VO Open: Data has helped researchers investigate and create precision medicine for the individualized treatment of cancer and heart patients. While evidence for genetic drivers of kidney diseases is substantial, much larger data sets are needed to untangle the complex interactions that lead to kidney injury. By combining clinical and genetic data from diverse patients, scientists can better understand how to make a more precise diagnosis of kidney disease in individualized care. Precision nephrology is an aspiration that would afford the right patient gets the right treatment at the right time. With me today is Dr. Ali Gharavi, Professor of Medicine at Columbia University and Chief of Nephrology. He’s the principal investigator for the Frenova Renal Genomics Registry. Ali, thanks for joining me today.
Gharavi: Thank you. Thank you for inviting me. It’s a pleasure to be here.
Maddux: Thanks. How can genomics help us actually get closer to precision kidney care?
Gharavi: Well, I think it’s very clear that when you talk to our patients with kidney failure, about one-third of them will tell you that they have a family member who also has some kidney disease. So, right off the bat, you can understand that there are hereditary factors involved in the development of kidney disease and until recently, we were struggling to figure out how much this contributed overall in terms of the molecular level. So, about 20 years ago, we got the first draft of the human genome. There are 3 billion letters or nucleotides in the human genome and from then on, then the real challenge was to sequence individual genomes at scale and in a cost effective manner and so, now, we have a technology to sequence individual genomes for under $1,000 and so, hundreds of thousands of individuals have been sequenced and from there, now, we can try to figure out what are the typos in this genomic alphabet that may lead to disease and so, now, a lot of kidney patients have been sequenced and we get a pretty good understanding that maybe roughly about 10% of cases have a diagnosable genetic disease that can then impact sort of their clinical care and can inform treatment in the long-term.
Maddux: Let’s step back for a second and just for the audience, describe the distinction between a genomics and genetic sequencing. To what degree are these omics broader than simply the pure base pair path?
Gharavi: So, I think genomics refers to the science of the genome. So, analysis of the nucleotide letters, how they come together, how they are-- how it helped the organization of the genome, the regulation of the genome, and so, all of this gets interpreted to make the building blocks for the cells in a human body and the organisms. Genetics refers to-- it’s a more clinical term referring to how variation in this genetic alphabet can lead to a particular trait or phenotype or sometimes to disease and so, it has a much more clinical connotation than the general overall science of genomics.
Maddux I know that you and I have had a discussion previously about whole genome sequencing versus exomes. Tell us a little bit about the distinction between what kind of research is done on exomes versus the whole genome.
Gharavi: So, as I mentioned, there are 3 billion nucleotides in our genome, but about 1% of them codes for protein. So, there are 20,000 genes in the genome, the building blocks of the genes that wind up coding for protein are called the exons and there are methods to selectively sequence the exons in a genome, so that 1% of the genome, and that’s called exome sequencing and that’s the portion of the genome that we can best interpret in the context of medical genetics and so, that’s the area where most of the focus on medical genetics that’s focused in terms of discovery of new disease genes or identification of previously undiagnosed, clinically undiagnosed genetic disorders. When we’re talking about the analysis of the entire genome, the rest of the genome is much more complicated to interpret right now in terms of its medical context, although there’s a lot of progress being made in terms of understanding how these non-coding regions of a genome regulate the overall organization and transcription of the DNA into RNA and then finally into protein.
Maddux: Over time, do you think our scientific interrogation of the whole genome will continue to evolve and improve so we understand the sort of epigenetic phenomena that might be impacted by the non-coding regions?
Gharavi: Absolutely. I think this is a large area of investigation across many groups around the world. It’s going to be challenging, but as we were accumulating more data across many human populations across different species, when we’re getting a lot better appreciation about the function of the non-coding region of the genome, again, just the sequence of variation in the sequence itself can alter sort of the expression of genes, how the genome and the DNA structure can fold together and large distant regions can potentially interact with one another to, again, impact gene regulation and from there, also understanding the epigenetics, as we say, how this genome can be modified by additional pathways and modification steps, which can then, again, alter the expression of genes and the function of the genome. So, I’m very optimistic that this will happen, but I think this will take a while and probably several decades until we get a better handle on it.
Maddux: Why is it so important to have sort of an ethnically diverse population of people to study and understand?
Gharavi: So, we’re all very diverse. Of the 3 billion nucleotides in the genome, maybe 1% of them is-- 1 in every 200, 300 nucleotides is variable across different humans and also, there’s a lot of variation across different populations. Humans evolved and developed in Africa somewhere around 100,000 years ago. A subgroup of humans then moved out of Africa. That’s just a small population compared to the remaining population in African and then as humans migrated across different parts of the world, something called genetic drift happened, where certain genetic variants were lost in some populations. Others accumulated over time and so, if we just sample one human population, you’re losing a large amount of variation. Having that understanding of that variation can really help in interpretation of the genome, understanding what’s the common variant, what’s the rare variant, and figuring out to better distinguishing what could be disease causing versus what could be essentially a neutral variant or a benign variant from that part and understanding all of this genetic variation can also tell you about the regulation of the genome and how also some areas of the genome have changed over time to, in some populations, to adapt to local environment as well and sort of some good examples of this relevant to kidney disease that has also been a very informative for us to understand the impact of genomics on kidney diseases.
Maddux: Does our genome change as we age or is it strictly the way the genome is expressed that might change as we age?
Gharavi: So, we have our germline sequence is-- we are born with a certain sequence. However, there are somatic mutations that occur in cells that grow and divide over time and so, these can have, for example, an immune system evolves over time and there are some adaptations that occur. For example, that’s how generate antibodies, by having somatic mutations that result in generation of specific antibodies to pathogens that we encounter. Somebody’s somatic mutations unfortunately can also lead to the deleterious consequences, such as development of cancer. So, yes, our genomics dynamic, though what we are born with is a certain-- is usually uniform sequence which can then change during different cell types across a lifespan.
Maddux: How far behind do you think kidney disease care is compared to oncology in creating this sort of genetic-driven precise treatment?
Gharavi: We’re probably, I would say, a decade or more behind in terms of-- that’s not unusual for-- oncology has been far ahead because they’ve understood that there’s a genetic basis to disease a long time ago. There were a number of germline mutations such as BRCA that were identified many years ago that sort of fueled the development of genomically driven research and therapeutics. But other areas of medicine have sort of lagged behind and nephrology is among them. I don’t think we’re farther behind than many other subspecialties, but we’re catching up quickly and I think that’s going to be one of the areas that’s going to be one of the most promising in the next few years for nephrology.
Maddux: Yeah. One of the things that’s been part of the sentinel components of the registry we’re developing is the belief that we’ll need large numbers of patients than we’ve been able to study so far. Can you just speak to the importance of patients and the interest in research for us to get to the number of patients that we need to get to have to do these interrogations?
Gharavi: It’s very clear that there’s a lot of diversity and heterogeneity in patients. As we know, we have people that will have diabetes or high blood pressure or different forms of kidney disease that we currently lump together for clinical trials, for research studies, and for clinical care, but we know everybody comes with a different path, a different story, and responds differently to medication. So, among those individuals, there’s a lot of heterogeneity and when we start doing genomic analysis, we realize that there are also a lot of people who have genetic diseases and there are lots of different genetic diseases, rare genetic diseases that can cause kidney failure, many of which are not recognized in the clinic based upon our standard clinical workup and so, by doing genome sequencing, we’re able to identify those, but because each one of these diseases individually are rare, we need large, large numbers to be able to put everything together and get past this heterogeneity and then try to follow up and see what happens to people in the long-term and that’s the power of having large data sets is that you need to have power to get past all of this heterogeneity and differences among individuals to get to an answer that might apply to large groups or also answers that might apply to those small groups of people who have a specific genetic signature.
Maddux: So, we’ve had a lot of interest in the renal community talking about APOL1 and those kinds of things. What are some of the genetic variations that you think have the greatest potential impact right now and sort of what excites you about particular genetic variants that you’ve seen? Any right now that come to mind?
Gharavi: Absolutely. I think APOL1 is arguably the best story in nephrology for the past half a century, perhaps. We hear every time-- among the dialysis service now, every time you go to a dialysis clinic, you see that there’s an over-representation of African Americans among the dialysis population no matter where you are in the country and so, that’s been attributed to many different factors over the years, but it’s very clear that this genetic variation, APOL1, really is a major contributor to the excess number of end-stage kidney failure among African Americans and so, this discovery about ten years ago, it changed our approached to understanding pathogenesis. Now, there’s a molecular handle to try to understand how these variants that evolved actually as a protection against sleeping sickness, but then are counterbalanced in a way, have come with a detrimental aspect if we should have two copies of these variants, you have a much higher risk of kidney failure. Martin Pollak’s group really identified this particular genetic variation and made a huge contribution to our understanding of kidney disease. So, we’ve evolved in ten years from having this understanding of this genetic diagnosis to now multiple companies having genotype-driven therapies available to see whether in fact if you knock down these particular pathogenic variants, you’re able to reduce the progression of kidney disease and treat patients. So, that’s really what’s going on in oncology. There’s genotype-driven therapy available to target the people with a specific type of genetic variation so that you can impact folks who really have-- who really have a specific molecular cause of their disease. So, I think that’s really exciting. The other area which has been really exciting, so that means there’s progress, has been in this disease that I’ve been studying called IGA nephropathy, which is a very complex disease, but based on genomic analyses that we’ve done in the multi-ethnic populations and over thousands of individuals involved, now, we have a better handle in terms of the pathogenesis and as a result of the genetic data, there are now clinical trials that are ongoing looking at specific molecular pathways that have been identified in the genetic data, the complement pathway looking at the IGA secretion pathway, looking at the intestinal IGA secretion pathway and so, I’m optimistic that there will be targeted therapy available this way as well.
Maddux: I’ve thought that our pathologic classification of diseases has been inadequate to kind of understand these multiple hits to the kidney that come from different pathways and mechanisms of injury. Do you think we’re going to change the way we talk about kidney disease from the traditional sort of nomenclature to something that’s more precise?
Gharavi: Absolutely. I think let me just say first off, the pathological classification is really a strength in nephrology, where we’re able to really distinguish people who have different forms of kidney disease. Let’s say we’ll present with peritinonium and we could say whether they have diabetes, FSGS, and other forms of glomerulopathy and then we can have some precise, more precise therapy available. But what we’re understanding now based on genetic testing is that even within these pathologic subtypes, there’s a lot of heterogeneity. For example, for FSGS, there are two dozen genes or more that have been identified that can cause genetic forms of FSGS and those are typically indistinguishable based on histopathological reading. So, I don’t think we’re going to completely do away with histopathologic reading because there’s a lot of information that we still get, for example, the extent of scarring and so forth in iso pathology, but coupled with genomic diagnostic, I think we’re going to have a much more powerful approach to be able to design clinical trials, target care for our patients, and think about their long-term outcomes.
Maddux: Yeah. So, it feels to me like a layering effect. We’ll get progressively more precise as we layer on more information to that histopathologic diagnosis.
Maddux: Final question before we end today and that is the work that needs to be done to research the genomic impact on the kidney requires a lot of computational biology, computational genetics. How are we training people and do we have enough people that have the skillsets that are needed to do the kind of research that is going to address these really complicated areas of science?
Gharavi: So, I think overall, there are a lot more-- there’s a lot more emphasis on bioinformatics, on data sciences nowadays across all fields, but probably not enough. I think nowadays, if you’re doing any kind of PhD in biology, you have to become familiar with bioinformatics tools with some computer coding to be able to really do the high-level science that one can aspire to make a difference and so, I think that’s going to have to be incorporated into any basic science curriculum going forward. Do we have enough folks who are focused on this in nephrology? Not enough, I would say, but that’s just because nephrology is just emerging as an area of interest where we see that there is major contributions from genomics and so, I think that will change over time and I think as people see the opportunities, they will gravitate to nephrology and that’s, I think, our job is to also advertise this as a major frontier for discovery and having impact on care in general. So, I think that’s-- undoubtedly, that’s what we need as a workforce.
Maddux: I hope that programs like this will help encourage people to think about that as they look at their career choices. Before we end, any final thoughts that you have as we at Fresenius Medical Care are sort of embarking on trying to help catalyze and be more sensitive to the development of these genomic and genetic resources?
Gharavi: I think that being able to engage patients in entering into any kind of studies and genomic studies, one, but also clinical trials is going to be really important in the field of nephrology and I think groups like Fresenius can make a huge impact on this. I think the way we made the comparison to oncology, we should be able to enroll all of our patients into some sort of registries so that we can learn something from the clinical course and then hopefully offer them also targeted therapies down the line so that we know exactly who to go back to to offer them some therapies and I think that would be a really important step for us to go through in molecular diagnostics and provide the next generation therapeutics for our patients.
Maddux: Well, Ali, I want to thank you for being with me today and for helping us sort of expose and understand a little more about the genomics of kidney disease.
Gharavi: Thank you. It was a pleasure talking to you.