Studying the unique chemical fingerprints left behind by cellular processes is an area of opportunity in the fight against chronic kidney disease. In this episode of Global Medical Office Dialogues I talk with biochemist Nadia Grobe, supervisor of lab research at the Renal Research Institute, and an expert in metabolomics. Dr. Grobe and her colleagues are exploring ways metabolomic signatures can be used to deliver more personal, precise care for patients living with kidney disease. By learning from these unique biomarkers researchers are at the forefront of developing the future of personalized medicine.
Dr. Maddux: Let’s start with your sort of assessment of what metabolomics is and can you describe for our audience a little bit about this particular discipline?
Nadja Grobe: Absolutely. Thank you, Dr. Maddox. I’m really excited to be here today and talk to you about metabolomics. It’s a very, very exciting field of study for me and for the community. It’s basically a large-scale study of chemical processes involving metabolites and these are the actual small molecule substrate, intermediates, and products of metabolism. So, we measure those small molecules within the cells, biofluids, tissues, or even within organisms and their interaction with the biological system, which we call all together the metabolome.
Dr. Maddux: Tell me a little bit about, I understand there are two primary techniques for measuring the metabolome. There’s traditional mass spectrometry and there’s magnetic resonance spectroscopy. What do you use and just describe your lab setup a little bit.
Nadja Grobe: At the RRI Laboratory, we went with mass spectrometry. It is due to many different reasons. Also, it doesn’t require a huge magnet and I’m also more trained in that technique and I believe that the sensitivity of the mass specs you can purchase nowadays is a little bit better than using nuclear NMR. But both techniques are absolutely feasible and sometimes even a combination of both would be very advantageous in studying the metabolome.
Dr. Maddux: Tell me about your background. Your background is in biochemistry, I understand. Is that correct?
Nadja Grobe: That’s correct. Yes. I actually-- I’m a biochemist by training, got my master’s and PhD in biochemistry and actually started working as a, believe it or not, plant physiologist and slowly went into animal biomedical research and then ended up here now with always a focus on kidney disease with biomedical research, once that started, but also, really using mass spectrometry. For some reason, all these years, I’ve used this to discover biological function or use it for our research project.
Dr. Maddux: So, the human metabolome database has about 40,000 metabolites registered in it at this point. Do you think our patients with kidney disease have any distinct metabolites compared to the general population or what’s unique about kidney disease in wanting to look at what the metabolomic profiles look like?
Nadja Grobe: Yeah. Absolutely. It’s great that you bring up the database. It’s actually a free database and I heard that last time I checked, there were even more metabolites and the kidney, as you know, has a really large impact on levels of circulating metabolites. So, with that database, you can actually-- either you can do targeted or untargeted so that these are two strategies that you follow in doing metabolomic analyses. For targeted, you would probably, in terms of kidney disease, look more into perhaps uremic toxins and known biomarkers that are indicative of progressing kidney disease or even treatment and with untargeted, you basically do an unbiased examination of the whole metabolome of a patient and you would not necessarily test a hypothesis, but you really characterize biological fluids, for example, from individuals with kidney diseases and then those metabolites that are identified may even characterize novel metabolite perturbations that we didn’t even know about before doing these types of analyses.
Dr. Maddux: So, the metabolome is part of this larger group of omics that we are studying today. So, we talk about not only metabolomics, but proteomics and genomics and transcriptomics and I’m just curious how you see all of this fitting in in this concept that I know I’ve been interested in around precision medicine and the ability to actually understand the individual retreating. Can you talk a little bit about the clinical relevance of this?
Nadja Grobe: That’s a great, really good question. I actually think that all together, these omics technologies-- we call this a systems biological approach-- could really benefit the current practice when we diagnose diseases or treat patients to see how therapy works with each patient. Metabolomics is really the youngest technology of all these omics approaches and I am a big supporter of looking into all and using all to really characterize disease. I think that these omics approaches, you actually allow the analysis of large cohorts across different phenotypes and really correlate them to longitudinal outcomes. What I really like about metabolomics is that it’s actually really close to the phenotype, molecular phenotype of a patient. So, the metabolites are known to have various functions within the biological system. They are involved in the structure and signaling and have effects on enzymes, are sometimes cofactors of enzymes and those things are really very well-reflected when you characterize the balance of metabolites. So, if you just look at diabetic patients, they have a problem in their body that causes their blood sugar levels to rise higher than normal and we currently consider hypoglycemia as a biomarker for diabetes and that’s similar to metabolomics or even other omics approaches that a set of very well-studied metabolites or biomarkers could be used to diagnose diseases in these perturbed pathways that cause the disease.
Dr. Maddux: It’s clear that as we develop more precise methods for all the types of injury and mechanisms of injury to the kidney that we are expecting, we’re going to need to have a much larger number of biomarkers, where we can track whether we’re actually positively impacting a particular pathway or a particular mechanism. What are some of the things that you all are working on today that might actually be relevant in the short-term for our patients with kidney disease? How do you see your metabolomic research directly affecting our kidney patients in the next couple of years?
Nadja Grobe: We have a few studies ongoing right now where we use metabolomics actively to not only characterize patients, but also incorporate these metabolic signatures into predictive modeling. So, if I may come up with an example, one of our biggest, most ambitious projects right now is in this space of peritoneal dialysis, patients that have negative clinical outcomes could be potential characterized according to their metabolic fingerprints in their dialysate, in their effluent and we are using machine learning algorithms to look at all these different metabolic signatures and then assign them to different clinical phenotypes so they actually predict when a patient changes into some sort of negative clinical outcome. This is pretty exciting and we are fortunate to have this type of large cohort study set up, being set up at the RRI.
Dr. Maddux: Well, I know that in taking care of patients on peritoneal dialysis for years, so much of the issue that clinically we have to deal with is their loss of membrane function over time and this ability to understand what are the phenotypes that match to certain metabolic profiles that would tell us that a patient is either losing membrane function or has less likelihood for prolonged membrane performance in peritoneal dialysis. I think these are really quite relevant to what we as clinicians would be seeing. So, it’s very interesting. Do our patients with kidney disease over time change their metabolic profile? We know that their kidney disease itself is a metabolic insult to their system, but do we actually see that their metabolomic profile changes as they increase their vintage on renal replacement therapy? Is there any study of that that’s gone on?
Nadja Grobe: Yeah. Absolutely and there is literature already starting to look into those other research groups and also, we are looking into this. In fact, we have some data to support that longitudinally, those patients actually progress into different stages of metabolic signatures, which would reflect their overall health or pathologies or perhaps even comorbidities on underlying diseases that they have to deal with. We also have some studies where we look at metabolites that actually are removed throughout dialysis, in this case hemodialysis and we are lucky enough to work with our RRI clinics that are in New York and have done some studies with patients during dialysis so we can clearly see those profiles, that they are not only changing during dialysis, but also over several months of dialysis over the vintage of these patients and that’s quite interesting to see how these profiles are changing, which we don’t know enough yet to really characterize different signatures to understand these changes, but again, lucky again, we have all the data available of these treatments. We know a lot about the patients. Dialysis patients are monitored. I think they are the most monitored patients in the world, at least in our facilities and this opens up so many opportunities for correlation analysis or data analysis, advanced analytics.
Dr. Maddux: What are the difficulties? What are the challenges that you and your team face in actually trying to uncover this field to where it’s understood better? Because it is really a fairly new area of analysis for patients with kidney disease.
Nadja Grobe: The biggest limitation I’ve seen so far is that different labs, different research groups have different ways of doing things and that, as you can imagine, causes maybe some limited synergies between identified patterns. But I think there is a lot of effort in the community to create repositories like the human metabolome database that you mentioned, also reference materials that could be used by all these research groups to really fine-tune their system and say “Hey, if I find these thousand metabolites, then I’m good to go with my central analysis.” I also think that one biggest challenge is that we do a study. We publish that study, and then we move on to the next study. I think we are in a unique position. We just started a clinical trial that will go over three to four years where we follow patients repeatedly, draw samples and that will be just such a unique situation, where we just don’t do a study, publish, and we are done. This is a commitment on our side to improve patient outcomes since we have access to all the other data.
Dr. Maddux: I think you’re describing what I think is one of the unique features of Fresenius Medical Care and RRI as an institution is it’s not-- it is about the science and advancing the science, but it’s really about how you apply the science and your unique team has the opportunity to look at how do we create applied actions to metabolomic information as it’s developed and I think that’s a great opportunity to really blend the pure science aspect of this, which there’s plenty of, to the actual need that our patients have in trying to let us improve their lives and the treatment, safety, and all the things that go into the care that we give them. Tell me a little bit about what you’ve been doing during the pandemic at this time and how you’ve kept things going.
Nadja Grobe: We were ready to be virtual. So, we had to split up our teams and we started-- what you just said is really important, Dr. Maddox. It’s something that is really dear to my heart but also my team’s heart is to apply our skills to improve patient outcomes. We just don’t want to do research with no bigger impact. So, when COVID-19 happened, it was very clear to all of us in the lab that we really want to contribute. So, we have a very diverse team. We have a biomedical engineer, who started 3D printing screens and then she teamed up with the clinical research coordinator, who helped with this effort, and then we have lots of skills in terms of molecular biology, but also protein biochemistry. So, we started moving into COVID-19 research. We teamed up with our mathematics professional, Alhaji Cherif, and started looking into pool testing and published an article around that. So, it was challenging to be in two different-- basically, in two different shifts and schedule our way around lab work because we are bench people. We have to be in the lab to do work. We can’t just sit like on the computer and do data analysis. Even for that, you need data, right? So, we were-- we managed. It took us a few weeks, but we were really excited about the opportunities and also, what this has given us in terms of feedback and reward with all the projects we were involved during COVID-19.
Dr. Maddux: I think that the pool testing work that you all did and the publication that you had in JAMA Open Networks was a pretty sentinel paper with regard to this and extremely timely, as it wasn’t just a week or so after that article came out that there was quite a bit of public discussion around the concepts of pool testing and I think that will continue to be quite relevant as this pandemic evolves over the coming months and years. I think everybody is very proud of the work that you all are doing and I appreciate the chance. I’m here today with Dr. Nadja Grobe, who leads the metabolomic research team at Renal Research Institution. Thanks so much for being with me today.
Nadja Grobe: Thank you very much for this wonderful opportunity.