Faster Cures, More Genetic Treatments: A Q&A With NIH’s Joni Rutter

You might have seen the name on a listing of NIH institutes and centers: the National Center for Advancing Translational Sciences (NCATS). “What the heck are they up to?” you may have thought. To answer that question, MedPage Today Washington Editor Joyce Frieden sat down with NCATS Director Joni Rutter, PhD. The interview, which was conducted with a public relations person present, has been edited for length and clarity.

MedPage Today: Hi, Dr. Rutter. Thanks for sitting down with me today!

Joni Rutter, PhD: Sure! Glad to be here.

MPT: Tell us what exactly NCATS does.

Rutter: NIH writ large is 27 different institutes and centers, and most of the NIH is really built upon the basic science work, and all the different biomedical fields. And most of the institutes and centers have a mission and focus on a disease or a system or organ, for example.

NCATS is one of the few institutes and centers that doesn’t have that focus. We don’t have the basic science focus, nor do we have the organ system disease focus. Instead, we take basic science and move it into the translational space. We do development of the preclinical work from target identification, assay development to test those targets, and then push them through the sort of preclinical pipeline for drug development.

MPT: So why can’t the institutes themselves do that work?

Rutter: The other institutes primarily focus on their disease of interest. And we make the distinction between translational science and translational research. Translational research is taking a particular project or program — like a very specific disease — and a target for that disease that’s been identified, and moving that along the translational pipeline. Moving it from one step to the next is translational research — and that, everybody [at the other institutes] does.

What we bring to the table at NCATS is, we do specific projects. But our goal of doing those projects isn’t just to advance them across the translational pipeline — it is to understand what is slowing it down from moving across that pipeline faster, better, and more efficiently. So our goal is to take the basic findings and turn them into health solutions. That’s our basic mission, and our goal in doing [translational science] is to bring more treatments for all people more quickly.

Since we’re not disease-focused, we study them all. We don’t want to be tied to any one particular organ or system. So even though we might study a particular organ or system or disease for [a specific] program, it is really to help us understand what the bottlenecks are. And then we tackle those bottlenecks, and other projects move forward and follow suit. So we’re trying to really work at dismantling those bottlenecks.

And it’s not just always scientific bottlenecks. We also take aim at operational bottlenecks — you know, developing contracts and the agreements that support contracts. We try to “template” those as much as possible, and to learn from our interactions with working with industry or other other biotech companies or partners to make sure that what we can do next time is shave months off of those agreements.

MPT: Tell us about your work with rare diseases.

Rutter: There are over 10,000 diseases, and most of those are rare. Quantifying the actual number of diseases is really hard, and with the advent of whole genome sequencing, we’re finding more rare diseases every year. One of our key approaches is to understand how we can address rare diseases by looking at them in a “more than one disease at a time” approach. And that allows us to kind of do one thing, so to speak, for different kinds of phenotypes.

The advantage of doing that is we get insights into common diseases. So, for example, hypercholesterolemia was a rare condition caused by a genetic mutation in families — they have just super-high cholesterol. So because of that, the drugs that are used to treat that disease — statins — now those aren’t used just for those rare disease problems in hypercholesterolemia, but they’re also used for [other] people who take a statin every day. The advantage there was that we were focused really on a rare disease, but actually, it turns out that we’ve been able to address a common disease.

GLP-1 agonists are another example; they’ve gone through an evolution. We started with Ozempic [semaglutide] and now we have things like Mounjaro [tirzepatide]. And there are more drugs in that pipeline, we have [about] nine pharma companies working in this space, and there are over a dozen of these drugs out there, and they’re coming in the pipeline.

So one might think, well, do we need all these drugs? Yes, I think what we’re seeing — and this still needs to be borne out in terms of the data — but we do see differences in who might respond well to those drugs and why. Ultimately, what we might start to see is that actually, these very common diseases like type 2 diabetes might have these very nuanced, rare kinds of sub-phenotypes in them that we can learn from.

So I think there are really advantages of understanding and following common diseases for the rare disease understanding, and then there are reasons to think about rare diseases and how they might impact common diseases. It’s really this cycle that I think is critical, where we we can play a role here.

MPT: I hear you have a project called “Translator” in alpha testing that looks very promising.

Rutter: The Translator is a compendium of publicly important biomedical databases. Things like PubMed, things like ClinVar — a compendium of all the variants that are associated with some clinical phenotype, the genetic variants. So there are a variety of genetic databases that are in there, and protein databases, for example. Translator puts all of these data together in an enclave database, and really smart people who are computational scientists are thinking about how we can leverage knowledge that’s in these databases — what’s been published, what’s known about particular diseases or genes, and then it looks for other similar knowledge … in other databases that might align with the particular one in a paper.

And by doing so you can ask a question: “OK, I’m thinking about Disease X; what compound might be effective for Disease X?,” and Translator essentially looks at the 150 or so databases in there and uses this AI [artificial intelligence] algorithm to look for those keywords and create knowledge graphs of those words and how they might intersect. And then it returns back information at the level of “Here are some possibilities that might answer your question.” And then it gives back perhaps compounds that can do so.

So one example [of its use] was this little girl with SHINE syndrome — it’s a very, very, very rare disease named because of its symptoms: S is for sleep disturbances, H is hypotonia, I is for intellectual disabilities, [N is neurologic disorder], and E is epilepsy. This disease looks to be caused by a mutation in the DLG4 gene.

And so the clinicians came to the Translator team and said, “Hey, I have this child with SHINE syndrome. And I want to know what might be helpful for it,” [but because] they knew the mechanism and that was very helpful, rather than putting in the disease, they put in DLG4 and then they said, “What can upregulate DLG4?,” so it was a very specific question.

Translator came back with a variety of answers. And the physician who was working with this patient saw that guanfacine, which is a normally a blood pressure drug, was on the list. And [the research team] understood that it’s a safe drug, and it wasn’t necessarily a clinical trial — just an off-label use of the drug to see what would happen.

And so the mother sent in photos of the child’s [picture] coloring, and before treatment, the coloring was very erratic — outside the lines and kind of all over the place — but after treatment, she sent in a coloring picture of Peppa Pig, and the mother’s words were, “We’ve seen such great improvement in Peppa Pig.” When I saw it, I thought, “Wow, those are inspiring words.” So that’s one example.

Another example is a teenage young woman who had periods of cyclic vomiting — this is also a rare condition. So this was terrible and she couldn’t keep weight on. And [her doctors] did the same thing [and sent in a question].

How your brain works in thinking about vomiting, there are immediate early genes that are upregulated. So Translator returned this idea that if you could inhibit that upregulation, that would prevent the cycle from completing and you could prevent the vomiting. Well, it turns out that isopropyl alcohol was identified as one of these [inhibitors]. So all they did to test this was they took an isopropyl alcohol swab, and they just had her sniff it, and that was enough to trigger the stoppage of the vomiting.

We’re in the alpha stage [with Translator], and it’s still a little clunky — we can’t ask it really hard questions yet, and we’re continuing to develop and refine the algorithms. But in the alpha phase, we’re encouraging people to look at two specific types of questions that they can ask. And by limiting it to the kinds of questions that can be asked, that’s how we’re starting to really refine the algorithms for moving forward.

MPT: So is Translator similar to ChatGPT?

Rutter: Well, it doesn’t “hallucinate,” because it’s based on real data, so it’s not making up data. Now, that said, I’m a huge ChatGPT fan, because I think it’s really powerful for some instances — but for this one, not so much.

MPT: I understand that gene therapy is another NCATS focus.

Rutter: Yes, there are a variety of programs that we have in — I like to call it the gene-targeted therapy space, because gene therapy is kind of limited to putting a gene into an AAV [adeno-associated viral] vector and then delivering it. But gene-targeted therapies are more about the idea of impacting a specific gene in some way. So it could be gene editing, it could be gene replacement, it could be some other ASO — antisense oligonucleotide — way of affecting gene function.

So we have three major gene-targeted therapy programs. One is through the NIH Common Fund program — this is something that no one institute or center could do alone, but we have to work together to generate something for all of NIH. And the Common Fund program has one on somatic cell gene editing, where we’re working with a lot of the big names in the field to understand how we can work in small animals and large animals to develop the technology for gene editing, including the molecular environment to be able to do the edits in cells, putting those into a delivery vehicle to get them to the right place to edit the right cells.

Right now it’s early in the development of these and so their fidelity for doing the edits is not great. But this field is moving fast, and I have no doubt that this will just get better and better.

Then there’s another program called the Bespoke Gene Therapy Consortium, which is an accelerated medicines partnership. This is a partnership that NIH has with the Foundation for the NIH. So NIH brings in partners from industry and other organizations, and also comes to the table with our own institutes and centers. We work together to identify, “What’s our big goal here as an AMP — accelerating medicines partnership — kind of group?” And for this AMP consortium, we have recognized that the way we deliver genes into cells is through an AAV vector.

You take an adeno-associated virus, which is a DNA virus, and you essentially take the virus out of the virus, so it’s empty. And then you put in the therapeutic gene, so to speak, to replace the gene that has the mutation in the cell. And then you infuse that into the patient.

And the AAVs have 12 different flavors, and they each have their favorite parts of the body that they go to. So we’re working on these AAVs, developing them so that it enhances the biology, because one of the key problems with gene therapy and AAV technology is that those empty capsules — infusing the genes into them isn’t highly robust. So we’re trying to understand the biology a bit more to get more of the therapeutic genes into those viral capsules.

The advantage of doing this is we don’t actually think AAV is going to be the technology of the future, but it is the technology of now. And so we can learn a lot from using this as a way of helping us enhance and streamline delivery approaches for the future, like mRNA or lipid nanoparticles or other kinds of delivery vehicles. If we can work out the processes for AAV — we know we have approvals at the FDA with AAV gene therapies, so we have a path that works now. If we can streamline that path even more, then that will be learnings we can carry over to other kinds of delivery when they’re more robust and accessible for us to consider.

The third program is PaVe-GT [Platform Vector Gene Therapy]. It’s kind of a sneaky, cool approach — PaVe-GT is modest, but mighty. We’ve chosen four diseases: two are organic acidemias — these are sort of metabolic types of diseases — and two of them are neuromuscular diseases, so those two diseases can be caused by multiple genes.

We’re picking two genes — one for each of the two diseases in each category — that cause those two diseases. And the idea of the platform is that we want to create one AAV and [separately] put the two different genes in them, and then test them [individually] in the individuals with that specific mutation in their genes. And what we hope then [is to see whether] two points make a platform.

The goal is to manufacture the one AAV capsid and put those two genes separately in them, and then deliver them to the clinical trials. And if we can show that we can streamline the vector manufacturing — same manufacturer, same processes to create the vector, same clinical design — then that’s a platform that can be used. And so if we can get that manufactured, we can actually streamline a lot of the work in the regulatory space.

MPT: Last question: Are our readers seeing any results of your work in their medical offices, hospitals, or clinics right now?

Rutter: Most of the work that they have probably seen has been in the pandemic. One in particular that physicians might be seeing or hearing about is ACTIV-6 [Accelerating COVID-19 Therapeutic Interventions and Vaccines] — it’s one of the clinical trials that was conducted by this public-private partnership that we had over the course of the pandemic. It was studying repurposed medications that are now being sent to a person’s home for them to then take them at home and answer questionnaires. It’s all very low-touch and decentralized — they don’t have to come to a clinic or be seen by a doctor, and if they have mild symptoms, they don’t need to be admitted [to a hospital] or anything.

We’ve tested six different medications over the course of the program. We started with ivermectin, so that’s where they may have heard quite a bit about this. Ivermectin was the one that was first picked, because there was just data all over the map, plus people were prescribing it without really knowing. And so we wanted to do a definitive clinical trial, and we found that it didn’t have any benefit over other medications.

Now we’re doing the last one, which is on metformin. Metformin does seem to have these sort of anti-inflammatory effects, but nobody really knows what’s going on. In a very small study, it was shown to be effective for preventing long COVID and maybe even COVID, as well as reducing symptoms for COVID. So we’re doing a more definitive test for metformin in 3,000 people, and we’re looking at reducing COVID symptoms over the course of 28 days, [and also at] hospitalization and mortality.

Then we’re also going to look at long COVID effects — we’re going to follow these participants for 6 months to understand how their experience has been related to long COVID.

The other big thing that we have done during COVID is that we amassed the largest collection of electronic health records to evaluate over the course of COVID. We now have over 20 million records within this database, and we have CMS [Centers for Medicare & Medicaid Services] data in it as well. So we can really look at a lot of different data, and it’s all anonymized, and I think about 9 million right now are COVID-positive. So it’s a good number of COVID-positive patients.

We actually have 110,000 patients who were prescribed metformin after they got a COVID diagnosis. And so we are able to look at — a “clinical trial emulation” is what we call it. It’s all just data from their electronic health records. And we’re able to follow them and what their experiences are when they’re on metformin by just looking at their electronic medical records.

We think we might have a way to actually add to the body of literature in using this kind of new tool, and I think that’s super exciting.

MPT: Well, thank you for speaking with us today about all of this!

Rutter: My pleasure.

Please enable JavaScript to view the comments powered by Disqus.