaps I shouldn't give up on that wish quite yet but that's, that's my hope that people will be reminded, and perhaps we will painfully see that some of the states that were more science driven and more aggressive were able to perhaps flatten the curve a little bit more effectively than others. There are other things coming down the pike, as you know, in addition to vaccines that maybe will be slightly less contentious for those folks that use some of the neutralizing antibodies that are being tested now that maybe could serve at least as a stopgap until we get to definitive vaccines. 

Dr. Maddux: 
As we get towards the end of this particular dialogues interview with you. I would certainly want to state that for Fresenius medical care we take care of end stage kidney disease patients around the world in over 4000 clinics and your life's work that led to the Nobel Prize and other tremendous prizes and in medicine is fundamentally going to have an impact on how we care for patients and, and how we improve the quality of life that they're living, and hopefully reduce some of the risks that they're exposed to. And so I would thank you very much for the time today to talk to me and just ask if you have any other closing thoughts or comments before we end the interview?

Dr. Kaelin:  
Well first of all thank you I've enjoyed this tremendously and thank you for your comment you know I get asked frequently what was it like to win the Nobel Prize and what does it mean? and of course it's wonderful but as you just said in the end of the day, it's participating in the discovery process and seeing some of the knowledge you generated hopefully help people, which is really to me what was what's been most meaningful and so I'm just very excited to see this work hopefully help patients with chronic kidney disease.